Optimum Currency Area (OCA) Theory the Effectiveness of Exchange Rate Changes As a Mac- Optimum Currency Area (OCA) Theory Originates Roeconomic Policy Instrument

Copyrighted Material

however, and the move to European Monetary theory (OCA) area currency optimum Union (EMU), which began in the early 1970s, generated practical interest in currency areas that has continued to grow. Despite the huge subsequent literature and increased practical applicability, the theory of currency areas has remained centered on the Mundell- McKinnon approach of identifying the area within which the macroeconomic flexibility gains from having an independent currency are significant. This article focuses on OCA theory and the empirical evidence, while the complementary article on common currency analyzes actual experience with shared currencies. Variants of OCA Theory In OCA theory the key issue is the extent to which macroeconomic policy can be effective in an open economy. An external shock, such as a recession induced by a drop in export demand, can be countered by cutting interest rates to stimulate investment. An economy with capital mobility and a fixed exchange rate does not have monetary policy independence, however, because the cut in interest rates will lead to capital outflows and an unsustainable balance of payments deficit. Mundell (1961) identified the point at which currency independence becomes worthwhile as the point at which factors cease to be mobile. The boundary of the OCA is set by breaks in factor mobility, that is, points at which the movement of labor or capital becomes more difficult. The openness of an economy also can determine optimum currency area (OCA) theory the effectiveness of exchange rate changes as a mac- Optimum currency area (OCA) theory originates roeconomic policy instrument. In very open econo- from two seminal articles in the early 1960s by the mies, currency devaluation is ineffective because economists Mundell (1961) and McKinnon (1963). prices and wages immediately increase to remove any These articles drew on contemporary debates about competitive advantage. Openness undermines the fixed versus flexible exchange rates, treating a com- money illusion (or failure to distinguish nominal mon currency as the extreme case of a fixed exchange from real price changes) that permits the exchange rate. At the time, under the Bretton Woods system of rate to be an effective policy instrument. Larger fixed exchange rates, the choice of exchange rate was economies have lower ratios of trade to gross do- seen as a theoretical rather than practical issue, and mestic product (GDP), and currency devaluation applicability of the OCA analysis to the choice of can have a positive impact on output. Thus very currency domain was limited by the almost universal open economies are suboptimal currency areas, one country–one currency rule (Pomfret 2005). The but for a larger and less open economy the effec- shift to generalized floating of the major currencies, tiveness of the exchange rate as a policy tool increases;

869 Copyrighted Material

piu urnyae OA theory (OCA) area currency optimum at some point the currency area becomes optimal currency area becomes large enough to have well- (McKinnon 1963). functioning foreign exchange (including forward) In the decades following the original contribu- markets. Thus the increasing efficiency of financial tions of Mundell and McKinnon, subsequent writers markets becomes an argument for OCAs becoming lengthened the listofcriteria that might be relevant to smaller because it reduces the benefits via reduced an assessment of the optimality of a currency area. transactions costs of a common currency. This may Recent contributions have argued that the trade-off explain why it is feasible to have a small currency may be mediated by history and by geography domain, as in countries such as Iceland or New (Alesina and Barro 2002). With multiple criteria a Zealand, without incurring huge costs, but whether unique ranking of suitability for currency union is the trade-off between macropolicy flexibility and unlikely to exist and the theory becomes difficult to transactions costs is optimal for these small states is test, but the essential structure of OCA theory has another question, and one which is impossible to remained remarkably robust since 1961. answer without a good understanding of the nature OCA theory characterizes the choice of currency and size of the transactions costs associated with in- area as a cost-benefit analysis trading off microeco- dependent currencies. Before the beginning of the nomic efficiency against macroeconomic flexibility 21st century little research had been done on the (Krugman 1993). Microeconomic efficiency would transactions cost benefits of a common currency, but be maximized with a global currency. Thus, sub- this has changed since the controversial article by the global OCAs imply the existence of distortions, such economist Andrew Rose was published in 2000. as nonclearing labor markets that lead to involuntary Testing OCA Theory The OCA literature has unemployment, whose negative effects can be re- dominated the theoretical explanation of currency duced by macroeconomic policy. Another benefit of areas for almost a half century. There have, however, a larger currency area is that disturbances are likely to been few systematic tests of OCA theory and little be offsetting, so that exchange rate changes are positive support for OCA theory as a useful way of smaller, with less feedback on domestic prices. The explaining the composition of existing currency areas greater price stability is usually ascribed to random or of predicting changes in currency domains. One shocks being offsetting, but other mechanisms in- empirical challenge was the rarity of changes in cur- clude reduction in the impact of outliers in the rency areas during the 1970s and 1980s. In the early consumer price index, or CPI (i.e., goods with a 1990s it was argued that the stability of actual cur- particularly large weight in a region’s consumption rency arrangements may be explained by switching bundle), and reduction in the ratio of trade (or rather costs (Dowd and Greenaway 1993), but the many transactions denominated in potentially volatile changes in currency arrangements in Eastern Europe foreign currencies) to GDP. in the 1990s and the introduction of the euro in The cost-benefit perspective suggests an equal role Western Europe suggest that the mechanics of for microeconomic and macroeconomic analysis, changing currency arrangements are neither difficult but the currency area literature has been dominated nor especially costly. by macroeconomists, in part because the micro With multiple OCA criteria, the theory becomes benefits of reduced transactions costs from a com- difficult to test: a small open economy has the biggest mon currency are simple to visualize but difficult to potential gain from joining a larger currency area in measure. Although the benefit from a common order to reduce transactions costs, but it may also be currency (or fixed exchange rate) in terms of lower most vulnerable to external shocks and hence has the transactions costs has long been accepted as the most to lose from giving up the exchange rate as a overwhelming argument for mini states (such as macropolicy instrument. The economists Kreinin Luxembourg or Brunei) not to have independent and Heller (1974) synthesized the various criteria currencies, the argument becomes less potent as the into the single question of whether a country could

870 Copyrighted Material

better deal with external imbalance through devalu- seriesdataforcorrelationsbetweenchangingcurrency theory (OCA) area currency optimum ation or through adjustment of domestic demand. union status and bilateral trade flows, Glick and Rose Their conclusion was that Italy, Sweden, and Swit- (2002) estimatedthatdissolutionof a currency union zerland were the three Organisation for Economic halves bilateral trade. Currency union breakup is, Co-operation and Development countries most however, usually associated with other events that likely to abandon their national currencies. Thirty disrupt trade; out of some 60 cases of post-1947 years later only one of the three had done so, while currency union dissolutions in the Glick-Rose data ten of the ‘‘less likely’’ countries had abandoned their set,more thantwo-thirds broke upwithin a decade of national currencies. the end of a colonial relationship, and the end of the If currency areas become ‘‘optimal’’ ex post, then ruble zone, which is not in the data set, would in- OCA theory may be untestable. A common currency crease the percentage still further. In tranquil cur- might promote closer trade links and more syn- rency union changes, notably Ireland’s secession chronized cycles, both of which are OCA criteria; from its currency union with the United Kingdom closer trade ties increase the benefit from a common in 1979 and subsequent participation in the process currency, and synchronized cycles reduce the cost of leading to the euro, the impact on bilateral trade is giving up independent currencies. This is not a the- unclear. The weaker the link between currency union oretical result but a hypothesis to be tested empiri- and bilateral trade, the less convincing is the claim cally because more bilateral trade could promote that OCA criteria become self-fulfilling ex post. interindustry specialization and less synchronized In practice, irrespective of whether the criteria cycles. Using various measures of bilateral trade in- may become endogenous, most of the literature on tensity and cycle synchronization for 21 developed currency domains treats OCA theory as having pre- economies, Frankel and Rose (1998) find a robust dictive capability. Yet, the general track record of relationship between the two variables, which they OCA theory in explaining the monetary history of interpret as evidence that as a common currency the post-1945 international economic order has been promotes bilateral trade it also increases cycle syn- miserable. Despite the increasing openness of na- chronization. Thus actual currency areas fit OCA tional economies and increasing capital mobility, criteria betterthanpotentialcurrency unions,andthe both unambiguous pressures for larger currency areas OCA criteria are endogenous. according to OCA theory, the number of currencies The impact of currency union on bilateral trade has increased substantially and the geographical size flows has been the subject of a burgeoning litera- of currency domains has shrunk correspondingly. ture initiated by the economist Rose (2000). Using The sole significant exception is the introduction of a gravity model, Rose found that currency union has the euro, but in Europe and the former Soviet Union a large effect on bilateral trade, which he interpreted as a whole there were more currencies in 2002 than a as evidence that a common currency substantially decade earlier.Globally, over the last half century, the reduces transactions costs. Although it is plausible exogenous increase in the number of countries drove that a common currency reduces transactions costs the number and size of currency areas, and the OCA and stimulates trade, the magnitude of the common criteria were irrelevant to explaining this pattern. In currency effect is hotly debated. In Rose’s study the sum, although OCA theory has dominated the countries in currency unions are not from a ran- analysis of currency domains, the empirical support dom draw; several authors have shown that currency for the theory is weak. union members are smaller and more open than their natural comparators and that history (usually in the See also Bretton Woods system; capital mobility; common form of colonial background) matters. currency; dollar standard; dominant currency; euro; Eu- The debate has been conducted with analysis of ropean Monetary Union; exchange rate regimes; impos- currency area changes over time. Analyzing time- sible trinity; multiple currencies

871 Copyrighted Material

FURTHER READING Mundell, Robert. 1961. ‘‘Theory of Optimum Currency Alesina, Alberto, and Robert J. Barro. 2002. ‘‘Currency Areas.’’ American Economic Review 51: 657–65. The Unions.’’ Quarterly Journal of Economics (May): 409–36. original OCA theory. A recent restatement of OCA theory. Pomfret, Richard. 2005. ‘‘Currency Areas in Theory and De Grauwe, Paul. 2000. The Economics of Monetary Union. Practice.’’ Economic Record 81 (253): 166–76. A recent 4th rev. ed. Oxford: Oxford University Press. The best review of the evidence on OCA theory. textbook treatment of monetary union. Rose, Andrew. 2000. ‘‘One Money, One Market: The Ef- Dowd, Kevin, and David Greenaway. 1993. ‘‘Currency fect of Common Currencies on Trade.’’ Economic Policy Competition, Network Externalities, and Switching 30 (April): 9–45. An innovative and controversial cross- Costs: Towards an Alternative View of Optimum country study of the impact of a common currency on Currency Areas.’’ Economic Journal 103: 1180–89. An bilateral trade flows, and implicitly on transactions costs. influential explanation of the costs of changing currency Tower, Edward, and Thomas Willett. 1976. ‘‘The Theory arrangements, although to some extent countered by the of Optimum Currency Areas and Exchange-Rate Flex- relative lack of technical problems associated with the ibility.’’ Princeton Special Papers in International Eco- introduction of new currencies in the former Soviet nomics, No. 11. International Finance Section. Prince- Union and the introduction of the euro. ton, NJ: Princeton University (May). A classic survey of Frankel, Jeffrey, and Andrew Rose. 1998. ‘‘The En- OCA theory. dogeneity of the Optimum Currency Area Criteria.’’ RICHARD POMFRET Economic Journal 108 (July): 1009–25. An attempt to introduce systematic cross-country evidence into the debate about OCA criteria. Glick, Reuven, and Andrew Rose. 2002. ‘‘Does a Currency Union Affect Trade? The Time-Series Evidence.’’ European Economic Review 46 (June): 1125–51. A fol- low-up to Rose’s 2002 paper, which used cross-country data. Kreinin, Mordechai, and H. Robert Heller. 1974. ‘‘Ad- justment Costs, Optimal Currency Areas, and Interna- tional Reserves.’’ In International Trade and Finance: Essays in Honour of Jan Tinbergen, edited by Willy Sell- ekaerts. London: Macmillan, 127–40. A pioneering attempt to introduce empirical evidence into the debate over the OCA criteria, although limited to European countries. Krugman, Paul. 1993. ‘‘What Do We Need to Know about the International Monetary System?’’ Essays in Interna- tional Economics No. 190. International Economics Section. Princeton, NJ: Princeton University (July). An important reminder not to neglect transactions costs and to remember the microeconomic benefits from a well- functioning international monetary system (and a common currency). McKinnon, Ronald. 1963. ‘‘Optimum Currency Areas.’’ American Economic Review 53: 717–25. One of the two articles initiating OCA theory.

872