North American Monetary Union

North American Monetary Union January 25, 2003 A report prepared by Woodrow Wilson School Graduate Policy Workshop 591D: Regional Currency Unions Princeton University Authors: Jungmin Lee Matina Madrick Claudia Maldonado Joy Milligan Bertrall Ross Ye Tao Juan Manuel Valle Facilitated by: Professor Peter B. Kenen

Table of Contents Executive Summary i I. Introduction 1 II. Exchange Rate Regime and the Theory of Optimum Currency Areas. 3 III. The United States Perspective 17 IV. The Mexican Perspective 27 V. The Canadian Perspective 47 VI. Conclusion... 65 VII. References 69 VIII. List of Interviews 73

Executive Summary In North America, NAFTA has been widely perceived as a success, and some observers have asked whether integration will be deepened. With the European model setting the precedent, an evolution from a free trade area to a customs union to a currency union seems to be viewed as the natural progression. Thus, the question arises, could North America ever become a currency union? The question briefly became a pressing one in Mexico following the 1994-1995 peso crisis. In the latter part of the 1990s, the steadily weakening Canadian dollar also led Canadians to consider the pros and cons of an alternative monetary arrangement. This report is the result of an investigation into the prospects for a currency union in North America. As well as reading the analytic literature on currency unions, we visited Mexico and Canada to speak with leading government officials, economists, and businesspeople about a currency union a term that we use generically throughout this report to denote the whole set of arrangements involving a common currency, ranging from a monetary union like EMU to unilateral dollarization. The focus of this report is on the Canadian and Mexican perspectives, in order to provide guidance to US policymakers if they should have to confront the issue of currency union in the future. The report attempts to answer the following questions: What does economic research tell us about the benefits and costs of a currency union? What form might it take in North America? What would be the implications for the United States? What is the present state of the debate in Mexico and Canada? What has been and should be the attitude of the United States? Exchange Rate Regimes The choice of exchange rate regimes affects countries monetary policies and cost of trade. Throughout the world, countries use a wide range of regimes, from floating regimes to i

Executive Summary intermediate regimes to fixed regimes. Currently, Canada, the United States, and Mexico all have floating exchange rates; each country conducts an independent monetary policy and has its own currency. Furthermore, the three countries enjoy free trade as a result of NAFTA. A deepening of NAFTA or change in the exchange rate regimes of North America would most likely result in a regional currency union. Furthermore, given the asymmetry in the sizes of the economies in North America and the lack of interest in creating a larger political community, dollarization, wherein Mexico and Canada unilaterally adopt the US dollar, is the most reasonable and likely type of currency union for the region. Is Dollarization the Only Option? Supporters of a currency union in North America usually use the European Monetary Union as benchmark. However, fundamental differences separate North America and Europe. The first and most important is that the European Monetary Union is not merely a consequence of economic integration or of the recognition of the economic benefits of a single currency, but the result of a wider political agreement. Neither our interviews in Canada and Mexico nor our review of the current literature revealed any similar political motive in North America. Furthermore, the largest country in the European Union accounts for only 23 percent of the area s GDP and the output per capita of that country is only 43 percent higher than that of the country with the lowest per capita income; in North America, by contrast, the United States accounts for almost 90 percent of the area s GDP and its output per capita is 284 percent that of Mexico. These two factors lead most observers to conclude that a currency union in North America would be simply the dollarization of Canada and Mexico. Costs and Benefits of a Currency Union The benefits of a currency union, such as dollarization, include the elimination of exchange rate risk and the reduction of transactions costs between trading partners. It has been ii

Executive Summary hypothesized that the elimination of currency risk and transactions costs can lead to large increases in trade, which would be a main objective of further integration [Frankel & Rose 2002]. The costs of giving up a national currency, however, include the loss of an independent monetary policy, a lack of exchange rate flexibility, a loss of seignorage income and the loss of a discount window. These costs would increase the vulnerability of the Mexican and Canadian economies to internal imbalances (output and employment fluctuations) and to external imbalances (current account fluctuations). What, then, are the conditions under which those countries could safely sacrifice exchange rate flexibility and independent monetary policies to reduce transactions costs and exchange rate risk? According to optimum currency area theory, countries that possess one or more of the following characteristics are best positioned to be able to give up an independent monetary policy in exchange for the benefits provided by a currency union: Synchronized business cycles and structural similarities ensuring similar responses to economic shocks; Domestic price and wage flexibility and/or high mobility of capital and labor across the borders within the currency union; High levels of trade with one another; Diversified domestic economies; A common fiscal transfer mechanism. Country Positions: The United States For its part, the United States has been passively neutral toward dollarization, adopting a policy of benign neglect [Cohen 2002]. The United States appears to be averse to assuming any responsibility for dollarized economies and is not ready to assume seignorage costs, the lender of last resort function, or a change in its policy domain. At the same time, the United iii

Executive Summary States does not actively discourage dollarization, because there are implied microeconomic benefits such as increased seignorage, the greater power and prestige associated with wider use of the US dollar, and increased trade with dollarized countries. Country Positions: Mexico At the time of the 1995 peso crisis in Mexico, an intense debate on unilateral dollarization or the adoption of a currency board was prevalent both in the private and public sectors. The perceived benefits of a credible monetary policy namely those associated with inflation control and a reduction in interest rates appeared to outweigh the expected costs of surrendering monetary autonomy as a buffer to asymmetric shocks and the loss of sovereignty. In response to the crisis, Mexico adopted a floating exchange rate and inflation targeting, accompanied by measures intended to strengthen the financial sector. As a result, the financial authorities and institutions were able to restore their credibility. Because of its relative success since the crisis, the current arrangement is now favored over a potential currency union. One of the main arguments in favor of the current regime and against dollarization, is the role of a flexible exchange rate as shock absorber, and the fact that it has been useful to the Mexican economy in the face of economic shocks. Although business cycles between Mexico and the United States have become increasingly aligned since the implementation of NAFTA, there are still significant differences in the economic structures of the two countries that make asymmetric shocks a real possibility. Dollarizing would necessarily mean the loss of the flexible exchange rate, and given the asymmetries between the two economies, many in Mexico are very reluctant to sacrifice exchange rate flexibility. There is also a consensus that, for Mexico, a currency union or dollarization is politically impossible in the near future. A monetary union not on the agenda of the Government or Congress, due to the political costs it would entail and the urgency of other forthcoming reforms. Mexicans see NAFTA as a first step towards a more modern and competitive economy, but they iv

Executive Summary believe that the next steps lie in the realm of internal structural reform and eventual harmonization, as opposed to changes in the exchange rate regime and monetary system. The objective of these reforms is to maintain Mexico s competitiveness as a manufacturing center and as a magnet for investors willing to use Mexico as a gateway to the US market. There are also several issues on the bilateral or trilateral agenda that Mexicans perceive as essential to the process of gradual integration and as necessary preconditions to any consideration of a currency union in North America: Greater mobility; Mechanisms to overcome the asymmetries of the three economies, such as development funds; Trilateral agencies or institutions to cope with common problems within a well-defined framework. Country Positions: Canada Despite the gains from integration, currency union as a potential next step is not currently on the Canadian agenda. During the late 1990s, some Canadian economists proposed that Canada either adopt the US dollar or fix its currency to the US dollar, with the ultimate goal of creating a joint US-Canada monetary system. But that plan has been highly controversial in Canada. The current system of floating exchange rates and inflation targeting has been hotly defended. The advocates of change say that the steadily weakening Canadian dollar and disappointing Canadian productivity growth are evidence that the floating exchange rate has done more harm than good. But most of the officials, economists, and businesspeople we interviewed praised the current monetary order and thus opposed a currency union. The four main elements of the current monetary order discussed in Laidler and Robson [2002], are an independent central bank with the power to create money, a floating exchange rate, a credible inflation target, and an institutional v

Executive Summary arrangement that holds policymakers accountable for their decisions. This regime is similar to the one that currently exists in Mexico. Several reasons were given for opposing dollarization or the creation of a monetary union. As in Mexico, differences in business cycles and economic structure gave rise to concerns in Canada that, without a flexible exchange rate, the Canadian economy would suffer real losses in the face of an asymmetric shock. Politically, monetary policy would no longer be accountable to the Canadian electorate. Dollarization imposes a further cost by diluting Canadian political identity, and the disappearance of the Canadian dollar would have consequences for the perceived relationship between the United States and Canada. Canadians in the business sector, government, and academia appear to agree that a common currency is neither the next step in US-Canada integration nor the most direct means of strengthening Canadian productivity growth or raising living standards. Rather, they set out an agenda for changing both internal Canadian policies and the framework for trade with the United States. The internal changes are aimed at faster productivity growth, increasing the competitiveness of the Canadian economy, and stimulating innovation. The list of priorities also includes improving the institutional apparatus of NAFTA, eliminating border risk, and removing remaining impediments to trade. Conclusion The United States policy of benign neglect may not be the most advantageous position for North America in the long run. Future changes in the international or domestic arenas may make greater monetary integration more palatable and beneficial for all three countries. For example, the euro could mount a challenge to the US dollar for pre-eminence as an international currency. Should this challenge materialize, it may be in the United States best interest to reconsider its policy on dollarization, particularly for Canada and Mexico. The widespread use of the US dollar vi

Executive Summary can generate power and prestige for the United States and help to maintain the US dollar s position as the main international currency. Mexico and Canada may also find dollarization more appealing at some point in the future. Renewed economic uncertainty, especially in Mexico, would no doubt bring the issue back to the political forefront. And in Canada, any further erosion, or perceived erosion, of productivity could reinvigorate discussions of dollarization. Currently, however, it is extremely clear is that all three countries have other and higher priorities. The issue of further economic integration and a potential currency union could rise eventually to the top of the political agenda. Before that occurs, however, there are many other ways to strengthen and deepen existing agreements and relationships in North America. vii

Executive Summary viii

I. Introduction Regional economic integration appeared to be on a steady upswing throughout the 1990s. Free trade agreements, customs unions, and even monetary unions have been adopted worldwide, from the ratification of the North American Free Trade Agreement (NAFTA) in 1994, to the launching of the euro in 1999, up through the free labor movement agreement for MERCOSUR in November 2002. In North America, NAFTA has been widely perceived as a success, and some observers have asked whether integration should be deepened. With the European model setting the precedent, an evolution from a free trade area to a customs union and on to a currency union seems to be viewed as the natural progression. Thus, the question arises, could North America ever become a currency union? The question became a pressing one briefly in Mexico following the 1994-1995 peso crisis. Some Mexican government officials and members of the private sector suggested dollarization the unilateral adoption of the US dollar as a solution to Mexico s currency ills. Argentina s success with its currency board also spurred discussion of alternative currency arrangements. Argentina was able to halt persistently high inflation with an arrangement similar to dollarization, a currency board, and it appeared to be providing a solid base for Argentine growth. In the latter part of the 1990s, the steadily weakening Canadian dollar also led Canadians to consider an alternative monetary arrangement. As is the case with Mexico, most of Canada s trade is with the United States, and adoption of the US dollar was therefore suggested. A debate arose between those who thought the Bank of Canada s floating arrangement to be the right choice and those who predicted substantial gains from dollarization. The European Monetary Union (EMU) looms large over all these discussions. Economic integration has produced a monetary union complete with an entirely new currency, the euro, and a new monetary authority, the European Central Bank. Though the discussions in Canada and 1

Introduction Mexico have referred most often to dollarization, the example of Europe necessarily raised the possibility of a North American Monetary Union (NAMU). Presumably, NAMU would mirror EMU; Canada, Mexico, and the United States would have a joint monetary policy and possibly even a new currency. This report is the result of an investigation into the prospects for a currency union in North America. As well as reading the analytic literature on currency unions, we visited Mexico and Canada to meet with leading government officials, economists, and businesspeople and solicit their views about a currency union a term that we use generically throughout this report to denote the whole set of arrangements involving a common currency, from a monetary union like EMU to unilateral dollarization. The focus of this report is on the Canadian and Mexican perspectives, in order to provide guidance to US policymakers if they should have to confront the issue of currency union in the future. In the following chapters, we analyze the information we gathered in Canada and Mexico in light of the analytic literature. Essentially, we answer the following questions: What does economic research tell us about the benefits and costs of a currency union? What form might it take in North America? What would be the implications for the United States? What is the present state of the debate in Mexico and Canada? What has been and should be the attitude of the United States? We offer initial conclusions on these topics at the end of the report, based both on the current views in each country and on our sense of the potential for future events to produce revisions and new views on the case for currency union. 2

II. Exchange Rate Regimes and the Theory of Optimum Currency Areas Countries throughout the world use a range of different exchange rate regimes. Each type of regime affects the conduct of monetary policy and the cost of trade. The range of regimes can be organized into a continuum from floating regimes, which provide a country with the greatest amount of policy flexibility and independence, to the firmest of fixed regimes wherein countries share a common currency and monetary policy. This chapter begins with a brief discussion of the most likely alternatives to North America s current exchange rate regimes. It then examines the different exchange rate arrangements, focusing on the costs and benefits of each arrangement and how each arrangement may or may not be applicable to North America. The chapter turns to a closer examination of currency unions and subsequently discusses Optimum Currency Area theory, which identifies those characteristics that make countries good candidates for participation in a currency union. Currently, Canada, the United States, and Mexico all have floating exchange rate regimes; each country conducts an independent monetary policy and has its own currency. The three countries currently participate in NAFTA allowing free trade across their borders. Further economic integration for the region would most likely take the form of a currency union. There are two types of currency unions, dollarization and monetary union. Under both arrangements the independence of monetary policy would be restricted. However, both would eliminate exchange rate risk and reduce transactions costs between trading partners. When there is uncertainty about exchange rates, financial transactions may be hedged and contracts may be written to ensure favorable exchange rates. But these activities add costs and may serve as deterrents to trade. A single currency would eliminate these costs. Furthermore, the mere exchange of currencies, for goods and asset trading, results in transactions costs. Recent research suggests that the elimination of currency risk and transactions costs can lead to large increases in trade, which would be a main objective of further integration of the three countries [Frankel & Rose 2002]. 3

Exchange Rate Regimes and the Theory of Optimum Currency Areas The United States is by far the largest trading partner for both Mexico and Canada, making a currency union the arrangement with the greatest benefits. What Type of Currency Union for North America? In the aftermath of the completion of the European Monetary Union, it is impossible to discuss currency alternatives in North America without referring to the European example. However, there exist fundamental differences between North America and Europe. In 1999, Sirkka Hämäläinen, member of the Executive Board of the European Central Bank, stated that: the idea of introducing a single currency was originally motivated more by political than economic arguments. It was seen as a natural and necessary step in the process of European integration a process aimed at ensuring that the risk of war and crisis on the European continent would be avoided through the establishment of common institutions as well as economic, social and cultural integration. There is no evidence of a similar political agenda in North America; for the United States, Canada, and Mexico, the question of a currency union is primarily a question of economic benefit. Additionally, the balance of economic and political power in North America is much different than that of Europe. Within the European Union, the largest country accounts for only 23 percent of the area s GDP, while in North America, the United States accounts for almost 90 percent of the area s GDP. Given these inequalities and the lack of political support for monetary integration, those interviewed in Mexico and Canada did not view a monetary union as a likely outcome in North America. Furthermore, the United States has never given any indication that it is willing to share its monetary policymaking with either of its neighbors. Therefore, if a currency union were to evolve in North America, its most likely form would be that of dollarization the adoption of the US dollar by Canada, Mexico, or both. Even if the United States would agree to participate in a monetary union similar to that of Europe, its size and clout would surely marginalize its neighbors needs. As this chapter explores the details of different 4

Exchange Rate Regimes and the Theory of Optimum Currency Areas exchange rate regimes, it becomes obvious that the trade-offs required to create a currency union make this union of unequal partners a complicated and arduous task to accomplish. Exchange Rate Regimes In a recent lecture, Stanley Fischer [2001] organizes the continuum of exchange rate regimes into three categories: the float, the intermediate peg, and the hard peg. Floating regimes are those in which exchange rates are determined by market forces with only limited or no central bank intervention. They include the free float and the managed float, under which the central bank may intervene but not to achieve an explicit exchange rate target. Intermediate regimes require central bank operations to keep the exchange rate within a specific range. Examples include a pegged rate, a crawling peg, an exchange rate band, and a crawling band. And hard pegs are currency regimes that preclude exchange rate changes and may involve sharing a single currency; this category includes currency boards, monetary unions, and dollarization. Floating Rates Most currencies are traded on international exchange markets. However, the value of a floating currency will increase or decrease depending only on the world demand for goods and assets denominated in that currency. A free float is one in which the central bank does not buy or sell the currency in order to maintain its value. A managed float involves some interventions that are not designed to maintain a specific exchange rate for the currency, but may be used to smooth fluctuations or in times of crisis [Fischer 2001]. In either floating regime, however, the central bank maintains an independent monetary policy and can adjust its interest rate as desired, without necessarily taking into consideration its trading partners needs. The central bank also earns seignorage on its reserves and can act as lender of last resort to domestic banks. One of the major benefits of a floating regime is its ability to help an economy absorb shocks. Canada, for example, is a commodity dependent economy. When there is a worldwide 5

Exchange Rate Regimes and the Theory of Optimum Currency Areas drop in the price of commodities, the value of the Canadian dollar tends to decrease, making its commodity exports cheaper and therefore more competitive. This increase in competitiveness protects the commodity-producing sectors from the world price drop that may otherwise have resulted in damaging real effects within Canada, such as increased unemployment and reduced output. Intermediate Pegs Intermediate pegs are regimes in which the central bank intervenes in order to maintain the nominal exchange rate at a previously announced level. These regimes exist along a segment of the continuum and run from the more restrictive to the less restrictive, from a fixed peg, to a crawling peg, to an exchange rate band, to a crawling band. A fixed peg is an exchange rate that is unchanging or is altered only rarely. A crawling peg may change over time in accordance with a rule or a plan. An exchange rate band prescribes the limits within which the central bank is committed to maintain the exchange rate, and a crawling band is one where the band moves over time [Fischer 2001]. The European countries used an exchange rate band in preparation for moving to a single currency. The Bretton Woods arrangement was a pegged rate regime where the US dollar was fixed to gold and other currencies throughout the world were pegged to the dollar. Prior to the Great Depression, many of the world s currencies were fixed to gold. In order to maintain a fixed exchange rate system, central banks must buy and sell reserves when the exchange rate moves away from the agreed upon value of the currency. For example, if demand decreases for a currency on the exchange markets, the exchange rate will tend to fall. The central bank must therefore intervene using its foreign reserves to maintain the value of its currency [Abel & Bernanke 2001]. These types of interventions are mandatory under all of the intermediate regimes. 6

Exchange Rate Regimes and the Theory of Optimum Currency Areas Even when the exchange rate floats, interventions are sometimes undertaken to ensure exchange rate stability. Calvo and Reinhart [2002] argue that some countries claiming to have floating rates actually display rigidity because they have a very low tolerance for large variations in the values of their currencies. They may evince "fear of floating" because large depreciations can greatly increase the cost of servicing foreign currency debt, and the adverse effects on the creditworthiness of private-sector debtors can lead to large output losses, while large appreciations may discourage investment by foreigners. More generally, large fluctuations may raise questions about an economy's fundamentals, impairing its access to international capital markets. Nevertheless, Fischer [2001] shows that many countries around the world are abandoning intermediate regimes for corner solutions. They are moving either to hard pegs or floating regimes. His data demonstrate that from 1991 to 1999, the number of countries with hard pegs went up from 25 to 45 (or 16 percent to 24 percent), the number of countries with intermediate pegs went down from 98 to 63 (or 62 percent to 34 percent), and the number of countries with floating rates went up from 36 to 77 (or 23 percent to 42 percent). Floating regimes tend to be larger economies. Overall, Fischer foresees a world with fewer currencies, resulting from decisions to dollarize or form monetary unions. The costs of intermediate regimes can be high and may encourage countries to seek out corner solutions. One cost is the loss of an independent monetary policy. A country that could benefit from lower interest rates may be constrained to maintain high rates by its commitment to a pegged exchange rate. The central bank s ability to conduct monetary policy or alter the interest rate is limited by its primary commitment to the fixity of its exchange rate. Another cost is the susceptibility of intermediate regimes to speculation. Speculators may sell large amounts of a currency they believe to be overvalued. This will force the central bank to spend its reserves to buy back its currency in order to limit the available supply of its currency and maintain its value [Abel & Bernanke 2001]. If its reserves prove to be too small, 7

Exchange Rate Regimes and the Theory of Optimum Currency Areas the central bank may be forced to devalue its currency. After a devaluation, foreign investors may question the credibility of a country s commitment to an intermediate peg, and capital inflows may be reduced. Hard Pegs: Currency Board A currency board fixes a country s exchange rate tightly to another country s money. Let s suppose Mexico adopts a currency board based on the US dollar. The Mexican central bank must then maintain reserves in US dollars, usually US bonds, equal to the amount of the its outstanding liabilities. This means that citizens of Mexico can request at any time that their bank deposits be paid out in US dollars at the fixed rate. This result is similar to the effect of dollarization, except that the Mexican peso still circulates and the central bank still functions. However, the central bank s functions are greatly diminished and, Mexico has to import US monetary policy. Currency boards are often established in response to a crisis, such as hyperinflation, brought on when central banks are vulnerable to political influence [Williamson 1995]. If the central bank is subject to political pressure, it may have to print money for fiscal expenditure or debt repayment. This can lead to hyperinflation. Argentina established a currency board to stop the hyperinflation that was engulfing its economy in 1991 [Mussa 2002]. Under a currency board, not only is the central bank unable to print money freely, but the interest rates and monetary policy of another, presumably more stable, economy, are imported along with the value of its currency. A currency board is not considered an option by the North American countries, because, currently, these countries central banks do not suffer from a lack of political independence or credibility. Hard Pegs: Dollarization & Monetary Union 8

Exchange Rate Regimes and the Theory of Optimum Currency Areas As described earlier, dollarization and monetary union eliminate exchange rate risk and reduce transactions costs between participating countries. These advantages may greatly increase trade between countries, and this would be a key reason for considering further economic integration in North America. Dollarization replaces a country s currency, so that its own currency is extinguished. When there is no national currency, there is no exchange rate. To undertake dollarization, a central bank must use its foreign reserves to buy enough dollars to replace its circulating currency. Once the economy is dollarized, there is no need for a central bank. The elimination of the central bank makes it difficult, if not impossible, to reverse dollarization. As a result, dollarization indicates a very high level of commitment to the new monetary system [Kenen 2000]. This greater level of commitment, in turn, leads to higher levels of confidence in the economy. The regime is therefore much less susceptible to the banking crises that currency boards and intermediate pegs may experience due to lack of perfect confidence in the fixed nominal exchange rate [de la Torre & Yeyati 2002]. Having used its reserves to redeem its currency, the country will no longer have those foreign assets on which to earn interest (seignorage). Nor is there the ability to provide lender of last resort financing through money creation. Commercial banks will hold their own cash reserves, most likely with commercial banks in the United States. A monetary union is an agreement between two or more sovereign countries to share a currency and monetary policy. A monetary union requires: the establishment of new institutions, such as a supranational central bank; the establishment of new procedures for policymaking taking account of all of the participating countries needs; and the acceptance by each country of the sovereignty of these new institutions. In the beginning of 1999, twelve European countries abandoned their individual currencies and formed a monetary union by adopting a single monetary unit, the euro. At the center of the union is the European Central Bank, with representatives from each country, which sets monetary policy for all of the countries in the 9

Exchange Rate Regimes and the Theory of Optimum Currency Areas union. Exchange rate risk is eliminated and trade is expected to increase within the union. Much like dollarization, monetary union is difficult to reverse. Both regimes make it difficult for individual member economies to respond to asymmetric shocks. If the member economies have different economic structures, experience different shocks, and then require different policies, a single monetary policy may make one country suffer real losses (such as job losses and wage reductions). These issues can severely test the durability of dollarization or a monetary union and require a high level of political commitment to the regime. In the North American context, the differences between these two regimes are key. Under dollarization, the countries taking on the US dollar, Mexico and Canada, would also import the US interest rate; whatever expansionary or contractionary monetary policy the United States pursued would be imported by Mexico and Canada. Furthermore, under dollarization, Mexico and Canada would be subject to the movements in the US dollar s exchange rate. Again, this could adversely affect these countries industries. If, the US dollar appreciated, Canadian commodities and Mexican manufactured goods would become less competitive internationally, while the United States would benefit from cheaper imports. In a monetary union, by contrast, there would be a new currency that would fluctuate in response to conditions in the entire region not just those in United States, and asymmetric shocks would be less severe. Additionally, there would presumably be a new supranational central bank that would take into consideration the needs of all the partners and set policy accordingly. However, given the relative sizes of the three economies, it is likely that events in the United States would also dominate this arrangement. Asymmetric shocks would not be greatly mitigated, and monetary policy would be largely driven by the needs of the United States. Optimum Currency Area Theory 10

Exchange Rate Regimes and the Theory of Optimum Currency Areas As discussed in the previous section, currency unions can achieve greater microeconomic efficiency through reductions of transaction costs and the elimination of exchange rate risk. Frankel and Rose [2002] estimate that a currency union triples the trade of a country with its partners. Furthermore, they find that the increase in trade directly contributes to income growth. Every one percent increase in total trade (relative to GDP) raises income per capita by at least one third of a percent over a twenty-year period, and possibly by much more over the long run. If trade were the only issue to consider, these findings would suggest that the whole world should adopt a single currency to reap the largest possible gains from trade. Giving up a national currency, however, means the loss of an independent monetary policy and of the benefits of exchange rate flexibility. This increases the vulnerability of an economy to internal imbalances (output and employment fluctuations) and external imbalances (current account fluctuations). Monetary policy and the exchange rate can no longer be used as macroeconomic tools to stabilize the economy in face of adverse economic shocks. Consider the United States and Canada, which have independent national currencies and flexible exchange rates. Suppose that the United States starts to import more goods from Canada. Canada will experience a current account surplus and, as its output increases because of increased exports, it will suffer some inflationary pressure. Conversely, the United States will experience recessionary pressure as its output decreases due to decreased exports. With a flexible exchange rate, the Canadian dollar will appreciate as a result of increased demand for its exports and the US dollar will depreciate. As a result, Canadian exports will become more expensive, demand for Canadian exports will decrease, and US exports will become cheaper and more competitive. The change in the exchange rate will thus result in increased output and employment in the United States while easing the inflationary pressure in Canada. The initial balance will therefore be restored. Optimum Currency Area (OCA) theory thus addresses these questions: with a fixed exchange rate or a common currency, what alternatives to an independent monetary policy and exchange rate flexibility can countries or regions use to respond to asymmetric shocks? What 11

Exchange Rate Regimes and the Theory of Optimum Currency Areas are the conditions under which countries can safely sacrifice exchange rate flexibility and independent monetary policies in order to reduce transactions costs and exchange rate risk by forming a monetary union? The foundations of the theory were largely laid down in the 1960s by the work of Mundell [1961], McKinnon [1963], Ingram [1969] and Kenen [1969]. Mundell emphasized the role of factor mobility in countering asymmetric shocks, stating, the optimum currency area is the region-defined in terms of internal factor mobility and external factor immobility. McKinnon highlighted the implications of the degree of openness of a country, arguing that a small, open economy cannot use its nominal exchange rate to cope with asymmetric shocks and therefore cannot by itself be an optimum currency area. Kenen made two additions to the analytic framework: the role of fiscal transfers within a monetary union, and the role of economic or sectoral diversification, in affording protection against asymmetric shocks. It is now generally agreed that, the greatest benefits of a currency union accrue to partner countries that possess one or more of the following characteristics. They should experience symmetric shocks, have domestic price and wage flexibility, and/or have high mobility of labor or capital across the national borders within the currency union. The partners should trade extensively with each other, have diversified domestic economies, and have a common fiscal transfer mechanism. McKinnon [2001] also emphasized another way of coping with asymmetric shocks international risk sharing and portfolio diversification. These characteristics are discussed below. 1. Symmetric shocks or similar economic structures and business cycles Countries that have similar economic structures and business cycles tend to face symmetric shocks (shocks that affect output and employment of two countries similarly). For countries that share symmetric shocks, there is little need for nominal exchange rate flexibility as a macroeconomic stabilizer. The nominal exchange rate only offsets asymmetric shocks (shocks 12

Exchange Rate Regimes and the Theory of Optimum Currency Areas that affect output and employment of one country in the opposite direction or to a different degree from another). It is nevertheless possible that asymmetric shocks, which are caused by structural differences between economies, will be moderated by the creation of a currency union. This is the so-called, endogeneity of the OCA criteria. Frankel and Rose [1998] and McKinnon [2001] have emphasized this possibility. They believe that trade integration and a currency union can help to align business cycles and hence reduce the possibility of asymmetric shocks. So the act of joining a currency union can further align the partner countries and help them fulfill the OCA conditions. 2. Flexible prices and wages The nominal exchange rate is only useful as a macroeconomic lever in a Keynesian framework where prices and wages are sticky; therefore the more flexible are prices and wages, the smaller the need for a country to depend on its nominal exchange rate, and it becomes a better candidate for membership in a currency union. When there is perfect price and wage flexibility, prices and wages adjust immediately to counteract any asymmetric shocks. Hence, there is no need for the nominal exchange rate to change. Suppose US customers start to buy more goods from Canada and fewer domestic goods. If prices and wages are flexible in the two countries, the increased demand for Canada s goods will raise Canadian prices and the decreased demand for US goods will reduce its prices. As a result, the demand for Canada s goods will decline, the demand for US goods will increase, and the initial equilibrium will be restored. 3. High degree of factor mobility including labor and capital mobility Without a flexible exchange rate or flexible prices and wages, factor mobility can act as a macroeconomic stabilizer in the face of asymmetric shocks. This is the most important implication of Mundell s original analysis. To see this point, let us assume there is a switch in 13

Exchange Rate Regimes and the Theory of Optimum Currency Areas demand for Canadian goods to US goods. As a result there will be an excess supply of goods in Canada and the unemployment rate will increase in Canada. Correspondingly, there will be an additional demand for labor in the United States, reflecting the excess demand for its goods. With labor mobility, Canadian citizens can move to the United States to work, decreasing unemployment in Canada while meeting the extra demand for labor in the United States 1. Capital mobility works in the same way. Investors move from the prosperous country, in this case the United States, to the troubled country, in this case Canada, and create more demand and more employment opportunities there. In principle, labor mobility and capital mobility are perfect substitutes. Either one can neutralize asymmetric shocks. However, the restoration of equilibrium through a capital movement is problematic in practice. Investors invest in places where they expect the return on capital to be highest. It is difficult to convince them to invest in a country affected by adverse shocks, with high unemployment and low growth prospects. This is why the literature has traditionally placed much more weight on labor mobility. 4. A high degree of openness to trade in goods and services McKinnon [1963] argued that a small open economy, which is a price taker in international markets, cannot affect its terms of trade through changes in its nominal exchange rate. A devaluation of its currency will simply raise its domestic price level, instead of making the country s exports less expensive and therefore more competitive. Furthermore, the devaluation generates pressure to raise nominal wages and thereby impairs the purchasing power of the domestic currency. So a small open economy cannot be an optimum currency area in itself. An optimum currency area must have a large non-tradable sector, which can serve to 1 Furthermore, the Canadian migrants demand for US goods (initially foreign demand for US exports) is internalized and becomes part of the domestic demand for US goods, whereas the migrants demand for Canadian goods (initially domestic demand in Canada) is externalized and becomes part of the foreign demand for Canadian exports. Both internal balance (maximum output and employment) and external balance (current account balance) are therefore restored. This point is emphasized by Kenen [2000]. 14

Exchange Rate Regimes and the Theory of Optimum Currency Areas stabilize the home currency s purchasing power. McKinnon also argued that prices and wages tend to be more flexible in a relatively open economy. Finally, Frankel and Rose [1998] made an empirical point: the more open economies are to each other, the more closely aligned their shocks tend to be, so that there is less need for them to rely on the nominal exchange rate as an adjustment tool. In sum, the more open the economy, the less likely it is to lose by surrendering exchange rate flexibility and the more it will benefit from a common currency with its main trading partners. 5. A diversified structure of production and demand Kenen [1969] pointed out that a highly diversified economy is less likely to experience large aggregate shocks and the need for large shifts in its real exchange rate. As some sectors experience negative shocks, others will experience positive shocks. The net effect of the positive and negative shocks should lessen the aggregate effects of the shocks. Furthermore, when there is a shock to a single sector, the necessary adjustment in the real exchange rate will be smaller than that in a less diversified economy. This is because the depreciation of the real exchange rate will not only make that sector more competitive but will also enhance the competitiveness of other sectors, providing a stimulus to all of its exports. 6. Existence of a fiscal transfer mechanisms within the currency area Even if a currency area is sub-optimal because of price rigidities, a lack of factor mobility, and little diversification, inter-regional equilibrium can be maintained if there is a mechanism for making net fiscal transfers from currency union members experiencing positive shocks to those experiencing adverse shocks. The transfers will reduce the inflationary pressure in the enriched regions and boost output and employment in the troubled regions. There are, of course, such mechanisms in federal fiscal systems like those of the United States and Canada, but it would be difficult to manage them so as to make fiscal transfers supportive of a monetary union. 15

Exchange Rate Regimes and the Theory of Optimum Currency Areas 7. International Risk Sharing and Portfolio Diversification Building on a further contribution of Mundell in 1973, McKinnon [2001] has emphasized the role of international portfolio diversification and risk sharing in counteracting the effects of asymmetric shocks. He argued that international risk sharing through cross holding of financial assets can mitigate the impact of asymmetric shocks on a country s income. However, exchange rate volatility arising from a flexible exchange rate discourages the cross-border investment required for such risk sharing. Even with a credibly fixed exchange rate, full portfolio diversification is not possible because exchange-rate fixity will not completely eliminate investors home bias. A common currency, McKinnon believes, will be the only corrective to sub-optimal portfolio diversification across national borders. McKinnon s view leads to a seemingly paradoxical conclusion. The less diversified an economy, the stronger is the argument for it to join a currency union. First, it can gain more from risk reduction through holding the assets of other economies having different structures and industries. Furthermore, the flexible exchange rate of a less diversified economy is likely to experience considerable volatility, which will in turn discourage cross holding of financial assets, as well as reduce the liquidity value of the domestic money. Since its citizens know that the value of the domestic currency tends to be volatile, they will prefer to hold and use the currency of a more diversified foreign country. Essentially the decision to join a currency union involves a trade off between reaping the microeconomic efficiency gains from a single currency and losing macroeconomic flexibility. Countries with one or more of the above traits is most likely to find microeconomic gains outweigh the costs of giving up on independent monetary policy and exchange rate flexibility. 16

III. The United States Perspective Monetary Union or Dollarization? Monetary union and dollarization have different implications for the United States and these implications play a large role in determining why dollarization is the most feasible option for the North American region. For the United States, the economic benefits derived from forming a monetary union are similar to those it would derive from other countries decisions to dollarize. Measured in costs to the United States, however, the difference between the two regimes is striking. A monetary union would require greater concessions by the United States because the structure of the Federal Reserve System would have to be changed, imposing the costs of institutional change and the associated political costs. For example, the Federal Reserve System might be obliged to increase the number of Federal Reserve governors to include governors representing Mexico and Canada and giving them full voting rights [Truman 2002]. Alternatively, a supranational central bank could be established that would operate like the European Central Bank in the European Monetary Union, eliminating the Federal Reserve altogether [Grubel 2000]. If the monetary union adopted a new monetary unit, the changing cost or switching cost to the United States would be even higher. The switching costs do not just involve the administrative and cost of redenominating all contracts, changing vending machines, but also the psychological costs of having to compute prices with a new numeraire [Buiter 1999]. In any North American currency union, as mentioned in Chapter II, the United States will dominate monetary policy-making because of its economic size and influence [Grubel 2000]. Nonetheless, a North American monetary union would require the United States to take account of the economies of partner countries when making monetary policy. Considering the importance that Americans traditionally attach to the autonomy of their monetary policy, the idea of 17

The United States Perspective including partner countries in the Federal Reserve or transferring the main role of the Federal Reserve to supranational institution would be difficult for many Americans to accept. In sum, given the size and economic influence of the US economy compared to the Canadian and Mexican economies, and given the international importance of the US dollar, a North American currency union that replaced the dollar with a new currency and replaced or changed the Federal Reserve would not be feasible. Therefore, this chapter will focus on US views about dollarization. The chapter reviews the official US position and the economic and political implications of dollarization for the United States. The Official US Perspective Cohen [2002] has analyzed the important implications of dollarization from the US perspective, reviewing a vast amount of literature. The United States, he says, has been passively neutral toward dollarization, adopting a policy of benign neglect. It appears to be averse to assuming any responsibility for dollarizing economies. Yet, it does not want to forgo the potential benefits of dollarization. The United States is not ready to assume any financial costs resulting from other countries loss of seignorage, take on the lender of last resort function, or agree to a change in its policy domain. But the United States does not actively discourage dollarization because it does not want to forego the microeconomic benefits. In short, benign neglect means that the United States would like to enjoy the microeconomic benefits from dollarization, but at minimal cost. At Congressional hearings in April 1999, Lawrence Summers, the US Secretary of Treasury, said, We do not have an a priori view as to our reaction to the concept of dollarization. There are a variety of means and modalities for achieving it and we would expect to discuss these with any government seriously considering taking such a momentous step. Then, he made it clear that there will be no explicit commitment of any kind. 18