GEORGE S. TAVLAS
The international monetary system is characterized by a wide diversity of exchange rate arrangements. For example, in the aftermath of the breakup of the states of the former USSR, a key issue has been whether these countries should maintain the Russian ruble as their currency. Should they issue their own currency and allow it to float freely against other currencies, or should they adopt an exchange rate arrangement somewhere between these two alternatives (such as the limited flexibility of the currencies participating in the exchange rate mechanism of the European Monetary System)? This issue has revived the theory of optimum currency areas, once dismissed as primarily a scholastic discussion, because it offers guidance to countries choosing an exchange rate arrangement and sets forth the costs and benefits associated with various arrangements. In One
Market, One Money(1991), which evaluated the desirability of all countries of the European Community (EC) moving to a single currency, the EC used the optimum currency area approach as its standard of reference.
As the theory of optimum currency areas has taken on increased relevance, it has also been modified to take into account the lessons of recent economic history as well as recent contributions in the professional literature. This article will discuss what some of these changes are, and how the “new” theory might apply to countries faced with the decision as to whether or not to form a currency union.
Why the surge of interest?
By enumerating a set of criteria to guide countries in their choice of exchange arrangements and by stipulating the benefits and costs associated with different kinds of arrangements, this approach attempts to identify the optimal geographic area within which countries should adopt a monetary union. A monetary union is typically defined as the adoption among the participating countries of either a single currency or separate currencies with the following exchange rate arrangements: (1) the irrevocable fixing of parity rates; (2) the elimination of margins of fluctuation between or among exchange rates; (3) the total and irreversible convertibility of currencies (i.e., absence of exchange controls); and (4) the complete liberalization of both current and capital transactions. The optimum size for a currency area is defined in terms of attaining the macroeconomic goals of internal balance (low inflation and low unemployment) and external balance (a sustainable balance of payments position). That is, for a given country, the approach addresses the issue of what kind of exchange rate arrangement would be more likely to achieve low levels of inflation and unemployment, as well as external balance.
What accounts for the revival of interest in this subject? Two broad sets of factors have been primarily responsible for this outcome. The first set reflects developments on the international monetary scene, including:
the reinvigoration during the late 1980s and early 1990s of the movement toward European monetary integration with the aim of creating a single European currency by the end of the present decade;
the emergence of intense pressures within the exchange rate mechanism (ERM) of the European Monetary System in late 1992 and in early 1993, including the suspension of several currencies from the system and the first ERM realignments since January 1987, reignit-ing the debate over which countries should initially be included in the system if Europe is to proceed with monetary unification;
the breakup of the USSR and the issue of the demarcation of the optimum currency domain among the states of the former USSR; and
the emergence of a tripolar international monetary system, centered on the US dollar, the deutsche mark, and (to a lesser extent) the Japanese yen, and the implications of such a system for world welfare and the conduct of macroeconomic policies among the currency blocs.
The second set of factors contributing to the surge of interest in the optimum currency area issue involves developments in macroeconomic theory. These developments have allowed the results of the original optimum currency area approach to be cast in a new light. There is much from the old approach that has withstood the test of time—for example, the criteria set forth as guides in choosing a single currency are still relevant and illuminating. But the earlier results on the costs of joining a currency area—such as a nation’s inability to choose the long-run trade-off between inflation and unemployment—have been seriously challenged in view of recent developments in macroeconomic theory.
The early literature on optimum currency areas was concerned primarily with two issues. First, it sought to delineate the structural characteristics of a national economy that would render exchange rate changes vis-à-vis the currencies of other countries either ineffective, or unnecessary. Second, it studied the costs and benefits to a nation participating in a currency area.
With regard to the former issue, several criteria are relevant for choosing the likely members of a currency union.
The degree of labor mobility. Labor mobility between countries reduces the necessity of exchange rate variation as a means of restoring external competitiveness and eliminating external imbalances, since the mobility of labor from high- to low-unemployment regions will tend to bring about a convergence of wages and other costs (factor-price equality) among the countries. Because the degree of labor mobility is likely to be related to geographic distance, this criterion is often used in support of currency unions among neighboring states.
The degree of financial market integration. A high degree of financial market integration lessens the need for changes in relative costs and prices among countries via exchange rate adjustment, since capital movements can finance inter-regional external imbalances. Such balance of payments financing helps cushion the adjustment process and facilitates a spreading out of the adjustment process over a longer period of time.
The openness and size of the economy. The more open the economy, the more domestic prices are likely to change in line with exchange rate variations, inducing possible changes in labor and other costs. Consequently, fixed exchange rate arrangements are preferable for open economies since nominal exchange rate changes in such economies are less likely to be accompanied by significant effects on real competitiveness. As a corollary, the smaller the size of the economy, the more open it is likely to be and, thus, the more inclined to join in a currency union.
The degree of commodity diversification. Highly diversified economies are viewed as better candidates for currency unions than are less diversified economies, as diversification provides some insulation from a variety of shocks, forestalling the need for frequent changes in relative prices via the exchange rate.
Price and wage flexibility. When prices and wages are flexible between regions, adjustment of imbalances among countries is less likely to be associated with unemployment in one region and inflation in another, diminishing the need for exchange rate adjustment.
The degree of goods market integration. Countries that possess similar production structures are prone to similar terms-of-trade shocks, negating the effectiveness of exchange rate adjustment between the countries as an instrument to deal with such shocks. Consequently, countries with similar production structures are deemed to be better candidates for currency unions than are countries whose production structures are markedly different.
Fiscal integration. The higher the level of fiscal integration between two areas, the greater their ability to smooth out diverse shocks through fiscal transfers from a low-unemployment region to a high-unemployment region. In turn, fiscal harmonization usually implies that the members of a monetary union also enter some form of political union.
Benefits and costs
In addition to delineating the characteristics that make membership in a currency union desirable, the early literature endeavored to enumerate the costs and benefits of a nation’s joining a currency area. Early writers noted, for example, that a benefit of adopting a single currency is that it eliminates exchange rate risk. This risk is equivalent to a transaction cost to a risk-averse trader, who will sometimes bear an explicit cost (such as buying a hedging instrument) to avoid it. Also, a single currency enhances the role of money as a unit of account and decreases other transactions costs, including the costs of information, search, and calculation.
Further, the adoption of a single currency would eliminate the need for firms to maintain staff to look after currency exchanges within the area. Other decreases in costs to be derived from the move to a single currency include those associated with: (1) economizing on exchange reserves through the pooling of reserves; (2) elimination of speculative capital flows within the area; and (3) enlargement of the foreign exchange market vis-à-vis other countries, decreasing both the volatility of prices and the ability of speculators to influence prices and, thus, to disrupt the conduct of monetary policy.
The main cost (in addition to the loss of the exchange rate tool) traditionally ascribed to joining a monetary union is the loss of monetary policy independence. That is, when a nation joins a monetary union, the formulation of monetary policy is no longer made at the national level but at the level of the monetary union. If a country happens to have a high unemployment rate relative to the other members of the union, it no longer has the option of easing monetary policy to lower its unemployment rate unless the other members agree to ease monetary policy throughout the union. If, as many early writers on this subject believed, there is a long-run trade-off between unemployment and inflation (the so-called Phillips curve), a nation pays a high price in joining a currency union in that it cannot use monetary policy as a means of achieving its desired mix between inflation and unemployment.
The “new” theory
In recent years, the theory of optimum currency areas has been modified in line with developments in other areas of macroeconomic theory. Two developments are particularly relevant to the discussion. The first concerns the formation of inflation expectations, which has implications for whether a long-run (or even a short-run) trade-off actually exists between inflation and unemployment. If such a trade-off does not exist, then it is preferable for nations to aim for a low rate of inflation. The second development—which is the time-inconsistency issue—provides guidance as to how monetary authorities can best make credible their commitment to reducing inflation. These developments seek to clarify the benefits and costs of currency area participation so that a country can “get it right” in deciding whether to join a currency area. For one drawback from “getting it wrong” is the high cost of dissolving the currency linkages later on.
The vertical Phillips curve and policy ineffectiveness. The early view of a permanent (i.e., long-run) trade-off between inflation and unemployment was undermined by the experience of the late 1970s and early 1980s, which saw the coexistence of rising unemployment and higher inflation in a number of countries. The implication of the view that the unemployment rate cannot be lowered by raising the rate of inflation is that the only benefit of floating exchange rates and independent monetary policies is the ability to choose a different rate of inflation from other countries with no effect on employment. If money supply changes do not have real long-run effects on the economy, then monetary policy is considered to be “neutral.” This is the key issue now in the EC. Both Italy and the United Kingdom left the ERM because they wanted to have an independent monetary policy for short-run cyclical reasons.
The nominal anchor issue. A monetary union between two countries, achieved for example through the fixing of exchange rates, imposes identical interest rates (except for a risk premium) in the two countries. There are, however, many levels of nominal interest rates—and, thus, levels of the money supply—that can fulfill the fixed exchange rate commitment. Accordingly, the countries must agree on how monetary policy should be conducted. In other words, there must be some institutional mechanism that determines the money stock in the system—that is, the need to provide a nominal anchor to the system. This issue has implications for the ability of countries that are members of a currency union to conduct anti-recessionary monetary policy, for example in the face of adverse terms-of-trade shocks.
Two kinds of arrangements can be used to pin down the money stock in a monetary union: symmetric systems in which the members cooperate in reaching a policy solution, and asymmetric systems in which one country takes a leadership role. While each of these arrangements can serve to determine the money supply in a monetary union, each has its own particular kinds of problems. Symmetric (i.e., cooperative) monetary unions entail the loss of the “safety-valve” provided by resort to either suspension of currency convertibility or restrictions on trade and capital flows. In instances in which the members of a cooperative monetary union are hit by different shocks, competitive pressures will build for the discontinuation of cooperation and the adoption of independent, national monetary policies. Indeed, a number of commentators have suggested that it was precisely such pressures, emanating from different shocks, that led to the suspension of the pound sterling and the Italian lira from the ERM.
Asymmetric unions can exacerbate the domestic business cycle in peripheral (i.e., the nonleading) countries. For example, assume that only a peripheral country is hit by a negative terms-of-trade shock and a corresponding contraction in real income. Its money supply should be allowed to expand. But participation in the union prohibits the monetary authorities from taking expansionary action. In addition, the contraction of real income results in a decline in the demand for money, causing interest rates to fall and leading to capital outflows. Since the center country does not change its monetary policy, it sterilizes the capital inflow, keeping its interest rates unchanged. Hence, the capital outflow in the peripheral country effects a contraction of its money supply. The country not only loses its ability to conduct an independent monetary policy but also experiences money supply effects that exacerbate the business cycle. Thus, an important precondition for following a common monetary policy depends critically on the nature and the mix of shocks to which the participant countries are likely to be exposed.
The time-inconsistency issue. A protracted record of relatively low inflation (and low inflation variability) depends upon credible and stable government policies, particularly monetary policy. A country whose authorities have a reputation for pursuing inflationary policies will find it difficult to shed that reputation without a long and costly process of disinflation. To gain credibility, authorities must pursue a policy rule that is time consistent. The time consistency literature argues that the public recognizes that policymakers have every reason to assert that they will aim for low inflation, but it also recognizes that policymakers have a strong incentive subsequently to renege on these commitments once the public accepts the assertion at face value.
The reason why policymakers may have reason to renege is that if it can generate an unexpectedly higher inflation rate, it can generate a lower unemployment rate in the short run (but not permanently) and it can write off some of its debt. According to this view, the public will not believe that low inflation will be maintained unless policymakers can provide evidence that they are following a nonreversible policy rule that will yield low inflation. That is, if a particular policy rule is expected to become suboptimal in a future period, economic agents will assume that the authorities will change the rule, even if they have announced that they will not change the rule. As a result, the rule is not time consistent and lacks credibility.
One way to gain credibility is by “tying the hands” of the government by some kind of institutional change. The most drastic change would be to abolish national monetary sovereignty by joining a union with a low in flation country. By so doing, it may be possible for the high inflation country to reap the benefits of a low inflation reputation, without any loss of output and employment. If indeed there is no permanent Phillips curve trade-off, the high inflation country has little, if anything, to lose and much to gain in the long run by joining a currency union with a low inflation country. While, in principle, other domestic commitment mechanisms to contain inflation can be devised (e.g., the announcement of monetary growth targets), many countries have not been able to create domestic monetary institutions with a credible commitment to price stability. These arguments, however, do not mean that, for a small- or medium-sized country, joining a monetary union with a low inflation country is necessarily the road to follow. As discussed earlier, it is also important that the small- or medium-sized countries possess similar characteristics to the anchor currency country. If they do, it might make sense to join a currency union. But if the structural characteristics are different, then joining a monetary union may not prove to be a lasting exchange rate arrangement.
To summarize …
The “new” theory of optimum currency areas uses the earlier approach as a point of departure, stressing the criteria relevant for choosing a single currency domain and enumerating the costs and benefits of a single currency. As did the earlier theory, the new theory stresses that if a country is highly integrated with a geographic area in factor mobility, financial transactions, and commodity trading, and if the country is small, open, and has a diversified production structure, then fixed exchange rates (or a single currency) for that area may reconcile internal and external balance more efficiently than flexible exchange rates. The adoption of a single currency would have the added benefit of setting economies of scale into play, such as the decline in transactions costs involved in having a single currency rather than several currencies.
In sum, it is unlikely that any one exchange rate regime will be appropriate for all countries. The theory of optimum currency areas identifies characteristics that can be used to help in the choice of an exchange rate arrangement. The approach indicates, for example, that the smaller European countries that are members of the European Monetary System have relatively strong incentives to adopt some form of fixity in their exchange rate arrangements, or even to join a monetary union, given that such countries are relatively small and open, possess relatively high factor mobility with each other, and, in many cases, share similar production structures. It is important to note, however, that the criteria underlying the optimum currency area approach do not always point in the same direction. In such instances, the approach does not suggest which criteria should be assigned the highest priority. Finally, it should be remembered, the success of any monetary union is also likely to depend to a significant degree upon the political commitment behind it.