Abstract

In a world of increasing capital mobility and broadening and more diversified trade, many (but not all) developing and transition economies are likely to find it desirable to move from relatively fixed exchange rate regimes to regimes of greater exchange rate flexibility. This paper suggests why, and considers strategies that countries may consider for such a move. It reinforces this discussion with a review of experience from teh past two decades with alternative exchange rate regimes. The paper also identifies policies that can facilitate the transition to greater exchange rate flexibility for countries that wish to pursue this option.

Appendix I Criteria for Exchange Rate Regime Choice

There is an extensive literature on the choice of desirable and feasible exchange rate regimes, which includes a discussion of “optimum currency areas,” the choice between fixed and flexible regimes, and money versus exchange-rate-based stabilization. The focus here is on a short practical guide to exchange rate regime choice, rather than an exhaustive survey of the literature. Moreover, it needs to be recognized that there is no agreement on how precisely to quantify the various criteria, nor, to the extent that they conflict, on which should take priority.

Table A1 summarizes the implications for exchange rate arrangements of a number of criteria that have been discussed in the literature. Though not exhaustive, the set of exchange rate regimes listed at the top spans the range from flexible to perfectly fixed, with some overlap between managed flexibility, target bands, and crawling pegs. The criteria listed on the left are thought to influence positively (though not necessarily be prerequisites for) the success of the exchange rate regimes indicated with a dot. Conversely, the criteria are considered not to be compatible with successful operation of the other exchange rate regimes. For instance, high inflation is likely to be compatible only with flexible exchange rates or a crawling peg, while low inflation would permit a choice of any of the exchange rate regimes (but presumably make a crawling peg unnecessary).

Table A1.

Criteria for Choice of Exchange Rate Regime

article image

While a number of the criteria discussed below are intended to refer to structural characteristics of an economy, it has been pointed out that even these criteria are to some extent endogenous.26 For instance, a credibly fixed rate may lead to a structural break with inflation inertia, while the extent of trade and other links with partner countries may also adapt to joint exchange arrangements. As a result, countries that initially did not seem to be candidates for a fixed exchange rate arrangement would after a certain amount of time score highly on the relevant criteria.

An important factor in the choice of exchange rate regime not stressed by the traditional literature nor captured by the usual economic indicators is the objective function of the authorities, in particular as concerns the trade-off between a desire to control inflation (that is, provide a nominal anchor) and to limit fluctuations in competitiveness (and minimize output losses).27 Hence, two economies with the same structural features may (optimally) choose different exchange rate arrangements. Moreover, changes of government may involve different preferences concerning the trade-off of the two objectives, and hence a change of exchange rate regime.

In evaluating the predictive power of the criteria, two other points also need to be recognized. First, sound fiscal and monetary policies are a prerequisite for any exchange rate arrangement to function well (though a floating rate regime cannot “break down” in the same way as a fixed rate arrangement—its success is judged by reasonable stability in financial markets and good macroeconomic performance). So “optimality” cannot be judged independently of those policies.

Second, sometimes a stabilization program requires a regime change, to make a break with the past, so that questions of optimality are secondary. For instance, if the current regime lacks credibility and risks degenerating into a situation of accelerating inflation, a radical monetary reform—such as instituting a currency board—may be desirable.

Prevailing Rate of Inflation

A basic criterion for a fixed exchange rate is acceptance of a rate of inflation similar to that in the partner country. Too high (and persistent) inflation constrains the ability of a country to maintain an exchange rate peg against a stable (in the sense of low inflation) currency or basket of currencies. Hence, tolerance for inflation would make an extended commitment to a fixed peg or to narrow (fixed) exchange rate bands infeasible. The prevailing rate of inflation (averaged over a certain number of years) may give an idea of the tolerance for inflation.

However, there is evidence that at hyperinflation or very high rates of inflation, inflation inertia is quite low, permitting the use of an exchange rate peg to achieve rapid disinflation without significant costs in economic activity.28 In these circumstances, a pegged rate is likely to be a useful transitional policy, as part of an exchange-rate-based stabilization. In countries with high and persistent inflation (roughly above 30 percent a year), the use of the exchange rate as a nominal anchor has been somewhat less successful, due to inflation inertia and low credibility. Even with strong fiscal adjustment, exchange-rate-based stabilizations in these countries often have led to significant real appreciation and large current account deficits. Also, these experiences often have been characterized by an initial boom, led by consumption spending, followed by a recession associated with the cumulative effects of the real exchange rate appreciation. This contrasts with money-based stabilizations, which typically cause a recession early during the disinflation.

Whether a fixed exchange rate policy designed to stop inflation proves durable depends partly on the extent that the tolerance for inflation and residual inflation inertia has been eliminated. If not, then it is unlikely that inflation could be durably reduced to a level close to that of partner countries, and, as a result, a fixed peg would eventually lead to overvaluation and the need for more exchange rate flexibility. In these circumstances, adoption of a crawling peg and relatively wide bands may be appropriate to avoid significant and continuing real appreciation.

Level of Reserves and Capital Mobility

Without adequate reserves, the monetary authorities may not be able to maintain an exchange rate peg or a target band unless monetary policy (in particular, domestic interest rates) is clearly subordinated to the exchange rate objective. The level of reserves interacts with the extent of capital mobility to influence feasibility. High capital mobility makes it difficult to maintain a regime that is intermediate between the two extremes of floating rates and currency boards, unless stringent requirements of similarity of structures of the economies concerned and convergence of their policy objectives are satisfied. 29 Conversely, if reserves are not very large or the central bank is unwilling or unable to move interest rates in response to exchange rate pressures, then even with only moderate capital mobility it may be impossible to maintain some sort of pegged exchange rate arrangement.

Clearly, the extent of capital mobility and the availability of instruments to the private sector to bet against official exchange rate targets has greatly increased in recent years. As a result, as argued by Obstfeld and Rogoff (1995), exchange arrangements intermediate between complete flexibility and complete fixity have become less credible and more fragile. Nevertheless, in most currency crises industrial countries have had sufficient reserves to fully back the monetary base, and hence could have defended pegged rates (for instance, in the EMS in 1992–93) had they been willing to incur the cost that the resulting higher interest rates would have inflicted on the economy. Moreover, several smaller economics with a history of subordinating their monetary policies to a peg to the deutsche mark have long avoided any exchange rate pressures (Austria, the Netherlands). So the question of feasibility is not an absolute one, but also interacts with other considerations. It is also the case that greater capital mobility can add to the instability of flexible rates, leading to periods of large misalignments.

Extent of Labor Mobility and Nominal Flexibility

For a given level of wage or price stickiness, the costs of loss of flexibility associated with adopting a more rigid exchange rate arrangement would be lower, the greater the extent of labor mobility within a country and between partner countries. This is the fundamental argument of the seminal paper on “optimum currency areas.30 Otherwise, an asymmetric shock reducing the demand for labor in a country or region would lead to an increase in unemployment there. Labor mobility would instead permit (in principle) the workers affected by the fall in demand to migrate to a region where demand was higher.31 In the absence of labor mobility, the exchange rate could change relative prices in a direction of reducing the negative effects of shifts in demand.

In the polar case of flexible wages and prices, exchange rate flexibility would provide no additional policy instrument, since movements in relative prices would occur costlessly. In this case, there would be no reason not to fix the exchange rate, since it would impose stability of nominal magnitudes without leading to a loss of real flexibility. Thus, flexibility of wages and prices would provide a substitute for exchange rate flexibility. Flexibility is difficult to measure, but McKinnon (1963) argues that openness increases the responsiveness of wages and prices to the exchange rate and reduces the scope for using a devaluation to help labor to accept a cut in real wages. Hence, greater openness would argue for greater exchange rate fixity.

Production and Export Diversification

More generally, a country might want to avoid a peg, whether to a single international currency like the U.S. dollar or to a basket of currencies, if its production and exports were not diversified. Diversification makes the country less vulnerable to terms of trade shocks, and hence less likely to need exchange rate flexibility. In contrast, countries heavily reliant on a few commodities may need the flexibility of a floating rate to respond to changes in world prices in order to mitigate spillovers into nonresource sectors.

Assuming that a pegged rate were desired, the choice of a single currency peg rather than a basket would depend on how concentrated were trade (and capital account) links with a particular country and the similarity of their productive structures. For a country whose trade was diversified regionally, a basket peg would make the country less vulnerable to movements in cross rates, which would be important if trade links and production similarity vis-a-vis a particular country are not strong.

Fiscal Policy Flexibility and Sustainability

Fiscal flexibility is more important for exchange arrangements that incorporate more fixity, since the exchange rate instrument cannot be used as a shock absorber. This has two aspects. First, a country with a pegged rate may need to allow its own fiscal policy to respond more flexibly (including allowing automalic stabilizers to operate) in the face of shocks to the domestic economy. Second, in a monetary union it may be desirable to cushion shocks by using fiscal flows across countries in the union, for instance through a federal fiscal system of taxes and transfers.

Fiscal policy sustainability is also important for the survival of a fixed rate system. High public debt not only reduces the flexibility of fiscal policy but also increases the risk of default, a balance of payments crisis, and recourse to monetizaiion of fiscal deficits. High debt makes defense of a peg more costly, involving higher deficits and still higher debt, thereby increasing the risk that speculative attacks may be self-fulfilling. Therefore, the ratio of government debt to GDP may be an indicator of vulnerability of a pegged rate.32

Extent of Trade and Political Integration

The higher the trade integration with partner countries, the greater the benefits of a fixed exchange rate or common currency. The higher the level of international trade between two countries, other things being equal, the greater the reduction of transactions costs (including costs of hedging) that a regime with a high degree of fixity in the two countries’ bilateral exchange rate would bring. In addition, fixed rates, by making comparison of prices more straightforward, should increase their information content and increase economic efficiency. Such exchange rate arrangements could involve pegged rate systems of various types or go so far as the creation of a currency union. The latter needs to involve a high degree of common interests among the countries concerned, that is, a form of political integration, because otherwise the use of a common currency is unlikely to be durable, and hence not credible.

Symmetry and Type of Shocks

If shocks facing two countries are very different, then a fixed exchange rate linking their currencies may not be desirable. Thus, asymmetric shocks are likely to be a problem for pegs or currency unions, The type of shock facing the two economies is also important. A preponderance of domestic nominal shocks, in particular to money demand, suggests that more exchange rate fixity would be desirable, allowing anchoring to a stable foreign country’s price level. In contrast, in the face of real shocks (like the terms of trade shocks mentioned already, but also including domestic real shocks such as those to productivity) the economy would benefit from exchange rate flexibility, which would ease adjustment costs in the presence of stickiness of wages and prices.

The extent of dollarization of the economy wilt influence the exposure to monetary shocks. The associated currency substitution will make the demand for domestic money more interest elastic and sensitive to expectations of future changes in monetary policy. In these circumstances, domestic money demand shocks may dominate real shocks, and, as suggested above, a fixed exchange rate relative to the foreign currency (the “dollar”) that is used domestically may be appropriate.

Noneconomic Criteria

As has been pointed out by others, the optimum currency area criteria seem to have relatively little predictive power.33 Though they may have prescriptive implications, they may not well explain actual choices among exchange rate arrangements. Therefore noneconomic factors may dominate. In particular, lack of central bank credibility may make it difficult to attain low inflation on a sustained basis by running an independent monetary policy, making it attractive to “borrow” the credibility of a partner country central bank through an exchange rate peg,34 a currency board, or the abandonment of the domestic currency in the context of a currency union. The latter would be more attractive to the extent that the country concerned retained some influence over the common monetary policy in the context of some framework for political integration. Conversely, a floating rate may be chosen for noneconomic reasons, for instance as an assertion of monetary sovereignty.

Appendix II The Experience with Exits

This appendix summarizes historical experience with exits from currency pegs. It documents past experience with exits, establishing that these have typically been associated with exchange market crises and adverse macroeconomic outcomes.

It should be emphasized that no attempt is made here to isolate the effects of exits. Countries exiting from currency pegs have typically differed from other countries in a variety of other respects besides their decision to change their exchange rate arrangement. The fact that exits have typically taken place in an environment of crisis suggests that such countries may have suffered from an adverse shock or have been running unsustainable policies, either of which can influence the post-exit behavior of macro-economic and financial variables. Inevitably, simple comparisons of countries exiting from currency pegs with other countries reflect more than the effects of the exit, narrowly defined. Therefore, the goal here is more limited. It is to document the historical record with exits and thereby to motivate the discussion in the text of the need to identify better ways of undertaking them.

To this end, a comprehensive list of exits from currency pegs by developing countries was assembled, and a variety of macroeconomic and financial indicators were analyzed around the time of the exit and compared with those for control groups of nonexisting countries. Exits are defined as movements from a (single currency or basket) peg to a more flexible exchange rate policy.35 Since the problem of devising an exit strategy is particularly pressing for developing countries operating in an environment of international capital mobility, the sample was limited to the last two decades (generally, 1977–95, although for some variables the data end in 1992) and to countries considered in the World Bank’s World Development Indicators and Global Development Finance data tables.36 Short-lived experiences were eliminated (to be categorized as an exit, the exchange rate peg regime and the subsequent period of flexibility each had to last at least two years),37 along with unrepresentative economic and political cases (specifically, cases of hyperinflation and civil war where the behavior of macroeconomic and financial variables was so extreme that they would have dominated the results).

This left a total of 29 exits in which countries moved from single currency pegs or basket pegs to managed exchange rates or an independent float.38 Table A2 summarizes the exit cases that we were able to include in the analysis of each of the variables examined. Table A3 summarizes the countries in the sample, the size of any devaluation both before and after the exit, the nature of the exchange rate regime before and after the exit, and the date of exit.39 (See Figure A1) The relatively small number of exit cases is reason for regarding the results with caution. In addition, the relatively small number of exits in the sample prevented detailed comparisons of countries that exit in an orderly fashion and countries that exit as a result of a crisis.40

Table A2

Data Included for the Analysis of Exit Cases

article image
Table A3

Exit Cases in the Sample

article image
Notes: A = currency peg. B = basket peg, C = managed exchange rate, and D = independent float. Primes denote peg with occasional devaluations.
Figure A1

It is, however, possible to compare countries exiting from a peg with two control groups of countries: those that continued to peg without exiting, and other developing countries in the World Bank database aside from our exit cases.

Continuing peggers correspond to countries with lasting pegged exchange rate regimes between 1977 and 1996 (we accepted countries with occasional adjustments of the level of the peg). By examining the IMF’s sources described above, we identified 51 such countries (of which 36 were already in Frankel and Rose’s (1996a) database). For five of the variables examined, we included information for the period 1977–95: output growth, inflation, real interest rate, export value growth, and import value growth. For the remaining variables, we only included information for the period 1977–92.

Nonexits correspond to the data available in Frankel and Rose’s (1996a) database for the period 1977–92, excluding the exit cases in the above sample, and adding data for 1993–95 for five of the variables under examination: output growth, inflation, real interest rate, export value growth, and import value growth.

Figures A2A5 summarize the behavior of the relevant variables centered on the month or year of the exit.41 In each case, the observations for the exit cases are surrounded by two-standard-deviation bands for the estimated average.

Figure A2.
Figure A2.

Exchange Rate Indicators

(Centered on the month of exit)

1The base of the index is 100 for the month prior to exit.2Volatility is measured as the standard deviation of the monthly growth rate of the exchange rate over the last three months, averaged across exit cases.
Figure A3.
Figure A3.

Macroeconomic Indicators

(Centered on the year of exit)

Figure A4
Figure A4

Exchange and Trade Restriction Indicators

(Centered on the year of exit)

1The restriction index for a given country takes the value of one if there are payment restrictions and lero if there are no payment restrictions (as reported in International Monetary Fund. Annual Report on Exchange Arrangements and Exchange Restrictions)
Figure A5
Figure A5

External Debt and Other Indicators

(Centered on the year of exit)

The exchange rate indicators in Figure A2 suggest that exits are typically preceded by gradual nominal and real appreciation and followed by a step depreciation (and in the case of the nominal exchange rate by some further depreciation over time). Thus, though the definition of exits includes in principle both cases of appreciation and depreciation, in fact, the latter have dominated (see Table A3). Real and nominal exchange rate volatility jump up around the lime of the exit. Both remain an order of magnitude higher than before for several months after the event.42

Figure A3 suggests that exits have not been happy events. Typically, economic growth has slowed in the period leading up to the exit. In the year of the event, growth is actually negative, and significantly below that in both the nonexit cases and in countries with lasting pegs. The growth of exports (measured in current dollars) also slows, falling significantly below that in both control groups in the year preceding the exit. Subsequent to the event, output and exports recover.43 Logically, exports respond first to the change in the exchange rate; in contrast, output growth only begins to revive in the second post-exit year.

In the case of inflation and money growth, the choice of control group is important. Both variables are typically higher both prior and subsequent to exits than in countries with lasting pegs, since countries with lasting pegs have by necessity brought their inflation and money growth rates down to the levels prevailing in the industrial countries to which they peg. In contrast, inflation and money growth are significantly lower around the time of exits than in all nonexit cases, the alternative control group being dominated by high-inflation countries with floating rates. These contrasts suggest that countries that have exited from pegs have typically been less successful in bringing inflation and money growth down to the levels of the industrial countries to which they peg than have been countries with lasting pegs. But countries that have opted for pegs, temporary as well as lasting ones, have a preference or an ability to maintain relatively low rates of inflation. In other words, although inflation generally rises following an exit, countries that exit from pegs continue to run lower rates of inflation even after their move to greater flexibility than countries with longtime floating rates.

Three additional macroeconomic indicators, the government budget deficit, the current account deficit, and international reserves, behave as predicted by models of balance of payments crises: countries that exit have relatively large budget and current account deficits in the years leading up to the exit, and they tend to lose international reserves (measured in months of imports). But neither the budget, nor the current account, nor the level of reserves differs significantly from those for peggers or those for the entire group of nonexiting countries. Similarly, an indicator of the condition of the banking system (liquidity as measured by the ratio of liquid reserves to total assets) is little different in the exit cases from either control group of countries.44

The panels for the incidence of current and capital account restrictions, also in Figure A4, are striking. Countries that exit from pegged rates have a significantly lower probability of having maintained capital account restrictions in the period leading up to the event. This result should be regarded with caution, for there is reason to worry that the information on which it is based is subject to serious limitations.45

But we cannot rule out the possibility that countries that have undergone capital account liberalization tend to voluntarily opt for more flexible rates in order to better manage their exposure to international capital flows.46 Another interpretation is that countries with open capital accounts are more susceptible to being forced off their pegs.47

Figure A5, which considers various indicators of the external position, is consistent with this interpretation that exits historically have been forced on countries with a fragile external position. Countries that exit from their pegs tend to have more debt than the control group cases and to receive less inward direct foreign investment. They tend to have more variable rate debt than countries with lasting pegs.48 Thus, the composition of the capital account suggests external fragility on the part of countries that are forced or choose to exit from their pegs.

On average, then, exits from pegged exchange rates have not occurred under favorable circumstances. They have not had happy results in the short run. Countries have generally waited to exit until reserves are falling. Exits have been associated with disappointing economic growth, both before and after the fact, and with disappointing export performance in the preceding period.

Appendix III Country Cases

Chile

For over a decade Chile’s exchange rate policy has followed a relatively smooth process of gradual adaptation to changing conditions (Table A4). This process began after a turbulent period in the early 1980s, characterized by a balance of payments crisis, a severe recession, and a collapse of the banking system.49 A 15 percent devaluation in June 1982— which marked the end of an exchange-rate-based stabilization that failed to achieve rapid convergence of domestic inflation to industrial country levels— was followed by several turns in exchange rate policy: after the devaluation, the peso was linked to a basket of currencies and an exchange rate schedule covering the following 12 months was announced, only to be abandoned in August 1982, when the peso was allowed to float. A crawling peg was adopted in September of that year, but it was accompanied by several discrete devaluations. In June 1985, Chile adopted a system of crawling exchange rate bands.50

Table A4.

Chile: Foreign Exchange and Financial Systems

article image
Sources: Alexander and others (1997); International Monetary Fund, Staff Country Reports, Annual Report on Exchange Arrangements and Exchange Restrictions, and International Financial Statistics; Alexander, Balino, and Enoch (1995); Goldstein (1996, 1997); Goldstein and Turner (1996); Johnston and Pazarbasioglu (1995); Johnston, Darbar.and Echeverria (1997); Lindgren, Garcia, and Saal (1996); and various news sources.

The crawling band system remains in place today, although it has been modified gradually in response to a changing external environment and a shift in policy objectives. In general terms, the system consists of a central parity that is adjusted in line with the difference between past inflation and an estimate of world inflation, and parallel bands around the central parity. Band width was set at ±2 percent originally, and it has been widened several times since then.5l In November 1995, a real appreciation of 2 percent was built into the formula to compensate for faster productivity growth in Chile than in trading partners. In addition, on several occasions there have been discrete devaluations and revaluations of the central parity.

Two different periods can be distinguished in Chile’s experience with exchange rate bands: 1985–89 and the period starting in 1990.52 During the first period, exchange rate policy was largely aimed at restoring the losses of competitiveness accumulated since the late 1970s, in the context of a tight external constraint. Through several realignments (devaluations), widening of the band on two occasions, and with the exchange rate positioned near the upper (weaker) limit of the band most of the time, the system of crawling bands permitted the achievement of a 33 percent real depreciation in effective terms between 1985 and 1988.53 However, the pursuit of a competitive real exchange rate prevented a reduction of inflation to single digits, and inflation fluctuated between 15 percent and 30 percent (Figure A6). Control over the inflation rate was not lost mainly because Chile managed to maintain a very tight fiscal stance. A deficit of 3 1/2 percent of GDP recorded in 1985—largely the result of the crisis of previous years—was reduced gradually, and by 1988 Chile was recording sizable budget surpluses and had managed to reverse an upward trend in the external public debt ratio to GDP. At the same time, the current account deficit was lowered from 8 1/2 percent of GDP to near balance. In 1988, in anticipation of a national referendum—and subsequently of the first elections to be held in Chile in nearly two decades—macroeconomic policies were relaxed and, as a result, inflation picked up, the real exchange rate appreciated, and the current account deficit widened for the first time in five years.

Figure A6
Figure A6

Chile: Selected Macroeconomic Indicators

1In percent of GDP2Index 1990=100

The second period is characterized by a very different external environment and the reorientation of policy objectives. Chile no longer had a current account problem and, in the second half of 1990, a long period of limited capital mobility came to an end abruptly as the country resumed access to international capital markets. International investors, who were beginning to devote attention to emerging markets, found Chile attractive, especially given its high domestic interest rates, which reflected an important change in policy objectives. In 1989 the central bank had been granted legal independence and had been entrusted with the explicit objective of achieving price stability. Moreover, unemployment had declined steadily during the late 1980s and inflation was rising. In this context, the government gave increasing attention to the goal of lowering inflation and relatively less to the real exchange rate. To achieve these objectives white preserving the crawling band system, the authorities resorted, in addition to interest rate hikes, to discrete appreciations, widening of the bands, extensive sterilized intervention, and the use of capital controls.54 Although these policies did not prevent a real appreciation, they did succeed in reducing inflation to single digits while at the same time preventing the real appreciation warranted by market developments from leading to perverse dynamics in the current account.

As it was the case in the first period, it appears that a sound fiscal policy (budget surpluses were recorded throughout this period) played an important role in helping reconcile apparently conflicting objectives in the second period. An appropriate supervisory framework, implemented as part of the comprehensive restructuring of the banking system that took place in the 1980s, also ensured the smooth functioning of the banking system after 1990.

Israel

Israel fixed its exchange rate against the U.S. dollar as part of its July 1985 stabilization plan (Table A5).55 Since then, the exchange rate system has evolved gradually toward more flexible arrangements, to accommodate the authorities’ objective of providing the economy with a nominal anchor while at the same time preventing sizable appreciations of the real exchange rate that could lead to balance of payments problems. The currency was devalued in January 1987 and December 1988. In January 1989, the central bank introduced a ±3 percent horizontal band around the central parity. The parity was devalued four times between June 1989 and March 1991, and the band was widened to ±5 percent in March 1990. On December 17. 1991, following another discrete devaluation, a system of diagonal or crawling bands was introduced, with the slope of the bands determined as the difference between a target level for domestic inflation and a forecast of world inflation. Band width was originally kept at 5 percent, but was widened to ±7 percent on May 31, 1995, and in June 1997 by adjusting the limit on depreciation and the slope of the lower (appreciated) edge of the band. A small discrete devaluation was often introduced at the time of the announcement of the parameters of the new band.

Table A5.

Israel: Foreign Exchange and Financial Systems

article image
Sources: Intemstional Monetary Fund. Staff Country Reports and internationel financial Sttatics: Goldstein (1997); Lindgren, Garcia, and Saal (1996); Williamson (1996);and various news sources.

The stabilization program led to an important reduction in inflation (Figure A7), but not to the level observed in trading partners and, as a result, the real exchange rate appreciated considerably. In this context, the devaluations of 1987 and 1988 came only after considerable losses in competitiveness had accumulated. Thus, in order to prevent further losses in competitiveness and reduce the variability of the real exchange rate, the authorities adopted the horizontal band. The various realignments of the central parity of the horizontal band indicated above did stabilize the real exchange rate,56 but there were no further gains in disinflation during this period: inflation fluctuated between 16 percent and 20 percent from 1987 to 1991. The fiscal situation had worsened, as the surpluses recorded in 1985–86 vanished, and by 1989 a deficit of nearly 6 percent of GDP was recorded. A moderate deterioration of the current account balance also took place.

Figure A7
Figure A7

Israel: Selected Macroeconomic Indicators

1In percent of GDP2Index 1990= 100

The frequency of devaluations (four in less than two years) of the parity of the horizontal band led to an increase in interest rate volatility as markets began to anticipate, and possibly also precipitate, the subsequent devaluation. This fact, coupled with the recognition by the authorities that they were not pursuing a fast convergence of domestic inflation to the levels observed in partner countries, led the central bank to adopt a crawling band. As a byproduct of the band, the central bank also adopted explicit inflation targets. The move to a crawling band did lead to lower interest rate volatility and an improved inflation performance.57 However, in spite of robust growth throughout this period, there was only a temporary adjustment of the fiscal accounts in the early 1990s, followed by worsening later.

Although various measures aimed at deregulating the domestic financial system and relaxing restrictions on capital movements were implemented (with some reversals) in the 1970s and 1980s, net capital inflows were not significant until the 1990s, when there was a further move toward capital account liberalization. The presence of large capital inflows in the context of a lax fiscal policy has led to a real appreciation and a widening current account deficit in recent years. Nevertheless, Israel’s policy of moving gradually to more flexible exchange rate arrangements following an exchange-rate-based stabilization has functioned relatively well. The authorities have managed to reduce inflation, albeit slowly and to levels that are still high in relation to those observed in industrial countries, while preventing the development of significant balance of payments problems. They intend to remove all remaining capital account restrictions in the course of 1998.

Mexico

The crisis in Mexico in 1994–95 involved a sudden exit from a currency commitment that involved generalized loss of investor confidence and severe real and financial costs (Table A6). This occurred despite the fact that the exchange arrangement before the crisis was a crawling band that already had a degree of nominal flexibility. However, in 1994 it became a one-sided bet as the exchange rate moved to the most depreciated edge of the band, and stayed there persistently.

Table A6.

Mexico: Foreign Exchange and Financial Systems

article image
Sources: Alexander and others (1997); International Monetary Fund. Staff Country Reports, Annual Report an Exchange Afrarrgements and Exchange Restrictions, and International Financial Statics ; Alexander. Balifo and Enoch (1995); Goldstein (1997); Gonzales-Henmosillo, Pazarbasjoglij. arid Billings (1996); Kairiinsky and Reinhart (1996); Undgren. Garcia. and Saal (1996); and Odker and Pazarbasioglu (1995).

Following a period of occasional speculative pressures during 1994, on December 20 the Mexican authorities widened the band by devaluing its upper edge against the U.S. dollar by 15 percent and subsequently abandoned the crawling band mechanism that had been in effect since November 1991. The decision was taken against the background of continued market pressures, which had reduced the reserves of the Bank of Mexico to unsustainable levels, and fragility of the banking system, which had made raising interest rates an unviable option to defend the band.

The float marked a significant change in Mexico’s long-standing policy of employing fixed exchange rates in its stabilization efforts. Prior to the late 1980s, the Mexican authorities had adopted many different forms of a pegged exchange rate system, ranging from a pure peg to a preannounced crawling peg, alternating with brief periods of floating rates that followed episodes of balance of payments crises. From 1988 onward, however, Mexico followed a strategy of gradually adapting exchange rate policy to changing economic conditions and policy objectives. The authorities fixed the U.S. dollar exchange rate in early 1988, replacing it with the crawling peg mechanism in 1989, and moved to a gradually widening crawling band in November 1991. In doing so, the authorities attempted to combine nominal anchoring with sufficient exchange rate flexibility to maintain competitiveness while keeping inflation low, in the face of strong inflows of capital since the late 1980s, In practice, however, the Bank of Mexico did not use the flexibility offered by the band and conducted intramarginal intervention to keep the peso within a narrow inner band until early 1994.

Unlike previous currency crises in Mexico, the crisis of 1994–95 took place in the context of a broadly favorable macroeconomic environment following the authorities’ adoption of a comprehensive anti-inflation strategy in 1987 (Figure A8). Because of restrictive monetary and fiscal policies, a stable exchange rate supported by tight wage policy, and the liberalization of the trade and financial systems, economic growth picked up, inflation was reduced to single digit levels, fiscal accounts and public debt improved markedly, and external debt was reduced to more sustainable levels. By 1993, the peso was somewhat overvalued and the current account deficit was widening, but there was growing confidence in the economy since the deficit was being covered by strong inflows of capital. The fact that a significant part of the inflows was short term was to be the source of future problems. Dollar-denominated and short-term components of the external debt also rose, with the largest proportional increase reflecting a rise in nonresident holdings of dollar-linked nea-sury bills (Tesobonos),

Figure A8
Figure A8

Mexico: Selected Macroeconomic Indicators

Sources: International Monetary Fund, International Financial Statistics, Information Notice System, and World Economic Outiook; United Nations, Monthly Bulletin of Statistics; and IMF staff estimates.1In percent of GDP2Index 1990 = 100

In addition, deregulation of the financial system was undertaken without adequate preparation, with supervisory powers and commercial banks lacking the necessary expertise and tools to assess credit and market risks and to monitor lending. As a result, banking sector credit expanded substantially and asset quality deteriorated rapidly. Signs of increasing fragility became apparent from late 1992 and played a critical role in the authorities’ failure to respond with sufficiently higher interest rates to the exchange market tensions during 1994, which were triggered by a number of adverse domestic and externa) shocks. These tensions caused a shift in market sentiment, and loss of confidence was amplified by the surprise announcement of a devaluation of the band in December 1994. Given the short term and speculative nature of the capital inflows, there was a substantial flight of capital out of Mexico as maturing government debt was not rolled over. Continued selling pressure on the peso and the subsequent crisis in the foreign exchange market was followed by a severe banking crisis, exacerbated by the sharp depreciation of the peso, the downturn in economic activity, and sharply higher interest rates. More than half of the Mexican banks received financial support from the government, and a bank restructuring program was put in place in early 1995.

It is noteworthy that exchange market tensions did not cease with the floating of the peso, which lost more than half of its value over a period of a year. Although there is little disagreement on their role in the crisis, the deterioration in Mexico’s economic fundamentals did not seem to justify the size of the exchange rate depreciation that occurred. It seems that the sudden loss of confidence following the devaluation may have stoked self-fulfilling fears of an imminent default on short-term debt (given the low level of reserves), a collapse of the banking system, and a spiraling depreciation of the peso.58 The confirmation of the availability of adequate external financing and a comprehensive program put in place by the authorities (including a bank restructuring strategy) helped restore investor confidence, but only gradually. The Mexican experience is important in demonstrating that the implementation of significant reforms and maintaining prudent fiscal policies are not sufficient to ensure the sustainability of large external imbalances in a liberalized financial system, unless such prudence is accompanied with a healthy financial system, as well as with timely and adequate policy responses to shocks.

Poland

The Polish experience with exchange rates since 1990 is an example of a timely and flexible adaptation of exchange rate arrangements to changing economic conditions (Table A7). It also demonstrates that exchange rate policy cannot, and should not, be viewed in isolation, and should be accompanied by financial, incomes, and structural policies that are consistent with it.

Table A7.

Poland: Foreign Exchange and Financial Systems

article image
Sources : Alexander and others (1997). International Monetary Furd, Staff Country Reports, Annual Report on Exchange Arrangements and Exchange Restrictions (various issues), and International Financial Statistics’, Alexander, Balifio, and Enoch (1995); Ebrill and others (1994); Goldstein (1997);and Lindgren, Garcia, and Saal (1996)

In 1990, Poland embarked on a comprehensive reform program with a view to stabilizing the economy and to putting it on the transition path to a market economy. Monetary, fiscal, and incomes policies were tightened, the foreign exchange market was substantially liberalized, most restrictions on capital transactions were abolished, and a comprehensive set of financial reforms was implemented, including the freeing of interest rates and deregulation of the banking system. As a result, inflation was reduced substantially with a remarkable growth performance, imbalances in the external account were substantially eliminated, and external debt was reduced markedly. The success of Poland’s overall macroeconomics policy has been attributed to a “multiple-anchor, multiple-indicator” change rate, interest rates, fiscal and incomes policies, and structural reforms all playing a major role.

Within this strategy, exchange rate policy has played a dual role, acting both as an inflation anchor and responding as well to developments in international competitiveness. The zloty was initially fixed against the U.S. dollar to brake hyperinflationary pressures, while the parity was set at a level that would restore and maintain competitiveness. The currency was then pegged to a basket in May 1991, to address the continued loss of competitiveness in view of the still high inflation relative to trading partners. Further flexibility was introduced in October 1991, when the regime was modified to a preannounced crawling peg as deterioration in competitiveness persisted; the anchor role of the exchange rate was maintained, however, by choosing an “active” crawl that allowed the currency to depreciate at a rate smaller than the projected inflation differentials. Flexibility of the exchange rate was further advanced in May 1995 when the authorities decided to allow a market-driven appreciation within a formal crawling band of ±7 percent around a central parity. This system would maintain partially the anchor role for the exchange rate, while at the same time providing a scope to deal with various shocks and capital flows.

Although the Polish approach to monetary and exchange policy has so far served the country well, the authorities have had to deal with tensions associated with trying to attain the competing goals of competitiveness and disinflation. In order to address the competitiveness problem, Poland moved progressively toward more flexible variants of “fixed” exchange regimes (as described above), as well as allowing for step devaluations (three devaluations in 1991–93). The pass-through of the resulting depreciation to inflation was reduced by choosing an “active” crawl that allowed for less-than-perfect accommodation of inflation differentials; the rate was also reduced successively (in five steps in 1991–96). More important, exchange rate policy was complemented with additional anchors, including tight fiscal and (tax-based) incomes policies, which helped alleviate pressures on both inflation and competitiveness.

Implementation of monetary policy has been further complicated by a substantial increase in capital inflows since 1994 against the background of rising domestic interest rate differentials and liberalization of the capital account. Continued need to progress in disinflation required higher domestic interest rates, but the latter stimulated further inflows, risking a loss of control over monetary aggregates and inflation. As sterilization of these inflows became increasingly difficult and costly, the authorities used further reductions in the crawl rate and relaxed most controls on capital outflows. Faced with continued inflows, they increased the flexibility of the exchange rate through a switch to the crawling band regime; as in Mexico, however, the band has not been used much in practice, except for a few small step appreciations and a reduction in the crawl rate. The authorities finally adhered to interest rate reductions as the cost of sterilization became a serious policy concern; lower interest rates, however, caused a recent acceleration in domestic credit expansion, and interest rates have been increased since Sate 1996 (Figure A9).

Figure A9
Figure A9

Poland: Selected Macroeconomic Indicators

1In percent of GDP2Index 1990 = 100

Poland’s banking system was also not free of troubles. In fact, similar to the experiences of several of the other countries discussed in this section, insufficient preparation for liberalization had resulted in a sharp increase in bank lending and a subsequent deterioration in loan quality. Banking system problems were managed without a subsequent crisis, however, as the government support to problem banks was successfully tied to a comprehensive program of bank restructuring. Prudential regulations and supervisory capacity have also been strengthened. As a result, the situation in the banking sector stabilized, with a substantial improvement occurring in the ratio of nonperforming loans.

Thailand

After more than a decade of exchange rate stability, the Thai baht came under severe speculative pressure in May 1997, forcing the authorities to impose controls on international capital transactions (Table A8). Despite heavy intervention by the Bank of Thailand and significant intervention support provided by the regional monetary authorities, the pressure on the baht did not subside, and on July 2, 1997. the authorities abandoned the policy that had been in effect since November 1984, of pegging the baht to an undisclosed basket of currencies (dominated by the U.S. dollar). The decision to float was taken against the background of growing concerns about Thailand’s economic fundamentals, a sharp fall in official reserves, and concerns about the adverse effects of a required increase in interest rates on a fragile economy and financial system. By early October 1997, the baht had depreciated by about 30 percent and financial markets had not yet calmed down despite the announcement of an emergency financing package with the IMF. The turmoil also spilled over quickly to several other Asian currencies, forcing their authorities to adopt more flexible exchange rate regimes.

The crisis in Thailand occurred despite an impressive record of economic growth over more than a decade (Figure A10). The authorities’ commitment to macroeconomic stability, prudent financial policies, and increasingly open trade and financial systems had contributed to strong economic performance, as evidenced by single-digit inflation rates, improved fiscal accounts, and a marked decline in its public debt since the late 1980s. There were growing signs of overheating in the economy after 1994, however, despite tight financial policies followed by the authorities, as inflation picked up and the current account deficit widened significantly; the latter in part reflected Thailand’s loss of competitiveness brought about by the baht’s close link with the appreciating dollar. While the deficit was more than financed by strong inflows of capital, a growing component of these inflows (about 60 percent) was short term. Access to short-term foreign capital had been facilitated by capital account liberalization since 1993, including the opening of the Bangkok International Banking Facilities. The rising share of short-term external debt made the continued financing of Thailand’s large external imbalance vulnerable to a sudden shift in market sentiment.

The growing size and volatility of these inflows also complicated the implementation of monetary policy in an environment of a fixed exchange rate, an increasingly open capital account, and scarcity of indirect monetary instruments. The authorities attempted to manage these inflows with a combination of monetary and prudential measures, and later with some price-based controls. They refrained from further liberalization of the capital outflow controls or from allowing greater flexibility in the determination of the exchange rate, which could have helped discourage speculative inflows and put downward pressure on inflation.

Similar to the experiences of Mexico and Poland, the liberalization of the financial system in Thailand was associated with a rapid expansion in lending. A growing number of finance and securities companies increased their exposure to the real estate sector and provided foreign currency loans to borrowers who were only partly hedged in an environment secured by apparent exchange rate stability. This expansion in turn led to a gradual deterioration in asset quality. As the markets started in 1996 to observe growing indications of the solvency and liquidity problems encountered by several finance and securities companies, the authorities’ lack of transparency in disclosing data on their financial situation precipitated market fears that the health of the financial system was much worse than had so far been apparent. In the event, a significant number of institutions were asked to suspend operations, several to merge with stronger partners, and some to increase their capital in the aftermath of the baht’s float. The authorities provided emergency liquidity to the financial system and subsequently adopted a comprehensive restructuring strategy.

Table A8.

Thailand: Foreign Exchange and Financial Systems

article image
Sources:Alexander and others (1997);International Monetary Fund, Staff Country Reports, Annual Report on Exchange Arrangements and Exchange Restrictions and International Financial Statistics: Goldstein (1996. 1997); Goldstein: and Turner (1996); Johnston and Pazarbasioglu (1995); Johnston. Darbar, and Echeverria (1997); Kaminsky and Reinharr. (1996); Lindgren, Garcia, and Saal (1996); Mehran, Laurens.and Quintyn (l996);Warr and Nidhiprabha (1995); IBCA Reports on Thailand’s Banks; Standard and Poor’s Credct Analysis Service Reports; and various news sources.
Figure A10
Figure A10

Thailand: Selected Macroeconomic Indicators

1In percent of GDP2Index 1990 = 100

References

  • Aghevli, Bijan B., Mohsin S. Khan, and Peter J. Montiel, 1991, Exchange Rate Policy in Developing Countries: Some Analytical Issues, IMF Occasional Paper 78 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Alesina, Alberto, Ricardo Haussman, Rudolf Homines, and Ernesto Stein, 1995,Budget Institutions and Fiscal Performance in Latin America,OCE Working Paper.

    • Search Google Scholar
    • Export Citation
  • Alesina, Alberto, and Roberto Perotti, 1996,Budget Deficits and Budget Institutions,IMF Working Paper 96/52 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Alexander, Wiliam E., Tomás J.T, Baliňo, and Charles Enoch, 1995, The Adoption of Indirect Instruments of Monetary Policy, IMF Occasional Paper 126 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Alexander, William, 1997, Systematic Bank Restructuring and Macroeconomic Policy (Washington: International Monetary Fund).

  • Balino, Tomas J.T., and Charles Enoch, 1997, Currency Board Arrangements: Issues and Experiences, IMF Occasional Paper 151 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Bayoumi, Tamim, and Barry Eichengrecn, 1992.Is There a Conflict Between EC Enlargement and European Monetary Unification?Greek Economic Review. Vol. 15, No. I, pp. 131 -54; also published as CEPR Discussion Paper 646 (London: Centre for Economic Policy Research).

    • Search Google Scholar
    • Export Citation
  • Bayoumi, Tamim, and Barry Eichengrecn, 1996, “Exchange Rate Volatility and Intervention: Lessons from the Theory of Optimum Currency Areas,paper presented to a conference on Optimum Currency Areas: The Current Agenda for International Monetary and Fiscal Integration, Tel Aviv, December 4–6, 1996.

    • Search Google Scholar
    • Export Citation
  • Buiter, Willcm H., 1987.Borrowing to Defend the Exchange Rate and the Timing and Magnitude of Speculative Attacks,Journal of International Economics, Vol. 23 (November), pp. 221 -39.

    • Search Google Scholar
    • Export Citation
  • Chuhan, Punam, Stijn Ciaessens, and Nlandu Mamigni, 1993,Equity and Bond Flows to Latin America and Asia: The Role of External and Domestic Factors.Policy Research Working Paper 1160 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation
  • Cole, Harold L., and Timothy J. Kehoe, 1996,A Self-Fulfilling Model of Mexico’s 1994-95 Debt Crisis,Journal of International Economics. Vol. 41 (November),pp. 309 -30.

    • Search Google Scholar
    • Export Citation
  • Collins, Susan M., 1996,On Becoming More Flexible: Exchange Rate Regimes in Latin America and the Caribbean,Journal of Development Economics. Vol, 51 (October), pp. 117 -38.

    • Search Google Scholar
    • Export Citation
  • Cotlarelli, Carlo, and Curzio Giannini, 1997, Credibility Without Rules? Monetary Frameworks in the Post-Bretton Woods Era. IMF Occasional Paper 154 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Ebrill, Liam P., and others, 1994, Poland: The Path to a Market Economy, IMF Occasional Paper 113 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Edwards, Sebastian, 1996,The Determinants of the Choice Between Fixed and Flexible Exchange-Rate Regimes,NBER Working Paper 5756 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation
  • Edwards, Sebastian,, and Alejandra Cox 1991, Monetarism and Liberalization: The Chilean Experiment (Chicago: University of Chicago Press).

    • Search Google Scholar
    • Export Citation
  • Edwards, Sebastian, and Fernando Losada, 1994,Fixed Exchange Rates, Inflation and Macroeconomic Discipline.NBER Working Paper 4661 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation
  • Eichengrecn, Barry, 1994, International Monetary Arrangements for the 21st Century (Washington: Brookings Institution).

  • Eichengrecn, Barry, Andrew K. Rose, and Charles Wyplosz, 1995,Exchange Market Mayhem: The Antecedents and Aftermath of Speculative Attacks,Economic Policy, No. 21 (October), pp. 249 -312.

    • Search Google Scholar
    • Export Citation
  • Eichengrecn, Barry, and Albert Fishlow, 1996,Contending with Capital Flows: What Is Different About the 1990s?Council on Foreign Relations, Occasional Paper (New York).

    • Search Google Scholar
    • Export Citation
  • Fernandez-Arias, Eduardo, 1994,The New Wave of Private Capital Inflows: Push or Pull?Policy Research Working Paper 1312 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation
  • Frankel, Jeffrey A., and Andrew K. Rose, 1996a.Currency Crashes in Emerging Markets: Empirical Indicators.NBER Working Paper 5437 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation
  • Frankel, Jeffrey A., and Andrew K. Rose, 1996b.The Endogeneity of the Optimum Currency Area Criteriapaper presented to a conference on Optimum Currency Areas: The Current Agenda for International Monetary and Fiscal Integration, Tel Aviv. December 4–6, 1996.

    • Search Google Scholar
    • Export Citation
  • Frenkel, Jacob, Morris Goldstein, and Paul Masson, 1991, Characteristics of a Successful Exchange Rate System, IMF Occasional Paper 82 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Giavazzi, Francesco, and Alberto Giovannini, 1989. Limiting Exchange Rate Flexibility: The European Monetary System (Cambridge, Massachusetts: MIT Press).

    • Search Google Scholar
    • Export Citation
  • Goldstein, Morris, 1996,Presumptive Indicators/Early Warning Signals of Vulnerability to Financial Crises in Emerging Market Economies(unpublished; Washington: Institute for International Economics).

    • Search Google Scholar
    • Export Citation
  • Goldstein, Morris, 1997. The Case for an International Banking Standard (Washington: Institute for International Economics).

  • Goldstein, Morris, Philip Turner, 1996, Banking Crises in Emerging Economies: Origins and Policy Options, BIS Economic Paper 46 (Basle: Bank for International Settlements).

    • Search Google Scholar
    • Export Citation
  • González-Hermosillo, Brenda, Ceyla Pazarbaşjoğlu, Robert Billings, 1997.Determinants of Banking System Fragility: A Case Study of Mexico,Staff Papers, International Monetary Fund. Vol. 44 (September),pp. 295 -314.

    • Search Google Scholar
    • Export Citation
  • Goodhart, Charles A.E., 1995.The Political Economy of Monetary Union,in Understanding Interdependence: The Macroeconomics of the Open Economy, Kenen, Peter B. (Princeton, New Jersey: Princeton University Press).

    • Search Google Scholar
    • Export Citation
  • Helpman, Elhanan, Leonardo Leiderman, and Gil Buf-man, 1994.New Exchange Rate Bands: Chile. Israel and Mexico.Economic Policy, Vol. 9 (October), pp. 260 -306.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund. 1995a. International Capital Markets: Developments. Prospects, and Policy Issues, World Economic and Financial Surveys (Washington).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund. 1995b,Using Exchange Rate Anchors in Adjustment Programs: When and How?IMF Survey, Vol. 24 (November 20). pp. 361 -63.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund. 1997. World Economic Outlook, October 1997: A Survey bv the Staff of the International Monetary Fund, World Economic and Financial Surveys (Washington).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund. Annual Report on Exchange Arrangements and Exchange Restrictions (Washington), various issues.

  • Johnston, R. Barry, and Ceyla Pazarbaşjoğlu, 1995.Linkages Between Financial Variables, Financial Sector Reform, and Economic Growth and Efficiency,IMF Working Paper 95/103 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Johnston, R. Barry, Salim M. Darbar, and Claudia Echeverria, 1997,Sequencing Capital Account Liberalization: Lessons from the Experiences in Chile. Indonesia, Korea, and Thailand,IMF Working Paper 95/157 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Kaminsky, Graciela L., and Carmen M. Reinhart, 1996,The Twin Crises: The Causes of Banking and Bal-ance-of-Payments Problems.International Finance Discussion Paper 544 (Washington: Board of Governors of the Federal Reserve System).

    • Search Google Scholar
    • Export Citation
  • Kaminsky, Graciela, Saul Lizondo, and Carmen Reinhart, 1997,Leading Indicators of Currency Crises,IMF Working Paper 97/79 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Kenen, Peter B., 1969,The Theory of Optimum Currency Areas; An Eclectic View,in Monetaty Problems of the International Economy Robert A. Mundell and Alexander K. Swoboda, (Chicago: University of Chicago Press).

    • Search Google Scholar
    • Export Citation
  • Klein, Michael W., and Nancy P. Marion, 1994,Explaining the Duration of Exchange-Rate Pegs,NBER Working Paper 4651 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation
  • Krugman, Paul, 1979,A Model of Balance of Payments Crises,Journal of Money, Credit and Banking, Vol. 11 (August),pp. 311 -25.

  • Lindgren, Carl-Johan, Gillian Garcia, and Matthew I. Saal, 1996, Bank Soundness and Macroeconomic Policy (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Masson, Paul, Miguel Savastano, and Sunil Sharma, 1997.The Scope for Inflation Targeting in Developing Countries.IMF Working Paper 97/130 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • McKinnon, Ronald, 1963.Optimum Currency Areas.American Economic Review. Vol. 53 (September), pp. 717 -25.

  • Mehran, Hassanali, Bernard, Laurens, and Marc Quintyn, 1996, Interest Rate Liberalization and Monev Market Development: Selected Country Experiences (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Mundell, Robert, 1961,A Theory of Optimum Currency Areas,American Economic Review, Vol. 51 (September), pp. 657 -65.

  • Mussa, Michael, 1986,Nominal Exchange Rate Regimes and the Behavior of Real Exchange Rates: Evidence and Implications,” in Real Business Cycles, Real Exchange Rales and Actual Policies, Karl Brun-ner, Allan H. Meltzer, (Amsterdam: North Holland).

    • Search Google Scholar
    • Export Citation
  • Mussa, Michael,, and others 1994, Improving the International Monetaiy System: Constraints and Possibilities. IMF Occasional Paper 116 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Obstfeld, Maurice, 1994,The Logic of Currency Crises,NBER Working Paper 4640 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation
  • Obstfeld, Maurice, 1995,International Currency Experience: New Lessons and Lessons Relearned,Brookings Papers on Economic Activity: I, Brookings Institution, pp. 119 -211.

    • Search Google Scholar
    • Export Citation
  • Obstfeld, Maurice, Rogoff, Kenneth, 1995,The Mirage of Fixed Exchange Rates,Journal of Economic Perspectives, Vol. 9 (Fall),pp. 73 -96.

    • Search Google Scholar
    • Export Citation
  • Ötker, Inei, and Ceyla Pazarbaşjoğlu, 1995Speculative Attacks and Currency Crises: The Mexican Experience.Open Economies Review, Vol. 7, Supplement I, pp. 535 -52.

    • Search Google Scholar
    • Export Citation
  • Ozkan, F. Gulcin, and Alan Sutherland, 1994,A Model of the ERM Crisis,CEPR Discussion Paper 879 (London: Centre for Economic Policy Research).

    • Search Google Scholar
    • Export Citation
  • Rose, Andrew, 1994,Are Exchange Rates Macroeconomic Phenomena?Federal Reserve Bank of San Francisco Economic Review. No. l. pp. 19 –30.

    • Search Google Scholar
    • Export Citation
  • Sachs, Jeffrey, Aaron Tornell, and Andrés Velasco, 1995,The Collapse of the Mexican Peso: What Have We Learned?Harvard Institute of Economic Research Discussion Paper 1724 (Cambridge, Massachusetts).