Staff Response to the Evaluation by the Independent Evaluation Office of Prolonged Use of Fund Resources
- International Monetary Fund. Independent Evaluation Office
- Published Date:
- April 2003
1. This first evaluation conducted by the Independent Evaluation Office presents a useful analysis of the prolonged use of Fund resources over the past three decades and is a welcome addition to the set of independent evaluations of Fund policy and practices. The common objective of the Fund staff and the IEO is to strengthen the Fund’s lending decisions and improve the design and implementation of Fund-supported programs, so that they foster lasting improvements in members’ growth and welfare. The underlying issues raised in the report—how to enhance program ownership, how to help countries establish a track record of reform, and when to give the authorities the benefit of the doubt—are ones with which Fund staff and management grapple every day. In this regard, many of the report’s useful recommendations—such as those pertaining to streamlining conditionality and enhancing program ownership—are being implemented under initiatives undertaken by management or the Board, while others will require further consideration.
2. Although the staff agrees with much of the report, there were also issues where it would take a somewhat different position. The purpose of this note is not to address every such issue, but to provide an initial reaction to some important issues the report raises.
Nature and incidence of prolonged use
3. One of the important contributions of the report is to raise the question of when prolonged use is a problem, and if so, why. Prolonged use has emerged in part due to a recognition of the benefits that longer-term involvement may be able to offer the Fund’s members in some cases. This is true, for example, of the Fund’s involvement in low-income countries, where there has been an increasing appreciation of the protracted nature of their balance of payments needs. For other developing countries engaged in a transition to the point when they can benefit from durable access to international capital markets, the process of developing the macroeconomic policy framework and institutions necessary to deal with the risks that come with integration can be protracted. Given the challenges associated with this transition, and the new challenges that have accompanied global capital market integration, many countries have faced the need for periodic engagement with the Fund.
4. These examples would not justify the suggestion that there are no problems associated with prolonged use, and here, the report correctly identifies several issues. The report’s basic premise, which is widely accepted—and the staff would strongly endorse—is that the Fund is not intended to be a continuous source of financing for its members over a protracted period of time. As Executive Directors have noted on a number of occasions,1 the Fund’s capacity to provide support under a succession of arrangements or purchases over time is not intended to enable it to become a source of continuous financing, but to maintain its ability to respond to its members’ temporary needs for balance-of-payments financing as and when they arise.
5. The definition of a prolonged user of Fund resources presented in the report is that the member has been under Fund-supported programs for seven or more years in a ten-year period. On this basis, 44 members (29 of them PRGF-eligible) were prolonged users at some point during the period 1971–2000; a further seven members were prolonged users if precautionary arrangements are included. But in reporting on the overall incidence of prolonged use, grouping together GRA and non-GRA users may give a misleading picture: about 40 percent of PRGF-eligible members would be classified as prolonged users, while among middle-income members the ratio is closer to 20 percent (or about 25 percent if precautionary arrangements are included).
6. The evidence on the incidence of prolonged use presented in the report is broadly consistent with the findings of other work by the staff. The preparatory work done for the review of Fund facilities in 2000 found relatively few cases of members drawing resources in the credit tranches year after year, but many more cases where members entered repeat arrangements while making few purchases, because the programs went off-track relatively soon.2 This indicates that, in general, the fundamental issue is not that of tying up Fund resources for excessively long periods, but of negotiating programs that, for whatever reason, were not implemented. Consequently, our approach has not been to tackle prolonged use per se, but to promote more successful program design and implementation and enhance ownership by modifying Fund policies in the areas of conditionality, surveillance, and technical assistance.
7. The report finds that the rising incidence of prolonged use is largely accounted for by the increase in such use by PRGF-eligible members. For these members, the causes of prolonged use and the issues it raises are generally somewhat different than for the rest of the membership. The report rightly notes that the balance of payments needs of many low-income members are of a protracted nature, and thus the Fund’s assistance needs to be sustained over a considerably longer period than for other members. Under the ESAF, and now the PRGF, current Fund policies recognize that prolonged support is legitimate (in contrast to the situation under the earlier Trust Fund and Structural Adjustment Facility (SAF)). The guiding principles for the use of our concessional resources are governed by the terms of the PRGF Trust instrument, rather than the general provisions of the Articles. In these cases, the central problem is not the prolonged nature of the Fund’s involvement, but the failure of some PRGF-supported programs to produce the intended results. At the same time, one must bear in mind that some prolonged users among the low-income members have very strong records of successful program implementation, and indeed, it is because their records are strong that their programs have been supported with successive Fund arrangements. Uganda is one such example. While we do not intend to suggest that prolonged use of this kind has no potential difficulties associated with it, such differences should be borne in mind in understanding the phenomenon of prolonged use and in devising strategies to address it.
8. The report also notes that prolonged use by low-income members is related in part to the need of aid donors and other providers of funds to have assurances that they are disbursing into a sound macro-economic policy environment. Whether a PRGF arrangement is in every case the appropriate vehicles through which the Fund should signal its endorsement of a member’s macroeconomic and poverty reduction policies is a good question, and one that will be explored further in a note on the Fund’s role in its low-income members that will be ready for discussion sometime after the Annual Meetings. In addition, shortly after the Annual Meetings we plan to present a paper revisiting the more general issue of the use of Fund arrangements as a signaling device, and the extent to which other vehicles, such as staff-monitored programs or some form of enhanced surveillance, could be used for this purpose.
Consequences of prolonged use
9. In evaluating prolonged use as a whole, it is important to take account of the important tradeoffs involved. It is important to recognize that a sustained period of engagement is not always exclusively detrimental, but in a number of other cases, prolonged use is an indication that the adjustment programs that the Fund has been supporting may not have been effective in delivering the balance of payments adjustment anticipated. The latter cases are more clearly problematic, even recognizing that a variety of factors, including adverse shocks and political developments, contribute to program outcomes.
10. The report presents a body of analysis characterizing prolonged users and the associated economic outcomes, concluding also that prolonged use of Fund resources is associated with widespread adverse economic effects. However, it does not seem to find such effects in cases of prolonged use of concessional Fund resources, where there is a significant positive coefficient for output per capita. It should be noted that, in attributing such effects, there is an important identification problem, since the correlation may simply reflect countries subject to larger shocks and slower growth being more likely to need to turn to the Fund. The econometric techniques employed do not provide completely convincing results on the direction of causation.3 However, some other negative effects, such as the impact of successive program negotiations on the process of economic policy formulation, are likely to be significant in some cases, even if they cannot be statistically identified.
11. The potential problems associated with prolonged use are most clearly illustrated by the report’s critical and insightful look at three case studies (Pakistan, Philippines, and Senegal). The discussion of the selected case studies highlights discrepancies between the time frame of programs and the ambition of their objectives, the understatement of program risks associated with weak ownership and political commitment, insufficient analysis of the key risks to the program, and relative absence of ex post assessments of programs. In addition, the report argues that structural conditionality was often poorly prioritized, so that compliance with a subset of conditions did not ensure that the most critical problems were being addressed.
12. While these case studies highlight many issues that are both valid and concerning, the staff’s overall assessment of these country cases is somewhat different from that in the report. For example, the 1986–97 programs with the Philippines achieved some measure of success, laying the foundation for its to return to the capital markets—access which has been sustained without interruption (even during the Asian crisis) for nearly a decade. Indeed, of the Asian crisis countries, the impact on output growth was the smallest in the Philippines, in part because of the wide-ranging structural reforms that the country had undertaken under successive Fund-supported programs.
13. It is also important to recognize the element of selectivity in choosing these case studies. To take another example, the report could have included a study of at least one case, such as Bulgaria or Latvia, in which Fund support, although prolonged under the report’s definition, has been more clearly and overwhelmingly positive. Indeed, the willingness of member countries to request precautionary arrangements—and pay the associated commitment fee—suggests that engagement with the Fund carries substantial economic value.
14. In those cases in which prolonged use is motivated by the prolonged nature of the structural and institutional factors underlying the country’s balance-of-payments problems, the report rightly notes that the objectives of Fund-supported programs have not always been clearly articulated. In particular, when the SAF and ESAF were created, both were envisaged as one-off operations to support adjustment in low-income members during a relatively short period and with limited resources. These constraints were gradually lifted in several steps during the 1990s, culminating in the transformation of the ESAF into the PRGF. However, the intended duration of Fund involvement in individual cases often remained unclear, because the program objectives were not sufficiently clearly defined. The need for a clear articulation of program objectives is an important lesson.
15. Given that prolonged use reflects virtually every aspect of the way the Fund interacts with member countries, the report’s twenty-two recommendations are wide-ranging. Here we focus on a few of those recommendations that are most directly related to prolonged use—specifically the suggestions of greater selectivity in extending financial support, developing alternative vehicles to deliver a seal of approval, adopting explicit exit strategies, and establishing a higher rate of charge for prolonged users. In the following section, we discuss these recommendations in relation to the staff’s existing work.
16. Exercising greater selectivity in extending financial support is a long-standing idea, which was most recently considered in the context of the Fund’s conditionality review; the merits of greater selectivity are recognized, in particular, in the draft of the new conditionality guidelines. Weighing the consequences of a refusal to support the member’s policies against the need to give the authorities the appropriate degree of benefit of any doubt, especially when a new government takes office, or when there is a lack of sufficient track record, will always be a difficult matter.
17. A related proposal is the establishment of higher charges as a financial incentive for the country to curb prolonged use. It is not clear whether such incentives would be effective. This issue was considered during the 2000 review of Fund Facilities when the Board approved new surcharges on large amounts of credit outstanding, and stated its intention not to review Fund charges for a period of four years.
18. Whether there are viable alternatives for the “seal of approval” function of Fund-supported programs—whether staff-monitored programs or enhanced surveillance could fulfill this function—remains an open question. There is a general perception, however, that the credibility of the Fund’s approval is significantly greater when the Fund commits its resources to the member’s program. This perception is bolstered by questions about the quality of the “seal of approval” provided by staff-monitored programs (SMP), given that, as noted during the 2002 biennial review of surveillance, they do not carry Fund endorsement, they do not have to meet the standards of upper-credit-tranche conditionality, and reporting on performance under SMP is often limited. As regards providing a “seal of approval” for other donors and the Paris Club, this topic is perhaps best discussed in the con-text of the review of the role of the Fund in low-income countries; to date, however, bilateral donors and lenders, including the Paris Club, have typically insisted on the financial presence of the Fund as a condition for their support (and, in response, the Fund has at times reduced access to token amounts).
19. The report also calls for greater selectivity in program content, concluding that the reform agenda in many countries was overloaded. In other contexts, the report argues that the Fund has been too accommodative of program slippages, which is one of the factors behind prolonged use—implying that conditionality (including prior actions) should have been enforced more rigorously. A key impetus of the current conditionality review—expected to be codified in the new conditionality guidelines—is to streamline and focus conditionality on those measures deemed critical to the program’s objectives. An important element of this streamlining process is enhanced collaboration with other multilateral institutions, especially the World Bank; to this end, an operational framework for Bank-Fund collaboration has recently been formalized and a staff guidance note issued.
20. Finding the right balance between realism and ambition remains a difficult challenge to the institution. In this context, the recent initiative to improve the Fund’s analysis of public and external debt sustainability emphasizes the need to discipline projections and to undertake adequate sensitivity tests of scenarios. While this will not, in itself, necessarily reduce instances of prolonged use, it should result in a better articulation of realistic program objectives and projections. Recognition that a member’s reform agenda will take many years to complete may argue for more explicit recognition that external support, including that of the Fund, will need to be prolonged, rather than doing anything to reduce prolonged use.
21. There is widespread agreement that national ownership of reform programs is essential to their success, although the report glosses over the difficulty of achieving and sustaining national ownership of sound macroeconomic and structural adjustment policies. The PRSP process, initiated in 1999, is intended to help promote stronger ownership of PRGF programs, and in 2001, the Board discussed operational procedures aimed at increasing national ownership of Fund-supported programs more generally. In any case, nationally-owned policies should not mean not weaker, but better, programs.
Chairman’s Summing Up, “Modalities for Special ESAF Operations in the Context of the HIPC Initiative and Other ESAF Issues,” Executive Board Meetings 97/5, 97/8, and 97/10.
Review of Fund Facilities—Preliminary Considerations, EBS/00/37, March 2, 2000, pages 29–30, and Annex IV.
The econometric work presented in an annex to the paper is an extension of recent work by Barro and Lee (2002). While the annex does not present enough information to permit a full assessment, it would appear to share the shortcomings of that paper, including with regard to the choice of instrument and variables (the possibility that they may be correlated with the error term; their explanatory power; and their interpretation), and of the treatment of PRGF programs in the sample. It is also not clear in the report how the analysis took account separately of the endogeneity of the existence of a Fund-supported program versus the prolonged nature of the use of Fund resources, given that both aspects are likely to be correlated with the error term.