54956 IDA16 IDA's Performance Based Allocation System: Review of the Current System and Key Issues for IDA16 International Development Association IDA Resource Mobilization Department (CFPIR) May 2010 Abbreviations and Acronyms AfDB African Development Bank AfDF African Development Fund ADB Asian Development Bank CPIA Country Policy and Institutional Assessment EVI Economic Vulnerability Index FSF Fragile States Facility GNI Gross National Investment HAI Human Assets Index IBRD International Bank for Reconstruction and Development IDA International Development Association IEG Independent Evaluation Group MDB Multilateral Development Bank MDGs Millennium Development Goals MDRI Multilateral Debt Relief Initiative MTR Mid-Term Review PBA Performance-Based Allocation PCPI Post Conflict Performance Indicators PPR Portfolio Performance Rating WDR World Development Report Table of Contents EXECUTIVE SUMMARY ................................................................................................ I I. INTRODUCTION ........................................................................................................ 1 II. A COMPREHENSIVE REVIEW OF THE CURRENT PERFORMANCE-BASED ALLOCATION SYSTEM ................................................................................................ 1 A. The Core PBA System............................................................................................................2 B. Exceptions to the PBA System ...............................................................................................4 C. Implementation and Allocation Outcomes for IDA15 ...........................................................6 III. THE BALANCE BETWEEN PERFORMANCE AND NEEDS ................................... 9 A. Accounting for Current Measures of Needs ..........................................................................9 B. Including Vulnerability in the PBA Formula: Implications and Key Conclusions .............11 IV. SPECIAL FINANCING CONSIDERATIONS FOR IDA16 ...................................... 12 A. Strengthening Financing for Fragile and Conflict-Affected Countries ................................13 B. Strengthening IDA's Financial Assistance to Small States ..................................................22 V. KEY ISSUES FOR DISCUSSION.............................................................................. 26 REFERENCES .............................................................................................................. 27 Annexes Annex 1. Evolution of the PBA formula ...................................................................................... 29 Annex 2. The PBA Formula during IDA15 and Detailed Allocation Steps ................................. 31 Annex 3. CPIA Criteria and Portfolio Performance Rating Method ............................................ 33 Annex 4. Mechanics of IDA Post-Conflict Allocation ................................................................. 34 Annex 5. Estimated IDA15 Country Allocation Results by Allocation Steps ............................. 35 Annex 6. IDA Resource Management Within a Replenishment Period ....................................... 37 Annex 7. Simulation Results of Including Vulnerability in the PBA Formula ............................ 38 Annex 8. How African Development (AfDB) and Asian Development Bank (ADB) Determine Post Conflict Allocation and Phasing Out .................................................................................... 44 Tables Table 1. Current Schedule of Exit from the Exceptional Allocation Window ............................. 14 Table 2. Projected IDA16 Allocation Impact of a Longer Phase-out (Option 1) ......................... 17 Table 3. Application of the Phase Out Extension Criteria to Current Post Conflict and Re- engagement Countries ................................................................................................................... 19 Table 4. Financial Implications of Proposed Measures for Fragile and Conflict-Affected Countries (IDA16) ........................................................................................................................ 21 Table 5. How Much Infrastructure Does the Current Base Allocation "Buy"? ............................ 23 Table 6. Allocation Impact of Eliminating the Maximum Per Capita Ceiling and Raising the Base Allocation from SDR1.5 Million to SDR3 Million Per Year .............................................. 25 Figures Figure 1. A Schematic Illustration of the Current PBA Formula ................................................... 3 Figure 2. Steps in Allocating IDA Resources to Eligible Countries............................................... 6 Figure 3. Estimated IDA15 Allocation Results by Country Groupings1 ........................................ 7 Figure 4. The Performance Orientation of the PBA ....................................................................... 8 Figure 5. Impact of Capping and Exceptional Allocations on Redistribution of Resources Across Regions (FY09) ............................................................................................................................. 10 Figure 6. Estimated IDA15 Allocation Results by Region1 ......................................................... 10 Figure 7. Estimated Resource Allocation to Fragile and Conflict-Affected Countries in IDA15114 EXECUTIVE SUMMARY i. As part of the IDA16 replenishment discussions, IDA Deputies have requested Management to provide a comprehensive review of IDA's allocation system. In addition, several Deputies requested that the review also consider how the allocation system can be modified to respond to the needs of fragile and conflict-affected states and small states. This paper responds to these requests. ii. The current Performance Based Allocation system. The main factor determining a country's IDA allocation is its performance, as measured by the Country Policy and Institutional Assessment (CPIA) and Portfolio Performance Rating (PPR). While country performance is the main determinant, country needs are also taken into account through population size and GNI per capita. In addition, each IDA country receives a fixed minimum allocation, referred to as the "base allocation", which is set at SDR1.5 million during IDA15. While the bulk of IDA resources are allocated through the PBA system, there are five exceptions--capped allocations to blend countries with large populations, special allocation for post conflict and for re-engaging countries, funding for arrears clearance, special provision for regional projects, and exceptional natural disaster-related allocation. iii. Overall, the PBA system has continued to function well during the IDA15 period. The bulk of IDA resources­ 84 percent ­has been allocated either through the core PBA system or through capped allocations to blend countries with large populations. The rest has gone to exceptional allocations: 7.4 percent to post-conflict and re-engaging countries; 4.3 percent to regional projects; 3.7 percent to the Pilot Crisis Response Window (funded mainly through redeployment of arrears clearance funds and additional donor contributions); and 0.5 percent to natural disaster-related exceptional allocations. iv. The system has continued to direct more resources to countries that are performing better. During IDA15, IDA countries in the top performance quintile received about 2.7 times in allocations per capita than those in the lowest quintile. This is higher than the comparable ratio of 2.3 during IDA14, confirming that the performance orientation of the system has indeed slightly improved. Furthermore, countries with a higher per capita IDA allocation have on average experienced greater improvements in their development outcomes. At the same time, the system has continued to balance performance with needs by allocating, consistent with performance, about 50 percent of IDA15 resources to Sub-Saharan Africa, the region with the greatest needs for IDA resources. The capping of India and Pakistan more than doubles sub-Saharan Africa's share from 20 percent to 49 percent, while special allocations for the post conflict and reengagement countries further increase this share to 50 percent. v. Strengthening financing for fragile and conflict-affected countries. IDA's capacity to support fragile and conflict affected countries has been enhanced in recent replenishments. Beginning with IDA13, countries emerging from conflict or disengagement have received exceptional IDA allocations that support their special recovery and reconstruction needs. During IDA14 and 15, further enhancements were made to the post conflict and re-engagement allocations, including extending the duration of the special allocation; linking the share of post-conflict allocations to changes in the overall replenishment size; introducing a systematic approach to arrears clearance; and strengthening the existing review process for, and moving towards public - ii - disclosure, of the Post Conflict Performance Indicators (PCPI) to increase transparency around the special allocations. vi. This paper proposes three measures to strengthen support for fragile and conflict-affected countries during IDA16. First, Sudan and Zimbabwe are likely to re-engage with IDA during the IDA16 period, and supporting their re-engagement, including through arrears clearance operations, would be important. Second, it is proposed to extend the phase-out period either uniformly for post conflict countries (by two years) and re-engaging countries (by one year) or adopt a case-by-case approach where the phase-out period is extended for individual countries against a specified set of criteria. Either option will strengthen financing for post-conflict and re-engaging countries during IDA16. Finally, it is proposed that the current three-country requirement for regional projects be relaxed to allow two countries, of which at least one is a fragile or conflict-affected country, to be eligible for financing for regional IDA projects. vii. Strengthening financing for Small States. Small states, particularly those with populations less than 1.5 million, face a number of challenges owing to their higher vulnerability to economic shocks and natural disasters. In view of these vulnerabilities, IDA has strengthened its financial assistance to these countries in recent replenishments, but many are still looking towards IDA for enhanced financial support. This paper proposes two measures to strengthen IDA support for small states during the IDA16 period. First, it is proposed to eliminate the maximum per capita allocation ceiling, currently set at SDR19.8, which has constrained the allocations of several small states. Second, given the importance of the base allocation for small states, it is proposed that this be raised from the current SDR1.5 million per year to SDR3 million per year. Both steps would have little impact on the performance orientation of IDA's PBA system, but high benefits for several small states. viii. Deputies' views and guidance are sought on the following key questions. With respect to financing for fragile and conflict-affected countries: · Do Deputies agree to extend the phase-out period for post conflict countries uniformly for all countries in the window (Option 1) or on a case-by-case basis against a specified set of criteria (Option 2)? · Do Deputies agree to modify the current three-country requirement for regional projects to allow two countries, of which at least one is a fragile or conflict affected country, to be eligible for IDA regional financing? ix. With respect to strengthening support for small states: · Do Deputies agree to eliminate the maximum per capita allocation ceiling in order to better support small IDA states? · Do Deputies agree with raising the base allocations to all countries from the current SDR1.5 million per year (or SDR4.5 million per replenishment) to SDR3 million per year (or SDR9 million per replenishment) for the benefit of small states? I. INTRODUCTION 1. IDA's Performance Based Allocation (PBA) system has formed the basis for allocating resources among IDA countries since the late 1970s. Over the years, the PBA has undergone significant changes, reflecting both new analytical insights and implementation experience. Through successive IDA replenishment negotiations, IDA donors have introduced several changes and adjustments to the formula underlying the PBA system, which have helped increase its performance orientation and improve its transparency. 2. Several changes were made to the PBA system during the IDA15 replenishment.1 First, the functional form of the country performance rating was simplified and the methodology for calculating the portfolio performance rating was modified to lower volatility in IDA allocations. Second, the base allocation was increased to SDR1.5 million per year, a change that benefits small states. Third, to encourage countries with small allocations to participate in regional projects, country contributions to regional projects were capped at 20 percent of a country's annual IDA allocation. Fourth, IDA's exceptional allocations were extended from 7 to 10 years for post conflict countries, and from 3 to 5 years for re-engaging countries to support their special development needs. 3. At the IDA15 Mid-Term Review (MTR) and at the first IDA16 replenishment meeting, IDA Deputies requested Management to provide a comprehensive review of IDA's allocation system.2 Several Deputies also requested that the review consider how the system can be modified to respond to the needs of fragile states and small states, and examine the emphasis placed on needs and vulnerability in the allocation formula. Some Deputies also suggested that, in light of the positive operational and financial impacts of the lengthened phase-out period discussed at the MTR, consideration be given to lengthening this period even further in IDA16. 4. This paper addresses these requests. Section II describes the current allocation system, including the core PBA formula and its main components and the exceptional allocations made outside the PBA system. Section III examines the current financing mechanisms for post conflict and re-engaging countries and presents recommendations to strengthen the financing mechanism for fragile and conflict affected countries during the period of IDA16. It also presents proposals that would allow IDA to enhance its support to small states. Section IV lays out the issues for discussion. II. A COMPREHENSIVE REVIEW OF THE CURRENT PERFORMANCE-BASED ALLOCATION SYSTEM 5. Eligibility for IDA financing depends on two criteria: (i) relative poverty defined as GNI per capita below an established threshold;3 and (ii) lack of creditworthiness to borrow on IBRD and market terms and therefore a need for concessional resources to finance the country's development program. As of FY10, 79 countries are eligible for IDA financing. 1 Refer to IDA (2008), "IDA: The Platform for Achieving Results at the Country Level." Additions to IDA Resources: the Fifteenth Replenishment. Annex 1. 2 Refer to IDA (2009d) "Chairperson's Summary: IDA15 Mid-Term Review Meeting", Washington, DC, November 18-20. and IDA (2010), "Chairperson's Summary: IDA16 First Meeting", Paris, March 2-3. 3 The IDA-operational cutoff is updated annually. For FY10, the IDA operational cutoff was US$1,135. -2- 6. IDA allocates the bulk of its resources among eligible countries using the PBA formula. The PBA formula has evolved over time (see Annex 1 for a summary of the evolution since the 1990s). The core PBA system under implementation during IDA15 is discussed in Section A below. 7. While the bulk of IDA resources are allocated through the PBA system, five considerations have so far merited exceptions. First, some blend countries with large populations and with access to IBRD resources receive less than their PBA allocation due to their broader financing options (i.e. their allocation is capped). Second, countries emerging from severe conflict or after prolonged disengagement with IDA can, under certain conditions, be provided with additional resources. Third, arrears clearance operations for eligible IDA countries are separately funded through the systematic approach for arrears clearance. Fourth, there is a special provision for funding selected regional integration projects. Finally, additional resources can be provided in the aftermath of natural disasters when the existing allocation does not allow for a sufficient response. These exceptions are discussed in Section B below. 8. Overall, the PBA system has continued to function well during the IDA15 period. The system has continued to direct more resources to countries that are achieving faster development outcomes.4 At the same time, the system has continued to balance performance with needs by allocating, consistent with performance, about 50 percent of IDA15 resources to Sub-Saharan Africa, the region with the greatest needs for IDA resources. The PBA implementation process and the estimated allocation outcomes for the IDA15 period are discussed in Section C below. A. THE CORE PBA SYSTEM 9. The main factor determining a country's PBA allocation is its performance in implementing policies that promote economic growth and poverty reduction. The PBA formula during IDA15 can be written as: PBA = f(CPR5, Pop, GNIpc-0.125) where CPR is the country performance rating, Pop is population, GNIpc is the gross national income per capita. Given that IDA's resource allocation system is a zero-sum-game (i.e. one country's gain in terms of more allocations would mean availability of fewer resources for others for a given level of the resource envelope), the PBA formula measures a country's own performance relative to those of all other IDA countries. Annex 2 provides further details on the exact functional form of the formula. Figure 2 provides a schematic representation of the PBA formula during IDA15. 4 Refer to IDA (2009c), "IDA Performance Based Allocation and Development Results: An Update", IDA15 Mid-Term Review Paper, October. -3- he Figure 1. A Schematic Illustration of the Current PBA Formula esource Source: IDA Resource Mobilization Department · The CPIA measures the extent to which a country's policy and institutional framework supports sustainable growth and poverty reduction, and consequently the effective use of development assistance. It consists of 16 criteria grouped in four equally weighted clusters (see Annex 3).5 In calculating the Country Performance Rating (CPR clusters A CPR), to C, which measure macroeconomic and structural policies, carry a weight of 24 percent; while cluster D, which assesses governance and public sector management and The institutions, carries the weight of 68 percent. The weights reflect the importance placed on governance for development and in terms of mitigating fiduciary risks to aid funds.6 g Following the recent evaluation and recommendations of the Independent Evaluation Group (IEG), the CPIA is currently under review ( (Box 1). · The PPR reflects the health of the IDA projects portfolio, as measured by the percentage country. of problem projects in each country. Similar to the CPIA, it measures performance and implementation factors that are broadly under the control of recipient countries. In calculating the CPR, the PPR ccarries a weight of 8 percent. Annex 3 provides further determined. details on how the PPR is determine · The CPR is calculated as: CPR = (0.24 * CPIAA-C + 0.68 * CPIAD + 0.08* PPR) where CPIAA-C is the average ratings of CPIA clusters A to C and CPIAD is the rating of CPIA cluster D. 5 For each of the 16 criteria, countries are rated on a scale of 1 (low) to 6 (high) annually by World Bank staff. 6 Term discussion During the IDA14 Mid-Term Review and the IDA15 replenishment, there were some discussions on the weight of governance in the PBA formula where, as part of the PBA simplification exercise, Management proposed to slightly reduce the weight of governance from 0.68 to 0.6. However, Deputies requested that Management retain the weight on governance at 0.68 for all simplification options. -4- Box 1. IEG Evaluation of the CPIA: Key Findings and Recommendations The Independent Evaluation Group (IEG) recently completed an evaluation of the World Bank's Country Policy and Institutional Assessment (CPIA) (see IEG 2009). The evaluation confirmed the usefulness of the CPIA as a broad indicator of development effectiveness. The evaluation also found that the contents of the CPIA were largely relevant for growth and poverty reduction; that there is no strong evidence of upward bias for either the IDA or IBRD scores; and that they mapped well with the policies and institutions that are identified in the literature as relevant for growth and poverty reduction. The evaluation recommended that a thorough review of the CPIA be made and the criteria revised as necessary. The evaluation contained specific recommendations on a few CPIA criteria, such as the criteria covering trade, financial sector, and equity of public resource use. In these areas, the evaluation recommended that the criteria be revised. In other areas such as the governance cluster, the evaluation recommended that the criteria be streamlined. After assessing gaps in coverage, the IEG evaluation noted that the CPIA is missing an assessment of disadvantaged socioeconomic groups other than gender. The evaluation also recommended that the CPIA guidelines clarify which criteria should take into account the stage of development and how the adjustments should be made. Management has taken these recommendations on board and is currently undertaking a thorough review of the CPIA in the context of IDA16. The review will be informed by the recommendations of the IEG evaluation and is expected to be completed by the time IDA16 is launched (i.e. at the end of FY11). Source: IEG 2009 and "Management Response", October 2009. 10. While country performance is the main determinant, country needs are taken into account through population size and GNI per capita. Population affects allocations positively, with a 1 percent increase in population resulting in roughly a 1 percent increase in allocation (except for capped blend countries), all else remaining constant. Similarly, countries with lower GNI per capita receive a higher allocation (i.e. a 1 percent decrease in a country's GNI per capita leading to a 0.125 percent increase in IDA allocations to the country, all else remaining constant). An analysis carried out during the IDA13 replenishment7 showed that even large changes in the weight of GNI per capita induced relatively small changes in a country allocation. 11. Finally, IDA's PBA has a fixed component referred to as the "base allocation". The base allocation amounts to SDR1.5 million, which IDA allocates each year to all countries irrespective of their performance and needs. This base allocation was deemed as the minimum amount necessary to maintain a program in a country. In terms of per capita allocation, introducing a base allocation largely benefits small states. B. EXCEPTIONS TO THE PBA SYSTEM 12. As noted above, there are five exceptions to the PBA system: (i) capped blend countries, whose allocations are capped at levels below what the PBA formula would allocate to 7 Refer to IDA (2001), "Weighing Poverty in the IDA Allocation Formula", IDA13 Replenishment Paper, November. . -5- them; (ii) post-conflict and reengaging countries, where IDA can provide special support for their recovery and development needs; (iii) arrears clearance operations; (iv) regional projects that play a key role in regional integration and/or regional public goods; and (v) exceptional allocations in the aftermath of natural disasters. 13. Allocations to capped-blend countries. Historically, IDA has set caps on its allocations to IDA-eligible countries with large populations which have access to IBRD and other market resources. Currently, allocations to India and Pakistan are capped at 11 percent and 7 percent, respectively, of total IDA resources over a replenishment period.8 The rationale for this capping is two-fold: first, without capping, these countries would receive the bulk of IDA resources due to their large populations;9and second, these countries have access to broader funding options. 14. Allocations to post-conflict and re-engaging countries.10 These countries have benefited from exceptional IDA allocations since IDA13 in order to support their special needs. These exceptional allocations are roughly double the allocations received by other IDA countries (on a per capita basis) for post-conflict countries,11 and slightly less for re-engaging countries.12 The total duration of these allocations is up to ten years for post-conflict countries (of which six are phase-out) and up to five years for re-engaging countries (of which three are phase-out). Allocations to countries eligible for this support are determined based on performance as measured by the Post Conflict Performance Indicators (PCPI),13 and where available, portfolio rating. Annex 4 provides further details on how post-conflict allocations are determined. 15. Arrears clearance operations. During IDA15, IDA donors endorsed a systematic approach to arrears clearance in IDA countries,14 whereby IDA finances arrears that exceed a country's capacity to pay. The preconditions for re-engagement and arrears clearance support from IDA include strong international cooperation, expectation of continued country stability, a successful macro program, and a comprehensive arrears clearance plan. Eligibility is limited to countries that were in arrears at end-2006 and that are (or could become) HIPC-eligible during the forthcoming replenishment cycle.15 Any unused arrears clearance funds are rolled over into the next replenishment; conversely, any funding shortfalls are compensated in the context of the next IDA replenishment. 16. Allocations to regional projects. IDA has dedicated funding for regional projects promoting regional integration and/or the supply of regional public goods.16 The criteria for 8 The fixed percentages are applied to total IDA resources net of set-aside arrears clearance funds. 9 This capping also shifts IDA resources towards Sub-Saharan Africa (from 20 percent without capping to 50 percent with capping), a region with the greatest needs for IDA resources. 10 See Section IV below for more on post conflict and re-engaging countries. 11 Excluding India and Pakistan whose allocations are capped. 12 Refer to IDA (2007b), Operational Approaches and Financing in Fragile States, World bank Operational Policy and Country Services (OPCS) and Resource Mobilization Department, Washington, D.C., June. 13 A disclosure of PCPI ratings is planned by the end of IDA15. 14 Refer to IDA (2007c), Further Elaboration of a Systematic Approach to Arrears Clearance, Resource Mobilization Department, Washington, D.C., June. 15 Current eligible countries are Somalia and Sudan. Zimbabwe and Myanmar could become eligible upon being grandfathered for HIPC eligibility. Sudan and Zimbabwe are expected to re-engage with IDA during IDA16. 16 Refer to IDA (2003), Pilot Program for Regional Projects, World Bank Resource Mobilization Department. Washington, DC. -6- funding from the regional top up include: (i) a minimum of three participating countries, (ii) border evidence of cross-border benefits, (iii) country and regional ownership, (iv) coherence with regional strategy, and (v) potential to contribute to policy harmonization. The regional funds thirds cover two-thirds of the project costs, and the remainder is contributed by the participating country's PBA allocation. Funding for regional projects are allocated to regions on an annual out basis, with about 80 percent of the funding currently going to projects in Africa, where the demand for regional solutions has so far been the greatest. The remaining 20 percent has been shares of IDA allocated to other regions and countries based on three factors: the regular PBA sh resources for each region, the level of demand, and the quality of project proposals. 17. disasters. Allocations in the event of natural disasters. In cases where the existing allocations would not allow for a sufficient response, IDA has provided an exceptional allocation in the aftermath of major natural disasters. The scope of such assistance is narrowly defined and related limited to natural disaster-related incidents. The specific level of IDA allocation is based on a needs assessment following the disaster, how much other donors plan to contribute to the for recovery (i.e. "burden share") and whether conditions are in place for IDA resources to be used effectively. In practice, IDA has used the authority to grant such allocations sparingly given that they are taken out of a fixed IDA commitment authority and they reduce allocations for all IDA term countries' long-term development n needs. C. IMPLEMENTATION . Figure 2. Steps in Allocating IDA Resources to Eligible Countries AND ALLOCATION OUTCOMES FOR IDA15 18. The amount of IDA resources available for PBA allocation depends on IDA's commitment authority, which in turn is determined by donor contributions and internal resources. The ring-fenced exceptions discussed above do not constitute part of the resources allocated through the PBA system and as such are deducted from the total IDA Source IDA Resource Mobilization Department. Source: commitment authority in order to determine the "PBA allocable" resource envelope. -7- (i) Allocation steps 19. Once the PBA allocable resource envelope is determined, six steps are applied to determine the volume and terms of the country-by-country IDA allocation. These steps are summarized in Figure 2 above; and further details on the steps are provided in Annex 5. 20. Considerable complexity has been added to the allocation system since IDA14. Since IDA14, debt distress risk has been the primary grant eligibility criterion.17 Similarly, with the implementation of the MDRI, two additional steps have been added to the allocation system: first, deduction of an eligible country's annual debt service forgiven from its annual PBA allocation (also known as MDRI "netting out"); and secondly, the reallocation of the compensatory donor resources that IDA receives using the PBA system. Together, the grant allocation framework system and the MDRI netting out mechanism have added several steps to the PBA system. Figure 3. Estimated IDA15 Allocation Results by Country Groupings1 Other exceptional allocations 16.1% Post-conflict & Re- Regular IDA engaging countries countries 65.9% 7.4% Regional projects 4.3% Capped Pilot CRW countries 3.7% 18.1% Natural disaster- related allocation 0.5% 1. Based on actual allocation data for FY09 and FY10, and FY11 indicative allocations. Estimated shares of capped countries include hard-term credits. (ii) Allocation outcomes for IDA15 21. Under IDA15, it is estimated that the bulk of IDA resources­ 84 percent­are allocated either through the core PBA system or through capped allocations to blend countries with large populations (Figure 3).18 Specifically, (i) 66 percent of the resources are estimated to go to the regular IDA countries, through the core PBA system; (ii) 18 percent are estimated to go to the capped-blend countries (including hard-term credits); and (iii) 16 percent 17 Refer to IDA (2004a), Debt Sustainability and Financing Terms in IDA14, , Washington, D.C., June 24; and IDA (2004b), IDA. 2004b. Debt Sustainability and Financing Terms in IDA14: Technical Analysis of Issues and Options, September 22. 18 The estimates are based on actual allocation data for FY09 and FY10, and indicative allocation for FY11. -8- he are estimated to go to other exceptional allocations. Within the exceptional allocations, 7.4 post-conflict and re-engaging countries; 4.3 percent of IDA resources are estimated to go to post engaging Crisis percent to regional projects; and 3.7 percent to the pilot-Crisis Response Window. The pilot- through Crisis Response Window was funded through redeployment of unused arrears clearance funds set aside during IDA15, internal resources, and additional donor contributions. Special allocations in the aftermath of natural disasters during the IDA15 period have to date remained .4 million. relatively small at SDR134.4 million Annex 5 presents estimated IDA15 country allocations by the allocation steps described above. 22. The PBA system has continued to direct more resources to better performers (Figure 4). During IDA15, IDA countries in the top performance quintile received about 2.7 times in allocations per capita (SDR9.8) than those in the lowest quintile (SDR3.6). This is higher than performance the comparable ratio of 2.3 during IDA14, confirming that the performance orientation of the in-depth system has indeed slightly improved. Furthermore, a recent in depth analysis of the relationship between average IDA allocations per capita and changes in development outcomes (e.g. HDI, IDA15 Mid-Term Review,19 shows that countries immunization, etc) , conducted as part of the ID with higher per capita IDA allocation have on average experienced greater improvements in development. This is true for both the full sample of IDA countries, which includes post conflict and capped countries, as well as a sample excluding the latter two categories of countries. The countries correlation becomes stronger as we move from the full sample of IDA countries to the sample conflict without post-conflict and capped countries, as doing so would effectively correct for "distortions" caused by exceptions to the PBA system. . Figure 4. The Performance Orientation of the PBA Source: IDA Resource Mobilization Department 23. Finally, IDA allocations are now disclosed to IDA Executive Directors at the end of each fiscal year. As agreed during IDA15, Management has started to make allocations and information commitments available to IDA's Executive Directors for information at the end of each fiscal commitments year starting in FY09. The FY09 allocations, together with the corresponding commitments, 19 Refer to IDA (2009c), "IDA Performance Based Allocation and Development Results: An Update", IDA15 Mid Term Review Paper, October. -9- were shared with the Executive Directors in August 2009. Similar reporting is expected at the end of each fiscal year going forward, increasing the transparency of IDA's allocation system. Annex 6 discusses IDA resource management practices within a replenishment period. III. THE BALANCE BETWEEN PERFORMANCE AND NEEDS A. ACCOUNTING FOR CURRENT MEASURES OF NEEDS 24. While country performance is the main determinant of country allocations, IDA also considers country needs in its resource allocation system in a variety of ways: through the IDA GNI per capita cutoff, the GNI per capita and population components in the PBA formula, the minimum base allocation, capping of allocations to blend countries with large populations as well as exceptional allocations to post-conflict and re-engaging countries with special needs. 25. First, the IDA eligibility criteria grant access to IDA concessional resources to countries whose GNI per capita per annum in FY10 is below US$1,135 (excluding small island exceptions). This ensures that over 80 percent of IDA-eligible countries are low-income countries. 26. Second, within the group of IDA-eligible countries, the PBA allocation factors in population and GNI per capita as measures of needs. The population factor ensures that allocations are granted proportionately to a country's population for all levels of poverty. Similarly, the inclusion of GNI per capita in the formula ensures that a country with a lower GNI per capita receives a higher allocation, all else being equal. 27. Third, each IDA country receives a minimum base allocation of SDR1.5 million per annum. In terms of per capita allocations, this benefits small states, which have a unique set of needs, not captured by the GNI per capita indicator, such as greater vulnerability to shocks and periodic recurrence of natural disasters.20 28. Fourth, the capping of IBRD creditworthy countries with large populations (i.e. India and Pakistan) shifts resources to countries with greater development needs. For instance, using the FY09 performance data, if all funds were allocated using the PBA formula, 72 percent of resources would go to South Asia and Sub-Saharan Africa would receive only 20 percent (Figure 5). With capping of India and Pakistan, however, Sub-Saharan Africa's share in IDA resources more than doubles from 20 percent to 49 percent. Similarly, special allocations for the post conflict and reengagement countries further increase this share to 50 percent. 20 For further details, refer to IDA (2007a), IDA's Performance-Based Allocation System: Options for Simplifying the Formula and Reducing Volatility, Resource Mobilization Department, Washington, D.C., February. - 10 - Figure 5. Impact of Capping and Exceptional Allocations on Redistribution of apping edistribution esources Resources Across Regions (FY09) ECA, LCR and MNA 120% Percent of Total IDA Resources 3% 7% 6% 100% SOUTH ASIA 80% 33% 34% 72% 60% 11% 11% EAST ASIA 40% 5% 49% 50% 20% 20% SAHARAN SUB-SAHARAN AFRICA 0% Pure PBA PBA with caps PBA with caps and post-conflict and re-engaging countries Department. Source: IDA Resource Mobilization Department 29. For IDA15 as a whole, the PBA system is estimated to have directed about half (50 Sub-Saharan Africa, the region with the greatest needs (Figure percent) of IDA resources to Sub , ( ), 6), without compromising the performance orientation of the system. South Asia is estimated to receive 33 percent; East Asia and the Pacific 10 percent; Europe and Central Asia 4 percent; Latin America and the Caribbean 2 percent; and the Middle East and North Africa 1 percent. Figure 6. Estimated IDA15 Allocation Results by Region1 . 1/ Based on actual allocation data for FY09 and FY10, and FY11 indicative allocations. Estimated regional shares is hard include allocations through the regional projects, the pilot-Crisis Response Window (CRW), and hard-term credits for the relevant countries. Source: IDA Resource Mobilization Department - 11 - B. INCLUDING VULNERABILITY IN THE PBA FORMULA: IMPLICATIONS AND KEY CONCLUSIONS 30. In recent years, there has been much academic and policy debate about whether and how to modify IDA's PBA system to increase its development effectiveness. One suggestion has been to reform IDA's PBA formula so that it explicitly includes development outcomes (Kanbur, 2005). The implications of this proposal were explored in detail at the IDA15 Mid-Term Review.21 The results of that analysis showed that a results-based allocation system that uses outcome levels as a basis to allocate resources would shift IDA resources from LDCs to the relatively better off low income countries, while a system that uses changes in outcome levels raises serious concerns in terms of aid volatility, policy incentives, and the time lag between outcome data and allocations. 31. Another recent suggestion has been to modify the PBA formula to take into account vulnerability in order to capture structural handicaps to growth, in particular economic vulnerability and lack of human capital (Guillaumont 2008, Guillaumont 2009, and Guillaumont et al. 2010). The broad argument for this is to capture the equity or needs aspect of the PBA more broadly. Specifically, it has been suggested to replace the current country performance rating (CPR) by a weighted average of (i) CPR, (ii) a structural Economic Vulnerability Index (EVI)22 and (iii) a Human Assets Index (HAI).23 The EVI and HAI are compiled by the UN for the purposes of identifying least developed countries. 32. Four alterative formulations of a modified Country Performance Rating (CPR) are considered: Functional Form Simulation Modified CPR1 = 0.75 CPR+ 0.25 EVI Simulation 1, S1 Modified CPR2 = 0.50 CPR+ 0.50 EVI Simulation 2, S2 Modified CPR3 = 0.33CPR + 0.33 EVI + 0.33 HAI Simulation 3, S3 Modified CPR4 0.33CPR + 0.33 EVI + 0.33 HAI plus reduce the Simulation 4, S4 population exponent from 1 to 0.5 These simulations (S1-S4) are compared to the FY09 allocation outcome under the current system (before grant discount and MDRI netting out and excluding disaster-related exceptional allocations), which have already been disclosed to the Board. The remainder of this section summarizes the key conclusions of this exercise. Further details of these scenarios are presented in Annex 7; and Table A2 of that annex provides the allocation impact at the country level. 33. The introduction of vulnerability into the PBA formula would have several distributional implications. By performance quintile, the introduction of vulnerability would 21 Refer to IDA (2009c), "IDA Performance Based Allocation and Development Results: An Update", IDA15 Mid-Term Review Paper. October. 22 This is a composite indicator of seven components, three related to the size of recurrent exogenous shocks, either natural or external, and four related to the exposure to these shocks. The components are the instability of exports of goods and services, the instability of agricultural production and the homelessness due to natural disasters, the smallness of population (log), remoteness from world markets, export concentration and the share of agriculture in GDP. 23 This is a composite index of health and education conditions/indicators. - 12 - reduce the performance orientation of the current PBA system. As the weight on vulnerability increases in the PBA formula, the per capita allocation to countries in the lower performance quintiles increases relative to the present system. By country groupings, under S1-S2, India would receive a much higher allocation (nearly two to three times its current allocation) while most post-conflict and re-engaging countries would receive less than their current allocation. Only under S3, where the weight on vulnerability is significantly increased to two-thirds (and that of the CPIA kept at one-third), would these countries receive allocations comparable to their current levels.24 By region, Sub -Saharan Africa's allocation share is reduced under S1 (relative to the current case) but steadily increased in all other scenarios; while South Asia's increases in S1 but steadily declines in other scenarios. The allocation share of East Asia and Pacific declines under all simulations, and the picture for the other regions is much more mixed. 34. Introducing vulnerability will also raise several data and implementation issues. First, the vulnerability data are available every 3-4 years, which limits variations in allocations within those years (except due to changes in CPIA). In addition, the official data at the UN website do not cover all IDA-eligible countries (Armenia, Azerbaijan, Bosnia-Herzegovina, Georgia, Kosovo, Kyrgyz Republic, Moldova, Tajikistan, and Uzbekistan). Second, the data have significant time lags (3 to 5 years),25 which makes the allocation process backward looking. Finally, the vulnerability data exhibit significant fluctuations over time. EVI data from the two most recent rounds­2006 and 2009­vary from a decrease of 30 percent (Afghanistan) to an increase of 58 percent (Bosnia-Herzegovina).26 Such fluctuations would cause significant allocation volatility, which would place the vulnerability adjustment at odds with aid predictability. IV. SPECIAL FINANCING CONSIDERATIONS FOR IDA16 35. During the IDA15 MTR and later at the first meeting of the IDA16 replenishment, IDA Deputies requested that a review of IDA's PBA system consider two principal issues: (i) how to strengthen financing for conflict affected and fragile states, including a consideration to further extend the phase out period in light of the positive operational and financial impacts of the longer phase-out discussed during the IDA15 MTR;27 and (ii) how to strengthen financial assistance to small states. This section focuses on these issues. Options to strengthen IDA financing for conflict-affected and fragile states are discussed in Section A below. Proposals to strengthen IDA's financial assistance for small states during IDA16 are discussed in Section B below. 24 Even under S3, however, three countries would receive less than their current allocation (Côte d'Ivoire, -SDR34 million; Liberia, -SDR8 million and Republic of Congo, -SDR6 million relative to their actual FY09 allocation). 25 The 2009 HAI, for example, was based on the 2003-05 undernourishment data from FAO, with about four years of lag which did not capture the 2008 food crisis. 26 In addition, some changes--such as significant decreases in vulnerability in Georgia (-17 percent), Guinea (-13 percent), Ethiopia (-12 percent) and increases in Azerbaijan (44 percent), Angola (32 percent), and Bolivia (23 percent)--may be difficult to reconcile with actual country circumstances on the ground. 27 Refer to IDA (2009a), "IDA's Exceptional Allocation: A Review of the Implementation Experience with Lengthened Phase Out", IDA15 Mid-Term Review Paper, Washington, DC. October. - 13 - A. STRENGTHENING FINANCING FOR FRAGILE AND CONFLICT-AFFECTED COUNTRIES 36. Assistance to countries facing situations of fragility and conflict has been an important part of IDA's strategy: fragile and conflict-affected countries28 account disproportionately for gaps in achieving the MDGs; assistance to these countries is one of the six pillars of the Bank's strategy; the World Bank's Executive Board has discussed and endorsed some important measures to enhance strategy and operational effectiveness in fragile states; and the Bank is currently preparing a World Development Report on conflict, security, and development. Assistance in IDA15 has contributed to achievement of important results in some fragile countries,29 particularly taking into account the extraordinary challenges faced in such countries and the need to keep expectations, in particular about the pace of change, realistic. 37. IDA capacity to support conflict-affected and fragile states has been substantially enhanced in recent replenishments. Beginning with IDA13, countries emerging from conflict have received exceptional IDA allocations that support their special recovery and reconstruction needs. During IDA14, the special category of re-engaging countries was introduced. During IDA15 a number of enhancements were made to the post conflict and re-engagement allocations, including extending the duration of the phase-out period; linking the share of post-conflict allocations in overall allocations to changes in the overall replenishment size; introducing a systematic approach to arrears clearance for eligible countries (which are all fragile states); and strengthening the existing review process for, and moving towards public disclosure, of the Post Conflict Performance Indicators (PCPI) to increase transparency around the special allocations. 38. IDA allocations to fragile states during IDA15 are estimated at SDR2.5 billion, or about 10 percent of IDA15 resources (Figure 7). Within fragile states, post-conflict and re- engaging countries account for about 8 percent30 (excluding regional project funds, which are difficult to breakdown by country groupings) of IDA resources, while other fragile states account for 2 percent. Assuming that there are no changes to the current rules or that no new countries enter the special allocation window, the projected share of fragile states is set to decline from 10 percent in IDA15 to 6.7 percent in IDA16, as exceptional allocations phase down and countries return to their PBA allocations.31 While post-conflict and re-engagement countries have continued to receive substantially higher allocations than they would have received under the regular PBA system,32 the projected decline is owing to the fact that most of these countries are in the phase-out period (Table 1). Without a further extension of the phase out period, most of 28 Fragile states are defined as countries with CPIA less or equal to 3.2. 29 Refer to IDA (2009b), "IDA's Support to Fragile and Conflict-Affected Countries: Progress Report 2007-09", IDA15 Mid-Term Review paper, Washington, DC, October. 30 The share of post conflict and reengaging countries goes up slightly from 7.4 percent in Section II to 8 percent in this section due to the exclusion of regional project funds. 31 These projections assume: a 6 percent nominal growth of IDA envelope in IDA16 (no real growth); countries' performances are held at their FY10 level; Azerbaijan graduates from IDA at the end of IDA15; the MDRI netting out is capped at 30 percent of gross PBA allocation; post-conflict and re-engaging countries exit the window as per the IDA15 timetable; and there will be no re-engagement of Sudan or Zimbabwe during IDA16. 32 During IDA15, the post conflict and re-engaging countries received about 150 percent higher than their PBA allocation; furthermore, they together would have received only about 3.1 percent of total IDA15 resources under the normal PBA rules, a substantially lower amount than the 8 percent share they actually received under the exceptional allocation mechanism during IDA15. - 14 - the countries--with the exception of Côte d'Ivoire and Liberia-- would exit the special allocation window during IDA16. Figure 7. Estimated Resource Allocation to Fragile and Conflict-Affected Countries in IDA151 Post-conflict and re- Non-fragile states engaging countries2 Fragile states3 8% 90.0% 10% Other fragile states4 2% Note: 1/ The IDA15 estimates are based on actual allocation for FY09 and FY10, and indicative allocation for FY11. 2/ Afghanistan, Angola, Burundi, Côte d'Ivoire, Democratic Republic of Congo, Republic of Congo, Eritrea, Liberia, and Timor- Leste. 3/ Central African Republic, Haiti, and Togo. 4/ Based on the 2008 CPIA score, Chad, Comoros, Djibouti, Guinea, Guinea-Bissau, Kiribati, Sao Tome and Principe, Sierra Leone, Solomon Islands, Tajikistan, Tonga, and the Republic of Yemen. Source: IDA Resource Mobilization Department. Table 1. Current Schedule of Exit from the Exceptional Allocation Window IDA16 FY11 FY12 FY13 FY14 Haiti Congo Republic Angola Afghanistan Eritrea Burundi Central African Rep Congo, DR Timor-Leste Togo Source: IDA Resource Mobilization Department. 39. This note discusses three financing issues relevant for conflict affected and fragile states during IDA16: (i) supporting a possible re-engagement during IDA16 of Sudan and Zimbabwe; (ii) extending the existing exceptional allocation mechanisms for post-conflict and re-engaging countries, in view of the longer recovery and reconstruction support that these countries need; and (iii) increasing flexibility in the rules governing IDA regional projects with a view to better supporting fragile and conflict-affected countries. The remainder of this section explores the rationale and financial implications (during IDA16) of these considerations. (i) Supporting the re-engagement of Sudan and Zimbabwe 40. During the IDA15 discussions, donors endorsed a systematic approach to help finance the clearance of arrears for eligible IDA countries. The approach covers IDA countries with arrears to IBRD and/or IDA as of December 31, 2006 and grandfathered for - 15 - eligibility under the HIPC Initiative but that have yet to reach the HIPC decision point.33 While all the countries eligible for support under this approach are fragile states, their economic and financial capacity vary widely. The design of arrears clearance packages, including with respect to the share of arrears that each country would be expected to finance from its own resources, will take into consideration such differences. In that regard, the level of support to be provided by IDA under this approach will be determined through a detailed assessment prepared by Bank staff. 41. Currently, four countries (Myanmar, Somalia, Sudan and Zimbabwe) are eligible or could become eligible for exceptional IDA support under this approach. The arrears to IBRD and IDA from these four countries amount to SDR1.3 billion at end-March 2010. There is uncertainty surrounding the timing for the re-engagement of these countries: · At this time, the situation in Myanmar and Somalia is such that a re-engagement during IDA 16 does not seem likely. · In the case of Sudan, the timing and manner for re-engagement with the Bank hinges critically on the outcome of the January 2011 Southern Sudan Independence Referendum. Assuming that all requirements are satisfied, the re-engagement with the Bank could take place during FY12. · At present Zimbabwe is not eligible for support under the systematic approach for arrears clearance; however, it could become eligible as an exception or modification of the HIPC initiative eligibility criteria.34 Given that context, it is assumed that the re-engagement with the Bank could take place during the IDA16 period. 42. The estimated cost of providing exceptional support for arrears clearance to Sudan and Zimbabwe is about SDR770 million.35 Both Sudan and Zimbabwe are assumed to re- engage at the beginning of IDA16 (i.e. FY12). However, to determine the actual level of exceptional arrears clearance support from IDA, Bank staff will undertake in due course an assessment of these countries' capacity to pay--including availability of domestic resources, the debt sustainability outlook and the level of the financing gap. In addition to arrears clearance, the countries would receive about SDR300 million in special IDA allocation (Sudan as a post- 33 Eligible countries would receive exceptional arrears clearance support after meeting the Bank's requirements for re-engagement, including a medium term growth-oriented reform program, satisfactory performance under an IMF program, and a financing plan that provides for the full clearance of arrears to the World Bank. To date, Liberia, Côte d'Ivoire and Togo have benefited from such support. 34 Zimbabwe has not been grandfathered for HIPC Initiative debt relief because at end-2004 Zimbabwe was a blend IDA/IBRD country, not an IDA-only and PRGF-eligible country as required under the criterion. Hence, even if Zimbabwe meets the HIPC Initiative indebtedness criterion, it could only be added to the list of countries grandfathered for HIPC eligibility upon a modification of, or exception to, the Bank's HIPC income eligibility criteria. 35 Sudan's estimated arrears to IDA (as of April 1, 2010) are SDR383 million. Zimbabwe's arrears (as of April 1, 2010) are SDR88.9 million to IDA, and SDR411.9 million to IBRD. For the purposes of IDA16 financing, and without prejudging the actual amounts, it is assumed that IDA will finance about two-thirds of Sudan's arrears and a 100 percent of Zimbabwe's arrears. The arrear amounts are estimated to grow by about SDR20 million between end-April 2010 and the beginning of the IDA16 period, when the countries are assumed to re-engage with IDA.IDA's arrears clearance and allocation amounts presented for Sudan are for the whole of Sudan, and would be appropriately apportioned in the event of possible independence of Southern Sudan. . - 16 - conflict country and Zimbabwe as a re-engaging country) to support their re-engagement and recovery during IDA16. (ii) Extending the phase-out period for post conflict and re-engaging countries 43. In response to the need to avoid a substantial decrease in allocations for post-conflict countries, during the IDA15 Replenishment the phase-out period for exceptional allocations was increased from three to six years resulting in ten years of support for post-conflict countries which would have otherwise exited the window during IDA15.36 Similarly, the phase-out period for re-engaging countries (Haiti, Central African Republic and Togo) was extended from two to three years (for a total five years of exceptional allocation). During IDA16, only two countries (Côte d'Ivoire and Liberia) would continue to receive exceptional allocations if there is no further extension of the phase out period. 44. The experience of post conflict and re-engagement countries shows, however, that transition periods vary substantially across countries and that in many cases exceptional support beyond the existing 10-year period may be warranted. Compared to the period immediately before entering into post-conflict or re-engagement status, many countries have made progress during the post conflict or re-engagement period. Trends in the CPIA ratings for the post-conflict countries and re-engaging countries show that while progress on the whole has been gradual, some countries have exhibited improvements (e.g., Angola) while others have shown deterioration (e.g. Eritrea). Despite the significant progress, none has yet exited from fragility (i.e. achieved a CPIA rating above 3.2), suggesting the seriousness and complexity of the challenges faced by these countries and need for a longer support for recovery and transition. 45. In light of this variability across countries, for some countries, slowing the pace at which exceptional allocations are reduced and phased out makes good sense from an aid effectiveness perspective, as large discrete reductions in aid (particularly for countries whose CPIA/PCPI performance has not improved rapidly due at least in part to factors beyond the government's control) can be counterproductive and de-stabilizing. This is particularly true for IDA reductions, given that this form of aid is fundamentally supportive of state-building (which is at the heart of post-conflict reconstruction). The extension of the phase-out is important because it increases allocations for these countries thus enabling them to fund their emergency needs and essential public services, including safety nets to poor and vulnerable groups. 46. For other countries, extending the phase out period may not be necessary ­and indeed early exit from the special allocation window may be warranted. When the phase out period was extended in IDA15, it was stipulated that under certain circumstances, countries could revert to regular PBA allocations before the end of their phase out period. These circumstances would include:37 (i) attaining a "gap" country status, which would mean that per capita income would have exceeded IDA's operational cutoff; (ii) attaining "blend" country 36 Refer to IDA (2009a), "IDA's Exceptional Allocation: A Review of the Implementation Experience with Lengthened Phase Out", IDA15 Mid-Term Review Paper, Washington, DC, October. 37 Refer to IDA (2007b), "IDA's Support to Fragile and Conflict-Affected Countries: Progress Report 2007-09", IDA15 Mid-Term Review paper, Washington, DC, p.33. - 17 - status, which would suggest creditworthiness and access to other sources of financing; or (iii) violation of IDA's non-concessional borrowing policy. 47. With respect to extending the phase out period during IDA16, Management proposes two options for consideration by the IDA Deputies: Option 1: Extend the length of the phase-out period uniformly by two years for post conflict countries and by one year for re-engaging countries (with a total support of 12 years for post conflict countries and 6 years for re-engaging countries). 48. The financial impact of Option 1 is estimated at SDR376.4 million and at about SDR330 million if Angola is excluded on the grounds of its high per capita GNI of US$3,450 (Table 2). The proposed extension will benefit most countries during IDA16 (with the exception of Côte d'Ivoire and Liberia which are not expected to exit the special allocation window during IDA16). Table 2 details the impact of such extension by country. Table 2. Projected IDA16 Allocation Impact of a Longer Phase-out (Option 1) (SDR million) Country IDA15 estimate IDA16 allocation with no IDA16 allocation Change in allocation lengthened phase out1 with lengthened due to lengthened phase-out phase out Post-conflict countries Afghanistan 359.2 185.0 259.5 74.4 Angola 195.1 114.6 163.1 48.5 Burundi 152.2 85.2 123.8 38.6 Congo, DR 723.8 379.6 567.7 188.1 Congo, Republic of 39.3 15.9 24.2 8.3 Côte d'Ivoire 268.4 240.4 245.2 4.7 Eritrea 43.4 33.3 40.3 6.9 Liberia 93.5 82.6 83.5 0.9 Timor-Leste 8.8 7.4 8.3 0.9 Re-engaging countries Central African Republic 25.4 18.2 18.9 0.7 Haiti1 52.4 41.8 41.1 -0.7 Togo 80.8 40.8 45.8 5.0 Total 2042.3 1244.9 1621.2 376.4 Post-conflict countries 1883.7 1144.1 1515.4 371.4 Re-engaging countries 158.6 100.8 105.8 5.0 % share in IDA resources Post-conflict countries 7.4% 4.3% 5.7% 1.4% Re-engaging countries 0.6% 0.38% 0.4% 0.0% Note: 1/ Haiti is projected to exit the window in FY11; and extension of the phase-out period by 1 more year will not benefit the country during IDA16 (FY12-14). However, a special allocation is being proposed for Haiti for the IDA16 period in light of the earthquake related reconstruction needs­ see the companion paper "The Demand for IDA16 Resources and the Strategy for their Effective Use", IDA Resource Mobilization Department, May 2010. Source: IDA Resource Mobilization Department. - 18 - 49. The advantage of Option 1 is equity and a uniform treatment of all countries in the window. The disadvantages are that: (i) it does not reflect or capture the diversity of post-conflict experience across countries in terms of the time required for transition, or the special problems that certain post-conflict countries may face, often due to factors beyond their control; (ii) it entails a higher cost of financing as the extension has to be made for all countries in the window; and (iii) there are no incentives for countries to exit the window nor any criteria as to when the phase out should stop, and even if the phase out is extended for IDA16, the same question remains in future replenishments. Option 2: Beyond 10 years of support for post-conflict countries and 5 years of support for re- engaging countries, extend the phase out period for individual countries based on a case-by-case assessment against a specified set of criteria. 50. Under this option, for countries that have already received 10 years of exceptional support, the progress and pace of post-conflict transition would be regularly reviewed in order to justify whether further extension of the phase out period is warranted or not. The advantages are that (i) it reflects or captures the diversity of post-conflict experience across countries in terms of the time required for transition, and adjusts special support to their particular circumstances; (ii) it entails a lower cost of financing than Option 1; and (iii) it puts in place a medium term framework for assessing whether a country's special support needs to continue. The disadvantage is a perception of inequity of treatment, as the approach does not treat all countries in the window identically. 51. The specific criteria against which to assess further extension are: (i) Limited economic status and financing options: the country's GNI per capita (Atlas method) is not above the IDA operational cutoff and the country does not have access to IBRD financing (i.e. the country is IDA-only with limited economic and financing capacity); (ii) The presence of clear factors slowing down the transition, most notably a resurgence or continuation of conflict in parts of the country particularly if this includes regional or global drivers beyond the country's control; (iii) Continued exceptional treatment by other MDBs (as demonstrated, for example, by special allocation) in recognition of the country's special development needs ­which will be essential in harmonizing approaches with other multilateral organizations. Annex 8 summarizes the current practices of African and Asian Development Banks with regard to post conflict allocation and phase-out mechanisms. 52. Based on these criteria, the decision rule would be as follows: (i) Do not extend a country's phase out period if condition (i) above is not met; (ii) Extend the country's phase out period if it meets (i) above and at least one of the two criteria listed under (ii) to (iii) above. - 19 - 53. It is proposed that this assessment be conducted by Management and a determination made in the context of each IDA replenishment negotiation (and the financial costs presented as part of the Ask paper). If the decision is reached to extend a country's phase out period, the extension will cover the next replenishment period (for both post-conflict and re- engaging countries), subject to the condition that the cumulative support does not exceed 20 years for a post conflict country and 10 years for a re-engaging country. The financial benefit to the country judged eligible for further extension will be limited to that arising from a longer phase out, with no other changes to its level of allocation. Table 3. Application of the Phase Out Extension Criteria to Current Post Conflict and Re- engagement Countries Country Criterion (i): Criterion (ii): Criterion (iii): Preliminary IDA-only Slow transition due to Continued Decision 1/ country? continuation of conflict exceptional or other exogenous treatment by other factors? MDBs? Post-conflict countries Afghanistan Yes Yes Yes Extend Angola No No No Do not extend Burundi Yes No Yes Extend Congo, DR Yes Yes Yes Extend Congo, Republic of No No No Do not extend Côte d'Ivoire 2/ Yes No Yes NA Eritrea Yes No No Do not extend Liberia 2/ Yes No Yes NA Timor-Leste No No Yes Do not extend Re-engaging countries Central African Republic Yes No Yes Extend Haiti 2/ Yes Yes Yes NA Togo Yes No Yes Extend Note: 1/ Preliminary at this stage as it is based on desk review. The final decision will be informed by a more rigorously application of the criteria, and in consultation with the respective country authorities. 2/ The decision as to whether to extend the phase out is not under consideration for Côte d'Ivoire and Liberia because they do not need extension during the IDA16 period; and for Haiti because it is expected to exit the window in FY11, and a separate financing proposal is being presented for IDA16. Source: IDA Resource Mobilization Department, staff estimates. 54. Based on the above criteria and decision rule, a preliminary desk review shows that four countries--Angola, Congo Republic, Eritrea and Timor-Leste--would have to exit the window without any further extension to their current phase out period (Table 3). As of FY2010, Angola's per capita Gross National Income was US$3,450, above the IDA operational cutoff, and much higher than the rest of the countries in the window. In addition, the country has recently been declared creditworthy for IBRD lending.38 Congo Republic is a hardened term country with a GNI per capita above the IDA operational cutoff for at least two years in row. In FY10, its GNI per capita stood at US$1,970. Timor-Leste's GNI per capita in FY10 was US$ 38 This is based on the Bank's creditworthiness mission to Angola in August 2009. The government would need to formally request IBRD if it wished to start borrowing from IBRD. - 20 - 2,460 and it is expected to move to the hardened term category from FY11. Although Eritrea is an IDA-only country, it would exit the window without any further extension because it does not meet both criteria (ii) and (iii). The rest of the countries--Afghanistan, Burundi, Congo DR, Central African Republic, and Togo--would meet the criteria for further extension of their phase out period during IDA16. 55. The preliminary financial cost of Option 2 is SDR275 million, significantly less than the amount required under Option 1. If IDA Deputies endorse this case-by-case approach, Management will present an updated financing estimate associated with the lengthened phase out at the third IDA16 replenishment meeting. (iii) Enhancing IDA's Capacity to Address Regional Issues of Fragility and Conflict 56. Regional spillovers of conflicts and their adverse impacts across national borders call for an effective IDA response through regional projects. Some modest adjustments to the existing IDA regional window can be made to address the regional dimensions of fragility and conflict. Regional IDA funds can be used to support projects involving landlocked fragile countries where investments are required in one or more neighboring countries to provide better trade and communications logistics with the rest of the world. Similarly, regional projects would be a good vehicle to address sub-national conflicts in border areas that, either directly or in terms of their immediate effects, cover two or more countries. 57. Specifically, the following change in the criteria governing the use of regional IDA resources is proposed: · the current three-country requirement for regional projects be relaxed to allow two countries, of which at least one is a fragile or conflict-affected country, to be eligible for financing for regional IDA projects. The three country criterion will be retained for other countries. 58. Possible examples of where these changes would help enable and facilitate IDA support for beneficial regional activities involving fragile and conflict-affected states include the following: (i) infrastructure improvements through Kenya beneficial to Sudan; (ii) strengthening infrastructure access for landlocked Zimbabwe via one or more neighboring countries; (iii) a set of investments and programs on both sides of the Afghanistan-Pakistan border to address drivers of conflict that spill over between the two border areas; and (iv) infrastructural and communications investments enhancing access of other landlocked fragile states such as the Central African Republic. Furthermore, when funding is considered for fragile and conflict-affected countries through the regional IDA window, an effort would be made to measure or capture the potential peace-building impacts of regional development and economic projects, not just their economic and development benefits. (iv) Summary of Financial Implications 59. Overall, with no changes to the current rules or that no new countries enter the special allocation window, the projected share of fragile and conflict affected countries is 6.7 percent in - 21 - IDA16 (Table 4). However, with the possible re-engagement of Sudan and Zimbabwe, the projected allocation share of fragile and conflict affected countries in IDA16 would increase from 6.7 percent to 10.6 percent including arrears clearance. The cost of arrears clearance is estimated at about SDR770 million; in addition, the two countries are estimated to receive about SDR300 million in special IDA allocations to support their re-engagement and recovery. 60. The financial cost of the proposed extension of the phase out is about SDR330 million (or 1.1 percent of IDA resources) under Option 1, and SDR275.0 million under Option 2, which depending on the option chosen will further increase the projected allocation share of fragile and conflict affected countries in IDA16 to 11.9 percent or 11.7 percent of IDA resources. Table 4. Financial Implications of Proposed Measures for Fragile and Conflict-Affected Countries (IDA16) Proposed Changes Estimated IDA16 allocation to fragile and conflict-affected countries Amount % share in IDA (SDR million) resources 1. Estimated allocation with no changes to the IDA15 rules 1,791.8 6.7% 2. Cost of re-engagement of Sudan and Zimbabwe 1,070.0 4.0% Of which: Arrears clearance 770.0 2.9% IDA country allocations 300.0 1.1% 3. Cost of extended phasing-out Option 1 330.0 1.2% Option 2 275.0 1.0% 4. Cost of a two-country flexibility for regional projects 1/ 0.0 0.0% Total with Option 1 phase out 3,191.8 11.9% Total with Option 2 phase out 3,136.8 11.7% Note: 1/ Although the implementation of this additional flexibility is expected to benefit fragile and conflict-affected countries, the allocation impact is hard to predict at this stage as it would depend on the number of countries that choose to take advantage of the new rule. Source: IDA Resource Mobilization Department. 61. The proposal to modify the rules of IDA regional funding from the current three-country to two-country (of which one is fragile or conflict-affected) requirement, if endorsed by Deputies, would provide additional flexibility to address regional spillovers arising from fragility and conflict but the costs will be accommodated through the regional project funding. As such, it does not have a separate financial cost during the IDA16 period. (v) Forward Looking Work Agenda 62. While the above measures would, for the IDA16 period, strengthen financing for fragile and conflict-affected countries, further work is needed going forward. The WDR 2011 on fragility and conflict, currently in progress, is expected to generate useful knowledge and understanding about the causes and types of fragility, and the operational and policy lessons that could inform IDA's engagement in these countries. Management intends to draw on these findings, and if warranted develop further financing proposals for fragile and conflict affected - 22 - countries (e.g. modification and/or simplification of the allocation framework) for IDA Deputies' consideration at the IDA16 Mid-Term Review.39 B. STRENGTHENING IDA'S FINANCIAL ASSISTANCE TO SMALL STATES 63. Small states face a number of challenges owing to their higher vulnerability to economic shocks and natural disasters.40 These include higher per capita costs of production and basic services provision, higher vulnerability to economic shocks and significantly more volatile growth rates than larger countries. Many small states are also prone to natural disasters-- cyclones, droughts, hurricanes, tsunamis, and volcanic eruptions--which compound the human, economic, and environmental costs and the financial burdens of replacing lost infrastructure. Given these vulnerabilities, it is critical for small states to have the financial capacity to react quickly and flexibly to adjust to shocks, provide infrastructure and public services, and generate effective information flows. The most vulnerable small states lacking market access and IBRD creditworthiness also experience more difficultly in mobilizing official development assistance.41 64. IDA has continued to look for ways to strengthen its financial assistance to small states--defined as countries with a population of 1.5 million or less-- in light of their vulnerability. As pointed out earlier, the PBA formula has a `fixed' component (or base allocation), which favors small states by allocating a lump sum amount to each country. By providing a base allocation, irrespective of population or performance, many small countries are guaranteed a considerably greater allocation than they would receive if the entire allocation were based on population and performance ratings. In addition, although many small states have GNI per capita far above the IDA operational cutoff, they continue to access IDA resources in recognition of their vulnerability and special circumstances. 65. IDA currently supports 17 small states, while three more small states42 are seeking a reclassification from IBRD to IDA-eligibility (Table 6). During IDA15, the base allocation was raised from SDR1.1 million to SDR1.5 million per year, and the ceiling on per capita allocation was raised from SDR13.2 to SDR19.8 to adjust for inflation since IDA9. In addition, a 20 percent ceiling was put on country contributions to regional projects to help countries with small allocations take part in these projects. 66. Still, two key challenges remain. First, some small states' allocations continue to be constrained by the maximum per capita allocation ceiling. In the current PBA model, the per capita allocation ceiling supersedes the minimum base allocation provision, which means that small states with a population of less than 75,800 would be constrained by the ceiling before they reach their full base country allocation of SDR1.5 million. While Dominica is the only country 39 For further discussion on this, please refer to the fragile and conflict affected countries section of the "Special Themes Paper", being presented concurrently with this paper for discussion at the June meeting. 40 Refer, for example, to Commonwealth Secretariat/World Bank (2000), Toward an Outward Oriented Development Strategy for Small States: Issues, Opportunities, and Resilience Building--A Review of the Small States Agenda, April. 41 Refer to IDA (2004), "Supporting Small and Vulnerable States", IDA14 Replenishment Paper, December, pp. 2- 3. 42 The Marshall Islands and the Federated States of Micronesia are seeking a reclassification back to IDA in light of their lack of IBRD creditworthiness. Tuvalu has just become a member of IDA, and would likely seek access to IDA resources under the small states exception. - 23 - currently receiving less than its base allocation, some small states currently in the process of seeking IDA-eligibility--in particular the Marshall Islands and Tuvalu ­may face the same problem because of their low populations.43 Furthermore, six other countries would receive higher allocations than they do at present if the per capita allocation ceiling were to be relaxed. 67. Second, the current base allocation is too small to maintain an effective country program. While the base allocation was revised upwards during IDA15 from SDR1.1 million to SDR1.5 million, it is clear that for many small states the increase was not significant enough. Table 5 provides an illustrative calculation of how much infrastructure investment can be made with the current base allocation. Based on available unit cost estimates, the current base allocation can only fund 6 kilometers of two-lane paved road; 3 km of rail road; 1,200 kilowatt of electricity generation; or 5,700 households connected to sanitation (excluding maintenance costs). Table 5. How Much Infrastructure Does the Current Base Allocation "Buy"? What the current base allocation 1 Sector Unit Unit costs (US$) of SDR1.5 million can "buy" Electricity Per kilowatt of generation capacity 1,900 1,200 kilowatt Roads Per kilometer of two lane paved road 410,000 6 kilometers Railway Per kilometer of rail, 900,000 3 kilometers Sanitation Per connected household 700 3,257 households Water Per connected household 400 5,700 households Mainlines Per line 400 5,700 mainlines Mobile Per subscriber 580 3,931 mobile lines Note: 1/ Unit costs for mobile phone are from Pyramid Research; for mainlines from: Ruzzier, Kennet, Benitez, and Estache (2000). Other unit costs are Bank staff estimates. Source: Fay and Yepes (2000). (i) Lifting the per capita allocation ceiling 68. Two options can be considered to address the per capita allocation ceiling constraint. One option is to increase the per capita allocation ceiling for IDA16. Another option is to eliminate the per capita allocation ceiling altogether. · Under the first option, the level of per capita ceiling would need to be raised from the current SDR19.8 to SDR40, in order to allow all current small states to receive their full allocation without being constrained by the ceiling. However, even if this is done, a potential future eligibility of some micro states such as Tuvalu would mean that this ceiling will have to be adjusted again in the future. · Under the second option, all small states would clearly be able to receive their full PBA allocation and no future revision of the ceiling will be needed. The total additional resources going to small states as a result of eliminating the ceiling (relative to the current 43 Marshall Islands and Tuvalu would, respectively, receive up to SDR1.2 million and SDR0.2 million per year under the current per capita cap. - 24 - system with cap) is about SDR4.6 million during the IDA15 period. While the original rationale for the maximum per capita ceiling may have been to maintain fairness across countries in terms of the support they receive from IDA, this ceiling has almost always constrained small states, with no instances of mid or large population countries reaching the limit. Furthermore, for the small countries that have already reached the per capita ceiling, this constraint has meant that there is no longer a link between performance and allocations. Eliminating the per capita ceiling can thus not only increase these countries' IDA allocation but also strengthen the performance incentive for small states and reduce the complexity of the PBA system. 69. Given the importance of base allocations in small states, Management proposes eliminating the per capita allocation ceiling. Such a step would have little impact on the performance orientation of IDA's PBA system, but will have a relatively high benefit for several small states (as a share of their current allocation), in particular micro states, whose allocations are currently constrained by this ceiling. - 25 - Table 6. Allocation Impact of Eliminating the Maximum Per Capita Ceiling and Raising the Base Allocation from SDR1.5 Million to SDR3 Million Per Year Estimated IDA15 allocation (SDR m) with base Pop. 2008 with base allocation of Country CPIA 2008 (thousand) allocation of SDR1.5 m and with base % Increase SDR1.5 m and no per capita allocation of with SDR3 m per capita cap of cap SDR3 m and no and no per SDR19.8 per capita cap capita cap Small states that exceed the per capita cap Dominica 3.85 73.2 4.3 6.6 10.9 151% Tonga 1/ 3.19 103.6 4.9 4.9 8.4 74% Grenada 3.72 105.6 6.3 7.1 11.3 79% St. Vincent and the Grenadine 3.83 109.1 6.7 7.4 11.7 75% St. Lucia 3.88 170.0 9.9 10.1 14.3 45% Samoa 3.99 190.0 11.3 11.8 16.2 45% Small states that do not exceed the per capita cap Kiribati 3.02 97.0 5.7 5.7 10.1 79% Sao Tome and Principe 2.98 161.0 4.6 4.6 8.2 77% Vanuatu 3.30 231.1 7.9 7.9 12.4 56% Maldives 3.43 310.5 11.0 11.0 15.4 40% Cape Verde 4.16 498.7 25.1 25.1 29.3 16% Solomon Islands 2.76 507.0 6.0 6.0 9.5 60% Comoros 2.34 643.6 5.7 5.7 9.2 63% Bhutan 3.87 686.8 24.7 24.7 28.7 16% Guyana 3.44 763.4 8.3 8.3 12.1 45% Djibouti 3.12 847.7 10.7 10.7 14.2 33% Timor-Leste 2.81 1,098.4 8.8 8.8 11.5 30% 2/ Potential IDA-eligible small states Marshall Islands NA 59.7 Micronesia, Fed. Sts. NA 111.3 Tuvalu NA 12.4 Total small states 6,597 161.7 166.3 233.3 44% % Share in IDA resources 0.6% 0.7% 0.9% Note: 1/ Tonga crossed the ceiling starting in FY10. 2/ The Marshall Islands and the Federated States of Micronesia are seeking a reclassification back to IDA in light of the recent IBRD creditworthiness assessment that concluded that the two countries are no longer IBRD-eligible. Tuvalu is in the process of applying for IDA membership and could be eligible for IDA allocation under the small countries exception. (ii) Raising the base allocation 70. In light of the higher per capita costs of production and basic services provision, an option to consider would be to raise the base allocation from the current level of SDR1.5 million to SDR3 million per year (or from SDR4.5 million to SDR9 million per replenishment). With 79 IDA countries, the implication is that SDR355 million (= 79*SDR4.5 million), or about 1.4 percent of IDA15 resources, would be allocated outside the performance based system-- although most countries' allocations would exceed this basic allocation. Yet, this measure, combined with the elimination of the per capita ceiling, will increase the resources available to - 26 - small states significantly--by an average of 44 percent of their PBA allocation during IDA15 (Table 6). Table 6 details the impact of such measure by country. 71. Given the high per capita costs of production and basic service provision, Management proposes raising the base allocation to SDR3 million per year. This would increase the allocations of small countries, without causing a significant departure from the current performance orientation of the PBA system or significant changes in country allocations of non-small countries. V. ISSUES FOR DISCUSSION 72. With respect to financing for fragile and conflict-affected countries: · Do Deputies agree to extend the phase-out period for post conflict countries uniformly for all countries in the window (Option 1) or on a case-by-case basis (Option 2)? · Do Deputies agree to modify the current three-country requirement for regional projects to allow two countries, of which at least one is a fragile or conflict-affected country, to be eligible for IDA regional financing? 73. With respect to financing for small states: · Do Deputies agree to eliminate the maximum per capita allocation ceiling in order to better support small IDA states? · Do Deputies agree with raising the base allocations from the current SDR1.5 million per year (or SDR4.5 million per replenishment) to SDR3 million per year (or SDR9 million per replenishment) for the benefit of small states? - 27 - REFERENCES African Development Bank. 2010. "Proposed Adjustments to the Enhanced Approach to Fragile States", AfDF-12 Replenishment Discussion Paper, Forthcoming. Asian Development Bank. 2010. "ADF Allocation to Afghanistan", Forthcoming. Commonwealth Secretariat/World Bank. 2000. Toward an Outward Oriented Development Strategy for Small States: Issues, Opportunities, and Resilience Building--A Review of the Small States Agenda, Commonwealth Secretariat/World Bank Joint Task Force Report, April. Fay, Marianne and Tito Yepes. 2000. "Investing in Infrastructure: What is Needed from 2000 to 2010". Guillaumont, P. 2008."Adapting Aid Allocation Criteria to Development Goals" An Essay for the UNDevelopment Cooperation Forum, FERDI Working Paper, May. Guillaumont, P, S. Guillaumont Jeanneney, and D. Vencatachellum. 2009a. "Accounting for Vulnerability ofAfrican Countries in Performance Based Aid Allocation", Working paper series, African Development Bank Group, Working paper n° 103, October 2009, 24p. Guillaumont, P. 2009b. "An Economic Vulnerability Index: Its Design and Use for International Development Policy", Oxford Development Studies, Vol. 37, n° 3, p. 193-228. September. IDA. 2001. "Weighing Poverty in the IDA Allocation Formula", IDA13 Replenishment Paper, November. IDA. 2003. Pilot Program for Regional Projects, Resource Mobilization Department. Washington, DC. IDA. 2004a. Debt Sustainability and Financing Terms in IDA14, , Washington, D.C., June 24. IDA. 2004b. Debt Sustainability and Financing Terms in IDA14: Technical Analysis of Issues and Options, September 22. IDA. 2006a. IDA14 Mid-term Review of the IDA Pilot Program for Regional Projects, Washington, D.C. IDA. 2006b. "Chairman's Summary: IDA14 Mid-Term Review Meeting." World Bank, Washington, D.C., November. IDA. 2007a. IDA's Performance-Based Allocation System: Options for Simplifying the Formula and Reducing Volatility, Resource Mobilization Department, Washington, D.C., February. - 28 - IDA. 2007b. Operational Approaches and Financing in Fragile States, Operational Policy and Country Services (OPCS) and Resource Mobilization Department, Washington, D.C., June. IDA. 2007c. Further Elaboration of a Systematic Approach to Arrears Clearance, Resource Mobilization Department, Washington, D.C., June. IDA. 2007d. IDA's Performance-Based Allocation System: Simplification of the Formula and Other Outstanding Issues, World Bank Resource Mobilization Department , Washington, D.C., September. IDA. 2008. "IDA: The Platform for Achieving Results at the Country Level." Additions to IDA Resources: the Fifteenth Replenishment. IDA. 2009a. "IDA's Exceptional Allocation: A Review of the Implementation Experience with Lengthened Phase Out", IDA15 Mid-Term Review Paper, Washington, DC. October. IDA. 2009b. "IDA's Support to Fragile and Conflict-Affected Countries: Progress Report 2007- 09", IDA15 Mid-Term Review paper, Washington, DC. . October. IDA. 2009c. "IDA Performance Based Allocation and Development Results: An Update", IDA15 Mid Term Review Paper. October. IDA. 2009d. "Chairperson's Summary: IDA15 Mid-Term Review Meeting", Washington, DC, November 18-20. IDA. 2010. "Chairperson's Summary: IDA16 First Meeting", Paris, March 2-3. IEG. 2009. The World Bank's Country Policy and Institutional Assessment (CPIA): An Evaluation, July. - 29 - Annex 1. Evolution of the PBA formula IDA's performance-based allocation (PBA) formula has evolved over time (see Table A1). In particular, the emphasis on governance has evolved considerably because of its importance for improving the development performance of partner countries and for mitigating fiduciary risks to aid funds. Table A1: Evolution of IDA's Performance-Based Allocation (PBA) Formula and the Adjustment for Governance Country Performance Rating Elements in the performance- (CPR) Governance adjustments in IDA Period based allocation formula1/ Portfolio allocation formula CPIA performance rating 2/ 1991-1996 GNIPC-0.25, CPR1.8 100% 0% N.A. GNIPC-0.125, CPR0.5 CPR < 2 2 < CPR < GNIPC-0.125, CPR1.6 1997 2.9 100% -- N.A. GNIPC-0.125, CPR >2.9 CPR1.95 GNIPC-0.125, Governance discount (introduced in CPR < 3 (CPR/3)1.75 FY00): For countries with 3 or more ratings of 2 or below out of the 6 CPIA governance 1998-2000 80% 20% criteria, and over 30% of projects with GNIPC-0.125, CPR2 CPR > 3 deficient procurement practices (according to the ARPP rating), the CPR is cut by one-third. 2001-2003: governance rating = average of 7 governance criteria (6 CPIA criteria plus procurement criterion of the ARPP CPR = (0.8*CPIA + portfolio rating). 0.2*ARPP) *governance 2004-2007: governance rating = average factor 2001-2008 GNIPC-0.125, CPR2 of 6 governance criteria (5 CPIA criteria plus a 3-year moving average of the where governance factor = procurement flag of the ARPP portfolio (governance rating/3.5)1.5 rating). FY08: governance rating = average of 5 governance criteria in CPIA CPR = 0.24*CPIAA-C + From -0.125 5 0.68*CPIAD governance rating = average of 5 GNIPC , CPR FY09 + 0.08*Portfolio governance criteria in CPIA performance rating 3/ Notes: 1/ GNPPC = Gross National Income Per Capita; CPR=Country Performance Rating; ARPP = Annual Review of Portfolio Performance. 2/ From 1998 through FY07, this was represented by the projects at risk rating in the ARPP. Projects at risk consist of actual and potential problem projects. 3/ CPIAA-C refers to CPIA clusters A (economic management), B (structural policies), and C (policies for social inclusion/equity), and CPIAD refers to CPIA cluster D (public sector management and institutions). Source: IEG (2009) Beginning in 1998 with IDA12, and in response to IDA Deputies' suggestions, IDA allocations have been adjusted by the country's performance in the governance-related CPIA criteria. The adjustment took into - 30 - account the ratings of the CPIA governance criteria and that of the procurement criterion of the Annual Review of Project Performance (ARPP). This was referred to as a "governance discount". In 2001, the "governance discount" was replaced by the "governance factor" to address the problem of discontinuity in allocations under the governance discount. However, adjusting the country performance rating by the governance factor rendered the allocation formula more complex. The CPIA governance cluster and the procurement flag from the ARPP were double-counted; and, the exponential multiplier (of 1.5) on the governance rating (to arrive at the governance factor) made the calculation and interpretation of the performance rating more complex. In 2004, the number of CPIA governance criteria included in the governance factor fell from 6 to 5, as the 2004 restructuring of CPIA had removed one of the governance criteria. This reduced the effective weight of the governance criteria in the country performance rating from 68 to 66 percent. In the light of the complexity of the allocation formula and at the Mid-Term Review of IDA14, IDA Deputies requested Bank management to "...simplify the allocation formula and reduce unwarranted ratings volatility."44 Beginning with IDA15 (FY09), the country performance rating has been simplified to make the weights of the components more explicit. CPR = (0.24 * CPIAA-C + 0.68 * CPIAD + 0.08* portfolio performance) where CPIAA-C refers to the average of the ratings of CPIA clusters A (economic management), B (structural policies) and C (policies for social inclusion/equity), and CPIAD refers to ratings of CPIA cluster D (public sector management and institutions). Correspondingly , the PBA formula for IDA15 was: PBA = f(CPR5, population, GNIpc-0.125) where GNIpc is gross national income per capita. These changes have made the IDA allocation formula more transparent. 44 IDA (2006), "Chairman's Summary: IDA14 Mid-Term Review Meeting." November. - 31 - Annex 2. The PBA Formula during IDA15 and Detailed Allocation Steps The specific PBA formula during IDA15 can be written in full as: Annual Country i Allocation = SDR 1.5 million + PBAi Where : (Country Performanc e Rating i )5 * Popi * (GNI per capita i )- 0.125 (i ) PBAi = n *Annual IDA Envelope [(Country Performanc e Rating ) * Pop * (GNI per capita ) i =1 i 5 i i - 0.125 ] (ii ) Country Performanc e Rating i = (0.24 * CPIA A - C + 0.68 * CPIAD + 0.08 * PORT ) and i = country and n = number of IDA eligible countries . This formula is applied in six steps to determine the country-by-country allocation of IDA resources and IDA's terms of support (i.e. credit, grants, or a mix): Stage One: Allocate Volumes According to the PBA. The PBA formula is used to determine the volume of IDA resources at the country level. The three-year envelope ­ with a firm allocation for the first year and indicative allocations for the last two years ­ feed into IDA Country Assistance Strategies (CASs). Stage Two: Determine Eligibility for Grants on the Basis of Debt-Distress Risk. Eligibility for grants is limited to IDA-only countries rated at medium or high risk of debt distress or in debt distress, on the basis of the mechanism described below. IBRD/IDA blend countries and gap countries are not eligible for grants, regardless of their risk of debt distress. Debt distress risk ratings are grounded in the joint Bank-Fund Debt Sustainability Framework (DSF) that contains two pillars: (i) a `static' pillar ­ a matrix of policy-dependent external debt burden thresholds; and (ii) a `dynamic' pillar ­ an assessment of debt distress risk based on forward-looking DSAs. As endorsed by IDA Deputies at the IDA15 meetings, IDA grant eligibility is based on the risk of debt distress emerging from DSAs under the `dynamic' pillar of the DSF. When DSAs are unavailable, grant eligibility still relies on the `static' pillar of the DSF ­ which involves a "static" comparison between its latest available relevant external debt indicators and the applicable external debt thresholds from the DSF. Debt-distress risk classifications are then be mapped to a "traffic light" system, on the basis of which a country's terms of assistance (i.e. grant and credit mix) are determined: Debt-distress risk classification Traffic Light Grant and credit assignment Low risk of debt distress. "Green light" 100 percent credits. Moderate risk of debt distress. "Yellow light" 50 percent credits, 50 percent High risk of debt distress or in debt "Red light" grants 100 percent grants. distress. Stage Three: Apply Volume Discounts on Grants. Grant allocations are subject to a 20 percent upfront reduction in volume.45 This volume reduction is aimed at achieving two distinct objectives: First, an incentive-related volume discount of 13 percent is applied to the grant allocations of all grant recipients (except for post-conflict countries eligible for exceptional IDA allocations) in order to help maintain the 45 Refer to Debt Sustainability and Financing Terms in IDA14: Further Considerations on Issues and Options. IDA/SecM2004-0779, December. - 32 - strength of IDA's incentive system in light of increased grant levels. Second, a charges-related volume discount of 7 percent46 is applied to all grant allocations, to help finance foregone charge income on IDA grants. The upfront volume reduction is therefore limited to only 7 percent in the case of post-conflict countries. Stage Four: Redistribute Resources Resulting from Volume Discounts. The resources created by the upfront volume discounts are then re-injected into the IDA pool, and redistributed back to eligible countries. The resources from the 13 percent incentive-related volume discount are redistributed to all IDA-only countries, excluding gap countries and those which are eligible for exceptional post-conflict allocations (given that this share of volume discount is not deducted from their allocation as described in the previous section). Terms of such reallocation are in accordance with countries' risk of debt distress. Amounts are allocated to eligible countries in accordance with the PBA system. For those countries receiving part or the totality of such additional resources as grants, no further volume discounting will take place. The resources from the 7 percent charges-related volume discount are made available on hard terms47 to blend countries with per capita income below the operational cut-off for IDA, but with an active IBRD lending program (India, Pakistan and Vietnam as of FY10). Annual hard-term country allocations are additional to regular PBA allocations and are determined in proportion to a country's PBA allocation. Stage Five: Apply MDRI "Netting-Out" Mechanism to Eligible Countries For MDRI recipient countries, annual MDRI debt relief is deducted from the allocation amounts they would otherwise receive from steps one through four above. Stage Six: Distribute MDRI-related Compensatory Resources to All IDA-Only Countries. Donor resources provided to IDA to compensate for debt service forgone as a result of the MDRI are reallocated to all IDA-only, non-gap, countries. The compensatory resources are distributed to countries in accordance with the PBA system. For those countries receiving part or whole of such additional resources as grants, no further volume discounting will take place.Finally, it is important to note that the application of Stages 2-4 is limited to IDA-only countries. All blend and gap48 countries receive their respective allocations in the form of credits, irrespective of their risk of debt-distress. Stage 5 is applicable to MDRI-eligible countries only. Stage 6 is applicable to all IDA-only countries. Stages 4 and 6 do not apply to countries whose allocations were independently determined or to inactive countries. 46 Foregone charge income equals 7 percent of the face value of a standard IDA credit. 47 For details, refer to IDA (2005), "Implementation Note: Lending at "Hard Terms" in IDA14." April. 48 Gap or hardened-term countries are IDA-eligible countries whose per capita incomes are above IDA's operational cutoff for more than two consecutive years but are not creditworthy for IBRD lending. In FY10 such countries included Angola, Armenia, Azerbaijan, Bosnia-Herzegovina, Bhutan, The Republic of Congo, Georgia, Guyana, Honduras, Moldova and Sri Lanka. - 33 - Annex 3. CPIA Criteria and Portfolio Performance Rating Method CPIA Criteria A. Economic Management 1. Macroeconomic Management 2. Fiscal Policy 3. Debt Policy B. Structural Policies 4. Trade 5. Financial Sector 6. Business Regulatory Environment C. Policies for Social Inclusion 7. Gender Equality 8. Equity of Public Resource Use 9. Building Human Resources 10. Social Protection and Labor 11. Policies and Institutions for Environmental Sustainability D. Public Sector Management and Institutions 12. Property Rights and Rule-based Governance 13. Quality of Budgetary and Financial Management 14. Efficiency of Revenue Mobilization 15. Quality of Public Administration 16. Transparency, Accountability, and Corruption in the Public Sector Portfolio Performance Rating Method IDA's portfolio performance rating is constructed from the Bank's Annual Review of Portfolio Performance (ARPP) in three steps. First, all problem projects are identified, and averaged over four quarters. Second, the percentage of problem projects in the total IDA country portfolio is adjusted for the average age of the entire IDA portfolio, as countries with young portfolio tend to have fewer problem projects simply because it is too early for problems to surface. For each year that a country's portfolio is younger than the average age, the percentage of projects at risk is increased by five percent. Finally, the resulting percentage of problem projects is converted into portfolio performance ratings using the following conversion scale: Projects at risk Rating 0% 4.5 1-4% 4.0 5-15% 3.5 16-26% 3.0 27-67% 2.5 68-100% 2.0 - 34 - Annex 4. Mechanics of IDA Post-Conflict Allocation Post-conflict (PC) allocation is conducted in four steps as described below. Step 1: Determine the norm allocation using the regular PBA system . ( ) = [ ( ) . ] where PBA is the PBA norm allocation and N is the number of IDA countries eligible for allocation. Step 2: Calculate the notional post-conflict envelope based on Post-Conflict Performance Index (PCPI) and population of the eligible countries = [ ] where the notional PC allocation per capita for an eligible PC country is determined by its PCPI and using a conversion table (Table 1) below. The notional allocation for the country is then obtained by multiplying the per capita allocation and its population. Summing up notional allocations for all eligible countries would then yield the notional PC envelope. This notional envelope is then deducted from the total IDA resource envelope. Step 3: Determine the actual post-conflict allocation based on Post-Conflict Performance Rating (PCPR), population, and GNI per capita relative to all post-conflict countries = (0.8 + 0.2 Portfolio rating ) ( ) . = [ ( ) . ] where PCPBA is the post-conflict performance based allocation and M is the total number of post-conflict countries; and Step 4: Apply a phase-out (proportional reduction) for countries in the phase-out period = +( - )*(1 - /7) where PCPBA is the post-conflict performance based allocation, PBA is the PBA norm allocation (step 1 above), and PY refers to the number of years (from 1 to 6) in the phase-out period. The phase-out formula ensures that the allocation is reduced gradually by 1/7 during each of the 6 phase-out years to support a smooth transition to regular IDA allocation, so that when a country exits the PC window in the 7th year after the phase-out period, its allocation is phased down to its regular PBA allocation (as in step 1). The re-engagement allocations are determined following similar steps, but the notional per capita allocation is half of that for post conflict countries with the same PCPI. Table 1. PCPI and Notional Post-Conflict Allocations per Capita Allocation maximum PCPI score (US$ per capita per annum) 2.0 to 2.5 3.4 2.5 to 3.0 6.0 3.0 to 3.5 8.5 3.5 to 4.0 11.9 4.0 to 4.5 14.4 4.5 to 5.0 17.0 Source: IDA (2007b). - 35 - Annex 5. Estimated IDA15 Country Allocation Results by Allocation Steps Step 1 Step 2 Step 3 Step 4 Step 5 Step 6 Country Gross Grant Apply Redistribute Apply Redistribute Final allocation share volume resources from MDRI MDRI-related allocation before discount on grant discount netting compensatory grant grants out resources discount Capped countries India 2921.7 0% 0.0 133.7 0.0 0.0 3055.4 Pakistan 1859.3 0% 0.0 85.1 0.0 0.0 1944.4 Regular PBA countries Armenia 106.4 0% 0.0 0.0 0.0 0.0 106.4 Azerbaijan 140.6 0% 0.0 0.0 0.0 0.0 140.6 Bangladesh 2347.3 0% 0.0 60.6 0.0 155.5 2563.4 Benin 173.1 50% -17.3 4.3 -30.6 11.1 140.7 Bhutan 25.0 0% -0.8 0.2 0.0 0.4 24.7 Bolivia 178.1 0% 0.0 0.0 -56.5 0.0 121.6 Bosnia-Herzegovina 67.2 0% 0.0 0.0 0.0 0.0 67.2 Burkina Faso 411.9 100% -69.7 10.5 -25.3 27.1 354.5 Cambodia 139.9 50% -14.0 3.5 0.0 9.0 138.5 Cameroon 210.1 0% 0.0 5.3 -32.6 13.5 196.3 Cape Verde 24.6 0% 0.0 0.2 0.0 0.3 25.1 Chad 40.0 50% -5.5 0.9 -3.6 2.3 34.2 Comoros 6.8 100% -1.4 0.1 0.0 0.2 5.7 Djibouti 12.5 100% -2.5 0.2 0.0 0.5 10.7 Dominica 4.3 0% 0.0 0.0 0.0 0.0 4.3 Ethiopia 1610.0 50% -161.0 41.3 -33.9 105.7 1562.2 Gambia, The 24.5 100% -4.9 0.5 -8.8 1.3 12.6 Georgia 156.4 0% 0.0 0.0 0.0 0.0 156.4 Ghana 911.1 0% 0.0 23.4 -116.8 59.9 877.5 Grenada 6.3 0% 0.0 0.0 0.0 0.0 6.3 Guinea 79.7 100% -13.2 1.9 -7.7 5.0 65.8 Guinea-Bissau 14.7 100% -2.5 0.3 -1.4 0.7 11.9 Guyana 14.6 0% -0.4 0.1 -6.1 0.2 8.3 Honduras 147.1 0% 0.0 0.0 -37.2 0.0 109.9 Kenya 750.8 0% 0.0 19.2 0.0 49.2 819.4 Kiribati 5.6 0% 0.0 0.0 0.0 0.1 5.7 Kosovo 30.5 50% -3.0 0.7 0.0 2.0 30.1 Kyrgyz Republic 85.4 50% -8.5 2.1 0.0 5.4 84.4 Lao PDR 74.2 100% -14.8 1.8 0.0 4.7 65.9 Lesotho 45.6 50% -4.6 1.1 0.0 2.7 44.9 Madagascar 535.8 0% 0.0 13.7 -78.5 35.1 506.1 Malawi 317.2 50% -31.7 8.1 -25.8 20.7 288.4 Maldives 10.4 0% 0.0 0.2 0.0 0.4 11.0 Mali 296.4 0% 0.0 7.5 -50.5 19.0 272.3 Mauritania 45.6 50% -4.6 1.1 -17.6 2.7 27.1 Moldova 89.6 0% 0.0 0.6 0.0 1.4 91.6 Mongolia 48.9 0% 0.0 1.1 0.0 2.9 52.9 Mozambique 472.0 0% 0.0 12.1 -58.4 31.1 456.8 Nepal 430.9 50% -43.1 10.9 0.0 27.8 426.6 Nicaragua 126.0 50% -12.6 3.1 -19.0 8.1 105.6 Niger 245.2 50% -24.5 6.2 -19.2 16.0 223.8 Nigeria 1826.9 0% 0.0 46.9 0.0 119.9 1993.8 - 36 - Step 1 Step 2 Step 3 Step 4 Step 5 Step 6 Country Gross Grant Apply Redistribute Apply Redistribute Final allocation share volume resources from MDRI MDRI-related allocation before discount on grant discount netting compensatory grant grants out resources discount Papua New Guinea 75.5 0% 0.0 0.0 0.0 0.0 75.5 Rwanda 268.9 50% -36.5 6.8 -6.8 17.3 249.7 Samoa 11.1 0% 0.0 0.2 0.0 0.4 11.1 Sao Tome and Principe 6.6 100% -1.3 0.1 -0.8 0.1 4.6 Senegal 258.9 0% 0.0 6.5 -78.1 16.6 203.9 Sierra Leone 54.6 50% -5.5 1.3 -7.9 3.3 45.7 Solomon Islands 7.2 100% -1.4 0.1 0.0 0.2 6.0 Sri Lanka 326.4 0% 0.0 0.0 0.0 0.0 326.4 St. Lucia 9.9 0% 0.0 0.0 0.0 0.0 9.9 St. Vincent and the 6.7 0% 0.0 0.0 0.0 0.0 6.7 Gren. Tajikistan 63.2 100% -12.6 1.5 0.0 3.8 55.9 Tanzania 1200.7 0% 0.0 30.7 -105.0 78.2 1204.7 Tonga 5.9 100% -1.2 0.0 0.0 0.1 4.9 Uganda 792.3 0% 0.0 20.3 -111.4 51.9 753.1 Uzbekistan 237.6 0% 0.0 0.0 0.0 0.0 237.6 Vanuatu 7.6 0% 0.0 0.1 0.0 0.2 7.9 Vietnam 2280.0 0% 0.0 123.2 0.0 36.9 2440.1 Yemen, Republic of 279.4 100% -55.9 7.1 0.0 18.3 249.0 Zambia 207.0 0% 0.0 5.2 -50.4 13.5 175.4 Post-conflict and re-engaging countries Afghanistan 376.3 100% -23.0 0.0 -1.0 6.9 359.2 Angola 195.1 0% 0.0 0.0 0.0 0.0 195.1 Burundi 162.2 100% -11.4 0.0 -3.5 4.8 152.2 Congo, DR 758.9 100% -45.9 0.0 -9.5 20.3 723.8 Congo, Republic of 43.9 0% -1.7 0.0 -3.3 0.4 39.3 Cote d'Ivoire 282.3 100% -19.8 0.0 0.0 5.9 268.4 Eritrea 44.7 100% -3.1 0.0 0.0 1.9 43.4 Liberia 100.4 100% -5.8 0.0 -2.5 1.4 93.5 Sudan 0.0 0% 0.0 0.0 0.0 0.0 0.0 Timor-Leste 14.3 0% -5.9 0.0 0.0 0.4 8.8 Central African Republic 37.7 100% -7.5 0.5 -6.6 1.3 25.4 Haiti 77.9 100% -15.6 1.3 -14.7 3.5 52.4 Togo 98.1 100% -19.6 0.6 0.0 1.7 80.8 Zimbabwe 0.0 0% 0.0 0.0 0.0 0.0 0.0 IDA total 25361.5 -714.2 714.2 -1030.9 1030.9 25361.5 Capped countries 4781.0 0 218.8 0 0 4999.8 Regular PBA countries -555.0 492.8 -989.8 982.5 18319.3 Post-conflict and re-engaging countries -159.2 2.5 -41.1 48.4 2042.3 Note: The IDA allocation figures exclude allocations for regional projects, arrears clearance, disaster-related exceptional allocation, and allocations under the Pilot Crisis Response Window. Also, MDRI netting out is kept at 30% of gross allocation in FY11. Finally, hard-term credits are included in the country allocations to India, Pakistan, and Vietnam. - 37 - Annex 6. IDA Resource Management Within a Replenishment Period IDA allocates its resources annually based on the PBA system. Since country performance and grant eligibility are assessed every year, firm allocations are provided only for the upcoming fiscal year and indicative allocations are provided for the outer years to facilitate planning at the country level.49 The IDA resource management has built-in flexibilities within a replenishment period to ensure effective utilization of IDA resources based on country circumstances and operational realities on the ground. Front and back loading of allocations. Clear guidelines are established in advance of each fiscal year on front and back loading of resources. Under normal circumstances, limited front-loading of allocations in the first two years of a replenishment period of up to 30 percent is possible. However, because of the unusual circumstance of the global financial crisis and its impact on many low income countries, the front-loading rule was relaxed in FY09-10 to allow countries to frontload up to 50 percent of their annual allocation for programs or projects that respond to the impact of the global financial crisis. Small countries (with populations of less than 1.5 million) can frontload up to about 80 percent of their replenishment allocation in light of their small allocations. Capped countries (India and Pakistan) have no limits on front-loading but they have to stay within the three-year cap on allocations. Resources can be back-loaded in the first two years of the replenishment period. The flexibility in front- and back-loading means that, in any given year, commitments may exceed PBA allocations for some countries (front- loading) while commitments may be lower than the PBA allocations (back-loading) for others. However, country commitments are required to stay within their indicative three-year cumulative PBA envelope for the replenishment period as a whole. Reallocation. In advance of the final year of a replenishment period, if it becomes evident that the allocated resources will not be committed in any country, then these dormant funds are returned to the IDA pool and redistributed to other countries using the PBA system. In addition, if a region identifies more dormant funds during the course of the final year of a replenishment period, then it could reallocate up to 10 percent of the total regional PBA allocations from countries where IDA allocations cannot be absorbed to better-performing countries with higher absorptive capacity. 49 Allocations for the outer years are subject to a number of uncertainties including the country's performance, its relative performance compared to other countries, the size of the available IDA envelope, traffic lights, MDRI status, reactivation of countries, and other reasons. - 38 - Annex 7. Simulation Results of Including Vulnerability in the PBA Formula This section explores the allocation and redistribution impact of modifying the PBA formula to take into account structural vulnerability, as measured by an Economic Vulnerability Index (EVI) and a Human Assets Index (HAI). It has been argued that the introduction of these indices could provide a simplified and uniform allocation formula that could accommodate the varied needs of IDA countries (regular, post- conflict, capped IDA countries) as under the current PBA system but without segregating allocations into different country groups (Guillaumont 2009, Guillaumont 2008, and Guillaumont et al. 2010). This section presents allocations under alternative formulations in which the country performance rating is modified to include these indices, and compares such allocations to the current PBA allocations. EVI and HAI are two indices used by the UN to define the list of Least Developed Countries.50 EVI is designed to reflect the risk posed to a country's development by exogenous shocks and incorporates a combination of seven indicators: (i) population size, (ii) remoteness, (iii) merchandise export concentration, (iv) share of agriculture, forestry and fisheries in GDP, (v) homelessness owing to natural disasters, (vi) instability of agricultural production, and (vii) instability of exports of goods and services. HAI incorporates a combination of four indicators on health and education: (i) the percentage of population undernourished, (ii) under 5 child mortality, (iii) the gross secondary school enrolment ratio, and (iv) the adult literacy rate. The UN updates these indices every three years, and the last two updates were conducted in 2006 and 2009. Four alternative formulations of a modified Country Performance Rating (CPR) are considered: CPR Functional Form Simulation Modified CPR1 = 0.75 CPR+ 0.25 EVI Simulation 1, S1 Modified CPR2 = 0.50 CPR+ 0.50 EVI Simulation 2, S2 Modified CPR3 = 0.33CPR + 0.33 EVI + 0.33 HAI Simulation 3, S3 Modified CPR4 0.33CPR + 0.33 EVI + 0.33 HAI plus reduce the Simulation 4, S4 population exponent from 1 to 0.5 S1 and S2 examine the impact of raising the weight of the EVI; S3 further introduces HAI; while S4 explores whether the current capped allocations in the PBA system can be replaced by reducing the exponent of population in addition to the introduction of EVI and HAI. For example, Guillaumount et al. (2010) argue that introducing the vulnerability index can secure sufficient allocation to post-conflict and re-engaging countries with just one uniform formula; and a lower exponent of population can eliminate the need for base allocations and capped allocations in the current PBA system while achieving similar patterns of allocations as under the current system. These simulations (S1-S4) are compared to the allocation outcome under the current system, using the FY09 actual allocations data (before grant discount and MDRI netting out and excluding disaster-related exceptional allocations).51 The FY09 CPR has a correlation of -0.25 with EVI and a correlation of 0.41 with HAI. The remainder of this section summarizes the key findings of this exercise. 50 The component indicators are converted into the indices through a max-min procedure and rescaled to 1 to 100. For further details, please see http://www.un.org/esa/policy/devplan/profile/criteria.html. For inclusion in the modified CPR formula, the indices have been rescaled to 1 to 6. 51 The gross allocation before grant discount and MDRI netting out is most appropriate to serve as the baseline to compare the impact of vulnerability-related modifications without any other distortion. - 39 - Allocation impact of introducing vulnerability into the PBA formula · By country groupings, India's allocation--a capped country under the current system--would increase significantly, by between SDR310 million to SDR2.5 billion.52 On the other hand, most post-conflict and re-engaging countries would receive less than their current allocation under S1- S2. Only under S3, where the relative weight on vulnerability is significantly increased to two- thirds (and the relative weight of CPIA in performance kept at one-third), would these countries receive allocations comparable to their current levels. Even under S3, however, three countries would receive less than their current allocation (Côte d'Ivoire= -SDR34 million; Liberia=-SDR8 million and Republic of Congo= -SDR6 million relative to their actual FY09 allocation). Modifying the PBA to include vulnerability thus does not seem to serve the needs of post-conflict and re-engaging countries as well as the current system does. Figure A1. Baseline and Simulated FY09 Allocation by Group (SDRm) Baseline S1 S2 S3 S4 7,000 6,000 5,000 4,000 3,000 2,000 1,000 0 Capped countries Regular PBA Post-conflict and countries re-engaging countries · By region, it will lead to significant redistributions across regions. The East Asia and Pacific region's allocation share would decline in all scenarios (relative to the current case). Sub - Saharan Africa would lose its share in S1 but steadily gain in all other scenarios; while South Asia would gain in S1 but steadily lose in other scenarios. The picture for the other regions is much more mixed. 52 In the current PBA system, India's allocation is capped below the level based on the PBA-formula; and thus moving to a uniform formula using S1 leads to significant increase of its allocation; in S2 to S4, the increase of removing the cap is offset by the higher weights given to EVI, HAI, and lower weights given to population. - 40 - Figure A2. Baseline and Simulated FY09 Allocation by Region (SDRm) Baseline S1 S2 S3 S4 (SDRm) Baseline S1 S2 S3 S4 7,000 800 700 6,000 600 5,000 500 4,000 400 3,000 300 2,000 200 1,000 100 0 0 Europe and Latin America and Middle East and Africa South Asia East Asia and the Central Asia the Caribbean Noth Africa Pacific Table A1. Regions with Allocation Gains/Losses S1 S2 S3 S4 Gains SAR AFR, ECA, LAC, AFR, MNA AFR, ECA, LAC, MNA MNA Losses AFR, EAP, ECA, SAR, EAP SAR, EAP, ECA, SAR, EAP LAC, MNA LAC · By performance quintile, the introduction of vulnerability would, as expected, reduce the performance orientation of the current PBA system. As the weight on vulnerability increases in the PBA formula, the per capita allocation to countries in the lower performance quintiles increases relative to the present system. In the extreme case of S4 where the exponent of population is reduced in addition to including vulnerability, the allocation per capita in the 4th quintile would exceed that of the top performers in the 1st quintile.53 53 The comparison is based on simple average across 14-15 countries in the respective quintile in each scenario. In S4, the higher per capita allocation for an average country in each quintile does not mean that all IDA countries get more allocation. The picture would be different if a population-weighted average was used instead. - 41 - Figure A3. Simulated FY09 Allocation by IDA CPR Quintile in 2007 1/ 2/ (SDR per capita) 30.0 25.0 Baseline 20.0 S1 15.0 S2 S3 10.0 S4 5.0 0.0 Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 (Top) (Bottom) 1/ CPR 2007 data were the most recent rating available at the time of FY09 allocations in June 2008; 2/ Simple averages of allocation in SDR per capita for the 14-15 countries in each quintile. Data issues and implementation challenges Modifying the PBA formula to include vulnerability will raise several data and implementation issues. · First, the vulnerability data are available only every 3-4 years, which limits variations in allocations within those years (except due to changes in CPIA). In addition, the official data at the UN website do not cover all IDA-eligible countries (Armenia, Azerbaijan, Bosnia-Herzegovina, Georgia, Kosovo, Kyrgyz Republic, Moldova, Tajikistan, and Uzbekistan). · Second, the data have significant time lags. For example, the 2009 HAI used the 2003-05 estimates on undernourishment data from FAO, about four years of lag and without capturing the food crisis in 2008. The significant time lag between allocation and vulnerability indicators (3 to 5 years) makes the allocation process very much backward looking. · Finally, the EVI data seem to fluctuate significantly over time. EVI data from the two most recent rounds­2006 and 2009­vary from a decrease of 30 percent (Afghanistan) to an increase of 58 percent (Bosnia-Herzegovina). In addition, some changes are difficult to explain in light of realities on the ground. Examples include significant decreases in Georgia (-17 percent), Guinea (-13 percent), Ethiopia (-12 percent) and significant increases of vulnerability in Azerbaijan (44 percent), Angola (32 percent), and Bolivia (23 percent). While this fluctuation might not be a significant problem to define the set of Least Developed Countries (LDC) based on thresholds, it would create significant allocation volatility if it were to be used in determining IDA allocations. - 42 - Table A2. Baseline and Simulated FY09 Gross Allocations with the Introduction of EVI 1/ (SDRm, unless noted otherwise) Baseline Simulated gross allocation Change from the baseline % change gross Country allocation S1 S2 S3 S4 S1 S2 S3 S4 S1 S2 S3 S4 Regular PBA countries Armenia 34.8 21.5 23.9 10.0 31.4 -13.3 -10.9 -24.8 -3.4 -38% -31% -71% -10% Azerbaijan 48.1 52.5 102.9 44.8 83.1 4.4 54.8 -3.2 35.0 9% 114% -7% 73% Bangladesh 727.9 318.3 218.1 265.4 114.3 -409.7 -509.8 -462.5 -613.7 -56% -70% -64% -84% Benin 56.9 45.7 66.2 69.1 124.6 -11.2 9.2 12.1 67.7 -20% 16% 21% 119% Bhutan 8.3 6.5 10.9 5.3 35.3 -1.8 2.6 -3.1 27.0 -22% 31% -37% 323% Bolivia 55.5 44.9 65.5 20.0 35.1 -10.6 10.0 -35.5 -20.4 -19% 18% -64% -37% Bosnia-Herzegovina 23.0 13.7 14.9 5.6 15.3 -9.3 -8.1 -17.4 -7.7 -41% -35% -76% -33% Burkina Faso 127.0 100.8 141.9 171.3 241.6 -26.2 14.9 44.3 114.6 -21% 12% 35% 90% Cambodia 40.4 63.0 168.2 93.6 133.5 22.6 127.8 53.2 93.1 56% 316% 132% 230% Cameroon 73.5 43.5 44.8 54.7 68.9 -30.0 -28.8 -18.8 -4.7 -41% -39% -26% -6% Cape Verde 8.1 5.2 7.2 1.7 13.0 -2.9 -0.9 -6.3 4.9 -36% -11% -78% 61% Chad 14.5 25.9 80.9 151.4 250.1 11.4 66.4 136.8 235.6 78% 457% 941% 1620% Comoros 2.3 1.7 5.8 4.5 30.9 -0.6 3.5 2.2 28.7 -25% 154% 99% 1265% Djibouti 4.3 3.5 7.7 6.6 39.1 -0.8 3.4 2.3 34.8 -18% 79% 54% 814% Dominica 1.4 0.7 1.3 0.2 3.7 -0.8 -0.2 -1.2 2.3 -53% -12% -87% 159% Ethiopia 542.1 313.4 306.9 619.6 377.7 -228.7 -235.2 77.5 -164.4 -42% -43% 14% -30% Gambia, The 8.7 10.5 25.8 21.1 87.5 1.8 17.1 12.4 78.8 21% 196% 142% 904% Georgia 47.8 33.9 43.4 36.5 94.4 -13.8 -4.3 -11.3 46.6 -29% -9% -24% 98% Ghana 297.0 219.1 283.5 130.5 146.0 -78.0 -13.6 -166.6 -151.0 -26% -5% -56% -51% Grenada 2.1 0.6 0.8 0.2 2.8 -1.5 -1.3 -2.0 0.6 -72% -63% -92% 31% Guinea 24.9 13.7 13.7 29.9 53.0 -11.2 -11.3 5.0 28.1 -45% -45% 20% 113% Guinea-Bissau 4.9 7.6 25.6 26.9 112.4 2.6 20.6 21.9 107.4 54% 417% 443% 2172% Guyana 4.5 3.3 6.5 1.1 7.1 -1.1 2.1 -3.3 2.6 -25% 46% -75% 58% Honduras 53.2 33.8 38.0 12.4 25.3 -19.4 -15.1 -40.8 -27.9 -36% -28% -77% -52% Kenya 247.3 85.2 41.7 48.3 42.8 -162.1 -205.6 -199.0 -204.6 -66% -83% -80% -83% Kiribati 1.9 1.0 3.7 0.7 11.8 -0.9 1.8 -1.2 9.9 -47% 94% -63% 520% Kosovo 9.7 6.6 8.3 2.6 10.0 -3.1 -1.5 -7.1 0.3 -32% -15% -73% 3% Kyrgyz Republic 26.3 28.7 57.5 21.5 50.9 2.4 31.1 -4.8 24.6 9% 118% -18% 93% Lao PDR 18.8 31.2 90.4 41.1 92.1 12.4 71.6 22.3 73.3 66% 380% 119% 390% Lesotho 14.6 13.2 23.4 11.3 43.1 -1.4 8.8 -3.3 28.5 -9% 60% -23% 195% Madagascar 173.6 112.1 125.7 122.9 150.2 -61.5 -47.8 -50.7 -23.4 -35% -28% -29% -13% Malawi 103.4 125.1 264.6 185.5 269.5 21.7 161.1 82.0 166.1 21% 156% 79% 161% Maldives 3.5 2.5 5.3 1.0 10.1 -1.0 1.7 -2.5 6.6 -29% 50% -71% 188% Mali 111.0 82.7 109.3 137.1 211.6 -28.3 -1.8 26.0 100.5 -25% -2% 23% 91% Mauritania 14.9 13.9 25.3 16.8 51.5 -1.0 10.4 1.9 36.5 -7% 70% 12% 245% Moldova 27.3 18.0 21.7 7.7 22.7 -9.3 -5.6 -19.6 -4.7 -34% -20% -72% -17% Mongolia 19.0 18.9 35.7 9.0 30.2 -0.1 16.7 -10.0 11.2 0% 88% -53% 59% Mozambique 145.4 143.9 252.2 334.2 391.9 -1.5 106.8 188.8 246.5 -1% 73% 130% 170% Nepal 160.5 100.2 108.2 79.8 81.5 -60.3 -52.3 -80.7 -79.0 -38% -33% -50% -49% Nicaragua 39.8 31.0 44.1 14.8 33.9 -8.8 4.3 -25.0 -5.9 -22% 11% -63% -15% Niger 73.3 71.0 123.0 208.1 299.5 -2.3 49.8 134.9 226.2 -3% 68% 184% 309% Nigeria 618.2 549.4 860.3 707.1 314.9 -68.7 242.2 88.9 -303.2 -11% 39% 14% -49% - 43 - Baseline Simulated gross allocation Change from the baseline % change gross Country allocation S1 S2 S3 S4 S1 S2 S3 S4 S1 S2 S3 S4 Papua New Guinea 25.4 23.6 40.8 29.0 62.5 -1.8 15.4 3.6 37.1 -7% 61% 14% 146% Rwanda 96.0 103.4 197.3 189.3 328.9 7.4 101.3 93.3 232.9 8% 106% 97% 243% Samoa 3.6 2.7 5.8 1.0 11.8 -0.9 2.2 -2.7 8.2 -26% 60% -74% 227% Sao Tome and Principe 2.1 0.9 2.0 0.7 9.5 -1.3 -0.1 -1.5 7.3 -59% -6% -68% 342% Senegal 94.8 62.0 71.2 80.4 123.7 -32.8 -23.6 -14.4 28.9 -35% -25% -15% 30% Sierra Leone 20.6 25.2 56.7 97.8 220.3 4.6 36.2 77.3 199.8 22% 176% 376% 972% Solomon Islands 2.4 1.8 5.6 2.6 19.7 -0.6 3.2 0.1 17.3 -25% 133% 6% 717% Sri Lanka 104.2 60.8 61.1 14.5 17.6 -43.4 -43.1 -89.8 -86.7 -42% -41% -86% -83% St. Lucia 3.3 1.5 2.3 0.4 4.9 -1.7 -0.9 -2.9 1.7 -53% -29% -89% 51% St. Vin. and the Gre. 2.4 0.9 1.2 0.3 4.1 -1.5 -1.1 -2.1 1.7 -64% -48% -89% 72% Tajikistan 22.4 27.6 62.0 37.1 77.4 5.1 39.6 14.7 55.0 23% 176% 65% 245% Tanzania 430.2 218.6 182.8 229.7 198.7 -211.6 -247.4 -200.5 -231.5 -49% -58% -47% -54% Tonga 1.8 0.7 2.4 0.3 5.9 -1.1 0.6 -1.5 4.0 -60% 32% -81% 220% Uganda 265.0 272.2 492.2 311.2 303.4 7.3 227.2 46.3 38.4 3% 86% 17% 15% Uzbekistan 63.7 47.4 63.2 34.4 36.0 -16.3 -0.5 -29.2 -27.7 -26% -1% -46% -43% Vanuatu 2.4 1.5 4.1 1.3 14.8 -0.9 1.7 -1.1 12.4 -38% 71% -46% 521% Vietnam 702.3 316.8 226.4 63.2 37.1 -385.5 -475.9 -639.2 -665.2 -55% -68% -91% -95% Yemen, Republic of 82.0 81.3 143.9 109.8 125.8 -0.7 62.0 27.8 43.9 -1% 76% 34% 54% Zambia 63.1 73.0 149.8 131.9 207.1 9.9 86.7 68.8 144.0 16% 138% 109% 228% Capped countries India 969.7 3531.1 1551.4 1298.2 210.0 2561.4 581.7 328.5 -759.7 264% 60% 34% -78% Pakistan 617.1 374.8 231.5 313.2 133.2 -242.3 -385.6 -303.9 -483.9 -39% -62% -49% -78% Post-conflict and re-engaging countries Afghanistan 164.4 44.0 87.6 294.9 287.4 -120.4 -76.7 130.5 123.0 -73% -47% 79% 75% Angola 67.9 36.4 93.0 155.2 204.0 -31.5 25.1 87.3 136.0 -46% 37% 128% 200% Burundi 63.6 42.8 122.5 182.0 338.4 -20.8 58.9 118.3 274.8 -33% 93% 186% 432% Central African Republic 12.1 8.2 19.3 37.1 96.6 -3.9 7.2 25.0 84.5 -32% 60% 207% 699% Congo, DR 295.5 160.8 427.3 819.8 568.2 -134.8 131.8 524.3 272.7 -46% 45% 177% 92% Congo, Republic of 18.2 9.2 21.4 11.8 32.9 -9.0 3.2 -6.5 14.7 -50% 17% -35% 80% Cote d'Ivoire 90.0 20.6 27.1 52.0 64.2 -69.5 -62.9 -38.1 -25.8 -77% -70% -42% -29% Eritrea 15.9 17.3 51.9 57.1 140.7 1.4 36.0 41.2 124.8 9% 227% 260% 787% Haiti 31.7 27.2 74.6 79.6 139.2 -4.5 42.9 47.9 107.5 -14% 136% 151% 339% Liberia 31.6 0.2 9.7 21.8 61.0 -31.4 -21.9 -9.8 29.4 -99% -69% -31% 93% Timor-Leste 4.9 3.3 10.0 6.4 33.4 -1.7 5.0 1.5 28.5 -34% 103% 30% 581% Togo 35.1 9.6 22.4 29.6 62.6 -25.5 -12.7 -5.5 27.5 -73% -36% -16% 78% IDA total 8425.2 8425.2 8425.2 8425.2 8425.2 0 0 0 0 0 0 0 0 Regular PBA countries 6007.5 4140.0 5675.6 5066.6 6053.4 -1867.5 -332.0 -940.9 45.9 -31% -6% -16% 1% Capped countries 1586.8 3905.8 1782.9 1611.4 343.2 2319 196 25 -1244 146% 12% 2% -78% Post-conflict and re- engaging countries 830.9 379.4 966.7 1747.2 2028.6 -451 136 916 1198 -54% 16% 110% 144% 1/ Data refer to gross allocations before grant discount and MDRI netting out. - 44 - Annex 8. How African Development (AfDB) and Asian Development Bank (ADB) Determine Post Conflict Allocation and Phasing Out African Development Bank The African Development Fund has a special allocation system, known as Fragile States Facility (FSF). The FSF has three pillars: Pillar I supplementary financing, which allocates a times-two multiplier of AfDF-10 PBAs while continuing to link resources to preceding cycle performance. During AfDF11, nine countries (Burundi, Central African Republic, Comoros, Congo Democratic Republic, Côte d'Ivoire, Guinea-Bissau, Liberia, Sierra Leone and Togo) were eligible for FSF supplemental support. Pillar II arrears clearance financing, where funds are set aside to provide timely and adequate arrears clearance support to countries that may qualify for the FSF. Pillar III targeted support, where a limited pool of resources are allocated based on an analysis of eligible countries' demand for technical assistance and taking into consideration the amount of resources that are required for Pillars I and II. The FSF support is provided for two replenishment cycles. The AfDB is currently consulting with donors, in the context of AfDF-12, the possibility of beginning a phase out of the special allocation based on three simple parameters--CPIA (greater than 3.2), GNI per capita on PPP terms (greater than US$500), and length of re-engagement (greater than 6 years). If a country meets all the three parameters, its allocation will be reduced by 50 percent over a three-year period; if it meets only two, its allocation will be reduced by a third; and if it meets only one condition, its allocation will be phased down by 25 percent over a three-year period. Asian Development Bank The Asian Development Bank began its post-conflict allocation in 2002 with a disaster and emergency relief program in Afghanistan. Two post conflict countries are supported by ADB: Afghanistan and Timor-Leste. In the case of Afghanistan, ADB set aside US$400 million during 2005-08. During the ADF10 replenishment, donors agreed that the ADB system should be harmonized with that of IDA, with the same phasing out system. Since 2009, Afghanistan has been in the phase-out period. ADB has proposed several options (extending the phase-out by 2 years, extending the phase-out by 4 years, or suspending the phase- out) for discussion with donors to address Afghanistan's declining allocation. In the case of Timor-Leste, the ADB has provided concessional support, with up to 40 percent of the country's allocation provided as grants. The total indicative ADF allocation to Timor-Leste during 2009­ 2010 is estimated at US$25.7 million, of which about 30 percent is in grants. Source: African Development Bank (2010) and Asian Development Bank (2010).