Exchange Rate Regimes in Developing Countries and Emerging Markets

Jeromin Zettelmeyer

Over the past two to three years, there has been a resurgence of interest in exchange rate regimes for developing and emerging market countries. In the early 1990s, attention focused mainly on stabilization from high levels of inflation and on the right regimes for the transition economies in Eastern Europe and the former Soviet Union. The emerging market crises of 199598 completely changed this focus. Sustainability and crisis prevention began to be viewed as key criteria for judging exchange rate regimes. Much more attention was directed at the “hardness” of alternative pegged regimes. The standard dichotomy of floats versus pegs was replaced by a three-way distinction: floats, intermediate regimes, and very hard pegs. More attention was also given to the distinction between exchange rate regimes “de jure” and “de facto,” as revealed by ex-post behavior. Two main areas of research have arisen in this new context. The first concentrates on the sustainability of intermediate regimes, while the second deals with comparing and evaluating alternative regimes. This survey briefly traces the debate since 1997, and discusses some recent research conducted at the IMF.

In the aftermath of the Asian crisis, two propositions about exchange rate regimes attracted widespread support. The first asserted that either “extreme” (floats or hard pegs) was preferable to the “middle” (soft pegs, bands, and crawling pegs), largely on the grounds that intermediate regimes were hard to sustain and more crisis-prone than either extreme.¹ Prior to the mid-1990s, the mainstream literature had basically argued the opposite: in essence, because satisfying several objectives—flexibility versus commitment, inflation stabilization versus competitiveness, and insulation from monetary shocks versus insulation from real shocks—inevitably pointed to compromise solutions between hard pegs and pure floats.² The second proposition argued that, given the choice between the two extremes, floating exchange rates were best for most emerging market economies, with hard pegs reserved for “unusual circumstances.”³ Underlying this was the belief that only very hard pegs were sustainable, that is, currency unions and currency boards supported by a strong national consensus, and, as such, unfeasible or overly constraining for many emerging market countries.

What appeared to be a new consensus did not last long, however. First, several authors reasserted the case that “intermediate solutions are often more likely to be appropriate than corner solutions.”⁴ Second, the feasibility of floating was questioned by several authors, particularly with respect to emerging markets in Latin America. This “fear of floating” school argued that, because of credibility problems, worries about inflation pass-through, and dollarization in the domestic financial system, central banks sharply curtailed movements in the exchange rate even when officially floating. This resulted in “flexible regimes that are managed as if they were fixed, but without the benefits of precommitment.”⁵

Two recent survey papers from the IMF can be viewed as proposing compromises between the initial “bipolar” view and arguments in favor of intermediate regimes.⁶ Fischer (2001) differs from the original bipolar view in two respects: first, by recognizing that developing countries which are not very exposed to international capital flows still face a wide range of intermediate exchange regime options; second, by defining flexible arrangements suitable for emerging markets more broadly to include “crawling bands with wide ranges.” Mussa and others (2000) go somewhat further in defending intermediate regimes by making a distinction between the sustainability and the desirability of pegs. The fact that soft pegs may not be sustainable for many countries need not imply that they cannot play a useful role for a limited period of time (for example, as a nominal anchor during stabilization from high inflation). Of course, this assumes that these countries find a way to safely “exit” from the peg without a crisis.⁷

As an empirical matter, do intermediate regimes show signs of “vanishing”? Fischer (2001) presents evidence for emerging markets and developing countries indicating that the proportion of countries with intermediate regimes has indeed declined during the period 199199. But is this trend likely to continue until intermediate regimes become extinct? This question is addressed by Masson (2000), who estimates Markov transition probabilities between the three states (hard pegs, floats, and intermediate regimes) for various periods and sets of countries.⁸ His evidence suggests that intermediate regimes will continue to exist in steady state or, at least, take a very long time to disappear.⁹

No compromise seems to have developed, so far, in the central debate on whether floating regimes or hard pegs are preferable for the majority of emerging market economies. There is by now a substantial body of empirical work, however, that sheds light on specific aspects of the debate.

• Studies that link exchange rate regimes and economic performance tend to find favorable effects of pegs, particularly hard pegs. Ghosh and others (1995, 1997) find that inflation tends to be lower under pegged regimes (which are defined to include both soft and hard pegs), while output volatility is higher under pegs.¹⁰ Ghosh, Gulde, and Wolf (1998, 2000) find that currency boards exhibit better inflation performance than soft pegs, mostly due to a credibility effect.¹¹ Regarding growth, both currency boards and floats seem to do better than soft pegs. A separate set of papers finds no systematic evidence linking pegs and banking crises; if anything, hard pegs seem to be associated with a lower probability of banking crises than both floats and soft pegs.¹²

• Full dollarization is likely to imply a further credibility effect and lower real interest rates.¹³ In addition, there is evidence that currency unions have very large positive effects on trade.¹⁴

• Floats, on the other hand, really do seem to have some of the insulating properties that are traditionally attributed to them. Borensztein, Zettelmeyer and Philippon (2001) find much smaller reactions in domestic interest rates, to both U.S. interest rate shocks and shocks to emerging market risk premia, for Singapore (managed float) than for Hong Kong (currency board).¹⁵ A similar comparison between Mexico and Argentina is not as clear cut; in particular, interest rates in both countries seem to react to common shocks to emerging market risk premia with about equal force. However, this is not easily attributable to a “fear of floating” because the Mexican exchange rate also shows a very large reaction to the same shocks.¹⁶

In summary, intermediate regimes—including soft pegs, bands, and crawling pegs—are unlikely to disappear, and they remain appropriate for a wide range of developing countries without large exposure to international capital flows, and as temporary regimes. When comparing soft pegs with more flexible arrangements, the latter seem to do better on growth and real volatility, while the former may be useful for stabilization purposes. As permanent arrangements for emerging markets, however, the choice is between floats and very hard pegs. How countries resolve this choice depends on how they trade off the benefits of some monetary autonomy and perhaps reduced real volatility on the one hand, and credibility, commitment, and integration on the other.

Systemic Banking Crises: Causes, Consequences, and Policy Lessons

Enrica Detragiache

The last two decades have seen a proliferation of systemic banking problems around the world. Banking crises have threatened macroeconomic stability through their effects on monetary control, the fiscal consequences of bank rescue packages, and the knock-on effects on capital outflows and the external balance. Banking sector problems have quickly assumed a prominent role in the IMF’s operational work of surveillance and technical assistance, and have often posed difficult challenges to the design and implementation of IMF-supported programs. These developments have stimulated a large body of IMF research.

Recent research at the IMF on banking sector distress encompasses a variety of issues and methodologies. Some work remains close to operational experience, and approaches policy questions in a pragmatic way, while other work is more abstract, and arrives at policy recommendations through modeling and econometric analysis. In the first category are several case studies that review the causes and effects of banking crises and the policy responses.¹ These papers show that, while common elements are present, every banking crisis is different, and policymakers will always have to deal with unexpected circumstances.

Another group of studies asks more specific questions that can be addressed through econometric analysis. Gonzáles-Hermosillo and others (1997) study the determinants of the 1994 Mexican crisis using bank-level data. Ramos (1998) and Schumacher (2000) explore the behavior of Argentine banks during the Tequila episode.² An issue that has received special attention is the “credit crunch,” namely whether the decline in bank credit observed after a crisis is due to bank financial distress or to weak demand.³ Using aggregate data, Pazarba¸sioglu (1997) finds some evidence of a credit crunch in Finland in 199192, while Ghosh and Ghosh (1999) reject the hypothesis for the recent Asian episodes. A negative result is also reported by Woo (1999), who finds that the shortage of capital affected credit supply in Japan only in one year. Using firm-level data for Korea, Borensztein and Lee (2000) show that bank lending policies reflected financial distress at the corporate level.

While each banking crisis is unique, case studies often point to a number of similar factors as proximate causes of distress. Cross-country econometric studies of banking crises test which (if any) common factors can be identified.⁴ Using a multivariate logit econometric model, Demirgüç-Kunt and Detragiache (1998, 1999, 2000) find that low output growth, high interest rates and inflation, and rapid credit expansion tend to be associated with banking crises. Also, financial liberalization and generous deposit insurance, when accompanied by poor institutions to regulate and supervise banks, tend to contribute to fragility. Hussain and Wilhborg (1999), using evidence from six Asian countries, show that bankruptcy procedures favoring corporate debt restructuring lead to a speedier crisis resolution. Gonzáles-Hermosillo (1999) integrates macroeconomic variables with bank-level data in a sample consisting of three regions in the U.S., Mexico, and Colombia, and shows that indicators of individual bank weakness are useful for predicting systemwide crises.

While most cross-country empirical papers focus on the determinants of crises, Demirgüç-Kunt and others (2000) study how the economy and the banking sector are affected by distress.⁵ Using both macroeconomic and bank-level data, they find that depositor runs are not a typical feature of banking crises, and that output growth recovers quickly while credit remains depressed for a longer period of time. Also, there is evidence that banks, even the healthier ones, reallocate their asset portfolios away from loans and into safer assets.

Research on banking crises at the IMF has also tackled a number of questions at a theoretical level.⁶ Inspired by the Asian crises, Chan-Lau and Chen (1998) develop a model in which a small change in fundamental variables may induce banks to stop monitoring their customers, triggering a collapse in financial intermediation. Challenging conventional wisdom, Cordella and Levy-Yeyati (1998, 1999) show that forcing banks to disclose their risk exposure does not necessarily increase bank stability, and that bank bailouts, by increasing bank charter value, may reduce excessive risk-taking. Detragiache (1999) shows that, when the bank deposit market becomes more integrated internationally but significant barriers remain on the lending side, bank fragility will increase, and the economy may be worse off. Huang and Xu (2000) analyze the conditions under which the interbank market may break down, causing a systemic crisis, while Goodhart and Huang (2000) develop a model of an international system under a fixed exchange rate. They argue that although an international interbank market can improve liquidity sharing among all the banks, it also leads to international financial contagion, which can be contained by an international lender of last resort.

Policy responses to banking crises have also been a focus of IMF research.⁷ This work, which typically draws more from practical experience than from theoretical or empirical results, has either offered general evaluations of the set of policies available and the trade-offs involved, or focused on specific policies. Among the latter, Woo (2000) evaluates asset management companies versus decentralized corporate workouts, while Garcia (2000) discusses the introduction of temporary blanket deposit guarantees during a crisis.