Staff Response to the Report by the Independent Evaluation Office on the Role of the IMF in Recent Capital Account Crises

Kevin Barnes, Ali Mansoor, Benjamin Cohen, and Shinji Takagi
Published Date:
September 2003
  • ShareShare
Show Summary Details

Executive Board Meeting 03/50

May 30, 2003

1. This report by the Independent Evaluation Office on the role of the IMF in recent capital account crises presents many lessons for improving the efficacy of the Fund’s efforts at crisis prevention and resolution, complementing previous studies undertaken both within and outside the Fund. Following up on the staff’s response to the report (SM/03/171, Sup 1),1 this statement addresses the report’s main recommendations. Board discussions are scheduled on many of the areas covered by the recommendations, and staff will reflect the conclusions of the Executive Directors on this report in staff documents for these meetings.

2. The report focuses mainly on the Fund’s involvement in the early stages of the crises. While this phase clearly holds important lessons—and has been the subject of much previous attention, both inside and outside the Fund—it is not the whole story. Most dramatically, in Korea, economic activity was recovering vigorously by the second half of 1998—less than a year after the worst of the crisis—following the restoration of market confidence and a resumption of private capital flows which permitted a substantial easing of policies; in 1999, real GDP rebounded by nearly 11 percent. In Brazil, the situation began to improve within a few months of the abandonment of the exchange-rate peg in January 1999, with the restoration of voluntary capital flows and some monetary easing in the context of inflation targeting; in 2000, growth reached over 4 percent. But with the immediate crisis over, the debt dynamics remained fragile, contributing to the country’s vulnerability when market pressures reemerged in 2002. Even in Indonesia, where the crisis was more severe, progress was made in tackling the fundamental problems in the financial and corporate sectors as early as the second half of 1998 and, as the government’s ownership of the IMF-supported program strengthened and as the policies took hold, the economy’s performance improved markedly. The current report would have been all the stronger, and provided greater value-added over existing literature, if it had focused, not just on the initial crises, but also the later successes and challenges in restoring confidence.

3. While we agree with the general thrust of the report, there are some specific issues on which the report’s conclusions differ somewhat from our own. The Executive Summary focuses almost entirely on what went wrong, without giving any sense of how the Fund responded to the challenges posed by the crisis; it also fails to reflect the complexities of program ownership and implementation which are amply examined in the main body of the report. As detailed in the staff’s own extensive reviews of the crisis cases, there are cases where we conclude that programs put in place at the outset of the crises could have been improved. The IEO report echoes many of these conclusions: while it is broadly supportive of the overall strategy followed, it notes a number of aspects of the programs that did not work as planned. However, some of the report’s criticisms of initial judgments in program design do not in our view provide a sufficiently complete sense of the feasibility and costs of the alternative policy options available at that time.

4. Although the report provides an in-depth analysis of the IMF’s policy advice, it does not sufficiently explore why these crises were so severe. As a result, it overstates the contribution of individual aspects of policy design to the intensity of the crises. In the case of Indonesia, for example, the report, particularly the Executive Summary, could leave the impression that poor policy advice from the Fund (e.g., relating to bank restructuring) was a major factor magnifying the severity of the crisis. However, while with hindsight some aspects of the program might have been designed otherwise, the basic policy response advocated to address the early stages of the crisis was generally appropriate for the evolving circumstances. In the event, a confluence of factors overwhelmed the early program, causing what was initially viewed as a mild case of contagion to degenerate into a full-blown crisis. These factors included, most notably, a worse than expected deterioration of the crisis in the region; a complete loss of monetary control caused by massive liquidity support to the banking system after the government decided against closing additional banks; uncertainty over the commitment of senior political authorities to the policy program; political uncertainty caused, amongst other things, by the President’s ill-health; and the dynamic and self aggravating instability caused by the massive foreign exchange exposure in the corporate sectors

5. The report provides useful recommendations about crisis prevention and management, but it also confirms the impression that every crisis is unique in the problems it poses. Anticipating crises will always require difficult judgments in the context of great uncertainty, and the capacity to prevent them will always depend principally on the actions of our members. The authorities and Fund staff faced enormous analytical and practical challenges as they sought to deal with the onset of crisis. Policy making during the crises frequently involved painful trade-offs—notably those associated with countering market pressures on the exchange rate in the presence of bank and corporate balance sheet weakness. Much of the economic trauma that followed was unavoidable and our collective capacity to contain the damage was limited by problems that were beyond the control of the member and the Fund. We should not expect future crises to be any less challenging, but we will be able to benefit significantly from a systematic effort to incorporate into the work of the institution the experience gained from these cases. With these general thoughts as a backdrop, this statement considers each recommendation of the IEO report in turn.

6. Recommendation 1: To increase the effectiveness of surveillance, Article IV consultations should take a “stress testing” approach to the analysis of a country’s exposure to a potential capital account crisis. The staff supports this recommendation, especially on the need to integrate various dimensions of vulnerability assessments and stress testing into the regular surveillance role of the Fund. Considerable effort is now being made to bring stress testing and other analytical techniques to bear in the Fund’s work. For instance, stress tests are an integral part of FSAPs, the vulnerability exercise, and the debt sustainability framework (which the Board will review in June 2003). There may be other areas to which it would be beneficial to apply this approach, such as in the area of liquidity risk and for low-income countries particularly vulnerable to external shocks, both of which are the topic of forthcoming staff papers. Issues regarding financial vulnerabilities will be addressed in an informal Board seminar on the balance sheet approach in June 2003 and further work incorporating analytical developments in various areas will be undertaken in the context of the 2004 Biennial Surveillance Review (BSR). The critical challenge of course is not simply to explore the resilience and sustainability of a member’s policy framework in the face of various types of shocks, but to identify the types of policy actions that can be taken in advance and in the event of crisis to mitigate those risks. These issues should rank high in the hierarchy of surveillance priorities. A related recommendation is that surveillance pay more attention to social and political constraints on policy making and on market perspectives on policies: the September 2002 Guidance Note on Surveillance calls for particular attention to be paid to these issues in Article IV consultations, and staff will follow this matter up in the BSR in 2004.

7. Recommendation 2: Management and the Executive Board should take additional steps to increase the impact of surveillance, including through making staff assessments more candid and more accessible to the public and providing appropriate institutional incentives to staff. Improving the focus and candor of staff assessments and encouraging more systematic public release of staff reports and the analytical work that supports them can improve the impact of surveillance. The staff sees considerable room for further progress in these areas. The Board will have the opportunity to address many of these issues in its review of transparency policy in June 2003. The issue of greater independence for teams conducting surveillance may be discussed by the Board in the July 2003 discussion on fresh perspectives in surveillance; the issue will be revisited in the 2004 BSR.

8. Recommendation 3: A comprehensive review of the Fund’s approach to program design in capital account crises cases should be undertaken. Program design obviously plays an important role in determining the success of programs, recognizing that a broader range of factors ultimately plays the decisive role. Balance sheet interactions and the uncertainty associated with projecting the path of key variables in capital account crises are recognized as presenting important complications in the initial design of program strategy and reinforce the importance of using the flexibility provided in the program architecture to adapt the strategy as events unfold. Building on the work undertaken since the emerging market crises of the 1990s, the staff have initiated an examination of various dimensions of program design. As part of this effort, PDR is preparing a paper distilling lessons from capital account crises and the implications for program design. This paper will give us the opportunity to explore in detail the various recommendations included in the IEO report. The need to incorporate better assessments of financial vulnerabilities into staff analysis could be taken up at the June 2003 Board seminar on the balance sheet approach. The revised Guidelines on Conditionality, as the IEO report recognizes, specify that conditionality should be streamlined and focused; the 2004 review of Conditionality will include an assessment of implementation of these new guidelines. Staff will also continue undertaking internal reviews of the experience with crisis cases, for instance with the forthcoming paper reviewing Lessons from the Crisis in Argentina.

9. Recommendation 4: Since restoration of confidence is the central goal, the Fund should ensure that the financing package, including all its components, should be sufficient to generate confidence and also be of credible quality. The level, terms, timeliness, and quality of official financing can be critical to the success of a program. The IEO offers valuable reminders about the uncertainty and damage to credibility created by some of the official financing packages that were announced in association with Fund arrangements in past crises. The staff supports the IEO’s recommendations in this context, although in some cases their implementation depends on the actions of other official creditors. The periodic access reviews, as well as the forthcoming review (early 2004) of the experience in applying the new framework for exceptional access decisions, will provide an opportunity to consider experience in these areas in the future.

10. Recommendation 5: The Fund should be proactive in its role as crisis coordinator. The key recommendations offered by the IEO in this context—that management should play a more proactive role in identifying circumstances where concerted efforts can be useful in overcoming “collective action” constraints, that management should provide candid assessments of the probability of success (of a program), and that the technical judgment of the staff should be protected from excessive political interference—are welcome. The specifics of the Fund’s role will have to be determined on a case-by-case basis, but several recent cases offer valuable lessons on how the Fund can be more effective in this area. The new framework for exceptional access decisions provides a mechanism for encouraging more systematic early consideration of circumstances in which the success of a program would be enhanced by voluntary efforts to address collective action problems among private creditors and where steps to address an unsustainable debt burden need to be part of a strategy to restore growth and financial viability. Steps have been taken to strengthen the Fund’s institutional knowledge in this area, including through the establishment of the International Capital Markets Department, and the Capital Markets Consultative Group provides an important new vehicle for improving the Fund’s dialogue with the private sector.

11. Recommendation 6: Human resource management procedures should be adapted further to promote the development and effective utilization of country expertise within the staff, including political economy skills, and to ensure that “centers of expertise” on crisis management issues allow for a rapid application of relevant expertise to emerging crises. The proposed approach for establishing institutional arrangements to deliver a rapid response is, as the report notes, being reflected in the reorganization of MAE (with steps taken to provide dedicated and consistent support on crisis resolution matters), as well as recent changes within PDR (with the establishment of the Crises Resolution Issues Division), and is also being taken up in the review of Area Departments. Steps are being taken to ensure that staff have the necessary political economy skills. A Working Group is examining the role of resident representatives, including their involvement in surveillance and program design. The proposal to ensure that staff are protected from complaints from the authorities is welcome. Although there are no internal guidelines in this regard, there may be a need for greater positive recognition for candor.

    Other Resources Citing This Publication