6 Exchange Rate Management in the Reform Process
- Richard Bart, Chorng-Huey Wong, and Alan Roe
- Published Date:
- September 1994
Exchange rate management is one of those subjects in economics on which consensus is always difficult to obtain, even in the IMF, where it is at the core of the institutional work. The difficulty of achieving consensus arises because of two very complex links, one between exchange rate policy and other economic policies, and the other between exchange rate policy and economic performance in terms of output growth, employment and inflation levels, and balance of payments developments. The complexity of these relationships leads to different views on the way exchange rate policy operates when it is integrated with macro- and microeconomic policies, and also on the role it plays in terms of ultimate policy objectives, which are, of course, sustained growth, full employment, a viable external payments position, and low inflation.
Although my topic is exchange rate management in the reform process—that is, in economies in transition from central planning to market-based regimes—I want to make the point that exchange rate policy is a subject that applies uniformly to all types of economies: industrial and developing as well as those in transition. But while the basic principles of exchange rate policy are universally applicable, the workings of a particular policy will reflect the structure of the economy in which it is being used. A fixed exchange rate, for instance, is more likely to succeed in a relatively open, flexible, and competitive economy than in an economy with high trade barriers and few exports. A country’s stage of development may also limit its options in terms of exchange regimes.
To simplify my analysis, I have divided it into two parts: the macroeconomic aspect and the microeconomic aspect. The macroeconomic aspect is concerned with domestic financial stability, that is, with price and exchange rate stability. The microeconomic aspect is concerned more with the competitiveness of the economy and the sustainability of the real exchange rate.
The dilemma created by the different vantage points provided by the micro- and macroeconomic angles surfaces in the context of industrial economies. It can be seen, for instance, in regimes like the European Monetary System (EMS), where controlling inflation and protecting competitiveness are critical aims. It can also be seen in many developing countries, where capital inflows can endanger efforts to control inflation, thereby making economic competitiveness difficult to maintain. And, of course, this dilemma is present in the economies in transition. The essential question can be formulated as a straightforward choice between achieving a nominal objective—namely, price level stability—or meeting a real objective—competitiveness. The complexity lies in the difficulty of using economic policy in general, and exchange rate policy in particular, to attain these goals simultaneously.
Though complex, the issue is nothing more, I suggest, than a new version of the age-old conflict between internal and external balance that was discussed in the literature many years back. Should policy be aimed at domestic price stability at full employment (thus making the external variables endogenous), or should it be directed at achieving a balance of payments at a given exchange rate level (in which case it would be the domestic variables that would become endogenous)? The problem is essentially the same, although one important element has changed since the old controversy was debated: the setting in which national policies are implemented. Today, the world economy is closely interconnected in terms of trade and current account flows as well as capital movements. This interconnectedness, I will argue, is one of the most important modern economic developments for national authorities, because it limits not only the scope of exchange rate policy options but, more broadly, those of economic policies.
The Choice of an Exchange Rate Regime
The first key issue in determining how to manage exchange rate policy is choosing an exchange regime. The choice tends to be made in the context of a national economy’s relations with the international economic community. This relationship turns on two important points: the point of intersection between national economies, or the balance of payments; and the terms of the intersection, or the exchange rate and exchange regime.
Further, two basic principles are behind the choice of an exchange rate regime. The first I call the preference for either a relatively open or a relatively closed national system. I identify a fixed exchange rate regime with a preference for an open national system, for a variety of reasons. First, a fixed rate puts a clear constraint on domestic policymaking that is the direct consequence of the recognition or acceptance of the endogenicity of national policies. And second, a fixed rate implies a willingness to live within a system that requires its members to accept the spillover effects of each other’s policies. In this sense, a national regime based on a fixed exchange rate is open because it does not seek to insulate itself from the international system. Of course, this openness requires that a country be willing to live with intrusions from policies elsewhere, but on the other hand, it also allows intrusions, so to speak, in the policy domains of others.
A flexible exchange rate can be identified with a preference for a relatively closed national system, because of the main point made by advocates of this type of system: that flexibility provides autonomy in national economic policymaking, particularly when it comes to domestic monetary policies. In a sense, the national economy is closed, because the flexible exchange rate is ostensibly expected to make it possible for a country to contain the effects of its own policies and to provide a barrier to spillover effects of policies from other countries. In this sense, flexible exchange rates are a means of insulating, or “closing,” national economies and preventing the dispersion or dissemination of policy effects across countries.
The second fundamental principle that lies behind the choice of an exchange regime I call the degree of domestic policy activism. This principle is not equivalent to policy independence; I will come back to that later. The emphasis is on activism. From that standpoint, a fixed exchange rate constrains the scope of domestic economic policy aims much more than a flexible exchange rate. Thus, a fixed exchange rate regime suggests a limited degree of national economic policy activism. In contrast, a flexible exchange rate regime implies the active use of domestic economic policies, something that requires relative isolation.
In discussing what flexible rates provide, I purposely do not use the term “independence” because I do not think that, in practice, national economic policy is independent under a flexible exchange rate. For example, in today’s world the major currencies—those of the United States, Japan, and the European countries—float against one another, but no one suggests that the U.S. monetary authorities can be oblivious to what happens to interest rates—or exchange rates, for that matter—in Europe or Japan, and vice versa. The United States is not independent; Japan is not independent; Europe is not independent; and yet their exchange rates are flexible in relation to each other. For this reason, I prefer not to make too much of a point of the degree of economic policy independence that can be gained through exchange rate flexibility.
These are the two fundamental choices of exchange regimes—“fundamental” because they involve more than just purely technical economic considerations. The choice between an open and a closed system involves many criteria beyond those of economics. But there are technical standards that can be used in determining which exchange rate regime is best. The three that I consider follow.
The first relates to the type of disturbance that economies are often exposed to. Numerous sources in the literature focus on the importance of distinguishing between real and nominal shocks and on the merits and demerits of fixed and flexible exchange rates in relation to each type of shock. The conventional wisdom is that a fixed exchange rate is best for a country that faces primarily nominal shocks—such as those originating in the domestic money market—and that a flexible exchange rate is best for a country experiencing real shocks, such as those originating in the domestic goods market.
Now, despite the logic of the argument, I am skeptical of this criterion, for two reasons. First, it is difficult for policymakers to determine whether they are facing a nominal shock, a real shock, a domestic financial market shock, or a domestic goods market shock quickly enough to alter exchange rate policy in response. Secondly, even if they were able to do so, exchange rate regimes should not be changed frequently. It is not practical to change an exchange rate regime in response to periodically recurring shocks.
Another criterion that technically can be used in choosing a regime relates to the structural characteristics of an economy, or the degree of its openness. In the context of the so-called “optimum currency area” literature in the early 1960s, Robert Mundell argued that if factors of production among economies are mobile, then those economies will be in a better position if they fix their exchange rates.1 This idea was then extended to openness measured in terms of both traded goods and the diversification of output.
The final criterion is the commonality of policy aims or of economic policies themselves. The reasoning here is that when countries have similar aims—for example, vis-á-vis inflation—they probably should fix the exchange rate among themselves. Similar reasoning applies more broadly to the degree of commonality that characterizes domestic financial policies—that is, the extent of convergence of fiscal and monetary policies.
These two last criteria, openness and commonality of aims or policies, are quite operational. They can be used in choosing a regime more easily than the criterion based on types of shocks.
Reform and Exchange Rates
Like exchange rate policy, the reform process has two components: the macroeconomic aspect and the microeconomic aspect. How do they interact in the transition economies? In the reform process in many countries (if not all), the macroeconomic aspect involves the need for stabilization. Inflation, whether open or repressed, needs to be reduced to an acceptably low level. Microeconomic issues also need to be addressed, in particular the numerous distortions that exist in the economy because of previously prevailing price and other types of controls. Both aspects of the economic imbalance need to be confronted, as one cannot be solved unless the other is addressed as well. Policymakers must decide on the measures that will be included in the reform process and determine the role exchange rate management will play in it.
In terms of the macroeconomic aspects of exchange rate policy, one line of reasoning maintains that the exchange rate can be used as an instrument to control inflation and thus to obtain a measure of domestic price stability. This reasoning is behind those approaches that use the exchange rate as a nominal anchor for the economy by fixing the rate at a certain level in the hope that it will help bring inflation under control. Since price liberalization is also taking place in many transition economies, the exchange rate is often anchored after the initial price adjustments have taken place in order to ensure that the level at which the rate has been fixed can be sustained.
Of course, using a nominal anchor means that any relative price distortions still prevailing in the economy will have to be resolved through adjustments in other variables. An interesting empirical issue arises here, because a strong argument can be made that a flexible exchange rate in these economies is a better instrument for attaining and maintaining the competitiveness that is a key element in successful reforms. Actually, it can be (and has been) contended that the only way to assure competitiveness when inflation is high is through a flexible exchange rate. As always, there are good arguments on both sides of the issue; therefore, there is no single “right” answer.
However, I do have a preferred answer, even though it often proves contentious. An exchange rate anchor, once set, helps not only to control the rate of change in the overall price level but also to translate the structure of relative prices prevailing in world markets into domestic currency terms. This factor is important in transition economies, because most of them have highly distorted relative price structures. Thus, an anchor’s indicative value in providing a link between the relative price structure abroad and the corresponding structure at home is an important advantage that is not available—at least, not in the same way—with a flexible exchange rate.
However, the real issue in choosing an exchange regime is the potential impact that regime will have on output and employment. Clearly, if prices and costs were flexible both upward and downward, as is typically posited in the context of a fixed exchange rate, the choice of an exchange rate system would not be so relevant. The exchange rate could be fixed, or it could float, because if all other policies were right, the output and the employment levels would remain essentially the same under either system. But prices and costs—especially wages—in transition and other economies tend to be rigid in the downward direction. This empirical finding is probably the strongest argument that can be made in favor of exchange rate flexibility during the reform process. To the extent that domestic prices, domestic costs, and domestic wages are downwardly rigid, a fixed exchange rate policy will likely be difficult to sustain.
Whatever the situation, even when prices and wages are downwardly rigid, one thing is clear: exchange rate flexibility cannot substitute for appropriateness in other economic policies. In other words, a government can try to use the exchange rate as an escape valve for inadequate domestic financial policies but, in doing so, risks increasing the incentives for capital flight. And when capital flees the country, the government will have trouble maintaining those policies. In principle, there is little, if any, trade-off between the degree of exchange rate flexibility and appropriate domestic policies.
The argument often made in favor of a flexible exchange rate in the presence of downward rigidity in domestic prices and costs is based on what economists call “money illusion.” This proposition argues that cutting real wages during an adjustment effort is not acceptable if the process involves nominal wage reductions but that similar real wage cuts are tolerable if they come about through exchange rate depreciation. If this phenomenon—money illusion—is at work, the question soon becomes how long it is likely to persist. If economic agents were rational, the real issues in adjustment would be the real wage, the real factor price, and the output and employment levels that could be sustained; how these are brought about (that is, through domestic wage or exchange rate flexibility) would seem less important. This reasoning suggests that in the end, the ability of flexible exchange rates to safeguard competitiveness is not unlimited.
The exchange rate regime has implications for other policies as well, particularly for monetary policy. When the exchange rate is fixed, monetary policy independence lessens. Essentially, the exchange rate, through its impact on balance of payments outcomes, will determine the rate of monetary expansion. The transmission mechanism operates this way because, if the domestic source of monetary expansion—central bank or banking system lending—deviates from the desired change in cash balance holdings, balance of payments deficits or surpluses will eliminate the difference through movements in international reserves.
When the exchange rate is flexible—and here is where the argument for independence applies—essentially there is only one source of domestic expansion, which is central bank or banking system lending. This is because, if the exchange rate is allowed to float freely, the balance of payments is always in equilibrium, and the level of international reserves will neither rise nor fall. Clearly, reality is never that clear-cut; no country allows its exchange rate to float that freely, nor does any economy fix its exchange rate that permanently. So balance of payments outcomes generally do have some effects on the domestic money stock.
Options in Exchange Rate Management
What are the choices, then, for exchange rate management? Since no country either fixes or floats its exchange rate indefinitely or entirely cleanly, examples abound of many “combination” exchange rate regimes. Basically fixed regimes may peg to a major currency or a basket of currencies; peg within a very narrow margin to either another currency or a basket; or maintain an adjustable peg (the Bretton Woods system). There may even be a preannounced—but fixed—rate of adjustment. I call this last case “fixed” because, when the rate of change is preannounced, the exchange rate level will vary over time, but the rate at which it changes remains the same. Conceptually, this type of arrangement is fixed.
Among systems that are basically flexible are the freely floating system; those that allow the exchange rate to move on the basis of changes in certain indicators; systems with a rate that is allowed to move within very wide margins (relative to a certain currency or a basket of currencies); and systems with a rate that is allowed to move in response to the balance of payments.
Among the basically fixed exchange rate systems is the currency board regime, which is often discussed in the context of the reform process and which Estonia, for example, has adopted. With a currency board arrangement, the only source of monetary expansion in the system is through the central bank’s purchases of foreign exchange. The central bank extends no credit; instead, it has a single window for buying or selling foreign currency at a set price in terms of domestic money. The merits of this procedure are its simplicity and transparency. The presumption is that since the rest of the economy is willing to sell its foreign exchange holdings in order to purchase domestic currency, increases in the money supply will not be inflationary. According to this theory, people must want to hold domestic currency if they are willing to part with foreign exchange to obtain such currency.
The disadvantage of this arrangement is that, like a fixed exchange rate, it is easy to introduce but not to maintain without appropriate support from other policies. It requires an adequate stock of international reserves, particularly if the initial conditions in which the currency board is set are not balanced. In addition, if questions exist about the direction of economic policy, or if the reform package is not fully credible, then the currency board system may put pressure on the economy, as it may lead to heavy selling of domestic currency for foreign exchange rather than the opposite.
Incidentally, the currency board option is the external modality of an old proposal for domestic banking systems. Many years back, at the University of Chicago, it was argued that the optimal banking system was one in which a 100 percent reserve requirement prevailed. The idea underlying this proposal was that the two primary banking functions—deposit taking and lending—should be separated to ensure stability. The only way to guarantee that deposits were completely safe was to have a 100 percent reserve requirement, thereby preventing deposit banks from lending. Only investment banks would undertake risk lending, and their liabilities would not be guaranteed. With this separation of banking activities, the public would also become more aware of the risks involved in bank lending and, it was argued, would act accordingly—that is, people would be careful in choosing the banks where they would place their funds. In theory, this idea has a good deal of appeal, but it has yet to be put into practice.
Currency convertibility is an important part of the reform process. How has the IMF approached this issue over time? When the IMF was first set up, the world economy was characterized by severely disrupted trade flows, and the institution’s aim was therefore twofold: to restore these flows and to liberalize the trading systems of member countries. The emphasis in the IMF Articles of Agreement was on establishing convertibility of the national currencies with regard to current account transfers and payments—that is, on liberalizing current account transactions, including trade. But there was no similar commitment with regard to capital account convertibility, for two reasons. First, as I have mentioned, initially the IMF’s primary objective was to restore freedom to trade and current account transactions. The second reason related to the need for a degree of national economic policy independence. Policymakers had two fears at the time: that free capital mobility could lead to capital flight, and that it would also reduce the degree of national economic policy independence.
The policy environment of the early postwar period was quite different from what it is today. Then, the government was perceived as the primary agent capable of handling the economy and was expected to stabilize economic cycles, intervening to jump start the economy when it was flagging and to hold it back when it heated up. There was a certain amount of faith in the government’s ability to steer the economy in the right direction, and perhaps by implication, there was some suspicion of the workings of market forces. By contrast, in the current world environment, the policy environment is almost the opposite. Most countries have moved in the direction of becoming free-market economies or, at the least, have implemented significant market reforms. And the free play of market forces places constraints on economic policy and on the size of government, as well.
This fact is clear in the transition economies of Eastern Europe and the former Soviet Union. For example, although the Bretton Woods system was established in the mid-1940s, it was not until the late 1950s and early 1960s that the industrial countries were able to establish current account convertibility—a fairly long time. In contrast, most reforming economies of the 1990s have made convertibility an urgent priority in their reform programs. The urgency these countries attach to establishing convertibility for their currencies is an important aspect of the reforms that cannot wait for 15 or 20 years, as it did after Bretton Woods.
Another difference in today’s policy environment is that in many reforming economies, the notion of convertibility is a complete one, including capital transactions. I, for one, have never been comfortable with the idea of aiming for current account convertibility without also aiming for capital account convertibility. My unease about the narrower approach is both logical and institutional. Logically, capital controls are just as distortionary as any other type of restrictions, so that I argue against capital controls just as strongly as I argue against trade and current account controls. From an institutional standpoint, it is awkward to stress the merits of free trade in relation to current transactions and not in relation to capital transactions, especially in today’s world. In the West, at least, there is now a much larger degree of capital mobility than there was in the 1940s, 1950s, or even the 1960s.
Capital flows are important, in part because of the constraints they impose on national economic policy. The events in the EMS at the end of 1992 and in mid-1993 provide a clear example of the relevance of the constraints of free capital flows in national economic policy settings. Nevertheless, a trend has prevailed for some time as governments adopt policies and make other efforts aimed at promoting market forces. Both domestic financial systems and international capital transactions have been liberalized, resulting in a truly global capital market. This market has, of course, limited countries’ ability to use national policies for purely domestic objectives without regard to international considerations.
How quickly should convertibility be implemented? The answer to this question lies in the sequencing and pacing of appropriate reforms. In terms of sequencing, policymakers must decide which to liberalize first: the domestic financial sector or international capital movements. An argument can be made here for external liberalization that is similar to the one I made in favor of fixing the exchange rate in order to improve relative price structures. A fixed exchange rate creates a link between the domestic price structure and world market prices. Likewise, liberalizing external capital transactions introduces the terms of the international financial system into the domestic system. In this sense, liberalizing capital transactions can help the process of overall domestic financial liberalization that the transition economies have undertaken.
Issues of sequencing involve a great deal of judgment. Again, there is no “right” answer to the question of the order in which reforms should be introduced, since all countries begin from different initial conditions. This problem leads directly to the issue of pacing. Some economists argue that no constraints—be they exchange rate fixity or capital account liberalization—should be placed on domestic policymaking until the time is right. And the right time is often understood to be the moment when there is no domestic imbalance, the external payments position is strong, and inflation is under control—essentially when the economy is finally sound. Certainly this argument has its merits. However, waiting until all the fundamentals are optimal is likely to mean that nothing is ever done, and thus the question of constraints need never arise. The opposite school of thought is that constraints should actually be used as a stimulus for domestic policy action. If a country with an imbalance, high inflation, and a weak balance of payments fixes its exchange rate and opens its capital account, the constraints placed on the economy will be so tight that policies will have to adapt to them.
The optimal path probably lies somewhere between these two schools of thought. However, my own preference would be not to wait too long. I tend to be among those who argue for relatively fast-paced policy actions during the reform process, but the costs of moving quickly are difficult to measure. Weighing the merits of gradualism against the merits of speed often means choosing between reduced output, employment, and growth today and reduced output, employment, and growth tomorrow. Gradualism argues for policies that do not have an overly adverse impact on certain basic objectives of economic policy now. The key question is whether protecting the status quo has a price tag attached. And to answer that question, it is necessary to understand clearly whether gradualism really encourages persistence in the right policies or simply urges delays in their implementation.
One very recent example of this quandary can be seen in developments within the EMS between the fall of 1987 and September 1992. During these five years, the policies of the participating countries were not sufficiently convergent to maintain consistency with the currency exchange rate grid. Fixing exchange rates for a period as long as five years implies that for this period, all participating countries are operating as though they were in a currency union. And yet the policies of the individual countries within the EMS were not consistent with that particular constraint. It can be argued that the markets allowed for a five-year period as a means of ensuring that the European governments had time to align their policies and take the actions required to make their economic policy grid consistent with their currency grid. And yet it is clear that all the markets saw during that period was continual delays of the necessary adjustments.
My perception that this type of foot-dragging is what gradualism often leads to explains why I am for speed and for applying constraints—including exchange rate constraints—relatively rapidly. The move from central planning to a market-based system, especially in the former Soviet Union and Eastern Europe, offers even stronger support for this position. The essence of reform there is to redefine the role of government and reintroduce market forces in all sectors of the economy. Thus, many goals must be pursued simultaneously. If some reforms are postponed, the markets receive signals indicating that certain things will not change, and these signals bring the credibility of the reform process into question.
And typically people challenge these things. For example, if reform requires that enterprises be subjected to budget discipline, and such discipline is introduced only gradually, people will wonder if it will ever be established. My preference is for clear signals. And from that perspective, in talking about exchange rate management, a fixed exchange rate is a very clear signal. One thing, however, should be clear with both fixed and flexible exchange rates: exchange rate flexibility does not allow other necessary policies more margin to depart from their appropriate path than a fixed exchange rate does.
My remarks will focus on the issue of fixed versus flexible exchange rates. On a theoretical basis, it is difficult to disagree with Manuel Guitián. In the real world, however, I believe few countries actually make the choice between a fixed and a floating exchange rate on the basis of whether it is better to be insulated from or to accept the constraints of the international economy.
On the side of fixed exchange rates are the members of the CFA franc zone, a very special monetary union supported by the French Treasury, and the core members of the European Monetary System, which is striving for monetary union. On the other side is the exchange rate system that incorporates the dollar, the deutsche mark, and the yen. It is practically impossible to fix exchange rates among these key currencies because of the magnitude of capital flows. For the countries involved, floating is not so much a choice made on a doctrinal basis as a fact of life. In practically all other cases, choices are made on a pragmatic basis. Some countries, like the Scandinavian economies, allow their currencies to float but maintain that pegs will be established as soon as possible. Other countries have pegs but change them every few years—in some cases every few months or even every few weeks. In this pragmatic world, the distinction between fixed and flexible exchange rates is a tenuous one indeed.
There is no simple answer to the question of whether an economy gains much in terms of discipline by pegging. If it is thought that the exchange rate is going to be pegged for a few months or a few years at best, the resulting discipline is rather limited.
What are the factors that determine whether a country is closer to pegging than to floating? The first factor is how much money it has—i.e., how large its foreign exchange reserves are. For a country like the United States or Germany, the cost of maintaining foreign exchange reserves may be close to zero, but for a smaller, poorer country, the cost of holding foreign exchange reserves is high. A country with very low reserves cannot peg, even for a short time, without introducing restrictions on foreign trade and capital flows, among other things. In general, a country considering pegging should ask whether its circumstances will allow it to choose a reasonable exchange rate that can be maintained for some time. Without the resources to defend its rate, a country’s decision to peg is not very meaningful.
Under what circumstances might a country be able to peg its exchange rate, at least for a certain period? Let me take an easy case: a relatively low-inflation country, such as Morocco, that is not experiencing abrupt structural changes and with terms of trade that vary occasionally but are not subject to wide swings. For some years, Morocco has had a conservative monetary policy. It has experienced some changes in its terms of trade, but its export receipts from manufacturers, its tourism earnings, and its remittances from Moroccans abroad have provided a stable flow of foreign exchange. Over time, its foreign exchange reserves have become sizable. A country like Morocco can choose a reasonable rate for pegging that can be maintained for some time. The stability of the exchange rate may, in turn, strengthen the overall policy framework. An additional factor contributing to the success of Morocco’s pegging was the authorities’ willingness to change the peg when circumstances warranted it (for example, in the late 1980s).
Let us consider more common cases. A country with an inflation rate far higher than those in the industrial world—as much as five or even ten times higher—may decide to peg and, simultaneously, to adopt tight monetary and budgetary policies to cut the inflation rate sharply. While it may take a long time to reduce inflation to levels enjoyed by Germany and the United States, the expectation is that any loss of competitiveness during this effort will be regained later on by driving the inflation rate even lower than it is in Germany or the United States. Now, it would be a major achievement for a country with an annual inflation rate of 10–50 percent to reduce that rate to industrial world inflation rates relatively quickly; in fact, only a few countries have ever succeeded. To plan to drive the inflation rate even lower to regain what has been lost during adjustment is unrealistic. Thus, in my opinion, a country with a high inflation rate will find that pegging its exchange rate as a source of discipline will not succeed.
Even if a country’s inflation rate is not extremely high, the economy may be experiencing structural changes that make the choice of a pegged rate a poor one. Take the extreme case of the Central and East European countries, or of some African countries, all of which are undergoing a strenuous process of structural change. In such cases, the appropriate real exchange rate—that is, the amount of price competitiveness needed—is going to change drastically during this process. In most Central and East European countries, and even in many African countries, the real exchange rate that was appropriate at the beginning of the process has very little to do with the real exchange rate that is appropriate now or that will be suitable in three years.
A country may try to fix the nominal exchange rate and change the real rate by reducing inflation. My argument in this case is much the same as it was in the earlier one. Few countries can be sure that they will be able to reduce inflation sufficiently. The task is made even more difficult because it must be carried out in the presence of additional problems—exports that are mainly primary commodities, for instance, and, as I mentioned, terms of trade that change by extremely large amounts from year to year.
My viewpoint may be too negative, and I have already mentioned one country, Morocco, where pegging worked reasonably well. There are also countries where, because of abnormal circumstances, the real exchange rate initially depreciated by an extremely large amount, probably overshooting. In such cases, it may be useful to peg the nominal exchange rate as part of a stabilization program. Why? Because for a period of time, even though the inflation rate remains high, real exchange rate appreciation may correspond to the gradual correction of the initial overshooting. A stable rate may help the overall stabilization effort without giving rise to a balance of payments problem.
The last point I would like to make is that even when circumstances appear favorable, monetary authorities need to think very carefully before pegging the exchange rate. The key challenge in Central and Eastern Europe, and in the developing world, is to expand and diversify exports. With a pegged exchange rate, countries tend to resist devaluing because of the political costs that may be involved. Ultimately, the rate is changed, but often the change is long in coming. Therefore, for a sustained period, these countries find themselves with overvalued exchange rates that damage efforts to expand and diversify exports.
Jacques Artus’s comments indicate clearly that economists are not all of the same view on this subject, as they are not on many other topics. However, I find that on practical grounds, I do not really disagree with what Mr. Artus says. One of the reasons I stick to theory is that theory is much easier to deal with than reality. It is clear that the choice of an exchange rate regime will have to reflect the conditions in which the country finds itself. If a country has a weak balance of payments, a large imbalance, or no reserves, there is no point at all in trying to fix the exchange rate.
My sense of the advantages of fixed rates really applies mainly to large countries, even though I agree that among the U.S. dollar, the yen, and the European currencies, exchange rate fixity does not seem to be on the horizon for the near future. However, I do see merit in fixing the relation between these major currencies and the associated commitment of the countries involved to harmonize their national policies with the requirements of the international system.
In the case of a small open economy, the availability of the choice of whether to float or to fix is in fact more apparent than real: such an economy reflects conditions in the rest of the world, no matter what its exchange rate regime. In a mediumsized economy, however, the choice does matter, and in practical terms, everything Mr. Artus has said applies. The only points I would make in this regard would be the following. If, in a specific country, the flexibility provided by floating promotes structural reform; absorbs some terms of trade changes; helps diversify export products; and achieves all these objectives simultaneously, the implication is that the country is following the right policies and, by all means, should keep its flexible exchange rate. However, to the extent that a floating rate allows a country to maintain policies that are not conducive to adjusting to changes in the terms of trade, diversifying exports, and making all the important real changes in the economy, then exchange rate flexibility is certainly less appealing.
Most of the problems of countries that have experienced high inflation are typically not the result of fixed exchange rates, but of other policies. If a floating exchange rate helps correct those policies and eventually becomes stable, then it should be used. But my emphasis is much more on the policies a country is following and the extent to which those policies have resulted in low reserve levels and a lack of diversification in exports. And again, I agree with what Mr. Artus has said: that in many cases overvaluation has been part of the problem and that devaluations, which are occasionally necessary under fixed regimes, have a political cost. I also do not advocate using fixed exchange rates to establish credibility, exploit it, and then have it vanish when the peg cannot be maintained. Choosing to fix an exchange rate is a very fundamental decision that requires flexibility in domestic costs and wages. Mr. Artus has said that such flexibility is typically nonexistent and that the success of a peg therefore depends on the economy’s ability to increase productivity, a task that may be difficult. Essentially, we are not in disagreement, even though we have different perspectives. To use an old adage, I look at the glass, and it is half full; he looks at the glass, and it is half empty.
The last point I would make is that in today’s world, the importance of market forces has become so great that economic policy in most countries is very tightly constrained. Thus, the real issue under discussion involves more than the choice of a flexible exchange rate. If the conditions in a country are such that a fixed exchange rate will help control inflation—causing the economy to become somewhat less competitive for a time, at least until appropriate policies restore competitiveness—then, in some circumstances, a peg should be used. Actually, if limited loss of competitiveness helps instill financial discipline or becomes the means of creating greater flexibility in labor markets, a fixed exchange rate is appropriate. However, if the loss of competitiveness is too great and cannot be offset with wage flexibility or increased productivity, then clearly fixing the exchange rate is not going to be useful.
In essence, the issue is that in fixing its exchange rate, a country is giving up one avenue of adjustment. And in giving up one avenue of adjustment, it has to find another variable of adjustment, typically an internal variable. If a country cannot find this variable, it cannot fix the rate. The issue is that simple.
The discussion reflected the two approaches to exchange rate management as presented by Manuel Guitián and Jacques Artus, with the two speakers continuing to argue for their respective points of view. Participants were interested in the fact that Mr. Guitián and Mr. Artus were able to arrive at radically different policy conclusions while seeming to agree with each other’s analyses. This apparent anomaly was found to be mainly the result of the difference between approaches to exchange rate management: Mr. Guitián bases his policy conclusions strictly on theoretical considerations, while Mr. Artus takes into account a number of practical aspects.
The discussion then turned to capital account liberalization. It was pointed out that countries thinking about liberalizing their capital accounts need to consider not only their domestic economic conditions but also the potentially disruptive circumstances of the world economy. Mr. Guitián did not see this uncertainty as a reason for delaying capital account liberalization in some countries, such as Egypt, where most of the domestic fundamentals seem correct. Rather, he argued that liberalization programs should be designed to limit some of the dangers. Mr. Artus argued that given the problems in the world economy, the real question to ask in a country like Egypt is whether the domestic economy has anything to gain by retaining capital account restrictions. In general, he asserted, the answer will be no.
This topic led directly to a discussion of the potentially destabilizing effects of short-term capital flows. Some participants maintained that the possibility of destabilization posed by these flows is another reason for countries to take a gradual and cautious approach when opening up their capital accounts. Mr. Guitián disagreed, arguing that capital account restrictions are the equivalent of having more than one exchange rate—a well-known source of inefficiency. In support of this proposition, some participants argued that short-term flows typically reflect the fundamentals of a country’s economic situation. However, this argument was countered with examples such as the attack on the French franc in 1992–93. These attacks, it was maintained, should sound a cautionary note for other countries.
On the issue of exchange rate pegging, it was pointed out that a nominal peg will not be credible in high-inflation countries like Russia; in fact, in these countries, such a peg will only make it easy for speculators to earn huge profits. Even economies with relatively stable prices, such as the United Kingdom, will not be able to maintain a nominal peg without appropriate, sustainable supporting policies. The United Kingdom has not pegged its nominal exchange rate within the European Monetary System since the markets challenged the sustainability of the policies necessary to defend the peg. It was pointed out that a far more credible course for countries with moderate inflation is a regime based on a crawling or adjustable peg. However, more than one participant mentioned the problems involved in determining both the level of the peg and the speed of the crawl.
Some participants criticized Estonia’s currency board arrangement, which Mr. Guitián had mentioned favorably. Although it has succeeded in halting inflation, the cost seems unacceptably high—about 50 percent of total output. The related issue of squaring export incentives with macroeconomic concerns was also debated. Participants agreed that a policy of establishing currency convertibility early in the reform process and then adjusting the exchange rate will result in lower costs in terms of real output than a policy that attacks inflation through a pegged exchange rate. Bolivia’s recent success was cited here, as was the failure of policy in some neighboring countries. Again, appropriate supporting policies must be in place, as they were in Bolivia. It was also noted that credible long-term price incentives are essential to a structural adjustment based on exports. However, in some circumstances, long-term credibility might well be served by the maintenance of a relatively fixed exchange rate. The need to be competitive does not necessarily argue for a flexible rate.
Argentina was mentioned as an example of a country that fixes its exchange rate primarily to provide an anchor for inflation. This type of regime de-emphasizes the role the exchange rate can play in improving and maintaining competitiveness. Some participants questioned the wisdom of fixing the exchange rate and allowing other policies, such as domestic labor market reforms, to assume responsibility for ensuring economic competitiveness. Mr. Artus observed that this approach can work, but only if initial economic conditions and relevant political circumstances are appropriate. Unfortunately, he added, it has often been adopted by countries that do not meet these criteria and so have little chance of using it successfully. Generalizing about this approach on the basis of the very few cases in which it has succeeded is dangerous.
One contributor summed up the discussion by observing that the advice seemed to be that countries should remain flexible as long as they must but should peg as soon as they can.