IDA13 Grants in IDA13 Summarizing the Options International Development Association March 2002 Table of Contents Current Options for Grants in IDA .......................................................................................1 Costs of IDA Grants.................................................................................................................4 Providing Borrowers with a Choice of Grants or Credits ...................................................6 Implementing Grant Financing in IDA13 .............................................................................7 Summary...................................................................................................................................7 Table 1: Grants for specific uses for all IDA-only countries...........................................................2 Table 2: Grants for IDA-only countries with US$360 GNI/Capita and below..................................3 Table 3: Grants for specifics Uses for IDA-only countries with US$360 GNI/Capita and below.......4 Table 4: Loss of Repayments to IDA due to grants........................................................................5 Table 5: Equivalent Grant Amount...............................................................................................6 Chart: Estimated Annual Donor Contributions to IDA................................................................5 Grants in IDA13 ­ Summarizing the Options 1. Over the past several months, the issue of grants in IDA has been discussed in several fora; and most intensively in successive meetings of IDA Deputies. In this process various possibilities for the use of grants, as well as a number of combinations of such uses, have been discussed. The purpose of this note is to restate for IDA Deputies the options that have been put forward ­ to take stock of where we are, in order to facilitate the next phase of discussions leading to completion of the IDA13 replenishment. The paper also provides a brief review of the costs that donors would face to provide to IDA additional contributions to compensate for lost reflows through the provision of grants. Current Options for Grants in IDA 2. The dialog on grants has moved forward on two axes. One axis deals with specific `uses' for which grants are deemed to be the most appropriate instrument, and the other axis focuses on grant eligibility in the context of a country's level of poverty (with per capita GNI being a proxy for this measure). Using these two axes simultaneously, several combinations for the use of grants are possible, and are also described below. 3. Discussions by Deputies and the donor community about grants have been based on projections of lending across countries and sectors during the IDA13 period, and these same estimates form the basis of the figures in this note. The actual amounts of grants in IDA13 could differ materially however. Usage of the instrument will depend upon IDA's performance based resource allocations, especially for countries with a per capita income equal to or less than $360 (to the extent country income level is a criterion for grants), CAS priorities, the existence of strong programs in the sectors eligible for IDA grant financing, and other factors including countries' absorptive capacity, and the programs of other development partners. 4. The `Uses' Axis. In this line of proposals, grant eligibility would be determined according to specific uses and be available to all IDA-only countries (i.e., the poverty level would not be a criterion of eligibility). The approach ­ which builds upon ideas set out in IDA13 papers1 and discussed by Deputies ­ was taken up by some donors in / recent months, in the context of searching for a compromise that could form the basis of an agreement on IDA13. Some donors have continued to prefer a restrictive view, limiting grants to areas where there is a strong public-goods and externality argument, for example post-conflict support and regional disease programs. A variant of this approach proposed focusing grants in areas which were pre-requisites for development, such as early recovery efforts in post-conflict countries, the fight against communicable diseases 1/ "Grants in IDA13", September 2001. "Grants and Concessionality", November 2001. - 2 - at country as well as regional level, and recovery from major natural disasters. Others have proposed using grants also in sectors that would yield financial returns only in a long term perspective (such as primary education), or for capacity building in order to improve governance. 5. Table 1 sets out possibilities for grants to be deployed on the basis of uses. It draws upon preferences expressed by a range of donors in recent months in terms of the areas where grants would be seen as most appropriate, and tries to quantify those uses based upon IDA's most recent data. The table also reflects an apparent consensus developing among donors that in three areas ­ post-conflict, regional health, and natural disaster recovery ­ grants would be an appropriate instrument for IDA to deploy, and thus in all the scenarios presented 100% of the uses in column 1 would be covered by grants. As the table shows, overall grants in IDA constructed on the full range of such uses might range between 16% to 28%. Other combinations could also be constructed. For example, a stricter approach, as in the first column, would restrict grants to about 5%. Some discussions have stressed the high economic returns to education, and questioned the incentive effects of grant-based social protection: if these were to be removed from the scenario presented, the amounts indicated would be in the 10-18% range. On the other hand, to reach a level close to 40%, grants would need to be extended to IDA operations beyond areas related to emergency assistance and social sectors. Table 1: Grants for specific uses for all IDA-only countries a/ (Percentage) Post-Conflict, Regional Health, and Subtotal Capacity Natural Basic Social Social Building / b/ c/ Disaster Health Education Protection Water Sectors Governance Subtotal Total 1 2 3 4 5 6=2+3+4+5 7 8=6+7 9=1+8 1.1 100% grants for specific uses 5% 5.6% 5.6% 4.9% 3.5% 20% 2.9% 23% 28% (Cumulative Total) (5%) (11%) (16%) (21%) (25%) (28%) 1.2 100% grants for health, and education sectors; 50% for social protection, water, and governance. 5% 5.6% 5.6% 2.5% 1.8% 16% 1.4% 17% 22% (Cumulative Total) (5%) (11%) (16%) (19%) (21%) (22%) 1.3 50% grants for specific uses 5% 2.8% 2.8% 2.5% 1.8% 10% 1.4% 11% 16% (Cumulative Total) (5%) (8%) (11%) (13%) (15%) (16%) a/ Based on IDA13 resource needs. b/ Financed by 100% grants in all cases. c/ Figures in total and subtotal columns have been rounded. 6. The Country Poverty Axis: On this axis, grant eligibility would be based on national income level, and grants would finance a defined share of the country program not linked to specific uses. Many possibilities have been explored by donors over the course of the negotiations, including setting different grant shares for different income - 3 - groups. Much of the discussion among donors seems to have converged on a simple GNI cutoff for IDA grant eligibility at around $360 per capita ­ which would represent income of a dollar a day. Thus countries above this level would not have access to grants. IDA- only countries below this threshold would finance a certain portion of their IDA programs through grants. Table 2 sets out the share of overall grants in IDA13, assuming various levels of grants to IDA countries with GNI of $360 per capita or less. Table 2 - Grants for IDA-only Countries with $360 GNI/capita and below (Percentage) Grants Share Overall grants in IDA Allocation in IDA13 to eligible countries 2.1 50% 21% 2.2 40% 17% 2.3 20% 8% 2.4 10% 4% 7. Combining Poverty and Uses Axes. Grant eligibility would be for IDA-only countries with per capita incomes less than or equal to $360, and for specified uses. Several possibilities for the use of grants along these lines have also been proposed and discussed in recent weeks, focussing on sectors/areas where donors appear to have some interest in providing grants (as set out in Table 1 above), and in the context of countries with per capita income of $360 or less (illustrated in Table 2 above). 8. Table 3 illustrates these discussions by providing a set of "combination" scenarios. It assumes, as in Table 1, that funding for regional health/communicable diseases, natural disaster recovery and post-conflict (which are all areas where there appears to be broad donor agreement on the use of grants) would be financed by IDA grants regardless of countries' per capita incomes, while the use of grants for other types of operations would be linked to countries' poverty levels. In the scenarios provided below, the overall share of grants range from 12% to 18%. Once again, other combinations are also possible: for instance if scenario 3.1 were expanded to permit exceptional grant financing for all HIV/AIDS programs in IDA-only countries, irrespective of GNI, the overall level of grants would quite possibly rise to around 20%. - 4 - Table 3: Grants for specific uses for IDA-only countries with $360 GNI/capita and below a/ (Percentage) Post-Conflict, Regional Health, and Subtotal Natural Basic Social Social b/ c/ Disaster Health Education Protection Water Sectors GovernanceSubtotal Total 1 2 3 4 5 6=2+3+4+5 7 8=6+7 9=1+8 3.1 100% grants for specific uses 5% 3.3% 3.3% 2.9% 2.0% 11% 1.7% 13% 18% 100% grants for health, and education 3.2 sectors; 50% grants for social 5% 3.3% 3.3% 1.4% 1.0% 9% 0.8% 10% 15% protection, water, and governance. 3.3 50% grants for specific uses 5% 1.6% 1.6% 1.4% 1.0% 6% 0.8% 7% 12% a/ Based on IDA13 resource needs. b/ Financed by 100% grants in all cases. c/ Figures in total and subtotal columns have been rounded. Cost of IDA Grants 9. Donors have also discussed in recent weeks the cost of providing grants, since the fact that grants do not generate reflows will ceteris paribus reduce the ability of the institution to make new loans in the future. In order not to impair IDA's capacity to provide future assistance to its borrowers, donors could replace the loss of the principal repayments and service charges (commitment charges on IDA credits are currently set to zero) provided under standard IDA credit terms. 10. Table 4 sets out the cost to donors of IDA providing three hypothetical levels of grants, if donors were to finance "grant costs" on a pay-as-you-go (PAYG) basis.2/ As the table shows, for 40% grants, donors would need to provide IDA with cash injections of the order of $30 million a year for the first ten years, $570 million a year for the second decade, $1.8 billion a year for the third decade and $3.5 billion a year for the fourth decade, totaling $59 billion.3 This "back-loaded" structure of costs has prompted / discussion of ways to ensure that they are recognized and managed from the outset, rather than postponing action to a future when their dimensions may be unmanageable. For illustration, the Chart therefore shows how donor contributions would need to increase, over the 40-year period, to provide these resources by means of commitment to a steady and sustained annual increase in donor contributions to IDA. 2/ As in all discussions on IDA grants and costs, these proceed from the "base case" financial assumptions used by Deputies throughout the replenishment negotiations, under which IDA lending is maintained in real terms at the IDA13 level, while donor contributions remain broadly stable in nominal terms with rising repayments making up the difference. 3/ These estimates assume encashment of these additional contributions at the time they are provided. - 5 - Table 4. Loss of Repayments to IDA due to Grants (USD, Nominal Value) Share of Grants Average Annual Loss of Repayments due to Grants 1/ Total Loss of Repayments due to Grants First Decade Second Decade Third Decade Fourth Decade Over the Next 40 years 10% IDA Grants 8 m 140 m 0.5 b 0.9 b 15 b 20% IDA Grants 15 m 280 m 0.9 b 1.7 b 30 b 40% IDA Grants 30 m 570 m 1.8 b 3.5 b 59 b Estimated Annual Total Donor Contributions to IDA 1/ 8,000 8,000 Average annual growth rate: 2%2/ 7,000 7,000 6,000 6,000 Average annual growth rate: 1.2% 5,000 5,000 Average annual growth rate: 0.7% Million 4,000 4,000 SDR SDR 3.3 billion 3,000 3,000 2,000 2,000 1,000 1,000 - 2003 2005 2007 2009 2011 2013 2015 2017 2019 2021 2023 2025 2027 2029 2031 2033 2035 2037 2039 2041 Estimated donor contributions including compensation for loss of commitment authority due to 40% grants and HIPC 3/ Estimated donor contributions including compensation for loss of commitment authority due to 20% grants and HIPC Estimated donor contributions including compensation for loss of commitment authority due to 10% grants and HIPC Estimated donor contributions if there is no loss of repayments due to grants or HIPC 1/ Estimated annual donor contributions include SDR 3.3 billion from the base-case scenario and an additional amount corresponding to the loss of commitment authority due to loss of repayments from grants and HIPC, both of which encashed under the standard schedule. 2/ Represent average annual growth of total donor contributions to IDA assuming full replacements of lost repayments from grants and HIPC. Actual donor increases would vary from year to year, and would have to be slightly front-loaded in the earlier years to deal with IDA's losses from HIPC. 3/ Based on actual FY01 lending data, assuming 50% for IDA-only portfolio represents approximately 40% of IDA's overall program. - 6 - Providing Borrowers with a Choice of Grants or Credits 11. An approach that would be resource neutral to IDA, and hence its donors, would be to provide IDA borrowers with a choice of drawing down either credits or a financially equivalent amount of grants. This approach had come up briefly in the context of Deputies' discussions about increasing the overall concessionality of IDA, but is set out in more detail here. The grant amount that could be provided in lieu of a credit would be smaller, with the difference set aside to cover (i.e., defease) the cost of future reflows and service charges that would have accrued to IDA had the financing been provided as a credit. Thus, the net grant amount would depend on the investment rate at which the defeasance portion could be invested to eventually cover the lost reflows and service charges as they occur. Table 5 below sets out the grant amount that would be equivalent to a credit at different assumed investment rates. If the investment return were around 6%, the grant amount would be roughly two-thirds of the available credit amount, while an investment rate of 4% would yield a net grant amount of slightly less than half of the credit amount. In order to reduce the dependence of the grant equivalency on interest rates -- and/or to increase the concessionality of the financial transfer to IDA clients -- one variation of this approach would be to fix the credit/grant equivalency in advance, with donors undertaking to make up the shortfall, if any, in the defeasance portion. Table 5: Equivalent Grant Amount 1/ Net Grant Amount Corresponding to an Investment Return Assumption 2/ SDR 100 IDA Credit 4% 47 5% 56 6% 63 6.3% 3/ 64 Note: 1/ IDA would invest the difference between 100 and the net grant amount, and use this fund to defease the expected loss of repayments (both principal and service charges). 2/ The average return on IDA's liquidity portfolio for the last three years is approximately 5%. 3/ Equivalent to 2000 SDR composite CIRR rate, used in the IDA grant discussions as the discount rate for grant element estimates. This is also the rate used for concessionality/PV calculations in the HIPC context. 12. Providing borrowers with such a financial mechanism to permit choice between grants or credits could help ensure that IDA's finances are not adversely affected by the extension of grants, and in turn would not require additional contributions from donors for grants. Replacing credits with grants could be useful in country settings where debt sustainability concerns limit even concessional borrowings, or alternatively where - 7 - countries have access to amounts of concessional resources from other sources which permit adequate financing of development strategies even with smaller volumes from IDA than would be available on a credit basis. In other IDA countries, however, urgent and immediate development needs could reduce the flexibility for countries to forego credits in return for a smaller amount of grants. Implementing Grant Financing in IDA13 13. While the grant schemes set out above are implementable, they would bring a considerably greater level of complexity to the management of IDA resources and operations. To the extent a country poverty cutoff is agreed for grant eligibility, decisions would need to be made each year, on the basis of the most recent GNI and population data, regarding countries that qualify for grants, and also methods would need to be devised to handle eligibility problems at the cut-off boundary. IDA's aggregate resources (credits and grants) should of course continue to be allocated to individual countries on the basis of performance (and reflected as such in CASs). Grants for regional public goods/natural disaster would, however, not be allocated through this process, but would need to be kept to a small proportion of IDA's overall financial assistance. The use of grants in programmatic operations will also need to be clarified. To the extent that specific sectors are eligible for grants, it would be reasonable to expect that programmatic operations that support such sectors could be provided, at least in part, by grants. Summary 14. This note has recapitulated a range of issues on grants currently under discussion among donors, with estimates of financing needs updated on the basis of IDA's current projections. As noted earlier, the grant estimates provided are indicative, and actual resource allocation in IDA13 will depend on a number of factors. Grants could also be made available to borrowers as a possible option, in lieu of credits. In that instance, there would be no special need to specify a share of grants in IDA in advance ­borrowers would choose the instrument best suited to their needs, at no additional cost to IDA or its donors. Some donors have also suggested that it would be useful to take a gradualist approach, in which IDA13 could earmark a relatively modest amount of IDA resources specifically for grants, and also provide some degree of choice to borrowers in their access to grants. On the basis of experience with grants gained in IDA13, a more comprehensive approach could be adopted in IDA14 and subsequent replenishments.