41148 IDA15 IDA'SPerformance-BasedAllocation System: Simplificationofthe Formula and Other OutstandingIssues InternationalDevelopment Association Resource Mobilization (FRM) September 2007 Abbreviations and Acronyms ARPP Annual Review o f Portfolio Performance CPIA Country Policy and InstitutionalAssessment GNI Gross National Income H P C Heavily IndebtedPoor Countries IBRD International Bank for Reconstruction andDevelopment IDA International Development Association IEG IndependentEvaluationGroup LAC LatinAmerica and Caribbean MDRI Multilateral DebtRelief Initiative ODA Official Development Assistance PBA Performance-Based Allocation SDR Special Drawing Rights TABLEOFCONTENTS Executive Summary.......................................................................................................................... i I Introduction......................................................................................................................... . 1 I1. 1 A.Components o fthe Current PBA formula ..................................................................... IDA's Current PBA System................................................................................................ 1 B.Stressing Country Performanceand Accounting for Needs........................................... 4 C.Managing IDA Resources Within a ReplenishmentPeriod .......................................... 7 I11. Simplifyingthe PBAFormula............................................................................................ 8 Iv. Small States and IDA's Financial Assistance ................................................................... 11 A.Challenges Facedby Small States ............................................................................... 11 B.What do These Challenges Mean for Small States...................................................... 13 C.IDA's Role in Small States.......................................................................................... 13 D.IDA's Financing inSmall States.................................................................................. 14 E. Strengthening IDA's Financial Assistance to small states.......................................... 15 V. Regional Projects.............................................................................................................. 16 VI. Issues for Discussion ........................................................................................................ 18 Figures Figure 1 IDA Country Performance Rating. FY08 ....................................................................... . 2 Figure2: EstimatedFlow o f IDA Resources by Income Groups inIDA15 ................................... 5 Figure3: EstimatedFlows ofIDA Resources by Region inIDA15............................................... 6 Figure 4: Estimated Grant Element o f Regional Flows IncludingMDRIand HIPC Transfers in IDA15 ............................................................................................................................. 6 Figure 5: Projected Impact o f a Ceiling o f 20 percent on Country Contributions to Regional Projects.......................................................................................................................... 18 Tables Table 1: Estimated Flows o f IDA Resources by Eligibility Criteria in IDA14 .............................. 5 Table 2: Small States are. on Average. Subject to Greater Volatility than Other Countries ........ 13 Table 3: How Much InfrastructureDoes the Current MinimumAllocation Buy? ...................... 15 Annexes Annex 1: Evolutiono f the IDA PerformanceBased Allocation Formula (IDA 11-13) ............... 19 Annex 2: IDA15 Projections Under the Current PBA Formula and SimplificationOptions .......22 Annex 3: Simulated Impact o f a Floor and Base on Country Allocations .................................... 25 Annex 4: Pipeline of Regional Projects for the IDA15 Period..................................................... 28 Annex 5: Share o f Country Allocations to Finance Regional Projects in IDA15.......................... 29 EXECUTIVE SUMMARY 1. Background. Eachyear, IDA allocates the bulk o f its resources usingthe Performance-Based Allocation (PBA) formula. Modifications to this formula over the past several replenishments have helped direct more concessional resources to countries where results are beingachieved. However, these modifications have also made the formula more complex. 2. Objectives. This paper follows up on discussions relatedto the simplification o f the PBA formula that took place at the first IDA15 replenishment meetingheldinParis in March 2007.' As requested by the Deputies,this paper provides additional information on the impact o fproposed simplification options on allocations at the country level and will propose apreferred option. Apart from follow-up work from the Paris meeting, this paper also discusses two additional allocation-related issues: (i) strengtheningIDA'S financial support to small states; and (ii) enabling countries with small IDA allocations to participate inregional projects. 3. Simplification. After weighing both options -the geometric and additive formulas -to simplify the current PBA formula, management proposes usingthe additive formula going forward. The proposed formula is: Country Performance Rating= (0.24 * CPIA A-c+0.68 * CPIA D+0.08 * PORT) IDA countryallocation =f (Country performance rating 5.0,Population'.', GNI/~apita''.'~~) While both options simplify the current PBA formula considerably, maintain the current weights o f components, and track current allocations very closely, management proposes the additive formula because it is simpler and more transparent. It could be understood easily bypolicy makers inIDA countries, which is the aim o f the simplification exercise. 4. Small states. IDA-eligible small states, with populationsbelow 1.5 milli~n,~ face several challenges including highper capita costs o f production, lack o f diversification and openness leadingto higher vulnerability to shocks, and periodic recurrence o fnatural disasters which compound human, economic, and environmental costs. Giventhe structure o f their economies and the challenges they face, it i s all the more important for governments insmall states to react swiftly and flexibly to mitigate the impact o f shocks, implement structural reforms, provide infrastructure andpublic services, and generate effective information flows. 1 IDA (February 2007). "IDA s Performance-Based Allocation System: Optionsfor Simplifjiing the ' Formula and Reducing Volatility. '' Cluster D o f the CPIA assesses a country's public sector management and institutions. The other three clusters, A-C, assess economic management, structural policies, and policies for social inclusiodequity respectively. 3 The World Bank does not have a formal category o f small states. The Commonwealth Secretariat and the World Bank Joint Task Force defines small states as those with populations o f 1.5 million or less (see Small States: Meeting Challenges in the Global Economy, Commonwealth SecretariatlWorld Bank Joint Task Force report, April 2000). - .. 11- 5. IDA'soverallrole insmall stateswas discussed duringthe IDA14replenishment meetings4and in a report bythe IndependentEvaluation Group (IEG) in2006.5 The IEG report notedthat the scope and nature o fthe World Bank's activities insmall states should be "strategic and selective." The report also finds that the World Bank canplay a limited, but useful role inthese countries by : (i) helping governments coordinate and harmonize donor activities to maximize the development benefits o f Official Development Assistance (ODA); (ii) supporting cooperation among small states to address development issues on a regionalbasis, especially where there are asymmetries inresources, skills, andbargainingpower; (iii) preparing cross-country analytical work on issues o f common concern andwhere possible, tailoring them to country circumstances; and (iv) helpingharmonize donor practices to lower transactions costs. 6. Inadditionto providing a "platform" for making overallaidmore effective, IDA's role insmall states canbe enhancedthrough the learningthat comes from implementingprograms andprojects ina country.6 A review shows that IDA's financial assistanceto small states has eroded over the past replenishments. Since base allocations matter for small states, management proposes increasing them from SDRl.1millionper year (or SDR3.3 millionper replenishment period) to SDR1.5 million per year (or SDR4.5 millionper replenishment period) for all countries. This increase adjusts for inflation and, together with the increase agreed upon duringthe IDA14replenishment discussions, helpsmaintain the base allocation constant inreal terms since IDA9. Management fkrther proposes raising the cap on per capita allocations from SDR13.2 (US$20) to SDR19.8 (US$30), to also adjust for inflation since IDA9. Such an increase would also benefit the small states disproportionately and would be inline with the performance orientation o f IDA'SPBA system. This i s becausewell-performing small states are more likely to reachthe cap, andtherefore stand to benefit. Inaddition, some micro states with very small populationswill also benefit. 7. Regionalprojects. IDA's regional pilot program began inIDA13 and continues duringthe IDA14period. Demand for regionalprojectshasbeenstrong so far, especially inSub-SaharanAfrica, andthis strong demandis expected to continue to grow duringthe IDA15 period. Giventhe externalities associated with regional projects and the need to demonstrate ownership by participating countries, 1/3 o f the cost o f an eligible regional project i s charged to the country envelope to leverage 2/3 o f the cost from the regional envelope. While the 1/3:2/3 split betweenthe country allocation and topping-up fund has beenworking well, implementation experience has shown that there have beeninstances where country participationinregional projectswas constrained by a small IDA allocation. IDA (2004). "Supporting Small and Vulnerable States." IEG(2006). "Small States: Making the Most of Development Assistance. A Synthesis of WorldBank Evaluation Findings." IDA (2007). "The Role of IDA in the Global Aid Architecture: Supporting the Country-Based Development Model. " 8. Therefore, management proposes placing a cumulative ceiling on an individual country's cost o fparticipationinregional projects at 20 percent of its annual a~ocation.~ Hence, a country would finance its 1/3 share o f participation inregionalprojectsuntilthe cost ofparticipation reaches 20 percent o f its annual allocation, with the regional project envelope financing the country's costs beyondthis ceiling. While this provision applies to all countries, limitingthe costs borne by an individual country as a share o f its IDA allocationbyplacinga ceiling would benefit countries with small IDA allocations where this ceiling would likely bebinding. This i s corroborated by experience with regional project program so far andby an examination o f the pipeline o fproposed projects in IDA15. Moreover, such a ceilingwould disproportionately benefit small states with populations below 1.5 million, where regional integrationi s important. At the IDA15 Mid-Term Review, management will present a review o f the regional project program, report on experience with the scaling up, and propose adjustments to the proposed ceiling on country contributions ifnecessary. 9. Issuesfor discussion. Management proposes the following issues for discussion. 0 D o Deputiesagree that Option 11,the additive formula, i s the preferred option? 0 D o Deputiesagree with increasing the base allocations to all countries from SDRl.1millionperyear (or SDR3.3 millionperreplenishment) to SDRl.5 millionper year (or SDR4.5 million per replenishment) and the per capita cap on allocations from SDR13.2 (US$20) to SDR19.8 (or US$30) for the benefit o fthe small states? Do Deputies agree with a ceiling o f 20 percent on the country contributions to regionalprojects for the benefit o f countries with small IDA allocations? 7 After grants discount and Multilateral Debt ReliefInitiative (MDRI)netting out. x IDA's Performance-BasedAllocation System: Simplificationof the Formula and Other OutstandingIssues I.INTRODUCTION 1. Background. Each year, IDA allocates the bulk o f its resources usingthe Performance-Based Allocation (PBA) formula. The PBA formula, while emphasizing country performance as the main determinant o f IDA allocations, also takes needs into account. This formula was modified over the past several replenishment discussions to incorporate new analytical insights, evolving donor priorities, and lessons learned during implementation. While these modifications have helpeddirect more concessional resources to countries where results are being achieved,' they have made the formula more complex. 2. Objectives and Layout. This paper follows up on discussions relatingto the simplification o fthe PBA formula that took place at the first IDA15 replenishment meetingheld inParisinMarch2007 (Section III).9 by Deputies at this As requested meeting, this paper will provide additional informationon the impact o f the proposed simplification options on country allocations (Annex 2) and will recommend a preferred option. Apart from the follow-up work from the Paris meeting, this paper will also discuss two other allocation-related issues: (i) strengthening IDA's financial engagement insmall states (Section IV); and(ii) facilitating participationo f IDA countries with small allocations inregionalprojects (Section V). Finally, issues for discussion are presented inSectionVI. Butbeforegettinginto these topics, thispaperbeginswith abrief descriptiono f IDA's PBA system and a discussion o fhow IDA resources are managed duringthe replenishment period(Section 11). 11. IDA'SCURRENTPBASYSTEM A. Components of the current PBA Formula 3. ThePBA system. This section provides a short description o f the main components o f the PBA system.lo Inthis system, country performance ratings play a major role. Figure 1shows the three main components o fthe country performance ratings - the Country Policy and Institutional Assessment (CPIA), the portfolio performance ratings and the governance factor. Each year, a rigorous set o fpeer- reviewedCPIA ratings are prepared for all IDA countries, which assess the quality' o f a ' See IDA (March 2007). "Selectivity and Performance: IDA 's Country Assessment and Development Effectiveness.I' IDA (February 2007). "IDAs Performance-Based Allocution System: Optionsfor Simplibing the Formula and Reducing Volatility.'' This section is partly drawn from IDA (2006), "IDA'SPerformance-Based Allocation System: a review of the governancefactor. " "Quality" refers to how conducive that framework i s to fostering poverty reduction, sustainable growth, and effective use o f development assistance. - 2 - country's present policy and institutional framework. Portfolio performance ratings, as captured annually inthe Bank's Annual Review o f Portfolio Performance (ARPP), reflect the percentage o f actual IDA-fundedproblemprojectsineach country." The governance factor, calculated usingthe five indicators o f cluster D13o f the CPL4, places extra emphasis on public sector management and institutions ina country to mitigate fiduciary risks to aid funds.14 The CPLA, portfolio performance ratings and governance factor feed into the calculationo fthe country performanceratings, all o fwhich are now publicly disclosedby IDA. Figure I:IDA Country Performance Rating, FY08 v IDA Country Performance Rating B e g i m n gFY08, followmg a decision by the Deputies mthe first IDA15 meeting held inParis in March 2007, three changes were introduced m the calculation o f the portfolioperformance ratmgs to lower unwarranted volatility These are: (1) usmgactual problemprojects instead o f actualpZus potential problem projects,(ii)usinga quarterly average of actual problemprojects rather than an end- year snapshot; and (iii) usinga revised scale to convert percent o fproblem projects to a rating. Cluster D o fthe CPIA assesses a country's public sector management and institutions. The other three clusters, A-C, assesses economc management, structural policies, and policies for social inclusiodequity respectively. Note that the procurement rating i s no longer used to calculate the governance factor beginning FY08 as a consequence o fthe decision by the Deputies inthe IDA15 Paris meeting to drop potential problem projects from the calculabon o fthe portfolio performance ratings Potentialproblem projects are identifiedby 12 nsk flags - one o f which is the procurement flag. So dropping potential problem projects from portfolio ratings also meant that the procurement flag would no longer be included inthe governance factor. - 3 - 4. The equation for calculatingthe countryperformance rating for eachcountry is shown below. The weighted average o f the CPIA rating (80 percent) and the portfolio performance rating (20percent) is multipliedby the governance factor. Equation (I):Country Performance ratingformula Country performance rating = (0.8 CPIA + 0.2 portfolio performance rating) * * * governance factor 5. Thegovernancefactor. As seen inequation (l),governance factor is applied the as a multiplier. It is made up o f five criteria drawn fiom Cluster D o fthe CPIA ratings. The governance factor for each country is calculated as follows: Equation (2): Thegovernancefactor Governance Factor = (average rating of 5 governance criteria / 3.5)'.5 The averageratingis dividedby 3.5, which i s the mid-point ofthe CPIA scale, andthen raised to an exponent o f 1.5, forming the governance factor. So for governance scores above 3.5, the rating i s increased while for scores below 3.5, it i s decreased. 6. Other elementsof theformula -population and GNIper capita. Inaddition to the country performance ratings calculated as described above inequation 1, IDA also uses measures o fpopulationand GNIper capita in allocatingresources. Population affects allocations significantly -the relationship i s linear with the populationterm, whereby ahigherpopulation results inaproportionately increased allocation. Moreover, while all IDA countries arepoor, there is an additionalmodest biastowards countries with lower GNIper capita. Finally, each country also receives abase allocation o f SDR 1.1millionper year, which favors the small countries. Equation 3 shows that a country's performance is the dominant determinant o f IDA allocations - a score twice as high would result infour times the allocation, other things remainingconstant. Equation (3): PBAformula IDA country allocation per annum=base allocation + f (Country performancerating 2.0, Population'.O, GNVcapita-O.'25) 7. Two additional steps in arriving at IDA allocations. Once PBA allocations are determined usingthe above formula, two additional steps are requiredto arrive at a country's "final" allocation. First, grant allocations are discounted by 20 percent (9 percent for post-conflict c~untries)'~and 11percent o f this discounted amount i s reallocated to all IDA-onlycountries, excluding gap and post-conflict countries. Second, for countries eligible for debt cancellationunder the MDRIinitiative, the debt service due 1s Grandcreditcompositionis set based on a country's risk of debt distress. For details see IDA (2005). "Additions to IDA Resources: Fourteenth Replenishment. Working together to Achieve the Millennium Development Goals. I' - 4 - inthe relevant fiscal year is nettedout from that year's allocation. These netted-out amounts are then redistributedto IDA-onlycountries, excluding gap countries.'6 B. StressingCountry PerformanceandAccountingfor Needs 8. Performance is important. Country performance i s the maindeterminant o f IDA allocations. The stress on performance was put inplace duringthe past replenishments becauseresearch shows that it helps direct more resources to countries where results are beinga~hieved.'~ described above, the formula also giveshighweight to governance As becauseit i s important for the development performance o f a country as well as for mitigating fiduciary risks to aid funds.'* 9. And needs are addressedtoo. Inaddition to factoring incountry needs by includingpopulation and GNIper capita into the allocation formula, IDA's allocation system addresses country needs inthe following ways. Needsproxied by per capita income: IDA's eligibility criteria ensure that resources are allocated to meet the needs o fthe poorest countries. Infact, around 90 percent o fIDA's resources duringthe IDA14 period are estimated to go to countries withper capita incomes below IDA's operational cut-off, o fwhich 72 percent are estimated to meet the needs o f countries that are not creditworthy to sustain IBRD financing (Table 1). Further, even within this group ofpoor countries eligible for IDAresources, the PBA system with its ring-fenced exceptions directs funds to the poorest among them. For instance, ifallocations were purely drivenby the PBA formula, only 12 percent o f fundswould go to countries where people earn less than a dollar a day andonly 23 percent o fresources would go to countries earning less than two dollars a day (Figure 2). Bycapping allocations to some creditworthy countries (because they have access to other sources o f financing), providing exceptional allocations to some post-conflict countries to meet their extraordinary reconstruction needs, andby topping up funds for regional projects with positive externalities, allocations to countries where people earn less than a dollar a day would go up to 32 percent while allocations to countries earning less than two dollars a day would double to around 48 percent. 16 IDA (2005). "The Multilateral Debt Relief Initiative: Implementation Modalitiesfor IDA. " Also see IDA (March2007). "Selectivity and Performance: IDA 's Country Assessment and Development Effectiveness." 18 See IDA (October2007). "IDA's Performance-Based Allocation System: A Review of the Governance Factor. " and the 2006 GlobalMonitoringReport. - 5 - I N o Yes (Blends) 56 IDA-only countries that lack 2 Creditworthy blends below credit-worthiness to sustain IBRD IDA's operational cut-off Per capita financing, will receive an estimated (India and Palustan) have their income No 72 percento f IDA14 resources. allocations cappedand will aboveIDA'S receive an estimated 18 operational Dercent of IDA14resources. cut-off for 4 "Gap" counties I(Angola, 5 creditworthy counties morethan Georgia, Honduras, and Sri Lanka) (Albania, Armenia, Azerbaijan, two plus6 small islands4willreceive Bolivia, and Bosnia & Yes an estimated4 percento f IDA14 Her~egovina)~ Indonesia & resources. (allocation capped) plus4 small islands4will receive an estimated 6 percentof IDA14 resources. Notes: 1. Gap countries have GNIper capita above IDA'Soperational cut-off for more than two years but are not creditworthy for IBRDlending. 2. Blendcountries have accessto JBRD, butaccessmay be limited ("notional blends") because o f country-specific circumstances that make them non-creditworthy for IBRDfinancing. At present, Papua New Guinea, Uzbekistan, and Zimbabwe are classified as blends but do not receive any IBRDfinancing. 3. Unless a blend country i s capped, its IDA allocations are determined using the regular PBA formula. 4. Small islands have exceptional access to IDAbecause o f their vulnerability. These include: Cape Verde, Kiribati, Maldives, Samoa, Tonga, and Vanuatu (above IDA's operational cut-off butnot creditworthy for IBRD),pZus Dominica, Grenada, St. Lucia, and St. Vincent and Grenadines (above IDA's operational cut-off and classified as blends). Figure2: EstimatedFlow ofIDA Resourcesby IncomeGroupsinIDA15 0Percapital mcomeaboveIDA- operational Cutoff 0PerCapitaIncomeBetweenIDA Operational Cutoff and Two Per Capita Income under Two Per Capita Incomeunder One Allocatrons D e t e m e d PBA systemwith Dollar a Day i I by the PBA Formula exceptions (caps, post- i conflict, regional projects) - 6 - 0 Meeting Africa 's needs: The selected exceptions (caps, post-conflict allocations and regional projects) also helpmeet the needs o f the poorest regions that have difficulty accessing other sources of financing. For instance, if all funds were allocated using the PBA formula, 72 percent ofresources would go to South Asia and A h c a would get only 20 percent (Figure 3). The exceptions help direct half of IDA'S assistance to the Africa region. Africa's share would go up further ingrant element" terms, with HIPC and MDRIassistance included (Figure 4). Figure 3: Estimated Regional Flows by Region inIDA15 100%7 0 ECA,LACandMNA 2 90%- 22 =I 80%- $ 70%- 0 SOUTHASIA 60%- 2 5 0 % ~ 1 40% EASTASIA rc 0 30% 5 20% 7 AFRICA 0% Allocations Determinedby PBA Systemwith the PBA Formula Exceptions (Caps, Post- Conflict and Regional Projects) Figure 4: Estimated Grant Element of Regional Flows Including MDRI and HIPC Transfers in IDA15 0ECA,LACandMNA 3; m 90% 80% 70% 0 SOUTHASIA < 60% 0 E S 50% 40% 0 EASTASIA c 25 30% g 20% AFRICA c, 10% 0% PBA Systemwith PBA Systernwith Exceptions (Caps,Post- ExceptionsIncluding Conflict and Regional MDRland HIPCin Grant Projects) Elemnt Terms l 9 The grant element i s an assumed economc gain to IDA credit recipients and an assumed economc loss to IDA due to the differential between the market and IDA'Sinterest rates. Estimated grant element of regional flows was calculated assumng that IDA credits have a grant element of 60 percent and all projected HIPC and MDRIassistance were treated as grants. - 7 - C. ManagingIDA ResourcesWithin a ReplenishmentPeriod 10. Managing IDA resources. Every year, IDA resources are allocated according to principles agreed with donors duringthe replenishmentdiscussions and tables showing individual country allocations are preparedin advance o fthe fiscal year. Since country performance and grant eligibility are assessed every year,Jirm 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.20 11. Meeting country-specific needs. While IDA resources are allocated in accordance to principles agreedwith donors each year (PBA-driven allocations, capped allocations, post-conflict allocations, regional project allocations), operational realities on the ground inIDAcountries require flexible management o fresources duringa replenishment period. For example, a large and lumpy infrastructure project may require commitments inexcess of annualPBA allocations at the country level. Or it could be the case that not all allocated resources are committed ina country ina givenyear because o f country-specific circumstances and allocations are carried forward to the next year. It could also bethe case that annual allocations for small countries may be insufficient for a viable project and there may be a need to commit their three-year allocations ina single year. For all these reasons, inany givenyear, commitments may exceed PBA allocations for some countries (front loading) while for others, commitments may be lower than the PBA allocations (back loading). So although annual commitments could deviate from performance-based allocations for operational reasons, management keeps track of commitments ineach country to ensure that they are inline with overall performance- basedallocations for each country within each replenishmentperiod. 12. Front and back loading of allocations. Clear guidelines are established in advance o f each fiscal year on front andback loading o fresources. Limited front loading o f allocations inthe first two years o f a replenishment periodo f up to 30 percent i s possible. Small countries (withpopulations o f less than 1.5 million) can frontload by a higheramount giventhe small size oftheir allocations. Capped countries (India, Indonesia, and Pakistan) can also choose to frontloadhigher amounts as long as they stay within the three-year cap on allocations. Resources canbeback loaded inthe first two years o f the replenishment period, ifcountry conditions are not amenable for the full utilizationo f IDA allocations. Country commitments are requiredto stay within their indicative three-year cumulative PBA envelope for the replenishment period as a whole. Ifresources are front loadedinthe first two years ofareplenishment period, then resources available for commitment inthe final year will be commensurately lower. 13. Final year of the replenishmentperiod. Inadvance o fthe final year o f a replenishment period, ifit becomes evident that the allocated resources will not be committed inany country, then these dormant funds are returned to the IDA pool and redistributedto other countries usingthe PBA system. In addition, if a region identifies *' Allocations for the outer years are subject to a number o f uncertainties including the country's performance, its relative performance compared to other countries, the size o f the available IDA envelope, traffic lights, MDRIstatus, reactivation o f countries, and other reasons. - 8 - more dormant funds duringthe course of the final year o f a replenishment period, then it could reallocate up to 10 percent o f the total regional PBA allocations from countries where IDA allocations cannot be absorbed to better-performingcountries with higher absorptive capacity. 14. Going forward, management will continue to manage its resources flexibly in IDA15 to meet the emerging needs of the countries, while ensuring that overall commitments are inline with overall performance-based allocations for the replenishment period as a whole.21 111. SIMPLIFYING THEPBAFORMULA 15. Background. At the first IDAl5 meeting held inParis inMarch 2007, Deputies discussed the two proposals put forwardby management to simplifythe PBA formula.22 While agreeing that bothproposals greatly simplify the current PBA formula, Deputies indicatedthe "need to reflect further on the informationprovided inthis meetingas well as inputs from further work on IDA'Sresource allocation framework prior to deciding on a specific option."23 Specifically, to be able to make a choice between the two options, some Deputiesindicated that they would like to see the impact o f the two options on individual country allocations. This additional country-level information i s provided in Annex 2. Before comparingthe two options, this section beginswith a summary o fthe discussions inthe Marchpaper. 16. Increasing complexity of the PBAformula. IDA'SPBA formula has evolved over time and was modified several times duringthe IDAl1-14period (Annex 1). While such modificationshave helpeddirect more resources to countries that have achieved results, they also contributed to the complexity o f the formula, making it difficult to disentangle the impact o f each component on allocations at the country level. 17. Simplification is necessary. Since 2006, IDAhas been at the forefront in enhancing the transparency o f its resource allocation systemthrough public disclosure o f the ratings that feed into the PBA formula. However, given the complexity o f the formula, the way inwhich the ratings affect the final allocations i s not readily evident. This i s becauseof: (i) double counting o f governance; and (ii) structure o fthe the the formula, which includes a combination o f additive and multiplicative/geometric elements. Once the performance-based allocations are determinedusingthe formula described in Section 11, further steps involve grants-related discounts and the MDRI 21 IDA'Sallocation systemis one o f30business processes beingreviewedby IEGinthe on-going assessment o fthe IDA controls framework, mandated as part ofthe IDA14 replenishment arrangements. Preliminary findings o f the assessments, which will be discussed at the third IDA15 meeting in October 2007, suggest that the allocation system is appropriately designedand operating effectively. 22 IDA (February 2007). "IDA 's Performance-Based Allocation System: Optionsfor Simplijjing the Formula and Reducing Volatility. '' 23 IDA (2007). "Chairman's Summary: IDA Deputies Meeting, Paris, France, March 5-6, 2007.'' - 9 - netting-out mechanism. These two additional steps add to the complexity o f the allocation exercise. The focus inthis section is only on the PBA formula. Simplifying the formula would enhance its transparency, thus makingit easier for partner countries andcountry teams to better understand what drives changes intheir allocations. 18. Guidingprinciplesfor simplification. The two options for simplificationfocus on the country performance rating, a major component o fthe PBA system. At the IDA14 Mid-Term Reviewmeeting, Deputiesprovided two guiding principles for simplification. First, they requested that simplified options retain a weight o f governance similar to the current formula. Second,they asked that the new options track closely the allocations from the current formula to minimize disruptions at the country level. 19. The currentformula. As agreedinParis with the Deputies, beginningFY08, the current formula incorporates the three technicalchanges inthe calculation o f portfolio performance ratings. These changes were introducedto lower unwarrantedvolatility. The changes are: (i) actual problemprojects instead of actualplus potential using problemprojects; (ii) usingan average o f quarterly data on actual problemprojects instead of an end-year snapshot; and (iii) revisingthe scale usedto convert the percent o f projects at risk into a rating. A s pointed out inthe paper presented at the Paris meeting, dropping potentialproblem projects from the calculation o f the portfolio ratings has an additional consequence for the governance factor. Since the procurement ratingi s based on a risk flag drawn from the potentialproblemprojects, it i s no longer included inthe calculationo fthe governance factor. 20. Two simplification options. Inaccordance with the guidance providedby Deputies (paragraph 18) and the changes introducedto the current formula (paragraph 19), two options -a geometric formula (Option I) and an additive formula (Option 11)- were proposed to simplify the formula.24 Current Formula Countryperformancerating = (0.8 "CPIA + 0.2* PORT) * (CPIA /3.5)'.5 D IDA country allocation= f (Country performance rating*.O, Population'.', GNI/~apita-O.'~~) Option I Country performance rating= (CPIA A-C) 0.24 * (CPIA D) o.68* (PORT) '.Os IDA country allocation = f (Country performance rating5.0, Population'.o, GNI/~apita-'.'*~) 24 Inthe current formula, country performance ratings are more dispersedbecause ofthe governance factor. Incontrast, ratings are less dispersed inOptions Iand 11. Therefore, increasing the exponent o n the country performance ratings from 2 to 5 inthe proposed option, results in similar shares and therefore similar levels o f allocations per capita as the current formula. - 10- Option 11 Country Performance Rating= (0.24 CPIA* A-c + 0.68 * CPIA D+ 0.08 * PORT) IDA country allocation =f(Country performance rating '.',Population`.',GNV~apita-'.'~') 21. Comparing the Options. Both options address the main problems inthe current formula by doing away with: (i) double counting of governance and (ii) structure the the o f the formula, which includes a combination o f additive andmultiplicative/geometric elements. Inaddition, both options comply equally well with the Deputies' guiding principles o f retaining the weight o f governance, and closely tracking the allocations from the current formula to minimize disruptions at the country level. 22. Both optionsproduce similar results. Applyingthe options to IDAl5 projections shows that they both track the current formula very closely. As requestedby the Deputies, country-specific information for the IDAl5 period comparing the two options is shown inAnnex 2, along with assumptions that underpin these simulations. Allocations generated by Options Iand I1differ from the ones generated by the current formula on average by 0.32 percent and 0.50 percent respectively over the entire IDA15 period. Over the same period, under Option I(Option 11) the greatest fall inallocations is around 3.3 percent (1.1 percent) and the greatest increase is around 1.3 percent (4.2 percent). Inaggregate, for all countries, this translates into an overall shift o f SDR26.7 million (0.10 percent oftotal resources) among all IDA countries under Option Iand SDR68.6 million (0.26 percent o f total resources) under Option 11. Interms o f allocations, both options produce similar results and the differences between the two are minor at the country level. 23. Option II is simpler of the two options. Although Option Itracks the current formula slightly more closely, Option I1is preferable because it is simpler and more transparent. It has the following advantages: It is more easily understoodbypolicymakers inIDA countries. IDAhas made considerable progress inensuringgreater transparency and disclosure o f IDA'SPBA system. IDAhaspublicly disclosed detailed CPIA scores for all IDA-eligible countries since 2006. This disclosure has allowed IDA countries to benefit from open scrutiny o f comparative performance. Inaddition it has served, through consultations at the country level, as a diagnostic tool to strengthen partnership between the country policy makers and IDA. A further simplification o fthe PBA formula, by choosing Option 11, would help IDA countries take the next step infiguring out how changes intheir CPIA ratings would affect their IDAresource allocation, other things remaining constant. 0 The impact o fincremental changes inthe CPIA and portfolio ratings on performance-based allocations i s straightfonvard and readily understood - all components have the same incremental impact on the final country performance rating regardless o f the level o f performance. Instead, inOption I, wouldneedtobefamiliarwiththeconceptsofelasticityand one - 11- logarithmic transformations to arrive at the incremental impact o f changes in CPIA and portfolio ratings on country performance rating. The latter adds a further step when calculatingthe effect of component changes on the final ratings, thereby makingit a bit more complex. 24. Inview ofthe advantagesdescribed above, management recommends OptionI1to simplify the PBA formula. IV. SMALL STATESANDIDA'sFINANCIAL ASSISTANCE 25. Small states. The term "small states" generally refers to sovereign countries with fewer than one and a half millionpeople.25 While small states can be found in every geographic region, they are largely concentrated inAfrica, Caribbean and Pacific regions. Together, the small statesare home to 29 millionpeople, 0.5 percent o f the total population o f developing countries. Their sizes differ greatly, from micro states with fewer than 100,000 people eachto those with more thanone millionpeople. 26. IDA eligiblesmall states. DuringIDA14, 20 IDAborrowers hadpopulations o f around 1.5 millionor less, with per capita incomes rangingfrom below IDA operational cutoff inFY08 (Bhutan,Comoros, Djibouti, Gambia, Guinea Bissau, Lesotho, Sao Tome and Principe, Solomon Islands, Timor Leste) to those above IDA's operational cutoff (Cape Verde, Dominica, Grenada, Guyana, Kiribati, Maldives, Samoa, St. Lucia, St. Vincent and the Grenadines, Tonga, and Vanuatu).26 While small states account for a fourth o f IDA-eligible countries, their combined populationi s small at around 0.5 percent o fpeople living inIDA countries. A. ChallengesFacedby SmallStates 27. Challenges. Small states face a number o f well-known challenges because o f their size.27 These include: Higher per capita costs ofproduction. Like all countries, small states have to provide a range o fpublic goods and services. These include the central functions ~ ~ ~ 25 See also IDA (December 2004), "Supporting Small and Vulnerable States. " 26 Several small island economies have special access to IDA despite highper capita incomes because o f their vulnerability. These are: Cape Verde, Dominica, Grenada, Kiribati, Maldives, Samoa, St. Lucia, St. Vincent and the Grenadines, Tonga, and Vanuatu. 21 Recent research indicates that these challenges, outlined inthe 2000 Commonwealth Secretariat/World Bank Joint Task Force report presented to the Development Committee, have endured-and indeed that "new challenges have emerged for small states inthe recent past, including faster than anticipated preference erosion for traditional exports; a rapid rise inthe debt burdenfor many small states; increased environmental susceptibilities; rising concerns with respect to youth employment; security and crime; and the HIV/AIDS pandemic" (see the Joint Commonwealth Secretariat/World Bank report Toward an Outward-Oriented DevelopmentStrategvfor Small States: Issues, Opportunities, and Resilience Building-A Review of the Small States Agenda Proposed inthe Commonwealth Secretariat/World Bank Joint Task Force Report o fApril 2000). - 1 2 - o f government, regulatory activities, tax administration, provision of education, health and social services, ajudicial system, foreign relations, and security. To the extent that there are indivisibilities inthe provision o f such public services, the fixed costs per capita are higher in small states. This is corroborated by empirical evidence, which shows that small states have higher costs o f producingpublic goods.28 Ln addition, the costs o f providing physical infrastructure are particularly highinsmall states. Moreover, indispersed islandstates such as inthe Pacific, the development o f even a small domestic market is constrained by high transportation costs. Distance is ofien compounded by relatively small volumes o f cargo, making transportation expensive. A shows that small and island states are amongst countries facing the highest transportation which are higherby25-100percent compared to eventhe mediancost o fdeveloping countries.31 Highinternational transport costs in turn increase the cost o f exports, thereby reducing their competitiveness and export returns. Similarly, the costs o f imports are increased, resulting inconsumer welfare losses. 0 Economic vulnerability is high. Second, because their economies tend to be undiversified and highly dependent on external trade, they are more vulnerable to economic shocks and have significantly more volatile growth rates than larger countries. With a given level of shock, the impact on consumption will be more rapidly felt ina small state than ina medium or large state. Consumption volatility i s not only higherthan output volatility, but also a more relevant measure o f welfare, especially o f the poor inthese countries (Table 2). 0 Also vulnerable to natural disasters. Finally, many small states are prone to natural disasters such as tropical storms, rise in sea level, cyclones, droughts, hurricanes, tsunamis, and volcanic eruptions. These natural disasters typically affect the entire economic, human and physical environment of small states. Often, periodic recurrences o f these events compound the human, economic, and environmental costs and the financial burdenso f renewingor replacing infrastructure. 28 See Alesina, Albert0 and Romain Wacziarg (1998). "Openness, Country Size and the Government, " Journal o f Public Economics, Vol. 69, Issue 3 (September), 305-321.) 29 Atkins, J P, Mazzi, S, A & Easter C D, Commonwealth VulnerabilityIndexfor Developing Countries, Economic Paper, No. 40, Commonwealth Secretariat, London, 2000. 30 Transport costs are measured as a ratio of insurance and freight debits to merchandise imports. The ratio to imports, rather than exports, is used since merchandise imports are often much larger than exports for many small states. 31 Insurance and freight debits as a percent o f merchandise exports for some small states: Kiribati 26 percent; Comoros 24 percent; SBo Tome and Principe 23 percent; Tonga 22 percent; Vanuatu 20 percent; Gabon 20 percent; The Gambia 20 percent; Solomon Islands 20 percent; Equatorial Guinea 19 percent; Trinidad & Tobago 18 percent; and Seychelles 18percent. The median percentage for developing countries as a whole i s 14 percent. - 13 - Table 2: Small States are, on Average, Subject to Greater Volatility than Other Countries (Standard deviations of annual growth rates, inpercent; 1960-2000) Output Terms o f Private Private and public trade consumption consumption Small states 5.8 5.6 12.6 7.7 Other developing countries 4.9 4.2 8.2 8.7 Industrial countries 2.5 1.5 2.6 2.2 Source: M.Ayhan Kose and Eswar S. Prasad, Finance and Development, December 2002, Volume 39, Number4. B. What do These ChallengesMeanfor Small States? 28. Policy responseshould be swvt andflexible. Given the structure o f their economies and the nature o f challenges they face, it is all the more importantfor governments in small states to react swiftly and flexibly to mitigate the impact of shocks, implement structural reforms, provide infrastructure and public services, and generate effective information flows. 29. Regional integration is important. Inaddition, growth andpoverty reduction in the small IDA-eligible states depends on their ability to take full advantage o fincreased regional and global integration. Inaddition, integration and alliances o f small states with similar economic structures (for example, the Pacific Forum and the Caribbean Community) have provided opportunities for poolingto reduce the costs o f providing public goods and services. C. IDA's Role in SmallStates 30. IDA's overall role in small states. A comprehensive report on the Bank's support to small states - covering knowledge work, lending, and humanresources-was provided at the IDA14 replenishment meetings.32 Meanwhile, the Independent Evaluation Group (IEG) also prepared a synthesis o f findings o fWorld Bank evaluations insmall states.33 The IEGreport notes that, while the World Bank has a comparative advantage interms o f its vast knowledge and experience, it is relatively constrained inits activities in small states because o f the lack o f field presence and because many small states have access to substantial grants from other donors. 31. Lessons learned. As such, the IEGreport called for the scope and nature o f World Bank's activities insmall states to be "strategic and selective." 34 This is based on its findings that Bank lending was effective when it was focused infew critical sectors and underpinned by equally focused analytical work o f highquality. The report also found that continuity o f Bank involvement was necessary to make the sustainability o f 32 See IDA (December 2004). "Supporting Small and Vulnerable States." 33 IEG (2006). "Small States: Making the Most of Development Assistance. A Synthesis of WorldBank Evaluation Findings. 'I 34 The IEGreport also finds that IDA's administrative costs are three times higher per dollar lent in small states as compared to the Bank average during FY00-05. - 14- project benefits likely. Going forward, the report notes that the World Bank can play a limited, but useful role insmall states by: (i) helping governments coordinate and harmonize donor activities to maximize the development benefits o f Official Development Assistance (ODA); (ii) supportingcooperation among small states to address development issues on a regional basis, especially where there are asymmetries inresources, skills, andbargainingpower; (iii) preparingcross-country analytical work on issues o f common concern and where possible, tailoring them to country circumstances; and (iv) helpingharmonize donor practices to lower transactions costs. 32. Inadditionto providinga"platform" for makingoverallaidmore effective, IDA's role insmall states can also be enhanced throughthe learningthat comes from implementingprograms and projects ina country.35 This section focuses on strengthening IDA's financial assistance to small states, which has eroded over the past replenishments. D. IDA's Financingin SmallStates 33. Base allocation and cap onper capita allocation. IDA's PBA systemhas a `fixed' component, which gives a base allocation o f SDR 1.1million each year to all countries irrespective o f their performance assessment. This base allocation was deemed as the minimumamount necessaryto maintaina program ina country. Inaddition, there i s also a cap on per capita allocations placed at US$20 (or SDR13.2), to ensure equitable treatment o f countries. 34. Impact on small states. The base allocations as well as the cap on per capita allocations largely affect small states. Small states, incomparison to large countries, have small overall allocations (base allocationplusperformance-based allocations) and the base allocation forms a larger share o f this overall allocation. Inaddition, the cap on per capita allocations also largely affects small states. InFY07, for instance, seven small states reached the cap either because they were goodperformers (Dominica, Grenada, St. Lucia, St. Vincent and the Grenadines, and Samoa) or have extremely small populations (Kiribati and Tonga). Some large countries, like Ghana and Tanzania, are also close to the cap because o f their strong performanceratings, but it i s relatively uncommon for large countries to approach the cap on per capita allocations. 35. Erosion of base and cap in real terms. A review o f IDA allocations shows that the base allocation and cap on per capita allocations have not been fully adjusted since IDA9 and have therefore eroded inreal terms.36 Moreover, giventhe highper capita cost o fprovidingpublic goods inthese countries, as well as the lumpynature o f some infrastructureinvestments where IDA's support i s most strongly sought inthese states, the current base allocationi s too small. For instance, IDA's current minimumlump sum allocation o f SDR 1.1million allows very little infrastructure investment inthese countries. Table 3 provides an illustrative calculation o f what the current minimum 35 IDA (2007). "The Role of IDA in the Global Aid Architecture: Supporting the Country-Based Development Model. " 36 The base allocation was adjusted slightly inIDA14,but this adjustment did not go far enough. - 15- (SDR1.1 million a year) can "buy" interms o f infrastructure. At most, based on available estimates of average unit costs for all countries, this minimumallocationbuys very little infrastructure: 4 kilometers o f two-lane pavedroad, or 2 kmo frail road, or 895 kilowatt ofelectricity generation, or 4,700 households connected to sanitation, or about twice as many households connected to water pipeline. These calculations do not include the additional minimummaintenance costs neededto ensure the full hnctionalityo fthe infrastructure network. Table 3: How MuchInfrastructureDoes the Current MinimumAllocation"Buy"? What the current Unit costs* minimumallocation Sector ($) Unit can "buy" Electricity 1,900 Per kilowatt o f generating 895 kilowatt o f capacity, including generation ................................................................................................................................................................................................................................................................F associated network cost Roads 410,000 Per lulometer o f two lane 4 kilometers of two lane ................................................................................................................................................................................................................................................................O pavedroad p,avedroad Railway 900,000 Per kilometer o f rail, 2 kilometers o frail including associated rolling ............................................................................................. ............................................................................................................................................................... stock Sanitation 700 Per connected household 2,430 households ................................................................................................................................................................................................................................................................8 connected Water 400 per connected household 4,250households ................................................................................................................................................................................................................_.................................................................................................... connected ................................................................................................................................................................................................................................................................' Mainlines 400 per line 4,250 mainlines Mobile 580 per subscriber 2,93 lmobile lines Note: 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. Note that the unit costs are averages for all countries and that the unit costs in small states could be higher. Source: Marianne Fay and Tito Yepes (2000). "Investing in Infrastructure: What is Neededfrom 2000 to 2010." E. StrengtheningIDA's FinancialAssistanceto SmallStates 36. Increase in base allocations. Giventhe importance o fbase allocations insmall states as discussed inparagraphs 34 and 35, Management proposes increasingthem from SDRl.l millionper year (or SDR3.3 millionper replenishment period) to SDRl.5 millionper year (or SDR4.5 millionper replenishment period) for all countries. This increase adjusts for inflation and therefore helps maintain the base allocation constant in real terms since IDA9. 37. Raising the cap onper capita allocations. Management further proposes raising the cap on per capita allocations from SDR13.2 (US$20) to SDR19.8 (US$30), to adjust for inflation since IDA9. Such an increase would also benefit the small states disproportionately and would be inline with the performance orientation o f IDA's PBA system. This i s becausewell-performing small states are more likely to reach the cap, - 16- andtherefore standto benefit from it. Inaddition, some micro states with very small populations will also benefit from it.37 38. Annex 3 shows the impact o f changing the base and cap at the country level. Small countries would benefit from the two proposals, and the burden on other countries to compensate for this increase is negligible. It also causes minimal departure from the current formula. 39. Inaddition, inview ofthe importance ofregional integrationfor small states, management also proposes changes inregionalproject allocations as discussed inthe next section. V. REGIONAL PROJECTS 40. Background. Duringthe IDA14 period, SDR200 million inFY06 and SDR250 million ineach o f FY07 and FY08 was set aside for regional projects. Giventhe externalities associated with regional projects and the need to demonstrate ownership by participating countries, 1/3 o f the cost o f an eligible regional project is charged to the country PBA envelope to leverage 2/3 o f the cost from the regional envelope. Thus while providing an important incentive for regional integration, this formula also provides a clear link to performance-based allocations, and helps ensure country ownership and commitment. 41. Strong demandfor regionalprojects. The demandfor regional project envelope was strong inthe first two years o f IDA14, with approved projects using SDR235 million and SDR233 million respectively ofthe topping-up resources, leaving SDR233 million for FY08. This i s insufficient for meeting the demandfor topping-up funds inFY08, where the 16project proposals received so far add upto SDR465 million intopping-up funds. The excess demand for topping-up funds for regional projects will therefore spill over to IDA15 and this i s corroborated by the strong pipeline of regional projects, especially in Sub-Saharan Africa. As requested by the Deputies at the IDA15 ReplenishmentmeetinginMaputo, Annex 4 shows the pipelineofregional projects for the IDA15 period. 42. Ceiling on country contribution. While the 1/3:2/3 split between the country allocation and topping-up fund has beenworking well, implementation experience has shown that there have beeninstances where country participation inregional projects was constrained by a small IDA allocation. Therefore, management proposes placing a ceiling on an individual country's cost ofparticipationinregional projects at 20 percent o f its annual a l l o c a t i ~ n .A~ ~ country would finance its 1/3 share of participationin 37 Insome micro states, performance-basedallocations could be lower than base allocations. This is because either their performance or their small population or both could cause them to reach the cap thus constraining their allocations below the base allocation. 38 After grants discount and MDRInetting out. - 17- regionalprojects untilthe cost o fparticipationreaches 20 percent of its allocation, with the regional envelope financing the country's costs beyond this ceiling. 43. Rationaleforplacing the ceiling at 20percent. An analysis o fthe regional projects approved duringIDA14 shows that the average country contribution has been around 20 percent (Annex 5),39with countries with small allocations above this average and countries with larger allocations well below. Therefore, placing the ceiling at 20 percent will ensure that countries with small allocations, for which the country contribution represents a serious constraint, will be able to participate to a greater extent to regionalprojects. Infact, applying this ceiling to the IDA15 regional projectspipeline shows that limiting the costs borne by an individual country as a share o f its IDA allocation by placing a ceiling would mainly benefit countries with small IDA allocations, especially the small states (also see paragraph 29). While the same objective could have been achieved by limitingthe group o f countries to which a ceiling would apply, having one ceiling for all countries provides a simple and transparent rule that avoids creating yet another subset o f countries inIDA.40 44. Benefits for countries with small IDA allocations. As shown inFigure 5,41a 20 percent annual ceilingwould benefitprimarilycountries with small allocations, and in particular small states. Based on the IDA15 pipeline, o fthe 20 countries that would likelybenefit by limiting their costs o fparticipatinginregionalprojects to 20 percent, 16 are projected to have allocations o f less than SDRl1million, with the remaining country having an average allocation o f about SDR43 million.42 Because these are mainly countries with small IDA allocations, the actual additional costs imposedon the regional envelope are small - estimated at about SDR61 millioninIDA15. 45. Checks and balances. There are a number o f checks andbalances inplace that will help ensure that burdensharing inregional projectscontinues to reflect benefits. Currently each region is requiredto manage a share o f the fixed regional envelope and set regional priorities. As such there would not be a strong incentive for a region to encourage large projects to move forward that would cause a breach o f this cumulative ceiling, as this would more quickly deplete the limitedregional envelope. FRM also clears each individual regionalproject and hence would carefully monitor any unexpected impacts from this cumulative ceiling. At the IDA15 Mid-Term Review, management will present a review o f the regional project program, report on experience with the scaling up o f regional projects, and propose adjustments to the proposed ceiling on country contributions ifnecessary. 39 The analysis relates to FY06 and FY07 for which the regional project data is more relevant. FY08 figures are less representative because o f the constraints relative to the final year o f the IDA replenishment: countries that have frontloaded have fewer resources to contribute to regional projects. 40 Definingcountries with small allocations by usinga cutoff could also create problems during implementation. For example, if countries below SDRlO million are designated "small allocations countries," a country with SDRl1million would not benefit from the ceiling even though its allocation is not substantially different from the countries with SDRlO million. 4` Figure 5 reports FY09 and FYI0 for which the regional pipeline data is more concrete. 42 Countries are likely to reach the ceiling because o f the limited size o f their allocations, size o f regional project, and/or due to the number o f regionalprojects inwhich they participate. - 1 8 - Figure5: ProjectedImpact of a Ceilingof20 percent on Country Contributionsto Regional Projects FY09 100 ~ -- 80 v1 -- E 60 -2.- E Small - Allocation Size , Large Country contribution to regional project----+---- Share o f allocation s *E 100% - $ 2 80 .? a 80% m -; -3.- c +3 s7i 60 o s .$, 60% % e E 40 40% 2 3o) cA E 20 c Ag 20% 0% Small Allocation Size , 0 Large 1- Country contribution to regional project----+----o Share f allocation 1 VI. ISSUES FOR DISCUSSION D o Deputiesagree that Option 11, the additive formula, i s the preferredoption? Do Deputiesagree with increasing the base allocations to all countries from SDR1.l millionper year (or SDR3.3 millionper replenishment) to SDRl.5 millionper year (or SDR4.5 millionper replenishment) and the per capita cap on allocations from SDR13.2 (US$20) to SDR19.8 (or US$30) for the benefit o f the small states? Do Deputiesagree to a ceiling o f 20 percent on the country contributions to regionalprojects for the benefit o f countries with small IDA allocations? - 19- ANNEX1. EVOLUTION THEIDA PERFORMANCEBASED OF ALLOCATION FORMULA (IDA11-13) 1. This annex provides describesthe modifications inthe PBA formula since the IDA11period.43 FY97-98 Share = (IDArating i)'.*x Population x (GNYcap i i)-0'25 . Where IDA Rating=CPIA andbase allocationo f SDR3 million for 3 years. FY99 IfIDAPerformance Rating<=2.00, then Share = (IDArating i)'.' x Population i x (GNYcap i)-.I2' If2.00 2.90, then Share = (IDA rating i)1.95Population x (GNYcap x i i)-.I2' . Where IDA Rating=CPIA and base allocation o f SDR 3 million for 3 years. moo-01 IfIDAPerformance Rating<=3.00, then Share = (IDA rating i)1'75x Population x (GNYcap i i)-'125 IfIDAPerformance Rating>3.00, then Share =(IDA rating i)2'oo Populationi x (GNYcap x i)-'125 Where IDA Rating for Country =(0.8 x CPIA + 0.2 x ARPP i i i)x Governance Discount Other: (i) Base allocation o f SDR 3 million for 3 years. Inaddition, there was apopulationadjustment. IfIDAPerformance Rating<=3, nopopulationadjustment is applied IfIDAPerformance Rating>3, then: 43 This annex draws from Moorty, L.andA. Orzan(2007). "DoIDA Resources Flow to Better- Performing Countries?" Manuscript,The World Bank. - 20 - Ifpopulation< lmillion, populationadjustment = 1.2 (thismultiplier is applied to the formula above) Iflmillion<=population<20million, populationadjustment = 1+[0.2 - (population inmillionllOO)] Ifpopulation=>20m, no populationadjustment isapplied. . FY02 IfIDAPerformanceRating<=3.OO, then Share = (IDA ratingi)1'75 x Population i x (GNI/cap i)-.12' IfIDAPerformanceRating>3.00,then Share = (IDA ratingi)2'oo x Population i x (GNVcap i)-.12' where IDA Rating Country i=( 0.8 x CPIA + 0.2 x ARPP x Governance Discount i i) Other: (i)Base allocation o f SDR 3 million for 3 years; (ii) population adjustment. No FYO3-FY05 Share = (IDA rating i)2'oo Population i x (GNI/cap x i)-.l2' where IDA Rating Country i=( 0.8 x CPIA + 0.2 x AFWP x Governance factorj j i ) where Governance Factori= (average ratingo f 6 governance criteriai/ 3.5)'.' Other: (i)Base allocationo f SDR 3 million for 3 years; and (iii) population No adjustment. FY06-FY07 Three-year allocations. Allocation Country = SDR3.3 million + Performance-Based Allocation (PBA i i) Where: (IDA ratingi)2x Population i x (GNI/cap PBA =________________________________________----------_-------------------------- i)-.12' x Envelope C i=1-81 [(IDA rating i)2x Populationi x (GNVcap i)-.12'] (i) IDARatingCountry i= ( 0.8 x CPIA +0.2 x ARPP x Governance factor i j) i (ii) Governance Factor = (average rating o f 6 governance criteria / 3.5)'.5 i i - 21 - (iii) The Envelope = IDA three-year envelope, after deduction o f the otherwise determinedblendallocations as well as the allocations to eligible post-conflict countries. (iv) The countryperformance-based allocationi s subject to a maximum o f $20 per capita per annum. FY08 Three-year allocations. Allocation Countryi = SDR3.3 million +Performance-Based Allocation (PBA i i) Where: (i) IDARating Country =(0.8 x CPIA + 0.2 x ARPP i i i)x Governance factor i (ii) Governance Factor i= (average rating o f 5 governance criteria i/ 3.5)'.' Procurement factor dropped from the calculation of the governance factor due to changes inthe way the portfolio performance ratings are calculated. (iii) The Envelope =IDA three-year envelope, after deductiono fthe otherwise determinedblendallocations as well as the allocations to eligible post-conflict countries. (iv) The countryperformance-based allocation i s subject to a maximum o f $20per capita per annum. (v) The portfolio ratings are calculated using(a) usingactual problem projects instead of actual pluspotentialproblemprojects; (b) usingthe average data on actual problem projects at the end o f four quarters rather than an end-year snapshot; and (c) revising the conversion scale usedto convert the percent o f projects at risk into a rating. - 22 - ANNEX2: IDA15 PROJECTIONSUNDERTHE CURRENTPBA FORMULA AND SIMPLIFICATION OPTIONS Option I Option I1 comparedto compared to IDAlS albl IDA1S a/b/ IDAlS a/b/ Current Current Current Option I Option I1 o/o % Africa Africa Regional 960.0 960.0 960.0 0.0% 0.0% Angola 156.0 156.0 156.0 0.0% 0.0% Benin 153.8 154.4 153.0 0.4% -0.5% Burkina Faso 320.9 320.1 318.3 -0.3% -0.8% Burundi 157.2 157.2 157.2 0.0% 0.0% Cameroon 192.8 193.4 192.1 0.3% -0.4% Cape Verde 22.0 21.9 21.9 -0.3% -0.2% CentralAfrican Republic 18.0 18.1 18.0 0.0% 0.0% Chad 27.3 27.3 27.3 -0.2% 0.1% Comoros 3.8 3.7 3.8 -2.2% 0.8% Congo, DR 719.8 719.7 720.1 0.0% 0.0% Congo, Republic of 52.2 52.2 52.2 0.0% 0.0% Cote d'Ivoire 214.4 214.4 214.3 0.0% 0.0% Eritrea 41.5 41.3 41.6 -0.6% 0.1% Ethiopia 1562.7 1570.1 1551.2 0.5% -0.7% Gambia, The 13.2 13.3 13.1 0.6% -O.!)% Ghana 851.0 852.0 848.4 0.1% -0.3% Guinea 33.0 33.1 33.1 0.3% 0.4% Guinea-Bissau 11.2 11.1 11.1 -0.9% -0.5% Kenya 857.3 855.4 851.1 -0.2% -0.7% Lesotho 39.6 39.7 39.3 0.4% -0.7% Liberia 68.6 68.6 68.6 0.0% 0.0% Madagascar 445.6 447.8 442.2 0.5% -0.8% Malawi 278.2 279.0 275.9 0.3% -0.8% Mali 361.7 363.3 358.9 0.4% -0.8% Mauritania 19.1 19.1 19.2 0.1% 0.7% Mozambique 316.6 315.6 315.7 -0.3% -0.3% Niger 232.3 233.1 230.3 0.4% -0.8% Nigeria 1680.7 1678.9 1692.8 -0.1% 0.7% Rwanda 221.9 222.4 221.1 0.2% -0.4% Sao Tome and Principe 3.O 3.0 2.9 0.3% -0.4% Senegal 266.8 268.4 263.9 0.6% -1.1% Sierra Leone 56.7 56.6 56.3 -0.3% -0.8% Somalia 0.0 0.0 0.0 0.0% 0.0% Sudan 118.0 118.1 118.1 0.1% 0.1% Tanzania 1521.9 1523.2 1517.6 0.1% -0.3% Togo 71.7 71.8 71.7 0.0% 0.0% Uganda 778.0 776.5 786.2 -0.2% 1.1% Zambia 166.0 166.3 164.5 0.2% -0.9% Zimbabwe 10.4 10.4 10.4 0.0% 0.0% - 23 - Option I Option I1 compared to compared to IDA15 a/b/ IDA15 a/b/ IDA15 a/b/ Current Current Current Option I Option I1 % % Arrears Clearance 355.9 355.9 355.9 0.0% 0.0% Subtotal 13381.O 13392.0 13355.6 0.1Yo -0.2% East Asia East Asia Regional 48.0 48.0 48.0 0.0% 0.0% Cambodia 130.9 130.7 131.6 -0.2% 0.5% Kiribati 4.1 4.1 4.1 0.0% 0.0% Lao PDR 49.3 49.2 49.6 -0.2% 0.7% Mongolia 51.8 51.8 51.5 0.0% -0.6% Myanmar 0.0 0.0 0.0 Papua New Guinea 66.9 67.7 67.3 1.3% 0.6% Samoa 7.9 7.9 7.9 0.0% 0.0% Solomon Islands 5.5 5.4 5.6 -2.O% 1.2% Timor-Leste 9.1 9.1 9.1 0.0% 0.0% Tonga 3.2 3.2 3.2 -0.3% 0.1% Vanuatu 5.9 5.9 5.9 0.3% -0.3% Vietnam 2508.5 2511.6 2497.5 0.1% -0.4% Subtotal 2891.1 2894.7 2881.3 0.1Yo -0.3% South Asia Afghanistan 483.8 483.8 483.9 0.0% 0.09.b Bangladesh 1918.4 1908.8 1953.7 -0.5% 1. w o Bhutan 21.5 21.6 21.4 0.4% -0.7% India 2829.6 2829.6 2829.6 0.0% 0.0% Maldives 11.2 11.2 11.1 0.2% -0.5% Nepal 448.0 448.5 444.9 0.1% -0.7% Pakistan 1800.0 1800.0 1800.0 0.0% 0.0% Sri Lanka 442.2 444.3 438.7 0.5% -0.8% Subtotal 7954.7 7947.9 7983.3 -0.1 Yo 0.4% ECA ECA Regional 48.0 48.0 48.0 0.0% 0.0% Armenia 111.3 111.2 111.7 -0.2% 0.3% Azerbaijan 158.6 158.4 159.8 -0.1% 0.8% Bosnia-Herzegovina 74.8 74.6 74.3 -0.3% -0.7% Georgia 148.1 147.8 147.4 -0.2% -0.4?'0 Kosovo 19.8 19.8 19.8 0.0% 0.0% Kyrgyz Republic 66.4 65.6 68.8 -1.1% 3.7% Moldova 73.6 73.6 74.0 -0.1% 0.5% Tajikistan 62.5 61.6 65.1 -1.4% 4.2% Uzbekistan 143.8 142.1 145.7 -1.2% 1 4% Subtotal 906.8 902.6 914.7 -0.5% 0.9% LAC Bolivia 115.2 111.4 114.1 -3.3% -0.9% - 2 4 - Option I Option I1 compared to comparedto IDA15 albl IDA15 dbl IDA15 a/b/ Current Current Current Option I Option I1 % % Dominica 2.8 2.8 2.8 0.0% 0.0% Grenada 4.2 4.2 4.2 0.0% 0.0% Guyana 7.1 7.1 7.1 ', -0.2% 0.4% Haiti 30.9 30.4 32.0 -I.6% 3.7% Honduras 153.6 153.7 152.9 , 0.1% -0.5% Nicaragua 109.1 109.3 108.6 0.2% -0.4% St. Lucia 6.5 6.5 6.5 0.0% 0.0% St. Vincent & the Cienadines 4.7 4.7 4.7 0.0% 0.0% OECS Countries 48.0 48.0 48.0 1 0.0% 0.0% Subtotal 482.2 478.2 481.2 1 -0.8% -0.2% MNA Djibouti 9.4 9.4 9.5 -0.7% 0.4% Yemen, Republic o f 231.8 232.0 232.3 j 0.1Yo 0.2% Subtotal 241.2 241.3 241.7 . 0.1% 0.2% Regional Set Aside non-AFR 96.0 96.0 96.0 0.0% 0.0% ~ Hard Term Lending 346.1 346.4 345.4 I 0.1% -0.2% Total 26299.2 26299.2 26299.2 I 0.0% 0.0% Notes: a/ All amounts are net of grant discounts and o f debt service due (MDRI). Projected country allocationspresented under the current formula differ from those presented in the June meeting because they are based on updated performance ratings, incorporate newly agreed assumptions on post-conflict and re-engagement country allocations, take into account anticipated graduations of countries from IDA and include the proposed scale up of regional projects. . hi IDA15 projectionsassume a 20 percent increase incommitment authority and the saine country indicators as in FY08. - 25 - ANNEX3: SIMULATEDIMPACT OFANINCREASE INTHE BASEALLOCATIONONSMALL STATES Minimum Minimum Minimum Minimum Allocation Allocation Allocation Allocation Raised to Raised to Raisedto Raisedto SDR1.5 SDR1.5 SDR1.5 SDR1.5 Million Million Million Million and Cap and Cap and Cap and Cap at at at at Country Current SDR13.2 SDR19.8 SDR13.2 SDR19.8 Performance Formula pcpa pcpa pcpa pcpa Population Ratings SDR m SDR m SDRm Percent Percent millions Africa Africa Regional 960.0 960.0 960.0 0.0% 0.0% Angola 156.0 156.1 156.1 0.1% 0.1% 16.4 1.5 Benin 153.8 154.2 153.5 0.3% -0.2% 8.7 3.3 Burkina Faso 320.9 320.8 319.5 0.0% -0.5% 13.6 3.6 Burundi 157.2 157.3 157.3 0.1% 0.1% 7.8 2.1 Cameroon 192.8 0.2% 2.6 Central African Republic 18.1 18.0 0.0% 0.0% 4.1 1.2 Chad 27.3 28.3 28.1 3.6% 2.7% 10.0 1.5 Comaros 3.8 4.7 4.7 24.4% 23,8YrJ 0.6 I,3 Congo, DR 719.8 718.4 718.4 -0.2% -0.2% 59.3 1.5 Congo, Republic o f 52.2 52.4 52.4 0.4% 0.4% 4.1 2.0 Cote d'lvoire 214.4 214.4 214.4 0.0% 0.0% 18.5 1.5 Eritrea 41.5 42.6 42.2 2.4% 1.7% 4.5 1.8 IGhana Guinea 33.0 33.9 33.6 2.7% 1.7% 9.6 1.9 Kenya 857.3 855.6 852.1 -0.2% -0.6% 35.1 3.4 Liberia 68.6 68.6 68.6 0.0% 0.0% 3.4 Madagascar 445.6 445.0 442.9 -0.1Yo -0.6% 19.1 3.4 Malawi 278.2 278.2 277.0 0.0% -0.4% 13.2 3.2 Mali 361.7 361.5 359.8 -0.1% -0.5% 13.9 3.6 Mauritania 19.1 20.0 19.9 4.8% 3.9% 3.2 2.5 Mozambique 316.6 316.5 315.0 0.0% -0.5% 20.2 2.8 Niger 232.3 232.5 23 1.5 0.1% -0.3% 14.4 2.8 Nigeria 1680.7 1676.2 1669.5 -0.3% -0.7% 144.7 2.3 Rwanda 221.9 222.1 221.2 0.1% -0.3% 9.2 3.5 Saa Tome and ~ r i ~ ~ i p ~3.o 3.9 3.9 31.2% JO.S% 0.2 2.3 Senegal 266.8 266.7 265.3 0.0% -0.6% 11.9 3.8 Sierra Leone 56.7 57.6 57.3 1.5% 1.O% 5.6 2.2 Somalia 0.0 0.0 0.0 0.0% 0.0% 8.5 Sudan 118.0 118.0 118.0 0.0% 0.0% 37.0 1.3 Tanzania 1521.9 1520.6 1542.7 -0.1% 1.4% 38.3 4.4 Togo 71.7 71.8 71.7 0.0% 0.0% 6.3 1.2 - 26 - Minimum Minimum Minimum Minimum Allocation Allocation Allocation Allocation Raised to Raised to Raised to Raised to SDRl.5 SDRl.5 SDRl.5 SDRl.5 Million Million Million Million and Cap and Cap and Cap and Cap at at at at Country Current SDR13.2 SDR19.8 SDR13.2 SDR19.8 Performance Formula pcpa pcpa pcpa pcpa Populatio Ratings SDR m SDR rn SDR rn Percent Percent millions Uganda 778.0 776.2 772.6 -0.2% -0.7% 29.9 3.6 Zambia 166.0 166.5 165.6 0.3% -0.3% 11.9 2.9 Zimbabwe 10.4 10.4 10.4 0.0% 0.0% 13.1 0.7 IDA Arrears Clearance 355.9 355.9 355.9 0.0% 0.0% Subtotal 13381.0 13375.7 13391.4 0.0% 0.1% East Asia East Asia Regional 48.0 48.0 48.0 0.0% 0.0% Cambodia 130.9 131.6 131.0 0.5% 0.1% 14.4 2.1 Indonesia 0.0 0.0 0.0 0.0% 0.0% 223.9 3.2 4.1 4.1 5.7 0.I 2.7 Lao PDR 49.3 50.0 49.8 5.8 2.1 Mongolia 51.8 52.6 52.4 1.7% 1.7% 2.6 3.2 Myanmar 0.0 0.0 0.0 0.0% 0.0% 51.0 Papua New CJumea 66.9 67.8 67.5 1.3% 0.9% 6.0 2.4 Samoa 7.9 7.9 11.6 0.0011 46.0?40 0.2 4.7 Solomon Islands 5.5 6.4 6.4 16.7% 16.2% 0.5 1.9 Timor-Leste 9.1 9.7 9.7 6.1% 6.1% 1.0 I.x 'Tonga 3.2 3.4 4.1 5.O% 28.2'K 0.1 2.1 Vanuatu 5.9 7.1 7.0 19.3% 18.9% 0.2 2.6 Vietnam 2508.5 2501.2 2491.1 -0.3% -0.7% 84.1 3.8 Subtotal 2891.1 2889.7 2884.4 -0.2% -0.2% South Asia Afghanistan 483.8 483.8 483.8 0.0% 0.0% 28.8 1.4 Bangladesh 1918.4 1913.1 1905.4 144.4 2.5 Bhutan 21.5 22.1 22.3 0.6 4.4 India 2829.6 2829.6 2829.6 0.0% 0.0% 1124.7 4.1 Maldives 11 2 12.3 12.3 10.0% 9.6% 0.3 3.7 Nepal 448.0 447.4 445.6 -0.1% -0.5% 27.7 2.9 Pakistan 1800.0 1800.0 1800.0 0.0% 0.0% 159.0 3.1 Sri Lanka 442.2 441.7 439.8 -0.1% -0.5% 19.7 3.5 Subtotal 7954.7 7950.0 7938.9 -0.3% -0.2 Yo ECA ECA Regional 48.0 48.0 48.0 0.0% 0.0% Armenia 111.3 112.1 111.6 0.7% 0.2% 3.0 4.6 Azerbaijan 158.6 159.1 158.5 0.4% -0.1Yo 8.5 3.3 - 27 - Minimum Minimum Minimum Minimum Allocation Allocation Allocation Allocation Raised to Raised to Raised to Raisedto SDRl.5 SDRl.5 SDRl.5 SDRl.5 iMillion Million Million Million and Cap and Cap and Cap and Cap at at at at Country Current SDR13.2 SDR19.8 SDR13.2 SDR19.8 Performance Formula pcpa pcpa pcpa pcpa Population Ratings SDR m SDR m SDR m Percent Percent mllions Bosnia-Herzegovina 74.8 75.6 75.3 1.2% 0.7% 3.9 3.4 Georgia 148.1 148.7 148.1 0.4% 0.0% 4.4 4.3 Kosovo 19.8 19.8 19.8 0.0% 0.0% 2.0 Kyrgyz Republic 66.4 67.I 66.8 1.1% 0.7% 5.2 2.6 Moldova 73.6 74.5 74.2 1.2% 0.8% 3.4 3.3 Tajikistan 62.5 63.2 63.0 1.2% 0.8% 6.6 2.2 Uzbekistan 143.8 144.4 143.8 0.4% 0.0% 26.5 1.6 Subtotal 906.8 912.5 909.0 1.0% 0.2% MNA Djibouti Yemen, Republic o f Subtotal 241.2 242.4 241.4 0.7% 0.1 Yo LAC OECS 48.0 48.0 48.0 0.0% 0.0% Bolivia Haiti 30.9 31.9 31.6 3.1% 2.4% 8.5 1.7 Honduras 153.6 154.0 153.2 0.3% -0.2% 7.2 3.8 109.7 109.2 5.5 7.1 Subtotal 482.2 485.7 491.9 4.4% 2.0% Total Unallocatedregional 96.0 96.0 96.0 0.0% 0.0% Hard term lending 346.1 347.2 346.2 0.3% 0.0% Grand Total 26299.2 26299.2 26299.2 0.0% 0.0% Notes: All amounts are net of grant discounts and of debt service due (MDlU) IDA15 projections assume a 20 percent increase incommitment authority and the same country indicators as inFYO8.In addition, it takes into account the anticipated graduations of countries from IDA, the proposals made in accompanying papers on fragile states and arrears clearance, and the proposed scale up ofregional projects. - 28 - ANNEX4: PIPELINE OFREGIONAL PROJECTSFORTHEIDA15 PERIOD Proposed IDMIBRD Project Size Region Project Name $millions Participating Countries Africa West Africa Power Pool - Phase I11- 1 100 Cote d'Ivoire and Mali for this APL Africa Central Africa Backbone (CAB - 40 CAR, Chad, Cameroon, Nigeria Africa West Africa Road Transport 135 Mali, Burkina Faso, Ghana Africa WA Reg. Project for Fisheries 90 Mauritania, Senegal, Gambia, Guinea-Bissau, Guinea, Cape Verde, Sierra Leone, Ghana Africa Africa ScienceTechnology 80 Burkina Faso, Nigeria, Tanzania Africa West & Cent. Air Trans. Safety APL. 3 50 SelectedW. and Central African Countries Africa S R B Multi-Modal Transport Project - Phase I1 130 Mali, Mauritania, Senegal, Guinea Africa SouthernAfrica - Trade & Transport Comdor 150 SADC member states Africa NELSAP RusumoFalls MP SIL 119 Rwanda,Burundi, Tanzania EuropeiCentralAsia Sangtuda IHydroelectric Project and transmission lines 40 Tajikistan, Kyrgyz Republic, Afghanistan, Pakistan Europe/Central Asia Regional Energy/Transport 15 Balkans (to be determined) LAC/OECS Regional E-Government 16 Dominica, Grenada, St. Lucia, St. Vincent Subtotal-pipeline in FY09 965 Africa West Africa Agricultural Productivity Program - APLII 45 Burkina Faso, ECOWAS countries Africa West Africa Multi DiseaseSurveillance Project 60 Ecowas plus R4auntania. Africa West Africa RegionalTelecom Program 80 Senegal, Cote d'lvoire, Guinea, Guinea-Bissau Liberia, Sierra Leone Africa SouthAfrica Power Pool - APL 111 '50 Zambia-Tanzania Africa Abidjan-LagosTrans.& Transit-ALTTFP 242 Nigeria, Benin, Togo, Ghana, Ccte d'lvoire Africa EasternNile Watershed Mgmt ProgramSIL 35 Ethiopia, Sudan (Egypt) Africa Zanibezi Project($25KVC) 120 to be determined Africa West Africa Power Pool - Phase IV - 1 90 Guinea, Guinea-Bissau, Gambia, Senegal Central Asia The Central Asia Infectious Diseases Control Project 16 Tajikistan, Uzbekistan,Kyrgyz Republic Subtotal-pipeline in FYlO 822 Africa EasternNile ENSAP Flood Preparedness SIL 35 Ethiopia, Sudan (Egypt) Africa East Africa Power Market APL - Phase I1(EAPP) 200 to be determined Africa SouthernAfrica Corridor HIV/AIDS Project 120 to be determined Africa Niger River Basin - APL2 120 Niger, Nigeria Kenya, Uganda, Rwanda, Burundi, Tanzania, Mozambique, Africa Telecommunications RCIP I11(APL 3) Madagascar, Zambia, Malawi, Ethiopia, Zimbabwe, Botswana, Lesotho, Swaziland, Sudan Djibouti, Somalia, South Afnca, Mauritius, Eritrea. Seychelles, Comoros Africa Southern Africa Agricultural Productivity Project 75 SADC member states Africa West Africa Power Pool - Phase I11- 1 60 BurkinaFaso, Ghana Africa West Africa Power Pool - Phase IV -2 50 Cote d'Ivoire, Guinea, Liberia Subtotal-pipelinein FYll 760 - 29 - ANNEX5: SHARE OFCOUNTRYALLOCATIONTO FINANCE REGIONAL PROJECTSIN ---- IDA14 140% - Fy06-FY07 ______ 7 100 1 so Small Allocation Size , Large Ii Country contribution to regional project + Share o f country allocation ,