- Ales Bulir, Marianne Schulze-Gattas, Atish Ghosh, Alex Mourmouras, A. Hamann, and Timothy Lane
- Published Date:
- February 2002
The capital account crises in emerging markets confronted both the affected countries and the IMF with a new set of challenges. The central feature of all these crises was the rapid reversal of capital inflows, bringing about a large and abrupt current account adjustment with pervasive macroeconomic consequences. This presented a striking contrast to the typical situation in which countries turn to the IMF for support, in which macroeconomic imbalances associated with policy slippages result in a more gradual deterioration on the external side, and these imbalances are corrected gradually as policies are adjusted.
Although the reversal of capital outflows was a common feature of all the crises, the factors that account for the reversals differed, and these factors needed to be reflected in the policy response. To varying degrees, concerns about the sustainability of precrisis exchange rate pegs (either formal or de facto) played a role, but the changes in market sentiment that precipitated the reversals were ultimately rooted in the growing evidence of underlying vulnerabilities in public or private sector balance sheets. In the Asian crisis countries, vulnerabilities were rooted in the financial and corporate sectors, associated with the interaction of unhedged foreign currency exposures, maturity mismatches, and asset price bubbles. In Mexico, the risky public debt management strategy was of central importance. In Turkey and Brazil, fiscal policy faced an uphill battle against adverse public debt dynamics. In addition, in all cases, political uncertainties played a role, in part because they undermined confidence that the authorities would be able to tackle the underlying problems.
Once they started, these capital account crises displayed very distinctive patterns of macroeconomic events—which differed from those of other countries that have turned to the IMF for support. The reversal of capita] flows was typically abrupt, and was associated with wide overshooting of the exchange rate. In many cases, dynamic instability associated with self-reinforcing processes came into play: for instance, in Asia, given the unhedged foreign currency exposures, currency depreciation generated widespread insolvency that further undermined confidence and led to more currency depreciation. The crises generally had large effects on economic activity: the recessions were in most cases associated with a collapse of private domestic demand, but there is also evidence, at least initially, of major supply-side effects—attributable, among other things, to the drying up of financing for imported inputs. In most cases, the recessions turned out much more severe than predicted, reflecting the larger-than-anticipated capital outflows—suggesting that beyond a confluence of exceptional factors in each case, the catalytic effect of official financing, on which the programs were predicated, was systematically over estimated. At least in some cases, the recovery of spontaneous market access, bringing reserve accumulation, a recovery in exchange rates, and an easing of the current account, has typically also been relatively swift.
Given the nature of these crises, the principal objective of the IMF-supported programs was to stem the capital outflows and mitigate their effects by addressing the vulnerabilities that had triggered the adverse market reactions and putting in place adequate financing. Macroeconomic policies played an important role, but only to the extent that they could contribute to this. In a setting where macroeconomic imbalances were typically not the root cause of the crisis, the traditional financial programming approach, with its emphasis on the correction of monetized deficits, had relatively little mileage in determining the policy response.
This basic logic was reflected in several aspects of the programs that were implemented—but in each case, there are questions of whether it was reflected enough:
The financing packages were exceptionally large by international standards. They were also exceptionally front-loaded, particularly after December 1997 when the Supplemental Reserve Facility (SRF) was created. The size and front-loading sacrificed some policy conditionality for availability to deal with market forces. This made sense under the circumstances: large financing available up front was essential in restoring confidence and mitigating the effects of the outflows. But the amounts actually available, especially at the early stages of the programs, were nonetheless in most cases small in relation to the total commitments or to the actual (let alone potential) capital outflows.
Attempts at private sector involvement—moral suasion by creditor central banks, agreements among major bank creditors, de facto standstills—were generally modest. Moreover, in most cases, concerted private sector involvement was eschewed at the outset on the grounds that it might exacerbate capital outflows, as well as concerns about possible contagion. In Korea the introduction of private sector involvement was a turning point, setting the stage for recovering stability in the markets, although the direct impact on the capital account was small in relation to overall capital flows.
Structural reforms were an essential part of the solution in many of these cases. The main purpose of these reforms was to address the vulnerabilities that had contributed to the shift in market sentiment. In the Asian crisis countries, the programs sought both to address the financial system weaknesses that were at the heart of the crisis and to clean up the wreckage of insolvency that had resulted from the exchange rate and interest rate movements associated with the crisis itself. In Mexico, reforms were needed to restore the health of the banking system, although they were not included in the IMF-supported program and were mainly addressed with the help of bilateral assistance. In Brazil, the primary objective of structural reform was to underpin fiscal sustainability—and, indeed, doubts about the authorities’ commitment to undertake these reforms was a major reason for the loss of market confidence. But, notably in the case of Indonesia, the structural reform agendas were not sufficiently well focused, especially in the early stages of the crisis.
Given the large exchange rate movements associated with these crises, which significantly overshot any reasonable estimate of equilibrium levels, monetary policy faced a challenge of maintaining or regaining nominal stability. Monetary policies were thus tightened at some stages in all of these cases, but in many cases with significant hesitation. (In the case of Indonesia, monetary tightening was preceded by a period of several months of unbridled money and credit expansion with deeply negative real interest rates, in the context of a collapsing banking system.) Once monetary policy was tightened in a determined way, market conditions stabilized, and the period of high real interest rates was typically short-lived; nominal stability was preserved or restored. But at the same time, in most of these crises, monetary policies were handicapped by the loss of a nominal anchor, as they were driven off a previous exchange rate peg: Brazil responded by introducing formal inflation targeting soon after floating the real, but in other cases the authorities relied on a less formal price stability objective.
Fiscal policies played a different role in different countries. In some cases—such as in Brazil, Argentina, and Turkey—establishing fiscal sustainability in the face of adverse debt dynamics or a fragile public debt structure was essential to build confidence. In other cases, notably the Asian crisis countries, fiscal sustainability was not a major issue, and efforts at fiscal adjustment (which were short-lived in Indonesia and Korea) were unsuccessful in boosting confidence. Taken together, the evidence suggests that fiscal adjustment is an essential part of the solution if, and only if, fiscal unsustainability is a major part of the problem to begin with. Fiscal adjustment that is unnecessary from a medium-term perspective is unlikely to have a favorable confidence effect. Finally, the macroeconomic analysis in the paper suggests that the crises were characterized by a shifting mix of supply and demand shocks, as balance sheet imbalances worked themselves out; in this setting, the scope for fiscal fine-tuning may be limited.
Markets proved unforgiving of any lapses in program implementation or policy indecision. In a number of cases, firm political commitment to the reform process crystallized only after initial hesitations resulted in renewed bouts of capital outflows.
In reviewing these crises, it is difficult to see how they could have been managed without considerable pain once they had broken out. The crises were characterized by an over-adjustment of external current accounts in relation to what was needed for any reasonable means of sustainability. This over-adjustment was associated with severe macroeconomic disruptions. To avoid these consequences, substantially more financing—in the form of private sector involvement, official packages, or both—would have been needed. The scope to deliver such financing is quite limited, however. It is difficult to contemplate official financing packages several limes the size of recent ones, while extensive work on private sector involvement suggests that, short of draconian measures that could jeopardize a country’s market access for years to come, there is no simple way to stop the exit of capital once a crisis breaks. Moreover, even if available financing were abundant, resolving the underlying balance-sheet disequilibria would still involve some dislocation. It would be far better, if at all possible, to address these imbalances before a crisis breaks out. In short, these crises make a compelling case for improving prevention.
Beyond the importance of crisis prevention, the experience of these countries suggests a number of lessons for program design in the context of high capital mobility—such as the appropriate roles for monetary, fiscal, and structural policies. Yet their experience also underscores the difficulty of robust program design when stock imbalances can trigger sudden and massive capital outflows.
The capital account crises that broke in the mid and late-1990s have often been termed the first of the “twenty-first century” crises; no doubt they will not be the last. They posed particular challenges for the provision of financing and for the design of macroeconomic and structural policies. In many cases, they also severely tested the political commitment of the authorities to be resolute in undertaking reforms. Whether their lessons for crisis management and program design have been learned remains to be seen.