VI An Exit to What?
- Barry Eichengreen, Inci Ötker, A. Hamann, Esteban Jadresic, R. Johnston, Hugh Bredenkamp, and Paul Masson
- Published Date:
- August 1998
When an existing regime of relative exchange rate fixity is exited in a period of tranquility or upward pressure on the currency, it may be desirable, as previously argued, to move gradually to a regime of greater exchange rate flexibility and even to signal somewhat in advance how this process will develop. When a regime of relative exchange rate fixity is abandoned in a crisis, a gradual move to greater exchange rate flexibility is rarely, if ever, a relevant issue. The initial drop in the foreign exchange value of the domestic currency is usually substantial and, if anything, there is a tendency for the exchange rate to overshoot in the shorter term because of the general loss of policy credibility. While gradualism is not an issue, the means and degree of resistance that should be mounted against potentially excessive depreciation is an important question. If it is perceived that the authorities see devaluation as an easy way out of their economic difficulties, the result can easily be a massive depreciation that proves to be self-justifying through its inflation effects and by its undermining of the financial and economic position of institutions and enterprises that have relied (perhaps somewhat unwisely) on the authorities’ previous commitment to a regime with a high degree of exchange rate fixity. Typically, following the exit from a pegged regime under conditions of strong capital outflows, the exchange rate will drop sharply before stabilizing, perhaps briefly, at a lower level. What is important, in these circumstances, is to put in place quickly the policies needed to avoid a further round of depreciation that may become self-reinforcing.
Under the Bretton Woods system in which all countries maintained fixed-but-occasional-adjustable exchange rate regimes, it was understood that when it became necessary to correct a “fundamental disequilibrium” in a country’s balance of payments through an adjustment of its exchange rate parity, it was also important to avoid an excessive depreciation of the country’s currency. With the move to floating exchange rates at least among the largest industrial countries, explicit concern to avoid this outcome has surely diminished if not completely disappeared from policy discussions. Experience suggests, however, that the economic problems of large exchange rate movements in a balance of payments crisis have not disappeared. In its precipitous depreciation between December 1994 and May 1995, the Mexican peso lost 40 percent of its real value against the U.S. dollar, contributing to a massive upsurge of inflation, a sharp rise in nominal and real interest rates, a collapse of domestic demand, and a very rapid reduction of the current account deficit, amounting to about 8 percent of Mexican GDP in a period of six months. And the effects of Mexico’s crisis both spilled over through the normal trade linkages to slow significantly the growth of the U.S. economy and spread through other mechanisms of contagion to afflict a number of other Latin American economies. The crises in the ERM during 1992–93 also exhibited spillover and contagion. Recent events, beginning with the float of the Thai baht in early July 1997 and the spread of crisis to other Asian countries, dramatically demonstrate the potential, in certain environments, for spillovers and contagion to go global.
The principal means for avoiding overly large exchange rate movements in a crisis is to introduce a comprehensive program of firm monetary and fiscal policies and to announce and implement relevant reforms in the objectives and institutions of economic policy, including measures to deal effectively with problems in the financial sector. Such a policy program will help to bring about the reduction in the external payments deficit that generally is needed to accommodate a sharp decline in external financing that typically occurs in a balance of payments crisis. It helps to spread the burden of this necessary adjustment more evenly across sectors of the economy, and it diminishes the inflationary and financial risks associated with leaving the entire burden of adjustment to be absorbed by the exchange rate. Such a policy program is also essential both to restore some of the credibility that typically is lost with the abandonment of the old exchange rate regime and to establish a new anchor for expectations about future economic policies.
If policies have already been tightened as part of the defense of the old exchange rate regime, this will help, but some further tightening may still be needed to reinforce credibility. The appropriate degree of tightening needs to be carefully judged in light of the economic situation and the state of financial markets. If fiscal excess is a key contributor to the payments imbalance, fiscal policy should be tightened aggressively. If the budget was not a problem before the crisis, then tighter fiscal policy should be expected to make a more moderate contribution to correcting the external imbalance. Monetary policy needs to be sufficiently firm to contain the inflationary consequences of depreciation and to resist any major overshoot of the exchange rate. Indeed, in the immediate post-exit period, a fairly tight monetary stance may be needed temporarily. Weaknesses in the economy and in the financial sector imply significant risks from a tight monetary policy; but these risks must be weighed against the danger, demonstrated in a number of cases, that a premature easing of monetary conditions can undermine confidence and lead to deeper and more prolonged economic and financial difficulties.
In support of these policy efforts, but not as a substitute for them, some intervention in the foreign exchange market may well be appropriate. Such intervention should be (partially) nonsterilized, so that monetary policy firms progressively in the effort to forestall unwarranted depreciation. Official intervention in such circumstances, after the exchange rate has already depreciated to or below the level apparently consistent with economic fundamentals, is very different from sterilized intervention to defend an established exchange rate that is potentially overvalued. Indeed, an important reason to avoid intervening too extensively to support an existing but potentially unsustainable exchange rate regime is to maintain an adequate stock of reserves that can be used to credibly defend a significantly depreciated exchange rate that is in line with, or even somewhat below, the level consistent with economic fundamentals.
If a country has been forced to exit in a crisis, it is an indication that the new exchange rate arrangement needs to embody more flexibility, unless a greater commitment to supporting policies is made. The presumption must be that the circumstances that gave rise to one disorderly exit can occur again—unless for some reason the authorities are prepared to undertake a much firmer commitment of the entire economic policy regime to resurrecting and maintaining a relatively fixed exchange rate. In the midst of a crisis following abandonment of the old regime, conditions in the foreign exchange and financial markets are likely to remain turbulent, and it is probably unwise for the authorities to attempt to do more than actively resist (through monetary policy and intervention) depreciation that appears substantially excessive. Aside from this, market forces should be allowed considerable room to move the exchange rate.
When the market situation calms down, the authorities may want to become more active in their efforts to influence the behavior of the exchange rate—as part of the new, more flexible exchange rate regime they intend to pursue. There is nothing wrong with this. As noted in Section II, developing and transition countries are generally not good candidates to be absolutely unfettered floaters. For these economies, the exchange rate is usually an important economic policy concern, and its behavior provides important signals about the government’s current policies and policy intentions.
But, in the new regime, what weight should be given to the exchange rate? In what way and how can the authorities’ commitment to its objectives under the new regime be credibly established? There is no unique answer to these questions that is relevant for all developing and transition countries. Instead, the answers depend on the circumstances of the country, on its history, traditions, and institutions, and on the preferences of policymakers and the general public (see Appendix 1). A general principle to keep in mind, however, is that a government faces the critical task of providing a reasonable and credible basis for the formation of expectations about what its policies will deliver, so that there are advantages in making explicit the degree of commitment to, or concern about, the level of the exchange rate.
For several of the more advanced transition economies of central and eastern Europe that have strong trade links with the European Union (EU), especially those that may be candidates for eventual membership, a unilateral link to the euro may provide an attractive way of anchoring their monetary policies. While these countries would not enjoy the support of intervention by the European System of Central Banks, since the new exchange rate mechanism (ERM2) is designed only for EU members that are not yet members of EMU, a similar rationale for such an exchange link would apply to them as for EU members. Depending on the country’s initial rate of inflation and other features of its economy, that link might be tighter or looser, involving a fixed peg or a crawl, and a narrow or wide band.
Developing and transition countries with diversified trade and financial relations across the major currency areas have, in many instances, experienced considerable difficulty maintaining single currency pegs in the face of wide swings of major currency exchange rates. Such difficulties, especially if they contributed importantly to a forced exit from a single currency peg, should be taken as a signal that if there is to be a return to a pegged exchange rate regime, an appropriate basket of currencies might usefully be the foundation for the peg. For a number of countries that find basket pegs attractive, the additional flexibility of moderately wide bands may have particular value. In particular, regional groups of countries, such as the emerging market economies of East and Southeast Asia, have important trade and financial relations with both Japan and the United States, and also increasingly compete with each other in their own and in third markets. A common single-currency peg for the members of such a group creates difficulties when major exchange rates move, but it imparts important stability to cross rates between members of the group. If all could agree on a common basket peg, this would have the same virtue without the defect. However, such an arrangement has other disadvantages: it might prove difficult to agree on a common basket, since the best basket for one might differ significantly from that for another: and, as recent events amply demonstrate, if one member of the group gets forced off of its peg in a crisis, pressures tend to spread to other members of the group. All of this suggests that in these circumstances, a basket peg with the additional flexibility of moderately wide bands merits serious consideration.
For countries where, for whatever reasons, inflation cannot be brought down to near the low rates expected to be sustained in the industrial countries, a pegged exchange rate even with moderately wide bands is unlikely to be sustainable for long. Some form of pegged exchange rate may, nevertheless, be relevant as a temporary expedient and as part of an effort to bring inflation down to somewhat lower rates. The exchange rate regime in such circumstances should not be portrayed as a long-term peg. and the authorities should not rest their credibility on the notion that the exchange rate will be long sustained. Rather the exchange rate regime should be understood as involving a temporary peg (with a band), with the recognition that the nature of the regime will probably need to be altered in a timely manner before significant pressures develop.
However, as discussed previously and in the appendices, fixed-but-adjustable regimes and temporary pegs have substantia] problems. Shifting an exchange rate peg is usually a difficult affair, as speculative pressures tend to build in anticipation of the shift and there is a tendency for the authorities not to recognize and implement a shift in the peg in a timely and relatively nondisruptive manner. Therefore, a country that expects to run for some time an inflation rate meaningfully above that in the industrial countries should consider an exchange rate regime that involves some form of crawl, where the persistent need to adjust the nominal exchange rate to maintain reasonable international competitiveness can take place without frequent and potentially disruptive breaks in the regime. Moreover, a crawl with bands communicates with reasonable clarity the policy intentions and commitments of the authorities. The parameters of the regime (the central parity, the rate of crawl, and the width of the bands) may need occasional adjustment, but presumably with less frequency and disruption than under a fixed-but-adjustable regime with the same long-term average inflation rate. The experiences of countries such as Chile and Israel speak well of properly managed crawling bands. On the other hand, it should be recalled that Mexico was pursuing a crawling band prior to its December 1994 devaluation and subsequent float, showing that exogenous shocks and policy inconsistencies can lead to the disruptive collapse of any exchange rate regime.
Beyond the crawling band lies the (pure or relatively) free float in which the exchange rate is allowed to fluctuate in response to market forces without any indication by the authorities of limits to its range of fluctuations. As indicated at the beginning of this paper, a free float is not the same as an absolutely unfettered float where the authorities exercise benign indifference toward the exchange rate. Rather, as in a country like Canada, the exchange rate can be viewed as a vitally important variable that substantially influences overall monetary conditions in the economy. Accordingly, economic policies, especially official interest rates, may be adjusted quite aggressively in light of exchange rate developments; and such adjustments will surely affect, and generally tend to limit, fluctuations in the exchange rate. With a pure free float, the intention of policy adjustments is not specifically to limit exchange rate movements, but to pursue other policy objectives. With a relatively free float, in contrast, specific concern about the exchange rate, along with other policy objectives, may be the motivation for policy adjustments.
In either case, the authorities need to be clear about their policy objectives and to communicate a credible commitment to achieve those objectives. One approach, successfully practiced by several industrial countries in recent years, is inflation targeting, where the primary objective (sometimes the only officially stated objective) of monetary policy is to keep the inflation rate in a very low and narrow range. The credibility of the commitment to this objective can be significantly enhanced by giving operational independence to the central banks with a clear mandate to accomplish the inflation target. Most important, credibility is established by consistently delivering inflation within the target range.
As discussed above, it is questionable whether a freely floating exchange rate and an inflation target objective for monetary policy are feasible, advisable, or fully credible for many developing and transition economies. Relative to their size, these economies are often subject to substantially larger internal and external shocks than industrial countries, and the transmission mechanisms through which monetary policy affects the economy and the price level tend to be less certain and reliable for many developing and transition countries. Even with the full dedication of monetary policy, inflation may simply not be as controllable as in industrial countries. Also, it may be important to retain significant flexibility for monetary policy to serve other objectives.
In particular, limiting the movement of the exchange rate from what would occur if market forces were allowed completely free rein may be a crucial objective that requires some support from monetary policy apart from its pursuit of an inflation target or other objectives. Facing large shocks, with relatively thin markets, poorly developed institutions, and absent a long track record for stable policies, there may not be an adequate basis for reasonably stable behavior of the exchange rate unless the authorities provide some guidance and support to the market. In this situation, there is a fundamental ambiguity about the objectives of the authorities’ exchange rate and monetary policies. Accordingly, there cannot be the same clarity of commitment of economic policy as under either a pure fix, where policy is fully dedicated to the exchange rate objective, or an absolutely unfettered float, where policy ignores the exchange rate and is fully dedicated to something else. Keeping inflation reasonably low and stable must be a primary, and sometimes overriding, objective of monetary and exchange rate policy, even when this requires a rise in interest rates or an appreciation of the exchange rate that many find uncomfortable. Sometimes concern with the exchange rate, international competitiveness, and the external payments position will need to take center stage. These occasions of conflicting objectives are the tough limes for economic policy and policy credibility. There can be no fixed rule for how to handle them. People must understand, and must be persuaded by experience, that policy will draw a reasonable balance between conflicting objectives.
In this regard, it is essential to recognize that an economic policy regime, which necessarily and desirably involves serious policy commitments, can collapse in a disruptive crisis that damages policy credibility for three fundamental reasons: because of large domestic or external shocks that no policy regime with a useful degree of commitment to good performance in more normal times could reasonably have withstood; because of policy errors and inconsistencies that significantly undermine a regime that might otherwise have survived troubled times; or because the overly ambitious policy objectives to which the regime became committed in relatively easy times never could plausibly have been achieved on a sustained basis. It is important for the authorities to avoid the temptation to undertake, in easy times when growth is good, inflation is low, and net exports strong, certain de jure or de facto policy commitments (such as pegging the exchange rate) that cannot be sustained when the economic weather turns more stormy.