Article

IMF Policy Paper: 2018–19 Review of Facilities for Low-Income Countries—Reform Proposals; Review of the Financing of the Fund’s Concessional Assistance and Debt Relief to Low-Income Member Countries

Author(s):
International Monetary Fund. Strategy, Policy, & Review Department;International Monetary Fund. Finance Dept.;International Monetary Fund. Legal Dept.
Published Date:
June 2019
Share
  • ShareShare
Show Summary Details

Introduction

1. The 2018–19 Review of Facilities for Low-Income Countries (the “LIC Facilities Review”) is the second review of the Fund’s concessional facilities available to LICs since the current framework was established in 2009.1, 2 The objective of this review is to take stock of experience and assess whether modifications to these facilities for PRGT-eligible countries are warranted to better serve members’ needs in an evolving global landscape.

2. The Fund has two types of lending facilities: (i) facilities financed through the General Resources Account (GRA), available to all member countries and providing loans on non-concessional terms; and (ii) facilities financed through the Poverty Reduction and Growth Trust (PRGT), available only to PRGT-eligible member countries and providing loans on a concessional basis. PRGT-eligible countries are countries with relatively low per-capita income levels and limited access to international financial markets.3 In this paper, the terms “Low-Income Countries” and “PRGT-eligible countries” are used interchangeably.

3. Demand for resources from the PRGT has receded since the global financial crisis but remains substantial, with over half of PRGT-eligible countries having had at least one Fund-supported program during 2010–18. Nearly three-quarters of these countries had repeat Fund program engagement over the period and over half had at least two disbursing arrangements, with engagement in Fund-supported programs for this sub-group averaging close to five years.4

4. This review takes place against the backdrop of evolving conditions in LICs:

  • Continued growth in output, trade, and the external financing needs of LICs has resulted in the erosion of access levels relative to these metrics of potential demand for use of Fund resources.
  • An increasing number of LICs have tapped international capital markets, leaving them exposed to repricing and rollover risks should global financial conditions tighten.
  • There has been a marked increase in public debt ratios in most LICs, with more than two-fifths of LICs now assessed to be at high risk of, or already in, debt distress.
  • One-half of LICs are in fragile and conflict-affected situations (FCS), where civil conflict and/or weak institutions create special challenges for implementing economic stabilization and reform programs.
  • Climate-related natural disasters are increasing in both intensity and frequency, creating enhanced demand for emergency financial support, particularly in small states.
  • Global interest rates remain low, reducing the concessional element of PRGT loans.

5. This paper proposes a package of reforms to the IMF’s concessional facilities. The first stage of the 2018–19 LIC Facilities Review in July 2018 sought Board guidance on a variety of options.5 Directors agreed that the basic architecture of facilities for LICs remains broadly appropriate. They saw merit in reassessing selected features of the existing facilities, although views on some of the specific reform options varied. Staff sought further inputs from Executive Directors in an informal session in March 2019, which discussed a possible reform package and its implications for PRGT finances.

6. The LIC Facilities Review is part of a wider review of policies relating to Fund support for LICs. A companion paper on the Review of the Financing of the Fund’s Concessional Assistance and Debt Relief to LICs analyzes the impact of the proposed reforms on the PRGT financing framework. The parallel Review of the PRGT Interest Rate Structure proposes the alignment of the SCF interest rate with the ECF interest rate. The upcoming Review of Eligibility to Use the Fund’s Facilities for Concessional Financing examines countries’ readiness for graduation from eligibility for concessional financing. The LIC Facilities Review also follows the May 2019 Board paper on Building Resilience in Developing Countries Vulnerable to Large Natural Disasters (IMF, 2019a).

7. This paper is informed by the staff findings from the 2018 Review of Program Design and Conditionality (ROC). During Board discussion of the LIC Facilities Review in July 2018, many Directors called for more information on the effectiveness of PRGT-supported programs in meeting their objectives, the experience with repeated use of PRGT facilities, and the evolution of debt vulnerabilities under PRGT-supported programs. They underscored that assessments and proposals from the LIC Facilities Review should be informed by the analysis in the ROC. Executive Directors were briefed on selected issues relating to the ROC in February 2019; the ROC paper will be discussed by the Board in early May 2019.6

8. The reforms to the LIC facilities proposed below seek to enhance support for LICs within the self-sustained PRGT financing framework. The central objectives are to:

  • Provide LICs with higher levels of access to concessional financing, favoring the poorest and most vulnerable PRGT-eligible members;
  • Respond to the specific challenges faced by FCS and by countries vulnerable to natural disasters, including through higher access limits under the RCF; and
  • Improve the flexibility of PRGT facilities through reforms to the SCF and the ECF to allow better tailoring of program design to countries’ diverse circumstances.

9. The reforms proposed in the report take account of views expressed by Executive Directors in successive rounds of consultations and constitute an integrated package. The reforms do not include options presented in the first-stage paper that received little support, including: maintaining the normal cumulative access limit unchanged while making the exceptional access criteria more flexible; modifications to the PSI; introducing a short-term ECF; and widening access to the Catastrophe Containment and Relief Trust (CCRT) to provide relief on debt service to the Fund for members hit by large natural disasters.

10. The paper also proposes changes in access limits under the RFI, an emergency instrument available to all Fund members. This would expand the scope for providing emergency financial support to those FCS and countries hit by natural disasters that are not eligible for concessional financing, while preserving broad harmonization of access limits across the RFI and the RCF.

11. The remainder of this report is organized as follows. Section II discusses findings of the ROC that are relevant for the LIC Facilities Review and examines the catalytic role of Fund-supported programs in LICs. Section III presents the proposed reforms to the LIC facilities and the RFI. Section IV summarizes the implications of the proposed reforms on the PRGT financing framework (analyzed in the companion Board paper), Section V discusses the impact of the proposed reforms on the Fund’s risk profile, and Section VI identifies implementation issues. The final section suggests issues for discussion.

Stocktaking of Experience—Further Analysis

Further to the stocktaking of experience with the LIC Facilities presented in the July 2018 Board paper, this section provides additional analysis of program design issues, drawing on the 2018 Review of Program Design and Conditionality, and of the catalytic role of the Fund.

A. 2018 Review of Program Design and Conditionality7

12. Staff's analytical findings from the ROC informed the proposed reforms to the LIC facilities. Its findings on effectiveness of programs, repeat use of PRGT resources, and experience with debt vulnerabilities are summarized below.

Program Performance

13. Program engagement in many LICs takes place in an inherently challenging context. Half of PRGT-eligible countries are FCS; a large number of LICs have concentrated export structures, with high reliance on commodities whose prices are volatile; and weak institutional capacity often compounds development challenges.

14. The performance record of PRGT-supported programs during 2011–17 shows that three-quarters of these were at least partially successful.8 About 25 percent of PRGT-supported programs are assessed to have been successful, 50 percent partially successful, and 25 percent unsuccessful—a pattern broadly similar to the outcomes for GRA-supported programs.9 Unsuccessful programs typically involved either (i) weak program implementation, with programs going off-track (text chart); 10 or (ii) adverse shocks (including civil conflict, terms of trade, natural disasters, and epidemics).11 In contrast, the preponderance of programs that remained on-track were successful (38 percent) or partially successful (47 percent). Analysis showed that program completion raised the probability of success by 40 percentage points, while negative commodity price shocks increased the likelihood of an unsuccessful program by 7 percentage points.

PRGT Program Success and Completion Rates

(Percent) 1/

Sources: MONA and IMF staff calculations.

1/ MONA and IMF staff calculation.

Successor Programs

15. Successor arrangements in LICs were common, consistent with the protracted nature of countries’ balance of payments (BoP) problems. Over half of PRGT-eligible members had at least one Fund-supported program during 2010– 18, of which three-quarters had repeat programs, and over half had repeat disbursing programs. Repeat use of the ECF is expected, as the purpose of the facility is to enable members with a protracted balance of payment problem to make significant progress towards stable and sustainable macroeconomic positions consistent with strong and durable poverty reduction and growth.12 There were 13 cases of Fund-supported programs followed by new programs within two years during 2011–17. Nine of these successor programs were at least partially successful. Of the remaining four, three were unsuccessful due to high debt vulnerabilities, including both of the back-to-back unsuccessful cases—highlighting the need to strengthen focus on debt vulnerabilities in program design. Program completion was associated with greater program success: of the 10 successor programs that concluded all or most (all but one or two) reviews, four were successful and another four partially successful.

PRGT: Successor Arrangements

(Number of arrangements)

Sources: MONA and IMF staff calculations.

Debt Vulnerabilities

16. Programs helped contain or reduce debt vulnerabilities.

  • Across all PRGT-supported programs, debt sustainability risks remained well-contained at low or moderate risk levels or improved in three-quarters of cases.13 This number rises to 91 percent for the sample of countries that completed most or all reviews (Figure 1).14
  • Over the program period, there was a net reduction of 8 percentage points in the number of countries at high risk of debt distress or in debt distress (this number rises to 17 percentage points for countries with competed or largely completed programs).15 This compares favorably to the evolution of debt vulnerabilities in the broader group of PRGT-eligible countries, where between 2013 and 2018, the share of countries at high risk of debt distress or in debt distress increased by 18 percentage points (text chart).16
  • In most cases where program countries experienced a deterioration in debt risk ratings, exogenous shocks and/or severe governance issues (e.g. undisclosed borrowing, misappropriation of funds) were important underlying causes, although fiscal slippages also contributed.

Figure 1.LIC DSF Transition Matrices for Debt Vulnerabilities

Sources: IMF country reports and IMF staff calculations.

PRGT-Eligible Countries: Risk of Debt Distress

(Percent)

Sources: IMF country reports and IMF staff calculations.

Conclusions

17. The design of future Fund-supported programs will be informed by lessons drawn from the ROC and the upcoming Review of the Debt Limits Policy. The ROC’s analysis yielded recommendations on strengthening program design and conditionality. Those recommendations, if endorsed by the Board, will be incorporated into relevant staff guidance, including the Operational Guidance Note to IMF Staff on the 2002 Conditionality Guidelines, to help inform future program design. They will also be reflected in an updated version of the Handbook of Facilities for Low Income Countries, which will also incorporate reforms adopted in the context of the LIC Facilities Review. Key program design lessons from the ROC include:17

  • the importance of operating with realistic program projections;
  • the need to protect the quality of fiscal adjustment, including the potential role for more granular fiscal conditionality such as a floor on budgetary revenues;
  • the need to give greater attention to debt transparency, the adequacy of coverage of public sector debt, and improving debt management policies;
  • the need for better tailoring and prioritization of structural conditionality, coupled with realism in implementation timelines; and
  • the need to build and support program ownership by national authorities and the wider public.

18. Reforms proposed in the LIC Facilities Review are aligned with key staff findings of the ROC. The enhanced emphasis on debt vulnerabilities is reflected in the proposals for new safeguards for high access programs. The importance of customizing program design to the specific circumstances of FCS is reflected in the proposal to use the full flexibility of the ECF to focus on immediate reform priorities for countries emerging from conflict or facing domestic instability. The merits of applying greater realism in implementation timetables for structural reforms is reflected in the proposal to allow ECF arrangements with an initial duration of five years under selected circumstances.

B. Catalytic Role of the Fund

19. While official development assistance (ODA) has declined steadily as a share of recipient GDP for most aid recipients, Fund-supported programs continue to have an important catalytic role in LICs. Median ODA as a share of PRGT-eligible members’ GDP remained broadly unchanged at 11 percent until the late 2000s, but subsequently fell to about 7 percent in the last five years, with the drop in ODA more pronounced for better-off LICs and non-fragile states. But donor financing continued to play an important role in Fund-supported programs and was more sizeable for ECF-supported programs that remained on-track throughout most of the program period.

Figure 2.Fund Catalytic Role 1/

Sources: OECD ODA database, WEO, and IMF staff calculations.

1/ ODA flows are measured on a gross basis, and include both DAC and non-DAC members.

2/ Non-prospective blenders are poorer PRGT-eligible members, who do not meet income or market criteria for blending.

Reform Proposals

A. Access Policies and Financing Terms

Access Policies18

Proposal: A generalized increase of one-third in access limits and norms for all concessional facilities.

20. At the Board discussion in July 2018, most Directors expressed openness to an appropriately calibrated generalized increase in access limits and norms for all concessional facilities. Directors underscored that any increase in access limits and norms should be compatible with maintaining the financial sustainability of the PRGT. Subsequent informal consultations indicated that higher concessional access limits and norms were generally seen as needed to avoid access erosion. However, a variety of views were expressed on the appropriate size of the increase, including considering greater frequency of adjustments in access limits rather than periodic discrete jumps.

21. Higher concessional access limits and norms are needed to avoid access erosion and preserve the potential financing contribution of Fund program engagement in LICs. The staff proposal for a generalized increase of one-third (33.3 percent) applicable to access norms and to annual and cumulative global PRGT access limits and RCF sub-limits would broadly restore current access limits in relation to GDP and trade exposure to the levels achieved when past generalized access increases took place (2009 and 2015). In relation to gross financing needs (GFN), access would still be below the levels after previous increases, especially for comparatively better-off LICs that meet the income or market access criteria for blending (Figure 3, “prospective blenders”). The increase is calibrated to be consistent with preserving the PRGT’s self-sustainability in conjunction with the other reforms proposed in this paper.19

Figure 3.Access Erosion Metrics

Source: WEO, and IMF staff calculations.

Note: “Prospective blenders” is the group of PRGT-eligible countries that meet the income or market access criteria for blending, irrespective of their debt risk rating. This is a broader set of PRGT-eligible countries than the group of “presumed blenders,” which also meet the income or market access criteria for blending but exempt some countries from blending due to their risk of debt distress.

22. The proposed increase in access limits and norms is sufficiently ambitious that interim adjustments before the next review of LIC facilities are not envisaged at this juncture. A generalized increase of a third now would address access erosion at an earlier point than in previous episodes (see text chart), implying higher average access over a multi-year period. The case for greater frequency of adjustments (of a more modest size) could be examined at the next LIC facilities review; any such adjustments would need to be linked to an assessment of consistency with the PRGT’s financial self-sustainability.20

Safeguards21

Proposal: To strengthen safeguards against credit risk, modify high access (HA) procedures by establishing an additional threshold on projected outstanding credit to the PRGT, and provide for more timely and better-informed Board engagements for both HA and exceptional access (EA) requests.22

23. Given rising debt vulnerabilities, staff proposes strengthened safeguards to accompany the generalized access increase. At the informal session in March 2019 views were expressed that the LIC Facilities Review should take account of rising debt vulnerabilities in LICs, given that rising debt vulnerabilities may affect the capacity of borrowing countries to repay the Fund. While recognizing that the primary safeguard of Fund resources comes from the specification of program design and conditionality, staff proposes to strengthen existing procedural safeguards for HA requests (Box 1) along two dimensions.

24. First, it is proposed to introduce an additional trigger for applying HA procedures to address concerns about a possible increase in credit risks to the PRGT. Currently, HA safeguards are triggered when access to PRGT resources over any 36-month period exceeds 135 percent of quota; this “flow trigger” allows closer scrutiny of financing requests that involve relatively large use of scarce concessional resources. Staff sees merit in adding a “stock trigger” for high overall exposure of a country to the PRGT as a complementary metric justifying closer scrutiny of program requests.

Box 1.Informational Requirements for Informal Board Engagement Under HA 1/

  • The factors underlying the large BoP need, taking into account financing from donors
  • A brief summary of main policy measures and the macro framework
  • The expected strength of the program and capacity to repay
  • An analysis of debt vulnerabilities
  • A reference to the impact on the Funds concessional resources
  • The likely time table for discussion with authorities
  • An SEI table
  • If possible, DSA charts

Informational requirements are the same as those required for exceptional access under the PRGT.

1/Modification of Access Policies for the Poverty Reduction and Growth Facility and the Exogenous Shocks Facility (IMF, 2009a) para. 25 and A New Architecture of Facilities for Low-Income Countries (IMF, 2009b) para. 87 and 2017 Handbook of IMF Facilities for Low-Income Countries (IMF, 2017b).

Specifically, it is proposed that the HA procedure apply to all new financing requests (including augmentations) where (i) the amount requested exceeds 60 percent of the cumulative normal access limit (180 percent of quota, after the proposed access increase) over a 36-month period or (ii) outstanding credit to the PRGT exceeds or is projected (over the lifetime of the existing or proposed arrangement) to exceed a threshold, set at the equivalent of three-quarters of the cumulative normal limit (225 percent after the proposed access increase).23 The case for this specific threshold level is that the borrowing member is “running out of space:” a successor PRGT arrangement at a conventional moderate level (the low access norm for a three-year ECF arrangement) would not be possible within the normal access limits. 24

Projected Net Credit Outstanding (PRGT Countries, 2023) 1/

(In percent quota)

Sources: FIN Query Database and IMF staff calculations.

1/ Net PRGT Credit Outstanding through 2023 based on commitments under existing PRGT arrangements and scheduled repayments, augmented by a new PRGT arrangement at the new low-access norm (75 percent of quota).

25. Second, it is proposed that informational requirements for informal HA Board meetings be strengthened to enhance assessment of debt sustainability and capacity to repay. Staff judges that existing HA informational requirements are broadly adequate but sees room to strengthen safeguards by clarifying best practice with respect to the analysis of debt sustainability and capacity to repay. This would typically involve requiring DSA charts (which are currently to be provided “if possible”); a preliminary assessment of the risk of debt distress facing the member, along with discussion of any deficiencies in the quality/transparency of public debt data; and an assessment of capacity to repay the Fund, including an updated capacity-to-repay table. The enhanced analysis would also help strengthen the analytical and information content of the internal review process at an early stage.

26. There is also a need for greater clarity on the appropriate timing for the informal-to-engage Board session for HA cases. The current policy requires that the Board engagement takes place early and before reaching ad referendum understandings on a PRGT-supported program with the authorities. Recent experience with HA cases indicates that informal Board engagements often took place only a few weeks before staff reports were circulated to departments, which suggests that discussions on PRGT-supported programs had reached an advanced stage at the time of informal HA Board sessions—potentially limiting the scope to incorporate Directors’ feedback on program design and access levels. It is therefore proposed that informal Board engagements be required to take place as soon as management agrees that a new or augmented financing request involving HA could be appropriate, similar to GRA practices for EA.25 The proposed changes to informational requirements and timing of informal Board briefings will also apply to EA requests under the PRGT.

27. Rising debt levels in many countries underscore the importance of careful scrutiny of debt sustainability for countries requesting Fund financial support. The new LIC Debt Sustainability Framework (LIC DSF) contains several features intended to more fully capture debt vulnerabilities in PRGT-eligible countries, including “realism” tools to assess the credibility of and risks associated with the program baseline and an enhanced focus on the correct measurement of public sector debt. The latter includes identification of data weaknesses and areas outside the perimeter of measured public debt where public liabilities (including contingent liabilities) may reside. These risk factors should feature as inputs in the discussion of capacity to repay the Fund. The upcoming review of the Fund’s debt limit policies will consider whether the policy has been effective in containing debt vulnerabilities and achieving its other objectives, and recommend reforms where shortcomings are identified.

Blending Policies26

Proposal: Remove the exclusion from presumed blending for higher-income LICs at high risk of debt distress provided they have substantial market access, including on a prospective basis.

28. At the July 2018 Board meeting and subsequent discussions, there was broad support for removing the blanket exemption from the presumption to blend for countries at high risk of debt distress that have substantial access to international markets. That said, views were expressed in March that case-by-case judgment should be employed in determining whether countries at high risk of debt distress should be presumed to blend.27 The approach laid out here allows for exercising judgment in assessing whether the blending requirement under the market access criterion that the country has prospective market access is met.

29. Rising debt vulnerabilities among higher-income LICs have reduced the targeting of concessional resources to the poorest LICs. Under the current blending policy, PRGT-eligible members are presumed to blend PRGT with GRA resources if they (i) meet either the market access or income criterion for blending; and (ii) are not classified as in debt distress or at high risk of debt distress under the LIC DSF. In recent years, there has been a significant increase in the number of LICs meeting the income and/or market access criteria that are now excluded from the presumption of blending because they have moved into the high debt risk category (text chart).

Number of Blenders – By Qualification Criteria

Sources: BEL database, IMF country report, WDI, WEO, and IMF staff calculations.

30. The demands on scarce PRGT resources arising from the debt-related blending exclusion of PRGT-eligible members meeting the income and/or market access criterion for blending are significant. The share of PRGT disbursements to this category of members rose sharply—from near zero in 2014 to 30 percent of all PRGT disbursements in 2017 and 2018 (text chart). Financial pressures could further intensify if more PRGT-eligible members move to high debt distress risk. Staff’s analysis of how the blending exclusion affects the outlook for the PRGT’s self-sustained capacity is contained in the companion paper Review of the Financing of the Fund’s Concessional Assistance and Debt Relief to Low-Income Countries (text tables 4 and 5).

PRGT Use Benefiting Exempted Blenders

(Share of all PRGT, percent)

Sources: FIN query database, MONA, WEO, and IMF staff calculations.

31. Staff proposes a removal of the blending exclusion for higher-income LICs at high risk of debt distress provided that they have substantial market access, including on a prospective basis. 28

  • Blending would be presumed for LICs that are assessed to be at high risk of debt distress (but are not in debt distress), if their per capita income is above the IDA operational cutoff threshold and they have substantial access to international financial markets on both a past and prospective basis.
  • The test for substantial past access to international financial markets in such cases would be closely aligned with that for graduation from the PRGT, i.e., the country has issued or guaranteed eligible external debt during at least three of the past five years in a cumulative amount equivalent to at least 50 percent of the member’s quota.
  • Assessment of prospective access to international financial markets would require judgement, based on such factors as the evolution of debt vulnerabilities in the context of the program DSA, the evolution of sovereign spreads and credit ratings over time, program assumptions on commercial financing, and the scale and evolution of nonresident holdings of domestic-currency debt. The quality of public debt data—including coverage of public sector entities outside central government and of publicly guaranteed debt, and transparency on terms and conditions—would also be an important factor in the assessment, given the threat to prospective market access from any significant debt surprises. Specific guidance on the assessment of prospective market access will be included in the Handbook on IMF Facilities for LICs.29
Market vs. IMF Borrowing
Market
Avg market rate (%)7.5%
10-year US Treasury (%) 1/2.9%
Average EM BI spread (bps) 1/463
Ghana441
Kenya446
Honduras268
Senegal406
Zambia822
Cote d’Ivoire393
IMF
GRA rate (%) 1/1.9%
ECF rate (%)0.0%
Blended GRA-ECF rate (%)1.3%
Sources: Bloomberg, Haver, and IMF staff calculations.

2018 averages.

Sources: Bloomberg, Haver, and IMF staff calculations.

2018 averages.

32. Several considerations weigh in favor of the proposed changes. By not excluding higher-income LICs with substantial market access from blending, the proposal would better target scarce PRGT resources to the poorer and more vulnerable LICs. Moreover:

  • The estimated increase in the average cost of accessing Fund resources for countries no longer benefitting from the exclusion is not large.30 The interest rate for blended PRGT-GRA financing averaged 1.3 percent in 2018, compared with zero percent for ECF financing. This is still very low compared to the cost of market borrowing (text table).31 Indeed, countries that have accessed international markets in significant volumes often continue to tap market-priced loans and international bond markets during a Fund-supported program, maintaining market access despite high risk of debt distress.32
  • Substantial market access is an indication that private creditors see debt vulnerabilities as manageable, even when classified as high risk. This provides some reassurance about access to market liquidity and rollover of existing debt. Moreover, countries with substantial past market access have significant amounts of debt on commercial terms (see text chart)—debt that can be restructured should the country enter into debt distress, providing flexibility in resolving debt problems.33
  • The proposal helps mitigate concerns that large-scale borrowing on non-concessional terms is de facto rewarded ex post through greater subsidization. Under the current blending policy, countries that have entered into high risk of debt distress because of the accumulation of substantial amounts of market-priced debt, including from international financial markets, are exempted from the blending requirement and obtain all Fund financing on PRGT terms. In effect, the PRGT would subsidize higher-income LICs that have relied heavily on market borrowing and thereby moved into a higher debt risk category.

Public Debt by Type

(Percent of total)

Sources: National authorities and staff estimates.

1/ Excludes publicly-guaranteed debt.

2/ Includes publicly-guaranteed debt.

3/ Staff estimates.

Interest Rate Setting Mechanism34

33. The companion paper PRGT—Review of Interest Rate Structure proposes to harmonize the SCF interest rate with the lower ECF rate schedule. This would make Fund financing somewhat more concessional while simplifying the PRGT financing framework. The revised interest mechanism, which would imply zero interest rates for all three PRGT facilities until the next biennial review in mid-2021, received broad support from Directors in July 2018.35

B. Supporting LICs in Fragile and Conflict-Affected Situations36

Proposal: In addition to the generalized access increase, double the annual RCF access limit under the regular window, while introducing some associated safeguards. For FCS facing substantial near-term uncertainties, make full use of the flexibility allowed under the ECF to focus attention on near-term targets fleshed out on an annual basis as the arrangement unfolds.

34. In the July 2018 LIC Facilities Review paper, two specific concerns regarding Fund financial support for low-income fragile states were discussed: (i) the access limit to the RCF was seen as unduly constraining financial support to countries unable to implement an upper tranche conditionality (UCT)-quality program; and (ii) the three-year time horizon for ECF arrangements was seen as rather long for FCS facing emerging from conflict and/or facing elevated short-term uncertainties.

35. Most Directors were open to addressing both concerns. On the first issue, there was broad support for increasing the annual access limit of the regular window of the RCF by more than the generalized access increase. On the second issue, the Executive Directors expressed mixed views as to whether changes to the design of the ECF were needed to address the specific challenges faced by FCS. At the March 2019 informal session, staff explained their preference for making use of the flexibility inherent in the ECF instrument to customize programs to the specific circumstances of FCS, rather than pursuing the option of introducing a new shorter-term variant of the ECF.37 The informal session suggested the need for further details on how this flexibility could be achieved (see below).

Changes to the RCF “Regular Window”38

36. To address the first concern, it is proposed to double the RCF annual access limit under the regular window, in addition to the generalized one-third increase. This would bring the annual access limit for the regular RCF window to 50 percent of quota (see Annex I) and provide greater flexibility to assist FCS in situations where a UCT-quality program is not yet feasible. Financial support for countries that do not have the capacity to implement UCT-quality economic programs should ideally be provided in the form of outright grants (rather than adding to debt burdens), implying that Fund loans should not be the main source of financing in such situations. However, there can be emergency circumstances (e.g., emergency post-conflict financing) where Fund financing can provide support at a vital time. Doubling the annual access limit would provide room for higher financial support in such situations.

37. Staff also proposes to introduce safeguards to ensure that the higher annual access limit does not become the de facto average access level under the regular RCF window. First, it is proposed to introduce a norm for annual access under the regular RCF window at one-half of the annual limit to underpin expectations that the regular RCF window would typically provide access of around 25 percent of quota; as with the ECF and SCF, the norm would be neither a floor nor a ceiling. Second, it is proposed to introduce a “per disbursement limit” at 25 percent of quota; annual access in excess of 25 percent of quota could occur only through an additional RCF disbursement within the year and would be linked to a track record of adequate macroeconomic policies (e.g., through an SMP), consistent with the current requirement for repeat use under the regular window.39 Third, the cumulative limit on access to the RCF via the regular window would increase only in line with the general increase, providing incentives to limit reliance on accessing resources via the RCF for an extended period.40,41

Flexible Use of the ECF for FCS

38. Regarding the second concern, FCS emerging from conflict and/or facing substantial domestic instability or uncertainties need to focus attention on near-term objectives. Elaborating detailed three-year economic plans can be ambitious in such situations and may distract attention from near-term imperatives. When a country lacks capacity to implement a UCT-quality program, the Fund can provide financial support through the RCF. Where a UCT-quality program is feasible despite fragilities and uncertainty, the ECF can be used to support a program with streamlined conditionality that focuses on near-term stabilization needs, guided by broad medium-term objectives, while fleshing out further program specifics as the ECF arrangement unfolds.

39. Existing policies provide the basis for such use of the ECF. Several elements underpin the ECF’s flexibility:

  • The core requirements of an ECF-supported program are that it supports the member in making significant progress towards a sustainable macroeconomic position consistent with strong and durable poverty reduction and growth, in line with the purpose of the ECF, and that it meets the UCT conditionality standard. A member requesting Fund support under an ECF arrangement is required to indicate how the program advances the member’s poverty reduction and growth objectives and present a detailed statement of the policies and measures it intends to pursue for the first twelve months of the arrangement, leaving scope for firming up policies beyond this period until later reviews.42
  • The Guidelines on Conditionality emphasize the need for tailoring programs to a member’s circumstances including consideration of its implementation capacity.43 This allows programs for FCS to focus on near-term reforms and, as implementation capacity strengthens and confidence in the near-term outlook increases, to flesh out a broader reform agenda.
  • The Staff Guidance Note on the Fund’s Engagement with Countries in Fragile Situations reinforces the message that conditionality should be tailored to the weak implementation capacity of FCS. Three principles specified in the Guidance Note are particularly relevant: (i) the pace of macroeconomic adjustment should be appropriately tailored, taking account of weak institutional capacity and supporting essential cohesion-building; (ii) there is a need for “quick wins” to deliver early successes and build support for the government’s reform agenda; and (iii) there should be a prioritized and gradual structural reform agenda.44

Box 2.Case Studies—Flexible and Tailored Approach of the ECF for FCS

Case studies illustrate that a flexible and tailored approach has been used in past ECF arrangements with FCS. We focus here on the examples of Chad (2014), Mali (2013), and Yemen (2010) (discussed further in Annex II).

  • Near-term focus. Each arrangement involved an explicit focus on immediate stabilization objectives (two of which were accompanied by large-scale donor support), with streamlined use of structural conditionality and medium-term policies specified in future reviews.
  • The ECF arrangement with Chad (2014) featured a focused reform agenda. There were only four structural benchmarks (SBs) at program initiation—much below the average of ten such conditions (SBs or prior actions (PAs)) in other ECF-supported programs. Parsimony and prioritization, in light of weak capacity and recurrent shocks, were also applied throughout the four completed reviews (after which the arrangement was cancelled and replaced following the collapse of oil prices) (text charts). Structural conditionality focused on two areas only—fiscal and debt management, though with a small rise in the number of conditions set in subsequent reviews.
  • Reforms in the ECF arrangement with Mali (2013) gradually increased with growing capacity. The program featured five SBs at inception, and a gradual increase in the number of structural conditions (SBs and PAs) over the program period. The functional areas covered also expanded: the program focused initially on fiscal management, but gradually added reforms in debt management, governance, and financial sector stability. The Mali case study also highlights the importance of a strong capacity development focus.
  • The three-year ECF arrangement with Yemen (2010) included two PAs and only three SBs. These conditions all focused on tax broadening and expenditure control measures that directly underpinned the fiscal adjustment effort to reduce reliance on central bank financing, which had led to sizeable reserve losses. While program implementation was derailed after approval by the escalation of civil conflict, program design was appropriate.

Chad (2014): Parsimonious Conditionality

Source: MONA, and IMF staff calculations.

Mali (2013): Evolution and Focus of Structural Conditionality

Source: MONA, and IMF staff calculations.

40. Programs supported under ECF arrangements with a near-term focus suitable for fragile states would typically contain the following elements, while meeting UCT standards and being consistent with the provisions under the PRGT Instrument:

  • At approval, the program should describe broad objectives for the full program period supported by a medium-term macroeconomic framework and DSA and include a detailed statement of the critical policies and measures the member intends to pursue for the first 12 months of the arrangement. The structural reform agenda and related conditionality would be streamlined to reflect capacity constraints and prioritization of immediate stabilization objectives. It is understood that the medium-term framework will be subject to higher uncertainty, and that medium-term policy objectives may have to be adapted as circumstances evolve.
  • Specific policies and measures after the first 12 months will be defined in the context of future reviews, in line with the current provisions of the ECF.
  • The medium-term program objectives should enable the member to make significant progress toward a stable and sustainable macroeconomic position consistent with strong and durable poverty reduction and growth. Such progress could be evidenced, for example (depending on the relevant macroeconomic vulnerabilities) by stronger fiscal or external positions, reduced debt vulnerabilities, higher levels of foreign exchange reserves, and/or more contained inflation.
  • The program should aim to strengthen institutional capacity to implement appropriate macroeconomic policies.
  • An Economic Development Document (EDD) would normally be expected by the time of the second review, but the member may request Board approval for an extension of this deadline to accommodate limited institutional capacity (per staff proposal in section D).

C. Supporting LICs Vulnerable to Natural Disasters

RCF Limits45

Proposal: Raise the cumulative RCF access limit for disbursements associated with large natural disasters to a level that is one-third higher than the cumulative RCF limit applicable to all other RCF disbursements.46

41. Recent informal consultations suggest that this proposal, which seeks to boost the Fund’s capacity to provide urgently needed liquidity after large natural disasters, would be viewed favorably by Executive Directors. The cumulative access limit under the RCF currently leaves limited room to support members with high outstanding RCF exposure when they are hit by a large natural disaster. A new annual access limit of 60 percent of quota was introduced in 2017 for urgent BoP needs arising from large natural disasters (where damage exceeds 20 percent of GDP). However, the cumulative limit of 75 percent of quota under the RCF was left unchanged, providing limited room to support members hit by a large natural disaster if they already have significant outstanding RCF exposure. Raising the cumulative RCF limit for disbursements associated with large natural disasters would increase the scope for providing Fund support. Given that the large natural disasters threshold is likely to be met mainly by small states, the impact on PRGT finances should be modest, and, in light of the large size of the BoP needs following a large natural disaster, Fund assistance would remain catalytic.

42. Other changes proposed in the LIC Facilities Review will facilitate Fund support to countries vulnerable to large natural disasters. First, the generalized access increase, including under the RCF, will allow higher levels of support. Second, the proposed changes to the SCF (see below) will make it a more flexible instrument to provide precautionary support and assistance for short-term BoP needs, including to countries vulnerable to large natural disasters. Third, ECF arrangements with an initial longer duration may better match the time-frame needed to implement strategies to build resilience to natural disasters in countries with protracted BoP problems where disaster vulnerability is an ongoing threat to macroeconomic stability (see below). Existing instruments can also be tailored to resilience-building strategies, as discussed in the Board paper on “Building Resilience in Countries Vulnerable to Natural Disasters.

Reforms to the Rapid Financing Instrument (RFI)

Proposal: Raise the annual and cumulative access limits under the RFI by one-third and raise the cumulative access limit for disbursements associated with large natural disasters by an additional one-third.

43. It is proposed that the increases in access limits envisaged for the RCF also apply to the RFI. Specifically, the annual access limit under the RFI would be increased from 37.5 to 50 percent of quota and the annual access limit under the large natural disasters window from 60 to 80 percent of quota, while the cumulative access limit would be raised from 75 to 100 percent of quota, except in the case of large natural disasters, where it would be increased to 133.33 percent of quota. The access safeguards proposed for the RCF (limits on individual disbursements under the regular window and an annual norm) do not apply to the RFI, which does not have separate “regular” and “exogenous shock” windows.

  • This approach is consistent with previous decisions since 2015 to increase access limits under the two instruments simultaneously.47
  • The rationale for increasing access limits under the RCF also applies to the RFI, namely to improve the Fund’s flexibility to support FCS and countries hit by large natural disasters.48

D. Enhancing Flexibility in Supporting Reform Programs

44. Program design needs to be tailored to country circumstances, including to support the most critical medium-term reforms. There is scope to enhance the flexibility of both the SCF and ECF to support tailoring of program design, including by allowing higher access in cases of precautionary support, extending the maximum length of SCF arrangements and the maximum initial duration of ECF arrangements, and giving more time for the preparation of national development strategies. Staff guidance on the use of LIC Facilities will be updated to reflect the lessons from the ROC (see paragraphs 12–14 above).

Precautionary Support Under the SCF49

Proposal: Extend the maximum duration of an SCF arrangement from two to three years.50 Abolish sub-limits on access for precautionary SCF use.

45. Directors broadly supported raising the maximum length of the SCF from 24 to 36 months in the July 2018 Board discussion. This reform would provide LICs with access to a three-year precautionary arrangement within the concessional facilities, while facilitating blending of SCF and SBA arrangements where called for under the blending policy (see (IMF (2018b), para. 47). The SCF and ECF facilities would remain legally and substantively distinct: the SCF is available to countries with short-term BoP needs that are expected to be resolved within “two years but in any case not more than three years” (IMF, 2009b?), while the ECF is for use by countries with protracted BoP problems, who are expected only to make significant progress toward a stable and sustainable macroeconomic position during the program period. The extension of the maximum length of an SCF arrangement by one year thus remains consistent with the purpose of the SCF.

46. Directors also broadly supported removing the sub-limits on the SCF’s precautionary access. Concerns at the time of the creation of the SCF that precautionary use could tie up large amounts of scarce PRGT resources have not materialized, as explained in IMF (2018b), para 47. There is no convincing case for maintaining the sub-limits, which act as a disincentive to LICs seeking precautionary (rather than disbursing) support from the Fund.

Five-Year ECF Arrangement51

Proposal: Extend the maximum initial duration of ECF arrangements from four to five years.52

47. There was considerable support from Directors for extending the maximum initial duration of ECF arrangements to five years, although some noted that back-to-back three-year arrangements could achieve broadly similar objectives and questioned whether demand would materialize for a longer-term ECF arrangement.

48. A longer ECF arrangement (i.e., up to five years from the outset) may be appropriate in circumstances when program success depends critically on longer-term reform efforts. This could be the case where major structural reforms, such as building a national tax system or achieving significant economic diversification, are needed to achieve a significant and sustained improvement in a country’s macroeconomic position and may require more than the norm of three years to plan and implement. In such cases, a five-year program would allow for more realistic timetables for critical reforms, enhancing the chances of sustained program success—although these reforms could also be supported in the context of successive three-year programs. The longer-term program is more likely to be the preferred choice when there are strong merits in anchoring the government’s program on a specific medium-term strategy (such as a five-year plan or a medium-term revenue strategy).

49. To justify a five-year program from the outset, a well-sequenced reform plan with strong ownership should normally be in place, supported by a coherent technical assistance program from development partners and grounded in the country’s development and poverty reduction plan. The presumption would remain that the length of an ECF arrangement would normally be three years; staff would need to make the case for a longer-term arrangement when it is proposed.

Strengthening Program Links to Poverty Reduction53

50. Countries receiving concessional financial support from the PRGT are expected to have in place a strategy for reducing poverty while supporting strong economic growth. This requirement is currently met by producing an Economic Development Document (EDD) that outlines the government’s specific plans for achieving these objectives. Prior to 2015, this requirement was met by producing a Poverty Reduction Strategy (PRS) document to be issued to the Board of the Fund within specified periodicities and which was the subject of staff analysis. The EDD was introduced as a replacement for the PRS document when the World Bank decided that it no longer needed the PRS documents for its concessional support.

Proposal: Rename the EDD to provide it with a title more closely linked to its objectives.

51. The EDD is proposed to be re-named the Poverty Reduction and Growth Strategy (PRGS), while retaining its current functions and required content.54 The label EDD is opaque to outsiders, providing no indications as to the purpose and contents of the document. By contrast, the title of “Poverty Reduction and Growth Strategy” is both clear as to the purpose and contents of the document and closely aligned with the objective of meeting the requirements of the PRGT.

Proposal: Standardize the use of the PRGS across the ECF, SCF and PSI by requiring a PRGS document to be produced whenever an ECF or SCF arrangement, or PSI, has an initial duration of more than two years. Allow greater flexibility on the timeline for producing a PRGS for countries with weak capacity to avoid compromising on ownership.55

52. Directors were supportive of harmonizing the requirement to produce an EDD across the ECF, SCF, and PSI, while providing more flexibility on the timeline for producing an EDD.

  • It is proposed that SCF arrangements with an initial duration exceeding two years have the same EDD (PRGS) requirements as the ECF arrangement.
  • To provide flexibility to country authorities, it is proposed to extend the current deadline for meeting the EDD (PRGS) requirement by the first program review to the second program review.
  • As noted above, there are circumstances where a country seeking support under the ECF may need to focus limited institutional capacity on near-term objectives and measures to enhance economic and political stability; examples include countries emerging from conflict or confronted with significant domestic instability. In situations where a country has limited institutional capacity for meeting EDD (PRGS) requirements by the second review, it may request Board approval of an extension of the deadline for meeting such EDD (PRGS) requirements up to the fourth review; and may make one further request for an extension up to the sixth review where (i) an adequate justification can be provided and (ii) the arrangement (when the second request is made) has a duration of at least four years.56, 57

E. Other Issues

53. A technical clarification to the criteria determining eligibility for exceptional access (EA) is proposed to eliminate an anomaly. Under the blending policy, countries are not presumed to blend on the basis of market access if their per capita GNI is at or below 80 percent of the IDA threshold. Under the current EA policy, countries are precluded from exceptional access if they meet the market access criterion: there is no linkage to GNI, implying that even very poor countries (the intended beneficiaries of EA) are not eligible for EA if they have market access.58 It is proposed that language be added to the PRGT instrument to explicitly align the market access criterion used for EA purposes with that used in blending, so that only countries with market access and GNI per capita above 80 percent of the IDA threshold are excluded from EA on grounds of market access.59 An explicit link to blending criteria will also clearly establish the definition of market access for purposes of EA.

54. Other related changes (Annex I). In line with the generalized increase in access, it is proposed to make necessary changes to relevant policies that contain an access threshold trigger. Specifically, it is proposed to (i) require a new DSA when a new financing request brings total access to more than 80 percent of quota (previously 60 percent of quota); and (ii) update the threshold from 12.5 percent of quota to 15 percent of quota above which Lapse of Time procedures for augmentation requests are not permitted.

Resource Implications of Proposed Reforms

55. The proposed reform package would be generally consistent with the self-sustained PRGT financing framework, with risks evenly balanced over the coming decade.60 In particular, the proposed reform package is projected to result in average annual demand of SDR 1.0–1.7 billion on a subsidy-use basis over the next decade. Based on this demand range, self-sustained capacity would reach SDR 1.1–1.4 billion by 2028. The costs resulting from higher access limits and norms, a lower SCF interest rate, and enhanced facilities flexibility would be partly offset by the proposed modification of the blending exemption.

56. The evolution of the PRGT’s self-sustained capacity requires careful monitoring given downside risks. The longer-term outlook is subject to considerable uncertainty, with downside risks from both supply and demand factors, such as low investment returns on PRGT assets or prolonged periods of high and rising aggregate financing needs. The evolution of capacity will thus need to be monitored carefully, and policies reviewed periodically to ensure that the outlook for capacity remains in line with the base envelope of SDR 1¼ billion. A range of policy options and contingency measures can be triggered under the three-pillar PRGT framework in the event of a sustained disequilibrium between supply and demand.

Impact on the Fund’s Risk Profile

57. While some aspects of the proposals may potentially generate new risks, these have been anticipated, and mitigation measures have been proposed where needed. Key risk areas that may be affected by the proposals include:

  • Strategic Direction and Reputation. Addressing access erosion, improving targeting and tailoring, and increasing the flexibility of LIC facilities will further align the Fund’s lending toolkit with members’ changing needs. These measures can help sustain the membership’s support for the Fund’s strategic agenda, thus mitigating strategic and reputational risks.
  • Use of Fund resources (UFR) and self-sustainability of the PRGT. Making the Fund’s lending facilities more suited to the evolving challenges faced by members, and thus better able to help deliver viable outcomes for program countries, meaningfully mitigates UFR risks. On the other hand, higher access norms and limits and longer arrangements may add pressures to the self-sustainability of the PRGT. These pressures are mitigated by the removal of the exemption from presumed blending for qualified countries at high risk of debt distress, and this risk can be addressed by regular monitoring and annual updates on the integrity of the PRGT’s self-sustained financing framework.
  • Credit risk. Rising debt vulnerabilities in LICs could increase credit risks to Fund resources. This is mitigated by stronger procedural safeguards, including by linking high-access procedures to the PRGT’s overall exposure to program countries, and raising the informational requirements when keeping the Board informed on HA cases. The new LIC DSF, which is equipped to better capture countries’ debt vulnerabilities, and a stronger focus on debt transparency and management in program design, provides another safeguard.

Implementation Issues

58. Required Board Majorities. Adoption of the proposed changes requires the approval of the Executive Board by a majority of the votes cast. The proposals do not modify protected provisions; accordingly, there is no need to seek consent from affected lenders and contributors to the Trust. Various changes are proposed as a unified package.

59. Impact on existing arrangements on the adoption of the proposed reforms. Adoption of the proposed modifications to the blending policy would not affect existing arrangements and emergency financing requests already made; they will apply to any future new arrangements and outright RCF/RFI disbursements, approved after the adoption of the proposed decision. Adoption of the modifications to the HA safeguards will apply to new financing requests (including augmentations under existing arrangements) that results in exceeding any of the thresholds proposed in para. 24. The proposals to extend the maximum length of SCF arrangements to three years, abolish sub-limits on access for the precautionary SCF, and standardize the use of PRGS will apply immediately upon their adoption by the Board. The proposal to extend the maximum initial duration of ECF arrangements to five years will be effective for new ECF arrangements. The proposed new access limits and the clarifications to the EA criteria would be immediately applicable; access levels under any future arrangements or changes in access levels under existing arrangements would need to be justified in accordance with the standard access policies, including considerations on the size of the member’s BoP need.

Issues for Discussion

  • Do Directors support the proposed package of reforms to the LIC facilities?
  • Do Directors agree with the proposal to make changes to the Rapid Financing Instrument?
Annex I. Proposed Changes in Norms, Annual, and Cumulative Limits
Table 1.Summary of Norms, Limits, and Procedural Safeguards(Percent of quota)
Access LimitsCurrent (14th General Quota Review) 1/Proposed Reform
Cumulative access limits
All PRGT facilities-normal225.00300.00
All PRGT facilities-exceptional300.00400.00
RCF 2/3/75.00100.00 (133.33 see footnote 3/)
Annual access limits
All PRGT-facilities-normal75.00100.00
All PRGT-facilities-exceptional100.00133.33
SCF (precautionary) – average annual37.50..
SCF (precautionary), at approval56.25..
RCF 2/18.7550.00
RCF (exogenous shocks window) 2/37.5050.00
RCF (large natural disasters window) 2/60.0080.00
Per-disbursement-limit
RCF (regular window)..25.00
Norms 4/5/
3-year ECF – High access 6/90.00120.00
– Low access56.2575.00
18-month SCF – High access 7/90.00120.00
– Low access56.2575.00
RCF (norm for annual access under the regular window)..25.00
Blending proportions applicable to members presumed to blend (PRGT:GRA) 8/1:2 with concessional access capped at the applicable norm (all GRA thereafter)1:2 with concessional access capped at the applicable norm (all GRA thereafter)
Triggers for procedural safeguards on high access requests
Total access in any 24-month period—for DSA 9/60.0080.00
High Access Safeguards: An informal Board Meeting in advance of new PRGT request 10/Total access in any 36-month period of 135 percent of quotaTotal access in any 36-month period in excess of 180 percent of quota, or total Fund credit outstanding to exceed 225 percent of quota at any point during the program period
Triggers related to Lapse of Time procedures
Augmentations above which Lapse of Time procedures are not permitted12.5015.00

The current access limits in effect from January 26, 2016 do not affect disbursements under arrangements approved prior to that date and any changes in access levels is to be justified by balance of payments needs in accordance with the standard policies for augmentation of access amounts. Outstanding PRGT credit in existence as of January 26, 2016 counts towards the current annual and cumulative PRGT access limits.

Any RFI access also counts towards these limits.

This limit is raised by one-third under the proposed changes for disbursements associated with large natural disasters.

Norms provide guidance on what may constitute an appropriate level of access under PRGT facilities, but they should not be misconstrued as access limits or entitlements.

High access norms apply if PRGT credit outstanding is less than 75 percent of quota (100 percent after the proposed quota increase). Norms are not applicable if PRGT credit outstanding exceeding150 percent of quota (200 percent after the proposed access increase). In such cases, access is guided by consideration of the cumulative access limit, the expectation of future need for Fund support, and the repayment schedule.

For four-year or five-year ECF arrangements, access for the fourth and fifth year is expected to be set in line with the average annual access corresponding to the norm that would otherwise have applied to a successor three-year ECF arrangement. For countries whose outstanding PRGT access is above 150 percent of quota (200 percent after the proposed access increase), the norms do not apply.

For SCF arrangements of any other length, the norms will be proportionately adjusted to keep annualized average access unchanged.

For the RCF, which has no norm, the cap on access to concessional resources is the annual limit, while for the SCF treated as precautionary this cap applies to the average annual access limit.

A new DSA is also required for any PRGT financing request if it involves exceptional access to concessional resources or involves a member with high risk of debt distress or in debt distress.

An early informal meeting is also required if the financial request would involve exceptional access to concessional financing.

The current access limits in effect from January 26, 2016 do not affect disbursements under arrangements approved prior to that date and any changes in access levels is to be justified by balance of payments needs in accordance with the standard policies for augmentation of access amounts. Outstanding PRGT credit in existence as of January 26, 2016 counts towards the current annual and cumulative PRGT access limits.

Any RFI access also counts towards these limits.

This limit is raised by one-third under the proposed changes for disbursements associated with large natural disasters.

Norms provide guidance on what may constitute an appropriate level of access under PRGT facilities, but they should not be misconstrued as access limits or entitlements.

High access norms apply if PRGT credit outstanding is less than 75 percent of quota (100 percent after the proposed quota increase). Norms are not applicable if PRGT credit outstanding exceeding150 percent of quota (200 percent after the proposed access increase). In such cases, access is guided by consideration of the cumulative access limit, the expectation of future need for Fund support, and the repayment schedule.

For four-year or five-year ECF arrangements, access for the fourth and fifth year is expected to be set in line with the average annual access corresponding to the norm that would otherwise have applied to a successor three-year ECF arrangement. For countries whose outstanding PRGT access is above 150 percent of quota (200 percent after the proposed access increase), the norms do not apply.

For SCF arrangements of any other length, the norms will be proportionately adjusted to keep annualized average access unchanged.

For the RCF, which has no norm, the cap on access to concessional resources is the annual limit, while for the SCF treated as precautionary this cap applies to the average annual access limit.

A new DSA is also required for any PRGT financing request if it involves exceptional access to concessional resources or involves a member with high risk of debt distress or in debt distress.

An early informal meeting is also required if the financial request would involve exceptional access to concessional financing.

Annex II. Flexible Use of the ECF for Countries in Fragile Situations: Examples

Three case studies elaborate on how a short-term focus was embedded within a three-year program, conditionality was streamlined and reflected prioritization of immediate stabilization objectives, and institutional capacity was strengthened over the program period to implement appropriate macroeconomic policies.

A. Chad (2014)

What Was The Program Context, And Source of Fragility?

1. Chad’s fragility stemmed from a nexus of security challenges, weak institutional capacity, and large development needs. Despite significant oil revenues, domestic armed conflicts plagued the country up to 2009. Weak institutions led to pro-cyclical fiscal policy, poor governance, and inefficient spending. Regional security issues acutely compounded fragility. Chad had to provide assistance to thousands of refugees fleeing conflict in the Central African Republic (C.A.R.), and strengthen security measures along its borders, particularly with Libya, the C.A.R., Nigeria, and Cameroon. Chad was ranked 184th out of 186 countries on the UN’s 2013 Human Development Index.

2. Following an SMP, Chad requested an ECF in the context of a reprieve from domestic political instability. A period of internal political stability, the longest since it became independent, paved the way towards a Staff-Monitored Program (SMP) in 2013 that improved economic management, helped by stepped-up technical assistance by the Fund. A three-year ECF (120 percent of quota) followed the SMP. To support state-building and development needs, the program included a strong catalytic response from bilateral and multilateral donors. A donor conference held in Paris in June 2014 produced pledges of US$2.1 billion for the National Development Plan 2013– 15 and the National Program for Food Security 2014–21. The program also served as an opportunity for Chad to reach its HIPC Completion point.

How Did The Program Design Allow For A Focus On Near-Term Priorities Within A Medium-Term Framework?

3. The immediate challenge in Chad was to restore social stability and effectiveness of the government in the aftermath of deep insecurity. The Fund-supported program catalyzed large-scale donor support to this end, including for providing food assistance, repairing infrastructure, and rebuilding institutions.

4. The near-term focus of the program was to stabilize public finances and strengthen government fiscal control, including completion of PFM and fiscal reforms started under the SMP, informed by a prioritized TA agenda. The staff report (¶23) acknowledged that “since reforms are needed in many areas, careful prioritization is required in line with institutional capacity and prospective technical support from international donors.”1

5. Structural conditionality initially reflected focused and prioritized measures to improve fiscal management. The four structural benchmarks due at the first review focused on fiscal and debt management issues—submission of the budget to Parliament, limits on emergency spending procedures, preparation and publication of budget execution data, and steps to improve debt management. A similar focus was present in structural conditionality for the second review.

Chad (2014): Parsimonious Conditionality

Sources: MONA and IMF staff calculations.

6. As initial fragility was amplified by further shocks that arose under the program, the program continued to maintain a prioritized focus, taking into account capacity constraints. Lower oil prices and other shocks worsened the fiscal situation, which reinforced the need to maintain a focus on budget management. In light of these problems, along with capacity constraints, the program avoided widening its scope outside of fiscal and debt management conditionality.

7. Thanks to extensive TA engagement under the prior SMP, some medium-term objectives were fairly clear already at program inception, while others were left to be specified later. Having stabilized the non-oil primary balance at 14¼ percent of non-oil GDP in 2015, the program was designed to broadly maintain the deficit at this level while increasing non-oil revenue to finance investment, spur development and diversification to non-oil growth, address poverty, and deepen reforms in key fiscal areas, as informed by prior TA.

8. The authorities’ ambition to meet completion point triggers was another factor that helped clarify the medium-term strategy for the program. Despite reaching the Heavily Indebted Poor Countries (HIPC) Initiative Decision Point in 2001, Chad had not achieved the Completion Point. Reaching this goal was a key medium-term priority of the authorities under the program, especially in light of the recent improvement of relations with the international community.

How Did The Program Perform?

9. Declining oil prices and the debt burden complicated program implementation.2 In response to sharply declining oil prices and security challenges, access was augmented at the first (19 percent of quota) and combined third and fourth reviews respectively (24 percent of quota). Four reviews were completed, and maintained a well-prioritized reform focus. However, the intensifying effect of the oil price declines and security shocks as well as a heavy external debt service burden, put the program objectives out of reach and required a recalibration of policies. The program was cancelled and replaced with a new three-year ECF arrangement of 160 percent of quota.

B. Mali (2013)

What Was The Program Context, And Source Of Fragility?

10. In 2013, Mali emerged from the most serious security and political crisis in its recent history. In 2012, insurgents took control of the north, which sent ripples throughout the country: half a million people were displaced; a military coup destabilized the domestic political situation and the economy suffered. The security situation improved by mid–2013 following the French-led military intervention, which succeeded in dismantling the insurgents’ bases and regaining control over the towns and in northern Mali. The arrival in 2013 of a UN security force helped restore law and order in the north and paved the way for a political transition.

11. Mali’s request for a three-year ECF arrangement (32 percent of quota) was approved against the background of this highly fragile environment. The low-access request coincided with a resumption of donor support: in May 2013, at the international donor conference in Brussels, donors pledged €3.25 billion ($4.4 billion) in financial assistance.

How Did The Program Design Allow For A Focus On Near-Term Priorities Within A Medium-Term Framework?

12. The immediate challenge facing Mali was to restore social and economic stability, supported by large-scale donor assistance to rebuild the country and restore institutions. The ECF arrangement was intended to catalyze large-scale donor assistance to this end.

13. The staff report at program initiation clearly identified the short- and more medium-term focus of the program. The staff report noted Mali’s “dual challenges” of “dealing with the legacy of the recent past”—needing to advance reconciliation between north and south, improving governance, and consolidating the nascent recovery—and the “second challenge to strengthen the foundations for robust, poverty-reducing growth, while ensuring macroeconomic stability.” It also noted that “given the early stage of Mali’s recovery from the recent political and security crisis, reforms will ramp up progressively over the course of the program.” The initial phase—through mid-2014—focused on strengthening institutional capacity and developing strategies to address the most pressing issues. The following phases of the program would involve rolling out policy actions in these areas. These actions were specified at the time of the first and subsequent reviews of the arrangement.

14. Structural conditionality was initially very streamlined, focused on building capacity, but gradually embraced broader reform areas. Structural benchmarks progressively increased in number and expanded into more areas including to address weaknesses in the banking sector, governance, and to improve debt management.

Mali (2013): Evolution and Focus of Structural Conditionality

Sources: MONA and IMF staff calculations.

How Did The Program Perform?

15. While the program got off to a mixed start, it was quickly brought back on track, allowing resumption of donor support. The program was extended and augmented twice (by 32 and 47.5 percent of quota respectively). The gradualism embedded in the program allowed for reform momentum to build, engagement to strengthen, and capacity to build, while locking in earlier gains. The reforms helped the authorities to create the foundations for solid growth and subdued inflation. However, meaningfully reducing poverty remained a challenge.

C. Yemen (2010)

What Was The Program Context, And Source Of Fragility?

16. Yemen suffered from high poverty and poor security. Heavy dependence on declining oil revenues contributed to growing macroeconomic imbalances, while other sectors remained underdeveloped amid widespread poverty and a water shortage. A difficult security situation complicated fragilities: periodic breaches in the existing ceasefire kept tensions high, contributing to a humanitarian crisis in the north. Further pressure came from increasing protests and civil unrest in the south.

17. Yemen approached the Fund amid acutely deteriorating fiscal and external balances. Risks to macroeconomic and financial stability had been present for some time but had been masked by high oil prices. The sharp drop in oil production in 2007 and weak oil prices in late 2008 and 2009 brought these risks to center stage. The 2009 fiscal deficit reached a record high, while a deteriorating current account led to a steady decline in foreign exchange reserves and considerable pressures on the exchange rate. Security concerns added to these pressures. The authorities requested a three-year arrangement under the ECF (100 percent of quota).

18. Despite the ambitions of the Friends of Yemen forum to catalyze donor support, actual support was low. The program was based on conservative (identified) external financing of 0.7 percent of GDP, as past experience argued for a cautious approach in programming donor support.3 Donor support beyond the committed amounts, however, would have automatically allowed for program adjustments.

How Did The Program Design Allow For A Focus On Near-Term Priorities Within A Medium-Term Framework?

19. The staff report for program approval clearly separated the specific policies for 2010 (“the 2010 program” in the report) from broader medium-term goals.

  • The “2010 program” focus was exclusively to rein in central bank financing and avoid further reserve losses. Initially, a sharp fiscal adjustment in 2010 was needed to reduce the government’s reliance on domestic central bank financing and consequently stem pressures on the exchange rate and foreign exchange reserves. Hence, the program for 2010 focused on revenue measures—a new General Sales Tax and abolishing of tax exemptions—and expenditure control—mainly a phased adjustment in fuel prices. Some of the savings from the fuel price increase were used to increase social transfers, and the floor on social spending increased. Limited donor support, however, was a constraint on program design.
  • Structural conditionality at program approval was streamlined and focused on the fiscal area, with completion of structural measures due at the first review. Two prior actions and three structural benchmarks featured at program initiation, three of which focused on expenditure control and two on tax broadening. All three structural benchmarks were due by the first review. No structural conditionality was set beyond measures due at the first review.
  • Medium-term program objectives for the program were expected to be fleshed out at later reviews. Program objectives beyond near-term stabilization included preparing the economy for the prospect of lower oil revenues while boosting public investment and social spending, re-building reserves, and improving competitiveness. The medium-term outlook relied on reducing the non-hydrocarbon primary deficit by about 1 percent of GDP annually to achieve fiscal sustainability, while shifting the balance from current to capital spending and social transfers. A stronger social safety net and sustained higher levels of public investment were envisaged to help support growth and development objectives. At program approval, the authorities signaled in the MEFP that this expenditure shift would be further underpinned by civil service reforms (intended for completion by end-2011), but steps were not fleshed out, nor included in conditionality at program approval. Other medium-term objectives in the MEFP included reforms to boost competitiveness such as simplifying business regulation, improving property rights protection, and clarifying and simplifying bankruptcy procedures to bolster financial sector development. These were left to later reviews to be translated into concrete policy measures.

How Did The Program Perform?

20. No reviews were completed. The authorities cancelled the ECF arrangement in 2012 following an intensification of the political crisis and civil unrest, which led to a fall in economic activity and policy paralysis. Nevertheless, the initial program design focus on short-term priorities was sensible. Yemen subsequently requested Fund assistance under the RCF for urgent balance of payments needs in 2012, for 25 percent of quota.

References

    International Monetary Fund2002Guidelines on Conditionality” (Washington).

    International Monetary Fund2003The Acting Chair’s Summing Up Review of Access Policy Under the Credit Tranches and the Extended Fund Facility and Access Policy in Capital Account Crises—Modifications to the Supplement Reserve Facility and Follow-Up Issues Related to Exceptional Access Policy” (Washington).

    • Search Google Scholar
    • Export Citation

    International Monetary Fund2009aModification of Access Policies for the Poverty Reduction and Growth Facility and the Exogenous Shocks Facility” (Washington).

    • Search Google Scholar
    • Export Citation

    International Monetary Fund2009bA New Architecture for Facilities for Low Income Countries” (Washington).

    International Monetary Fund2009c2010 Eligibility to Use the Fund’s Facilities for Concessional Financing” (Washington).

    International Monetary Fund2011The Chairman’s Summing Up—Macroeconomic and Operational Challenges in Countries in Fragile Situations” (Washington).

    • Search Google Scholar
    • Export Citation

    International Monetary Fund2015Selected Streamlining Proposals Under the FY16–FY18 Medium-Term Budget— Implementation Issues” (Washington).

    • Search Google Scholar
    • Export Citation

    International Monetary Fund2017aReview of Eligibility to Use the Fund’s Facilities for Concessional Financing” (Washington).

    International Monetary Fund2017b2017 Handbook of Facilities for Low-Income Countries” (Washington).

    International Monetary Fund2018aUpdate on the Financing of the Fund’s Concessional Assistance and Debt Relief to Low-Income Countries” (Washington).

    • Search Google Scholar
    • Export Citation

    International Monetary Fund2018b2018 Review of Facilities for Low-Income Countries” (Washington).

    International Monetary Fund2018cThe Acting Chair’s Summing Up on the 2018 Review of Facilities for Low-Income Countries” (Washington).

    • Search Google Scholar
    • Export Citation

    International Monetary Fund2019aBuilding Resilience in Developing Countries Vulnerable to Large Natural Disasters” (Washington).

    • Search Google Scholar
    • Export Citation

    International Monetary Fund2019b2018 Review of Program Design and Conditionality” (Washington).

    International Monetary Fund2019c2018 Review of Program Design and Conditionality—Supplementary Information” (Washington).

    International Monetary Fund2019d2018 Review of Program Design and Conditionality—Case Studies” (Washington).

1

The current framework comprises three concessional lending facilities under the PRGT—the ECF, which provides medium-term support to LICs with protracted balance of payments problems; the SCF to help members deal with short-term balance of payment needs; and the Rapid Credit Facility (RCF) to provide rapid financing with limited conditionality to help members deal with urgent balance of payment needs—and one non-financial instrument, the Policy Support Instrument (PSI). In addition, PRGT-eligible members have access to the Funds’ resources in the General Resource Account (GRA), as well as to the non-financial Policy Coordination Instrument (PCI).

2

The first review was undertaken in 2012–13. See IMF (2018c).

3

For a full discussion of the determinants of PRGT eligibility, see Eligibility to Use the Fund’s Facilities for Concessional Financing, 2017.

4

Fund-supported programs include those under the ECF, SCF, and PSI. Fund-supported programs under the PSI and precautionary use of the SCF are non-disbursing.

5

See The Acting Chair’s Summing Up on the 2018 Review of Facilities for Low-Income Countries (IMF, 2018c) and the related Board paper (IMF, 2018b), discussed by the Board in July 2018.

6

The 2018 Review of Program Design and Conditionality (IMF, 2019b), The 2018 Review of Program Design and Conditionality—Supplementary Information (IMF, 2019c), and The 2018 Review of Program Design and Conditionality—Case Studies (IMF, 2019d).

7

See The 2018 Review of Program Design and Conditionality (IMF, 2019b), The 2018 Review of Program Design and Conditionality—Supplementary Information (IMF, 2019c), and The 2018 Review of Program Design and Conditionality—Case Studies (IMF, 2019d).

8

PRGT instruments here include the ECF and SCF, and also the PSI (a policy support instrument for PRGT-eligible countries). The ROC examined 50 PRGT-supported programs between 2011 and 2017 for which adequate data were available to evaluate program success. About half of the assessed programs were in FCS.

9

The methodologies for distinguishing between successful, partially successful, and unsuccessful PRGT and GRA programs are described in IMF (2019c), Section III.

10

Programs going off-track includes programs that failed to complete at least the last three reviews, or were replaced by new arrangements.

11

For further discussion, see IMF (2019b) and IMF (2019c), Section III.

12

Use of the ECF is appropriate in cases where the resolution of the entrenched macroeconomic imbalances that underlie the balance of payments problem is expected to extend over the medium or longer term. For further discussion of these concepts, see A New Architecture of Facilities for Low-Income Countries, IMF (2009b).

13

The analysis focuses on the evolution of the debt sustainability risk rating, which, if unchanged, could still imply a moderate increase in the public debt-to-GDP ratio.

14

Completion of most reviews implies completion of all but the last one or two reviews.

15

In 18 percent of programs, risk ratings improved from “in debt distress” or “high risk of debt distress” to moderate or low risk ratings over the three years following program approval; risk ratings worsened in 10 percent of programs.

16

While performance on debt vulnerabilities under PRGT programs was better than in LICs as a group, vulnerabilities rose in a significant share of cases; the Review of the Fund’s Debt Limit Policies will explore these issues and the implications for debt-related program conditionality.

17

See IMF (2019b) for a comprehensive discussion of the lessons for program design.

18

See also IMF (2018b), paras. 30–35 for background discussion.

19

See Section IV and the accompanying paper on the Review of the Financing of the Fund’s Concessional Assistance and Debt Relief to LICs.

20

The LIC facilities architecture was initially expected to be reviewed on a three-year cycle; the length of the cycle has since been increased to five years or more as needed (see Selected Streamlining Proposals Under the FY16-FY18 Medium-Term Budget—Implementation Issues, IMF (2015)).

21

See also IMF (2018b), Box 2, paras. 35–39 for background discussion.

22

Unlike the GRA lending framework, the PRGT lending framework has procedural safeguards for HA, which kick in below the threshold for exceptional access.

23

A de minimis exemption from the HA requirement is envisaged where the new program involves low access levels (i.e., where the new program entails a total commitment of not more than 15 percent of quota).

24

Based on the experience of the past three years, the latter threshold would have been reached in the top 9 percent of PRGT financing requests.

25

GRA procedural requirements for EA specify that the informal Board engagement should take place once management decides that new or augmented exceptional access may be appropriate (IMF (2003), March 5, 2003).

26

See also IMF (2018b), paras. 43–44 for background discussion.

27

Recent informal consultations with Executive Directors revealed a preference that blending for countries at high risk of debt distress should be accompanied by high standards of debt transparency, prudent overall non-concessional borrowing limits, and good prospects for reducing debt risks over the course of the program.

28

A member is presumed to blend based on market access if, in addition to having a per capita GNI of at least 80 percent of the IDA cut-off, the member has past and prospective market access. A member is assessed as meeting the past market access criterion if the sovereign has issued or guaranteed eligible external debt during at least two of the past five years in a cumulative amount equivalent to at least 25 percent of the member’s quota.

29

Consistent with current practices, countries that are at low or moderate risk of debt distress are presumed to have prospective market access, based on (i) established past market access and (ii) limited debt vulnerabilities.

30

During 2014–18, the following members that would otherwise be presumed to blend were able to meet their financing needs through PRGT resources alone because of debt vulnerabilities: Cameroon (2017 ECF), Ghana (2015 ECF), Grenada (2014 ECF), and São Tomé and Príncipe (2015 ECF). Grenada and São Tomé and Príncipe did not meet the market access criterion for PRGT graduation and would not have been required to blend under the new proposal.

31

The incremental impact on budgetary interest payments would also be modest: for example, the exclusion from the blending presumption reduced Ghana’s interest bill in 2018 by about US$14 million, a saving of some 0.03 percent of GDP.

32

For example, Ghana accessed sovereign bond markets (US$ 3.75 billion during 2015–18) while at high risk of debt distress.

33

The existence of substantial amounts of commercial debt that is collateralized could undermine this argument— underscoring the importance of a careful assessment of the quality of public debt data in making a determination on blending.

34

See also IMF (2018b), paras. 40–42 for background discussion.

35

The Executive Board decision on this issue will be taken in the context of the Executive Board discussion on PRGT— Review of Interest Rate Structure.

36

See also IMF (2018b), paras. 66–74 for background discussion.

37

Establishing a shorter-term version of the ECF would have required a more fundamental change in the structure of the LIC facilities architecture and support from all PRGT contributors.

38

Access under the “regular window” of the RCF refers to RCF access for urgent BOP needs that is not provided under the exogenous shocks nor under the large natural disaster windows of the RCF.

39

The current limit of not more than two RCF disbursements in any 12-month period would also continue to apply.

40

Thus, the proposed reform would provide much greater flexibility in terms of the scale of annual lending under the regular window of the RCF; but cumulative access to RCF resources via the regular window would increase only in line with the generalized increase in access levels.

41

See Section C for discussion of the cumulative access limit available under the large natural disasters window of the RCF.

42

The 12-month statement of policies and measures is a rolling concept, such that at each review the member will present the policies that it will follow during the next 12 months.

43

“In helping members to devise economic and financial programs, the Fund will pay due regard to the domestic social and political objectives, the economic priorities, and the circumstances of members, including the causes of their balance of payments problems and their administrative capacity to implement reforms.” Guidelines on Conditionality (IMF, 2002), paragraph 4.

44

The Board has endorsed the objective of greater flexibility in program design for countries in fragile situations (see IMF (2011): The Chairman’s Summing Up—Macroeconomic and Operational Challenges in Countries in Fragile Situations, July 12, 2011).

45

See also IMF (2018b), paras. 58–59 for background discussion.

46

Given a generalized access increase of one-third, the cumulative access limit on RCF access under the regular and exogenous shocks windows would increase to 100 percent of quota. Countries seeking a disbursement under the large natural disasters window would be subject to a higher cumulative limit of 133.3 percent of quota; outstanding disbursements from all RCF (and RFI) windows would be included in calculating cumulative access.

47

While there should be no presumption of alignment of access limits under the PRGT and GRA, both RCF and RFI access limits were simultaneously increased by 50 percent in 2015; and annual limits under the large natural disasters window were increased for both in 2017. Staff guidance on the use of RCF and RFI financing will be updated, including on blending.

48

There have been 26 cases of RCF disbursements since inception of the facility in 2010, and 4 cases of disbursements under the RFI since it was approved in 2011. Two of the RFI cases were blended with RCF resource (St. Vincent and the Grenadines and Vanuatu) and two were used for non-PRGT-eligible members (Iraq and Ecuador).

49

See also IMF (2018b), paras. 45–49 for background discussion.

50

The current restrictions on use of the SCF for more than 2½ years out of a five-year period are proposed to be modified to allow for use in three years out of a six-year period, to accommodate SCF arrangements of up to three years. This limitation does not apply to precautionary use.

51

See also IMF (2018b), para. 75 for background discussion.

52

Under the current policy, an ECF arrangement can be extended up to a maximum of five years but the duration at approval must be three to four years.

53

See also IMF (2018b), para. 77 for background discussion.

54

The functions and required content of the EDD are discussed in the 2015 Board paper on Reform of the Fund’s Policy on Poverty Reduction Strategies in Fund Engagement with Low-Income Countries—Proposals.

55

The proposed modifications do not apply to PRS requirements under the HIPC Initiative. Countries under the HIPC Initiative would continue to be subject to the current PRS documentation requirements for purposes of reaching decision and completion points.

56

When the EDD (PRGS) requirements are met by the second review, the good practice on PRS implementation review (PIR) would be for it to take place by the fifth review. In the event of extensions, PIR is not an expected practice.

57

Request for extensions beyond the second review will not be permitted under SCF arrangements or under the PSI, given that countries with limited capacity to prepare an EDD (PRGS) would not normally be expected to request support under the SCF or PSI.

58

The 2009 Board papers that established the current LIC facilities architecture linked the market access criterion for qualification for EA under the PRGT to the market access criterion for blending, which did not originally contain the GNI requirement. However, a follow-up paper that proposed the addition of a minimum GNI per capita above 80 percent of IDA’s operational cutoff as a requirement to satisfy the market access criterion for blending did not propose a corresponding modification to the criteria for EA under the PRGT (See Eligibility to Use the Fund’s Facilities for Concessional Financing, IMF (2009c), paragraph. 13).

59

Currently only four countries with GNI per capita below 80 percent of the IDA threshold have market access, where market access is as defined in determining whether a country is expected to blend.

60

For details, see “Review of the Financing of the Fund’s Concessional Assistance and Debt Relief to Low-Income Countries.”

1

The program targeted to deliver a strong upfront fiscal adjustment-while accommodating security and investment needs—but also recognized the desirability of a smoother adjustment path. This was accommodated by built-in flexibility to relax the fiscal target by almost 1 percent of GDP, should additional resources become available.

2

Nevertheless, Chad reached the HIPC completion point in April 2015.

3

Total pledges at an earlier 2006 Consultative Group meeting were US$5 billion, which were to be disbursed over three years, but by 2010 total disbursements have been about US$400 million.

Other Resources Citing This Publication