Public Information Notice No. 05/16

Public Information Notice: IMF Executive Board Discusses Program Design

February 8, 2005

On December 17, 2004, the Executive Board of the International Monetary Fund (IMF) discussed program design in Fund-supported programs in light of experience over the past decade.

Background

A Fund-supported program is a package of policy measures which, combined with approved financing, is intended to accomplish specific economic objectives such as reducing inflation, achieving orderly external adjustment, and promoting growth and poverty reduction. Agreed policies are monitored by conditionality to give the country confidence that it will continue to receive financing from the Fund as long as it implements the policies envisaged under the program, while providing safeguards for the Fund's resources.

During the course of the 2000-02 Conditionality Review, Executive Directors requested that the next conditionality review address broader issues of program design. In response to this request, the 2004-2005 Conditionality Review consists of two parts. The first part examines the design of Fund-supported programs, their objectives, and the extent to which these objectives were achieved. It was discussed by the Executive Board on December 17, 2004 and, at the request of Board members, follow up work on this topic will be carried out in the near future. The second part of the Conditionality Review will consider the application of the Conditionality Guidelines which were introduced in 2002. The board in scheduled to discuss these issues in March 2005.

Executive Board Assessment

Executive Directors welcomed the staff's comprehensive empirical study of country experiences with Fund-supported programs approved during the period 1995-2000, which seeks to draw lessons for improving program design. Today's discussion and the forthcoming review in early 2005 of the application of the 2002 revised Conditionality Guidelines will constitute the two pillars of the 2004-2005 Conditionality Review.

Directors commended the study's careful and comprehensive analytical treatment of various aspects of program design in terms of program objectives and outcomes, analytical frameworks, and policies. This detailed review of the full spectrum of Fund-supported programs has produced a range of analytical findings, which Directors felt will be valuable in informing the Fund's ongoing work on program design issues.

Directors broadly supported many of the interesting findings of the study, some of which empirically confirm previous, more anecdotal assessments of program design and performance. They considered that public availability of the findings, based on the "stylized facts" developed in the study, should be helpful in improving public understanding of the experience of Fund policies in a key area of its responsibilities. At the same time, however, Directors considered that the study also raises new issues and questions, which will merit further analysis and inquiry by the Fund. In this spirit, and given the study's broad scope and breadth, Directors saw the study as a work in progress that will help enrich our ongoing learning and evaluation of program experience. Carrying forward this work will help to draw out more fully its operational implications for program design and conditionality, as well as for other aspects of Fund-supported programs.

Turning to the specific aspects of the study, Directors concurred broadly with the taxonomy and characterizations of the different types of Fund-supported programs and their objectives developed in the staff paper—classic, capital account crisis, transition economy, and low-income cases. Directors also noted that Fund-supported programs are in general tailored to members' individual economic conditions and objectives. At the same time, there could be merit in focusing on the purpose or objectives of Fund lending in each individual case, and analyzing country program performance in relation to those objectives.

Objectives and Outcomes

Directors agreed that external viability remains a core objective of Fund-supported programs. Fund resources are made available to member countries to correct balance of payments problems, while easing the burden of adjustment and helping to replenish international reserves. Directors noted that anchoring external adjustment in medium-term debt sustainability allows for a better balance to be struck between putting external debt dynamics on a more sustainable path, thus reducing vulnerability to future crises, and rapid external adjustment, which may entail adverse consequences for economic activity.

Turning to General Resource Account-supported programs, Directors noted that external adjustment targeted in these programs has been broadly in line with considerations of medium-term debt sustainability, with programs seeking to stabilize external debt ratios or reduce them to appropriate levels. Moreover, a given improvement in the current account balance, when associated with Fund financial support, leads to a smaller negative impact on growth. Directors also noted, however, that in a number of GRA-supported programs—especially, though not exclusively, cases of capital account crisis—private market financing has turned out to be much less abundant than expected, resulting in current account adjustment that is both larger than envisaged and larger than that required to stabilize external debt at moderate levels. The sharpness of the external adjustment in these countries has had pervasive macroeconomic consequences, including steep output declines and inflationary spikes.

Directors expressed a range of views on the underlying causes of these outcomes. Some Directors stressed that, in some cases, crisis countries' initial failure to implement credible policies may have undermined the attempt to catalyze a positive market response. Others considered that the sharpness of external adjustment points to a need to consider the adequacy of official financing packages, while noting that, in the absence of supporting domestic policies, a larger official financing package could have simply facilitated a faster exit by the private sector.

Directors called for further careful analysis of program design issues relating to capital account crises, as such cases represent a major challenge to global financial stability and require the largest share of Fund resources. In particular, they encouraged staff to undertake further analysis of the optimal mix between financing and adjustment as well as of the determinants of capital flows, which have tended to fall short of expectations, and of the catalytic effects of Fund-supported programs. In this regard, the aim should be to develop a better understanding of how markets evaluate and respond to the design and signaling aspects of Fund-supported programs. It was suggested that these considerations heighten the importance of a wider application of balance sheet analysis, especially in emerging market countries, which are vulnerable to capital account movements. Directors also attached importance to the need for sound judgment as well as close collaboration between staff and the country authorities in program design. In this context, Directors saw major analytical challenges in capital account crisis cases, where near-term projections, on which programs are based, have often proved inaccurate. They urged staff to examine further methods for modeling capital flows and their macroeconomic impact as well as for balance sheet approaches, including the contingent claims approach, where feasible, to identifying key vulnerabilities.

Directors considered the contrasting experience in Enhanced Structural Adjustment Facility/Poverty Reduction and Growth Facility-supported programs. While these programs typically built up reserves and liberalized import regimes, medium-term external debt sustainability (in the absence of debt relief) has neither been targeted nor achieved. Improvements in current account balances, on average, have been smaller than programmed, and insufficient to stabilize external debt ratios. Indeed, both programmed and actual improvements in the current account balance have been negatively related to the initial debt ratios, contrasting with the positive relationship for GRA-supported programs. While in part this may have reflected anticipated HIPC debt relief, in general, current account deficits have nevertheless been too large to stabilize external debt ratios, even at the lower levels prevailing after HIPC debt relief. Directors called for further reflection on how this phenomenon would be corrected in designing PRGF-supported programs. The low-income debt sustainability framework could play a role in this regard. Another avenue could be the introduction more broadly of limits on concessional borrowing. Several Directors considered that a key factor underlying external debt sustainability problems in low-income countries relates to an inappropriate financing mix, and a few Directors suggested that this situation should be addressed by increasing the share of financing provided in the form of grants. Directors also considered that greater attention should be given, in designing PRGF-supported programs, to the effects of exogenous shocks on achieving external adjustment and debt sustainability. These issues require further review and some will be taken up in the context of the upcoming discussion of PRGF program design.

Directors took note of the favorable findings relating to program outcomes for key macroeconomic targets. Most GRA-supported programs have seen sustained decreases in inflation and recovery of output growth rates to pre-crisis levels. In low-income countries, in contrast to the experience in the 1980s, output growth has risen significantly during the program period, with the improved performance largely sustained in the period following the program as well. This improved growth performance reflects greater macroeconomic stability, including lower inflation and smaller after-grants fiscal deficits, as well as a more benign external environment. Directors stressed the need to maintain this improved economic performance in low-income countries, while taking account of external debt sustainability concerns. They considered that the design of programs in low-income countries should be based on a full consideration of the implications of policies for poverty reduction, taking into account micro-macro linkages and Poverty and Social Impact Analyses. Directors called for future analytical work aimed at identifying those program elements that are shown to be most successful in helping to reduce poverty as well as better understanding the determinants of growth in low-income countries.

Directors noted that policies in precautionary arrangements are broadly similar to those in programs where the member expects to draw, and that there is no evidence of statistically significant differences in outcomes. However, it was noted that real GDP growth during the program period and the savings ratio over the longer term tend to rise under the reviewed precautionary programs, contrary to the experience under all other GRA-supported programs.

Analytical Framework

Directors agreed that no single model or analytical framework is universally applicable to policy formulation in Fund-supported programs. They welcomed the fact that national authorities, and Fund country teams, normally draw on a variety of models and methods for policy formulation and temper them with economic judgment. In this connection, Directors observed that the Fund's financial programming framework is more useful for providing a consistency check on macroeconomic projections and the formulation of policies underlying program design than for tying down the parameters of the economic program.

Directors observed that this eclectic approach to policy formulation has generally worked well in practice. With the exception of capital account crisis cases, near-term projections for output growth and inflation have not exhibited systematic bias. However, external adjustment has tended, on average, to be under-estimated in GRA-supported programs, and over-estimated in PRGF-supported programs. The relationship between macroeconomic instruments and targets assumed in Fund-supported programs has also been consistent with outturns, showing no systematic bias, except that program projections tend to under-estimate the positive impact of fiscal adjustment on growth and on the improvement of the current account balance.

Medium-term growth projections, by contrast, have been over-optimistic. Such overoptimism, Directors considered, risks undermining the reliability of debt sustainability assessments and the credibility of programs. Directors saw the proposed analytical "reality checks" on growth projections—for example, elaborating the demand side components of growth projections, and comparing these projections with a variety of analytical reference points—as useful in helping to guard against over-optimism. More systematic comparisons with forecasts by other analysts, where available, as well as greater use of cross-country analysis, were also recommended. More generally, Directors called for additional work on the sources of the errors in medium-term growth projections. They also saw a need to understand better the medium-term growth impact of structural reforms, including by seeking to identify the categories of reforms that may be more conducive to growth and more macrorelevant. This would also help inform the forthcoming review of the 2002 Conditionality Guidelines, in particular on how best to strike a balance between the principle of streamlining conditionality and the need for reforms to ensure medium-term growth and viability.

Policy Content

Turning to the policy content of Fund-supported programs, Directors noted that exchange rate regimes are no more likely to be altered at the outset of a Fund-supported program than at other times, and drew from this finding a variety of implications. In the broadest sense, Directors agreed that coherence between the exchange rate regime and macroeconomic and structural policies is critical, and cautioned that the Fund should avoid supporting policy mixes that do not sufficiently underpin the exchange rate regime around which a Fund-supported program is designed. Directors also observed that disinflation is achieved equally successfully under fixed or flexible exchange rate strategies, pointing to the finding that success with disinflation depends mainly on whether the targeted fiscal adjustment is achieved. Directors noted that the exchange rate regime remains a matter of choice for the national authorities, but assessing whether a regime is sustainable depends on a judgment about the feasibility of the adjustment burden placed on monetary policy and fiscal consolidation. In this context, some Directors suggested that Fund staff should not remain passive with respect to recommending the use of the exchange rate to achieve program objectives, pointing to the paper's finding that countries with more flexible exchange rates tend to achieve external adjustment at a lower output cost. The technical challenges of assessing the equilibrium level of the exchange rate were also acknowledged.

In reviewing the overall stance of monetary policy in Fund-supported programs, Directors noted that targeted monetary policies have been broadly aligned with program objectives, and there is no evidence that monetary programs have been set excessively tight, leading to lower output growth. Moreover, slippages in money growth rates—regardless of their source—have contributed to inflation targets being missed. Most Directors considered that money growth rates contain valuable information about future inflation performance, and requested staff to consider the implications of these findings for program design and policy formulation in the context of capital inflows or large-scale donor support.

As regards fiscal policy, Directors observed that program practice has been significantly more diverse and matches overall economic objectives more systematically than commonly assumed—as was also concluded in the recent Independent Evaluation Office report on Fiscal Adjustment in Fund-Supported Programs. Targeted fiscal adjustment has been broadly appropriate in light of initial economic conditions and program objectives. At the same time, program targets have often been missed, and fiscal adjustment has not been sustained, jeopardizing macroeconomic stabilization and public debt sustainability. This points to the importance of strengthening and improving debt sustainability analysis to better assess medium-term debt dynamics. In light of the empirical findings that slippages tended to be larger in later years and in programs targeting more ambitious adjustment, Directors stressed the need for greater focus on fiscal consolidation in program design, with emphasis on high-quality fiscal measures that could be politically feasible and sustainable. Several Directors suggested that future work on assessments of fiscal policy should take better account of both the magnitude and quality of fiscal adjustment, and a few Directors thought that a more independent process for debt sustainability analyses could be helpful. Greater analysis of the distributional impact of policies should also help throw light on their political feasibility. At the same time, the largest errors in projecting public debt dynamics stem from below-the-line items, suggesting that greater attention needs to be paid to possible contingent liabilities, including from financial sector restructuring costs. Several Directors also expressed concern about the risk of countries' adjustment falling primarily on investment rather than on consumption, and suggested that this issue be examined in future work.

Directors discussed the role of fiscal policy in promoting external adjustment. They noted that fiscal adjustment has generally contributed to improvements in the current account balance, and that instances where external adjustment was significantly sharper than anticipated mainly reflected a lack of external financing, particularly private capital inflows. In these cases, Directors observed, the appropriate role of fiscal policy has depended on whether capital outflows reflected concerns about public or private sector balance sheet vulnerabilities, and on the nature of the shock that led to the capital outflows. Directors also welcomed the finding that fiscal consolidation has not been associated with lower output growth, suggesting that confidence effects play a significant role.

Directors noted that, in designing programs, structural reforms are viewed as necessary to buttress adjustment efforts by enhancing efficiency and eliminating structural distortions that inhibit long-term growth, and to reduce vulnerabilities to financial crises. In reviewing structural measures in Fund-supported programs, Directors found broad alignment between reform measures and program goals. In particular, they noted that measures intended to underpin demand management seem to have contributed to sustained fiscal adjustment, and that measures geared towards enhancing efficiency have been associated with better output growth outcomes. While these initial indications were seen as useful, Directors underscored that the linkages between structural reforms and macroeconomic performance, including the confidence effects of these reforms and issues of appropriate sequencing of reforms, remain uncertain, and a more detailed analysis will be required before firm conclusions can be drawn. Moreover, several Directors pointed out that the frequent granting of waivers of structural program conditions might be symptomatic of possible flaws in program design, and called for further analysis of these issues. Directors noted that these considerations raise crucial issues regarding the effectiveness of structural conditions in programs. Some of these issues will be discussed further at the upcoming review of the 2002 Conditionality Guidelines.

Looking Ahead

The review of experience with macroeconomic and structural reforms constitutes an important step not only in advancing understanding of Fund-supported programs but also in identifying key questions for further analysis. Indeed, the study points to several areas where Fund program design has "got it right," but equally highlights those cases where program objectives are not being met. It is on the latter areas that the IMF will need to focus future analysis. Directors have offered a number of valuable suggestions for building on the present study, as well as some initial thoughts for drawing concrete operational lessons and policy conclusions from it, which staff has carefully noted. The staff will prepare a short note to form the basis for an informal discussion with Directors on how to arrange the follow-up work on program design issues.

On a key point of methodology, Directors have suggested that the more generalized approach of the present study be complemented with a broad array of other analytical approaches, including more in-depth case studies of disaggregated and homogenous groups, which could yield useful policy guidance for specific types of programs and country situations. Such groups could include capital account crisis cases, precautionary programs, and PRGF-supported programs. In addition, staff should continue to ensure that ongoing analysis benefits fully from the lessons drawn in several recent Independent Evaluation Office studies and other policy papers covering particular aspects of program design.

Directors have also made a number of suggestions for inclusion of additional specific issues in the staff's further analyses of program performance and design. Importantly, Directors have stressed that national ownership and the role of alternative policy options should be further reviewed. Directors have also suggested that follow-up work could usefully reflect the findings of the new ex post assessments. Yet another suggestion is to examine program design and policy implementation, contrasting the experiences of countries with successful exit programs and those with a succession of Fund-supported programs. The experience with precautionary arrangements is another area highlighted for further work. Finally, Directors have also suggested looking at the level of access and Fund financing in program design.