Journal Issue

New IEO report: Watchdog faults Argentina, but also IMF

International Monetary Fund. External Relations Dept.
Published Date:
August 2004
  • ShareShare
Show Summary Details

IMFSurvey:Why does the role played by the IMF in Argentina’s crisis—unlike other cases, such as Indonesia and Korea—deserve special attention?

Takagi: In Korea and Indonesia, the IMF wasn’t involved immediately prior to the crises, and these crises came as a surprise. But the IMF was involved in Argentina almost continuously for 10 years prior to the crisis.

IMFSurvey: Why wasn’t the IMF able to help Argentina prevent the crisis?

Takagi: In all fairness, there is only so much an international institution can do to influence a country’s policy choices. Ultimately, accountability for a country’s economic policy must rest with its authorities. But having said that, there are several reasons why the IMF wasn’t effective. First, IMF staff may not have fully appreciated the formidable political obstacles in Argentina. The authorities knew what they needed to do, but they didn’t have the political ability to deliver. At the same time, the IMF didn’t use the available tools from its surveillance and program relationships with Argentina as effectively as it could have. Conditionality was weak, and Argentina’s failure to comply with it was often accommodated.

The IMF—although it wasn’t alone in this—was also overoptimistic in its estimation of the impact of certain reforms that Argentina was undertaking and of the prospects for economic growth and manageability of debt. As a result, the IMF remained engaged in a program relationship too long, when the policies being supported were inadequate.

When the crisis unfolded and Argentina sought exceptional access, the IMF viewed its exchange rate or debt sustainability problems as manageable. Staff also had very optimistic assumptions about developments in the world economy.

IMFSurvey:To what extent did IMF surveillance help identify the vulnerabilities that led to the crisis?

Takagi: Surveillance was not sufficiently candid about the inconsistencies between Argentina’s choice of exchange rate regime and its other policies. The IMF recognized the need for fiscal discipline and structural reform—labor market reform in particular—to compensate for the fixity of the exchange rate and underpin the convertibility regime. But its surveillance underestimated the vulnerability that could arise from the steady increase in public debt—much of it was dollar-denominated and externally held—and did not consider exit strategies when it became evident that meaningful progress in structural reform was not going to materialize.

Part of the problem was that there were no accurate and timely data available to allow the staff to monitor provincial finances, where a lot of things were going wrong. Also, back then it was assumed that a debt-to-GDP ratio of up to 60 percent was acceptable—if it was good enough for Europe, it was good enough for Argentina. But that ratio was too high for an emerging market like Argentina with a history of defaults, extensive reliance on external borrowing, a relatively small export sector, and limited capacity to collect taxes. A more appropriate ratio, as staff now recommend, should have been, say, 30 percent.

IMFSurvey:How could the IMF have done a better job of dealing with the political aspects of the crisis?

Takagi: That’s a very difficult question. As an Argentine economist said to me, if a commercial bank were to lend to a sovereign borrower, it would first carefully assess the country’s political situation to determine its prospects of loan repayment. The IMF, in assessing whether loans to Argentina were justified, particularly in 2000 and 2001, didn’t do this.

IMFSurvey:But as lender of last resort of sorts, does the IMF have that luxury?

Takagi: As one of the members of the IMF’s management team indicated to us, political risk shouldn’t be overemphasized because, after all, the IMF is a crisis lender and, by definition, must be willing to lend when no one else is. And the IMF naturally tends to give the “benefit of the doubt” to its members. But at the same time, the IMF’s role as a crisis lender doesn’t justify its lending to a country in an unsustainable situation.

IMFSurvey:In the fall of 2000, when Argentina effectively lost access to the capital markets, it approached the IMF for a major increase in financial support under the March 2000 loan, which up to that time had been treated as precautionary. In response, from January to September 2001, the IMF made three decisions to provide exceptional financial support, raising its total commitments to $22 billion. In December, however, the fifth review of the program was not completed, marking the effective cutoff of IMF financial support. What is your assessment of the IMF’s crisis management strategy from late 2000 through the collapse of convertibility in early 2002?

Takagi: The IMF knew that there would be large exit costs and thus supported Argentina’s efforts to preserve the exchange rate regime. This support was justifiable up to January 2001 because financial support, combined with strong policy corrections, had some chance of success. But after that, further disbursements—made despite repeated policy inadequacies—only postponed the resolution of the crisis.

The Executive Board didn’t fully perform its oversight responsibility. It could have explored the potential trade-offs among alternative options, such as disengaging much earlier. Argentina’s economic situation would have deteriorated that much less and the resources used to try to preserve the peg could have been better used to mitigate some of the inevitable costs of the painful exit. Of course, this is all hindsight, and the exit itself was bound to have unpredictable political consequences, but I should also note that some within the IMF were advocating an exit option from the beginning of 2001.

IMFSurvey:One of your recommendations is that IMF support be predicated on a meaningful shift in policy when the sustainability of debt or the exchange rate is in question. But wouldn’t you agree that any assessment of sustainability and of the genuineness of a shift in policy is probabilistic and does not override the need for judgment?

Takagi: That’s true. The assessment of sustainability is always a judgment call. Certainly, the economics profession doesn’t have the tools to make a precise estimate of the probability that a country will default on its debt. However, the IMF’s decision, in September 2001, to continue to lend to Argentina had a low probability of success—and, in our assessment, was not justified.

IMFSurvey:How does the IEO’s assessment compare to the IMF’s own assessment of the crisis,“Lessons from the Crisis in Argentina,” released in October 2003?

Takagi: Largely, we have similar views, especially on the causes of the crisis and the IMF’s crisis management strategy. But we’re a bit more critical about earlier decisions, especially the May 2001 decision to complete the review. We also take a much more detailed look at the decision-making process.

IMFSurvey:Your report points out that the IMF should have been more attentive to whether Argentina’s convertibility regime—the hard peg to the dollar—was consistent with other policies. What kept the IMF from doing so?

Takagi: First, a country’s right to choose its own exchange rate regime is held very sacred in this institution, and staff and management are very reluctant to challenge the authorities’ choice. It’s part of the culture. Nevertheless, when it became evident that supporting policies weren’t being implemented, the IMF should have told the authorities that they were going down the wrong path and disaster awaited the country. Second, the authorities wouldn’t consider any other option because the public wouldn’t allow it.

IMFSurvey:Some critics have said that your report is somewhat unbalanced in that it focuses mainly on the role played by the staff and not so much on the roles of the IMF’s shareholders, its Executive Board, and its management—or even the Argentine authorities.

Takagi: If there is any bias, it’s because the overwhelming portion of the information we had related to the staff. We say very little about the role of the IMF’s shareholders because communication between management and individual shareholders isn’t documented. And if it were, we wouldn’t have had access to it. We did have access to the minutes of the Executive Board meetings as well as the minutes of informal meetings, which were provided by individual Executive Directors.

IMFSurvey: Some have accused the IMF of yielding to outside market and political pressures to support Argentina.

Takagi: The evaluation team is convinced that such a view is incorrect. It’s one thing to say that management consulted with major shareholders; it’s another to say they yielded to outside pressure against their own judgment. Those closely involved have emphasized that decisions were made in what was, in their view, the best interest of the international community, right or wrong.

The full text of the IEO report, “Evaluation of the Role of the IMF in Argentina, 1991–2001,” along with IMF management and staff responses and the summing up of the Executive Board’s discussion of the report are available on the IEO’s website at

Key IEO findings and recommendations

Major findings

The crisis resulted from the failure of Argentine policymakers to take necessary corrective measures sufficiently early. IMF surveillance failed to highlight the growing vulnerabilities in the authorities’ choice of policies and the IMF erred by supporting inadequate policies for too long. By 2000, there were concerns about exchange rate and debt sustainability, but recognizing the large costs of exit, the IMF supported Argentina’s efforts to preserve the exchange rate regime. This support was justifiable up to January 2001 because large financial support, combined with strong policy corrections, had some chance of success. However, subsequent disbursements, made despite repeated policy inadequacies, only postponed the fundamental resolution of the crisis. In retrospect, the resources used in an attempt to preserve the peg could have been better used to mitigate some of the inevitable costs of exit from the peg.

Precrisisperiod. The IMF correctly recognized fiscal discipline and structural reform, labor market reform in particular, as essential to the viability of the convertibility regime. However, surveillance underestimated the vulnerability that could arise from the steady increase in public debt, when much of it was dollar-denominated and externally held, and the staff did not consider exit strategies when it became evident that meaningful progress in structural reform was not forthcoming. Long-standing political obstacles proved formidable, but the IMF also did not use the available tools effectively. Conditionality was weak, and Argentina’s failure to comply with it was repeatedly accommodated.

Late2000. The IMF increased its commitment of resources to as much as $22 billion. The strategy viewed any exchange rate or debt sustainability problem as manageable with strong action on the fiscal and structural fronts, and it might well have worked if the underlying assumptions had materialized and the program had been impeccably executed. The authorities, however, proved unable to implement the policies as agreed, and the successive resignations of two ministers of economy in March 2001 shattered market confidence. Then the new minister of economy began to take a series of controversial and market-shaking measures. Yet the IMF, having no effective contingency plan, continued to disburse and augment funds in support of the existing policy framework.

Mid-2001. It should have been clear that the initial strategy had failed and that Argentina’s exchange rate and public debt could not be considered sustainable. However, the IMF did not press the authorities for a fundamental change in the policy regime and, in December 2001, effectively cut off its financial support to Argentina. The decision to call the program off-track was fully justified under the circumstances, but the way in which it was done meant that the IMF was unable to provide much help as the crisis unraveled. Exit from the peg would have been very costly regardless of when it was made, but an earlier shift in the IMF’s strategy could have mitigated some of the costs because Argentina’s economic health would have deteriorated that much less and more resources would have been available to moderate the transition process.

Key recommendations

IMFsurveillance needs to be strengthened further, by making medium-term exchange rate and debt sustainability the core focus. To fulfill these objectives, the IMF needs to improve tools for assessing the equilibrium real exchange rate, examine debt profiles from the perspective of “debt intolerance,” and take a longer-term perspective on vulnerabilities that could surface over the medium term. Systematic discussion of exchange rate policy must become a routine exercise on the basis of candid staff analysis.

The IMF should have a contingencystrategy from the outset of a crisis, including “stop-loss rules”—a set of criteria to determine if the initial strategy is working and to guide the decision on when a change in approach is needed. Where the sustainability of debt or the exchange rate is in question, the IMF should indicate that its support is conditional upon a meaningful shift in the country’s policy. High priority should be given to defining the role of the IMF when a country seeking exceptional access has a solvency problem.

The IMF should refrain from entering a programrelationship with a country when there is no immediate balance of payments need and there are serious political obstacles to needed policy adjustment or reform. Exceptional access should entail a presumption of close cooperation, and special incentives to forge such close collaboration should be adopted, including mandatory disclosure to the Board of any critical issue or information that the authorities refuse to discuss with (or disclose to) staff or management.

To strengthen the roleoftheExecutiveBoard, procedures should be adopted to encourage effective Board oversight of decisions under management’s purview; provision of candid and full information to the Board on all relevant issues; and an open exchange of views between management and the Board on all topics, including the most sensitive ones. These initiatives will be successful only insofar as IMF shareholders uphold the role of the Board as the prime locus of decision making.

The staff’s response

The IMF’s staff share the Independent Evaluation Office’s (IEO) basic diagnosis of the crisis, which is similar to their own assessment, presented in the October 2003 staff paper “Lessons from the Crisis in Argentina.” They also feel that the IEO report takes an important step beyond the staff paper in its detailed examination of how the IMF’s decision-making processes contributed to the course of these events; by doing so, it provides a fresh perspective on IMF governance.

While the staff agree with many of the recommendations and are already acting on them, they do not concur with some of the report’s interpretations and conclusions, some of which depend very much on hindsight. As the report itself notes, for example, it does not examine external influences on the IMF’s decisions, nor does it consider informal channels by which the Board may have been given information by the staff and management. It may therefore understate the information on which the Board’s decisions were based.

An important theme of the report is that the IMF should have taken a step back from the program relationship with Argentina to assess whether the economic policy strategy was on track to achieve its objectives. This is related to the need to strengthen surveillance in program countries, an issue stressed in the 2002 Biennial Surveillance Review. In light of that review, the IMF is taking greater care to ensure that Article IV consultations with program countries pay adequate attention to the issues that are most important from a medium-term standpoint.

A key area in which a more candid assessment of the economic policy strategy would have been desirable in the case of Argentina is the exchange rate regime and its consistency with other policies. The Argentine experience provides a graphic illustration of the need for a more pointed treatment of exchange rate issues in the context of surveillance—notably in staff reports, but also in staff discussions with the authorities and in the Board.

Other Resources Citing This Publication