51507 IDA15 MID-TERM REVIEW The IDA15 Mid-Term Review: Aid Effectiveness International Development Association Operations Policy and Country Services (OPCS) October 2009 ABBREVIATIONS AND ACRONYMS AAA Accra Agenda for Action AE Aid effectiveness AEAP Aid Effectiveness Action Plan AFR Africa CAS Country Assistance Strategy CAW Country analytic work CFP Concessional Finance and Global Partnerships DAC Development Assistance Committee DPL Development policy lending DPO Development policy operation EFA Education for All ESW Economic and sector work FCC Fragile and conflict-affected countries FDI Foreign direct investment FTI Fast Track Initiative GPF Global Program Fund HLF High Level Forum IBRD International Bank for Reconstruction and Development IDA International Development Association IEG Independent Evaluation Group IFC International Finance Corporation IHP International Health Partnership IMF International Monetary Fund IRS Internationally recruited staff ISN Interim Strategy Note JAS Joint assistance strategy JFA Joint Financing Agreement LEG Legal Vice Presidency LHI Legal Harmonization Initiative MDG Millennium Development Goal MOU Memorandum of Understanding ODA Official development assistance OECD Organization for Economic Cooperation and Development OPCS Operations Policy and Country Services PBA Program-based approach PFM Public financial management PIU Project implementation unit PMA Project management arrangement PRSP Poverty Reduction Strategy Paper PSIA Poverty and Social Impact Analysis SEETF South-South Experience Exchange Trust Fund SWAp Sectorwide approach VPU Vice-presidential unit WBI World Bank Institute WP-EFF Working Party on Aid Effectiveness IDA15 MID-TERM REVIEW: AID EFFECTIVENESS CONTENTS Executive Summary ....................................................................................................................... i I. Introduction ...............................................................................................................................1 II. Strengthening Country Ownership ........................................................................................3 A. Experience with Joint/Collaborative CASs....................................................................4 B. Improving Project Management Arrangements .............................................................7 C. Application of Conditionality.......................................................................................11 D. Country Ownership: Way Forward .............................................................................11 III. Building More Effective and Inclusive Partnerships .........................................................12 A. Aid Architecture and Nontraditional Partners .............................................................13 B. Legal Harmonization ....................................................................................................18 C. Leveraging IDA Resources ..........................................................................................20 D. Partnerships: Way Forward ..........................................................................................21 IV. Delivering and Accounting for Development Results .....................................................21 A. Aid Predictability .........................................................................................................21 B. Poverty and Social Impact Analysis.............................................................................25 C. Transparency, Accountability and Results: Way Forward..........................................26 V. Shaping the AE Agenda and Supporting Implementation ................................................26 A. Matrix of Actions and Action Plan ..............................................................................27 B. Regional Reporting ......................................................................................................27 C. Decentralization ...........................................................................................................27 D. Staff Incentives and Capacity .......................................................................................28 E. Implementation: Way Forward....................................................................................30 VI. Conclusions ............................................................................................................................31 Boxes Box ES-1. IDA15 Aid Effectiveness Commitments at a Glance................................................... ii Box 1. Monitoring the Use of PIUs: Bank Performance .................................................................8 Box 2. Examples of CPPRs and CASs that Address PMA Issues ...................................................9 Box 3. Emerging Lessons from PMA Experience .........................................................................10 Box 4. The South-South Experience Exchange Trust Fund ..........................................................17 Box 5. EFA and IHP: Country Priorities to Drive Global Programs .............................................18 Box 6. Areas Covered in MOUs ....................................................................................................19 Box 7. The Legal Harmonization Initiative ...................................................................................20 Box 8. Aid Predictability Measures and Bank Performance .........................................................22 Box 9. Main Areas of Focus for the IL Reform--Initial Phase ....................................................30 Box 10. OECD-DAC Working Party on Aid Effectiveness and the World Bank .........................31 Box D1. MOUs with Flexible Frameworks ...................................................................................53 ii Figures and Tables Figure 1. Collaborative Activities at the Country Level ..................................................................5 Figure 2. Work with Nontraditional Partners at the Country Level ..............................................17 Table 1. Value-added of Collaborative Country Strategy Activities ...............................................6 Annexes Annex A. The Surveys on Monitoring the Paris Declaration: Methodology and Bank Performance...................................................................................................................33 Annex B. IDA15 Replenishment Commitments -Monitorable Actions .......................................42 Annex C. Predictability and Aid-on-Budget Measurements ........................................................44 Annex D. Challenges with MOUs ................................................................................................51 IDA15 MID-TERM REVIEW: AID EFFECTIVENESS EXECUTIVE SUMMARY i. This report reviews the Bank's progress in meeting commitments for the IDA15 Replenishment related to aid effectiveness (AE), discusses what we have learned from the experience in selected areas--in the context of progress and challenges in the overall aid effectiveness agenda--and draws on this experience to point to some conclusions for the Bank going forward. ii. The Paris Declaration provided the context for the commitments made in January 2008 under the IDA15 Replenishment--since then the AE agenda has continued to evolve. Some eight months after the IDA15 Replenishment, the Third High Level Forum (HLF) was held in Accra, Ghana. The detailed commitments resulting from Accra reflect a strong message from partner countries calling on all development partners to move to a higher level of reform in order to make real breakthroughs on the aid effectiveness agenda. Together with the Paris Declaration, the Accra Agenda for Action (AAA) laid the foundation for the Bank's Aid Effectiveness Action Plan (AEAP), articulated earlier this year. The AEAP also builds on feedback from the Survey on Monitoring the Paris Declaration highlighting areas of strength as well as priority areas in which the Bank will need to take further action. iii. During the initial years of implementing the Paris Declaration, the Bank instituted changes in some of its key instruments, notably DPOs and CASs, as well as approaches, such as an operational results-orientation. These reforms resulted in some significant initial successes in achieving aid effectiveness objectives, but overall the Bank has demonstrated somewhat uneven progress across the Paris commitments, even though it has fulfilled IDA15 commitments. At this juncture, there is a need for further policy change--currently envisaged under the investment lending reform and other corporate initiatives--to move to a more systematic mainstreaming of AE on the ground. iv. Progress on IDA15 Replenishment Commitments. All 12 of the IDA15 Replenishment commitments related to aid effectiveness have been fully met, are in progress, or are ongoing activities (see Box ES-1 below). This report summarizes the work related to the 12 commitments, organized under the three major pillars of the AAA. Some commitments are treated only briefly (notably in areas where papers have recently been discussed by the Bank's Board and/or where specific topics are addressed in other IDA15 Mid-Term Review (MTR) papers) while other issues receive a more detailed examination. It draws key lessons and identifies actions related to specific issues as well as work needed internally to support implementation, building on the Bank's Aid Effectiveness Action Plan. ii Box ES-1. IDA15 Aid Effectiveness Commitments at a Glance Commitment Status Strengthening country ownership Update good practice note on joint/collaborative CAS In progress Implementation guidance on project implementation units Ongoing Review conditionality Completed Building more effective and inclusive partnerships Incorporate nontraditional partners and undertake analytic work on aid Ongoing architecture Survey IDA leverage through cofinancing Completed Work on harmonized approaches to MOUs in cofinanced operations Ongoing Delivering and accounting for development results Review predictability of IDA disbursements Completed Update good practice note on PSIAs Completed Shaping the AE agenda and supporting implementation Update progress on the Matrix of Actions and develop Bank Action Plan Completed Reporting by Regional Management to the Board Ongoing Enhance staff incentives and guidance on ownership, harmonization, alignment Ongoing Decentralization: Specific targets for the Africa Region Completed Continue to explore further options; complete analysis of different models Ongoing A. Strengthening Country Ownership v. Reflecting the spirit of the Accra agenda, the Bank is focusing its AE work on the country and its capacity to lead and manage the development process. It gives priority to strengthening country ownership and to supporting governments in developing their capacity to identify strategic priorities, set out and measure results, and lead donors; and it is working to increase its reliance on country financial management, procurement, and safeguard systems to implement development programs and manage aid. vi. Wide Range of Activities. The Bank has reviewed its experience with collaborative country strategy work and is updating its good practice note, emphasizing engaging in high- value-added activities in this area, with the country leading in each activity. To help implement projects through a greater use of government structures--thus increasing ownership and sustainability--the Bank will examine options to better support government sector programs; shift from supervision to implementation support under the IL reform, and support government capacity for project implementation; and better align salaries of PIU staff to that of civil servants under civil service reform. The Bank continues to prepare its development policy operations (DPOs) in line with its good practice principles for the application of conditionality: reinforcing ownership, harmonization, customization, criticality, and transparency and predictability. The AEAP also includes additional actions through which the Bank is supporting country ownership: (a) supporting government capacity to consult civil society and other institutions (such as local governments and parliaments); (b) continuing to disseminate good practices and support countries to make national development strategies more operational, including through better iii prioritization and stronger links to budget and results frameworks; and (c) monitoring the extent to which Country Assistance Strategies (CASs) and DPOs reflect national priorities. vii. Focus on Use of Country Systems. The use of country systems for procurement, financial management and environmental and social safeguards is a key area of focus for the Bank. In all three areas, the Bank has developed policy frameworks for its reliance on countries' systems, and it has set up work programs to move in this direction. The Bank will deepen the already substantial achievements related to financial management, giving particular attention to the development of and reliance on national audit capacity, in the private sector and in particular in national Supreme Audit Institutions. In procurement, the Bank will continue to encourage the greater use of national procedures under national competitive bidding, when feasible (especially in IDA countries). It will build its program of pilots in international competitive bidding and will develop a strategy--including assessments and implementation support--for countries not presently included in this pilot, especially IDA countries. It will continue to scale up its program to pilot the use of country systems to address environmental and social safeguard issues from the project to the country level, and has also developed an assessment tool (the Safeguards Diagnostic Review) which is carried out between the Bank and its clients and is made public. CASs will provide the vehicle to raise issues related to the use of country systems and to consider related capacity development efforts. To support the international agenda, the Bank has been very active in the OECD-DAC cooperation fora in the area of use of country systems, through which diagnostic tools have been developed that permit partner countries to evaluate their capacity building and policy reform against agreed international standards and identify areas for development. These tools lay the basis for donors to develop their policy frameworks for the use of country systems; it is critical that donors work in harmony to support capacity building in partners and then move to reliance on a country's system once it has reached the mutually agreed standards. B. Building More Effective and Inclusive Partnerships viii. The Bank continues its active coordination with traditional donors--at the country level (for example, via collaborative strategy and budget support activities) and at the institutional level (for example, via the Legal Harmonization Initiative)--and is deepening its partnerships with nontraditional partners through a range of activities at the country and global levels. Its review of selected country programs has shown that every dollar of IDA aid leverages from $1.70 to $2.50 in aid from other sources, and the increasing use of program-based approaches allows Bank country knowledge and technical expertise to be shared across a broad array of activities and partners. The Bank continues to focus on developing the capacity of nontraditional partners in their role as donors, including through knowledge sharing and supporting and facilitating learning from South-South partnerships (and notably support for the Colombia High Level Meeting on South-South Cooperation); working with new donor countries and global programs to improve country alignment; and working with recipient countries to support engagement with nontraditional partners. Going forward, the Bank will continue to consolidate experiences related to working with nontraditional development partners at all levels, and support donors and recipients in building new approaches to the development partnership. To improve partnerships in joint financing at the country level, the Bank will continue to contribute to the Legal Harmonization Initiative to propose common donor guidance on the use of Memoranda of Understanding (MOUs). This will complement the investment lending reform iv effort, which will examine options for a new programmatic or results-based financing instrument which may facilitate joint financing. And finally, as the AEAP notes, the Bank is actively investing in partnerships and harmonization in the context of fragile and conflict-affected countries, using CASs to emphasize selectivity and country-level division of labor. It will continue to work closely with the United Nations (through the UN-World Bank Partnership Framework for Crisis and Post-Crisis Situations), multilateral development banks, the International Monetary Fund, the European Commission, and the OECD-DAC in many aspects of its fragility and conflict work. C. Delivering and Accounting for Development Results ix. The Bank reviewed its record with respect to predictability as part of the overall agenda on transparency, accountability and results. This exercise has highlighted the importance of using a measure of predictability that appropriately reflects donor and country performance based on the needs of recipient governments. The Bank's performance on aid predictability has been relatively good, but it can and will improve the reliability, availability, and timing of disbursement data and better communicate this information to governments. With the ultimate objective of strengthening accountability, the Bank will work with international initiatives that focus on providing reliable and transparent aid data. These efforts to make donor funds more predictable need to be coupled with efforts to develop country capacity to better integrate aid management with budget management and with overall support to countries to improve budgeting systems and medium-term fiscal frameworks. x. Transparency, Accountability, and Results. Beyond predictability, the Bank is supporting many areas to further the transparency, accountability and results agenda. It has issued its updated good practice note on poverty and social impact analysis (PSIA), which focuses on building country capacity to carry out PSIA and encourages making PSIA findings available for informed public debate by a wide range of stakeholders. In addition, as the AEAP indicated, a revamped disclosure policy may soon bring the Bank to a new level of transparency, increasing the accessibility of information to diverse external actors. The Bank will also continue to participate in mutual accountability reviews at the country level, drawing on the good practices for these exercises that others distill. In parallel efforts, the Bank will continue to mainstream governance and anticorruption, for instance through the issuance of guidance notes as part of the GAC agenda and through implementation of the IDA controls review. Finally, the Bank is implementing the results agenda within the institution and is supporting governments in their efforts to better link aid effectiveness to development effectiveness, aided by the new Statistics for Results Facility. The report on results prepared for the IDA15 MTR showcases this agenda with a focus on country-level as well as agency effectiveness. D. Shaping the AE Agenda and Supporting Implementation xi. A key commitment for IDA15 and for improving Bank progress on the aid effectiveness agenda is related to Bank staff incentives and capacity to implement aid effectiveness, and deepening decentralization through the achievement of specific targets in Africa and the analysis of decentralization approaches: these are ongoing activities. As reported to the Executive Directors in June 2009, the Bank has met its commitments related to decentralization. For the Africa Region (AFR), the number of internationally recruited staff (IRS) based in the field v increased between FY06 and FY09 by over 50 percent. In addition, the percentage of tasks (appraisal, supervision, and analytic and advisory work) that AFR has devolved to the field has climbed nearly 47 percent in FY09. Since the end of FY06, IRS in IDA countries outside of Africa have increased by 21 percent; also IRS in fragile and conflict-affected countries (FCC) have increased by 28 (30 percent) and three new offices--in the Solomon Islands, Tonga, and Samoa--have been opened. The Bank is also exploring ways to develop the next generation approach to decentralization. The objective is to bring the Bank closer to clients and partners through greater field presence, while complementing existing services provided by World Bank Group offices with more emphasis on global expertise and staff. Issues currently being examined include the questions of costs, numbers of technical staff, recruitment, and knowledge sharing across countries. The Bank is placing particular emphasis on (a) increasing staff in and devolving tasks to country offices in FCC and low-income countries, which face particularly difficult development issues and have limited capacity to deal with them; and (b) creating better accessibility to staff with specialized knowledge and skills. xii. A survey of Bank staff undertaken for this report indicate that they have been highly motivated by the potential benefits of the AE agenda, and in particular by seeing concrete results at the country level. They feel they have a good understanding of the agenda and that largely they have the support of Management. However, they seek a more consistent signal on risk- taking in the institution and more guidance on how to implement the AE agenda. The Bank is acting on these challenges. It is striving to realign the risk/results approach to its operations: a significant effort has been made to improve the focus on results at the project, country, sector, and corporate levels and investment lending reform is focusing on defining a clearer typology of risk which will enable staff to better identify and mitigate different risks related to development objectives in lending operations. Changing the Bank's culture related to risks and results will provide an important signal to staff about Management's priorities. xiii. In parallel, increased efforts will be made to support staff in implementing the AE agenda, especially in IDA countries. As the international community moves from isolated examples of good practice to mainstreaming aid effectiveness, the Bank will focus on gathering the rich experience of staff, donors, and countries to facilitate implementation and increase traction at the country level. This is an area of emphasis in the AEAP, which notes that, the Bank is increasing its focus on knowledge and learning related to AE, and it will provide more updated examples of experience through its revamped AE website. The Bank will also continue with its internal monitoring of the AE agenda and provide feedback to staff showing where progress has been made and where areas for improvement remain. In addition, the Bank is working with the international community--notably in its close involvement with the Working Party on Aid Effectiveness (WP-EFF) hosted by the OECD-DAC, and with other global initiatives--and contributes to advancing technical issues and identifying global good practices through these efforts. E. Ongoing Effort xiv. The agenda going forward--reflecting the AAA--highlights the centrality of the country: country ownership, leadership and the development of capacity and systems. In AE efforts, donors need to keep in sight the primacy of country leadership, leveraging joint donor activities to strengthen country capacity. The relevance of different aspects of the AE agenda varies across vi client segments, as do the solutions; partner countries are seeking pragmatic solutions tailored to country conditions. Just as countries differ from one another, so do donors--some aspects of aid effectiveness are more relevant to some donors than others, and some aspects of donor coordination will yield more results than others. While the Bank has met or is meeting all the commitments under the IDA15 Replenishment, it will continue efforts by implementing the actions outlined in this paper and in the AEAP. Aid effectiveness comprises a broad agenda, and for the Bank it is not an isolated agenda: it cuts across and draws on major corporate initiatives such as reform of disclosure policy and investment lending, the results agenda, IDA controls, decentralization, and the fragile and conflict-affected countries agenda. These efforts will bring the Bank even closer in alignment with clients, realign the Bank's approach to risks and results, and make the Bank more transparent and accountable for concrete results--thus deepening and mainstreaming aid effectiveness. Going forward, the Bank will integrate new knowledge and good practice as it is developed--focusing on those priorities that will yield the highest value- added to countries, weighing costs and benefits, and balancing the ideal and the practical. IDA MID-TERM REVIEW: AID EFFECTIVENESS I. INTRODUCTION 1. This report reviews the Bank's progress in meeting commitments for the IDA15 Replenishment related to aid effectiveness (AE), discusses what we have learned from the experience in selected areas requested by the IDA Deputies--in the context of progress and challenges in the aid effectiveness agenda overall--and draws on this experience to point to some conclusions for the Bank going forward. 2. Evolving AE Agenda. Starting with the 2002 meeting in Monterey, Mexico, agreement among stakeholders on principles and approaches to aid effectiveness has been broadening and deepening. The Second High Level Forum (HLF) in Paris led to the endorsement of the Paris Declaration on Aid Effectiveness, which set out five principles of aid effectiveness: ownership, harmonization, alignment, managing for results, and mutual accountability. This provided the context for the commitments made in January 2008 under the IDA15 Replenishment--since then, the AE agenda has continued to evolve. 3. Some eight months after the IDA15 Replenishment, the Third HLF was held in Accra, Ghana, involving bilateral and multilateral donors, low- and middle-income country and fragile state representatives, civil society organizations, and representatives of other development partners. Building on the Paris Declaration but reflecting the broader aid architecture and increased engagement by developing countries, the Accra Agenda for Action (AAA) focused on three major pillars: (a) strengthening country ownership; (b) building more effective and inclusive country partnerships; and (c) delivering and accounting for development results. The detailed commitments resulting from Accra reflect a strong message from partner countries calling on all development partners to move to a higher level of reform in order to make real breakthroughs on the aid effectiveness agenda. The need for accelerating aid effectiveness progress has been underscored in the past year as the global crisis has made clear the necessity of reducing aid transaction costs and increasing the focus on results. Together with the Paris Declaration, the AAA laid the foundation for the Bank's Aid Effectiveness Action Plan (AEAP), articulated earlier this year. The AEAP also builds on feedback from the Survey on Monitoring the Paris Declaration (see Annex A) highlighting areas of strength as well as priority areas in which the Bank will need to take further action to move toward fulfilling its AE commitments. 4. Institutional Environment. During the initial years of implementing the Paris Declaration, the Bank instituted changes in some of its key instruments, notably DPOs and CASs, as well as approaches, such as an operational results-orientation. These reforms were framed in the spirit of ownership and outcomes. And they encouraged speed, adaptability, predictability and accountability--features which are key to the Bank's clients and to its effectiveness. They resulted in some significant initial successes to achieving aid effectiveness objectives, notably leveraging ownership and aligning with country priorities, which have been confirmed in the recent Retrospectives on CAS and DPLs. But overall the Bank has demonstrated somewhat uneven progress across the Paris commitments, even though it has fulfilled IDA15 commitments. 2 5. At this juncture, there is a need for further policy change to enable a leap forward on the aid effectiveness agenda and to move from ad hoc implementation on the ground to a more systematic mainstreaming. The institution will need to focus clearly on defining the results-risk profile of its activities: what is to be achieved with support from Bank services coupled with keen attention to the specific risks involved--and the main focus needs to be on the Investment Lending (IL) instrument.1 As will be noted in many sections of this report, the reform envisaged for this instrument will reduce the rules-based focus of the institution and shift to a risk-based focus--and it will shift from a focus on compliance and oversight during supervision to a focus on implementation support and institution building. Ultimately the reform agenda is expected to examine new approaches to lending to facilitate more programmatic and results orientation-- especially at the sector level--without losing critical attention to good fiduciary practices. In tandem, further refinements of the CAS and DPO instruments will be undertaken to strengthen their results frameworks and their use as instruments for harmonization and alignment. These fundamental changes in Bank policy and practices, together with work being currently undertaken in key corporate initiatives--Managing for Development Results, Fragile States, Disclosure Policy, Decentralization and the Governance and Anti-Corruption Strategy--will provide major support for mainstreaming the aid effectiveness agenda going forward. 6. Structure of the Report. The Bank's work to fulfill its IDA15 commitments has been part of a broader effort to move the AE agenda forward--as such, these commitments are reviewed in the context of the evolving post-Accra agenda. Chapters II, III and IV provide an overview of how the Bank has met or is meeting the 12 IDA15 commitments related to "harmonization and alignment", clustered around the Accra themes of ownership, partnership and results, with an additional section on implementation in the Bank. These chapters review progress and experience and suggest steps to be taken for the Bank's continuing work in these areas. In these chapters some commitments are treated only briefly (notably in areas where papers have recently been discussed by the Bank's Board and/or where specific topics are addressed in other IDA15 MTR papers2 while other issues receive a more detailed examination (particularly joint/collaborative CAS work, project management arrangements, aid architecture and nontraditional partners, legal harmonization, aid predictability, and staff incentives and capacity). Chapter IV draws conclusions for the overall agenda going forward. 1 Some good examples of policy changes related to IL do exist, for instance additional financing in which low risk leads to quick processing; the rapid response policy which recognizes that even where there are high risks the urgency of needed results requires accepting higher risks and has helped identify modes of collaboration (e.g. with the UN). 2 This report focuses on areas not treated in other IDA and recent Board papers including the IDA15 Results Measurement System: Mid-Term Review Report, October 2009, IDA's Support to Fragile and Conflict-Affected States--Progress Report 2007-2009, October 2009, A Review of the Use Output-based Aid Approaches, October 2009, 2009 Development Policy Lending Retrospective: Flexibility, Customization and Results (forthcoming), Country Assistance Strategies (CAS): Retrospective and Future Directions (SecM2009-0441), September 10, 2009, Good Practice Note: Using Poverty and Social Impact Analysis to Support Development Policy Operations, August 2008, Moving Ahead on Investment Lending Reform: Risk Framework and Implementation Support (SecM2009-0442), September 9, 2009, Toward a Global Bank: Discussion Note (SecM2008-0428), and Toward a Global Bank: An Update (SecM2009-0326), June 25, 2009. These papers should be read in parallel to this report. 3 7. The 12 commitments examined in the chapters below (summarized in Annex B) are as follows: Strengthening country ownership Update good practice note on joint/collaborative CAS Implementation guidance on project implementation units Review conditionality Building more effective and inclusive partnerships Incorporate nontraditional partners and undertake analytic work on aid architecture Survey IDA leverage through cofinancing Work on harmonized approaches to MOUs in cofinanced operations Delivering and accounting for development results Review predictability of IDA disbursements Update good practice note on PSIAs Shaping the AE agenda and supporting implementation Update progress on the Matrix of Actions and develop Bank Action Plan Reporting by Regional Management to the Board Enhance staff incentives and guidance on ownership, harmonization, alignment Decentralization o Specific targets for the Africa Region o Continue to explore further options; complete analysis of different models. II. STRENGTHENING COUNTRY OWNERSHIP 8. The Paris Declaration called for partner countries to exercise effective leadership over their development policies and strategies and to coordinate development actions. Similarly, it called for donors to respect partner countries' leadership and help strengthen their capacity to exercise it. The AAA marked a further step forward, as developing countries demonstrated significantly greater ownership of the aid effectiveness agenda. Thus the AAA supports ownership through capacity development to help governments: (a) identify strategic priorities, (b) develop and measure results, and (c) lead donors--as well as an increased role for local government bodies, civil society organizations, and parliaments. It also emphasizes greater reliance on country systems to implement development programs and manage aid. The IDA15 commitments in this area were to update the good practice note on joint/collaborative CASs; develop guidance on project implementation units; and review conditionality. This section reviews in detail issues related to experience with joint/collaborative CASs and improving project management. 4 A. Experience with Joint/Collaborative CASs 9. The Bank is updating its guidance note on good practice for collaborative CASs to fulfill the IDA15 commitment. In preparing this update, the Bank reviewed experience with collaborative country strategy work.3 This review found that the term "joint/collaborative CAS" goes well beyond the production of a strategy document to include a range of collaborative activities that are often sequential, each building on the preceding ones: Aligning donor priorities with country priorities. The country's articulation of its development strategy, higher-level goals, and strategic priorities are expected to be the basis for donors' articulation of their assistance strategies. A major challenge is where the country has no national strategy, or where government capacity is too weak to provide leadership and vision; in such a situation, donors can overwhelm the country with their own priorities and results matrices. Carrying out joint analytic work. Shared or joint analytic work can promote a coherent donor and country view of the country's potential and challenges; it can also reveal important differences in approach among the stakeholders, and can help reconcile these differences by providing an opportunity to reduce and/or eliminate conflicting recommendations. Good practice involves donors that have the capacity, mandate, and resources collaborating on a coherent program of analytic work, ideally under the management of and with participation from the country, timed to feed into country strategy development. Conducting country and sector fora. Country- and sector-level fora have potential payoffs, but they can become unworkable if they are too numerous, large, and elaborate. An inclusive forum for key stakeholders, with effective leadership and participation from high-level country officials focusing on country alignment, can enhance the coordination process. To best utilize country capacity, pragmatic mechanisms are needed to contain the proliferation of these groups, drawing on partners with technical advisory capacity while still retaining the ability to reach out to all donors and other stakeholders (e.g., civil society). Mapping donor activities. Donor mapping provides stakeholders with a clear understanding of "who is doing what" to help the government achieve its development goals. It can point to areas where donor programs contribute to fragmentation or are not well aligned with government priorities, and can help partners plan their own actions. The process requires the cooperation of all relevant aid agencies to provide timely information and keep information up to date. Participating in division of labor exercises. Division of labor among donors can have high returns to AE, but can be challenging to achieve. If done well, division of labor can reduce duplication and allocate aid more efficiently; but most donors are reluctant 3 The review covered all 55 CASs and 21 Interim Strategy Notes (ISNs) produced between FY06 and FY08 (50 were from IDA countries); in addition, more than 25 interviews were conducted with country directors, staff, and government officials involved in preparing CASs. 5 to cede participation in sectors and thematic areas that are important to their own constituencies. Thus, unless the country exerts strong leadership in deciding which donors do what, reaching agreement among only the donors themselves can be time- consuming and difficult. Good practice involves balancing country preferences with donor mandates and capacity. Developing a joint assistance strategy (JAS) document. Producing JAS documents has proven to be complex and cumbersome; and in some cases, depending on the country milieu, it has yielded low value-added relative to the benefit from other collaborative processes. The benefits of collaborative processes and dialogue surrounding development of a joint country strategy document are recognized as productive and useful to all participants. But in many cases it was found that the net value of producing a joint document was low or negative. 4 The underlying issue may be that CASs, as internal plans for the Bank, are not well suited to be joint products with other agencies that have their own mandates, policies, and timetables. Joint documents also make it difficult for the Bank to be accountable to its shareholders for its own deliverables. In many cases it may be useful to maintain the joint collaborative processes surrounding strategy development, but allow for greater flexibility in whether the final product of the collaboration is a single document. These collaborative activities can and often do provide a supportive framework for collaborative implementation, for example through joint financing. 10. Extent of Collaborative Activities. Figure 1. Collaborative Activities at the Country Among the 76 countries reviewed, 50 of which Level are IDA, the frequency and relevance of these activities varied widely (see Figure 1). All documents reviewed referred to aligning with government objectives and to participating in donor working groups. Evidence of shared or joint diagnostic work and donor mapping was also common, although not universal, in the documents. This review found that 68 CASs and ISNs referred to joint analytic work and donor mapping.5 This review also found that formal JASs were much less common; only five CASs and ISNs reviewed were identified as such--those for Gambia, Ghana, Kyrgyz Republic, Tanzania and Uganda. The country's interest in donor 4 Many participants, both inside and outside the Bank, echo this view of the difficulty, and mixed net benefits, of finalizing a single document. According to the 2009 Bangladesh CAE, IEG argues that "improved coordination is important, but the particular vehicle of a joint CAS has not been as useful as initially thought." The 2009 CAS Retrospective states that, "experiences have demonstrated that the transaction costs of developing and implementing fully joint country strategy documents are high: the process demands extensive preparation time, and leads to often lengthy country strategy documents that focused more on development partner processes, and less on a country strategy well aligned with country priorities than would be desired." 5 The 2008 Paris survey found that the Bank carried out 59 percent of its analytic work jointly with other donors and/or with the government, against the 2010 target of 66 percent. 6 collaboration and the size of the overall aid (and the Bank's) program in the country was related to the extent and relevance of collaborative practices: these practices were more prevalent in IDA countries. 11. Value of Collaborative Activities. The value-added of each type of collaborative activity varies, and depends on how well each process is carried out. Table 1 summarizes the value added from the different activities. Table 1. Value-added of Collaborative Country Strategy Activities Collaborative Benefits Costs Value-added practice Country articulation High Low High of development Provides key information and Unless strategy contains Provides goals, broad strategy, higher- coherence to stakeholders; high level of detail on strategy, macro-framework, level goals, and helps keep them on same projects, sectors. and key results. Country strategic priorities page. leadership is key. Shared and /or joint High Medium High analytic work Crucial to common Can be high to Bank and Especially when partners understanding of country other IFIs, and for task have capacity to contribute situation and prospects. managers. and country participates. Country/sector fora Medium Low-High Medium Some benefit from sharing Costs for both donors and Especially when country-led information and approaches to government rise sharply and frequency is limited. common objectives; enhanced as subgroups proliferate. More value for coordination by country leadership and of financing, e.g., debt high-level participation. reduction or budget support. Donor mapping of Medium-High Medium Medium activities Provides all stakeholders with Costs primarily to donors Valuable to illustrate universe of activities. both at outset and for fragmentation and lack of updating. alignment, and as first step toward selectivity. Selectivity and High High High division of labor in If based on comparative Very costly to negotiate If well done and led by donor programs advantage. among aid agencies and country. within any one agency. Joint assistance Low High Low strategy document Each donor needs a separate For joint documents high Negative for document itself; business plan for internal costs to all donors; lower value mainly as vehicle to reporting and accountability. if prepared by one launch other coordination agency. activities. 12. Need for Country Leadership. The single most important factor determining whether collaborative country strategy work will add significant value is country leadership exercised across the spectrum of collaborative activities. The underlying principles of collaborative country strategy work are to ensure the alignment of donors' input with the country's own priorities and to develop a common framework among the country and partners--and country leadership is key to this process. The most critical activity is for the country to identify its priorities and articulate 7 them in development strategies, which can serve as the basis for country leadership in donor alignment and division of labor, and help avoid overlap and conflicting efforts. While collaboration cannot ensure consensus among all partners on policies, approaches, and instruments, it can lead to more coherence in the development process. When the country lacks capacity to produce such a plan, donors sometimes develop parallel strategies, thereby undermining or overwhelming the country's ownership of the process. Thus, ensuring that the country remains at the center of the strategy process is critical for the success of that process. B. Improving Project Management Arrangements (PMAs) 13. The Bank has fulfilled its IDA15 commitment to assess the implementation of the guidance on project implementation units (PIUs). PIUs or more broadly PMAs, not integrated into government structures, can contribute to or undermine ownership and sustainability. There is considerable evidence that, especially in countries where there are many donor-financed projects, PIUs that are not integrated with national institutions have caused a number of problems. Avoiding the use of PIUs entirely, or integrating them with existing institutions, makes several desirable outcomes more likely: donor-financed activities will be consistent with overall government capacities; managerial and technical skills acquired by national staff working on donor-financed projects will be retained within government institutions; country systems associated with project implementation will be used and strengthened; fragmentation and the transactions costs of assistance will be reduced; government institutions' authority and sense of responsibility will not be undermined; and government institutions will have the motivation and capacities to sustain the operation of and the benefits generated by the donor-financed project even after donor financing ends. In short, using integrated PIUs is likely to contribute to long-term capacity development and sustainability; support the broader objectives of country ownership, alignment, and harmonization; and lead toward greater self-sufficiency of the recipient government. 14. Bank Performance. The Bank's performance in this area is good: by 2008 it had already met the 2010 Paris Survey target to reduce the stock of parallel PIUs. Based on the definition of parallel PIU used in this survey, the Bank supports relatively few parallel PIUs, mainly because the classification places considerable weight on PIUs' accountability to country authorities rather than to donors--an important element of project management that the Bank always applies (see Box 1). However, recognizing that more can be done to integrate the management of IDA-financed projects with existing government institutions, the Bank is exploring how to further increase its level of ambition. 8 Box 1. Monitoring the Use of PIUs: Bank Performance The Paris Declaration included in its monitoring framework an indicator on the number of parallel PIUs per country, with an objective to reduce that number. In this framework, a PIU is considered parallel when the answer to three or more of the following questions is "yes": Is the PIU accountable to the external funding agencies/donors rather than to the country implementing agencies (ministries, departments, agencies, etc.)? (Y/N) Are the terms of reference for externally appointed staff determined by the donor (rather than by the country implementing agencies)? (Y/N) Is most of the professional staff appointed by the donor (rather than the country implementing agencies)? (Y/N) Is the salary structure of national staff (including benefits) higher than those of civil service personnel? (Y/N) Bank-supported PIUs generally do not adhere to the first three criteria for "parallel" because they hire staff who are accountable to the government, not the Bank, with the TORs and appointments made by the government (to satisfy its fiduciary requirements, the Bank often requires that the government seek a no-objection on the TORs of PIU staff). However, these PIUs often meet the fourth criteria for a parallel PIU: staff in PIUs are paid more than civil servants. In only 4 of the 54 countries that participated in the survey are staff in PIUs under Bank- financed projects paid on a scale equivalent to that of civil servants: Bolivia, Indonesia, Nigeria, and Tonga and even in these countries this might not be universally true. 15. Bank Guidance. The Bank has reviewed its current (2005) guidance to staff on PMAs and found it to be fully aligned with aid effectiveness commitments and intentions. It recommends the "use of existing institutional structures as the default mode, and use of `enclave' PIUs as an exception." It recognizes that the appropriate PMAs for each project depend on country and project circumstances. It also recommends bringing the issue of country capacity development and project implementation arrangements into the country-level dialogue, establishing country-level guidelines on PMA (including PIU staff remuneration and other incentives) to minimize distortions while pursuing broader civil service pay reforms, and addressing broader institutional capacity development issues at the sector strategy level. 16. Bank Practice. Bank practice, however, may not be fully consistent with this guidance. It is difficult to make generalizations about Bank PMA practice, since PMAs vary considerably with country and project circumstances; also, adherence to the guidelines and improvements in Bank performance with respect to PMA has not been systematically monitored, and anecdotal evidence on Bank practices is mixed. On the one hand, many Bank staff claim they are not aware of the 2005 Guidance Note for Project Management, and interviews with Bank staff and reviews of Bank documents suggest that there has not been significant change in Bank practices with respect to PMAs;6 staff express concern that decreased use of parallel PIUs may increase fiduciary and safeguard risks and slow project implementation and disbursements. On the other hand, the increased use of program-based approaches and country systems points to the conclusion that the Bank is using more integrated PMAs. Programmatic lending tends to rely more on the existing civil service than on consultants in parallel PIUs, and increasing the use of 6 Because the Africa Region has recently issued new guidance on PMAs, this Region might be an important exception to this general observation. 9 country systems eliminates the need for the government to ring-fence Bank funds or to hire specialized staff with knowledge of Bank systems. 17. Country Dialogue. A review of CASs and selected Country Portfolio Performance Reviews (CPPRs) indicates that, in many countries, the Bank and the recipient government place lower priority on PMA issues than on other issues. CASs and CPPRs do not always contain discussions of PMA issues, and when they do it is usually in countries where there are many donors and external financing is large relative to the government budget. In these documents, there is usually recognition that activities financed by the Bank and other donors should be increasingly integrated with existing institutions (see Box 2). Box 2. Examples of CPPRs and CASs that Address PMA Issues Ethiopia Joint Country Portfolio Review, 2009, gives extensive treatment to issues of country systems and discusses in detail the variation in salary scales across PIUs and consequent staff turnover. Yemen CPPR, 2008, expresses the goal that PIUs should be phased out as the civil service is reformed and includes detailed guidelines for all IDA-supported PIUs. Nicaragua Country Partnership Strategy, 2007, includes a general recommendation that "project design should better reflect existing institutional capacity of the Government and include capacity strengthening components instead of relying on quick fixes such as PIUs." Madagascar CAS, 2007, states, "The use of the sector-wide approach in policy and investment lending and phasing out of project implementation units will contribute to [the increased use of country systems]." 18. Addressing the Challenges of PMAs. The Bank is working to address the challenges of PMAs at their root. In addition to the increased focus on the use of country systems generally, the Bank is pursuing three primary approaches: Options are being examined to better support government programs at the sector/subsector level focusing on strengthening government capacity at the sector level--including for project implementation--which would lead to less need for supplemental capacity to manage individual projects. The IL reform process is striving to generate a comprehensive framework to address shortcomings in the supervision process and shift the focus to implementation support--helping governments to implement projects and build capacity.7 This will, in turn, reduce the need to have parallel units with capacity to specialize in Bank requirements. The Bank's focus on public sector reform and governance highlights the need to align salaries of PIU staff to that of civil servants as part of the broader civil service reform. Overall, the Bank will continue to monitor PMA issues--especially through the Regional Quality Management Units and financial management staff--and to draw lessons from experience in this area (see Box 3). 7 See Section III of Moving Ahead on Investment Lending Reform: Risk Framework and Implementation Support (SecM2009-0442), September 9, 2009. 10 Box 3. Emerging Lessons from PMA Experience Any decision on appropriate PMAs must take into account the prevailing environment. There is no single solution: finding the best arrangement requires careful analysis of the country, executing agency, and project context. For example, some countries may have a legal or policy framework guiding PMAs that must be taken into account. Some may have a large number of donors, making it important to work collaboratively, including on PMAs, to support country objectives. The capacity and track record of the executing agency to oversee projects will necessarily influence PMAs and must be assessed. Finally, project characteristics are key. Projects supporting programmatic approaches that include, for example, support to ongoing service delivery, are usually implemented through existing institutions, while a highly technical project might justify a relatively autonomous PIU. PMA design could be unbundled into six elements, recognizing that each element represents a range of the extent of integration/independence of the PMA: Staff composition of core functions. The mix of civil servants and consultants, and the types of positions each would fill, is fundamental to PMAs. Assigning to civil servants such core functions as overall project implementation, project management, staffing decisions, and coordination with other parts of government and other donors may help strengthen institutional capacity and improve sustainability. Staff composition of specialized functions. Specialized functions--such as technical inputs, procurement, financial management, safeguards, and monitoring and evaluation--may need to be assigned to consultants if unusual expertise is required and adequately trained civil servants are not available. Using strong country systems for project implementation reduces the need for externally recruited consultants to undertake specialized tasks. If expertise in a highly technical area outside the normal purview of the executing agency is required, externally recruited consultants may be warranted. Salary structure. Civil service salaries that are high enough to attract and retain well-qualified staff would help ensure a strong executing agency with a staff that could be used for all but the most specialized functions. Inadequate salaries may justify salary top-ups for civil service staff, as well as a salary scale for consultants based on comparators outside of government. Source of funding. If successful project implementation requires salary top-ups and dedicated equipment, increasing the share of government funding for project administrative costs could help improve sustainability. Reporting channels. To avoid overlapping responsibilities and confusion about authority, reporting relationships should consider the existing institutional hierarchy of the overall government and of the executing agency(ies) rather than circumvent it through new institutions or unique reporting relationships. Systems and standards. PMA design needs to draw on existing assessments of country and agency financial management, procurement, social and environmental safeguards, and monitoring and evaluation systems to evaluate the extent to which they can be used for project implementation. 19. Coordinated Approach. The larger the donor presence, the more important it is for donors to work jointly with the government to agree on a common approach to PMAs and improve the collective impact of PMAs. The Bank and government must consider what other donor agencies are doing with respect to PMAs so that donors do not undermine one another's efforts to build country capacity. Donors should look for opportunities for more collaborative approaches, including establishing PIUs that serve multiple donor-financed projects. In addition, at the international level, there needs to be a broader dialogue with DAC donors and other partners on PMA issues, which would highlight seeking agreement on common principles and 11 practices, common terminology for PMAs,8 and a common improved approach to measuring progress toward the goals of the Paris Declaration and the AAA. C. Application of Conditionality 20. Work on the IDA15 commitment is completed. The Bank has reviewed recent Bank experience with conditionality in development policy operations (DPO),9 focusing on the effectiveness of Bank DPOs in supporting countries in the design and implementation of their medium-term development policy agendas. One of the main findings of the retrospective was that the Bank continues to prepare its DPOs in line with its good practice principles for the application of conditionality (reinforcing ownership, harmonization, customization, criticality, transparency and predictability). Feedback from in-country consultations revealed that Bank DPOs have been successful in reinforcing the ownership of reform programs, with a focus on the most critical aspects of these reforms, by providing customized support that responded to governments' priorities and country circumstances. In addition, stakeholders and development partners have recognized that the Bank has made good progress on harmonization, transparency, and predictability. The retrospective also found that the Bank's application of conditionality is broadly consistent with its commitments in the AAA: the disbursement conditions for Bank DPOs are limited in number,10 based on national development strategies (often PRSPs), and results-based. In addition, all disbursement conditions are publicly available in the Program Documents and the Completion Reports, which are disclosed. The Bank will continue to monitor its performance against its own good practice principles and the AAA, and will also contribute to the OECD-DAC work to promote the application of similar practices across a broader set of development actors. D. Country Ownership: Way Forward 21. Reflecting the spirit of the post-Accra agenda, the Bank continues to focus its AE work on the country and its capacity to lead and manage the development process. WBI plays a critical role in carrying forward the Bank's overall capacity development agenda, including its efforts to strengthen country and regional institutions, leaders, and coalitions. As this section notes, the Bank will focus on high-value-added activities related to collaborative country strategy work under country leadership. It will also emphasize country ownership of project management by examining options to better support government sector programs; shifting from supervision to implementation support under IL reform and supporting government capacity for project implementation; and aligning salaries of PIU staff to that of civil servants under civil service 8 Under the OECD/DAC guidelines for the Monitoring Survey, two types of PIUs, parallel and non-parallel, have specific definitions. But the OECD/DAC parallel PIU does not correspond to the stand-alone PIU referred to in the Bank's 2005 Guidance Note. 9 See 2009 Development Policy Lending Retrospective--Customization, Flexibility, and Results (forthcoming). The review looked at some 160 DPOs in 66 countries, for a total amount of $33 billion, covering FY06 through the third quarter of FY09. 10 Prior actions are a set of mutually agreed policy and institutional actions that are deemed critical to achieving the objectives of the program supported by a DPO, which are implemented before the Board approves a loan, credit, or grant. There has been a reduction in the average number of prior actions per operation: for both IDA and IBRD operations, the average number of prior actions has declined from above 30 in the mid-1990s to about 9 in IDA operations and 11 in IBRD operations in FY09. See Development Policy Lending Retrospective--Flexibility, Customization, and Results (forthcoming). 12 reform. In addition, the AEAP indicates several actions that the Bank will undertake to support country ownership: (a) supporting government capacity to consult civil society and other institutions (such as local governments and parliaments)--an area that is complemented by WBI's efforts to strengthen civil society, parliaments, and other country actors beyond governments; (b) continuing to disseminate good practices and support countries in making national development strategies more operational, including through better prioritization and stronger links to budget and results frameworks; and (c) continuing to ensure that CASs and DPOs are aligned with a country's own development strategy.11 22. Focus on Use of Country Systems. The use of country systems for procurement, financial management, and environmental and social safeguards is a critical component of the Paris Declaration and the Accra Agenda for Action, and is a key area of focus for the Bank. Bank performance as measured by the 2008 Paris survey for the use of country financial management systems and country procurement systems is 62 percent and 52 percent respectively (compared to the target of 50 percent for each) but the Bank will continue to focus in this area as it firmly believes that stronger country systems have a major impact on development effectiveness and that ultimately such systems can only be strong when they are used for all public transactions. In all three areas, the Bank has developed policy frameworks for its reliance on countries' systems, and it has set up work programs to move in this direction. The Bank will deepen the already substantial achievements related to financial management,12 giving particular attention to the development of and reliance on national audit capacity, in the private sector and in particular in national Supreme Audit Institutions. In procurement, the Bank will continue to encourage the greater use of national procedures under national competitive bidding, when feasible (especially in IDA countries). It will build its program of pilots in international competitive bidding13 and will develop a strategy--including assessments and implementation support--for countries not presently included in this pilot, especially IDA countries. It will continue to scale up its program to pilot the use of country systems to address environmental and social safeguard issues in Bank operations from the project to the 11 See Country Assistance Strategies: Retrospective and Future Directions (SecM2009-0441), September 10, 2009 and Development Policy Lending Retrospective--Flexibility, Customization, and Results (forthcoming). 12 Financial management is the only sector (or area) in which a Bank OP articulates that the use of country systems is the default option (when circumstances permit). In addition, guidelines for assessing fiduciary risks in the use of country financial management systems have been finalized. This framework provides a holistic and uniform approach to the assessment of fiduciary risks in the use of country financial management systems--that is, the Bank is focusing on assessing, strengthening, and using country financial management systems on a countrywide basis rather than negotiating their use under individual investment projects. See Assessment of Fiduciary Risks in the Use of Country Financial Management Systems in Bank-financed Investment Projects, Interim Guidance Note for FM Staff, OPCS July 2009. 13 The pilot program on the use of country procurement systems, approved by the Board of Directors in June 2008, focuses on international competitive bidding procedures and covers approximately 19 countries, 10 of which are IDA (IDA only or IDA/IBRD blend). The IDA countries involved are Bhutan, Burkina Faso, Ghana, India, Laos, Mongolia, Pakistan, Rwanda, Senegal and Vietnam. See Use of Country Systems in Bank-Supported Operations: Proposed Piloting Program--First Progress Report (forthcoming). 13 country level,14 and has also developed an assessment tool (the Safeguards Diagnostic Review) which is carried out between the Bank and its clients and is made public. As the CAS Retrospective suggested, the CAS will provide the vehicle to raise issues related to the use of country systems and to consider related capacity development efforts. 23. The Bank has been very active in the OECD-DAC cooperation fora in the area of use of country systems, and has been and remains a co-chair of the task teams (formerly Joint Ventures) in financial management and procurement. These groups have developed diagnostic tools that permit partner countries to evaluate their capacity building and policy reform needs against agreed international standards and to identify areas for development. These tools lay the basis for donors to develop their policy frameworks for the use of country systems and to provide appropriate plans for capacity-building support. The Bank does not require that all such needs be fulfilled before it can rely on the country system; rather, such support can be built into the operations the Bank supports and provided during implementation. Progress in these policy areas has not always been smooth, will be regularly reviewed and experiences shared with all partners. The Bank believes that it is critical for donors to work in harmony to support capacity building and move to rely on a country's systems once it has reached the mutually agreed standards; to this end the Bank will continue its very close cooperation with donors and partners through the OECD-DAC clusters. III. BUILDING MORE EFFECTIVE AND INCLUSIVE PARTNERSHIPS 24. The Paris Declaration called for donor actions to be more harmonized, transparent, and collectively effective. Specific commitments were related to common donor arrangements and simplified procedures, more effective division of labor, and incentives for collaborative behavior. The AAA pays relatively less attention to issues related to donor harmonization as an end in itself, but broadens the aid effectiveness partnership by embracing the involvement of non-DAC donors, global funds, foundations, the private sector, civil society, and other partners in the country-led model. Some areas of focus: recognizing the value of South-South cooperation and the development role of civil society organizations (CSOs); the need for country leadership to reduce aid fragmentation at the country level; and increased cooperation and collaboration in fragile situations, emphasizing flexible, rapid, long-term funding and tailored support for building the capacity for core state functions. The IDA15 commitments in this area were to incorporate nontraditional partners and undertake analytic work on aid architecture; survey IDA leverage through cofinancing; and work on harmonized approaches to MOUs in cofinanced operations. This section reviews in detail IDA15 commitments related to aid architecture and nontraditional partners, and harmonization of MOUs. A. Aid Architecture and Nontraditional Partners 25. Over the past several years, the Bank has placed considerable emphasis on coordination with traditional donors at the country level (for example, via collaborative strategy and budget support activities) and at the institutional level (for example, via the Legal Harmonization 14 See First Year Review of Implementation of Incremental Scaled-up Program to Pilot Use of Country Systems to Address Environmental and Social Safeguard Issues in Bank-supported Projects (SecM2009-0367), July 16, 2009. 14 Initiative). As one of its six strategic themes, the Bank has developed partnerships with key actors to coordinate activities in fragile and conflict-affected countries. The Bank will continue with this collaborative work. Under IDA15, the Bank committed both to undertake further analytic work on the changing aid architecture and to increase its efforts to incorporate nontraditional partners. The Bank recognizes that nontraditional providers of development assistance play an increasing role in international development cooperation.15 Both global program funds (GPFs) and funding from middle-income and emerging countries are increasingly important sources of development assistance. The total contributions from GPFs, estimated in 2006 at about 3 percent of official development assistance (ODA) and 5 percent of country programmable aid,16 are expected to grow as new GPFs are formed to deal with issues such as climate change and trade. Assistance from middle-income and emerging bilateral countries was at least 10 percent of total ODA in 2006 and is also growing.17 Nontraditional donors are also visible within the Bank; for instance, 23 non-DAC donors contributed 3.7 percent of the total amount received from donors for IDA15. The Bank has been carrying out analytic work on these trends, and it is working increasingly with both GPFs and middle-income and emerging bilateral donors at several levels: work at the global/regional level toward a more inclusive aid architecture; support to nontraditional partners in operational collaboration at the country level and in their roles as development partners; and work with recipient governments to support productive engagement with nontraditional partners. 26. Bank Analytic Work. The Bank is carrying out analytic work to facilitate a better understanding of the role of nontraditional partners in the aid architecture. This work has included a study on the GPFs' application of the Paris Declaration principles at the country level,18 which drew on several country case studies in Africa to show that GPFs are increasingly taking steps to integrate into country-level development programs. However, the study also showed the risk of allocative distortions caused by vertical funds, especially if funding is off- budget. Other analytic work builds on the increasing recognition of the important role private aid and CSOs play in international development: it includes a study on private donors in the African aid landscape and a study that seeks to quantify public and private aid flows channeled through CSOs. 1. Global and Regional Activities 27. The Bank supports global/regional activities that facilitate the integration of nontraditional partners in the aid architecture. 15 See Chapter 9: Implications of the Changing Aid Architecture, Board background documents related to the Third High Level Forum, August 2008. 16 Based on information from the largest 21 global program funds. 17 For the rest of this discussion, these middle-income and emerging bilateral donors will be referred to as non-DAC donors. This group includes Algeria, Argentina, Brazil, Chile, China, Czech Republic, Egypt, India, Indonesia, Korea, Kuwait, Malaysia, Saudi Arabia, Singapore, South Africa, Thailand, Tunisia, Turkey, Venezuela, and others. 18 Global Program Funds at the Country Level: What Have We Learned? Concessional Finance and Global Partnerships, World Bank, July 2008. 15 The Bank was one of the organizers of the Accra HLF, and coordinated the work for a roundtable on the new aid architecture featuring GPFs and other providers of development assistance outside the DAC. The AAA, which was influenced by this work, makes a commitment to more inclusive aid architecture and encourages all development actors to use the Paris Declaration principles as a point of reference in providing development cooperation. The AAA also recognizes the importance of South-South cooperation and of triangular cooperation. The Bank has increasingly facilitated South-South Cooperation as a component of its operational work. As part of its commitment to South-South Cooperation, the Bank is actively supporting the Colombia High-Level Meeting on South-South Cooperation and Capacity Development (Spring 2010), beginning with a workshop in Washington DC (September 2009) . To support emerging donors as a group, the Bank has been working with Russia to organize development aid seminars and international events such as the High-Level Emerging Donor Meeting (2006) and the forthcoming Moscow Conference on Development Finance and Emerging Donors (February 2010). In 2008 the Bank organized a workshop for EU New Member States (Bulgaria, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Slovenia and Slovakia) to discuss how emerging donors, formerly aid recipients, can most effectively assume their roles as development partners and shape their development assistance programs. 2. Support to Nontraditional Development Partners 28. The Bank provides channels for nontraditional partners' assistance to recipient countries, including trust-funded activities and the HIPC initiative, which draw on the Bank's experience and knowledge. Between FY04 and FY08, the number of agencies contributing to Bank-administered trust funds increased from 162 to 224. Much of the increase is due to: (a) rising use by non-sovereign agencies (especially private corporations contributing to the various carbon funds); and (b) countries that are not members of the DAC, which tend to focus their trust-fund contributions on global and regional programs and activities--for example, the Global Environmental Facility, the African Program to Control Onchocerciasis, the Consultative Group on Agricultural Research, or the Avian and Human Influenza Program--and trust funds supporting reconstruction efforts by Afghanistan, Iraq, and the Palestinian Authority. The HIPC initiative also provides an important framework for the debt relief efforts of nontraditional development partners: eight non-Paris-Club creditors have provided full relief, and 22 have provided partial relief. 29. Developing Capacity as Donors. The Bank assists larger emerging donors in developing their capacity as providers of development assistance.19 For example, the Bank and China are working together to identify specific areas for the country's participation in 19 Country Partnership Strategy for the People's Republic of China for the Period 2006­2010 (R2006-00057/7), May 23, 2006 and Russia Country Partnership Strategy for the Russian Federation for the Period FY 2007- 2009 (R2006-0215), November 20, 2006. 16 development cooperation, including increasing its participation in learning exchanges, disseminating its development experiences to other developing countries, and facilitating its involvement in global and regional initiatives. The China-Africa initiative has been an important contribution to this. The World Bank has been involved in policy advice and capacity building for Russia for its development aid since the 2006 G-8 Summit in St. Petersburg. Cooperation follows a learning-by-doing approach: by channeling development assistance through Bank Group instruments, Russian specialists become active participants in addressing urgent human and development needs in Africa (malaria control, access to energy, quality of primary education), Central Asia (reducing the risks of infectious disease) and producing international tools for public policy (learning assessment tools and international methodology for measurement of financial literacy). The Government has signed a $3 million fee-based service agreement with the Bank to build capacity for providing development assistance in education. With support from the UK Department for International Development, the Bank is also helping the Government develop critical policies and functions for effective donor assistance, such as monitoring and evaluation, statistical reporting, and communications. The Bank has been involved with emerging donors in other regions, such as the Middle East, and is in the process of completing a study on the size of ODA provided by Arab donors.20 It has also significantly expanded its work with Korea, including establishing two funds in the East Asia Region. Korea is actively leading a number of fora related to aid architecture and nontraditional donors and other issues leading up to Fourth HLF. 30. Knowledge Sharing. WBI provides an array of support for knowledge sharing among all providers and recipients of development assistance, with a particular focus on capacity development through South-South partnerships. WBI has worked with Brazil on several South- South initiatives, including support of knowledge exchanges to share Brazil's experience, both within the region and to other parts of the world, particularly Lusophone Africa.21 WBI is also supporting the growing African-Asian trade and investment flows, for instance by supporting a 2008 China-led forum focusing on Chinese experience related to growth, FDI, trade, and other issues relevant to Africa. A similar South-South experience-sharing program will bring African officials to India. WBI has also collaborated with Singapore, Malaysia, and China to share development experiences on Special Economic Zones and FDI with senior African officials. In addition, the Bank also manages a multi donor South-South Experience Exchange Trust Fund (SEETF), which provides practical, low-cost ways to fill knowledge gaps through the sharing of developing country experiences between practitioners (see Box 4). 20 See Arab Development Assistance: An Overview of Trends and Institutions (forthcoming). 21 Brazil Country Partnership Strategy for the Federative Republic of Brazil for the Period FY2008-2011 (R2008- 0065/4), May 6, 2008. 17 Box 4. The South-South Experience Exchange Trust Fund Launched in 2008, the SEETF responds to direct requests from IDA countries seeking knowledge, tapping into the accumulated expertise of partner countries by funding direct contact between developing countries and thus deepening South-South experiences. It will also create a website-based experience exchange library that will document, monitor, and disseminate results, as well as include a roster of developing country experts. Seven donors have pledged support to the SEETF: China, India, Mexico, Denmark, the Netherlands, Spain, and the UK. All funds from the initial round of pledges have been committed: 28 percent to AFR, 20 percent to LCR, 18 percent to ECA, 15 percent to SAR, and 10 percent each to EAP and MNA. Activities cover a wide range of topics from collaboration on hospital policy and management between Senegal and Morocco and Tunisia to the exchange of experiences on forest and carbon trading between Kyrgyzstan and Tajikistan and Moldova. 31. Country-Level Work. The Bank has been Figure 2. Work with Nontraditional Partners working increasingly with both GPFs and non-DAC at the Country Level donors to further partnership at the country level and align with country priorities. As evidence of this NonDAC bilateral donors 34 Nontraditional donors increasing partnership, these donors' contributions are being mapped into the donor matrices included Global program funds 29 in country strategies. Of the 76 CASs and ISNs Coporations and 19 reviewed, 53 mentioned various types of foundations nontraditional partners--often more than one type Civil society organizations 4 (see Figure 2). The Bank also works with these partners on a variety of specific initiatives at the 0 5 10 15 20 25 30 35 40 country level. Number (out of 76) 32. Global Programs. The Bank has been working directly with a number of GPFs to increase alignment with country priorities--for example, in Education for All (EFA) and the International Health Partnership (IHP) (see Box 5). The Bank has recently partnered with OECD to develop good practice guidance for improving GPFs' aid effectiveness at the country level. The guidance, drawn from an analysis of the advantages and challenges that growth in GPFs brings to partner countries, was based on case studies, workshops, consultations with partner countries, and discussions with the Global Partnership Learning Group of five large GPFs. The result of the analysis, summarized in a paper prepared for the HLF in Accra, is a set of guidance principles for GPFs. However, more could be done to consider global programs as part of the strategic dialogue with clients as country strategies are being developed. 18 Box 5. EFA and IHP: Country Priorities to Drive Global Programs EFA. In 2002, the Bank joined with other development partners in launching the Education for All Fast Track Initiative (EFA-FTI) to help low-income countries meet the Millennium Development Goal (MDG) on education. The EFA-FTI is a platform for collaboration at both the global and country levels: through a compact, developing countries commit to design and implement sound education plans, while donors align their funding to support the country. EFA-FTI is a global partnership built on the principles of country ownership, donor harmonization, and local-level empowerment. According to the EFA-FTI progress report The Road to 2015: Reaching the Education Goals, the EFA-FTI is a leader in implementing the Paris Declaration and has made good progress in primary school enrollment trends. IHP. Established in September 2007, the International Health Partnership (with related initiatives, IHP+) is part of a renewed global push to improve the results linked to MDGs 4 and 5 on health by increasing the quality of health services and broadening access to them. In accordance with the Paris Declaration, IHP+ partners--a coalition of international health agencies, governments, and donors including both the Global Alliance for Vaccination and the Global Fund to Fight AIDS, Tuberculosis, and Malaria--have pledged to improve health results by developing and supporting country-led compacts under which all investments are based on a single country health strategy that is results-based and is costed, with clear performance benchmarks. This country compact, which builds on existing country processes and agreements, is linked to the overall country development plan. The IHP+ signatories have explicitly committed to engaging with civil society as a full partner, stating that CSOs should participate in the design, implementation, and review of the partnership at both the global and country levels. With other partners, the Bank supports the coordination of the IHP+ and is both a member of the High Level Task Force and the Core Team, which supports the virtual secretariat to the task force. 3. Support to Countries 33. The Bank works with recipient governments to support productive engagement with nontraditional partners. For example, the Bank country program in Rwanda taps South-South experience: the Bank is facilitating a twinning arrangement between the Rwanda's Electrogaz and the Tunisian Power Utility. Rwanda was one of the participants in the recent China-Africa Experience Sharing Seminar in Beijing. And a study tour is being planned to enable Rwandan officials to learn lessons from successful efforts on targeted poverty programs. The Bank also supports or leads global programs and other initiatives--such as the Extractive Industries Transparency Initiative--which provide support to countries focusing on specific sector and thematic issues by bringing together diverse partners from the CSO, industry and donor communities. B. Legal Harmonization 34. The Bank has been working with other lenders and donors to develop common guidelines for Memoranda of Understanding (MOUs) in joint financing operations as per the IDA15 commitment. MOUs can be a useful vehicle for agreeing on the parameters of government/donor coordination, both overall and in conjunction with cofinancing. MOUs clarify and record the understandings of the partners at different stages of their discussions and set out processes for ongoing or future cooperation in a number of areas (see Box 6). They have been relatively straightforward to agree on when they set out a broad framework for cooperation among development partners. However, they can be more difficult to agree on when they outline a more specific framework of cooperation--for example, when addressing the specifics of fiduciary arrangements for donor assistance in the context of program-based approaches (PBAs). 19 Box 6. Areas Covered in MOUs MOUs can include one or more of the following: expected broad objectives and principles of cooperation to enhance effective implementation and reduce administrative burden; roles and responsibilities of each of the parties (government and donors), often including how to select the lead donor and how to share costs; processes for exchanging information and monitoring and evaluating the cooperation and the operation(s)-- carrying out joint reviews and sharing information; and fiduciary requirements for financing (procurement, financial management) and disbursement mechanisms. 35. Rising Use. The use of MOUs has risen significantly in recent years, largely driven by the increase in support for PBAs. The Bank has a significant portfolio of program-based support, which it provides through investment lending, DPOs, cofinancing trust funds, and global program funds. It increasingly supports PBAs in both IBRD and IDA countries, supporting a variety of sectors--many accompanied by MOUs. Data collected by the Bank in parallel to the 2008 Survey on Monitoring the Paris Declaration indicate that in 2007 the Bank supported 100 SWAps in just the 54 countries that participated in that survey. 36. Challenges. The increasing numbers of donors entering into these agreements bring diverse expectations related to MOUs as well as diverse policies, mandates, and authorizing environments--all of which adds to the complexity of agreeing on MOUs. A review of MOUs undertaken by the Bank has found that MOUs are often expected to serve as a platform to harmonize issues that represent fundamental institutional differences--issues related to policies and mandates under the purview of senior management or higher-level governance structures within each organization--especially those related to budget support objectives and disbursement, institutional mandates, and fiduciary issues in pooled funding. Moreover, the legal status of an MOU--the extent to which they are binding--and their terminology is often unclear (see Annex D for details). 37. Addressing the Challenges of MOUs. The majority of issues which have arisen regarding MOUs are in conjunction with investment lending and pooled financing, and not related to general budget support. A number of solutions have been employed to accommodate joint financing under the current IL instrument. These have often included more flexible frameworks which signal that donors have agreed on the ultimate objective of the operations-- supporting government priorities through a coordinated program--and have accepted the fact that differing institutional policies do not allow complete harmonization of donor procedures. Going forward, the Bank will examine options for a new programmatic or results-based financing instrument which may better facilitate joint financing22. In addition, in 2008, recognizing the need to jointly address impediments to entering into MOUs at the country level, the Bank and partners started the Legal Harmonization Initiative (see Box 7). The LHI is expected to ease a number of the difficulties with MOUs related to joint financing. It is expected to develop common donor guidelines on MOUs, focusing on "legal drafting" and operational issues that can be resolved at the level of individual MOUs, and not issues that require 22 See Investment Lending Reform (SecM2009-0026), January 29, 2009 and Moving Ahead on Investment Lending Reform: Risk Framework and Implementation Support (SecM2009-0442), September 3, 2009. 20 overarching policy changes subject to approval at the level of the Board or Senior Management. The Bank will continue to play a leading role in this initiative. Box 7. The Legal Harmonization Initiative Since 2008, the LHI--launched by OPCS and the Office of the Senior Vice President and General Counsel-- has been bringing together international financial institutions, bilateral aid agencies, and several UN agencies with the aim of coordinating and streamlining legal tools among donors and partner countries and removing critical legal impediments to harmonized approaches in support of effective operations at the country level. The LHI also provides a forum for legal, operational, and policy advisers to discuss and share knowledge across institutions on legal and policy issues relevant to the AE agenda. A working group of the LHI is now involved in analyzing the underlying challenges of MOUs and proposing common donor guidance on their use. A meeting in 2009 is expected to pave the way for the development of common donor guidance and possibly a common template for use by donors in joint financing operations. C. Leveraging IDA Resources 38. Work on this IDA15 commitment is completed; the Bank has carried out a survey of selected country programs to examine the financial leverage from cofinancing.23 In the past 10 years, IDA has significantly broadened its financial collaboration with other donors, from cofinancing and parallel financing of stand-alone projects to emphasizing program-based financial collaboration (program-based approaches, or PBAs)24 that features a much larger contribution from other donors. In FY06-07, for example, IDA's project funding had a cofinancing ratio of about 170 percent, meaning that every dollar provided by IDA in project funding leveraged 1.7 dollars in cofinancing. Cofunding ratios for PBAs other than multidonor budget support vary significantly across countries, but average approximately 250 percent.25 This is similar to the leverage of general budget funds: data from a survey on general budget support conducted by the Strategic Partnership with Africa Program show that every dollar provided by IDA in the context of multidonor budget support has been associated with 2.4 dollars provided by other donors. 39. Changing Nature of Nonfinancial Services. With the increasing use of PBAs, the nature of IDA's nonfinancial services has also changed considerably, allowing Bank country knowledge and technical expertise to be shared across a broad array of activities and partners. Under project-based cofinancing, IDA typically works with the government to prepare, appraise, and supervise individual project and provides fiduciary services when donor resources are channeled through Bank-administered trust funds. Under PBAs, the Bank supports a government sector wide program with its country and sector knowledge and its expertise on a range of cross- sectoral issues (e.g., public sector reform) as well as mobilizes donor resources in support of the PBA. Thus, IDA's contribution focuses more on assisting partner countries in formulating 23 IDA's Financial Collaboration with Other Development Partners at the Country Level, Resource Mobilization Department (FRM), World Bank, August 2008. 24 PBAs can take the form of direct budget support or nonbudget support, such as sectorwide approaches or other multidonor programs that support thematic, agency, or department-specific reforms; donors and countries differ significantly in the terminology they use for these instruments and modalities. 25 This average was derived from a small sample of six IDA countries, and the average differs significantly across countries. 21 appropriate macroeconomic, sectoral, or thematic strategies and policies--with a specific emphasis on strengthening country ownership--and coordinating the dialogue and support of participating donors. D. Partnerships: Way Forward 40. The Bank is building more effective and inclusive partnerships with nontraditional partners through a range of activities at the country, regional, and global levels. Going forward, the Bank will continue to consolidate these experiences, and support donors and recipients in building new approaches to the development partnership. Work will also continue to improve partnerships at the country level--especially those linked to joint financing arrangements-- through the LHI and the development of a new programmatic or results-based financing instrument that may facilitate joint financing under sector programs. And finally, as described in the AEAP, the Bank is actively investing in partnerships and harmonization in the context of fragile and conflict-affected countries, using CASs to emphasize selectivity and country-level division of labor. It will continue to work closely with the United Nations (through the UN- World Bank Partnership Framework for Crisis and Post-Crisis Situations), multilateral development banks, the International Monetary Fund, the European Commission, and the OECD-DAC in many aspects of its fragility and conflict work.26 IV. DELIVERING AND ACCOUNTING FOR DEVELOPMENT RESULTS 41. The Paris Declaration called for donors to help countries build their capacity to better manage for results, and for partners and donors to be mutually accountable for implementing AE actions and for development results. The AAA emphasizes accountability and transparency for results, and the need to increase the medium-term predictability of aid. Key AAA commitments include aligning results monitoring with country information systems and investing in strengthening national statistical systems; reaching agreement between the country and the donors on, and disclosing, a limited set of conditions for financing; systematically sharing three- to five-year planned aid commitments and disbursements; and conducting country-level mutual accountability reviews. The AAA also notes the need to deepen efforts to fight corruption, highlighting areas such as recovery of stolen assets. The IDA15 commitments in this area were to review the predictability of IDA disbursements, and update the good practice note on poverty and social impact assessment. This section reviews in detail the issue of aid predictability. A. Aid Predictability 42. The HLFs in both Paris and Accra highlighted the need to improve and monitor aid predictability. In the Paris Declaration, donors committed to provide reliable indicative commitments of aid over a multiyear timeframe and to disburse aid in a timely and predictable fashion according to agreed schedules. The AAA expands this commitment to cover the responsibilities of recipient countries and to spell out expectations for donors in greater detail. Under the IDA15 Replenishment, IDA committed to reviewing its own performance on aid predictability (see Box 8). To carry out this work, the Bank considered the measurement of 26 See IDA's Support to Fragile and Conflict-Affected Countries: Progress Report 2007-2009, October 2009. 22 predictability and partner countries' capacity to use aid flow data, and examined areas where the Bank can improve its provision of information to partner countries.27 Box 8. Aid Predictability Measures and Bank Performance Data from the Paris Survey was used to measure aid predictability in several ways: from a mixture of the donor and country perspectives; from the country perspective alone: and from the donor perspective alone. These three measures are defined as: The "mixed perspective" measure of aid predictability, used for the OECD-DAC Survey, is defined as the ratio of the amount of aid scheduled for the disbursement by a donor (ex-ante) and the amount of aid received and recorded in government systems by a country (ex-post). The "country perspective" measure of aid predictability is defined as the ratio of the amount of aid recorded in the annual budget by a country (ex-ante) and the amount of aid disbursements actually received and recorded by the country (ex-post) The "donor perspective" measure of aid predictability is defined as the ratio of the amount of projected disbursements by a donor (ex-ante) and the amount of actual disbursements recorded by the donor (ex- post). Applying these measures to derive Bank performance reveals quite different results. Measurements of Aid Predictability: Bank Performance (Target 71%) Mixed Mixed Country Donor perspective perspective perspective perspective Countries 2006 2008 2008 2008 51 countries 65% 64% 73% 32 countries 63% 70% 69% 76% 51 countries provided Bank data to the OECD-DAC for the 2008 Survey and 32 countries provided data for both the 2006 and the 2008 Surveys. Note: Annex C discusses the predictability and aid-on-budget measures for the Paris Survey--measures that are closely interlinked.28 43. The Bank's Performance. Data collected through the 2008 Survey on Monitoring the Paris Declaration can be measured from a mixed perspective (a combination of the donor and country perspectives, which is the survey indicator), the partner country perspective alone, or the donor perspective alone. The measure can also capture the data from all 51 of the 2008 respondents that submitted data on Bank aid flows, or only those 32 that also responded to the 2006 survey, to enable comparisons over time. As Box 9 shows, the different measures yield quite different results. Interesting to note is that about half of all countries participating in the 2008 survey on the Bank's assistance met the Paris target of 71 percent (25 countries when using 27 Aid predictability should be distinguished from aid volatility, which refers to variation in aid flows from one period to the next. 28 Aid on budget performance for the World Bank as measured by Indicator 3 of the Paris survey shows that-- against a 2010 target of 85% aid on budget--the Bank was at 62% in 2006, 71% in 2008 (using the 32 countries on common for both surveys), and 66% in 2008 (using the larger 2008 sample). This relates to the overall donor average of 42% in 2006 and 45% in 2008. 23 the "mixed" measure, 29 countries when using the "country perspective" measure, and 33 countries when using the "donor perspective" measure). 44. Which Measure? While the different measures provide different results on the Bank's performance (and the performance of other donors as well29), perhaps the greater relevance is to governments. For the recipient country, aid predictability is important for budget management and the ability to plan and implement development programs: governments need to be able to forecast and execute reliable budget information. When governments are unable to predict expected aid flows, they must mitigate the risk that aid will not be disbursed as planned either by cutting back on planned development programs, or by facing the risk of having to make difficult midcourse cuts in development programs later--in both cases reducing the potential benefits from development programs. This argues strongly for using a predictability measure based on recipient country expectations alone: it has operational implications as it helps identify challenges to governments' ability to make reliable budget forecasts of aid and account for expenditures. Of the three measures, the least credible from a technical point of view is the mixed perspective combining data from the country and donor perspectives (see Annex C).30 45. Disbursement Forecasts of Budget Support. For development policy lending, the Bank has improved its disbursement forecasts as well as the timeliness and accessibility of this information. Because DPO disbursements are typically aligned with the government's budget cycle on the basis of conditions agreed with and met by the government in advance, they provide a predictable source of assistance.31 In countries where the Bank's assistance included development policy operations, predictability from a country perspective was 71 percent. In addition, medium-term DPO disbursement forecasts are widely available in CASs, and budget support provided through certain Bank-managed global programs (such as Education for All) is also being more systematically incorporated into the Bank's CASs, so that annual and medium- term disbursement projections from these programs are becoming more readily available. The Bank needs to continue to ensure that this information is conveyed to appropriate government counterparts through its regular dialogue. 46. Predictability of Investment Lending. Improving aid predictability for Bank investment lending is more challenging. In countries where the Bank's assistance included only investment lending and trust fund operations, the estimate of the Bank's aid predictability was only 53 percent. Developing credible disbursement forecasts is difficult for a number of reasons. First, data are not systematically updated each year; often it is only updated following a formal change 29 The percentage for all donors in the 2008 survey is higher at 63% (compared to 42%) when country-only data are used, rather than the mixed data used in the DAC measure. 30 Ideally, disbursement data would be agreed upon between both donors and governments (as the Paris Declaration originally intended), so that these data reflect a shared donor-government forecast. Improving communication between donors and governments and working together to develop and use common forecasts is key. If disbursement forecast data is mutually agreed on, the problem of mixing donor and government data would be eliminated. This would help improve measures for both aid on-budget and aid predictability. 31 By definition, a DPO should be highly predictable, provided that the government implements the DPO actions as planned, since it disburses against actions completed; the main exception is when the operation has multiple tranches (such operations are rare in IDA countries). The level and timing of DPOs may also be subject to changes in country circumstances. Data on the Bank's own expected and actual DPO disbursements in 2007 indicate that DPOs were disbursed as planned in all but 4 of the 33 countries that participated in the 2008 Survey on Monitoring the Paris Declaration whose Bank program included DPOs. 24 in the project's status (e.g., restructuring, extension of closing date). Second, projecting disbursements of project financing requires taking into account a number of factors--for example, the pace of project execution and contracting capacity in the recipient agency--that vary by project. Noncompliance with terms of financing arrangements, the client's disbursement absorptive capacity, and delays related to signing and effectiveness are also challenges that affect project disbursement forecasts. The Bank is reviewing its standard disbursement profiles against recent historical data and is striving to provide more systematic updates of projected disbursements, and report these to countries. 47. Trust Funds. The Bank now requires each vice-presidential unit (VPU) to prepare an annual Trust Fund Management Plan. The goals are to strengthen the links among strategic priorities, results, and resources for the VPU by integrating all trust funds into planning, budgeting, and monitoring processes, and to be able to report to Senior Management and the Board at an institutional level across multiple sources of funds on progress in integrating trust funds into the Bank's business. In addition, since the implementation of the 2007 Trust Fund Management Framework, all CASs and ISNs have been required to fully reflect trust-fund- financed activities.32 However, there continue to be issues related to measuring trust fund disbursements, and particularly attributing disbursements when one donor administers a large amount of trust funds on behalf of other donors, as the Bank often does (see Annex C for details). 48. Country Use of Information. Looking at aid predictability from the perspective of the partner country helps to highlight countries where aid predictability matters most, and focuses efforts on helping countries develop their capacity to better use aid flow information. The extent to which aid predictability matters is related to the importance of aid to the budget: the greater the proportion of aid in the country's total receipts, the greater the impact on budget execution stemming from deviations between actual and expected aid flows. Thus predictability can be a serious concern for IDA recipient countries. In addition, for countries with weaker systems of budget planning or execution, improving predictability of aid flows is key--but it is only part of the equation. In these countries, efforts to improve disbursement data need to be coupled with efforts to develop the country's capacity to improve its budget processes and better integrate aid management with budget management. Also, most countries operate using annual budgets only- -although some have multiyear program budgets, and a few have well-developed medium-term fiscal frameworks.33 Thus it is critical that disbursement information beyond a one-year time horizon be accompanied by capacity development to improve these medium-term frameworks. In sum, for many countries--particularly IDA countries-- improving aid predictability will require improving annual and multiyear budget systems and medium-term frameworks as well as improving the reliability and accessibility of donor disbursement information. 49. Improving Use of Information. The Bank is helping countries improve their budget systems and develop medium-term fiscal frameworks that will enable them to better use 32 See A Management Framework for World Bank-Administered Trust Funds (R2007-0198), October 9, 2007, and Guidance note on Integration of Trust Funds in Country Assistance Strategy Products. 33 Of 62 IDA-recipient countries reviewed, 4 have medium-term fiscal frameworks that have served as essential building blocks for making their national development strategy operational through budget allocations. See Results Based National Development Strategies: Assessment and Challenges Ahead, World Bank, December 2007. 25 information on expected disbursements. It provides diverse support to partner countries-- particularly IDA-recipient countries--to improve their public financial management (PFM): the Bank is the leading donor in providing support for strengthening public sector management;34 PFM reforms constitute the dominant share of the Bank's public sector management portfolio; and PFM conditions account for more than half of the public sector governance actions in DPOs to IDA countries. Bank PFM analytic work--such as PFM Performance Reports using the Public Expenditure Financial Accountability program framework, and the institutional aspects of Public Expenditure Reviews and Country Financial Accountability Assessments--helps countries assess and identify the strengths and weaknesses of their PFM systems. 50. Institutional/Global Levels. The Bank works at the institutional and global levels to complement its country-level work to improve aid predictability. With the OECD-DAC Working Party on Aid Effectiveness, the Bank will take the lead on issues related to predictability, working in close partnership with diverse donors and partner countries. The work program will likely focus on measurement; use of data at the country level; country capacity development needs; institutional constraints of donors and potential for improvements in disbursement forecasts, emphasizing the differences across donor groups stemming from differences in their enabling environments; and aid transparency. This work will be undertaken in tandem with work by other agencies (e.g., DAC statistics) and other initiatives (e.g., the International Aid Transparency Initiative). B. Poverty and Social Impact Analysis 51. The Bank has met its commitment to update its good practice note on poverty and social impact analysis (PSIA). The note dovetails well with the AAA.35 It aligns the focus of PSIA with country priorities and builds ownership of reforms; focuses on building country capacity to carry out country-led PSIA; encourages making PSIA findings available for informed public debate by a wide range of stakeholders; and highlights the usefulness of PSIA for building demand for measuring the impacts of reforms, and strengthening the country's overall national monitoring and evaluation framework. It does not present a template or "best practice" for PSIA; instead, it identifies important dimensions of good practice PSIA, providing examples that show pragmatic solutions to real situations. Since completing the good practice note, the Bank has taken additional steps to enhance its PSIA. Two stocktaking measures--one on the use of PSIAs to inform DPOs (which was updated in the DPL Retrospective) and one on the effectiveness of the PSIA approach in supporting partner country policy and planning processes--found improved quality and increased use of PSIA. The Bank's Independent Evaluation Group also recently completed a study on PSIA, which, like the two internal studies, suggested the need to focus more on outreach and learning efforts to build Bank staff awareness of and capacity in the PSIA approach. To help consolidate good practice on PSIA, the Bank is initiating a multidonor trust fund, to become effective in FY10,that will support more routine use of PSIA and support partner governments in examining the poverty, social, and distributional impacts of key policy reforms and operational activities. 34 See Country-Based Scaling Up: Assessment of Progress and Agenda for Action (DC2007-0017), September 26, 2007. 35 See Good Practice Notes for Development Policy Lending, OPCS, October 2004, and Good Practice Note: Using Poverty and Social Impact Analysis to Support Development Policy Operations, August 2008. 26 C. Transparency, Accountability and Results: Way Forward 52. Overall, the Bank will aim to improve the reliability, availability, and timing of Bank disbursement data and better communicate this information to governments to help reduce discrepancies between what governments and donors forecast and report. With the ultimate objective of strengthening accountability, the Bank will work with international initiatives that focus on reliable and transparent aid data. These efforts need to be coupled with supporting countries in improving budgeting systems and medium-term fiscal frameworks, with the goal of getting aid on-budget so that its use can be transparently monitored. Beyond predictability, and as highlighted in the AEAP, the Bank's revamped disclosure policy may soon bring the Bank to a new level of transparency, increasing accessibility of information to diverse external actors.36. The Bank will also continue to participate in mutual accountability reviews at the country level, drawing on the good practices for these exercises that are distilled by others. In parallel efforts, the Bank will continue to mainstream governance and anticorruption, for example, through the issuance of guidance notes as part of the GAC agenda37 and through implementation of the IDA controls review to ensure an effective and efficient use of funds as intended. Finally, the Bank is focusing on implementing the results agenda within the institution and supporting governments in their efforts to better link aid effectiveness to development effectiveness. This will be aided by the new Statistics for Results Facility, supported by donors, to improve statistical capacity linked to country-led monitoring of PRSPs and other development strategies. The IDA15 MTR paper on results showcases this agenda, with a focus on Bank results, country results and core sector indicators.38 V. SHAPING THE AE AGENDA AND SUPPORTING IMPLEMENTATION 53. To mainstream aid effectiveness principles and practices and strengthen the Bank's ability to make further progress on the AE agenda, the Bank is looking at internal corporate incentives, staff capacity development and the location of staff and responsibilities. The IDA15 commitments in this area were to update progress on the Matrix of Actions and develop a Bank Action Plan; ensure reporting by Regional Management to the Board; enhance staff incentives and guidance on ownership, harmonization, and alignment; and deepen decentralization by continuing to explore options, completing analysis of different models, and meeting specific targets for the Africa Region. This section reviews IDA15 commitments related to staff incentives and capacity and decentralization. 36 The Bank proposes a paradigm shift in its approach to disclosing information, from an approach that spells out what documents the Bank discloses (a "positive list") to one under which the Bank would disclose any information in its possession that is not on a list of exceptions--a policy that would be more consistent with the Bank's expressed presumption in favor of disclosure, and more in line with recognized international best practice. This approach is consistent with the Bank's business model, which recognizes the importance of transparency as a critical tool for enhancing good governance, accountability, and development effectiveness. See Towards Greater Transparency: Proposed Revisions to the World Bank's Disclosure Policy (forthcoming). 37 OPCS has issued "GAC in Projects: Emerging Good Practices," a guidance note developed by the GAC in Projects Working Group following consultations with task teams and operations staff. The note provides an organizing framework to help task teams better manage GAC risks at the project level. It also contains examples of approaches, tools, and techniques that have been effective in some projects, and live links to other sources of information and guidance. 38 See IDA15 Results Measurement System: Mid-Term Review, October 2009. 27 A. Matrix of Actions and Action Plan 54. The Bank has met both parts of its twofold commitment on action plans: to update progress on the Matrix of Actions for Harmonization and Alignment, and to develop a draft World Bank Action Plan for aid effectiveness in preparation for the HLF in Accra. The Matrix of Actions, updated to reflect the findings of this review, is attached as Annex B. For the second commitment, the Bank has taken the messages from Accra and translated them into the Bank's Aid Effectiveness Action Plan (AEAP), reflecting the evolution of the agenda into one that focuses on Bank actions to strengthen partner countries' ownership, capacities, and structures across the full spectrum of AE agenda issues; strengthen partnerships, especially by engaging actively with nontraditional partners; and continue the emphasis on results and transparency. The AEAP discusses both areas of priority to deepen the AE agenda within Bank operations and the Bank's leadership role on behalf of the multilateral development banks and the entire international community in shaping and implementing the AE agenda. The AEAP, which was discussed with the Bank's Executive Directors in March 2009,39 will be implemented in tandem with--and leveraging--other critical Bank initiatives: investment lending reform, decentralization, the fragile and conflict-affected countries agenda, managing for development results, and others. B. Regional Reporting 55. This commitment represents an ongoing activity: Regional Management reports regularly to the Board on progress on the AE agenda, and will continue to do so. The focus of the 2009 Regional Strategies to the Board was more muted--but not silent--on the AE agenda. This was largely because of the prominence of the global financial crisis and its impact across client countries. All the Regions highlighted the importance of coordinating assistance with other donors, the Africa Region most expansively to include specific initiatives with non-traditional donors. But many other issues--such as building country capacity to articulate priorities and strengthening country systems to permit greater use of them in Bank financed activities--were also raised as medium-term goals. Most of the regional strategies emphasized the vulnerability of many clients to crises, whether financial, natural/climate-related, or political-- particularly for fragile states--and the need for the Bank to remain flexible in the face of these risks. Bank shareholders and Management have also stayed focused on the AE agenda in conjunction with CAS discussions. A review of recent CAS discussions suggests that Executive Directors often raise AE issues, particularly for IDA countries. However, while support for the overall agenda is strong, Board members and Senior Management have suggested that different areas of the agenda should be emphasized in different country circumstances. Also, the Board continues to send mixed messages on certain aspects of the agenda--particularly greater use of country systems. C. Decentralization 56. Decentralization is expected to enhance many aspects of the AE agenda by delegating daily operational matters and dialogue to field staff and getting expertise (e.g., sector, fiduciary 39 For a full discussion of the action plan, see Following up on Accra: A World Bank Plan on Aid Effectiveness (SecM2009-0107), March 5, 2009. 28 and safeguard expertise) closer to clients. As Management reported to the Executive Directors in June 2009, the Bank has met its commitments related to decentralization.40 For the Africa Region (AFR): The number of internationally recruited staff (IRS) based in the field increased between FY06 and FY08 by over 50 percent, and as of the end of FY09, the increase had already reached 57 percent. This effort will continue through the IDA15 period. In addition, the percentage of tasks (appraisal, supervision, and analytic and advisory work) that AFR has devolved to the field has climbed appreciably over the past year, from 39 percent in FY08 to nearly 47 percent in FY09. For other Regions, the Bank has continued to pursue decentralization, especially in IDA countries and fragile and conflicted-affected countries (FCC). Since the end of FY06, IRS in IDA countries outside of Africa have increased by 21 percent. The Bank has also committed to increase the number of IRS in FCC by 30, to open two new country offices, and, where appropriate, to place a second IRS in countries where there is only one. Since these commitments were made, IRS in FCC have increased by 28 (30 percent) and three new offices--in the Solomon Islands, Tonga, and Samoa-- have been opened. 57. Exploring a New Approach. As the development arena becomes even more dynamic with new partners providing expertise and financial options to clients, clients are demanding higher-quality and more customized solutions. In response, the Bank is exploring ways to develop the next generation approach to decentralization, as elaborated in recent reports.41 A working group is examining various elements that will allow the Bank to be closer to clients and partners through greater field presence, while complementing existing services provided by World Bank Group offices with more emphasis on global expertise and staff. Issues being examined include costs, numbers of technical staff, recruitment, and knowledge sharing across countries. The Bank is placing particular emphasis on: (a) increasing staff in and devolving tasks to country offices in FCC and low-income countries, which face particularly difficult development issues and have limited capacity to deal with them; and (b) creating better accessibility to staff with specialized knowledge and skills. The Bank is also progressing on key initiatives that will support a new decentralization model: knowledge management, which will enable the Bank to retain its position as a source of global knowledge; investment lending reform, which will shift some responsibilities for Bank decision-making; the Global Human Resources Framework, to address recruitment issues; and information technology solutions, to enhance knowledge sharing across countries. D. Staff Incentives and Capacity 40 For more details see Toward a Global Bank: Discussion Note (SecM2008-0428), October 8, 2008, and Toward a Global Bank: An Update (SecM2009-0326), June 25, 2009. 41 Idem. 29 58. The Bank considers its IDA15 commitment of enhancing staff incentives and guidance an ongoing activity. Bank staff have been highly motivated by the potential benefits of the AE agenda, and in particular by concrete results at the country level. They have provided feedback on the AE agenda through a variety of channels: staff interviews, focus groups, and survey responses. Their input indicates that the most important incentives relate directly to the partner country and its signals to the Bank: when the client country leads in designing and implementing its development program, and is clear about the role it wants the Bank to play, Bank staff respond. And when staff see concrete benefits at the country level, they are more committed to the AE agenda. Staff are also driven by the belief that implementation of the AE agenda will benefit clients: over three-quarters of the 200 Bank staff responding to a survey agreed that achieving the AE commitments would improve development outcomes. 59. Staff Understanding. Approximately three-quarters of staff surveyed indicated that they have a good understanding of the commitments in the Paris Declaration--for instance, working more effectively with donors to support country-led priorities. Staff also appreciate the newer elements of the agenda stemming from Accra, including domestic accountability (e.g., to Parliament and civil society) and collaborating more intensively with nontraditional donors. In some of these areas, the agenda is taking hold: almost three-quarters of staff surveyed suggested that the Bank helps governments to strengthen domestic accountability. In other cases, however, staff feel that more support is needed: while over two-thirds of staff surveyed feel that the Bank works effectively with traditional donors, only one-third suggested that the Bank works effectively with nontraditional donors. 60. Management Support. Staff also indicated that they have support from Management on this agenda. Both the Board and Senior Management have been consistent in conveying the importance they place on country ownership: staff understand that the country is the focus of the Bank's work, that CASs are to be aligned with government development strategies, and that there are mechanisms and instruments by which Bank activities are aligned with country priorities. Regional Management at all levels has provided strong encouragement to adopt AE principles, including introducing tools such as emphasizing AE in job descriptions, and has empowered staff with the flexibility to experiment with innovative solutions in areas such as aid coordination. Of the staff responding to the survey, 68 percent agreed that their managers consider aid effectiveness important, although close to half suggested the need for more resources to engage meaningfully in all aspects of aid effectiveness. 61. Areas for Improvement. Staff suggest a number of areas for improvement to bolster the aid effectiveness agenda. Broadly, they seek: More operational advice for implementation. As the international community moves from isolated examples of good practice to mainstreaming aid effectiveness, the Bank will focus on gathering the rich experience of staff, donors, and countries to facilitate implementation and increase traction at the country level. Clearer signals from Management especially on risks and results. Staff have suggested that the signals on risk are mixed. Interviews, focus groups, and informal feedback suggests a widespread staff perception that the institution is becoming increasingly risk- 30 averse and rule-based, and that staff and Management are rewarded for avoiding risks.42 Most important, only one-third of the staff responding to the survey thought that Bank leadership encouraged risk-taking when it contributes to improved development outcomes. Staff report that they feel pressure related to speed of delivery and quantity of deliverables (commitments, products, disbursements), which translates into perceptions of tensions between quantity and quality and between short-term lending commitments and long-term sustainability. Policies and procedures consistent with supporting the high-level principles of aid effectiveness. Staff want more clarity on aid effectiveness priorities (only 40 percent of staff responding to the survey agreed that these priorities are clear) and they want policies and procedures that are in sync with the high-level principles of aid effectiveness (only one-third agreed that Bank policies and procedures supported AE efforts). E. Implementation: Way Forward 62. The Bank is acknowledging the challenges described above and is beginning to act accordingly. To ease the tensions staff feel, the Bank is striving to highlight a more clear risk- results approach to its operations. The investment lending reform is focusing on defining a clearer typology of risk, signaling a fundamental shift toward a more sophisticated risk/results- based approach, which will enable staff to better identify and mitigate different risks related to development objectives in lending operations (see Box 9). The Bank has also made a significant Box 9. Main Areas of Focus for the IL Reform--Initial Phase The proposed risk based approach will: (a) help teams design operations that take into account the operating environment, (b) link risks to desired results and development objectives, (c) differentiate projects on the basis of risks and allow a closer link to resources, (d) create a better environment for learning about risks, and (e) allow the Bank to examine the risk attributes of the portfolio. Rebalancing the effort from preparation to implementation, will be brought about through a variety of means: (a) a culture change from supervision to implementation support; (b) measures to support implementation planning and team integration; (c) reforms to make support for implementation more flexible and efficient; and (d) changes in the staffing and enabling environment to provide staff with the tools they need and to ensure an appropriate geographic distribution of staff to achieve these objectives. effort to improve the focus on results at the project, country, sector, and corporate levels to steer the institution to a more outcome-oriented approach. All CASs are now results-based--with continuing work on strengthening the quality of results frameworks. The recently completed DPL Retrospective also pointed to the need to continue focusing on results matrices. Changing the Bank's culture related to risks and results will provide an important signal to staff about Management's priorities. To complement this, the Bank has deepened decentralization-- especially focusing on the number of internationally recruited staff and percentage of tasks 42 Staff report that they typically spend more than half of the preparation time for investment lending operations on fiduciary aspects, that this proportion is on the increase, and that more and more implementation support is for procurement-related issues. 31 devolved to the field--and is exploring a new approach that would expand decentralization in a more directed fashion than in the past. 63. Bridging the Gap between Understanding and Implementation. The Bank will also increase its efforts to support staff in implementing the AE agenda, especially in IDA countries. What is needed is more guidance in diverse areas of AE (e.g., partnering with nontraditional donors, common MOU practices) featuring operational advice for implementation. This is an area of emphasis in the AEAP, and the Bank has been increasing its focus on knowledge and learning related to AE and revamping the website to be able to provide more updated examples of experience--but it can do more to provide guidance, disseminate examples of good practice, and enhance staff training in key areas related to the AE agenda. The Bank will continue to monitor internal progress on the aid effectiveness agenda, and to provide feedback to staff on where progress has been made and where areas for improvement remain.43 In addition, the Bank is working with the international community--notably in its close involvement with the Working Party on Aid Effectiveness (WP-EFF) hosted by the OECD-DAC, and other global initiatives (see Box 10)--and contributes to advancing technical issues and identifying global good practices through these efforts. Box 10. OECD-DAC Working Party on Aid Effectiveness and the World Bank The WP-EFF promotes, supports, and monitors progress on aid effectiveness of donor practices and alignment with country-owned development strategies. The Bank, through its support to the WP-EFF and its working groups, continues to help shape the international aid effectiveness agenda at the global level and supports its implementation. Bank responsibilities include: Serving as co Vice-Chair of the WP-EFF and its Executive Committee. Providing technical leadership by chairing the financial management, procurement, and predictability/transparency task teams. Providing significant support to the task teams on conditionality, South-South cooperation, MfDR and monitoring of the Paris Declaration. Leading work on fragile states with the United Nations and others. Providing major support for the planning, preparation and execution of the Third HLF in Accra in 2008 and expected support for the Fourth HLF to be held in Korea in 2011. 43 Going forward, the Bank may benefit from WBI's recent "Capacity Development Results Framework", which may help with monitoring aid effectiveness and capturing lessons learned. 32 VI. CONCLUSIONS 64. The Bank's aid effectiveness agenda--reflecting the AAA--highlights the centrality of the country: the theme of country ownership is dominant. A direct corollary of ownership is the need to focus more on developing country capacity and systems. In AE efforts, donors need to keep in sight the primacy of country leadership, leveraging joint donor activities to strengthen country capacity. A recent evaluation of the implementation of the Paris Declaration confirms this, pointing to specific areas where donors can assist countries.44 65. Country Differences. Partner countries are seeking pragmatic solutions tailored to country conditions, which underscores the importance of identifying good practices from diverse country experiences. The relevance of different aspects of the AE agenda varies across client segments, as do the solutions. For example, countries with fewer donors and stronger leadership are less interested in coordination exercises and joint financing--issues that are extremely relevant to aid dependent countries. The relevance of aid predictability is less in countries where aid is a small part of the budget. Countries that have weaker budgeting capacity and processes need more capacity development in this area as a primary focus to improve aid predictability and aid on-budget. And countries with weaker financial management and procurement systems require more tailored approaches to project management. 66. Donor Differences. Just as countries differ from one another, so do donors. Emerging donors differ from established donors in their attributes and approaches, and multilateral donors operate under different mandates and policies from those of bilateral donors (such as lending authorization and fiduciary requirements). These differences suggest that some aspects of aid effectiveness are more relevant to some donors than others, and that some aspects of donor coordination will yield more results than others. Important measures such as decentralization and legal harmonization can improve donor coordination, but will not resolve underlying issues related to differences in institutional policies and mandates. 67. Ongoing Effort. While the Bank has met or is meeting all the commitments under the IDA15 Replenishment, it will continue efforts by implementing the actions outlined in this paper and in the AEAP. Aid effectiveness comprises a broad agenda, and for the Bank it is not an isolated agenda: it cuts across and draws on major corporate initiatives such as reform of disclosure policy and investment lending, the results agenda, IDA controls, decentralization, and the fragile states agenda. These efforts will bring the Bank even closer in alignment with clients, realign the Bank's approach to risks and results, and make the Bank more transparent and accountable for concrete results--thus deepening and mainstreaming aid effectiveness. Going forward, the Bank will integrate new knowledge and good practice as it is developed--focusing on those priorities that will yield the highest value-added to countries, weighing costs and benefits, and balancing the ideal and the practical. 44 The study also highlights that the indicator of country ownership defined in Paris and monitored by the survey--that countries have an operational development strategy--captures only part of a complex story. 33 Annex A. The Surveys on Monitoring the Paris Declaration: Methodology and Bank Performance 1. The Paris Declaration on Aid Effectiveness is translated into over 50 commitments and a set of 12 quantitative indicators with targets. It is underpinned by an internationally agreed process--Survey on Monitoring the Paris Declaration--to monitor the implementation progress of these indicators toward targets set for 2010. 2. Survey Rounds. OECD-DAC, in partnership with donors and country governments, has taken the lead on the surveys monitoring the Paris Declaration. Two rounds of monitoring have taken place. The first round was conducted in 2006, with a sample of 34 countries, on the basis of activities undertaken in 2005; it served as the baseline for measurement. The second round of monitoring was conducted in 2008, with a sample of 55 countries, on the basis of activities undertaken in 2007; it served as a midterm assessment of progress under these indicators. The next round is due to take place in 2010. Because the countries and donors participating in the Surveys are self-selected, not a random sample, they may not represent and reflect the overall population. 3. Comparability of Data. The purpose of the Survey is to analyze progress over time. However there are various reasons why it is difficult to compare data from the 2006 and 2008 Surveys to study trends. Sample coverage. The 2006 Survey is based on feedback from 34 self-selected countries, while for the 2008 Survey 54 partner countries volunteered to organize the survey in their own countries. About 33 countries are common to both samples. Time coverage. The 2006 Survey covers fiscal year 2005, which might not be the same for donor and country, using fiscal year 2004 data if the government's fiscal year 2005 is not complete at the time of undertaking the survey. The 2008 Survey covers calendar year 2007--which does not mirror Bank fiscal year reporting. Methodological differences. Between the two surveys the methodology (criteria and/or measurement of indicator) was refined for four of the indicators--Indicators 4, 5a, 6, and 9 (see below for more details). In addition, when aggregating data for a particular donor or a particular country on any of the indicators, using the weighted average produces very different results from using the unweighted average--and there is no concrete justification as to why one way of aggregating data is better or more accurate to reflect performance against these indicators. 4. Despite these challenges, the Survey is an important tool to track and encourage progress in implementing the Paris Declaration at the country level. The indicators are proxies for measuring the principles of the Paris Declaration--ownership, harmonization, alignment, managing for results, and mutual accountability--and are needed for standard comparison. To complement these measures and fully reflect the nuances of performance, each institution may want to develop additional means to measure and collect supplemental qualitative feedback. (For two of the indicators in particular--Aid on Budget and Aid Predictability--there has been a great deal of debate on the credibility/validity of the measurement; see Annex C for more detail). 34 5. Coverage of this Annex. Of the 12 indicators, 4 are based on independent reviews to assess the quality of country national development strategies and results-oriented frameworks, the reliability of public management systems, and the existence of a mutual accountability framework in a country (Indicators 1, 11, 2, and 12 respectively). The remaining 8 indicators are based on survey responses from the governments and donors (Indicators 3, 4, 5, 6, 7, 8, 9, and 10). This annex reviews the methodology of seven of these eight indicators.45 For these indicators, participating countries and donors provided feedback in response to questionnaires and guidance. Variations in the quality and completeness of data and in the interpretation of the guidance for data collection and reporting limit comparability across countries and across donors. Indicator 3: Aid flows are recorded on budget 6. The Paris Declaration commits donors to "base their overall support--country strategies, policy dialogues and development co-operation programs--on partners' national development strategies and periodic reviews of progress in implementing these strategies." The Paris Declaration includes in its monitoring framework an indicator to measure the "percent of aid flows to the government sector that is reported on partners' national budgets." This was later revised to measure "the gap between what was disbursed by donors for the government sector and what was actually recorded in the annual budget by government." The Paris Declaration surveys have thus collected data from two different sources--from donors on the amount of aid they disbursed for the government sector and compared this with data collected from governments on the amount of estimated aid recorded in government systems. The 2006 survey introduced a target to reduce by half the proportion of aid not disbursed within the year that it was scheduled. Based on the survey results, this translates to a target to reduce the percentage of aid not disbursed with the year that it was scheduled to 85 percent. Both the 2006 and the 2008 surveys use the same aid on budget criteria (see Box A1). Box A1. Aid on Budget Mixed donor-government perspective: Measurement of Indicator: Ratio of Aid disbursed for the government sector at the country level and estimated aid recorded in budget with the higher value as the denominator. Criteria: ODA disbursed for the government sector at the country level (donor) Funds disbursed by donors for disbursement in year x in the context of an agreement between donors and governments. Estimated ODA recorded in budget (government) ODA recorded in the annual budget as grants, revenue or ODA loans at the beginning of year x. Indicator 4: Technical Assistance is aligned and coordinated 7. The Paris Declaration commits donors to "align their analytic and financial support with partners' capacity development objectives and strategies, make effective use of existing capacities and harmonize support for capacity development accordingly." The Paris Declaration includes in its monitoring framework an indicator to measure the "percent of donor capacity- 45 Indicator 8 is a measurement for untied aid. For the Bank, all of its aid is untied. 35 development support provided through coordinated programs consistent with partners' national development strategies." The Paris Declaration set as a target that 50 percent of technical assistance (TA) flows are implemented through coordinated programs consistent with national development strategies. The indicator is defined as a measure of the proportion of TA that was disbursed through coordinated programs in support of capacity development for the calendar year. There has been a refinement in the methodology to assess coordinated technical cooperation. Four criteria for coordinated TA: had been identified 1) clear capacity development objectives are identified in national or sector strategies; 2) technical cooperation is aligned with country development objectives; 3) country authorities have control over technical cooperation; 4) if more than one donor is involved, country authorities coordinate technical cooperation. In 2007, only criteria 1, 2 and either 3 or 4 had to be met for TA to qualify as coordinated, while in 2005, all four criteria had to be met (see Box A2). Box A2. Coordinated Technical Assistance Donor Perspective: Measurement of Indicator: Amount of TA provided through coordinated programs in support of capacity development/ Total amount of TA provided Criteria: 1. Have relevant country authorities (government or non-government) communicated clear capacity development objectives as part of broader national or sector strategies? (Y/N) 2. Is the technical co-operation aligned with the countries' capacity development objectives? (Y/N) 3. Do relevant country authorities (government or non-government) have control over the technical co- operation? (Y/N) 4. If more than one donor is involved in supporting country programs, are there arrangements involving the country authorities in place for coordinating the technical co-operation provided by different donors? (Y/N) Indicator 5a: Use of Public Financial Management Systems 8. The Paris Declaration commits donors to "use country systems and procedures to the maximum extent possible. Where use of country systems is not feasible, establish additional safeguards and measures in ways that strengthen rather than undermine country systems and procedures."The Paris Declaration includes in its monitoring framework an indicator to measure the "percent of donors and of aid flows that use public financial management (PFM) systems in partner countries which either (a) adhere to broadly accepted good practices or (b) have a reform program in place to achieve these." The Paris Declaration set as a target of a one-third reduction in the percent of aid to the public sector not using partner countries' PFM systems for partner countries with a score between 3.5 and 4.5 on the PFM/CPIA scale of performance and a two third reduction for partner countries with scores of 5 and over. The indicator is defined as the measure of the volume of aid that uses national PFM systems as a percent of total aid provided for the government sector with a 2010 target set at 80%. This target was revised in the AAA and committed donors to aim to channel 50% or more of government-to-government assistance through country PFM systems. There has been a refinement of the methodology to assess use of PFM systems. In 2007, specific criteria were provided to assess the use of national budget execution, financial reporting and auditing procedures, with clear guidelines on how to apply them. In 2005, definitions were provided for the use of the three national procedures, without including criteria on how to assess their use (see Box A3). 36 Box A3. Use of national financial management systems Donor Perspective: Measurement of Indicator=((Amount of ODA disbursed for the government sector that used NBE systems +Amount of ODA disbursed for the government sector that used NFR systems+ Amount of ODA disbursed for the government sector that used NA systems)/3) / Total Amount of ODA disbursed for the government sector Criteria: Use of national budget execution procedures. 1. Are your funds included in the annual budget approved by country legislature? (Y/N) 2. Are your funds subject to established country budget execution procedures? (Y/N) 3. Are your funds processed (e.g. deposited & disbursed) through the established country treasury system? (Y/N) 4. You do NOT require the opening of separate bank accounts for your funds? (Y/N) Use of national financial reporting procedures. 1. You do NOT require maintenance of a separate accounting system to satisfy your own reporting requirements? 2. You ONLY require financial reports prepared using country's established financial reporting arrangements? (Y/N) Use of national auditing procedures. 1. Are your funds subject to audit carried out under the responsibility of the Supreme Audit Institution? (Y/N) 2. You do NOT under normal circumstances request additional audit arrangements? (Y/N) 3. You do NOT require audit standards different from those adopted by the Supreme Audit Institution? (Y/N) You do NOT require the SAI to change its audit cycle to audit your funds? (Y/N) Indicator 5b: Use of Procurement Systems 9. The Paris Declaration commits donors to "use country systems and procedures to the maximum extent possible. Where use of country systems is not feasible, establish additional safeguards and measures in ways that strengthen rather than undermine country systems and procedures." The Paris Declaration includes in its monitoring framework an indicator to measure the "percent of donors and of aid flows that use partner country procurement systems which either (a) adhere to broadly accepted good practices or (b) have a reform program in place to achieve these." The Paris Declaration set as a target of one-third reduction in the percent of aid to the public sector not using partner countries' procurement systems for partner countries with a score of 'A' on the Procurement scale of performance and a two third reduction for partner countries with a score of 'B'. The indicator is defined as the measure of the volume of aid that uses national country procurement systems as a percent of total aid provided for the government sector with a 2010 target set at 80%. This target was revised in the AAA and committed donors to aim to channel 50% or more of government-to-government assistance through country procurement systems. Both the 2006 and the 2008 surveys use the same use of national procurement systems criteria (see Box A.4) 37 Box A4. Use of national procurement systems Donor Perspective: Measurement of Indicator=Amount of ODA disbursed for the government sector that used national procurement systems/ Total Amount of ODA disbursed for the government sector Criteria: Donors use national procurement procedures when the funds they provide for the implementation of projects and programs are managed according to the national procurement procedures as they were established in the general legislation and implemented by government. The use of national procurement procedures means that donors do not make additional, or special, requirements on governments for the procurement of works, goods and services. Indicator 6: Use of Parallel Project Implementation Units 10. The Paris Declaration commits donors to "avoid, to the maximum extent possible, creating dedicated structures for day-to-day management and implementation of aid-financed projects and programs." The Paris Declaration includes in its monitoring framework an indicator to measure "strengthening capacity by avoiding parallel implementation structures" The Paris Declaration set as a target of a reduction by two-thirds the stock of parallel project implementation units (PIUs). The indicator is defined as the measure of the total number of parallel PIUs in a single country. There has been a refinement in the methodology to assess parallel PIUs. In 2007, an additional criteria for parallel PIUs--terms of reference for externally appointed PIU staff are determined by the donor-- was added to the three criteria identified in 2005: 1) PIUs are accountable to the donor; 2) PIU staff are appointed by the donor; 3) the salary structure of PIU staff is higher than that of the national civil service. In 2007, three out of these four criteria had to be met for a PIU to qualify as parallel. In 2005, no clear guidelines had been provided on the application of the assessment criteria (see Box A5). Box A5. Parallel PIUs Donor Perspective: Measurement of Indicator=Number of parallel PIUS Criteria: 1. Are the PIUs accountable to the external funding agencies/donors rather than to the country implementing agencies (ministries, departments, agencies etc)? (Y/N) 2. Are the terms of reference for externally appointed staff determined by the donor (rather than by the country implementing agencies)? (Y/N) 3. Is most of the professional staff appointed by the donor (rather than the country implementing agencies)? (Y/N) 4. Is the salary structure of national staff (including benefits) higher than those of civil service personnel? (Y/N) Indicator 7: Aid Predictability 11. The Paris Declaration commits donors to "provide reliable indicative commitments of aid over a multi-year framework and disburse aid in a timely and predictable fashion according to agreed schedules." The Paris Declaration includes in its monitoring framework an indicator to measure the "percent of aid disbursements released according to agreed schedules in annual or multi-year frameworks." The Paris Declaration set as a tentative target that 75 percent of aid be 38 released on schedule by 2010, with the target subject to revision. The target was later revised to 71 percent. The indicator is defined as the measure of "the gap between aid scheduled and aid effectively disbursed and recorded in countries' accounting systems." The Paris Declaration surveys have thus collected data from donors on the amount of aid scheduled for disbursement and compared this with data collected from governments on the amount of aid recorded in government systems. Both the 2006 and the 2008 surveys use the same aid predictability criteria (see Box A.6). Box A6. Aid Predictability Mixed donor-government perspective: Measurement of Indicator= Ratio of aid actually received and aid scheduled for disbursement with the higher value as the denominator Criteria: ODA scheduled for disbursement (donor) Funds scheduled by donors for disbursement in calendar year 2007 and notified to government within calendar year 2006. ODA actually received (government) ODA actually received in the context of agreements between donors and the government, as it was recorded in government accounting/reporting systems. Indicator 9: Program-based approaches 12. The Paris Declaration commits donors to "implement, where feasible, common arrangements at country level for planning, funding (e.g. joint financial arrangements), disbursement, monitoring, evaluating and reporting to government on donor activities and aid flows. Increased use of program-based aid modalities can contribute to this effort." The Paris Declaration includes in its monitoring framework an indicator to measure "percent of aid provided as program-based approaches (PBAs)." The Paris Declaration set as a target that 66% of aid flows are provided in the context of PBAs. The indicator is defined as the measure of ODA provided in support of initiatives adopting PBAs as a percent of total ODA. There has been a refinement in the methodology to assess whether donor support is part of a program-based approach (PBA). In 2007, to qualify as a PBA, a formal process for donor coordination and harmonization had to be in place for at least two of the following systems: reporting, budgeting, financial management and procurement. Similarly, donor support had to use at least two of the following local systems: program design, program implementation, financial management and monitoring and evaluation. In 2005, donor coordination and harmonization had to be in place for all systems, and donors had to make efforts to increase use of all local systems (see Box A7). 39 Box A7. Program-based approaches Donor Perspective: Measurement of Indicator = (Amount of direct budget support provided in support of PBAs+ Amount of other donor assistance in support of PBAs)/ Total amount of ODA disbursed Criteria: 1. Is the host country or organization exercising leadership over the program supported by donors? (Y/N) 2. Is a single comprehensive program and budget framework used? (Y/N) 3. Is there a formal process for donor co-ordination and harmonization of donor procedures for the following systems: (i) reporting, (ii) budgeting, (iii) financial management and (iv) procurement? (Y/N) 4. Does your support to the program use of the following local systems: (i) program design, (ii) program implementation, (iii) financial management and (iv) monitoring and evaluation? (Y/N) Indicator 10a: Joint missions 13. The Paris Declaration commits donors to "work together to reduce the number of separate, duplicative, missions to the field." The Paris Declaration includes in its monitoring framework an indicator to measure the "percent of field missions that are joint." The Paris Declaration set as a target that 40 percent of donor missions to the field are joint. The indicator is defined as the measure of the proportion of coordinated missions as a percent of the total number of missions for the calendar year. A coordinated mission is defined as one that is undertaken jointly by different donors or by one donor on behalf of another (see Box A8). Both the 2006 and the 2008 surveys use the same joint missions criteria. Box A8. Joint missions Donor Perspective: Measurement of Indicator= Number of coordinated missions/Total number of donor missions to the field Criteria: Missions to the field This includes (i) missions undertaken by, or on behalf of, a donor, including program developers, appraisers and evaluators, sector assessment teams commissioned by a donor; (ii) missions involved with international travel typically, but not exclusively, from donor headquarters; and (iii) missions where donors made a request to meet with government officials including local government. Coordinated missions This includes (i) missions undertaken by one or more donor jointly, or (ii) missions undertaken by one donor on behalf of another donor (delegated co-operation). Indicator 10b: Joint country analytic work 14. The Paris Declaration commits donors to "work together to reduce the number of separate, duplicative, diagnostic reviews." The Paris Declaration includes in its monitoring framework an indicator to measure the "percent of country analytic work (CAW), including diagnostic reviews that are joint." The Paris Declaration set as a target that 66 percent of country analytic work are joint. The indicator is defined as the measure of the proportion of country analytic reports or reviews that are coordinated as a percent of the total number of reports or reviews for the calendar year. The Survey defines coordinated CAW in three ways. To be considered coordinated, analytical work must be undertaken jointly with another donor, 40 undertaken by one donor on behalf of another or undertaken with substantive involvement from the government. Both the 2006 and the 2008 surveys use the same coordinated CAW criteria (see Box A9). Box A9. Joint country analytic work Donor Perspective: Measurement of Indicator =Number of coordinated CAW/Total number of CAW Criteria: Country analytic work This include major pieces of analytic work such as: (i) diagnostic reviews (e.g. Country Procurement Assessment Report, Country Financial Accountability Assessments etc.); (ii) country or sector studies and strategies; and (iii) country or sector evaluations. Coordinated country analytic work This includes (i) CAW undertaken by one or more donor jointly; (ii) CAW undertaken by one donor on behalf of another donor (including work undertaken by one and/or used by another when it is co-financed and formally acknowledged in official documentation); and (iii) CAW undertaken with substantive involvement from government. Bank Performance 15. The table and the figure below describe Bank performance in the two Surveys. Table A1 compares 2006 and 2008 data for the Bank against the 2010 target. It illustrates Bank performance for the 32 original countries participating in the 2006 Survey and the 2008 Survey as well as Bank performance for the entire sample of 51 countries participating in the 2008 Survey. There has been no guidance on what sample of countries is be used to compare the 2008 and 2006 Surveys and accurately report a trend. Some donors compare the countries which responded to both surveys (which for the Bank would be only the 32 countries in common for both surveys) and some compare all 2006 respondents with all 2008 respondents (for the Bank, the 32 countries for the 2006 Survey with the entire sample of 51 countries participating in the 2008 Survey). Figure A1 shows the results of the monitoring surveys for the Bank against the average performance of all Survey participants. Bank performance for IDA countries (IDA only and IDA/IBRD blend) is very similar to the overall Bank performance for the 2008 Survey. Table A1. Bank performance for the 2006 and 2008 Results 2006 2008 2008 Indicators 32 32 51 2010 Target countries countries Countries Aid flows are aligned on national 3 62% 71% 66% 85% priorities Strengthen capacity by co-ordinated 4 57% 86% 85% 50% support 5a Use of country PFM systems 42% 54% 62% 50% 5b Use of country procurement systems 40% 44% 52% 50% Strengthen capacity by avoiding Parallel 6 223 79 101 Reduce to 33% the stock of parallel PIUs PIUs 7 Aid is more predictable 63% 70% 65% 71% Use of common arrangements or 9 57% 56% 54% 66% procedures 10a Joint missions 21% 31% 31% 40% 10b Joint country analytic work 49% 56% 59% 66% 41 Figure 2. Survey on Monitoring the Paris Declaration Figure A1. Survey on Monitoringthe Paris Declaration A comparison between the 2005 and 2007 results A comparison between the 2006 and 2008 Results Indicators 2010 Targets* Indicator 1. Operational 8% 13% 75% development strategies Indicator 2. Reliable public 0% 50% of countries management systems 33% improve CPIA scores 42% Indicator 3. Aid flows 45% are recorded in 62% 85% country budget 66% Indicator 4. Technical 48% 59% 50% assistance is aligned and 57% coordinated 85% 40% Indicator 5a. Donors use 47% 50%** country PFM systems 42% 62% 39% Indicator 5b. Donors use 50%** 44% country procurement 40% systems 52% 100% Reduce to one-third, Indicator 6. Donors avoid 88% the stock of parallel parallel PIUs 100% PIUs *** 35% 41% Indicator 7. Aid is 42% 71% more predictable 63% 65% 75% Indicator 8. Aid is untied 90% 100% Progress over time 100% 43% Indicator 9. Donors use 44% coordinated mechanisms 66% 57% for aid delivery 54% 18% Indicator 10a. Donors 20% 40% coordinate their missions 21% 31% Indicator 10b. Donors 42% 42% 66% coordinate their country 49% studies 59% Indicator 11. Sound results 3% 35% framew orks 5% Indicator 12. Mechanisms 22% for mutual accountability 26% 100% 0% 20% 40% 60% 80% 100% Percent 2005 2006 All Survey Participants 2007 2008 2005 2006 World Bank 2007 2008 * These targets were set for the development community as a whole. **The targets on the use of country financial management and procurement systems were revised to 50% in Accra. The original Paris Declaration target for the two indicators was 80%. *** The actual target is "Reduce by two-thirds, the stock of parallel PIUs." Source: 2008 Survey on Monitoring the Paris Declaration: Making Aid More Effective by 2010, OECD, 2008; Results-Based National Development Strategies: Assessment and Challenges Ahead (SecM2008-0009), January 10, 2008; and Accra Agenda for Action, September 2008. CPIA scores are available at http://go.worldbank.org/S2THWI1X60. 42 Annex B. IDA15 Replenishment Commitments -Monitorable Actions Objectives Recommendations/actions Product/target date Comment SPECIAL THEME II: ENHANCING COUNTRY-LEVEL EFFECTIVENESS Harmonization · Issue guidelines on PIUs requiring that their integration into government · Guidelines on PIUs/FY09 Guideline completed, and Alignment structures become the default option for IDA projects. 2005--implementation ongoing · Review IDA's performance with respect to predictability of disbursements, as · A Report/IDA15 Mid-Term Completed well as the constraints to medium-term predictability of IDA disbursements, at Review the country level. · Carry out--jointly with other interested donors--a survey of selected country · Report of the survey/Accra High Completed, 2008 programs and the total amounts of cofinancing and of pooled and parallel Level Forum, 2008 financing they leverage through IDA's budgetary support and investment lending. · Reporting by Regional Management to the Board periodically, on IDA's · Regional Alignment and Ongoing actions at the country level to assist in the preparation and implementation of Harmonization reports/Ongoing country-led strategies or plans for harmonization and alignment. · Update good practice note on joint/collaborative CAS preparation. · Good Practice Note/2008 In progress · Update good practice note on PSIAs. · Good Practice Note/2008 Completed, 2008 · Further review application of conditionality. · Development Policy Operations Completed, FY10 Retrospective/FY09 · Work with other lenders/donors on common (harmonized) legal requirements · Conferences, analytic studies, Ongoing for MOUs in joint financing operations. Memoranda of Understanding/Ongoing · Update progress on the Matrix of Actions for Harmonization and Alignment. · A Report/IDA Mid-Term Completed Review · Facilitate and support country efforts to incorporate nontraditional partners ­ · Conferences, Memoranda of 2008 and ongoing vertical funds, non-DAC donors, and the private sector ­ in harmonization and Understanding, Analytical alignment actions and undertake further analytic work on the changing aid studies/2008 architecture. · Enhance internal staff incentives and guidance on ownership, harmonization, · Good Practice Note/ FY09 Ongoing and alignment. · Continue to implement and explore further options for decentralization. A Report/IDA15 Mid-Term Ongoing · Complete an analysis of different models of decentralization by types of Review Ongoing clients and services, and develop an approach to better define cost effectiveness of decentralization in light of budget implications, with special attention to fragile states. · The Africa Region is working toward all new and vacant internationally- Completed recruited positions to be based in the field, thereby increasing the number of internationally recruited staff in the field by over 50 percent by end-FY08 43 Objectives Recommendations/actions Product/target date Comment compared to FY06. · The Africa Region intends to continue to pursue this field-based recruitment Completed effort during the IDA15 period so that the number and percentage of internationally recruited staff in the field would increase. · The Region plans to significantly increase the percentage of projects and Completed programs managed in the field by continuing to move toward a model where a majority of tasks are managed in the field with task managers located in one country and working on two to three. · Develop a draft World Bank Action Plan for aid effectiveness in preparation Action Plan/September 2008 Completed for the High Level Forum on Aid Effectiveness to be held in Accra. 44 Annex C. Predictability and Aid-on-Budget Measurements 1. The Paris Declaration and the AAA include separate but interrelated commitments and indicators with respect to aid predictability and aid on budget. 2. Aid Predictability. The Paris Declaration commits donors to provide reliable indicative commitments of aid over a multiyear framework and to disburse aid in a timely and predictable fashion according to agreed schedules. The AAA expands the donor commitment to emphasize the provision of full and timely information on annual commitments and actual disbursements and regular and timely information on rolling three- to five-year expenditure and/or implementation plans, with at least indicative resource allocations. It commits developing countries to strengthen budget planning processes for managing domestic and external resources. The rationale behind these predictability commitments is that governments can make the best use of development assistance if they are in a position to plan and therefore optimize external and domestic resource allocation within and across sectors. 3. Aid on Budget. The Paris Declaration does not include a separate partnership commitment for governments to inscribe more aid in their annual budgets, but it does include commitments relevant to increasing the amount of aid on budget. It commits donors to provide timely, transparent, and comprehensive information on aid flows, and to rely to the maximum extent possible on transparent partner government budget and accounting systems; and it commits governments to publish timely, transparent, and reliable reports on budget execution and to strengthen as appropriate the parliament's role in national development strategies and/or budgets. The rationale behind these commitments is to ensure that by 2010, aid forecasts are recorded in partner countries' annual budgets so that country authorities can present accurate and comprehensive budget reports to their legislatures and citizens. A. March 2005 Paris Declaration Monitoring Framework 4. Aid Predictability. The Paris Declaration monitoring framework included an indicator to measure the percentage of aid disbursements released according to agreed schedules in annual or multiyear frameworks. Assessing predictability using this measure would essentially require donors and governments to agree and report together on disbursement forecasts, and would require donors to disburse in accordance with those agreements. It would require information on donor records of disbursements. The Paris Declaration set as a tentative target that 75 percent of aid be released on schedule by 2010, with the target subject to revision. 5. Aid on Budget. The Paris Declaration monitoring framework included an indicator to monitor aid on budget, linking it to a Paris Declaration partnership commitment that donors will base their support on national development strategies. The rationale for linking aid on budget to this donor commitment was to recognize the importance of transparency not only to domestic accountability, but also to improving the amount of aid flows that support a country's national development strategy rather than donor-driven priorities outside of the strategy. The aid-on- budget indicator sought to measure the percentage of aid flows to the government sector reported on partners' national budgets, setting as a tentative target that 85 percent of aid flows be reported on budget by 2010. 45 B. 2006 and 2008 Surveys on Monitoring the Paris Declaration 6. Aid Predictability. To underscore that meeting the Paris Declaration objective is not exclusively within the control of donors but is a responsibility shared between government and donors, the indicator was revised to measure "the gap between aid Box C1. Paris Declaration Monitoring Surveys: Aid scheduled and aid effectively disbursed Predictability Criteria and Formula and recorded in countries' accounting Mixed donor-government perspective: systems." Thus the Paris Declaration surveys collected data from donors on A: ODA scheduled for disbursement (donor) the amount of aid scheduled for Funds scheduled by donors for disbursement in "year x" and disbursement and compared this with notified to government within "year x-1." data collected from governments on the B: ODA actually received (government) amount of aid recorded in government Funds actually received in "year x" as recorded in government systems. Both the 2006 and the 2008 accounting/reporting systems. surveys used the same aid predictability criteria and formula46 Aid predictability = B/A when BA (see Box C1). Neither survey requested data on disbursement forecasts agreed between donors and the government. The surveys did, however, include a request that the donor disbursement forecasts reported be those that had been given to the government. The 2006 survey introduced a target to reduce by half the proportion of aid not disbursed within the year that it was scheduled. Using the survey methodology and results, this translates to reducing the percentage of aid not recorded as disbursed by governments within the year that it was scheduled by donors to 71 percent by 2010 from 41 percent and 42 percent from the 2006 and 2008 surveys. Thus, 71 percent has become the Paris Declaration target. 7. Aid on Budget. The survey also revised the aid on budget indicator--instead of measuring the percentage of aid flows reported in the annual budget, the surveys would measure "the gap between what was disbursed by donors" and "aid to be disbursed recorded in the annual budget by government." This measure mixes ex-ante data with ex-post data as well as government data and donor data. The survey guidelines suggested making progress in this area (given the indicator as it is formulated) would require donors to (a) provide accurate "ex-post" data on donor aid actually disbursed, and (b) share their disbursement forecasts with budget authorities, and would require governments to inscribe these forecasts in their budgets (which would require government and donors to work together to develop and use identical disbursement forecasts.) Mutually agreeing disbursement forecasts in this way would eliminate the problem that the measure mixes country and donor perspectives and data; however, it would not eliminate the fact that ex-ante and ex-post data are mixed. 46 The 2006 survey had used a weighted average to aggregate the results. The 2008 survey used an unweighted average, an aggregate that better captures impact at the country level and restated the 2006 results. A formula that flips the ratio depending on which variable is larger is used to prevent over- and under-predictions from canceling each other out. 46 8. The Paris Declaration surveys have thus collected data from donors on the amount of aid disbursed and compared this with data collected from governments on the amount of estimated aid inscribed in the budget at the beginning Box C2. 2006 and 2008 Surveys on Monitoring the Paris of the year. Both the 2006 and the 2008 Declaration: Aid on Budget Criteria surveys use the same aid on budget criteria and formula47 (see Box C.2). The 2006 A: Actual ODA disbursed for the government sector at the country level (donor) survey introduced a target to halve by 2010 the proportion of aid disbursements greater Funds disbursed by donors for the "year x"(ex-post) than what the government recorded in the B: Estimated ODA recorded in budget (government) annual budget, with at least 85 percent ODA recorded in the annual budget at the beginning of reported on budget. The 2006 survey "year x" (ex-ante) indicated that the level of actual Aid on budget = A/B when AB disbursements was 42 percent of what governments had inscribed in their budgets at the beginning of the year and the 2008 survey 45 percent. C. Measurement Issues 9. Aid Predictability. Donor and recipient expectations are often different. Because the costs to governments of over-budgeting external aid can be high when intrayear budget adjustments have to be made, governments often view donor projections as optimistic and apply a discount rate to donor forecasts. However, the survey measure of aid predictability implicitly assumes that governments accept donor forecasts as their own. In comparing donor expectations with government records of disbursements, the survey aid predictability measure essentially compares apples to oranges. 10. The survey measure is weak from both a government and a donor point of view. A low percentage of projected aid recorded as received could mean that donors made poor projections and did not disburse as promised, or that governments made poor projections or recorded disbursements differently than donors because of coverage or timing issues that affect data (these issues are explained at the end of this annex). Donors' work to reduce the gap between their forecasts and actual disbursements would not be reflected in an improved survey score if government records are not reconciled with donor records. Governments' work to increase the percentage of disbursements recorded through country systems would not be reflected in an improved survey score if donor forecasts are poor or differ from their own. 11. Clarity on the perspective from which predictability is viewed would eliminate the measurement issues and some (but not all) data issues and would help clarify the operational implications to both governments and donors of a low predictability score. Unless predictability can be measured from a mutual perspective--that is, it relies on data agreed between both donors and governments as the Paris Declaration originally intended--it should be measured from either a government or donor perspective. A donor perspective would measure predictability relative to 47 Just as with Aid Predictability, the 2006 survey had used a weighted average to aggregate the results for Aid on Budget. The 2008 survey used an unweighted average, an aggregate that better captures impact at the country level and restated the 2006 results A formula which flips the ratio depending on which variable is larger is used to avoid over and under predictions from canceling each other out. 47 donor expectations and donor records of disbursements; a government perspective would measure predictability relative to recipient expectations and recipient records of disbursements. 12. Aid on Budget. There are two problems with the current measure of aid on budget. First, as with aid predictability, the measure mixes donor and government perspectives, thus implicitly assuming them to be the same when they are not. In the absence of an agreed disbursement forecast, the surveys' aid-on-budget indicator would suffer from the same measurement and data issues as the aid predictability indicator. Second, it mixes ex-ante data (estimated aid recorded) with ex-post data (aid actually disbursed). A more accurate way of assessing a government's capacity to put aid on budget would be to use either ex-post actual disbursements received by governments with ex-post disbursements actually recorded by the government or use ex-ante disbursement forecasts developed by governments with ex-ante disbursement estimates recorded by the government. D. Alternative Measures of Predictability 13. Ultimately it is the government's responsibility to project both domestic and external revenue sources and document those expectations in the annual budget. Aid predictability from a government perspective, therefore, can be defined as the percent difference Box C3. Aid Predictability Best Alternative Criteria and Formula between the level of aid the government expects to receive and that Government perspective: actually received (see Box C3). Scores A. Disbursements recorded in annual budget (government) on this measure would have clear Funds inscribed in the annual budget ex-ante. implications for improving budgetary management. Looking at aid B. Disbursements actually received (government) Funds recorded in the budget ex-post. predictability from a government perspective, therefore, can help Aid predictability = A/B when AB governments and donors identify ways to improve the capacity of governments to make reliable budget forecasts of aid and account for aid expenditures. 14. Using a measure of predictability from the government perspective has important country-level repercussions for aid on budget. A high score on predictability as measured from a government perspective would likely encourage governments to inscribe more aid on annual budgets at the beginning of the year because, as governments gain confidence in their budget forecasts, they would expect to incur fewer costs from midyear budget readjustments due to inaccurate forecasts. 15. Data collected through the surveys as part of both the "aid-on-budget" indicator and the "aid predictability" indicator enable estimates of predictability from both a government and a donor perspective. Of the 54 countries that participated in the 2008 survey, 51 provided data on both Bank aid budgeted and Bank aid received and recorded in country systems; and 48 provided data on Bank forecasts and Bank actual disbursements. 16. Aid Predictability: Government Perspective. When the survey data are used to measure predictability from a government perspective, they indicate that aid predictability for all donors that participated in the survey and for the Bank are similar: recipient countries recorded receipt 48 of an average of 63 percent of the amount of aid budgeted at the beginning of the year, and an average of 64 percent of Bank aid reported as budgeted was recorded through country systems as received. This falls short of the 71 percent target defined in the 2008 survey. However, in more than half--29 of the 51 countries for which survey data is available--75 percent of the Bank's budgeted aid was recorded as received (see Box C4). Box C4. "Country Perspective" Measure: Countries that Reported as Disbursed at least 71 Percent of Aid Forecasted (2008) Afghanistan Cameroon Kyrgyz Republic Moldova Niger Albania Cape Verde Liberia Mongolia Sierra Leone Bangladesh Colombia Madagascar Morocco Uganda Burkina Faso Dominican Republic Malawi PNG Ukraine Burundi Ethiopia Mali Philippines Vietnam Cambodia Ghana Mauritania Nepal 17. The predictability of Bank disbursements, measured from a government perspective with the survey data, was better in the 32 countries where the Bank's portfolio included development policy loans.48 In these countries, predictability from a government perspective was higher than for the average of all countries: these survey participants recorded through their systems 71 percent of budgeted amounts of Bank assistance. Conversely, the Bank's aid predictability performance was not as strong in countries where the Bank's assistance included only project and trust fund disbursements: in these countries, survey participants recorded through their systems just 53 percent of budgeted amounts of Bank assistance. 18. The operational implications of these scores are clear. Donors and governments need to work together to improve budgeted forecasts of aid both to improve aid predictability and increase the amount of aid governments put on budget. For the Bank, supporting governments to improve budget forecasts of its disbursements will require a special focus on improving disbursement forecasts for projects and trust funds and communicating these to country budget officials during budget preparation. Similarly, donors must step up their support to strengthen and use government financial reporting systems. 19. Aid Predictability: Donor Perspective. Aid predictability from a donor perspective can be defined as the percent difference between donor records of aid disbursed and donor Box C5. Aid Predictability Alternative Criteria and expectations of disbursements (see Box C5). Formula The advantage of measuring predictability Donor perspective: from a donor perspective is that ascertaining A. Projected disbursements (donor) whether donors are making realistic forecasts can help donors with their cash Disbursement projections made in year x-1 for year x. management. The disadvantage of this B. Actual disbursements (donor) measure is that improved donor forecasts Disbursements recorded in the donor's systems. have no direct operational implication at the Aid predictability = A/B when AB country level. Donors may make good 48 The survey actually included 33 countries where Bank assistance included development policy loans, but one of these countries did not provide information on both aid budgeted and aid received. 49 forecasts, but if these forecasts are not adopted by governments and inscribed in the budget, donor realism will not have a positive effect on improving the realism of government forecasts, increasing aid on budget, or increasing the use of country systems to record aid disbursements. 20. When the survey data are used to measure predictability from a donor perspective, the Bank disbursed 73 percent of forecasted amounts. This exceeds the 2010 target that 71 percent of aid disbursements forecasted be disbursed. Also, in about two-thirds of the countries for which survey data is available--33--the Bank's assistance met the 71 percent target (see Box C6). Box C6. "Donor perspective" Measure: Countries that Reported as Disbursed at least 71 Percent of Aid Forecasted (2008) Afghanistan DR Congo Jordan Mauritania Niger Ukraine Bangladesh Egypt Kyrgyz Republic Moldova PNG Vietnam Benin Ethiopia Lao PDR Mongolia Senegal Yemen Burundi Ghana Liberia Morocco Sierra Leone Cambodia Haiti Madagascar Mozambique Tanzania Cameroon Indonesia Malawi Nepal Uganda E. Data Issues 21. The survey data have three shortcomings. The first--lack of comparability--could be addressed by using a new measure of aid predictability. The other two--inaccurate reporting and donor attribution--could be addressed through improved survey design and execution. 22. Lack of Comparability. Data collected from governments and data collected from donors often cover different aid modalities as well as different time periods. The aggregate data hide differences in coverage that cannot be identified and eliminated since detailed data were not collected. These differences are due to different reporting of project assistance and of trust funds. Given the large percentage of project and trust fund support in the Bank's portfolio, this has a significant impact on Bank data. For example, governments were requested to report disbursements recorded in country accounting systems; in some countries this would not include project support accounted for outside of country systems. Donors, however, were requested to include all modalities of expected aid that had been communicated to the government, regardless of whether the government was expected to use country systems to account for this aid. Donor projections, therefore, would have included project support systematically. The aggregate data hide timing issues that can arise because of the lag between when a donor disburses funds and when the recipient records the receipt of these funds. Although typically short, this lag can be relevant at the end of fiscal years, particularly when the fiscal years are not the same for donors and recipients. This is particularly relevant for the Bank, since its fiscal year is different from that of most survey participants. In addition, the data do not clarify the point at which funds should be reported as received from donors. Typically donors consider funds disbursed when they are released into a special account, whereas the government systems may not recognize funds as received until they are disbursed from this special account. Disbursements could have been forecast for one 50 year and released on schedule, but the recipient may have recorded them as received in the following year. In addition, disbursements as recorded by the donor are often in a different currency than the government uses, resulting in some differences in measurement due to exchange rates 23. Inaccurate Reporting. In more than one-third of countries, there were obvious errors in what the survey reports as aid received. For example, in two countries the survey indicates that the government reported that it recorded no Bank disbursements even when it had reported budgeting a figure similar to the Bank's own forecasts. In three countries, the survey indicates the opposite: no budgeted amounts even when Bank disbursements are reported. In addition, in 14 countries budgeted amounts of Bank disbursements are reported as identical or unrealistically close to recorded amounts, including 4 countries that reported no aid budgeted or received. 24. Trust Fund Issues. There have also been significant problems with inaccuracies in attributing predictability (or lack thereof) to one donor or another as well as in attributing disbursements when one donor administers a large amount of trust funds on behalf of other donors, as the Bank often does. In countries where one donor disburses aid on behalf of another, the survey guideline indicated that the donor that made the final disbursement to the government was to report these funds through the survey. In these cases, the predictability of trust-funded disbursements would be as dependent on the predictability of disbursements from the donor funding the trust fund as it is on the donor disbursing the trust fund. For example, governments may have expected to receive certain disbursements from the Bank at the beginning of the year and budgeted for these amounts. However, if another donor made a midyear decision to disburse through Bank-administered trust funds, these amounts would not have been budgeted by the government or attributed to the Bank. However, once they were disbursed the government accounting systems may have recognized the Bank as the source of the disbursement, and the amounts of budgeted Bank aid recorded as received would then exceed the amounts initially budgeted in these countries, even if all other Bank disbursements were recorded as expected. In some countries, the donor that disbursed aid to another donor reported these disbursements through the survey, and the survey inadvertently attributed these as disbursements directly to a country rather than through another donor. Reporting disbursements in this way contributed to misleading predictability results since the government would likely have attributed aid to the donor that disbursed the funds and recorded the amount of the actual disbursement it received rather than what was disbursed into the trust fund. 51 Annex D. Challenges with MOUs 1. A review of MOUs undertaken by the Bank has found that MOUs are often expected to serve as a platform to harmonize issues that represent fundamental institutional differences-- issues related to policies and mandates under the purview of senior management or higher-level governance structures within each organization--especially those related to budget support objectives and disbursement, institutional mandates, and fiduciary issues in pooled funding. Moreover, the legal status of an MOU--the extent to which they are binding--and their terminology is often unclear 1. Budget Support and Disbursement 2. MOUs covering collaboration under a general budget support program require agreement on a common policy framework--but this can be problematic, since donors' budget support objectives and disbursement requirements vary. The Bank's policy is to disburse directly into a country's budget on the basis of prior policy or institutional reform actions; for other donors, such as the European Commission, the policy is to disburse against outcomes/targets. To illustrate: the Bank may agree to support government actions under an educational policy reform; other donors may focus on increased school enrollment rates; and still others may agree with the government on investment-type conditions, such as the number of schools or hospitals constructed (for the Bank, physical infrastructure activities do not fall within the ambit of a DPO). An MOU cannot be used to resolve differences in donors' basic policies; instead, they should accommodate donors' different approaches toward a common country goal. 3. Problems in agreeing on MOUs can also arise when the Bank is supporting sector policy through a DPO and other donors are supporting the same sector through earmarked sectoral budget support. The Bank's DPOs disburse directly into the budget and funds cannot be earmarked to a specific sector; the prior actions/triggers are sector-focused, but the funds go directly into the general budget. Some donors, however, wish to earmark and channel their sectoral budget support for the financing of specific sector work plans and budgets. It is difficult to reconcile these different approaches through an MOU. 2. Institutional Mandates 4. MOUs face other challenges related to the fundamentally different mandates of the signatories. For example, the mandate of bilateral donors may require them to include political governance conditionality (such as conducting free and fair elections and upholding democratic principles) with their financing, while the Bank and other multilateral agencies are prohibited by their Articles of Agreement from including political considerations in their decision making or from interfering in the political affairs of their member countries. It is difficult to reconcile these very different mandates. In many cases the solution has been to include language explicitly recognizing that each donor will be bound by its own legislation or charter, but these solutions often come at the cost of time and good will. 3. Pooled Financing 5. Joint financing under SWAps sometimes carries with it expectations of pooled financing: many donors are eager to have all development partners' pool funding in conjunction 52 with joint operations. However, Bank financing of SWAps with pooled funding presents challenges in two areas related to the Bank's fiduciary and safeguard policies. When funding is pooled, the Bank's fiduciary and safeguard policies must apply to all funds in the pool since the Bank's funds cannot be disaggregated from the other funds. If the country systems meet the standards of the Bank's policies, the Bank and all donors can use country systems. But if the Bank's assessment indicates the need for additional measures to bring these systems up to the Bank's standards, yet some donors are willing to use country systems as they are, conflicts often arise. In this case, if either the government or any donor insists that the same procedures apply to all signatories of an MOU, without an exception clause or without dividing donors into groups such as pooled and nonpooled groups, reaching agreement can cause delays and tensions. Procurement. Among fiduciary issues, procurement is often the most problematic when funds are pooled. As a matter of policy approved by the Board, all Bank Loan Agreements with borrowers for investment operations and Trust Fund Agreements referencing goods and services incorporate the Bank's Procurement and Consultant Guidelines. However, bilateral donors in pooled funding operations frequently object to any references in MOUs to the Bank's guidelines, even though their purpose to is to minimize fiduciary risks for procurement.49 Even when the Bank uses national procurement systems, including thresholds for prior review still hold, many bilateral donors do not require prior review. Therefore, agreeing within an MOU on the appropriate language to take into account the differing policy frameworks on procurement has been problematic. Safeguards. MOUs do not always address safeguard issues. Some donors do not have environmental and social safeguard policies and requirements equivalent to those of the Bank's, and treatment of these issues is sometimes left out of MOUs even in the context of pool-funded operations that trigger the Bank's safeguard policies. As a result, these issues can be underestimated in the overall coordination and harmonization dialogue, raising the potential for difficulties during project supervision.50 6. The process of agreeing on MOUs is less contentious and more efficient if the documents avoid complexity and detailed prescriptions, using flexible frameworks rather than rigid prescriptions. In addition, built-in flexibility facilitates donor responses in times of crisis. Of course, some MOUs have been drafted with pragmatic solutions to build in flexibility; but this is not systematic practice, and it often comes at a cost in time and harmonious relationships. In these cases, donors have agreed on the ultimate objective of the operations--supporting government priorities through a coordinated program--and have accepted the fact that differing institutional policies do not allow complete harmonization of donor procedures. For sector program support that finances investments, one practical solution to addressing institutional 49 Many donors consider Bank guidelines to be incompatible with the use of country systems. In reality, these guidelines include provisions for using country systems (e.g., the provisions permitting the use of national competitive bidding procedures). 50 The Multilateral Financial Institutions-Working Group on Environment and the OECD-DAC Network on Environment and Development Cooperation are important venues for collaboration in this area. See First Year Review of Implementation of Scaled-Up Program to Pilot Use of Country Systems to Address Environmental and Social Safeguard Issues in Bank-Supported Projects (SecM2009-0367) July 14, 2009. 53 differences has been to divide the donors into groups--for example, a pooled group and a non- pooled group. In other cases, MOUs have used flexible language accommodating donors' differences (by using such phrases as "to the extent possible" or "to the extent permitted by individual donors' statutory frameworks"), or have included special "exception" clauses, excluding one or more donors from a particular clause or approach (see Box D1). Box D1. MOUs with Flexible Frameworks Examples of MOUs covering both pooled and non-pooled funds: Bangladesh within the HNP sector program (2003-10) has differing arrangements for each type of funding. Cambodia's second Health Sector program (2008-15) distinguishes between pooled and non-pooled funds. Ghana's Medium-Term PSD Strategy and Action Plan (2005-9) provides an option for donors to accept all Sections of the MOU (including Section 3, which relates to pooling of funds) or to accept only those sections that set up general coordination principles, irrespective of the financing modality. Examples of MOUs that containing flexible provisions or exceptions: Sierra Leone (2006) partnership framework supporting poverty reduction under joint budget support recognizes the different mandates of the development partners. Tanzania (2006) partnership framework supporting poverty reduction under joint budget support acknowledges that donors may have specific legal and statutory requirements and provides for independent audits by partners. Ethiopia (2004) pooled fund support for the public sector capacity building program contains a specific exception for the Bank, whereby its contribution cannot be used for law enforcement activities. 4. Legal Status of MOUs 7. Much about MOUs is unclear: their legal status and extent to which they are binding, and their terminology. While the Bank, like most other donors, recognizes the need to honor the understandings reached in MOUs, it does not treat them as binding legal agreements. However, some donors take other approaches. As a consequence, the legal status and binding nature of an MOU is often left ambiguous. In addition, although similar and sometimes identical terms are used (for instance, the term sector budget support), they have different meanings across donors and have caused confusion when left undefined. Hence, terms must be defined in a way that can be understood by all.