The G-8 Debt Cancellation Proposal and Its Implications for the Fund

The G-8 Debt Cancellation Proposal and Its Implications for the Fund


The G-8 Debt Cancellation Proposal and Its Implications for the Fund

I. Introduction

1. The G-8 has proposed that the Fund, the International Development Association (IDA), and the African Development Fund (AfDF) cancel 100 percent of their claims on countries having reached, or upon reaching, the completion point under the enhanced HIPC Initiative. The proposal was initially presented in the G-8 Finance Ministers’ Conclusions on Development, issued on June 11, 2005, and reaffirmed in the statement on Africa signed by G-8 Heads of State and Government at the Gleneagles Summit on July 8, 2005. A more detailed technical note from the G-8 was circulated to the Executive Board on July 15, 2005. The objective of the proposal is to complete the process of debt relief for HIPCs by providing additional resources to help these countries, most of which are in Africa, reach the Millennium Development Goals (MDGs). At the same time, G-8 countries have committed to ensure that the financial capacity of the international financial institutions (IFIs) is not reduced as the result of debt forgiveness. The cost of debt relief to IDA and the AfDF is to be met by bilateral contributions based on agreed burden sharing; the cost to the Fund is to be covered through the institution’s own resources, with a call for bilateral contributions to cover additional needs. The utilization of appropriate grant financing is expected to ensure that countries do not immediately re-accumulate unsustainable external debts and are eased into new borrowing.

2. In their initial discussions of the G-8 proposal on June 22, 2005, Executive Directors stressed that the Fund will continue to operate under existing policies and procedures until decisions to change these policies are taken by the required majorities. Directors pointed out that the G-8 proposal raises issues regarding the uniformity of treatment of members, burden sharing, conditionality, and the Fund’s financial role in low-income countries. Directors asked staff to prepare an assessment of the proposal, its financial, legal, and policy implications for the Fund, for Board discussion prior to the Annual Meetings. Directors emphasized that the Fund’s financing capacity should not be undermined, in particular, its capacity to help other low-income countries within the framework of its established policies.

3. The Executive Board had also expressed a range of views on debt relief beyond the HIPC Initiative in a March 2005 Board seminar. Directors saw merit in providing further debt relief to low-income countries but thought its benefits must be weighed against other potential uses of scarce resources. They emphasized that further debt relief should not be considered in isolation from questions regarding the role of the Fund in low-income countries. The Fund should only consider providing further debt relief as part of a broader international initiative and only if additional financing could be mobilized that would not jeopardize future PRGF lending operations.

4. This paper responds to the Board’s request for further analysis of the G-8 proposal and possible implementation modalities. Section II provides a description of the proposal as presented by the G-8. Section III estimates the financial implications of this proposal for the Fund. Section IV discusses legal and policy issues raised by the proposal. Section V discusses needed approvals and consents. Section VI presents a possible plan for taking the proposal forward and issues for discussion.

II. Description of the G-8 Proposal

5. The key elements of the G-8 proposal of debt cancellation, which has been formally submitted to the Fund, World Bank and African Development Bank for consideration ahead of the 2005 Annual Meetings of the Fund and the Bank, are:

  • 100 percent, irrevocable, Fund, IDA, and AfDF debt stock relief on debt contracted prior to January 1, 2005 for countries that reach completion point under the HIPC Initiative (Table 1).

  • Debt relief should be additional to previously committed assistance, with the additional assistance provided through IDA and the AfDF allocated to recipient countries on the basis of country performance. Good governance, accountability and transparency are described as crucial to releasing the benefits of debt cancellation.

  • All countries receiving the relief must be current with their repayment obligations to the IFIs. In addition, countries must not have experienced serious lapses, including in governance, such that their ‘IMF programs’ would be at risk.

  • For IDA and the AfDF, debt stock cancellation would be delivered by relieving post-completion-point HIPCs’ repayment obligations and adjusting their gross assistance flows downward by the same amount. Donors would make additional contributions to IDA and the AfDF to match dollar-for-dollar the foregone principal and service charge repayments of the debt canceled, based on agreed burden sharing. Additional funds would be allocated to all IDA-only and AfDF recipients on the basis of existing performance-based allocation mechanisms. The additional funds to cover the full costs due during the IDA-14 and AfDF-10 period would be made available immediately; and donors would commit to make additional contributions to the subsequent regular replenishments of IDA and the AfDF to cover the full costs until all obligations under the canceled loans are extinguished.

  • For the Fund, the proposal does not consider any immediate allocation of additional resources, on the assumption that the Fund will be able to cover fully with its existing resources the costs of canceling debt to current HIPCs (excluding the protracted arrears and potentially the sunset clause cases) without undermining the Fund’s financing capacity. The proposal also assumes that the self-sustained PRGF should be adequately equipped to be responsive to short-term adjustment needs and to meet future demand as assessed by the Fund.

  • However, donors would commit to provide the additional resources necessary to cover the cost of debt cancellation at the completion point for the three protracted arrears cases (Somalia, Liberia and Sudan). Donors would also commit to provide, if necessary and on a fair burden share basis, resources to cover the possible costs of debt relief to those countries that may become eligible for HIPC assistance under the extended sunset clause. In addition, subject to further analysis, G-8 countries have committed to provide, on a fair burden share basis, up to $350-500 million to the Fund for the interim PRGF.

  • The proposal invites voluntary contributions, including from oil-producing states, to a new trust fund to support poor countries facing commodity price and other exogenous shocks. This would complement an exogenous shocks window in the PRGF Trust.

  • The Bank and Fund are to report on improvements in transparency at both ends (i.e., for donors and recipients) and the drive against corruption so as to ensure that all freed resources are used for poverty reduction.

  • Looking ahead, the proposal has the objective of ensuring that countries do not immediately re-accumulate debt levels that could become unsustainable, and are eased into new borrowing. In addition to appropriate grant financing as agreed, mechanisms compatible with the Debt Sustainability Framework need to be in place to ensure that countries do not borrow excessively from sources other than IDA and the AfDF.

Table 1.

Country Coverage of the G-8 Proposal

article image

6. The proposal identifies specific financial resources to cover the Fund’s costs of providing 100 percent debt relief to current HIPCs (excluding protracted arrears and sunset clause cases). They include:

  • the “corpus” of the 1999 off-market gold transactions that were undertaken to finance the HIPC Initiative, now held in the Special Disbursement Account (SDA);

  • the accrued investment income from these profits not yet transferred to the PRGF-HIPC Trust, also currently held in the SDA;

  • cash resources already disbursed to members and set aside in the HIPC Umbrella Account—these resources have been assigned to cover, as they come due, specified obligations to the Fund of countries that either have reached the completion point or are receiving interim assistance between the decision and completion points; and

  • the share of the PRGF Trust’s Subsidy Account corresponding to the projected cost of subsidizing Fund lending to HIPCs until end-2006 under the interim PRGF.

7. The combined amount of these resources should be sufficient to cover the claims outstanding on current HIPCs (excluding protracted arrears and sunset clause cases, and assuming (without prejudging) a cut-off date of January 1, 2005 and an implementation date of January 1, 2006). If necessary, donors would commit to provide (on a fair burden share basis) resources to cover possible costs of relief for potential new sunset clause cases.

III. Financial Implications of the Proposal for the Fund

8. The G-8 proposal has potentially significant financial implications for the Fund and its ability to support its low-income members. As noted above, unlike for IDA and the AfDB, the proposal does not provide new resources to finance debt relief or future lending operations by the Fund. Rather, it envisages use of the Fund’s existing resources for this purpose, while committing to provide additional resources in certain areas (e.g., the protracted arrears cases). The two main financing sources envisaged for additional debt relief by the Fund are the resources in the Special Disbursement Account (SDA) resulting from the 1999-2000 off-market gold sales and part of the resources in the PRGF Trust Subsidy Account. As discussed further below, without new resources, this could substantially curtail the Fund’s ability to lend under the interim PRGF. It could also have implications for the Fund’s future concessional lending over the medium-term, its income position, and balance sheet that would need to be carefully considered.

A. Cost of the G-8 Proposal

9. Cost estimates for the G-8 proposal are sensitive to the assumptions made regarding implementation modalities. These include, among others, the choice of a cut-off date and the coverage of obligations falling due after the cut-off date but before the initiative is effective.

10. The cost estimates presented in this paper are based on a fixed cutoff date for all 35 HIPCs at end-December 2004. This implies a debt stock of SDR 4.2 billion, and would exclude any new lending from that date forward. Table 2 illustrates how the eligible stock of credit outstanding would differ if a different cutoff date were chosen.

Table 2.

Fund Credit Outstanding to 35 HIPCs, December 2004–December 2005

(In millions of SDRs)

article image

Repayments during January-April 2005 include settlement of short-term principal arrears incurred in 2004 amounting to SDR 3.2 million.

Includes disbursement of interest accumulated in the HIPC Umbrella Accounts.

Includes Board approved HIPC assistance as of end-April 2005 and projected new HIPC disbursements during May-December 2005.

Projected disbursements include both those under existing and anticipated new programs, consistent with the projections in EBS/05/89.

11. A second consideration is the treatment of debts falling due after the cutoff date but before the debt relief initiative becomes effective, or (in the case of countries not yet at the completion point) before a country reaches the completion point. The former is an issue even for existing completion point countries; in addition, a significant amount of time could elapse before some countries reach the completion point. Under the G-8 proposal, countries would be expected to remain current on repayments and would not be compensated for such repayments. The cost to the Fund would, therefore, be less than the credit outstanding to the 35 HIPCs as of end-2004.1 The discussion here is based on the assumption that the debt relief initiative becomes effective as of end-2005. Moreover, it uses the simplifying assumption that pre-completion point HIPCs would also receive full debt stock cancellation as of end-2005.2 As shown in Table 3, these assumptions imply a total cost to the Fund of SDR 3.5 billion.

Table 3.

Fund Credit Outstanding and Projected Principal Repayments of 35 HIPCs

(In millions of SDRs as of December 31, 2004)

article image

HIPC assistance is not included in credit outstanding.

Full debt relief is assumed to be granted based on end-2004 credit outstanding. For completion, decision, and pre-decision point countries, debt relief is assumed to be implemented at end-2005.

Includes outstanding SAF credit of SDR 0.3 million.

Includes outstanding SAF credit of SDR 36.4 million.

12.The incremental cost to the Fund would be less than the remaining eligible credit outstanding of the 35 HIPCs at end-2005. Part of the cost is already being financed through the enhanced HIPC Initiative. Of the 35 HIPC-eligible countries, 18 have reached completion point and received HIPC debt relief disbursements, which are held in escrow by the Fund in the HIPC Umbrella Account. Table 4 shows the stock outstanding and these balances as of end-April 2005. The available balance in the Umbrella Account at end-2005 is projected at SDR 0.3 billion, which could be used for financing debt relief. In addition, the Fund has already identified resources for future disbursements under the enhanced HIPC Initiative, the cost of which, as of end-2005, is currently estimated at about SDR 0.6 billion. Second, subsidy resources already available to reduce the interest rate on PRGF loans to 0.5 percent would be freed up by full debt stock cancellation. The estimated NPV amount of subsidy resources on the outstanding stock of Fund credit for the 35 HIPCs is SDR 0.4 billion. Adjusting for these factors, the incremental cost for the Fund of the proposal would be about SDR 2.2 billion (Table 5).

Table 4.

Fund Credit Outstanding to 35 HIPCs

(In millions of SDRs as of April 30, 2005)

article image

Includes outstanding SAF credit of SDR 0.3 million.

Includes outstanding SAF credit of SDR 36.4 million.

Table 5.

Incremental Cost of Proposed Additional Debt Relief to 35 HIPCs

(In millions of SDRs; Cutoff Date at Dec. 31, 2004; Effective Date at Dec. 31, 2005)

article image

In NPV terms as of end-2005.

13.The G-8 proposal does not cover two elements of potential costs relating to debt relief, but specifically includes a commitment to provide the additional resources needed to meet them:

  • the costs related to debt relief for the three protracted arrears cases (Liberia, Somalia and Sudan). The cost for the Fund of HIPC relief for the former group is estimated at SDR 1.5 billion, while the cost of full debt relief would be SDR 1.8 billion.

  • the cost of canceling the debt of countries that would become eligible for HIPC relief under the extended sunset clause.3 The exercise to determine possible additions to the HIPC-eligibility list is not yet finalized.

B. Financing Needs and Sources

14.As noted, the G-8 proposal envisages use of the Fund’s existing financial resources potentially available for all its low-income members to finance further debt relief for the 35 HIPCs. The financing would be obtained from the following sources:4

  • SDR 2.5 billion of cash resources held in the SDA, comprising the corpus of the 1999-2000 off-market gold transactions (SDR 2.2 billion) and investment income not yet transferred to the PRGF-HIPC Trust (SDR 0.3 billion).

  • cash resources set aside in the HIPC Umbrella Account (SDR 0.3 billion).5

  • the remaining amount is to be obtained from existing cash resources held in the PRGF Subsidy Account. Based on a total cost estimate of SDR 3.5 billion as noted above, this amount would be about SDR 0.7 billion.

15.Implementation of the proposal would thus fully drain the resources in the SDA and significantly reduce the available balance in the PRGF Subsidy Account. The latter would decline to around SDR 1.1 billion compared with SDR 1.8 billion at present.6 Additional inflows into the PRGF and PRGF-HIPC Trusts are anticipated amounting to less than SDR 0.2 billion, reflecting future expected bilateral grant contributions and income on deposit contributions. Thus, remaining subsidy resources, including projected inflows, could amount to close to SDR 1.3 billion.

16.The reduction in subsidy resources could curtail remaining interim PRGF operations. The remaining PRGF loan resources of SDR 2.7 billion were expected to be sufficient to continue interim PRGF operations through at least end-2006 or into 2007. If subsidy resources on the order of SDR 0.7 billion were earmarked for debt relief, however, the remaining subsidy resources would become a constraining factor on the life of the interim PRGF. In particular, the above-estimated subsidy resources of close to SDR 1.3 billion would need to cover the subsidy needs on: (i) outstanding PRGF credit for all non-HIPCs (of the order of SDR 0.4 billion); (ii) undrawn balances from existing PRGF arrangements, which would be assumed to continue (about SDR 0.3 billion); and (iii) outstanding PRGF credit of HIPCs, including pre-cutoff date debt until the effective date of cancellation (about SDR 0.1 billion).

Table 6.

Existing Resources Earmarked for HIPC Debt Reduction Operations and Concessional Lending

(In millions of SDRs as of end-April 2005)

article image

Including SDR 68 million in the PRGF-HIPC Trust, most of which can be used for current HIPC operations.

17.This would leave an amount of PRGF subsidy resources on the order of SDR 0.5 billion available to cover other needs. These include new PRGF arrangements with non-HIPCs, new PRGF arrangements for pre-decision and pre-completion point HIPCs, PRGF arrangements for the protracted arrears cases, and meeting the financing needs associated with exogenous shocks. Allowance would also need to be made for the needs of HIPCs for new financing from the Fund after debt relief. The above amount would allow total new lending of only about SDR 1.9 billion under the interim PRGF, thus reducing the scope for new initiatives such as a shocks facility and suggesting a need to advance the start date for the self-sustained PRGF.

18. The above constraints could be eased if additional subsidy resources were forthcoming. In this context, the G-8 proposal includes a commitment to provide the Fund up to $350-500 million in additional resources over the next three years to meet difficult-to-forecast costs. Preliminary staff estimates suggest that an amount of about SDR 0.3 billion would meet the subsidy needs of new PRGF lending associated with arrangements to bring the pre-completion point HIPCs to the stage where they would qualify for full debt relief. The G-8 proposal also invites voluntary contributions, including from the oil-producing states, to a new trust fund to support poor countries facing commodity price and other exogenous shocks. Also, as noted, the proposal includes a commitment to provide additional resources to address the needs of the protracted arrears and sunset clause cases.

19. The financing needed to maintain the Fund’s lending capacity beyond the interim PRGF would need to be assessed in light of the G-8 proposal. the Fund’s longstanding intention has been to place PRGF lending on a self-sustained basis once interim PRGF resources are exhausted. The self-sustained PRGF was to be financed through the revolving use of the resources in the PRGF Reserve Account (see Box 1), supplemented as needed by bilateral loan resources. Staff has previously estimated the capacity of the self-sustained PRGF at about SDR 0.6 billion per annum.7

The PRGF Reserve Account

The Reserve Account of the PRGF Trust is designed to (i) provide security to the lenders to the PRGF Trust Loan Account in the event of a delay or nonpayment by PRGF borrowers; (ii) meet temporary mismatches between repayments from borrowers and payments to lenders; and (iii) cover the Fund’s costs of administering PRGF operations. It has also been long envisaged that, after bilateral PRGF loan and subsidy resources are exhausted, a self-sustained PRGF would start through a revolving use of the Reserve Account resources. 1

The Reserve Account has been financed by reflows of SAF and Trust Fund repayments, which were originally financed by the proceeds from gold sales in the 1970s. As of end-April 2005, the Reserve Account held a balance of SDR 3.2 billion.

The Reserve Account balance is affected by several factors: (i) transfers to the GRA to meet the costs of administering the PRGF Trust; (ii) investment income on the account balance; (iii) reflows of SAF and Trust Fund loans primarily arising from the clearance of arrears by Liberia, Sudan and Somalia; (iv) PRGF repayment by Zimbabwe once arrears to the Fund are cleared; and (v) retransfers from the PRGF-HIPC Trust for the temporary use of Reserve Account resources.

The first of these factors—reimbursement of the GRA for PRGF administrative expenses—could have a crucial impact on the balance. In 1998, with the initiation of PRGF-HIPC operations, the Executive Board decided to forego reimbursement for FY 1998-2004 and instead, transfer an equivalent amount to the PRGF-HIPC Trust as a Fund contribution to the Trust. The Board subsequently approved two more decisions in 2004 and 2005 to defray the reimbursement of the PRGF administrative expenses from the GRA for FY 2005 and FY 2006. In FY 2005, for example, PRGF administrative expenses were SDR 54 million.

However, once a self-sustained PRGF begins, reimbursement must begin, as Article V, Section 12(i) requires that the SDA reimburse the GRA for expenses associated with administering the SDA, which would include the administration of resources in the Reserve Account. If the GRA can be reimbursed from the investment income generated by the SDA corpus, then there would be no impact on the self-sustained commitment capacity. However, if the SDA corpus is fully used, reimbursement would have to come directly from the Reserve Account balances.

The projected commitment capacity of the self-sustained PRGF is, of course, closely linked to the balance of the PRGF Reserve Account. The Reserve Account has historically provided reserve coverage to lenders at a ratio of 40 percent. Thus, the commitment capacity is affected by the profile of projected PRGF credit outstanding, which, in turn, is sensitive to interim PRGF demand projections and the assumed disbursement profile. In the absence of additional debt relief, current projections indicate a commitment capacity of about SDR 0.6 billion per year.

1/ The PRGF Trust allows transfers of up to SDR 250 million from the Reserve Account to be used to provide Trust grants or Trust loans from HIPC and interim PRGF operations. Transfers are to be made only when and to the extent that there are no other resources immediately available for this purpose. To date, there have been two transfers totaling SDR 43.5 million from the Reserve Account to the PRGF-HIPC Trust.

20. The G-8 proposal is designed to preserve the resources in the Reserve Account. This is important to minimize the impact on the future loan capacity of the self-sustained PRGF. However, the proposal does envisage the exhaustion of the resources in the SDA, which could affect the future loan capacity in three ways:8

  • First, the SDA would no longer be available to finance the cost of PRGF administrative expenses, which could therefore fall on the Reserve Account. Once resources in the Reserve Account are used for PRGF lending, as envisaged under the self-sustained PRGF, there is a legal requirement under the Articles for the GRA to be reimbursed for administrative expenses of the SDA that arise in connection with administration of the PRGF. While such reimbursement could be paid using either SDA resources or the Reserve Account, it was previously envisaged that the investment income on the SDA corpus could be used for this purpose in order to maximize the resources available in the Reserve Account for self-sustained lending. Under the G-8 proposal, this would no longer be possible and reimbursement would have to come from the Reserve Account. This could affect the self-sustained PRGF loan capacity, unless it is supplemented by new bilateral loan resources.

  • Second, the use of the SDA corpus reduces the Fund’s future flexibility in its financial support for low income countries. For example, SDA resources would have potentially been available to deal with the additional subsidy needs associated with a sharper-than-expected future rise in interest rates.9

  • Third, under current projections, about SDR 1 billion of the investment income on the SDA corpus would have been available, once earned, as a surplus in the PRGF-HIPC Trust that could potentially be used for future PRGF operations, and this flexibility would also be eliminated (see Box 2).

Financial Impact of Use of SDA Resources

The use of the gold corpus of SDR 2.2 billion in the Special Disbursement Account (SDA) as envisaged under the G-8 proposal would have an impact on the Fund’s finances in the following four key areas:

  • Use of the corpus of the profits from the 1999-2000 gold sales would result in the loss of an asset that would have been available for future use. Although no Board decision was made in 1999 on the ultimate use of the SDA corpus after generating the necessary investment income for PRGF-HIPC Trust operations, the assets could potentially have been used for future concessional lending, or transferred to the GRA to offset the enlarged liabilities in reserve tranche positions (see discussion below).

  • The SDA corpus was financed through an equivalent enlargement of reserve tranche positions, which remain as a liability on the Fund’s balance sheet. The increase of reserve tranche positions resulted from the receipt of gold, rather than usable currencies, in the off-market gold transactions. Use of the SDA corpus for the proposed debt forgiveness would therefore weaken the balance sheet.

  • The commitment capacity of a self-sustained PRGF would be lowered due to the required reimbursement of the GRA for PRGF administrative costs, which would have to be made from the Reserve Account. Self-sustained PRGF operations were anticipated to use SDA resources, rather than the Reserve Account, for annual reimbursements to the GRA. Investment income from the gold corpus could have been used for this purpose.

  • Potential resources in the order of SDR 1 billion would be foregone. The resources for HIPC (up to SDR 1.76 billion), to be generated from investment of the SDA corpus following the 1999 off-market gold transactions, were expected to be generated over time. Temporary sources of funds were therefore obtained through internal borrowings, within the PRGF-HIPC Trust, from resources that were otherwise designated for PRGF subsidy operations. As of end-April 2005, only SDR 164 million of the SDR 1.76 billion had been disbursed for HIPC assistance; of the remaining investment income of SDR 1.6 billion, an estimated SDR 0.5 billion was projected for HIPC disbursements, with the residual of about SDR 1 billion on a cash basis repayable on earlier borrowings, available for interim PRGF operations, or potentially as a surplus to the PRGF-HIPC Trust.

Projected HIPC costs and SDA resources under the current HIPC Initiative /1

(In billions of SDRs; on a cash basis)

article image

Projected amounts are highly sensitive to assumptions on interest rates and timing of delivery.

The current decision caps the use of investment income for HIPC at SDR 1.76 billion on a cash basis.

Comprising existing resources, expected grant contributions, and projected interest income from deposits and grant contributions.

21. The resource needs to support future PRGF lending will depend on both the level of expected loan demand in light of the G-8 proposal and the modalities of future lending. The former issue is discussed in the next section. As regards the modalities, as noted, a shift to the self-sustained PRGF involving lending from the Reserve Account would require the GRA to be reimbursed for SDA-related PRGF administrative expenses, thus lowering the self-sustained capacity. This capacity could be supplemented by additional bilateral loan resources. Alternatively, the Board could decide to use Reserve Account resources to subsidize new lending based on bilateral loan contributions while continuing not reimbursing the GRA for administrative expenses. Under this model, initial estimates suggest that Reserve Account resources would provide sufficient subsidy resources to allow an annual loan capacity of around SDR 0.6 billion on a permanent basis. However, increasing this capacity temporarily would require additional subsidy resources. For example, to increase the annual loan capacity to SDR 1 billion through 2011, while ensuring a self-sustained PRGF capacity of SDR 0.5-0.6 billion thereafter, could require subsidy resources of the order of SDR 0.7-1.1 billion.

22. This discussion also assumes that the current framework for lending to low-income countries through the PRGF Trust is maintained. The Board has previously considered and rejected an alternative approach of using the Fund’s general resources for financing the principal for future PRGF-type operations. This issue was last discussed in March 2004, at which time most Directors opposed such an approach on the grounds that it was not consistent with the monetary character of the Fund, would compromise the short-run revolving nature of GRA resources, could increase the Fund’s risk exposure, and would depart from the current institutional framework for providing concessional financing.10 This framework is maintained under the G-8 proposal.

23. In sum, significant resources would be needed for the Fund to continue to support its low-income members in the context of the G-8 debt proposal. First, resources must be provided to meet the financing needs associated with eventual debt relief for the protracted arrears cases (SDR 1.8 billion). Second, resources will be needed to meet the cost for the Fund of extending the HIPC sunset clause, which is presently unknown but could be sizable in the context of 100 percent debt cancellation. Third, subsidy resources will be needed to cover the PRGF loans associated with bringing those HIPCs that have not yet reached completion point to that stage (preliminary staff estimates put this amount at SDR 0.3 billion). And fourth, additional subsidy resources would be needed to allow a higher level of PRGF lending over the medium term than is possible using the resources in the Reserve Account alone. Estimates of the latter are highly sensitive to the assumptions made regarding loan demand and the modalities of future lending, including the need for reimbursement to the GRA of PRGF expenses, but could be of the order of SDR 0.7-1.1 billion. The commitments in the G-8 proposal to providing additional resources will therefore be important in this regard.

C. Mitigating the Impact on the Fund’s Finances

24. The use of the SDA corpus also raises new issues for the Fund’s General Department, specifically the GRA balance sheet and the Fund’s income position. The on-going effect of the 1999-2000 off-market gold transactions on the Fund’s income position is currently being burden-shared among debtor and creditor members through adjustments to the rates of charge and remuneration, with additional income being placed in the SCA-1. This results in a refundable mechanism being used to cover a permanent increase in the Fund’s costs. Once sufficient income has been generated to meet HIPC assistance needs, the Executive Board could in the future decide to make the SDA corpus available to address the impact on the GRA of the off-market transactions, either by making the GRA balance sheet “whole” again through a transfer of the corpus, or by meeting the ongoing cost through a transfer of investment income on the corpus. Under the G-8 proposal, this would no longer be possible, and an alternative solution would need to be found if it were to be determined that the permanent impact on the GRA should be addressed.

D. Future Fund Lending

25. G-8 countries have indicated their willingness to provide additional resources to cover potential shortfalls in the financing of Fund support to low-income countries, subject to further analysis by the Fund. This section considers the rationale and possible demand for Fund financial support after debt cancellation.

Future Fund Lending to HIPCs

26. While there are general arguments against new lending after debt cancellation, staff sees a rationale for continued Fund lending to post-completion-point HIPCs, depending on their financing needs. The main argument against new lending is that debt relief now may raise expectations of similar actions in the future. To this extent, the new relief would reduce incentives for prudent borrowing policies and raise prudential concerns for creditors. Under this reasoning, new financing should either be stopped or take the form of grants. However, these arguments do not fit strictly with the nature and motivation of the G-8 proposal.

  • First, from a prudential perspective, there is a difference between forgiving debt because a country is not expected to repay, and canceling it to provide additional resources in support of the MDGs. In the latter case, the action reflects a policy choice of the international community, and responding to it by constraining new financing—which would likely be the case if loans were discontinued—would go against the proposal’s original intention. That said, in so far as debt relief creates moral hazard, there is a case for strong safeguards in future lending operations, combined with careful assessments of the appropriateness of subsequent borrowing policies.

  • A second argument is that grant resources remain scarce and should be targeted to those countries that would not benefit from debt cancellation and which thus face a relatively larger risk of debt distress subsequently.

  • Finally, debt cancellation is no guarantee that a country will avoid balance of payments problems in the future, either because of an exogenous shock or policy slippages. Thus, consistent with its mandate, the Fund would still need to be in a position to respond flexibly to the balance of payments problems faced by its low-income members, irrespective of whether the member has benefited, or is about to benefit from debt relief. It would be difficult to exclude a group of members from accessing resources under the PRGF if they meet the criteria for access. A more substantive reconsideration of the financing role of the Fund in low-income countries—including a possible move to grant financing—would need to be rooted in more fundamental considerations related to the institution’s mandate or its longer-term strategy, rather than to this debt-forgiveness initiative.

27. The G-8 proposal’s impact on the demand from HIPCs for Fund financial support would depend on their stage under the Initiative.

  • In pre-decision-point countries, the demand for PRGF resources is unlikely to be affected noticeably by the G-8 proposal until they advance further. These countries would not yet benefit from reductions in debt-service payments and their policies are generally weak, suggesting few extra resources would be available to them from the additional IDA/AfDF allocations. The Fund is not disbursing any new loans to these countries unless their policies warrant support, in which case they are building up a track record that should move them relatively quickly to the decision point. Thus, there is little rationale for any change in the Fund’s financing role vis-à-vis these countries.

  • In interim countries between decision and completion point, PRGF-supported programs will remain instrumental to reaching completion point. Thus no change in demand for PRGF resources is likely. However, deeper interim assistance, up to a full forgiveness of obligations on pre-cutoff-date debt, may be justified by a desire to limit the servicing of debt that is meant to be canceled under the G-8 proposal. Similar to the HIPC Initiative, provision of interim relief could be capped at a given percentage of total expected relief, so as to preserve incentives for the country to reach completion point. Such additional interim relief would lower countries’ balance of payment needs, thereby reducing the demand for PRGF lending.

  • Similarly, post-completion-point countries could be expected to have a reduced financing need after the proposed debt cancellation, though this may not be a lasting outcome if the additional resources are spent. Considering that low-income countries are capital constrained, it can be expected that, as more financing becomes available, current account deficits will increase accordingly—the corresponding resource transfer is indeed the rationale for providing debt relief in support of the MDGs.11 The nature and speed of the transition to a higher current account deficit is, however, extremely difficult to project.

28. A further qualification arises from the way the additional resources from IDA and the AfDF—holding the dominant share of the debt proposed to be canceled—are to be distributed across low-income countries.12 According to the G-8 proposal, the magnitude of additional resources these two institutions would provide to an individual country would be determined by existing performance-based allocation criteria.13 As a result, the financing impact of the proposal would depend primarily on performance assessments rather than eligibility for debt relief, and would need to be assessed on a case-by-case basis—as is indeed current practice.

Estimated Resource Needs of the PRGF

29. Staff estimates of expected PRGF demand have been revised to gauge the potential impact of the G-8 proposal. Recent staff estimates for PRGF demand had suggested that total resource needs for current PRGF operations could average over SDR 0.8 billion per annum during 2007-2011.14 These estimates were based on country by country projections and took into account the policies governing access to the PRGF, including tapered access norms. To account for the G-8 proposal, these estimates were modified with the following assumptions:

  • all HIPC countries not having reached the completion point move toward their completion point according to a realistic, yet ambitious, time schedule;

  • there would be no further PRGF lending to post-completion-point HIPCs.

30. The last assumption would reduce expected annual PRGF disbursements by slightly over SDR 0.2 billion during 2006-11, of which SDR 123 million per year are on account of no further lending to the 18 countries that have already reached completion point. This compares with scheduled repayments to the Fund during 2006-2011 by those 18 countries of SDR 2 billion (an annual average of SDR 340 million), which they would not have to repay under the G8 proposal. Hence, the reduction in PRGF demand even under the assumption of no further lending to post-completion-point countries would be considerably less than the debt forgiven to those countries. Overall, as a result of these assumptions, projected annual PRGF disbursements for the period 2006-11 would decline by about SDR 0.2 billion relative to the earlier projections, to about SDR 0.7 billion.

31. In addition, member-by-member balance of payments estimates were prepared to assess what would be the unconstrained demand for PRGF resources in 2006-2011 if the G-8 proposal were to be implemented. The estimates relied on country by country projections, with a baseline scenario and an alternative scenario including the impact of the G-8 debt relief proposal (see Box 3). Assuming that the Fund fills the same share of the financing gap as in 2000-2004, the demand for future PRGF resources over the period 2006-2011 is estimated at SDR 1.5 billion. Taking into account country-specific assumptions regarding the expected nature of future Fund engagement, the annual demand declines to about SDR 1 billion, with a larger portion of PRGF resources being demanded by non–HIPCs.

Demand Estimates for PRGF Resources

Member-by-member balance of payments estimates were prepared to provide a tentative quantitative assessment of future financing needs after debt relief. For all PRGF-eligible countries, gross official financing requirements over the period 2006-2011 were estimated under: (i) a baseline scenario reflecting actual and expected HIPC debt relief; and (ii) an alternative scenario including the impact of the G8 proposal.1 A combined scenario was derived using the baseline scenario for non–HIPCs and the alternative scenario for HIPCs. Main findings include:

  • The financing requirements in the baseline scenario for all countries were higher than when last estimated using WEO projections from 2004 reflecting the larger financing requirements for many oil importers on account of higher oil prices and more ambitious growth objectives. The annual impact of higher oil prices for oil importing sub-Saharan African HIPCs is estimated at SDR 1.1 billion in 2005, more than offsetting the positive impact on oil exporters.

  • The financing requirements of HIPCs after debt relief were not very different over the medium term from the baseline scenarios, given that the bulk of the G8 debt relief was assumed to be used by countries over time to make progress toward the MDGs (the use of the G-8 resources and spill-over effects on the balance of payments were estimated based on individual country circumstances). This reflects the increased demand of imports associated with the higher growth rates prompted by the increased spending. Financing needs would have certainly been lower in the alternative scenario if all G-8 debt relief had been used to fill them and not to finance faster progress toward the MDGs.

Assuming that, in the period 2006-2011, the same share of the gross official financing gap would be filled by the Fund as during 2000-2004, unconstrained annual demand for PRGF resources during 2006-2011 would be of the order of SDR 1.5 billion. This estimate is higher than earlier ones for the same period of around SDR 0.9 and an upper bound, since some element of shock financing is included in the historical data. It also shows that the current tapered access norms are rationing access and succeeding in graduating countries from reliance on PRGF financing. Refining the projections with specific country knowledge and an assessment of the expected nature of future Fund engagement—including the likelihood of PSIs or low-access PRGFs for a large number of post-completion-point countries—would lead to a reduction in the annual demand estimate to about SDR 1 billion. In addition, the implementation of the G8 proposal would alter the composition of PRGF demand, with a larger portion of PRGF resources being demanded by non–HIPCs.

1/ The impact of the G-8 debt relief proposal, as reflected in the alternative scenario, was modeled by assuming that Fund, IDA/IBRD and AfDB/AfDF debt as of end-2004 was written off as of end-2005 for countries that have already reached their completion point. For HIPC countries not having reached their completion point, a completion point date was assumed based on country specifics and the debt write off was assumed to be effected at those dates. Tentative assumptions were made regarding additional allocations from IDA and the AfDF in line with the G-8 proposal.2/ Including the protracted arrears cases but excluding Nigeria and Pakistan.

Estimated Resource Needs of the Shocks Window

32. Estimates of financing needs of low-income countries facing exogenous shocks after debt forgiveness have been prepared to provide a tentative assessment of the future demand for Fund resources for shocks-related financing. Area department staff provided country-by-country estimates of the likely demand for Fund financing to respond to exogenous shocks. Historical data on shocks and use of Fund resources guided the projected future demand for Fund financing. Shocks covered for this exercise include terms of trade shocks, natural disasters, and other external crises outside the control of a country that result in a large financing gap. Country teams assumed that a new shocks window along the lines proposed in by staff and the G-8 debt-relief proposal would be in place.

33. These estimates suggest that PRGF-eligible countries will, on average, request shocks financing 1.5 times between 2005 and 2010 and that these requests will average 27 percent of quota. This implies a financing need of roughly SDR 400 million per year.15 There is considerable uncertainty surrounding these estimates; for example, larger-than-expected demand by a few countries with large quotas would result in much higher demand.16

34. There is some uncertainty regarding how much of this total estimated demand would be associated with the proposed new shocks facility and how much with PRGF augmentations. Unlike the projections in EBS/05/89, the estimates presented here cover demand for both windows. To the extent that further debt relief implies a shift from PRGF arrangements to signaling arrangements, a larger proportion of shocks-related financing would be covered by the proposed new shocks window.

VI. Legal and Policy Issues Raised by the Proposal

35. In their initial consideration of the G-8 debt relief proposal, Directors observed that it raises important policy questions for the Fund. These questions arise with respect to uniformity of treatment, conditionality, the impact on future Fund lending, and implications for the debt sustainability framework (DSF).

E. Eligibility and Uniformity of Treatment

36. In their earlier discussion, Directors noted that “the proposal raises issues of uniformity of treatment of members required by the Articles of Agreement. It calls for the use of the Fund’s resources for full debt relief to HIPCs, leaving other members with similar incomes with debt burdens.” This section considers these issues.

Background and General Principles

37. The Special Disbursement Account (SDA) is the main financing source envisaged under the G-8 debt relief proposal.17 The SDR 2.2 billion in profits derived from the 1999-2000 off-market gold sales were placed in the SDA and have remained there since that time. The investment income accrued on these profits also resides in the SDA (some of the accrued income has also been used to fund HIPC Initiative debt relief). Similarly, the PRGF Trust Subsidy Account, which has been targeted to provide some resources for the G-8 proposal, is funded in part with SDR 400 million that was transferred from the SDA in 1994.18 Accordingly, transfers from that account, at least with respect to amounts that are attributable to SDA transfers, would continue to be subject to the requirements governing SDA assets.19

38. The authority to use SDA-related resources to finance the contemplated debt relief is set forth in Article V, Section 12(f)(ii), which provides that assets held in the SDA may be used, inter alia, “for operations and transactions that are not authorized by other provisions of this Agreement but are consistent with the purposes of the Fund. Under this subsection (f)(ii) balance of payments assistance may be made available on special terms to developing members in difficult circumstances, and for this purpose the Fund shall take into account the level of per capita income.”20

39. As a general matter, the principle of uniformity of treatment applies to all use of the Fund’s resources under the Articles of Agreement.21 Under this principle, any decision by the Fund to differentiate between members must be based on the application of criteria that are relevant to the objective of the power being exercised. For example, when exercising its authority under Article V, Section 3(a) to make its general resources available to help resolve balance of payments problems, the Fund is authorized to differentiate the terms of financing provided (e.g., the length of the repurchase period and/or the rate of charge) on the basis of the nature of the balance of payments problem being financed. This has been achieved through the establishment of special facilities, whereby the eligibility to receive preferential terms has been limited to those members that are experiencing the specific type of balance of payments problem being financed (e.g., Extended Fund Facility).

40. Article V, Section 12(f)(ii) sets forth an important–but limited–exception to this general rule. Specifically, it authorizes the Fund to use SDA resources, inter alia, to provide balance of payments assistance on special terms only to developing countries; i.e., developed countries may be excluded from eligibility under this sub-clause even if they are experiencing the type of balance of payments problem being financed. However, among developing countries, the principle of uniformity of treatment applies.22

41. Drawing on both text of Article V, Section 12(f)(ii) and its objectives, the Fund has limited developing countries’ eligibility and qualification to receive SDA resources on the basis of the application of two criteria: (i) level of per capita income and (ii) nature of balance of payments problem.23 Specifically:

  1. Per Capita Income: As noted above, Article V, Section 12(f)(ii) provides that when making available SDA resources on special terms, the Fund shall take into account the per capita income of members. Relying on this authority, the Fund has used income level as the basis for limiting eligibility for all of the concessional financing operations that have taken place pursuant to this provision. Specifically, financing under the Structural Adjustment Facility, the Enhanced Structural Adjustment Facility (the latter of which became the PRGF in 1999) and the HIPC Initiative has only been available for members whose per capita income was below specified thresholds.24 This criterion may be used as a basis for either providing debt relief to only a subset of PRGF-eligible members that fall below a specified income threshold, or gradating the amount of debt relief provided to PRGF-eligible members on the basis of income.

  2. Nature and Extent of Balance of Payments Problems: Consistent with the approach that has been followed by the Fund with respect to the use of its general resources, the Fund has also relied on the nature of balance of payments problems as a basis for further limiting members’ qualification to receive financial assistance under Article V, Section 12(f)(ii), with the assistance provided being designed to address the relevant difficulties. For example, financing under the HIPC Initiative is only available to those PRGF-eligible members whose debt levels are unsustainable as defined under the PRGF-HIPC Trust and, accordingly, debt relief under that Initiative is intended to lower these countries’ debt to sustainable levels. Moreover, at the time the HIPC Initiative was enhanced (inter alia, by redefining debt sustainability at lower levels), all members who had not qualified under the original HIPC Initiative but qualified for debt relief based on the now more generous definition became eligible for debt relief—consistent with the principle of uniformity of treatment. Notably, during the March 2005 discussion of debt relief, most Directors noted that, insofar as further debt relief is advisable, the eligibility for and extent of relief should be centered on considerations of sustainability.

42. The fact that a certain category of members may have a particular balance of payments problem that makes them eligible to receive preferential financing does not mean that all of these members will receive the same amount of financing. First, the specific problem of some members may be larger than others (e.g., under the HIPC Initiative, the extent to which members’ debt levels exceed the debt sustainability thresholds has varied), and the amount of financing that can be provided on preferential terms must be linked to the size of the special need that justifies this financing. Second, because the Fund’s financial resources are finite, limits can be imposed to address the very common situation where the needs of the eligible members–if fully financed–would outstrip the financing capacity of the Fund. In these circumstances, the amount of financing may be further constrained by access limits calculated in terms of members’ quotas (e.g., access limits under the PRGF).

Assessing the G-8 Proposal

43. The central motivation for additional debt relief should determine eligibility. There are a number of possible reasons for providing further debt relief, including improving debt sustainability, eliminating rollover pressures, buffering exogenous shocks, or increasing available resources for achieving the MDGs. While these objectives are not mutually exclusive, the central emphasis is important for determining eligibility.

  • If the primary motivation is to ensure debt sustainability, debt relief should be allocated to those members with the highest risk of debt distress.

  • If the objective is to eliminate pressures to refinance existing debt-service obligations, the level of debt service would be the determining criterion.

  • If the goal is to provide balance of payment support that frees up additional resources to achieve the MDGs, then resources for debt relief should be allocated on the basis of some mix of need and performance to ensure their most productive use.

44. While the G-8 proposal appears to be primarily motivated by a desire to generate additional financing in support of the MDGs with a focus on Africa, it is not clear how the eligibility criterion—having reached the completion point under the HIPC Initiative—relates to achievement of this objective. Some HIPCs are not poorer than some non–HIPC low-income countries (Table 7): they therefore do not have a greater need than other low-income countries for resources to meet the MDGs. Nor are they significantly better performers (e.g., as measured by the Bank’s CPIA) such as would suggest a higher likelihood of using resources in an effective manner in their efforts to meet the MDGs. The key factor that distinguishes HIPCs from other low-income countries appears to be that they at one point had an unsustainable debt burden, which was indeed the qualification criterion for relief under the HIPC Initiative.

Table 7.

Policy Performance and Per-Capita Income of Selected Groups of Countries

article image

A CPIA ratio over 3.75 corresponds to a strong performance; a rating between 3.25 and 3.75, to a median performance; and a CPIA rating below 3.25, to a poor performance.

In current U.S. dollars, 2003.

Excludes small island economies eligible for PRGF assistance even though their per capita income is above the IDA cutoff.

45. More generally, with respect to the criteria that the Fund has used in the past to limit eligibility among developing countries, the following observations may be made:

Per Capita Income

46. The G-8 proposal does not rely on per capita income as the basis for determining eligibility among developing countries. The proposal would provide full Fund and IDA debt relief to some members that belong to the higher-income segment of the universe of PRGF-eligible countries (e.g., Bolivia, Honduras), but no relief at all to members with much lower per-capita income levels (e.g., Cambodia or Tajikistan).

Nature and Extent of Balance of Payments Problem

47. As noted earlier, eligibility may be limited to a subset of low income countries if there is a reasonable economic basis for concluding that the subset in question is experiencing balance of payments difficulties not being experienced by others, with the relief in question being designed to address these difficulties. Assessing the G-8 proposal under this criterion:

  1. The G-8 proposal does not appear to limit eligibility on the basis of the debt sustainability considerations that underpin the HIPC Initiative. Under the HIPC Initiative, all members are eligible to receive an amount of debt relief that is necessary to bring their external debt burdens within the indicative thresholds established under that Initiative. This same rationale could not be used to justify additional relief under the G-8 proposal, as the proposal aims explicitly at providing relief beyond what is needed to bring debt to sustainable levels as defined by the HIPC thresholds. Of course, further debt relief could be provided on the basis of revised indicative thresholds, but debt relief would then need to be available to all members meeting the new definition. Again, this does not appear to be what is envisaged under the G-8 proposal.

  2. The question therefore arises as to whether eligibility could be limited to the HIPCs on the basis of an assessment that these countries, even after receiving debt relief under the HIPC Initiative, continue to experience balance of payments difficulties arising from their past indebtedness that are not being experienced by other low-income countries and that the G-8 proposal is designed to address. The argument would be that, by virtue of their history of over indebtedness, these countries remain particularly vulnerable going forward and, for this reason, require additional debt relief. The difficulty with this rationale, however, is that as the HIPC Initiative has already sharply reduced previous cross-country differences in debt indicators, post-completion-point HIPCs are not necessarily more vulnerable now to shocks or debt-servicing difficulties than other low-income members. Indeed, further debt relief would leave them with significantly lower debt burdens than other low-income countries (Figure 1).25

  3. Could it be argued that that uniformity of treatment would be satisfied if the financing needs of the non–HIPCs were being fully met from other sources? Under this approach, the balance of payments difficulties being financed would be defined in general terms as the need for grant financing, including financing for debt relief. If it could be demonstrated that non–HIPCs are not experiencing any of this need, as defined, because they receive financing from other sources, it could be argued that they should not be eligible to receive Fund financing for debt relief; i.e., the financing could be targeted exclusively to the HIPCs. There are three difficulties with this approach, however. First, it would be difficult to conclude that the needs of the non–HIPCs, as broadly defined above, can be fully met by IDA and AfDF. As long as non–HIPCs were to continue to have some residual need for this financing, they should be eligible to receive such assistance from the Fund, albeit in an amount that does not exceed this residual need.26 Second, even if the needs of non–HIPCs could be satisfied from these other sources, there is no certainty that these needs will in fact be satisfied given, inter alia, financing considerations and performance-related conditions. Third, even if it could be concluded with certainty that the full needs of non–HIPCs would in the future be met with certainty from other sources, this should not necessarily preclude them from choosing to seek relief from the Fund rather than new financing from these other sources, if they deem this more beneficial.

Figure 1.
Figure 1.

Impact of G-8 Proposal on HIPCs, 2004 1/

(in percent)

Citation: Policy Papers 2005, 038; 10.5089/9781498331364.007.A001

Sources: World Bank, Global Development Finance, HIPC Initiative documents; IMF country reports; and World Economic Outlook.1/ After full HIPC assistance.

48. In sum, the key issue that arises when assessing the consistency of the G-8 proposal with the requirement of uniformity of treatment under the Articles is whether there is a reasonable economic basis to conclude that the proposal addresses balance of payments difficulties being experienced by the HIPCs that are not being experienced by other low-income members.

49. Finally, it should be noted that, even if non–HIPCs were made eligible for further debt relief, they could decide voluntarily to refrain from requesting such relief from the Fund. Under this approach, however, there would be no legal exclusion from eligibility. Accordingly, if a non–HIPC were to reconsider and request debt relief in the future, the Fund would be obligated to provide such relief (i) to the extent that there were resources available for this purpose, and (ii) assuming the other requirements established for the relief were met, including observance of the time frame within which relief may be provided. The fact that members are eligible for assistance under an initiative does not necessarily mean that financing sufficient to provide assistance to all eligible members must be secured in order for the initiative to become effective.27

F. Conditionality

50. The G-8 proposal specifies certain conditions for receiving debt forgiveness. The proposal mentions “100 percent IDA, AfDF and IMF debt stock relief for Completion Point HIPCs,” suggesting that the conditions for additional relief under the proposal are the same as those for reaching completion point under the HIPC Initiative. The proposal also mentions that countries must be “current on their repayment obligations to the IFIs” and “must not have experienced serious lapses, including in governance, such that their IMF programs would be at risk.” Finally, the proposal makes several references to transparency and country performance as criteria for the allocation of development assistance—though not for debt relief itself—and asks the Bank and Fund to report on “improvements on transparency on all sides and on the drive against corruption so as to ensure that all resources are used for poverty reduction.”

51. The implementation of the G-8 proposal would require that the conditions to receive debt relief be clarified and defined in more objective terms. While countries that have not yet reached their completion point could qualify to receive full debt relief when they reach it (i.e, without having to take additional measures prior to or at that time), conditions in countries that have already received full HIPC relief are quite varied (Table 8). Should post-completion-point countries with programs that are currently off track, or that do not have any Fund-supported program at all, qualify immediately for additional relief? For such countries, debt relief could be conditioned to certain indicators28 or on a minimum track record of performance prior to disbursement of relief (analogous to what was done with “grandfathered” cases at the time the HIPC Initiative was enhanced).

Table 8.

Current Program Status of Post-Completion-Point HIPCs

article image

Stand-By Arrangement; no PRSP in place.

Discussions toward new arrangement underway

Program cancelled in November 2004 following large extrabudgetary expenditures leading to a non-complying initial disbursement. Further engagement subject to resolution of pending data verification issues, in particular central bank audits.

52. Some Directors have suggested the use of ongoing conditionality for both pre-and post-completion-point countries to help ensure that the funds freed by reduced debt-service payments are spent in support of efforts to reach the MDGs. For example, the provision of relief could be phased in time and linked to the successful implementation of a Fund-supported program, or to positive recommendations in the context of annual Article IV consultations. However, given the amortization period for PRGF resources, this might require Fund monitoring for up to ten years.29 Such monitoring, if envisaged, might be more practically implemented over a shorter period (i.e., one to three years).

53. Another issue is whether it is feasible to monitor the use of the resources freed by debt cancellation given the fungibility of funds. Experience with assessments of public expenditure management systems shows the difficulty of establishing a one-to-one link between debt relief and specific government spending decisions, as well as judging how much the spending meaningfully contributes to poverty reduction.

54. Additional conditionality on governance or transparency would raise additional issues. For example, it would be difficult to apply to interim HIPCs: the criteria for completion point have already been set for those countries. It may also raise issues of consistency across HIPCs, unless broadly similar conditions were required of post-completion-point countries.

55. The choice between different criteria or conditions may entail a trade off between speed and certainty on the one hand, and the appropriate use of resources on the other. Strict conditions over an extended period would minimize the risk of inappropriate use of the resources freed by debt relief but would slow delivery. Political pressures may build up over time for relaxing conditionality and accelerating the path to completion point. Alternatively, if conditionality is focused on the provision of new assistance and policy support, rather than on the delivery of the relief itself, it would not have these disadvantages but could increase the risks of misuse.

G. Potential Implications of the Proposal for Fund Instruments

56. Staff does not expect the G-8 proposal to alter in a fundamental manner the Fund’s role in low-income countries. The proposed debt cancellation will certainly have an economic impact on the countries benefiting from it, but it will not transform them in a way that would affect the basic nature of their engagement with the Fund. In particular, these countries will still rely on the Fund to support them in various areas, including:

  • policy advice and assistance in the design of macroeconomic stabilization and structural reforms and capacity building in the Fund’s areas of expertise;

  • assessments of macroeconomic policies and expenditure frameworks including to multilateral and bilateral donors in the context of their stepped-up efforts to support the MDGs; and

  • financing, either in response to exogenous shocks, or to address remaining or emergent protracted balance of payments problems.

57. Debt cancellation may affect the relative demand for each of these services, but the overall impact is hard to predict, as it will vary across countries. Thus, the nature of the Fund’s engagement with each member, including post-completion-point HIPCs, will continue to be decided on a case-by-case basis. Members will continue to enter into a range of different forms of engagement with the Fund, which may include an economic program supported by a PRGF arrangement; assistance under the PRGF shock window; a program backed by the proposed Policy Support Instrument (PSI); and policy dialogue and monitoring in a surveillance context. The choice between these different forms of engagement will continue to depend upon the particular need and policies of the member at a given time, rather than whether the member has been or has not been a beneficiary of the debt relief initiative.

The Self-Sustained PRGF

58. The self-sustained PRGF is intended, over the medium-to long-term, to provide the main source and mechanism for financing balance of payments problems of low-income members. As mentioned in Section III, debt cancellation notwithstanding, it may be appropriate for the Fund to continue to provide financial support to its low-income members, either in response to exogenous shocks or protracted balance of payments needs. The Fund will also continue to provide policy advice and assistance in the design of macroeconomic policy framework and structural reform programs, and support for capacity building. Moreover, multilateral and bilateral donors are expected to continue to ask for assessments of macroeconomic policies and budgetary frameworks in the context of their stepped-up assistance efforts in support of the MDGs.

59. A PRGF arrangement would remain the appropriate form of Fund engagement with many HIPCs after the completion point, and one that most post-completion-point HIPCs have availed themselves of. Indeed, many beneficiaries of the additional debt relief may still not have reached sustained macroeconomic stability and have an important unfinished agenda of structural reforms: the most appropriate form of support for their reform program would be a PRGF arrangement. Depending on the financing constraint on the PRGF, and the extent to which the proposed debt cancellation reduces a country’s balance of payments problems, support could be in the form of a low access PRGF. However, if a protracted balance of payments problem remains after debt relief, the possibility of a normal access PRGF arrangement need not be excluded. In case of an external shock, access under an arrangement could be augmented in line with existing policies.

The Policy Support Instrument (PSI)

60. The proposed Policy Support Instrument (PSI) would complement the self-sustained PRGF; together they would provide appropriate modalities by which to support the varied needs of low-income members. The proposal for a PSI has been developed with a view to meeting the needs of “mature stabilizers,” namely low-income countries that have reached macroeconomic stability, are fairly advanced in their structural reform program and either do not need or want Fund financial assistance. In terms of policy content, programs supported by a PRGF arrangement and a PSI would share similar characteristics, notably in that they would meet the standard of upper credit tranche conditionality. While a PSI-supported program would not provide automatic access to financing in the face of an external shock, it could facilitate access to the PRGF window for shocks, as proposed by staff, or a separate concessional shocks facility under a new trust, as proposed by the G-8.

61. The PSI should not be considered as the only, or preferred, instrument for countries that have benefited from debt cancellation. Some of them may still be facing a large reform agenda and pending challenges in the area of macroeconomic stabilization and management. They may prefer the familiar framework of a PRGF and the ready access to Fund resources in case of need, and a low access PRGF would be most appropriate for them. The PSI in turn would be better at meeting the needs of “mature stabilizers” or at signaling “graduation” from the PRGF.

62. The number of potential users of the PSI may increase, nevertheless, after the proposed debt cancellation. Some low-income members which have achieved macroeconomic stability will have their balance of payments financing needs significantly reduced by additional debt relief and be ready to graduate from the PRGF earlier than would have been the case otherwise. These countries are likely to continue to need policy support and endorsement from the Fund, not least in view of their ongoing external development assistance needs. For some such members, the PSI would be the instrument of choice. Thus, the group of countries potentially interested in the PSI would likely be larger than previously expected.

63. Staff do not see a need to modify the envisaged structure of the PSI in the face of this potential higher demand. Its design is sufficiently flexible to meet the needs of an expanded group of countries. PSI program content would vary depending on members’ circumstances, while continuing to meet the upper credit tranche conditionality standard. Similarly, it seems appropriate to continue to require that a poverty reduction strategy be in place to qualify for a PSI, as well as to see the PSI as facilitating access to a new shocks window.

64. Either a PRGF or a PSI could be used to provide a track record of policy performance if this is considered a condition for debt cancellation. Similarly, either could provide a framework for possible monitoring of a country’s use of resources freed up by debt cancellation if such monitoring is envisaged.

The PRGF Shocks Window

65. Staff do not expect the G-8 debt relief proposal to have major operational impact on the proposed shocks facility. Operationally, lending for shocks is mostly for liquidity purposes, rather than to address protracted balance of payments problems. Since debt forgiveness under the G-8 proposal aims at helping beneficiaries meet the MDGs, the shocks window would still play a role in assisting these countries overcoming adverse exogenous shocks—even after full implementation of the debt relief proposal—because the provision of debt relief would not have altered the magnitude or frequency of shocks. The proposed facility is also more concessional than current alternatives (mainly resources in the credit tranches and under emergency natural disaster assistance) and likely more quickly disbursing than available grant mechanisms. More broadly, the link between financing and policy support remains a useful signal for donors that post-shock cases have adopted an adequate mix of adjustment and financing.

66. Even if it were to draw from complementary sources of financing, maintaining the shock window within the PRGF Trust would be recommended. Staff welcomes the G-8 intent to mobilize financing for the facility that would help meet an important identified need. The G-8 proposal suggests establishing a new trust fund for this purpose, to be filled through voluntary contributions. However, it might be more appropriate to make this financing available for the shock window within the PRGF Trust, rather than for a separate, additional facility. Having two separate facilities for the same purpose would be administratively costly and potentially confusing for recipients, with no clear benefits.

H. Implications for the Low-Income Country Debt Sustainability Framework

67. The G-8 proposal states that following debt cancellation, eligible countries should be “eased into new borrowing.” However, to the extent that the operation would significantly lower debt ratios of eligible countries, the recently endorsed low-income country debt sustainability framework (DSF) would signal a low risk of debt distress.30, 31 Thus, according to the DSF, countries would have considerable space for new borrowing, which may be at odds with the G8 proposal. While a number of options exist to adjust the DSF in a way that imposes tighter constraints on new borrowing after debt relief, any modifications would seem premature to consider before the implementation of the framework has seriously begun. An opportunity to modify the DSF, if needed in light of decisions on debt relief, will arise in the context of its review, planned for Spring 2006.

68. Indeed, a number of considerations suggest that the DSF already contains safeguards that would limit the risk of excessive borrowing after debt cancellation:

  • First, the G-8 proposal itself already provides eligible HIPCs with a higher share of grants. Since IDA and AfDF gross allocations to these countries would be reduced by an amount equal to the reduction in debt service, the overall financing package to these countries would consist of fewer loan and more grant resources (in the form of debt relief). This alone would limit new debt accumulation, relative to other low-income countries not benefiting from the write-off.

  • Second, the forward-looking framework is designed to capture any potential rapid accumulation of debt. DSAs will be conducted on a regular basis to monitor new debt creation and provide early warning signals if debt ratios are taking a worrisome trend.

  • Some post-completion-point HIPCs have relatively strong policies and institutions, and if their debt ratios decline significantly following debt relief, they would have a particularly low risk of debt distress in comparison with other low-income countries. Higher risks of debt distress will thus rest with those countries that will not benefit from debt cancellation, making them, from a debt sustainability perspective, the more appropriate candidates for grants.

  • Finally, the main risk for a rapid re-accumulation of debt is likely to stem from increased access to financing on non-concessional terms from sources other than multilateral institutions. Under current policies, countries will still have incentives to pursue sound borrowing policies to remain eligible for ongoing multilateral support. But tighter constraints on the availability of the funds would tend to increase the risk that countries will meet their financing needs through non-concessional borrowing from other sources.

69. Given these considerations, an adjustment to the DSF at this stage would appear premature. Official financing by multilateral creditors and grants provided by donors will likely remain the most attractive source of funds for low-income countries even after debt cancellation. Thus, the Fund will continue to play an important role in the design of appropriate borrowing policies in the context of both surveillance and program support. The pace of new debt accumulation should continue to be governed by assessments of countries’ capacity to use aid effectively, while maintaining macroeconomic stability and limiting potentially adverse effects on other macroeconomic variables, such as exchange rates and interest rates. Apart from maintaining existing safeguards available to limit nonconcessional debt creation in Fund-supported programs, the prevailing case-by-case approach to new borrowing, embedded in a comprehensive macroeconomic framework, is likely to offer the best approach to prevent excessive debt accumulation after debt relief.

V. Approvals and Other Actions Needed to Effect the Proposal

70. The full scope of the approvals, consents and amendments (including to the PRGF and PRGF-HIPC Trust Instruments) required to effect a debt relief plan will vary, depending on the precise form and modalities utilized for the debt relief. Accordingly, no definitive assessment of these issues can be made until there is greater clarity as to the debt relief plan that will ultimately be proposed for adoption by the Executive Board.

71. Despite the foregoing, the following minimum actions are likely to be required in connection with any debt relief scheme that has financing components similar to those envisaged under the G-8 proposal:

  • Executive Board approval by an 85 percent majority of the total voting power would be required for any use of SDA resources. This would include (i) the SDR 2.2 billion in profits from the 1999/2000 off-market gold sales, (ii) income that has accrued on such profits and not been used to date, and (iii) the portion of the PRGF Trust Subsidy Account that is attributable to the SDA.32

  • The proposed use of PRGF Trust Subsidy Account resources would additionally require amendment of the provisions of the PRGF Trust Instrument that governs both the use of resources in the Subsidy Account (Section IV, paragraph 4). As this is a protected provision under Section IX of the PRGF Trust Instrument, this would require the consent of all 43 contributors to the Subsidy Account. In addition, to the extent that this amendment is understood as expanding the purposes of the overall PRGF Trust Instrument, the consent of all contributors and lenders to the Loan Account of the PRGF Trust would also be required. This latter requirement does not impose a considerable additional hurdle, however, as there is only one contributor or lender to the Loan Account that is not a contributor to the Subsidy Account.33

  • Concerning the third source of financing envisaged by the G-8 proposal, no new action would be needed with respect to funds in the PRGF-HIPC Trust Umbrella Account’s subaccounts so long as the amount in each subaccount (i) is used solely in respect of the member for which the subaccount was established, and (ii) is used solely to offset that member’s obligations falling due to the Fund in accordance with the schedule established at the time of the decision point or completion point (the former in the case of interim assistance). Any changes to the subaccounts that adversely affect the members for whose benefit the accounts have been established would require not only a decision by the Executive Board (by a majority of the votes cast) but also the consent of the members concerned.34

72. General decisions (or amendments of the PRGF or PRGF-HIPC Trust Instruments) to effect non-financing elements of the G-8 proposal would normally require only a majority of the votes cast by the Executive Board. For example, a decision amending the PRGF-HIPC Trust to include Policy Support Instruments among those that count towards the completion point track record could be adopted by the Fund as Trustee by a majority of the votes cast.

73. Finally, it should be noted that additional consents (e.g., of all 93 PRGF-HIPC Trust contributors and lenders) could be needed if the new debt relief scheme were to be implemented by means of certain kinds of amendment of the PRGF-HIPC Trust Instrument.

VI. Next Steps and Issues for Discussion

74. Following a request from the Executive Board, this paper has presented a preliminary assessment of the financial, legal and policy implications of the G-8 proposal for debt cancellation. Because the Board’s consideration of the proposal is still at an early stage, staff has not attempted to estimate the implications it may have in terms of the use of staff resources. Informed by Directors’ discussion of this paper, staff intends to return to the Board before the 2005 Annual Meetings with a follow up paper. This paper would attempt to incorporate Directors’ views on further debt relief, with a view to a possible implementation of a debt relief mechanism by year’s end, contingent upon the mobilization of the required financial resources for the Fund. The preparation of this paper will be coordinated as needed with the World Bank and African Development Bank.

75. In light of the proposal for debt cancellation put forward by the G-8, Directors may wish to consider the following issues:

  • Do Directors agree with the proposal to grant HIPCs full relief on the claims that the Fund has on them?

  • What are Directors’ views on the criteria put forward by the G-8 for eligibility for such debt relief and how do they view the requirement of uniformity of treatment in that context?

  • What are Directors’ views on possible modalities of delivery of this relief? Should it be delivered upfront or in a phased manner? Should additional relief be conditioned?

  • What are Director’s views on the Fund’s future role in countries that would have benefited from the debt relief proposed by the G-8?

  • What do Directors consider the additional financing resources required to maintain the Fund's capacity to continue lending under the interim and the self-sustained PRGF as well as provide debt relief to the protracted arrears cases and any new HIPC, should the G8 debt proposal be implemented?

  • Do Directors wish the Fund staff to follow up on specific implementation modalities of the G-8 proposal before the Annual meetings?

Appendix. Summary of Bilateral Contributions to the PRGF and PRGF-HIPC Trusts

(In millions of SDRs as of end-January 2005)

article image
article image
article image

Less than SDR 5,000.

The term “as needed” refers to the nominal undiscounted sum of the projected delivery of HIPC assistance taking into account the profile of projected subsidy needs associated with PRGF lending during 2002–05.


For example, the 35 HIPCs are expected to make repurchases or repayments of almost SDR 650 million in 2005, of which less than half would be covered by existing HIPC assistance (Table 2).


This assumption is made, in part, owing to the uncertainties as to when these countries would reach their completion points and how interim HIPC assistance would be delivered. To the extent that these countries reach their completion points beyond end-2005, the cost to the Fund would be lower than estimated here.


Any expansion of the eligibility list to include non–HIPCs could also substantially increase the cost of the proposal, unless the extent of debt cancellation was reduced. For example, at end-2004, debt to the Fund of non–HIPC PRGF-eligible countries (including four members initially thought to be HIPCs but whose debt is below the HIPC threshold) amounted to SDR 3.3 billion, of which SDR 0.6 billion to the GRA and SDR 2.6 billion to the PRGF. Preliminary cost estimates for the potential new HIPCs are expected to be provided in the context of the annual HIPC update, currently scheduled for Board discussion in early September.


In NPV terms at end-April 2005. Table 6 shows account balances in cash terms as of April 30, 2005, the end of the fund’s financial year.


Based on an end-April 2005 balance of SDR 0.5 billion, with adjustments for Board-approved drawdowns from the Umbrella Account, new inflows, and anticipated disbursements for the remainder of 2005.


This includes a small balance (SDR 68 million) in the PRGF-HIPC Trust that has not been earmarked specifically for HIPC operations (see Table 6).


This estimate is sensitive to the assumptions made about future interest rates.


If the start-up date for the self-sustained PRGF is advanced, this could also marginally lower its commitment capacity.


In the event that subsidy resources are insufficient, it would no longer be possible to fully subsidize the PRGF lending rate to 0.5 percent.


See The Fund’s Support of Low Income Member Countries: Considerations on Instruments and Financing (2/24/04), especially pp. 52-54.


As will be discussed in “The Macroeconomics of Managing Increased Aid Inflows: Experience of Low-Income Countries and Policy Implications” (August 8, 2005), the nature and speed of transition to a higher current account deficit depends on many factors, including the macroeconomic policy response to higher external inflows.


As of 2003, the Fund accounted for only about 10 percent of outstanding multilateral debt of HIPCs. In terms of debt service, the Fund’s share was larger (33 percent) because of the shorter maturities of Fund claims.


The proposal for IDA and the AfDF can be interpreted as a two-step process. In a first step, the new allocations to countries receiving debt relief would be reduced one-for-one by the amount of the debt-service reduction. This can be interpreted as a shift from direct project financing to budget support, with unchanged net flows, because the debt-service savings create additional budgetary resources but at the cost of lower gross financing. In a second step, the additional resources arising from the reimbursement by donors will be distributed to all IDA-eligible countries on the basis of existing performance-based allocation mechanisms.


These estimates were on expected annual commitments. The corresponding average annual disbursements for the period 2006-2011 were SDR 0.9 billion, including for dealing with the protracted arrears cases. Two PRGF-eligible countries with large quotas, Nigeria and Pakistan, were assumed not to request PRGF arrangements during the projecting period. If they were included, the staff’s projection of potential demand for PRGF resources would increase significantly.


These estimates assume a limit on cumulative shocks financing of 50 percent of quota. They also assume that the new guidelines on blending use of GRA and PRGF resources would also apply to the shocks financing, and in particular that the norm for annual access to the shocks financing for such counties would be 12½ percent of quota.


These estimates include likely demand by Nigeria and Pakistan (but exclude India); however staff expects that neither of these countries will request shocks financing at all over the 2005-2015 period.


The other source of financing for the G-8 proposal (i.e., the PRGF-HIPC Trust Umbrella Account’s subaccounts) is also derived in part from the SDA, through the PRGF-HIPC Trust. However, most of these resources (i.e., other than those amounts disbursed as interim assistance) were irrevocably transferred for the benefit of members at the time they reached the completion point. The funds in each subaccount are held on behalf of the relevant member to meet its obligations to the Fund as they fall due in accordance with the schedule decided at the completion point. The governing Board decision makes clear that any resources remaining in a sub account after the discharge of all obligations of that subaccount must be transferred promptly to the member on whose behalf the subaccount was established. (Decision No. 11698-(98/38) ESAF, April 1, 1998, as amended.)


Decision No. 10531-(93/170), December 15, 1993. It should be noted that the use of resources from the Reserve Account of the PRGF Trust would, similarly, be subject to SDA requirements, as SDA transfers make up the bulk of resources in the Reserve Account.


Attribution of Subsidy Account transfers among, on the one hand, SDA resources and, on the other hand, amounts provided by contributors and lenders would be calculated on a pro rata basis. By way of analogy, the provisions governing termination of the Subsidy Account specify that, upon the completion of subsidy operations, any resources remaining in the Subsidy Account are to be distributed to the SDA and to donors and lenders that have contributed to the subsidy operation in proportion to their contributions (PRGF Trust Instrument, Section IV, paragraph 6).


Article V, Section 12(f)(ii) provides a basis for the use of both resources that are currently in the SDA and unused resources that have been transferred from the SDA but are still subject to its requirements.


The principle of uniformity of treatment does not apply in circumstances where the Fund, pursuant to Article V, Section 2(b), administers resources provided by others.


The explanation of Article V, Section 12(f)(ii) in the Commentary on the Second Amendment of the Fund’s Articles notes that SDA assets resulting from gold sales can be used “...(ii) for the benefit of members in need, on a uniform basis, by using the assets in the Special Disbursement Account in operations and transactions other than those authorized by other provisions, provided that the operations and transactions are consistent with the Fund’s purposes and are for balance of payments assistance. The Fund may make assistance available, however, on special terms to developing members in difficult circumstances. When making this latter assistance available, the Fund will have to take into account the level of the per capita income of developing members, and may take into account other appropriate criteria as well (Section 12(f)(ii)). Balance of payments assistance on concessionary terms can include, for example, subsidies for the payment of charges levied in connection with the use of the resources in the General Resources Account.” Proposed Second Amendment to the Articles of Agreement, A Report by the Executive Directors to the Board of Governors (1976), page 47.


In addition, and reflecting the finite nature of SDA resources, such assistance has also generally been limited to those countries that satisfy these criteria within a specific time period.


Decision No. 8240-(86/56) SAF (March 26, 1986), as amended.


The proposal also gives rise to potential unfortunate incentive issues (countries having pursued more prudent debt management policies would not receive debt relief) as well as consistency of treatment issues on the creditor side (it does not consider debt cancellation across all multilateral creditors to HIPCs, excluding for instance credits from the Inter-American Development Bank and the Asian Development Bank, among others). It also leaves countries with varying amounts of bilateral (non–Paris Club) and commercial debt.


The additional factors that could be taken into account when allocating finite resources in this context could include quotas, per capita income and policy performance.


Consistent with past practice, the initiative could become effective without full financing (e.g., on the basis of an expression of intent by some eligible members not to draw, as was done by some members (China and India) in the context of the SAF, ESAF and PRGF). A subsequent change in this intention and concomitant draw down of resources by such members would reduce the amount of assistance that could be provided to other members under the initiative, but it would not necessitate additional financing from the Fund (since, ultimately, access to assistance under such initiatives is approved subject to the availability of resources).


Possible indicators include minimum levels of transparency or the existence of strong public expenditure management systems.


As mentioned above, the amortization period for IDA and AfDF credits is even longer.


This outcome assumes that debt relief will be provided irrevocably, i.e., not as debt-service relief subject to ongoing conditionality, and may not hold for non-African HIPCs with large exposure to non-participating creditors.


The framework, which was approved by the Fund and Bank Boards in April 2005, consists of three pillars: (i) a forward-looking analysis of debt and debt-service dynamics; (ii) an assessment of debt sustainability guided by indicative policy-dependent debt-burden thresholds; and (iii) an appropriate borrowing (and lending) strategy that contains the risk of debt distress. See Operational Framework for Debt Sustainability Assessments in Low-Income Countries—Further Considerations (3/28/05).


Parliamentary action in some member countries may also be required before some Executive Directors could vote in favor of a decision involving use of SDA resources. This is a matter that can be authoritatively decided only by members, however, and not by the staff.


The appendix contains a list of the contributors and lenders to the PRGF Trust Subsidy and Loan Accounts, as well as to the PRGF-HIPC Trust. Separately, while the requirement for consent implies that each contributor or lender has a veto over the relevant amendment, this veto scenario could in principle be averted by returning in full the resources provided by a dissenting contributor or lender (i.e., the veto power would terminate once the particular party ceased to be a contributor or lender).


As discussed in footnote 26, most of the amounts represent resources that the Fund irrevocably disbursed to members at their completion points.