Policy Support and Signaling in Low-Income Countries

This paper explores how the Fund's instruments and practices might be adapted to support sound policies in low-income members, in particular those that do not have a need or want to use Fund resources.

Abstract

This paper explores how the Fund's instruments and practices might be adapted to support sound policies in low-income members, in particular those that do not have a need or want to use Fund resources.

Executive Summary

Fund activities relating to low-income members are centered on supporting their efforts to secure and maintain macroeconomic stability, growth, and poverty reduction—consistent with their efforts to achieve the Millennium Development Goals. To ensure that Fund support for low-income members remains as effective as possible, it must adapt as members’ needs evolve. This paper explores how the Fund’s instruments and practices might be adapted to support sound policies in low-income members, in particular those that do not have a need or want to use Fund resources.

Staff undertook a survey of PRGF-eligible members and of donors seeking their views on how the mechanisms of Fund support for low-income countries could be made more helpful. Overall, the survey identified a need for: (i) the development of a new instrument designed for members that do not need, or want, Fund financial assistance; and (ii) filling information gaps, especially in program cases, through more frequent and regular detailed assessment letters.

A number of low-income countries may not need Fund financial support but still want the Fund to support their program and endorse the quality of their policies. This paper proposes the creation of a Policy Support Instrument (PSI) for this purpose. The new instrument would: (i) be based on a member’s poverty reduction strategy, ensuring ownership; (ii) consist of a policy framework normally focused on consolidating macroeconomic stability and debt sustainability, while deepening structural reforms in key areas that constrain growth; and (iii) provide the basis for rapid access to concessional Fund resources in the event of shocks. Use of the new instrument would be approved by the Board, and hence provide a visible signal of Fund endorsement of a member’s policies.

For low-income members with a PRGF arrangement or a PSI, delays in completing reviews create a “gray period” during which information on performance is not readily available. Assessment letters should be used more frequently to fill this information gap, reducing the risk of an unwarranted interruption of donor support. The content of these assessment letters should be more detailed than is current practice.

While the precise staff resource cost will vary from country to country, the recommendations in this paper are likely to imply some additional resource costs in the near future, stemming primarily from the increased use of assessment letters, rather than the creation of the PSI. However, the PSI may require additional staff resources in the medium term.

I. Introduction

1. The Fund supports its low-income members directly through policy advice, capacity building, and financial assistance.1 To ensure that Fund support for low- income members remains as effective as possible, it must adapt as countries’ needs evolve. In particular, as more low-income members have experienced a marked strengthening in their macroeconomic situation, the perception of a need for a new, non-financial, form of Fund engagement has grown. These countries still often want to maintain close engagement with the Fund and signal the Fund’s endorsement of their policies.

2. To assess members’ changing needs, staff conducted a survey of PRGF-eligible members and of donors.2 These results were complemented by information obtained in the context of staff’s regular bilateral contacts with members. Overall, consistent responses were received in a number of areas that have been considered in previous Board papers. In particular, a need was identified for: (i) the development of a new instrument designed for members that do not need, or want, Fund financial assistance; and (ii) filling information gaps, especially in program cases.

3. This paper makes specific proposals about how the Fund’s instruments might be adapted to support sound policies, particularly in those low-income members that do not need or want to use Fund resources. In line with the mandate given by the IMFC for “further work on a policy monitoring arrangement to enhance the IMF’s signaling role for countries that do not need or want IMF financing,”3 this paper proposes to establish a new instrument, a Policy Support Instrument (PSI),4 that will allow the Fund to support low-income members in designing and pursuing policies that meet high standards, and to signal the strength of these policies. It also proposes some modifications of signaling procedures for members with traditional Fund arrangements in response to views expressed by low-income members and donors.

4. The paper is organized as follows: Section II discusses the need for a new instrument, and makes specific proposals for the main features of the PSI; Section III examines the scope for filling information gaps in program cases through a more active use of assessment letters; Section IV deals with resource implications, and Section V with next steps; finally, Section VI suggests issues for Board discussion.

II. Fund Support of Low-income Countries Which do not Need, or Want, its Financial Assistance

A. General Considerations

5. As their circumstances change, a number of low-income members may not need or wish for Fund financial assistance but still want the Fund to support their programs, providing explicit assessments to themselves, their donors, and creditors (see survey results).5 These countries, often called “mature stabilizers,” have usually achieved a reasonable growth performance, low underlying inflation (comparable to that obtaining in their major trading partners), an adequate level of official international reserves, and have begun to establish (external and net domestic) debt sustainability. Some of these members no longer need Fund financial assistance and have indicated their desire to “graduate from the PRGF,” while wishing to maintain a close relationship with the Fund, and endorsement of their policies and performance.6 This group seems to be limited at present to a handful of countries in Africa and Asia, but is expected to grow over time. In addition, some PRGF-eligible members that have never had a PRGF arrangement may see benefit in a close relationship with the Fund entailing endorsement of their policies. While Article IV consultations provide policy advice, support, and an assessment, these may not be frequent or specific enough, raising the question of whether a special instrument for policy support and signaling is needed.

6. A new policy support instrument for low-income members requires serious consideration for other reasons as well. Even if low-income members with an improved external position have less demand for Fund financial assistance, most of them still continue to rely heavily on donor assistance to meet their development needs, and they would welcome a new mechanism to provide signals of strong policies outside a traditional Fund arrangement. Moreover, no matter the degree of stabilization attained, low-income members are vulnerable to exogenous shocks, and might want relatively quick access to concessional financing from the Fund in the event that a sudden balance of payments need were to arise. But in the absence of such a need, if the Fund can send clear signals outside a financial arrangement, this would help preserve PRGF resources.

7. The issue of whether the Fund needs a new instrument allowing closer engagement than under Article IV consultations while sending clear signals on the strength of a member’s policies has been discussed previously by the Board. Past attempts to develop such instruments have not stood the test of time or have failed to gain Board support.7 The features of a Policy Monitoring Arrangement (PMA) were outlined in September 2004.8 In the discussion of the PMA, Executive Directors stressed the challenge of defining a minimum standard for policies to be endorsed, and many Directors raised concerns that there might be insufficient demand given the PMA’s relatively stringent conditions (including the application of the standard of upper credit tranche conditionality), as well as the fixed cycle proposed for assessments.

8. These drawbacks need to be taken seriously, but they are not insurmountable. Indeed, it has been suggested that the attraction of a new instrument specifically aimed at low-income members could be enhanced by: (i) emphasizing country ownership; (ii) focusing the structural reform component; (iii) streamlining modalities and documentation requirements; and (iv) allowing a quick transition to concessional financing in the event of a balance of payments shock.

B. A New Instrument: The Policy Support Instrument (PSI)

9. The PSI would be a non-financial instrument for low-income members seeking Fund support of their policy framework and an endorsement of its quality. It would be used alongside existing instruments and be available to meet the needs of members seeking to graduate from PRGF arrangements or members without a PRGF arrangement seeking Fund endorsement of their policies. While not requiring Fund financial assistance, these members continue to look to the Fund for support in designing strong macroeconomic policy frameworks, sound institutions, and good governance structures. The PSI would be designed to: (i) promote a close policy dialogue between the Fund and a member; (ii) provide more frequent Fund assessments of a member’s economic and financial policies; and (iii) deliver clear signals that could be taken into account by donors, creditors, and the general public on the strength of these policies.

10. The following features would be desirable to ensure the effectiveness of the new PSI: 9

  • Program content should be based on the member’s poverty reduction strategy to ensure ownership. A monitorable program with a well-articulated quantitative framework would be developed. The framework would meet a common high standard, equivalent to that of upper credit tranche conditionality. Documentation would be focused and PRSP-related procedural requirements would be streamlined compared to the current PRGF arrangements.

  • The policy framework would normally focus on consolidating macroeconomic and debt sustainability, while deepening structural reforms in key areas that are constraining growth. The macroeconomic framework should be designed to provide stability as well as an enabling environment for growth including that which might be supported by a scaling-up of external assistance. On structural reforms, the key areas would include financial sector reform, trade reform, transparency, and strengthening institutions, especially in public finance. As is the case for PRGF arrangements, the focus on growth-enhancing reforms would require very close Bank-Fund cooperation.

  • Rapid access to Fund resources would be available in the event of shocks. Accelerated access would be possible to a new stand-by like window on concessional terms within the PRGF, on the basis of the strong policy framework established under the PSI.10

11. The PSI would be endorsed by the Board, and thus it would provide credible signals by the Fund of strong country policies to donors, creditors, and the private sector (both external and domestic). Together with existing instruments such as assessment letters, a PSI could play a critical role in strengthening coordination with donors, increasing the predictability of aid flows, and could also provide a suitable framework within which to scale up assistance.

Eligibility11

12. The PSI would be available to all PRGF-eligible Fund members with a poverty reduction strategy in place and that had achieved macroeconomic stability and generally did not need Fund financial support.12,13 An eligible member would be seeking to sustain macroeconomic stability and debt sustainability and achieve an ambitious growth objective in the medium term in order to achieve its development goals, particularly regarding poverty reduction, and facilitate absorption of higher aid flows. Members that have never had a PRGF arrangement, but which are implementing a participatory poverty reduction strategy clearly laid out in a public official document, would also be eligible. For all countries seeking a PSI, their poverty reduction strategy documents would be issued to the Executive Board, together with an advisory note by the staffs of the Fund and the Bank.

Program Standard

13. For effective Fund endorsement of policies, the PSI needs to establish a standard, with monitoring focusing on the implementation of an agreed program approved by the Board.14 The issue is which standard. In light of past experience, it appears best to introduce a new instrument with the standard of upper credit tranche conditionality. This standard is so well known that any attempt to define either a lower or higher standard would be confusing. Other standards tend to be misinterpreted as equivalent—as happened with enhanced surveillance and with the signaling staff-monitored programs, discontinued in 2003—or be misunderstood, a problem with the now expired Contingent Credit Line.15

14. The standard of upper credit tranche conditionality should apply to PSIs.16 In the context of Fund financial support, this standard aims at ensuring that: (i) the likelihood the Fund gets repaid is high, allowing it to meet its fiduciary responsibilities; (ii) the Fund gets repaid relatively quickly, consistent with the revolving nature of its resources; and (iii) policies will allow the member to solve its balance of payments difficulties in a manner consistent with the Articles of Agreement. In the context of the PSI, this standard can be interpreted as signifying policies strong enough to constitute the basis for a Fund-supported program should a user request Fund financial assistance in the event of a shock.17

15. This standard, as applied by the Fund, implies that the member’s macroeconomic position is sustainable or that policies will achieve sustainability in the near future. While an alternative standard of “sustainability” may be relevant in the context of surveillance, Fund endorsement requires more than this. Policies meeting a sustainability test—based on the debt sustainability analysis that the Fund carries out—might stabilize the situation but in ways that could distort resource allocation or be unneighborly. A “sustainability” benchmark would: (i) have to be defined more precisely; (ii) risk becoming diluted; and (iii) risk being perceived as a signal of less strong performance.

16. A PSI, by utilizing the upper credit tranche conditionality standard—a well-known standard—would signal clearly Fund endorsement of a member’s policies. The proposed features of the PSI would also address many of the difficulties faced in past attempts at designing a successful signaling instrument. In particular, Board endorsement, the provision of on/off signals, and the presumption of a quick access to Fund financial resources through a new window in the PRGF Trust in the event of a balance of payments need should: (i) reduce the risk of a dilution of standards; and (ii) enhance the clarity and credibility of the signal, given that Fund resources would be potentially at stake.

17. There are other vehicles for providing Fund support of policies and close 18 engagement with members that are not seeking Fund endorsement of their policies. Recently, the label “intensified surveillance” (IS) has been used to characterize the staff’s close monitoring of the authorities’ program at higher-than-annual frequency in a few member countries, with a particularly close policy dialogue with the Fund staff, bearing in mind that the authorities’ program may not be the appropriate benchmark for assessing performance. Such monitoring does not involve Fund or staff endorsement of the authorities’ program; it simply allows the more frequent assessment of economic developments and policies. This closer policy dialogue can be mistaken by donors and markets for endorsement because: (i) such schemes resemble traditional Fund arrangements (i.e., quarterly targets, frequent staff missions); (ii) a “label” (e.g., IS) suggests a special relationship; and (iii) the countries themselves may foster this perception. Candid assessments could overcome many of these problems. Actually, a special procedure is not necessarily needed for close monitoring. For some years, staff has monitored and assessed economic developments in Lebanon at high frequency—reporting to the Board through semi-annual reports—without using a particular label or procedure, and quarterly reports have been used to inform the donor community of macroeconomic developments and policies in the West Bank and Gaza. Such high-frequency monitoring and assessing without a specific label could meet the needs of countries that want to maintain a close dialogue with the Fund, but would not require Fund endorsement of their policies.

Policy Content

18. The policy package supported by the PSI would be geared toward sustaining macroeconomic stability and debt sustainability and achieving an ambitious growth objective to reduce poverty. It should also facilitate the member's eventual access to international capital markets. In keeping with the upper credit tranche conditionality standard, it would include the following:

  • A macroeconomic policy framework that sustains macroeconomic and financial stability as well as enables accelerated growth, with sustainable domestic and external debt dynamics. PSI performance targets would need to focus on measures that would inter alia help countries deal with the macroeconomic challenge of the scaling up of external assistance. There would also be a fully articulated quantitative framework. Performance targets will need to protect as well medium-term debt sustainability, both external and domestic.

  • Key structural measures should be included if, and only if, critical to meet the objectives of the program.19 Parsimony and selectivity would also be key features. Since mature stabilizers would be the most likely users, critical measures are likely to focus on second-generation reforms to enhance the growth potential and the capacity of the economy to absorb higher aid. Performance targets for structural policies would generally focus on the areas of Fund expertise, in particular: (i) banking sector reform; (ii) macroeconomic aspects of trade reform; (iii) public expenditure management (to help ensure that aid is used effectively for achieving the MDGs); (iv) revenue mobilization; (v) foreign exchange and liquidity management (to help manage the large inflows from donors); (vi) public debt management (both external and domestic); (vii) institutional reform in areas of Fund expertise (central bank independence, tax and customs agencies, expenditure agencies and systems, statistics offices, etc.); and (viii) transparency (to facilitate good economic governance and eventual access to capital markets).

Modalities

19. The PSI would provide a clear “on/off” signal as well as a multidimensional assessment.20 It would indicate that, in the opinion of the Fund’s Executive Board, the member’s policies meet the standard of upper credit tranche conditionality, with clear quantitative targets, avoiding the risk of misinterpretation of signals and standards. It would be based on close policy dialogue with the member, meeting its need for policy support. The related documentation would also provide detailed information on macroeconomic and sectoral developments and policies and would, therefore, address the needs of the donor community. Fund policies related to financial arrangements would pertain where applicable.

20. Duration. The period covered by a PSI would be 1-3 years. Given the likely users and the desirability of a medium-term framework for donor assistance, requests are more likely to be for 2- or 3-year PSIs. The duration of a PSI could be extended up to a maximum of 4 years. Members could request a successor PSI.

21. Activation. Following an expression of interest in a PSI, the member and Fund staff would conduct discussions on the member’s chosen economic and financial policies. The member would request that its policies be monitored under the PSI, and send the Managing Director a Memorandum of Economic and Financial Policies (MEFP) setting out the authorities’ commitments in support of this request. Based on the Managing Director’s recommendation, the Executive Board would consider approval of the PSI, provided that the members’ policies, as described in the MEFP, met the upper credit tranche conditionality standard. In addition, to ensure that this standard is met and comparability of treatment between PSIs and financial arrangements for low-income countries: (i) the Fund's policy on safeguards assessments would apply to the PSI; (ii) the program would have to be fully financed; and (iii) the treatment of obligations to creditors would be identical to that required in Fund financial arrangements. This comparable treatment would also help allow accelerated access to the shocks window should such a need arise.

22. Reviews. Reviews by the Board could take place semi-annually, with the same degree of flexibility as PRGF reviews. A Board meeting for the annual review might take place in conjunction with the finalization or approval of the next year’s budget and the Article IV consultation, should one be scheduled during the year.21 The mid-year review, with a streamlined staff report, would be expected to be concluded on a lapse of time basis if developments were proceeding as envisaged. Reviews would provide the means to assess performance against the authorities’ commitments and, in cases of changed conditions or prospects, to offer recommendations on appropriate policy steps and waivers, if needed.22 If needed, specific measures could be required up front for a review to be completed. Typically there would be two full missions a year, with additional staff visits if needed to maintain the momentum of the policy dialogue.

23. The discussions and reviews of the PSI should be designed to fit, to the extent possible, with country-led donor harmonization efforts. A link will remain with donor support geared to the objectives of the PRSP. Missions and reviews should therefore be scheduled to fit into the cycle of medium-term expenditure framework or budget discussions, and resident representatives should be involved at early stages.

24. Documentation. MEFPs would be concise, and fuller use could be made of the authorities’ annual progress report on the implementation of their poverty reduction strategy. For Board meetings on activation of a PSI, an MEFP and staff report, which would contain an appraisal of the member’s policies, would be provided. The MEFP would cover the member’s current economic situation and outlook; its medium-term policy framework in the macroeconomic, structural, and institutional areas; and its specific policy objectives for the period of the instrument. The description of the policy objectives would include a quantified annual macroeconomic framework for the next 12 months (including, when warranted, a medium-term budgetary framework), with quarterly or semi-annual quantitative targets and, as needed, benchmarks on structural and institutional policies. Documentation for Board reviews would consist of a staff report and, at the member’s discretion, a supplement to its MEFP. Board meetings on activation and reviews would result in a Chairman’s statement. A poverty reduction strategy would need to be in place.

25. Publication of the member’s MEFP, the staff report, and a press release containing the Board’s assessment, should be voluntary but presumed.

26. The question arises whether PSIs should count towards longer-term program engagement (LTPE). Revision of the LTPE policy to include PSIs may be justified, since they could be seen as a form of “program tutelage” involving policies that meet upper credit tranche conditionality standards. However, PSIs have a separate and important policy endorsement and signaling aspect and they do not entail any availability of Fund resources (unlike the precautionary UFR arrangements currently covered by the LTPE policy). On balance, staff does not recommend that PSIs be included in the LTPE policy, but the Board may wish to consider whether such inclusion is warranted.23

C. Relation of the PSI with Fund Financial Arrangements

The PSI and the PRGF “shocks window”

27. Low-income countries are particularly prone and vulnerable to shocks, which include natural disasters, commodity price changes, and conflicts and crises in neighboring countries. Even those opting for the PSI could face the need for Fund financing at some point. Therefore, the PSI would be more attractive and useful to low-income members if it were to facilitate access to concessional Fund resources that helped them deal with shocks when necessary.

28. This latter feature could be achieved through the creation of a second window in the PRGF Trust instrument to address shock-related financing needs, which seems a natural complement to the PSI. In the event of a shock, this window would allow for quick-disbursing financing on concessional terms for PRGF-eligible members whose PSI was on track.24 The economic program to be supported by the “shocks window” would be directly drawn from the program endorsed under the PSI, with additional adjustment limited to that necessary to correct the immediate balance of payments need arising from the exogenous shock.25 While access to the “shocks window” for members with an on-track PSI would not be automatic, faster access would be possible, given that the PSI would already be supporting policies that met upper credit tranche conditionality standards before the shock. The features of the proposed new “shocks window” are presented in a companion staff report.

The PSI and the PRGF

29. The creation of the PSI should not have negative spillover effects on PRGF arrangements. The PSI would most likely be requested by members that currently have low-access PRGF arrangements. There is a risk that members continuing to use such PRGF arrangements might be perceived as having weaker policies or more serious economic problems than those of members using the PSI. To counter this perception, it should be made clear that the diversity of members’ circumstances would determine the choice between use of the PSI and low-access PRGF arrangements. Among the relevant circumstances would be the overall balance of payments financing need, the frequency of exogenous shocks, and the member’s desire to flag its “graduation from the PRGF.”

30. The PSI could fulfill the role that some had envisaged for a precautionary PRGF arrangement. A number of low-income members and donors in recent Board meetings and in the survey considered that precautionary PRGF arrangements would have a positive signaling value, seeing benefits in a label associated with successful graduation from Fund arrangements. Precautionary PRGF arrangements, however, have been explored previously and their creation was not endorsed by the Board. Several drawbacks were cited with regard to creating a formal precautionary PRGF arrangement: they would not provide a clear break from traditional Fund support, would conflict with Board’s views that precautionary extended arrangements should be rare, and would require the commitment of some scarce PRGF resources.

31. The PSI and low-access PRGF together offer members sufficient instrument flexibility. The PSI would signal graduation from the PRGF and would not encumber scarce PRGF resources. Combined with genuinely quick access to the shocks facility in the event of a shock, the PSI could provide most of the advantages of a precautionary PRGF arrangement without its main drawbacks. It would still remain legally possible, however, in cases where a Fund arrangement was sought or in place, for the member to decide unilaterally not to draw (i.e., to treat its PRGF arrangement as precautionary).26 Some low-income members that are prone to shocks may be unwilling to do without the safety net of a PRGF arrangement and its possible augmentation of access. These members’ needs could be met through a low-access PRGF arrangement, while low-income members that feel ready for graduation would likely opt for a PSI.

III. Filling Information Gaps: The Case of Low-income Countries with a PRGF Arrangement or a PSI

32. Donors often utilize the information provided in the context of a Fund-supported program in making their aid decisions, and many identified the information gap that can arise in the event that program approval or reviews are delayed (see Annex I). In some cases, this “gray” period can last several months, between the time when a review was expected to be completed, as indicated in previous reviews or the original arrangement, and the time when it becomes clear to interested parties that a program is on/off-track. Delays can be caused by a variety of problems for which the appropriate donor response may differ. In some cases, donors would consider it appropriate to maintain aid disbursements, but donors may not have all the information necessary to make that decision. Donors identified the lack of information during this gray period as a significant obstacle to the smooth delivery of their assistance, which ultimately can harm low-income members. Maintaining clear, regular, and multidimensional communication with donors in this context would ensure that the Fund does not promote excessive aid volatility (see Box 1).

Fund Signals, Aid Volatility, and Uncertainty

Because of its central role in assessing economic policy frameworks and the importance of sound macroeconomic environments for aid effectiveness, the Fund is often viewed as catalyzing official assistance to LICs. This role is explicit in the context of Paris Club debt treatments, which are generally conditioned on “the existence of an appropriate program supported by the Fund, which demonstrates the need for debt relief’1 and the implementation of reforms to resolve payments difficulties. It is also explicit for debt relief in the context of the enhanced HIPC Initiative, which requires that eligible countries establish a track record of reform and sound policies through Fund- and Bank-supported programs. More generally, the Fund also influences the decisions of official donors and creditors to provide bilateral assistance to LICs, especially through its own financial assistance and assessments of performance under Fund-supported programs. As mentioned in the 2002 IEO report on prolonged use of Fund resources, official donors often view the presence of a Fund arrangement as a clear, one-dimensional, and credible signal that macroeconomic policies are appropriate. With the current shift towards more direct budget support, more sensitive to Fund signals than project aid (see the survey results in Annex I), the overall reliance of donors on Fund signals is likely to increase.

Such a situation has two potential downsides: the risk that changing Fund signals will result in large swings in official financing;2 and, conversely, the risk that the Fund will be less explicit in its signaling because of the potentially large impact of its signals on assessed members. The IEO report cited evidence of the latter risk in its case studies of Pakistan, the Philippines, and Senegal. Malawi also provides a recent illustration of these two risks. Malawi is highly dependent on foreign aid, which represents about 75 percent of central government revenue. Recent analysis has shown aid to this country is very volatile, which raises concerns for macroeconomic stability. The recent ex post assessment of longer-term program engagement in Malawi alludes to negative Fund signals—in the form of delays in completing reviews due to policy slippages—in explaining shortfalls of external budgetary support. Conversely, it also mentions “the risk that, in providing a signaling role to other official donors and creditors, the Fund’s assessment of policies becomes hostage to concerns about the impact and policy response to a disruption of donor support, potentially undermining the credibility and usefulness of that role of the Fund.”

Empirical research has focused on the Fund’s catalytic role in official financing rather than its impact on aid volatility. Qualitative and quantitative studies have found limited evidence that the presence of a Fund arrangement has positive impact on official capital flows.3 Research in this area, however, is hindered by the absence of a counterfactual—what would have happened to official flows in the absence of a Fund arrangement? There are other identification problems that make it difficult to reach firm conclusions. For instance, it is difficult to determine whether official flows are affected by Fund signals or by the underlying macroeconomic situation and policies.

1 See Paris Club website: www.clubdeparis.org.2 The damaging effects of volatile and uncertain aid flows have been widely documented—see, for example, Bulir and Hamann (2003). Bulir and Lane (2004) mention Fund conditionality as a possible source of aid volatility, and discuss policy implications of this volatility, with a focus on fiscal management.3 See, for instance, G. Bird and D. Rowlands (2002 and 2004).

33. In those cases, or in similar circumstances in PSI cases, assessment letters should be used more frequently to fill the information gap, reducing the risk of an unwarranted interruption of donor support.27 In most cases, PRGF program reviews are scheduled every six months (as would be the case for PSIs). If this regular schedule is maintained, and reviews are synchronized with the country’s budget cycle, program or PSI documentation is likely to be sufficient to satisfy donor signaling needs. However, assessment letters could be issued in cases where it became clear that the expected timetable for the completion of a review would be missed. Staff would notify the authorities and the donor community of the likely delay and inquire about the need for an assessment letter. Enquiries could be done informally through resident representatives, and requests (from either the low-income member or its donors) for such a letter would be honored. The Board would continue to be informed of these developments through informal country matter sessions, and to receive copies of any such letters.

34. The content of these assessment letters should be more detailed than is current practice. A number of donors regard assessment letters now issued by Fund staff as insufficiently informative and would value more detailed and quantified assessments. This request could be accommodated easily under the current guidelines for assessment letters,28 which specify that they “should contain a clear and candid assessment of the member’s macroeconomic conditions and prospects, and of macroeconomic and related structural policies. This assessment should enable the recipient to form a clear view of the strengths and weaknesses of a country’s macroeconomic and related structural policies. Specifically, the assessment should (i) identify existing macroeconomic imbalances and structural distortions; (ii) indicate to what extent current and planned policies are dealing with (or, perhaps, contributing to) these imbalances and distortions; and (iii) highlight policy areas where there are major outstanding concerns.” To meet donor expectations, the letter should describe recent economic and structural performance against targets and present macroeconomic projections. The letter should indicate: (i) the reasons for revisions, if any, made to the projections; and (ii) whether the authorities and staff are in agreement with these.29 The letter should also be candid about divergent views on policies. While it should present both views, it should remain descriptive and make clear that the views expressed are not those of the Board. Standard Fund guidelines on confidentiality would continue to be applicable. Consistent with normal requirements, assessment letters will include confidential information only with the prior consent of the member or party that had provided the information to the staff.

IV. Staff Resource Implications

35. While the precise resource impact will vary across countries, the recommendations in this paper are likely to imply some additional resource cost in terms of staff time in the near future. Preliminary information suggests that the PSI would be first requested mostly by mature stabilizers that currently have low-access PRGF arrangements. For those cases, the PSI would imply no additional resource cost. There would be an additional resource cost, however, if a member without a PRGF arrangement, or some form of intensive monitoring, requested a PSI.30 The bulk of any additional resource requirements in the near future are likely to stem from the increased use of assessment letters. For members with a PRGF arrangement or a PSI, these letters would be produced upon request when completion of a review is delayed. Letters in such circumstances would likely utilize resources initially intended for the corresponding review. Therefore, the additional resource cost would be limited. More additional resources would be needed, however, for assessment letters requested for members without a PRGF arrangement or a PSI. Assuming that 15 additional detailed assessment letters were produced a year,31 and that each such assessment letter requires about one-third staff year to be produced and reviewed, the additional annual resources needed would be 5 staff years. Production of some of these letters may also involve additional staff travel, although this should be minimal.

36. In the current context of zero real budget growth over the medium term, consideration could be given to what other activities would have to be reduced, thereby providing offsetting savings to cover additional costs.

37. Alternatively, since PSIs and assessment letters are services to members, the Board might want to consider whether it would be appropriate to charge for such services.32 Without a user fee, the additional resource costs mentioned above would have to be borne by the Fund’s administrative budget. Assuming a desire to cover administrative costs arising from the use of PSIs and assessment letters, various bases for a user fee could be envisaged including a flat fee or cost recovery. Such an approach might be more appropriate should a decision be taken that donors, as the likely end-consumer of these signals, should bear these costs.33 This approach would minimize disincentives that low-income members may face in considering use of the PSI. Alternatively, if it is decided that low-income members should bear these costs directly, then it might be more appropriate to link user fees to members’ capacity to repay, by basing user fees on per capita income. Staff proposes to follow up on this issue, if warranted, in light of Directors’ views.

V. Next Steps

38. A formal decision to establish a new PSI could be proposed for Board adoption together with the other relevant decisions following Board discussion of this and companion papers.

VI. Issues for Discussion

39. The paper outlines a number of possible improvements to current arrangements aimed at providing a menu of options to meet better the varied needs of low-income members and donors. Do Directors agree that:

  • there is a case for a new instrument—the PSI (as outlined in paragraphs 9 to 26 and Box 2)—for low-income members which may not need, or wish to seek, a PRGF arrangement but which still want to signal Fund endorsement of their policies?

  • detailed assessment letters could be used more frequently (as set out in paragraph 33-34) in program and PSI cases where delays in reviews create a “gray period”?

  • an on-track PSI would facilitate accelerated access to Fund concessional resources under the new shocks window in the PRGF?

  • some additional resources, or offsetting savings, would be needed to implement these proposals?

Main Features of the PSI

Purpose. The PSI would be designed (i) to promote a close policy dialogue between the Fund and a member; (ii) to provide more frequent Fund assessments of a member’s economic and financial policies; and (iii) to deliver clear signals to donors, creditors, and the general public on the strength of these policies.

Legal basis. The legal basis for the PSI would be Article V, Section 2(b) of the Fund’s Articles of Agreements.

Eligibility. The PSI would be available to all PRGF-eligible Fund members with a poverty reduction strategy in place, which had achieved macroeconomic stability and no longer needed Fund financial support. The Fund's policy on safeguards assessments would apply to the PSI. The program would have to be fully financed. The treatment of obligations to creditors would be identical to that required in Fund financial arrangements.

Activation would:

  • require policies that meet the standard of upper credit tranche conditionality;

  • involve Executive Board endorsement of these policies; and

  • be for a period of 1 to 3 years.

Reviews. Use of the PSI would entail Board reviews according to the following modalities:

  • There would be a review to evaluate performance every six months. A full Board meeting would generally take place once a year, possibly in combination with the Article IV consultation, while the mid-year review would be expected to be concluded on a lapse-of-time basis.

  • Reviews would provide the means to assess the implementation of the member’s economic and financial policies against the initial MEFP (and, for subsequent reviews, the conclusions of the previous reviews and any supplement to the MEFP) and, in cases of changed conditions or prospects, to offer recommendations on appropriate policy steps.

Documentation should be streamlined. For Board meetings on activation of the PSI there will be an MEFP setting out the authorities’ commitments, and staff report which would contain an appraisal of the member’s policies. Documentation for Board reviews would consist of a staff report and, if necessary, a supplement to the member’s MEFP. Board meetings on activation and reviews would result in a Chairman’s statement.

Publication. Publication of the member’s MEFP, the staff report, and a press release containing the Board’s assessment would be voluntary, but presumed.

What are Directors’ views on:

  • whether charging should be considered for these services to members?

  • whether PSIs should count towards determining longer-term program engagement?

Survey Results and Methodology

Introduction

1. Fund activities in its low-income members34 are centered on securing and maintaining macroeconomic stability, growth, and poverty reduction, to support their efforts to achieve the goals established in their poverty reduction strategies. But in this process the Fund conveys information and judgments to outside audiences. The signals sent by the Fund can range from simple descriptions of economic developments and policies (in particular, for countries where information is otherwise not readily available) to more elaborate assessments of economic performance, the need for financing, and capacity to implement policies.

2. Fund signals take many forms and have multiple audiences. In part, this diversity reflects the fact that signals tend to be a by-product of activities designed to support members’ needs directly. These signals are sent by the Executive Board, management, or staff, and they are conveyed in various formats and through various media, including: staff reports produced in the context of surveillance activities or financial arrangements, and the related public information notices and press releases; technical assistance reports; joint staff advisory notes on PRSs and their implementation; assessment letters or statements produced at the request of donors; missions’ concluding statements; informal briefings by missions or resident representatives in the field; and management activities. The diversity of formats also reflects the diversity of possible recipients. The primary recipients are the authorities of low-income members and the donors providing them support, who are interested in distinct, but often complementary, information. The former are interested in signals that allow them to demonstrate the strength of their policies and reform efforts—including to domestic audiences—while the latter tend to want information regarding the strength of the macroeconomic environment, the extent to which structural problems are being addressed, the quality of public expenditure and financial management systems, and the implementation of the poverty reduction strategy. For much official assistance to low-income countries, the Fund plays a central role in assessing economic policy frameworks, given the importance of sound macroeconomic environments for aid effectiveness. This role could increase with the growing shift of donor assistance towards direct budget support, which tends to be more sensitive to Fund signals (see the survey results below).35 Beyond the official sector, signals are also received by the broader public and financial markets.

3. This annex outlines the results, and methodology, of a survey of members’, aimed at assessing whether they are satisfied with the current framework, and how to improve it.36 These results were complemented by information obtained in the context of staff’s regular bilateral contacts with members.

Results

4. Most low-income members seem satisfied with the current provision of signals by the Fund in the context of relations with their donors, and they do not seem to attribute aid volatility directly to these signals (Figure 3). Most respondents indicated that information is generally provided at appropriate times and in the right forms to support their activities. They mentioned as being particularly useful end-of-mission statements and published papers such as staff reports (Figure 1). They considered that the Fund should assess key structural reforms as well as macroeconomic frameworks (Question B10, Table 1). In this regard, they viewed as important a quantified medium-term macroeconomic framework, as well as specific quarterly or semi-annual targets for key variables. Low-income respondents were divided equally regarding the desirability of on/off signals over more multidimensional assessments. Those preferring on/off signals overwhelmingly saw a need for endorsement by the Executive Board (Question B9, Table 1). About half of low-income respondents saw a gap in the information/signals provided by the Fund (Questions B2 and B14-16, Table 1).

Figure 3.
Figure 3.

What Signals Should the Fund Ideally Send and in What Manner and Frequency?

Citation: Policy Papers 2005, 080; 10.5089/9781498331487.007.A001

Source: Fund staff calculations.
Figure 1.
Figure 1.

Types of IMF Signaling Mechanisms Used, Their Importance in Assistance Decisions and Types of Aid Affected

What IMF mechanisms are useful for signaling?

Citation: Policy Papers 2005, 080; 10.5089/9781498331487.007.A001

Source: Fund staff calculations.
Table 1.

Summary of Responses to Yes/No Questions

(SDR billions)

article image
article image

There were 32 responses to the donors’ questionnaire and 31 responses to the low-income members’ questionnaire. Response rate is defined as the number of answers to a given question in percent of returned questionnaires.

Results are significantly different for these questions if donors are weighted by the amounts of aid they provide. This exercise was done for bilateral donors, for which data are readily available. The results are that about 90 percent of bilateral donors see a signaling gap in Fund information on LICs, about 60 percent consider this gap could be filled by modifying current instruments, while about 40 percent see a need for new a instrument. In addition to the answers to the survey reflected in this table, a number of donors indicated a need for innovation in the Fund instruments and that the case for a new instrument was open.

5. A number of low-income members that do not need Fund financing, but would like the Fund to send signals on the strength of their policies, have expressed interest in a new, non-borrowing instrument.37 They consider that a non-financial relationship with the Fund could demonstrate country ownership of policies, help maintain substantive policy dialogue with the Fund, and send credible signals to the donor community. They also believed that such arrangements should be endorsed by the Board and have a policy commitment content that is comparable across countries.

6. The survey confirmed the common view that Fund signals, from both the Board and the staff, are used extensively by donors. Fund signals, particularly assessments of the macroeconomic situation, policies, and related structural reforms, are in general viewed as a very important source of information in donors’ decisions on providing financial assistance (Figure 1).38 They also constitute one of the main sources of information on low-income countries for almost all donor respondents (Question A1, Table 1).39 Respondents noted that Fund signals are conveyed mostly through Article IV consultation and program documentation, but also indicated that they rely on other modalities such as assessment letters and informal briefings from Fund missions and resident representatives (Figure 1). The use of Fund signals in aid decisions is generally not a formal requirement; rather it is often a matter of well-established policy (Question A2, Table 1).

7. Fund signals have a significant—but in general not mechanical—influence on direct budget (or program) assistance. According to the survey, budget support is more sensitive to Fund signals than project aid (Figure 1).40 In a majority of cases, however, donors use Fund signals to help them form a judgment on the overall macroeconomic situation. The way donors use Fund signals in making such judgments varies (Figure 1),41 and aid can be affected at the planning, allocation, or disbursement phase. In countries where a Fund-supported program goes off track, the decision to suspend budget support often depends on the nature of the problems. In particular, many donors that would suspend their budgetary support in the event of weakening macroeconomic performance would be inclined to reconsider it, but not necessarily stop it, if the problems were occasioned by specific structural issues or external shocks (Figure 2). Also, if an arrangement expired with no successor, and no need for one, about 60 percent of donors would continue budgetary support.

Figure 2.
Figure 2.

Impact of IMF Signals on Continuation of Donors’ Budget Support

(in percent)

Citation: Policy Papers 2005, 080; 10.5089/9781498331487.007.A001

Source: Fund staff calculations.

8. Around 40 percent of donors expressed a need for on/off signals, and a majority for multidimensional assessments. According to the survey, the Fund is expected to assess, first and foremost, macroeconomic performance and policies. Like low-income members, donors consider a quantified medium-term macroeconomic framework—with quarterly or semi-annual targets—to be essential for the assessment of policies and progress made. Most also expect the Fund to assess structural reforms that are either macro- economically critical, or within the Fund’s core areas (e.g., tax system, exchange system, financial sector). Some major donors view the Fund’s involvement in the structural agenda as very important to help them identify technical assistance needs and coordinate delivery. Given the range of issues and sectors they are interested in, about 60 percent of donors indicated a preference for multidimensional signals over just on/off ones (Question B4, Table 1), which is consistent with their intensive use of surveillance and program documentation.

9. Both low-income members and donors view timeliness and regularity of signals as key. In program cases, they pointed to an information gap in the event of delayed review completion or program approval. In some cases, this “gray” period can last several months, between the time when a review was expected to be completed, as indicated in previous reviews or the original arrangement, and the time when it becomes clear to donors that a program is on/off-track. Donors identified the lack of information during this period as a significant obstacle to the smooth delivery of their assistance. Delays can be caused by a variety of problems which, as mentioned above, may trigger different responses from donors. In some cases, donor assistance might be unaffected, but donors may not have all the information necessary to make that decision. Donors therefore stressed the importance of regular information flows and of timely signals synchronized with the budget cycle and donor operations (Figure 3).

10. Overall, the survey provided consistent responses in a number of areas that have been considered in previous Board papers. In particular, there is a need for: (i) the development of a new instrument designed for members that do not need, or want, Fund financial assistance; and (ii) filling information gaps, especially in program cases, through more frequent and regular detailed assessment letters. According to survey respondents, low-income members and donors value multidimensional assessments, even when on/off signals are sent. So provided multidimensional assessments are given, the concern that the contribution of Fund signals to aid volatility will increase in the future seems unwarranted. The next two sections outline possible improvements to current arrangements, drawing on the results of the survey, contacts with members and donors, and previous Board discussions.42

Methodology

11. In December 2004, questionnaires were sent to 78 low-income countries and 50 donors and creditors through the offices of IMF Executive Directors. While the questionnaires for donors and creditors and low-income countries differ, they both address three groups of questions, dealing with (i) the current provision and use of IMF signals; (ii) the desirable features of IMF signals; and (iii) donor coordination issues. The questions can also be classified by types of answers they require: (i) yes/no questions; (ii) quantitative questions (participants were asked to select from a range of values); (iii) multiple choice questions (in some of these, participants were asked to select only one of several options; in other questions participants were asked to mark all relevant choices); and (iv) textual responses, used mainly for clarification purposes.

Biases

12. Sampling: All (78) PRGF-eligible countries, and large donors and creditors (multilaterals and bilaterals, and for the latter both Paris Club and non-Paris Club) active in low-income countries were included in the survey; that is, a non-probabilistic method of sampling was chosen. Since PRGF eligibility is based on income level and the list of donors was extensive, these criteria have minimized the sampling bias.

13. Self-selection: All major donors have responded to the survey (see the attached list). In the case of low-income countries, however, the lower response rate raises the issue of a self-selection bias. Specifically, 31 (or 40 percent of) 43 low-income countries responded to the questionnaire, a response rate comparable to that of other similar surveys. Since the views of members that have not responded may differ from those received,44 the scope for reliably generalizing results to all PRGF-eligible countries may be reduced.

14. Anonymity: Given the lack of complete anonymity, biases in members’ responses cannot be entirely ruled out,45 but the impact is likely to be relatively minor, especially since members were guaranteed that their answers would be treated confidentially.

Survey Respondents

article image
article image

References

  • Bird, G. and D. Rowlands, 2002, “Do IMF Programmes Have a Catalytic Effect on Other International Capital Flows?,” Oxford Development Studies, Vol. 30, No. 3.

    • Search Google Scholar
    • Export Citation
  • Bird, G. and D. Rowlands, 2004, “Financing Balance of Payments Adjustments: Options in the Light of the Elusive Catalytic Effect of the IMF-Supported Programs,” Comparative Economic Studies 46, No. 3.

    • Search Google Scholar
    • Export Citation
  • Bulir, A. and Hamann, J., 2003, “Aid Volatility: An Empirical Assessment,” IMF Staff Papers, Vol. 50, No. 1, pp. 6489.

  • Bulir, Ales, and Timothy Lane, 2004, “Aid and Fiscal Management,” in Helping Countries Develop: The Role of Fiscal Policy, ed. by Gupta, S., Clements, B. and Inchauste, G. (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2003, “The Conclusions of the Task Force on Prolonged Use of Fund Resources,” Washington, DC.

  • International Monetary Fund, 2003, “Operational Guidance Note for Staff on Letters and Statements Assessing Members’ Economic Conditions and Policies,” Washington, DC.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2003, “Role of the Fund in Low-Income Member Countries Over the Medium Term—Issues Paper for Discussion,” Washington, DC.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2003, “Fund Assistance for Countries Facing Exogenous Shocks,” Washington, DC.

  • International Monetary Fund, 2004, “The Fund’s Support of Low-Income Member Countries—Considerations on Instruments and Financing,” and Chairman’s Summing Up, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2004, “Biennial Review of the Implementation of the Fund’s Surveillance and of the 1977 Surveillance Decision—Overview,” Washington, DC.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2004, “Background Paper for the IEO Evaluation Report on PRSPs and the PRGF—Summaries of Country Case Studies,” Washington, DC.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund, 2004, “Signaling by the Fund—A Historical Review,” Washington, DC.

  • International Monetary Fund, 2004, “The Role of the Fund in Low-Income Member Countries,” Washington, DC.

  • International Monetary Fund, 2004, “Policy Monitoring Arrangement,” and Chairman’s Summing Up, Washington, DC.

  • International Monetary Fund, 2005, “Review of the 2002 Conditionality Guidelines,” Washington, DC.

1

The different aspects of the role of the Fund’s role in low-income members have been discussed in IMF documents: The Conclusions of the Task Force on Prolonged Use of Fund Resources, Role of the Fund in Low-Income Member Countries Over the Medium Term—Issues Paper for Discussion, Fund Assistance for Countries Facing Exogenous Shocks, The Fund’s Support of Low-Income Member Countries—Considerations on Instruments and Financing, Signaling by the Fund—A Historical Review, and The Role of the Fund in Low-Income Member Countries.

2

Following Board discussion of a Policy Monitoring Arrangement in September 2004, the IMFC explicitly requested in its October 2004 communique “further work on these proposals, including the usefulness and potential demand, in close consultation with potential users, donors, and creditors.” The results of this survey are presented in Annex I.

3

See the April 16, 2005 communique of the IMFC (paragraph 12).

4

Other names have been used to designate a new instrument in the recent past, including policy monitoring arrangement in the IMFC communique. Staff recommends PSI to avoid possible confusion with (i) the Policy Monitoring Arrangement (PMA) discussed by the Board in September 2004; and (ii) terminology used in the context of traditional financial arrangements (e.g., program, arrangement).

5

As detailed in Annex I, a number of low-income members have expressed interest in a new, non-borrowing instrument. They consider that a non-financial relationship with the Fund[0] could demonstrate country ownership of policies, help maintain substantive policy dialogue with the Fund, and send credible signals to the donor community. They also believe that such arrangements should be endorsed by the Board and have a policy commitment content that is comparable across countries.

6

Many of these members have low-access PRGF arrangements with the Fund and have benefited from debt relief in the context of the enhanced HIPC Initiative.

7

Signaling by the Fund—A Historical Review identified the main difficulties in designing such mechanisms as: (i) the tension between the desire to influence donors (the Fund’s catalyzing role) and the desire to leave them to reach their own judgment (reducing the negative and unintended consequences of the Fund’s perceived gatekeeper role); (ii) for positive signals to be meaningful, negative signals must also be possible, but the Fund has been reluctant to send negative signals to protect the frankness of the dialogue with the authorities and avoid a sharp reduction in foreign financing; (iii) on/off signals are open to misinterpretation if the signaling mechanism sets a standard other than upper credit tranche conditionality; (iv) the credibility of signals is helped when backed by financial resources, although recourse to Fund resources can be a signal of need as well as strength; and (v) on/off signals tend to crowd out multidimensional ones.

8

See Policy Monitoring Arrangement and Chairman’s Summing Up.

9

The main differences with the Policy Monitoring Arrangement proposed to the Board in September 2004 are: (i) the rigid calendar of reviews would be dropped; (ii) there would be a presumption of access to a new shocks window in the PRGF trust instrument; (iii) consistent with this presumption, the new instrument is primarily aimed at low-income members, and a poverty reduction strategy would need to be in place; and (iv) the transparency requirements would be no more stringent than for Fund financial arrangements. The reasons for these changes are discussed in more detail in the next paragraphs.

10

See companion paper: Strengthening the Fund’s Ability to Assist Low-Income Countries Meet Balance of Payments Needs Arising from Sudden and Exogenous Shocks.

11

The PSI has been designed primarily to meet the needs of low-income members. Should middle-income (emerging market) members express interest in this type of instrument, the PSI’s eligibility criteria could be revisited or a different instrument developed tailored to their needs.

12

The legal basis for the PSI would be Article V, Section 2(b) of the Fund’s Articles, which provides that “if requested, the Fund may decide to perform financial and technical services … that are consistent with the purposes of the Fund.” This activity would therefore be voluntary for both the relevant member and the Fund.

13

Low-income members requesting a PSI could still have protracted balance of payments problems. However, they would either have no balance of payment need within the meaning of Article V, or any such need would be met by other donors. They would also have their development financing needs met by donors and other IFIs.

14

The survey showed that, among low-income countries and donors who prefer on/off signals, about half were of the view that the signal had to have a standard equivalent to upper credit tranche conditionality, and almost all that the signal had to be endorsed by the Board (See Annex I).

15

See Signaling by the Fund—A Historical Review. In the absence of a clearly defined benchmark, it is difficult to establish standards that can be applied uniformly across the Fund’s membership.

16

See Signaling by the Fund—A Historical Review for a discussion of this standard.

17

This does not exclude that additional adjustment would likely be needed to address the immediate balance of payments need arising from a shock (see Section II.C).

18

As recommended by the Biennial Surveillance Review.

19

In the survey, both donors and low-income members strongly supported Fund monitoring of structural reforms.

20

Fund assessments tend to provide complex, or multidimensional, signals about developments. In the case of members with or seeking an arrangement, they can also provide simple on/off signals. Multidimensional signals provide a comprehensive assessment of performance in both quantitative and qualitative terms, covering the authorities’ commitments, the context, and the outlook. In contrast, an on/off signal is taken to mean a binary assessment of whether the member’s program is “on” or “off’ track. When reviews are completed on time, the Fund is sending a simple “on” signal, which is essentially shorthand for a multidimensional assessment that “all is fine.” This generally satisfies the needs of low-income members and donors, given the extensive coverage of review documentation. However, in the converse situation, when “not all is fine,” rather than a simple “off’ signal, then a multidimensional assessment would provide donors with broader information on which to base assistance decisions. This would also suit low-income members as, given differing priorities, donors may choose to continue to give their assistance even when program reviews are delayed.

21

In line with the practice in program countries, it is proposed that the 24-month cycle for Article IV consultations be applied to members with a PSI with a duration of 2 years or more.

22

The authorities’ MEFP would need to specify which quantitative and structural performance targets would have to be met, or waived if missed, for completion of the review.

23

In the context of the forthcoming review of Ex Post Assessments, staff also intends to address the question of whether the LTPE policy should continue to include time spent under precautionary arrangements in which no purchases or disbursements are actually drawn.

24

If a member with a PRGF arrangement faced a similar need, augmentation of access would be the appropriate response.

25

In the event of access under the shocks facility, a member’s PSI would lapse automatically. A member with a PSI faced with a shock might consider requesting a PRGF arrangement rather than using the PRGF “shocks window.”

26

As noted in The Fund’s Support of Low-Income Member Countries—Considerations on Instruments and Financing, certain legal issues would pertain to the creation of an explicitly precautionary PRGF. However, such legal issues would not arise should members simply decide unilaterally to refrain from requesting disbursements under their PRGF arrangement.

27

In some cases, the information gap could be filled by holding the Article IV consultation on time, rather than delaying it in parallel with the program’s approval or review.

28

See Operational Guidance Note for Staff on Letters and Statements Assessing Members’ Economic Conditions and Policies. Under current guidelines, letters are circulated to the Board for information at the time of their dissemination to outside recipients and the authorities’ consent is sought for publication.

29

In case of disagreement, staff would present projections based on announced policies, indicating that these are staff projections.

30

Staff considers that such cases would be relatively few in the near future. However, this issue could arise more in the medium term. In addition, according to the companion paper on Financing the Continuation of the PRGF and the Establishment of a Window within the PRGF Trust for Balance of Payments Needs Arising from Exogenous Shocks, the number of members with a PRGF arrangement is expected to remain broadly unchanged. However, if some of the countries assumed to “graduate from the PRGF” were to request a PSI, additional staff resources would be required. Assuming that 5 of the 16 countries expected, in the companion paper, to remain without a PRGF arrangement or to graduate from the PRGF were to request a PSI in the medium term (2007-11), the additional staff resources requirements would be about 8 staff years. Additional staff resources requirements could also stem from the use of the “shocks facility” by countries without a PRGF arrangement or a PSI. These are estimated at 5 staff years in the medium term (see companion paper).

31

This corresponds to the production of an assessment letter (in addition to the current workload) for about 20 percent of PRGF-eligible members a year.

32

This would represent a change from the current policy for assessment letters.

33

It would be possible to establish a Fund-administered trust account where donors could place grant money to cover user fees for PSIs and assessment letters.

34

Modes of Fund engagement include: surveillance, with Article IV consultations in low-income countries typically focusing on the role that fiscal, monetary, and structural policies can play in fostering stability, growth, and poverty reduction; financial arrangements in support of members’ programs; technical assistance/capacity building, advice, and support on key issues relating to institutional and policy reform and implementation; and training.

35

As general budget support is not earmarked for specific uses, in effect it supports the implementation of the recipient’s overall policies. This highlights the need for a good policy environment—recent assessments of aid effectiveness demonstrate that aid is most effective in good policy environments—and the fact that general budget support is disbursed into existing government systems, which increases the focus of donors on the quality of both public financial management and overall macroeconomic management.

36

All major donors answered the questionnaire (see list in Annex II), and hence the survey’s results reflect their views comprehensively. Responses from 31 low-income members, over half from Africa, were received. The response rates of 40 percent for low-income members and 62 percent for PRGF users are higher than in previous similar surveys. For instance, the survey of PRGF users conducted for the Review of the Poverty Reduction and Growth Facility—Issues and Options had a 50 percent response rate and the survey of all the countries with Fund arrangements conducted for Strengthening IMF-World Bank Collaboration on Country Programs and Conditionality—Progress Report had a 40 percent response rate. The results for low-income members are not as comprehensive as for donors, however. In any case, the results should not be interpreted as hard statistical evidence, but rather as indicative of the views of low-income members and donors. The survey also covered other issues of Fund engagement with donors. The corresponding results are detailed in a forthcoming Board paper on the role of the Fund in the PRS process.

37

These views were often expressed in contacts with country authorities in the context of the survey. The survey itself showed that: 78 percent of the countries responding on whether a new instrument is needed favored its creation (Question B16, Table 1); and about 40 percent of donors were of the view that a multidimensional assessment alone was insufficient, and that an on/off signal would also be required (Question B3, Table 1). Of those, only about a third were of the view that it was necessary to have the Fund’s money on the line (Question B7, Table 1).

38

Almost all respondents (97 percent) stated that they use Fund signals in their assistance decisions. The exceptions were donors with a specific mandate, such as development funds dealing mainly with projects. While a number of major donors assigned a high importance in their decisions to Fund signals (the median was 7 and the average 6.3 on the scale 1-10), responses varied widely (from 2 to 10), with only a few donors indicating that they rely primarily on their own sources of analysis.

39

Other sources include the World Bank and other multilateral institutions, embassy networks, and official documents produced by aid recipients.

40

As confirmed by the survey, and outlined in Box 1, Fund signals also play an important role in debt-relief operations.

41

There are a few cases among the main donors where the presence of an on-track Fund-supported program is a requirement for providing budget support.

42

See The Fund’s Support of Low-Income Member Countries—Considerations on Instruments and Financing and Chairman’s Summing Up.

43

The response rate is even significantly higher for countries with a PRGF arrangement.

44

While the overall rate of satisfaction with IMF signaling seems high, few countries expressed its dissatisfaction with IMF signaling, but chose not to respond to the questionnaire.

45

Countries that do not seek IMF financial support may be more candid/critical in their views than countries that have a Fund-supported program. In the actual answers, however, several countries that have a current financial arrangement with the IMF expressed reservations about the IMF activities and signaling, indicating that overall the answers provide a balanced view.

Policy Support and Signaling in Low-Income Countries
Author: International Monetary Fund
  • View in gallery

    What Signals Should the Fund Ideally Send and in What Manner and Frequency?

  • View in gallery

    Types of IMF Signaling Mechanisms Used, Their Importance in Assistance Decisions and Types of Aid Affected

    What IMF mechanisms are useful for signaling?

  • View in gallery

    Impact of IMF Signals on Continuation of Donors’ Budget Support

    (in percent)