- Email: [email protected]
Alternative exchange-rate regimes: The options for Latin America
North American Journal of Economics and Finance 13 (2002) 195–211
Alternative exchange-rate regimes: The options for Latin America Eduard Hochreiter1 , Pierre L. Siklos∗,2 Department of Economics, Wilfrid Laurier University, 75 University Avenue, Waterloo, Ont., Canada N2L 3C5 Received 27 July 2002; received in revised form 6 September 2002; accepted 9 September 2002
Abstract The debate on the optimal exchange-rate regime has been rekindled by the arrival of the euro and the rash of crises in Latin America and elsewhere. This paper reviews the key issues and assesses the state of play in the debate. This provides the context for the other papers in the group, prepared for a conference on “ Monetary union: Theory, EMU experience, and prospects for Latin America,” which took place in Vienna in April 2002 and which was co-sponsored by the Banco de Chile, the Oesterreichische Nationalbank, and the University of Vienna. The papers selected for this issue deal with exchange-rate problems in the Western Hemisphere, and Latin America and the Caribbean (LAC), in particular. After assessing the debate over floating rates versus currency union, the paper examines the issue in the context of monetary relations between the U.S. and Canada. It concludes with an empirical measure of convergence among the countries in the hemisphere. © 2002 Elsevier Science Inc. All rights reserved. JEL classification: E3; E42; F02 Keywords: Exchange-rate regime; Convergence; Monetary union
1. Introduction The debate over exchange-rate regimes continues to preoccupy economists and policy makers. The recent discussion has been energized by completion of European Monetary Union (EMU) and by a succession of exchange-rate crises in Europe, Asia, and Latin America. EMU has stimulated interest in currency unification, while the crises have cast doubts on the viability ∗
Corresponding author. Tel.: +1-519-884-0710x2559; fax: +1-519-888-1015. E-mail address: [email protected] (P.L. Siklos). 1 Oesterreichische Nationalbank. 2 Wilfrid Laurier University and Viessmann Research Centre on Modern Europe. 1062-9408/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S 1 0 6 2 - 9 4 0 8 ( 0 2 ) 0 0 0 9 9 - 2
196
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
of the so-called soft pegs. The collapse of the relatively rigid fixed-rate regime in Argentina has now spread those doubts to the hard pegs. Both developments contribute to the further hollowing-out of the middle of the spectrum of exchange-rate regimes and add to the popularity of the “bipolar view” of corner solutions, according to which the extremes of the hard peg and the freely floating exchange rate appear to be the only viable options.1 The problem, however, is far more complicated than simply choosing between flexible and rigidly fixed rates. To be fully viable, for instance, floating rates need to be accompanied by appropriate monetary policy rules, among which inflation targeting has been a recent favorite among both developed and developing countries. Among emerging economies, systemic weaknesses and lack of monetary policy credibility have given rise to the “fear-of-floating” syndrome, in which excessive exchange-market intervention destabilizes the arrangement.2 When it comes to fixed exchange rates, there is broad agreement that few, if any, Latin American countries currently satisfy the basic requirements for a workable currency area. However, failure to satisfy static optimum currency area (OCA) criteria is not sufficient to rule out monetary harmonization and cooperation. New insights from the theory of self-validating or endogenous currency areas suggest that forward-looking monetary cooperation may over time bring about the changes needed to satisfy OCA criteria. The trick is how to create the dynamics that will generate such an outcome? The answer involves complex issues pertaining to the nature and sequencing of policy coordination in trade, finance, and monetary policy. The European approach to currency union provides one example, involving a multi-decade process beginning with trade liberalization, followed by market integration, harmonization of standards, and creation of institutional and legal structures and an extended period of policy harmonization, before the final step of monetary unification. This may not be the right model for Latin America, where many of the conditions that existed when Europe embarked on its grand enterprise, do not obtain. In Latin America, the initial question is much more one of optimal sequencing, of deciding whether monetary cooperation or market integration might usefully be attempted first. The key concern is to identify the most promising catalyst, the cooperative effort that would be best at jump-starting the process of economic, financial and monetary integration. While the full implications of recent exchange-market crises are still being distilled by economists and policy makers, an obvious early lesson is that no regime will survive if the rules that make it work are systematically violated. It has long been understood that violation of the rules governing monetary policy will bring down a soft peg. Now, Argentina has demonstrated that violation of the rules governing fiscal policy will bring down a hard peg. Hard pegs were popular in part because it was believed that this was the most visible means of signaling adherence to a rule that would ensure macroeconomic stability. Argentina has shown otherwise, as Edwards details below. A second, and equally important, lesson, however, is that a system can be too rigid in the context of underlying structures, institutions and macroeconomic tendencies and in the nature of disturbances. Kopits stresses the importance of institutions and policies and Hochreiter, Schmidt-Hebbel and Winckler examine the implications of fiscal versus monetary policy dominance. Argentina may have ignored both lessons. It is clear that decisions were made at both public and private levels that ignored the principle embodied in the first lesson, but it is equally clear
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
197
that Argentina’s version of a currency board was, in the end, incompatible with the realities of the country’s past, its contemporary politics, and the nature of its markets and institutions. The rest of this introductory overview is organized as follows. Section 2 assesses the debate on the pre-conditions for an OCA. Section 3 examines the Maastricht Treaty as a model for Latin America and the Caribbean (LAC). Section 4 concludes.
2. Optimum currency areas: static and dynamic considerations The resurgence of interest in currency union is the result in part, at least, of completion of EMU. The apparent success of the European experiment has prompted countries around the globe to take stock of their own exchange-rate regimes and to ask whether currency union offers a superior prospect? When faced with this question, economists and policy makers typically compare the microeconomic benefits with the macroeconomic costs. The microeconomic benefits are well-known and include lower transactions costs, elimination of currency risk, and efficiency and scale economies from integration of markets. The principal costs include foregone seigniorage and loss of policy independence. Economic costs and benefits, however, are not always the main drivers on the road to trade and financial integration. In Europe, for example, political considerations played a key role and enabled the unification process to continue even when the economic justification was weak. A major criticism of floating rates is that their allegedly high volatility and sustained misalignments discourage trade, and hence economic growth, and that currency union offers a superior alternative from this perspective alone.3 Excessive volatility is seen by some as especially threatening to emerging economies with significant external debt, where it contributes to the “fear of floating” and encourages destabilizing foreign exchange market intervention.4 Belke and Gros study the effects of exchange-rate volatility on investment and employment in this issue. Recently, some Canadian economists have raised new concerns about the effect of floating rates. Courchene and Harris (2000) blame the sustained depreciation of the Canadian dollar, caused by deflation in world commodity prices, for having protected not only domestic resource-based sectors, but “old” manufacturing and having undermined innovation, renewal and the rise of new manufacturing. This argument is noteworthy for its use of insights from endogenous growth theory to reverse the standard view that causality runs from growth to exchange rates. For Courchene and Harris, this is a key rationale for monetary union.5 In the debate on asymmetries, floating rates are generally viewed as particularly suitable for countries with idiosyncratic economic structures. The literature suggests that at least four elements are critical to the viability of currency union in the presence of asymmetric structures and shocks. They are: labor mobility: the free movement of labor smoothes regional disparities in unemployment rates; capital mobility: savings and investment will seek out the most profitable opportunities while barriers to capital movement prevent this; openness and regional interdependence: it makes sense to use the same currency if goods move freely between the regions and if a significant portion of trade is done within a specific region;6 wage and price
198
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
flexibility: resources can only be allocated to their best uses if wages and prices are sufficiently flexible. Otherwise, the exchange rate must perform those functions. While there is disagreement on theoretical grounds and conflicting empirical evidence, there is broad agreement among observers that there does not at present exist a grouping of countries in Latin America which could be argued to satisfy the OCA criteria. The key elements of the problem are laid out in Kopits and in Berg, Borensztein, and Mauro in this issue. Failure to satisfy the static criteria of traditional OCA theory, however, does not foreclose the potential for currency unification. This consideration shifts the focus to questions concerning the optimal mix and sequencing of regional policies designed to fulfill the OCA criteria at some future date. This is the key concern in the emerging literature on self-validating or endogenous currency areas. 2.1. Varieties of currency unions For our purposes, the two main options under monetary unification are unilateral currency union, as in dollarization, and coordinated currency union, in which several countries either adopt a large member’s currency or create an entirely new money. Dollarization takes two forms. The first is market or informal dollarization, in which private parties elect to transact their business in U.S. dollars. The second is formal or policy dollarization, in which a country’s government officially adopts the dollar as a parallel currency or as the only currency. Under official dollarization the country gives up all seigniorage and monetary independence (see Berg & Borensztein, 2000). There is considerable disagreement on the costs and benefits of dollarization. Some argue that it is a sensible option only in the most extreme conditions such as persistent economic mismanagement (Buiter, 1999). Others are skeptical about the ability of dollarization to cure systemic economic ills (Eichengreen, 2001, 2002). The crux of the matter is that dollarization, as well as other fixed-rate systems, provide an inflation anchor, but they do not ensure resolution of deep-seated structural and institutional problems. Dollarization may be too rigid for countries with emerging markets and evolving economic structures. The collapse in Argentina of a relatively rigid exchange-rate regime suggests that no fixed-rate regime, including dollarization, is sustainable if the authorities choose to violate the rules that make it work or if the regime imposes constraints that are incompatible with existing institutions and practices. The attribute that was believed to make currency boards and dollarization superior to soft pegs was the promise that the rules of the game would be more difficult to violate. Argentina has demonstrated that the rules of the exchange-rate regime are meaningless unless there is a national ability and commitment to abide by them. The Argentinean experience proves that the costs of exit from a hard peg are indeed high ex post, but evidently not high enough ex ante to keep private behavior and public policy in check. As noted before, there are at present in Latin America no obvious combinations of countries, with or without the United States, suitable for monetary unification. Hence, the alternative model of monetary integration under consideration in some parts of the hemisphere is unilateral monetary union with the United States, that is, dollarization. It has its advocates in Canada, Mexico, and in other parts of the region. There is at this juncture, however, very little experience with this system, and hence very little relevant evidence to draw upon.
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
199
The alternative to dollarization is formal, coordinated currency integration, which is economically and politically far more complex, as the European experience makes clear. It is more complex in part, because it requires more encompassing institutional and political adjustments. It also raises questions about the timing and sequencing of monetary unification relative to economic integration more generally. Kopits provides a comparison between the Eastern European accession countries and Latin America. 2.2. The nature of aggregate shocks As noted earlier, the choice of exchange-rate regime must take into account the existence of asymmetries among countries. Significant differences in economic structure and shocks strengthen the case for floating rates, unless there is sufficient flexibility of wages and prices and mobility of resources to accommodate adjustment. An important advantage of flexible rates is their ability to react to disturbances and thereby to ease the burden of adjustment that needs to be absorbed by wages, prices, and employment. This is the well-known buffer function of floating rates. The floating rate also gives domestic monetary policy more room to respond to disturbances, not only in developed countries but, as Berg, Borensztein, and Mauro show, among Latin American floaters. Shock symmetry refers to the distribution of the impact of some economic shock across the participating economies in a potential monetary union. If shocks are highly asymmetric—read uncorrelated—then the required policy responses will differ across countries and a floating exchange rate serves as a shock absorber and provides room for differentiated policy reactions. If the exchange rate is fixed, domestic monetary policy cannot move out of line with policy in partner countries, and hence the burden of adjustment must be absorbed in other ways. In a model of VAR simulations of inflation and output growth, Bayoumi and Eichengreen (1994) attempt to measure asymmetry among contemporaneous shocks across a range of countries. They identify permanent and transitory shocks using the Blanchard–Quah decomposition method. For Canada and the U.S., they find that supply shocks are not highly correlated relative to the degree of symmetry of shocks in regions within the U.S. In a slightly more sophisticated study that takes into account both supply and non-monetary demand shocks, Lalonde and St-Amant (1995), and DeSerres and Lalonde (1994) conclude that shocks affecting the Canadian economy have little in common with those that affect the U.S. Finally, DeSerres and Lalonde (1994) find that shocks affecting Canada and the U.S. are subject to significant asymmetries, whereas structural shocks hitting the nine regions of the U.S. are very similar. Melitz and Weber (1996), and Dupasquier, Lalonde, and St-Amant (1997) point out that, when dynamics are taken into account, the U.S. and Canadian economies exhibit much greater symmetry. They find correlations between Canadian and U.S. shocks that are not very different overall from what we observe for European countries. This conclusion differs from the findings of Lalonde and St-Amant (1995), Bayoumi and Eichengreen (1994) or DeSerres and Lalonde (1994), who report more marked symmetries among European countries than between Canada and the U.S. Hochreiter, Korinek and Siklos (2002), using a structural VAR, find that interaction between fiscal and monetary policies reduces the correlation of shocks. They show that controlling for unsystematic policies, that is, ones that are generally thought to have been anticipated, can affect the size of correlations among structural shocks. Such
200
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
findings suggest that the practice of extracting “shocks” alone from models may omit more systematic components that also affect the interpretation of empirical evidence regarding the suitability of monetary regimes (e.g., see Hoover & Jordá, 2001). In other words, the fact that shocks to monetary policy appear to explain only a small portion of business cycle movements suggests that the anticipated portion of monetary policies may matter. The likelihood that an adverse shock would have a major impact on an economy depends on the commodity composition of production. Countries, like Canada and the Antipodes, where agriculture or resource sectors make up a significant share of GDP, may prefer floating rates as a buffer against movements in world commodity prices.7 A negative commodity price shock causes the exchange rate to depreciate, thereby easing downward pressures on domestic commodity prices. But the depreciation also allows prices to rise throughout the economy and so contributes to domestic inflation. Courchene and Harris (1999) and Harris (1999) worry that the flexible Canadian exchange rate divides the continent horizontally, whereas the trend has been for trade to expand along North–South lines, implying that the important symmetries are organized vertically rather than horizontally. Laidler (1999) and Murray (2000) disagree and argue instead that the structural differences between Canada and the United States are still more important than the similarities. Canada continues to be far more dependent on commodities than the United States and this asymmetry is in their view a dominant one. 2.3. The role of fiscal policy While monetary policy often receives much of the attention in the exchange-rate discussion, fiscal policy is no less important, for it is well-known that fiscal mismanagement has been the cause of many an exchange-rate crisis. Argentina is merely the latest example of the destabilizing consequences of fiscal indiscipline.8 Edwards, Kopits, and Hochreiter, Schmidt-Hebbel, and Winckler provide critical assessments below. Begg develops a model to formally incorporate fiscal policy rules. While the importance of fiscal discipline is not under dispute, it is difficult to find compelling evidence on the existence of a systematic relationship between fiscal policy and the exchange-rate regime, as Obstfeld and Rogoff (1995) show. Crow (1999) argues that floating currencies prevent fiscal irresponsibility on the grounds that it cannot be floated off by depreciation. Grubel (1999) and Courchene (1999) maintain that a common currency would have a positive impact on fiscal discipline through the external constraint on debt and deficit spending. As noted earlier, however, it is probably not realistic to expect any exchange-rate regime on its own to impose fiscal discipline. A key role falls upon domestic statutes and institutions. The relationship between fiscal policy and the exchange rate takes on additional complexities in the inter-temporal context, as Tornell and Velasco (1995) point out. In such a setting, lax fiscal policy today can generate fears of future devaluation, and thereby destabilize the system. In such circumstances, the advantage of floating rates is that they allow bad fiscal policy to be reflected more quickly in current exchange-rate movements, whereas pressures are allowed to build and accumulate under fixed rates until they overwhelm the system. In Europe, the convergence criteria of the Maastricht Treaty and the Stability and Growth Pact recognize these dangers and hence were expressly designed to ensure sound and
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
201
sustainable fiscal policies in member countries. The efficacy and sustainability of this system of rules is still to be tested, however. While the Stability Pact enjoys widespread support for the time being, it has received its share of criticism. One key issue under discussion concerns the role of fiscal policy as a shock absorber during cyclical swings. Some critics have asserted that the function has been largely emasculated, while others maintain that within the context of a medium-term goal of rough fiscal balance or surplus, there is enough flexibility to offset “normal” shocks. Eichengreen and Von Hagen (1996) raise a related issue by pointing out that too much fiscal discipline in a monetary union can impede coordination of stabilization policies. It may prevent members from cooperating in the defense of a partner in trouble, for fear of violating the pact and/or undermining system-wide credibility. Unlike the United States, the EU is not designed for union-wide tax transfers and so risk-sharing at the fiscal level is rather limited (Gramlich & Wood, 2001). Counter-cyclical fiscal transfers are thus confined to the level of nation states. While this may concern some, others, including Kletzer and Von Hagen (2002) and Hochreiter, Schmidt-Hebbel, and Winckler, in this issue, note that the net benefits of such transfers are not clear. The overall lesson from these considerations is that no exchange-rate regime will be sustainable without coherent monetary and fiscal policies. Each system comes with a set of rules and disciplines covering monetary and fiscal policies. The widely accepted argument that soft pegs are unworkable simply reflects recognition that their rules are too easily violated. Begg takes a closer look at the implications in this issue. 2.4. Business cycle synchronicity In addition to asymmetries in shocks, differences in the timing of business cycles undermine the viability of currency union. An important question, however, is whether cycle synchronicity is a necessary pre-condition for adoption of currency union or whether currency union contributes to synchronization? Frankel and Rose (1998) suggest that fixed exchange rates, including currency union, promote economic integration and thereby increase correlations among members’ business cycles.9 Krugman (1993) and Bayoumi and Prasad (1996) make the opposite case. To the extent that economic integration fosters specialization among members, as traditional trade theory would suggest, the member countries will become more diverse and asymmetric in production structure and hence more exposed to dissimilar industry-specific shocks. This will tend to make business cycles more idiosyncratic and asymmetric. These arguments, however, are relatively traditional and ignore recent innovations in trading patterns and cross-border production sharing. If the countries of a monetary union participate in cross-border production sharing, sector-specific shocks will cross borders as well. While the severity of the repercussions may vary with the importance of the affected sector in each member’s overall economic activity, the result will nevertheless be to increase cyclical symmetries.10 While the theoretical debate continues, the empirical evidence on cyclical convergence is rather mixed. Lafrance and St-Amant (1999) find that business cycles may become more similar if demand shocks dominate, countries are subject to common external shocks, or intra-industry
202
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
trade dominates. As a consequence, monetary union between Canada and the U.S. appears more costly from the point of view of shock asymmetry than from the perspective of business cycle asymmetry. A possible explanation is that the U.S. business cycle is quickly transmitted to Canada, due to the size of the U.S. economy and the tight economic relationship between the two countries. Hence, the two countries’ business cycles are more correlated than previously believed. Berg, Borenszin, and Mauro conclude that cycles between any pair of Latin American countries are no more correlated than cycles between emerging countries generally. Corsetti and Pesenti (2002), develop an alternative version of the endogenous OCA argument by showing that firms adapt pricing strategies in ways that allow currency areas which are sub-optimal ex ante to become optimal ex post. In particular, depending upon the exchange-rate regime in place, firms choose the optimal pass-through and this dictates whether shocks render business cycles symmetric across countries. For example, if there is zero pass-through, there are no costs in giving up an independent monetary policy and monetary union may then be a feasible option. At both ends of the spectrum, therefore, exchange-rate regimes can be self-validating.
3. A Maastricht approach for the Western Hemisphere? Does the European approach to monetary union provide a model for countries in the Western Hemisphere? Central to the Maastricht model are the concepts of monetary dominance (cf., Hochreiter, Schmidt-Hebbel and Winckler, in this issue) and of “convergence,” as defined in Protocol 6 of the Maastricht Treaty. Hochreiter et al. (2002) have examined annual date for a number of countries relevant to the question of convergence. Tables 1 and 2 compare convergence data for the Americas and the Antipodes with the Euro-zone. Canada, as well as New Zealand and Australia, satisfy the applicable Maastricht convergence criteria. The exchange-rate criterion is not applicable, since all three are classified as having freely floating exchange rates. As for the LAC countries, presented in Table 2, the evidence suggests dramatic improvements in inflation control, but limited success in bringing down nominal interest rates. The available debt and deficit data present a mixed picture, with some countries within or close to the limits, while others significantly exceed them. As a further measure of the likely costs of monetary union we apply here an approach developed by Alesina and Grilli (1992), which focuses on the volatility of output growth among countries. The approach employs a standard official loss function, with inflation variability and output variability as the arguments. As the difference in output variances rises between two countries and the correlation between output growths falls, the cost of monetary union rises. These two factors are combined to define “economic distance” in the following way: 1/2 σC 2 2 (1) + (1 + ρi ) σT where σ C is the standard deviation of output growth in the candidate country, σ T the standard deviation of output in the target country, and ρ i is the simple correlation coefficient for output
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
203
Table 1 Maastricht convergence: the Americas and the Antipodesa Country year
Inflation (CPI)
Fiscal 1 (deficit)
Australia 1998 1999 2000
0.9 1.5 4.5
0.6 1.0 −0.2
33.0 26.1 26.6
5.5 6.1 6.3
New Zealand 1998 1999 2000
1.3 0.1 2.6
1.4 0.3 0.5
38.6 37.1 34.7
6.3 6.4 6.9
Canada 1998 1999 2000
0.9 1.7 2.7
0.5/1.0 1.6/0.8 3.2/1.8
Euro-zone 1998 1999 2000
1.8 (0.7) 1.3 (0.5) 2.5 (1.6)
−2.2 −1.3 0.3
Fiscal 2 (debt)
116.2/64.9 111.6/61.0 104.9/51.8 76.9 74.8 72.4
Interest rate
4.89 6.18 5.35 4.8 (4.8) 4.7 (4.8) 5.4 (5.4)
Sources: Hochreiter et al., 2002, Table 2. Note: Inflation: OECD Economic Outlook, Annex Table 16. Average inflation rate of European Union and average inflation rate of the three best performing EU countries is in parenthesis. For New Zealand from http://www.rbnz.govt.nz and for Australia from http://www.rba.gov.au. 1998: Sweden (0.4%), France (0.8%), Austria (0.9%); average: 0.7%. 1999: Sweden (0.3%), France (0.5%), Austria (0.6%); average: 0.5%. 2000: Sweden (1.3%), France (1.7%), Germany (1.9%); average: 1.6%. Deficit: OECD Economic Outlook, Annex Table 30. For Canada from Bank of Canada Banking and Financial Statistics, Table A2. Debt: OECD Economic Outlook, Annex Table 34 with the exception of New Zealand: Source: IMF Staff Country Report No. 00/139, Table 12 (these terms refer to fiscal years). For Canada from http://www.fin.gc.ca. Interest Rate: OECD Economic Outlook, Annex Table 38. Average interest rate of European area and average interest rate of the three best-performing EU countries with regard to the inflation rate in parenthesis. Terms refer to 10-year government bond yields. For Canada, from Bank of Canada Banking and Financial Statistics Table A2. 1998: Sweden (5.0%), France (4.7%), Austria (4.7%); average: 4.8%. 1999: Sweden (5.0%), France (4.6%), Austria (4,7%); average: 4.8%. 2000: Sweden (5.4%), France (5.4%), Germany (5.3%); average: 5.4%. a Protocol 6 of the “Maastricht Treaty” contains the convergence criteria. (1) Inflation criterion: an inflation rate not more than (1 21 % higher than those of the three best-performing EU countries over the latest 12 months). (2) Fiscal convergence criteria: these criteria restrict the government budget deficit and the government debt to certain levels. A country which wants to participate in the EMU may not have • a government budget deficit higher than 3% of GDP, • a government debt ratio of more than 60% of GDP or sufficiently rapidly approaching that level. (3) Interest-rate criterion: an average nominal long-term interest rate that does not exceed by more than two percentage points that of the three best-performing member states in terms of price stability. (4) Exchange-rate criterion: participation in the Exchange-Rate Mechanism (ERM) of the European Monetary System (EMS) within the normal fluctuation margin without severe tensions for at least 2 years.
204
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
Table 2 Maastricht convergence in Latin America Country or country grouping Mercosurc Argentina
Year
Inflationa (CPI)
1998 1999 2000
6.62 (5.35)d 4.03 (3.55)d 4.96 (4.29)d
Fiscal 2 (debt)
Interest rateb
−2.1 −4.2 −3.6
41.3 47.4 50.6
19.38 (9.37)d 16.12 (8.00)d 12.82 (4.21)d
−4.4 (1992)e −0.3 (1993) −0.6 (1994)
NAf NA NA
Fiscal 1 (deficit)
Brazil
1992 1993 1994
Paraguay
1991 1992 1993
−6.40 −33.6 −2.6
Uraguay
1998 1999 2000
−0.8 −3.7 −3.4
23.6 (92) 21.9 (93) 24.5 (94)
Chile
1998 1999 2000
5.11 3.34 3.84
0.4 −1.5 0.1
12.7 13.9 13.9
9.12 7.44 8.73
Mexico
1998 1999 2000
15.93 16.59 9.50
−1.4 −1.5 −1.3
27.9 25.6 23.2
26.89 24.10 16.96
12.1 13.3 12.8
Sources: International Financial Statistics; country central banks; Government Financial Statistics (International Monetary Fund). See Table 1 for definitions. Data for Fiscal 2 for Argentina are from Mussa (2002), “Argentina and the Fund: From Triumph to Tragedy” (Washington, DC: Institute for International Economics), Policy Analyses in International Economics, 67, July. a Annual rate of change b Money market rate, except for Chile (discount rate) and Mexico (bankers’ acceptances). c Mercosur consists of Argentina, Brazil, Paraguay, and Uruguay. d The figures represent the mean CPI inflation and short-term interest rates with standard deviations (across countries) in parenthesis. e Most recent year for which there were comparable data. f Signifies that data comparable across countries were unavailable.
growth between the candidate and target economies. Rising economic distance increases the cost of monetary union. Figs. 1–3 plot the relationship between the two components of Eq. (1) for the various country groupings. Fig. 1 reveals that the cost of monetary union has declined substantially over the years for Australia and New Zealand. In Canada, output growth correlations remain very high throughout, but output volatility has increased. Hence, the cost of monetary union with the U.S. has, on balance, risen over time. Finally, the rise in output growth correlations between Austria and Germany and the decline in output growth volatility suggest the presence of self-validating tendencies. The picture is more mixed for the Netherlands, with a fall in volatility offset by a sharp decline in correlation.
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
205
206
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
207
208
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
Table 3 Economic distance, 1991–2001 Candidate (partner)
Output growth
Inflation
New Zealand (Australia) Canada (U.S.) Austria (Germany) The Netherlands (Germany) Argentina (Brazil) Paraguay (Brazil) Uruguay (Brazil) Chile (Brazil) Mexico (U.S.)
1.62 1.11 0.59 1.20 2.50 0.86 1.97 1.88 2.63
0.71 2.04 0.83 0.81 0.34 0.10 0.23 0.11 12.59
Note: Economic distance is as defined in Alesina and Grilli (1992) and Eq. (1). Annual data were used. See Figs. 1–3 for sources of data and Tables 1 and 2.
Fig. 2 assumes that the U.S. is the partner country. Output growth correlations rise for Mexico, reflecting the impact of NAFTA and the float, but output volatilities remain high relative to the countries depicted in Fig. 1. Overall, none of the countries appears to be a good fit for monetary union with the U.S. The situation is somewhat better when Brazil is the partner country, as shown in Fig. 3. For the most recent decade (1991–2000), Paraguay is the best fit, with Chile in second place. The cost of monetary union for the other countries appears to be rather high. The measures of economic distance shown in Table 3 confirm the foregoing results. Economic distance in terms of output growth suggests that the LAC countries are not generally good candidates for monetary union. The evidence is considerably more favorable for inflation convergence. It is, on the other hand, interesting to note that economic distance in terms of inflation between Canada and the U.S. is relatively high, a reflection perhaps of the manner in which a floating exchange-rate regime has de-coupled relative inflation performance in the two countries.
4. Conclusion The successful launch of a single currency in Europe has prompted policy makers elsewhere in the world to ask whether some form of monetary integration might deliver macroeconomic and microeconomic benefits relative to alternative currency regimes. There is wide agreement, as the papers in this issue explain, that the countries of the region do not at present satisfy the economic criteria for currency unification. Furthermore, evidence provided in this paper suggests that inspite of convergence in some respects—notably inflation, the economic distance separating countries in the region militates against monetary union. As Europe has repeatedly demonstrated, economic integration requires political will and commitment. The self-imposed policy constraints manifest in the Maastricht Treaty and the Stability Pact are a useful reminder. Such political will and the institutional restraints that reinforce it appear to be absent in the Western Hemisphere at present. This does not mean that
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
209
the status quo is the preferred solution, but rather that the long-run goal of monetary union must be facilitated by increasing policy coordination and harmonization. Notes 1. See Bayoumi and Eichengreen (1994) for an early articulation of the bipolar view. It is worth noting that not everyone is ready to abandon the intermediate exchange rate regimes (e.g., Frankel, 1999; Kuttner & Posen, 2001; Von Hagen & Zhou, 2002, and Willett, 2001). 2. See, for example, Calvo and Reinhart (2002). 3. John Williamson (2002) has articulated this concern. Mussa (1979) is the classic reference for empirical work that fails to find any deleterious effects from volatile exchange rates under a floating regime. See Arndt (2002) for an assessment. 4. See Calvo and Reinhart (2002). 5. For an assessment of this argument, see Arndt (2002). 6. Lane (2001) is a recent survey of the connection between openness, trade, and real exchange-rate behavior. 7. See Murray (2000) and Blundell-Wignall and Gregory (1990). 8. See Mussa (2002) for a detailed assessment. 9. See also Rose and Engel (2000). However, recent evidence by Ballabriga, Sebastian, and Valles (1999) shows that the common currency area in Europe has not led to more synchronized business cycles across Europe. 10. See Arndt (2002) for further discussion. Acknowledgments Portions of this paper were presented at the conference Monetary Union: Theory, EMU Experience, and Prospects for Latin America organized by the Oesterreichische Nationalbank, the University of Vienna and the Banco Central de Chile, April 15–16, 2002 in Vienna. Andreas Eckner and Sarah Facey provided excellent research and programming assistance. Comments on a previous draft by Sven Arndt, Heinz Herrmann, Tom Mayer, Weshah Razzak and Joob Swank are gratefully acknowledged. Part of the research for this paper was conducted while Siklos was WLU University Research Professor and Visiting Professor at UTS, Sydney. Financial support from the Social Sciences and Humanities Research Council of Canada and Wilfrid Laurier University is gratefully acknowledged. Previous versions of this paper were presented at the University of Vienna, the first annual Viessmann Conference at WLU, and the Conference on Exchange Rates, Economic Integration and the International Economy at Ryerson University. References Alesina, A., & Grilli, V. (1992). The European Central Bank: Reshaping monetary politics in Europe. In M. Canzoneri, V. Grilli & P. R. Masson (Eds.), Establishing a Central Bank: Issues in Europe and lessons from the U.S. (pp. 49–77). Cambridge: Cambridge University Press.
210
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
Arndt, S. W. (2002). The pros and cons of North American monetary integration. Mimeo. Ballabriga, F., Sebastian, M., & Valles, J. (1999). Interdependence of EC economies: A VAR approach. Mimeo. Bayoumi, T., & Eichengreen, B. (1994). One money or many? Analyzing the prospects for monetary unification in various parts of the world. Princeton, NJ: Princeton Studies in International Finance, No. 76. Bayoumi, I., & Prasad, E. (1996). Currency unions, economic fluctuations and adjustment: Some new empirical evidence. Working Paper No. 96/81. Washington, DC: IMF. Berg, A., & Borensztein, E. (2000). The pros and cons of full dollarization. Working Paper No. 00/50. Washington, DC: IMF. Blundell-Wignall, A., & Gregory, R. G. (1990). Exchange rate policy in advanced commodity-exporting countries: Australia and New Zealand. In V. Argy (Ed.), Choosing an exchange rate regime: The challenge for smaller industrial countries (pp. 224–271). Washington, DC: International Monetary Fund. Buiter, W. (1999). The EMU and the NAMU: What is the case for North American monetary union? Journal of Canadian Public Policy, 25, 3. Calvo, G. A., & Reinhart, C. M. (2002). Fear of floating. Quarterly Journal of Economics, 117(2), 379–408. Corsetti, G., & Pesenti, P. (2002). Self-validating optimum currency areas. Working Paper No. 8783. Cambridge, MA: NBER. Courchene, T. (1999). Alternative North American currency arrangements: A research agenda. Journal of Canadian Public Policy, 25, 3. Courchene, T., & Harris, R. (1999). From fixing to monetary union: Options for North American currency integration. CD Howe Institute Commentary, 127. Courchene, T. J., & Harris, R. G. (2000). North American monetary union: Analytical principles and operational guidelines. North American Journal of Economics and Finance, 1(August), 3–18. Crow, J. (1999). Canadian exchange rate policy. Canadian Public Policy, 25, 3. DeSerres, A., & Lalonde, R. (1994). Symetrie des chocs touchant les régions canadiennes et choix d’un réime de change. Working Paper No. 94–99. Bank of Canada. Dupasquier, C., Lalonde, R., & St-Amant, P. (1997). Optimum currency areas asapplied to Canada and the United States. In Exchange rates and monetary policy (pp. 131–170). Proceedings of a conference held by the Bank of Canada, October 1996. Eichengreen, B. (2001). What problems can dollarization solve? Journal of Policy Modeling, 23(April), 267–277. Eichengreen, B. (2002). When to dollarize. Journal of Money, Credit and Banking, 34(February), 1–24. Eichengreen, B., & Von Hagen, J. (1996). Fiscal policy and monetary union: Is there a tradeoff between federalism and budgetary restrictions? Working Paper No. 5517. Cambridge, MA: NBER. Frankel, J. A. (1999). No single currency regime is right for all countries or at all times. Working Paper No. 7338. Cambridge, MA: NBER. Frankel, J. A., & Rose, A. K. (1998, July). The endogeneity of the optimum currency area criteria. The Economic Journal, 108, 1009–1025. Gramlich, E. M., & Wood, P. R. (2001). Fiscal federalism and European integration: Implications for fiscal and monetary authorities. In M. Braimbridge & P. Whyman (Eds.), Fiscal federalism and European integration. London: Routledge. Forthcoming. Grubel, H. (1999). The case for the Amero. The Fraser Institute, Critical Issues Bulletin. Harris, R. (1999). The NAMU debate. Canadian Public Policy, 25, 3. Hochreiter, E., Korinek, A., & Siklos, P. L. (2002). The potential consequences of alternative exchange rate regimes: A study of three candidate regions. International Journal of Economics and Finance. Forthcoming. Hoover, K. D., & Jordá, O. (2001, July/August). Measuring systematic monetary policy Federal Reserve Bank of St. Louis. Review, 83, 113–138. Kletzer, K., & Von Hagen, J. (2002). Monetary union and fiscal federalism. ZEI. Mimeo. Krugman, P. (1993). Lessons from Massachusetts for EMU. In F. Torres & F. Giavazzi (Eds.), Adjustment and growth in the European monetary union (pp. 241–261). Cambridge: Cambridge University Press. Kuttner, K., & Posen, A. (2001). Beyond bipolar: A three-dimensional assessment of monetary frameworks. International Journal of Finance and Economics, 6(4), 369–387. Lafrance, R., & St-Amant, P. (1999). Optimal currency areas: A review of the recent literature. Working Paper No. 99-16. Bank of Canada.
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
211
Laidler, D. (1999). Canada’s exchange rate options. Canadian Public Policy, 25(September), 324–332. Lalonde, R., & St-Amant, P. (1995). Optimum currency areas: The case of Mexico and the United States. Centre for Latin American Studies. Monetary Affairs (July–December), 93–128. Lane, P. R. (2001, August). The new open economy macroeconomics: A survey. Journal of International Economics, 54, 2335–2366. Melitz, J., & Weber, A. (1996). The costs/benefits of a common monetary policy in France and Germany and possible lessons for monetary unions. Working Paper No. 1374. CEPR. Murray, J. (2000, August). Why Canada needs a flexible exchange rate. North American Journal of Economics and Finance, 11, 41–60. Mussa, M. (1979, Autumn). Empirical regularities in the behavior of exchange rates and theories of foreign exchange rate markets. Carnegie-Rochester Conference Series in Public Policy, 11, 9–57. Mussa, M. (2002, July). Argentina and the fund: From triumph to tragedy. Policy Analyses in International Economics, Vol. 67. Washington, DC: Institute for International Economics. Obstfeld, M., & Rogoff, K. (1995). The mirage of fixed exchange rates. Journal of Economic Perspectives, 9(4), 73–96. Rose, A. K., & Engel, C. (2000). Currency unions and international integration. Working Paper No. 7872. Cambridge, MA: NBER. Tornell, A., & Velasco, A. (1995). Fixed versus flexible exchange rates: Which provides more fiscal discipline? Working Paper No. 5108. Cambridge, MA: NBER. Von Hagen, J., & Zhou, J. (2002). The choice of exchange rate regimes: An empirical analysis for transition economies. Working Paper No. B 03 2002. ZEI. Willett, T. (2001). Fear of floating needn’t imply fixed rates: Feasible options for intermediate exchange rate regimes. Working Paper. Claremont Graduate University. Williamson, J. (Ed.). (2002, November). An economic policy agenda for Latin America. Washington, DC: Institute for International Economics. Forthcoming.
Alternative exchange-rate regimes: The options for Latin America Eduard Hochreiter1 , Pierre L. Siklos∗,2 Department of Economics, Wilfrid Laurier University, 75 University Avenue, Waterloo, Ont., Canada N2L 3C5 Received 27 July 2002; received in revised form 6 September 2002; accepted 9 September 2002
Abstract The debate on the optimal exchange-rate regime has been rekindled by the arrival of the euro and the rash of crises in Latin America and elsewhere. This paper reviews the key issues and assesses the state of play in the debate. This provides the context for the other papers in the group, prepared for a conference on “ Monetary union: Theory, EMU experience, and prospects for Latin America,” which took place in Vienna in April 2002 and which was co-sponsored by the Banco de Chile, the Oesterreichische Nationalbank, and the University of Vienna. The papers selected for this issue deal with exchange-rate problems in the Western Hemisphere, and Latin America and the Caribbean (LAC), in particular. After assessing the debate over floating rates versus currency union, the paper examines the issue in the context of monetary relations between the U.S. and Canada. It concludes with an empirical measure of convergence among the countries in the hemisphere. © 2002 Elsevier Science Inc. All rights reserved. JEL classification: E3; E42; F02 Keywords: Exchange-rate regime; Convergence; Monetary union
1. Introduction The debate over exchange-rate regimes continues to preoccupy economists and policy makers. The recent discussion has been energized by completion of European Monetary Union (EMU) and by a succession of exchange-rate crises in Europe, Asia, and Latin America. EMU has stimulated interest in currency unification, while the crises have cast doubts on the viability ∗
Corresponding author. Tel.: +1-519-884-0710x2559; fax: +1-519-888-1015. E-mail address: [email protected] (P.L. Siklos). 1 Oesterreichische Nationalbank. 2 Wilfrid Laurier University and Viessmann Research Centre on Modern Europe. 1062-9408/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S 1 0 6 2 - 9 4 0 8 ( 0 2 ) 0 0 0 9 9 - 2
196
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
of the so-called soft pegs. The collapse of the relatively rigid fixed-rate regime in Argentina has now spread those doubts to the hard pegs. Both developments contribute to the further hollowing-out of the middle of the spectrum of exchange-rate regimes and add to the popularity of the “bipolar view” of corner solutions, according to which the extremes of the hard peg and the freely floating exchange rate appear to be the only viable options.1 The problem, however, is far more complicated than simply choosing between flexible and rigidly fixed rates. To be fully viable, for instance, floating rates need to be accompanied by appropriate monetary policy rules, among which inflation targeting has been a recent favorite among both developed and developing countries. Among emerging economies, systemic weaknesses and lack of monetary policy credibility have given rise to the “fear-of-floating” syndrome, in which excessive exchange-market intervention destabilizes the arrangement.2 When it comes to fixed exchange rates, there is broad agreement that few, if any, Latin American countries currently satisfy the basic requirements for a workable currency area. However, failure to satisfy static optimum currency area (OCA) criteria is not sufficient to rule out monetary harmonization and cooperation. New insights from the theory of self-validating or endogenous currency areas suggest that forward-looking monetary cooperation may over time bring about the changes needed to satisfy OCA criteria. The trick is how to create the dynamics that will generate such an outcome? The answer involves complex issues pertaining to the nature and sequencing of policy coordination in trade, finance, and monetary policy. The European approach to currency union provides one example, involving a multi-decade process beginning with trade liberalization, followed by market integration, harmonization of standards, and creation of institutional and legal structures and an extended period of policy harmonization, before the final step of monetary unification. This may not be the right model for Latin America, where many of the conditions that existed when Europe embarked on its grand enterprise, do not obtain. In Latin America, the initial question is much more one of optimal sequencing, of deciding whether monetary cooperation or market integration might usefully be attempted first. The key concern is to identify the most promising catalyst, the cooperative effort that would be best at jump-starting the process of economic, financial and monetary integration. While the full implications of recent exchange-market crises are still being distilled by economists and policy makers, an obvious early lesson is that no regime will survive if the rules that make it work are systematically violated. It has long been understood that violation of the rules governing monetary policy will bring down a soft peg. Now, Argentina has demonstrated that violation of the rules governing fiscal policy will bring down a hard peg. Hard pegs were popular in part because it was believed that this was the most visible means of signaling adherence to a rule that would ensure macroeconomic stability. Argentina has shown otherwise, as Edwards details below. A second, and equally important, lesson, however, is that a system can be too rigid in the context of underlying structures, institutions and macroeconomic tendencies and in the nature of disturbances. Kopits stresses the importance of institutions and policies and Hochreiter, Schmidt-Hebbel and Winckler examine the implications of fiscal versus monetary policy dominance. Argentina may have ignored both lessons. It is clear that decisions were made at both public and private levels that ignored the principle embodied in the first lesson, but it is equally clear
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
197
that Argentina’s version of a currency board was, in the end, incompatible with the realities of the country’s past, its contemporary politics, and the nature of its markets and institutions. The rest of this introductory overview is organized as follows. Section 2 assesses the debate on the pre-conditions for an OCA. Section 3 examines the Maastricht Treaty as a model for Latin America and the Caribbean (LAC). Section 4 concludes.
2. Optimum currency areas: static and dynamic considerations The resurgence of interest in currency union is the result in part, at least, of completion of EMU. The apparent success of the European experiment has prompted countries around the globe to take stock of their own exchange-rate regimes and to ask whether currency union offers a superior prospect? When faced with this question, economists and policy makers typically compare the microeconomic benefits with the macroeconomic costs. The microeconomic benefits are well-known and include lower transactions costs, elimination of currency risk, and efficiency and scale economies from integration of markets. The principal costs include foregone seigniorage and loss of policy independence. Economic costs and benefits, however, are not always the main drivers on the road to trade and financial integration. In Europe, for example, political considerations played a key role and enabled the unification process to continue even when the economic justification was weak. A major criticism of floating rates is that their allegedly high volatility and sustained misalignments discourage trade, and hence economic growth, and that currency union offers a superior alternative from this perspective alone.3 Excessive volatility is seen by some as especially threatening to emerging economies with significant external debt, where it contributes to the “fear of floating” and encourages destabilizing foreign exchange market intervention.4 Belke and Gros study the effects of exchange-rate volatility on investment and employment in this issue. Recently, some Canadian economists have raised new concerns about the effect of floating rates. Courchene and Harris (2000) blame the sustained depreciation of the Canadian dollar, caused by deflation in world commodity prices, for having protected not only domestic resource-based sectors, but “old” manufacturing and having undermined innovation, renewal and the rise of new manufacturing. This argument is noteworthy for its use of insights from endogenous growth theory to reverse the standard view that causality runs from growth to exchange rates. For Courchene and Harris, this is a key rationale for monetary union.5 In the debate on asymmetries, floating rates are generally viewed as particularly suitable for countries with idiosyncratic economic structures. The literature suggests that at least four elements are critical to the viability of currency union in the presence of asymmetric structures and shocks. They are: labor mobility: the free movement of labor smoothes regional disparities in unemployment rates; capital mobility: savings and investment will seek out the most profitable opportunities while barriers to capital movement prevent this; openness and regional interdependence: it makes sense to use the same currency if goods move freely between the regions and if a significant portion of trade is done within a specific region;6 wage and price
198
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
flexibility: resources can only be allocated to their best uses if wages and prices are sufficiently flexible. Otherwise, the exchange rate must perform those functions. While there is disagreement on theoretical grounds and conflicting empirical evidence, there is broad agreement among observers that there does not at present exist a grouping of countries in Latin America which could be argued to satisfy the OCA criteria. The key elements of the problem are laid out in Kopits and in Berg, Borensztein, and Mauro in this issue. Failure to satisfy the static criteria of traditional OCA theory, however, does not foreclose the potential for currency unification. This consideration shifts the focus to questions concerning the optimal mix and sequencing of regional policies designed to fulfill the OCA criteria at some future date. This is the key concern in the emerging literature on self-validating or endogenous currency areas. 2.1. Varieties of currency unions For our purposes, the two main options under monetary unification are unilateral currency union, as in dollarization, and coordinated currency union, in which several countries either adopt a large member’s currency or create an entirely new money. Dollarization takes two forms. The first is market or informal dollarization, in which private parties elect to transact their business in U.S. dollars. The second is formal or policy dollarization, in which a country’s government officially adopts the dollar as a parallel currency or as the only currency. Under official dollarization the country gives up all seigniorage and monetary independence (see Berg & Borensztein, 2000). There is considerable disagreement on the costs and benefits of dollarization. Some argue that it is a sensible option only in the most extreme conditions such as persistent economic mismanagement (Buiter, 1999). Others are skeptical about the ability of dollarization to cure systemic economic ills (Eichengreen, 2001, 2002). The crux of the matter is that dollarization, as well as other fixed-rate systems, provide an inflation anchor, but they do not ensure resolution of deep-seated structural and institutional problems. Dollarization may be too rigid for countries with emerging markets and evolving economic structures. The collapse in Argentina of a relatively rigid exchange-rate regime suggests that no fixed-rate regime, including dollarization, is sustainable if the authorities choose to violate the rules that make it work or if the regime imposes constraints that are incompatible with existing institutions and practices. The attribute that was believed to make currency boards and dollarization superior to soft pegs was the promise that the rules of the game would be more difficult to violate. Argentina has demonstrated that the rules of the exchange-rate regime are meaningless unless there is a national ability and commitment to abide by them. The Argentinean experience proves that the costs of exit from a hard peg are indeed high ex post, but evidently not high enough ex ante to keep private behavior and public policy in check. As noted before, there are at present in Latin America no obvious combinations of countries, with or without the United States, suitable for monetary unification. Hence, the alternative model of monetary integration under consideration in some parts of the hemisphere is unilateral monetary union with the United States, that is, dollarization. It has its advocates in Canada, Mexico, and in other parts of the region. There is at this juncture, however, very little experience with this system, and hence very little relevant evidence to draw upon.
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
199
The alternative to dollarization is formal, coordinated currency integration, which is economically and politically far more complex, as the European experience makes clear. It is more complex in part, because it requires more encompassing institutional and political adjustments. It also raises questions about the timing and sequencing of monetary unification relative to economic integration more generally. Kopits provides a comparison between the Eastern European accession countries and Latin America. 2.2. The nature of aggregate shocks As noted earlier, the choice of exchange-rate regime must take into account the existence of asymmetries among countries. Significant differences in economic structure and shocks strengthen the case for floating rates, unless there is sufficient flexibility of wages and prices and mobility of resources to accommodate adjustment. An important advantage of flexible rates is their ability to react to disturbances and thereby to ease the burden of adjustment that needs to be absorbed by wages, prices, and employment. This is the well-known buffer function of floating rates. The floating rate also gives domestic monetary policy more room to respond to disturbances, not only in developed countries but, as Berg, Borensztein, and Mauro show, among Latin American floaters. Shock symmetry refers to the distribution of the impact of some economic shock across the participating economies in a potential monetary union. If shocks are highly asymmetric—read uncorrelated—then the required policy responses will differ across countries and a floating exchange rate serves as a shock absorber and provides room for differentiated policy reactions. If the exchange rate is fixed, domestic monetary policy cannot move out of line with policy in partner countries, and hence the burden of adjustment must be absorbed in other ways. In a model of VAR simulations of inflation and output growth, Bayoumi and Eichengreen (1994) attempt to measure asymmetry among contemporaneous shocks across a range of countries. They identify permanent and transitory shocks using the Blanchard–Quah decomposition method. For Canada and the U.S., they find that supply shocks are not highly correlated relative to the degree of symmetry of shocks in regions within the U.S. In a slightly more sophisticated study that takes into account both supply and non-monetary demand shocks, Lalonde and St-Amant (1995), and DeSerres and Lalonde (1994) conclude that shocks affecting the Canadian economy have little in common with those that affect the U.S. Finally, DeSerres and Lalonde (1994) find that shocks affecting Canada and the U.S. are subject to significant asymmetries, whereas structural shocks hitting the nine regions of the U.S. are very similar. Melitz and Weber (1996), and Dupasquier, Lalonde, and St-Amant (1997) point out that, when dynamics are taken into account, the U.S. and Canadian economies exhibit much greater symmetry. They find correlations between Canadian and U.S. shocks that are not very different overall from what we observe for European countries. This conclusion differs from the findings of Lalonde and St-Amant (1995), Bayoumi and Eichengreen (1994) or DeSerres and Lalonde (1994), who report more marked symmetries among European countries than between Canada and the U.S. Hochreiter, Korinek and Siklos (2002), using a structural VAR, find that interaction between fiscal and monetary policies reduces the correlation of shocks. They show that controlling for unsystematic policies, that is, ones that are generally thought to have been anticipated, can affect the size of correlations among structural shocks. Such
200
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
findings suggest that the practice of extracting “shocks” alone from models may omit more systematic components that also affect the interpretation of empirical evidence regarding the suitability of monetary regimes (e.g., see Hoover & Jordá, 2001). In other words, the fact that shocks to monetary policy appear to explain only a small portion of business cycle movements suggests that the anticipated portion of monetary policies may matter. The likelihood that an adverse shock would have a major impact on an economy depends on the commodity composition of production. Countries, like Canada and the Antipodes, where agriculture or resource sectors make up a significant share of GDP, may prefer floating rates as a buffer against movements in world commodity prices.7 A negative commodity price shock causes the exchange rate to depreciate, thereby easing downward pressures on domestic commodity prices. But the depreciation also allows prices to rise throughout the economy and so contributes to domestic inflation. Courchene and Harris (1999) and Harris (1999) worry that the flexible Canadian exchange rate divides the continent horizontally, whereas the trend has been for trade to expand along North–South lines, implying that the important symmetries are organized vertically rather than horizontally. Laidler (1999) and Murray (2000) disagree and argue instead that the structural differences between Canada and the United States are still more important than the similarities. Canada continues to be far more dependent on commodities than the United States and this asymmetry is in their view a dominant one. 2.3. The role of fiscal policy While monetary policy often receives much of the attention in the exchange-rate discussion, fiscal policy is no less important, for it is well-known that fiscal mismanagement has been the cause of many an exchange-rate crisis. Argentina is merely the latest example of the destabilizing consequences of fiscal indiscipline.8 Edwards, Kopits, and Hochreiter, Schmidt-Hebbel, and Winckler provide critical assessments below. Begg develops a model to formally incorporate fiscal policy rules. While the importance of fiscal discipline is not under dispute, it is difficult to find compelling evidence on the existence of a systematic relationship between fiscal policy and the exchange-rate regime, as Obstfeld and Rogoff (1995) show. Crow (1999) argues that floating currencies prevent fiscal irresponsibility on the grounds that it cannot be floated off by depreciation. Grubel (1999) and Courchene (1999) maintain that a common currency would have a positive impact on fiscal discipline through the external constraint on debt and deficit spending. As noted earlier, however, it is probably not realistic to expect any exchange-rate regime on its own to impose fiscal discipline. A key role falls upon domestic statutes and institutions. The relationship between fiscal policy and the exchange rate takes on additional complexities in the inter-temporal context, as Tornell and Velasco (1995) point out. In such a setting, lax fiscal policy today can generate fears of future devaluation, and thereby destabilize the system. In such circumstances, the advantage of floating rates is that they allow bad fiscal policy to be reflected more quickly in current exchange-rate movements, whereas pressures are allowed to build and accumulate under fixed rates until they overwhelm the system. In Europe, the convergence criteria of the Maastricht Treaty and the Stability and Growth Pact recognize these dangers and hence were expressly designed to ensure sound and
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
201
sustainable fiscal policies in member countries. The efficacy and sustainability of this system of rules is still to be tested, however. While the Stability Pact enjoys widespread support for the time being, it has received its share of criticism. One key issue under discussion concerns the role of fiscal policy as a shock absorber during cyclical swings. Some critics have asserted that the function has been largely emasculated, while others maintain that within the context of a medium-term goal of rough fiscal balance or surplus, there is enough flexibility to offset “normal” shocks. Eichengreen and Von Hagen (1996) raise a related issue by pointing out that too much fiscal discipline in a monetary union can impede coordination of stabilization policies. It may prevent members from cooperating in the defense of a partner in trouble, for fear of violating the pact and/or undermining system-wide credibility. Unlike the United States, the EU is not designed for union-wide tax transfers and so risk-sharing at the fiscal level is rather limited (Gramlich & Wood, 2001). Counter-cyclical fiscal transfers are thus confined to the level of nation states. While this may concern some, others, including Kletzer and Von Hagen (2002) and Hochreiter, Schmidt-Hebbel, and Winckler, in this issue, note that the net benefits of such transfers are not clear. The overall lesson from these considerations is that no exchange-rate regime will be sustainable without coherent monetary and fiscal policies. Each system comes with a set of rules and disciplines covering monetary and fiscal policies. The widely accepted argument that soft pegs are unworkable simply reflects recognition that their rules are too easily violated. Begg takes a closer look at the implications in this issue. 2.4. Business cycle synchronicity In addition to asymmetries in shocks, differences in the timing of business cycles undermine the viability of currency union. An important question, however, is whether cycle synchronicity is a necessary pre-condition for adoption of currency union or whether currency union contributes to synchronization? Frankel and Rose (1998) suggest that fixed exchange rates, including currency union, promote economic integration and thereby increase correlations among members’ business cycles.9 Krugman (1993) and Bayoumi and Prasad (1996) make the opposite case. To the extent that economic integration fosters specialization among members, as traditional trade theory would suggest, the member countries will become more diverse and asymmetric in production structure and hence more exposed to dissimilar industry-specific shocks. This will tend to make business cycles more idiosyncratic and asymmetric. These arguments, however, are relatively traditional and ignore recent innovations in trading patterns and cross-border production sharing. If the countries of a monetary union participate in cross-border production sharing, sector-specific shocks will cross borders as well. While the severity of the repercussions may vary with the importance of the affected sector in each member’s overall economic activity, the result will nevertheless be to increase cyclical symmetries.10 While the theoretical debate continues, the empirical evidence on cyclical convergence is rather mixed. Lafrance and St-Amant (1999) find that business cycles may become more similar if demand shocks dominate, countries are subject to common external shocks, or intra-industry
202
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
trade dominates. As a consequence, monetary union between Canada and the U.S. appears more costly from the point of view of shock asymmetry than from the perspective of business cycle asymmetry. A possible explanation is that the U.S. business cycle is quickly transmitted to Canada, due to the size of the U.S. economy and the tight economic relationship between the two countries. Hence, the two countries’ business cycles are more correlated than previously believed. Berg, Borenszin, and Mauro conclude that cycles between any pair of Latin American countries are no more correlated than cycles between emerging countries generally. Corsetti and Pesenti (2002), develop an alternative version of the endogenous OCA argument by showing that firms adapt pricing strategies in ways that allow currency areas which are sub-optimal ex ante to become optimal ex post. In particular, depending upon the exchange-rate regime in place, firms choose the optimal pass-through and this dictates whether shocks render business cycles symmetric across countries. For example, if there is zero pass-through, there are no costs in giving up an independent monetary policy and monetary union may then be a feasible option. At both ends of the spectrum, therefore, exchange-rate regimes can be self-validating.
3. A Maastricht approach for the Western Hemisphere? Does the European approach to monetary union provide a model for countries in the Western Hemisphere? Central to the Maastricht model are the concepts of monetary dominance (cf., Hochreiter, Schmidt-Hebbel and Winckler, in this issue) and of “convergence,” as defined in Protocol 6 of the Maastricht Treaty. Hochreiter et al. (2002) have examined annual date for a number of countries relevant to the question of convergence. Tables 1 and 2 compare convergence data for the Americas and the Antipodes with the Euro-zone. Canada, as well as New Zealand and Australia, satisfy the applicable Maastricht convergence criteria. The exchange-rate criterion is not applicable, since all three are classified as having freely floating exchange rates. As for the LAC countries, presented in Table 2, the evidence suggests dramatic improvements in inflation control, but limited success in bringing down nominal interest rates. The available debt and deficit data present a mixed picture, with some countries within or close to the limits, while others significantly exceed them. As a further measure of the likely costs of monetary union we apply here an approach developed by Alesina and Grilli (1992), which focuses on the volatility of output growth among countries. The approach employs a standard official loss function, with inflation variability and output variability as the arguments. As the difference in output variances rises between two countries and the correlation between output growths falls, the cost of monetary union rises. These two factors are combined to define “economic distance” in the following way: 1/2 σC 2 2 (1) + (1 + ρi ) σT where σ C is the standard deviation of output growth in the candidate country, σ T the standard deviation of output in the target country, and ρ i is the simple correlation coefficient for output
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
203
Table 1 Maastricht convergence: the Americas and the Antipodesa Country year
Inflation (CPI)
Fiscal 1 (deficit)
Australia 1998 1999 2000
0.9 1.5 4.5
0.6 1.0 −0.2
33.0 26.1 26.6
5.5 6.1 6.3
New Zealand 1998 1999 2000
1.3 0.1 2.6
1.4 0.3 0.5
38.6 37.1 34.7
6.3 6.4 6.9
Canada 1998 1999 2000
0.9 1.7 2.7
0.5/1.0 1.6/0.8 3.2/1.8
Euro-zone 1998 1999 2000
1.8 (0.7) 1.3 (0.5) 2.5 (1.6)
−2.2 −1.3 0.3
Fiscal 2 (debt)
116.2/64.9 111.6/61.0 104.9/51.8 76.9 74.8 72.4
Interest rate
4.89 6.18 5.35 4.8 (4.8) 4.7 (4.8) 5.4 (5.4)
Sources: Hochreiter et al., 2002, Table 2. Note: Inflation: OECD Economic Outlook, Annex Table 16. Average inflation rate of European Union and average inflation rate of the three best performing EU countries is in parenthesis. For New Zealand from http://www.rbnz.govt.nz and for Australia from http://www.rba.gov.au. 1998: Sweden (0.4%), France (0.8%), Austria (0.9%); average: 0.7%. 1999: Sweden (0.3%), France (0.5%), Austria (0.6%); average: 0.5%. 2000: Sweden (1.3%), France (1.7%), Germany (1.9%); average: 1.6%. Deficit: OECD Economic Outlook, Annex Table 30. For Canada from Bank of Canada Banking and Financial Statistics, Table A2. Debt: OECD Economic Outlook, Annex Table 34 with the exception of New Zealand: Source: IMF Staff Country Report No. 00/139, Table 12 (these terms refer to fiscal years). For Canada from http://www.fin.gc.ca. Interest Rate: OECD Economic Outlook, Annex Table 38. Average interest rate of European area and average interest rate of the three best-performing EU countries with regard to the inflation rate in parenthesis. Terms refer to 10-year government bond yields. For Canada, from Bank of Canada Banking and Financial Statistics Table A2. 1998: Sweden (5.0%), France (4.7%), Austria (4.7%); average: 4.8%. 1999: Sweden (5.0%), France (4.6%), Austria (4,7%); average: 4.8%. 2000: Sweden (5.4%), France (5.4%), Germany (5.3%); average: 5.4%. a Protocol 6 of the “Maastricht Treaty” contains the convergence criteria. (1) Inflation criterion: an inflation rate not more than (1 21 % higher than those of the three best-performing EU countries over the latest 12 months). (2) Fiscal convergence criteria: these criteria restrict the government budget deficit and the government debt to certain levels. A country which wants to participate in the EMU may not have • a government budget deficit higher than 3% of GDP, • a government debt ratio of more than 60% of GDP or sufficiently rapidly approaching that level. (3) Interest-rate criterion: an average nominal long-term interest rate that does not exceed by more than two percentage points that of the three best-performing member states in terms of price stability. (4) Exchange-rate criterion: participation in the Exchange-Rate Mechanism (ERM) of the European Monetary System (EMS) within the normal fluctuation margin without severe tensions for at least 2 years.
204
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
Table 2 Maastricht convergence in Latin America Country or country grouping Mercosurc Argentina
Year
Inflationa (CPI)
1998 1999 2000
6.62 (5.35)d 4.03 (3.55)d 4.96 (4.29)d
Fiscal 2 (debt)
Interest rateb
−2.1 −4.2 −3.6
41.3 47.4 50.6
19.38 (9.37)d 16.12 (8.00)d 12.82 (4.21)d
−4.4 (1992)e −0.3 (1993) −0.6 (1994)
NAf NA NA
Fiscal 1 (deficit)
Brazil
1992 1993 1994
Paraguay
1991 1992 1993
−6.40 −33.6 −2.6
Uraguay
1998 1999 2000
−0.8 −3.7 −3.4
23.6 (92) 21.9 (93) 24.5 (94)
Chile
1998 1999 2000
5.11 3.34 3.84
0.4 −1.5 0.1
12.7 13.9 13.9
9.12 7.44 8.73
Mexico
1998 1999 2000
15.93 16.59 9.50
−1.4 −1.5 −1.3
27.9 25.6 23.2
26.89 24.10 16.96
12.1 13.3 12.8
Sources: International Financial Statistics; country central banks; Government Financial Statistics (International Monetary Fund). See Table 1 for definitions. Data for Fiscal 2 for Argentina are from Mussa (2002), “Argentina and the Fund: From Triumph to Tragedy” (Washington, DC: Institute for International Economics), Policy Analyses in International Economics, 67, July. a Annual rate of change b Money market rate, except for Chile (discount rate) and Mexico (bankers’ acceptances). c Mercosur consists of Argentina, Brazil, Paraguay, and Uruguay. d The figures represent the mean CPI inflation and short-term interest rates with standard deviations (across countries) in parenthesis. e Most recent year for which there were comparable data. f Signifies that data comparable across countries were unavailable.
growth between the candidate and target economies. Rising economic distance increases the cost of monetary union. Figs. 1–3 plot the relationship between the two components of Eq. (1) for the various country groupings. Fig. 1 reveals that the cost of monetary union has declined substantially over the years for Australia and New Zealand. In Canada, output growth correlations remain very high throughout, but output volatility has increased. Hence, the cost of monetary union with the U.S. has, on balance, risen over time. Finally, the rise in output growth correlations between Austria and Germany and the decline in output growth volatility suggest the presence of self-validating tendencies. The picture is more mixed for the Netherlands, with a fall in volatility offset by a sharp decline in correlation.
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
205
206
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
207
208
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
Table 3 Economic distance, 1991–2001 Candidate (partner)
Output growth
Inflation
New Zealand (Australia) Canada (U.S.) Austria (Germany) The Netherlands (Germany) Argentina (Brazil) Paraguay (Brazil) Uruguay (Brazil) Chile (Brazil) Mexico (U.S.)
1.62 1.11 0.59 1.20 2.50 0.86 1.97 1.88 2.63
0.71 2.04 0.83 0.81 0.34 0.10 0.23 0.11 12.59
Note: Economic distance is as defined in Alesina and Grilli (1992) and Eq. (1). Annual data were used. See Figs. 1–3 for sources of data and Tables 1 and 2.
Fig. 2 assumes that the U.S. is the partner country. Output growth correlations rise for Mexico, reflecting the impact of NAFTA and the float, but output volatilities remain high relative to the countries depicted in Fig. 1. Overall, none of the countries appears to be a good fit for monetary union with the U.S. The situation is somewhat better when Brazil is the partner country, as shown in Fig. 3. For the most recent decade (1991–2000), Paraguay is the best fit, with Chile in second place. The cost of monetary union for the other countries appears to be rather high. The measures of economic distance shown in Table 3 confirm the foregoing results. Economic distance in terms of output growth suggests that the LAC countries are not generally good candidates for monetary union. The evidence is considerably more favorable for inflation convergence. It is, on the other hand, interesting to note that economic distance in terms of inflation between Canada and the U.S. is relatively high, a reflection perhaps of the manner in which a floating exchange-rate regime has de-coupled relative inflation performance in the two countries.
4. Conclusion The successful launch of a single currency in Europe has prompted policy makers elsewhere in the world to ask whether some form of monetary integration might deliver macroeconomic and microeconomic benefits relative to alternative currency regimes. There is wide agreement, as the papers in this issue explain, that the countries of the region do not at present satisfy the economic criteria for currency unification. Furthermore, evidence provided in this paper suggests that inspite of convergence in some respects—notably inflation, the economic distance separating countries in the region militates against monetary union. As Europe has repeatedly demonstrated, economic integration requires political will and commitment. The self-imposed policy constraints manifest in the Maastricht Treaty and the Stability Pact are a useful reminder. Such political will and the institutional restraints that reinforce it appear to be absent in the Western Hemisphere at present. This does not mean that
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
209
the status quo is the preferred solution, but rather that the long-run goal of monetary union must be facilitated by increasing policy coordination and harmonization. Notes 1. See Bayoumi and Eichengreen (1994) for an early articulation of the bipolar view. It is worth noting that not everyone is ready to abandon the intermediate exchange rate regimes (e.g., Frankel, 1999; Kuttner & Posen, 2001; Von Hagen & Zhou, 2002, and Willett, 2001). 2. See, for example, Calvo and Reinhart (2002). 3. John Williamson (2002) has articulated this concern. Mussa (1979) is the classic reference for empirical work that fails to find any deleterious effects from volatile exchange rates under a floating regime. See Arndt (2002) for an assessment. 4. See Calvo and Reinhart (2002). 5. For an assessment of this argument, see Arndt (2002). 6. Lane (2001) is a recent survey of the connection between openness, trade, and real exchange-rate behavior. 7. See Murray (2000) and Blundell-Wignall and Gregory (1990). 8. See Mussa (2002) for a detailed assessment. 9. See also Rose and Engel (2000). However, recent evidence by Ballabriga, Sebastian, and Valles (1999) shows that the common currency area in Europe has not led to more synchronized business cycles across Europe. 10. See Arndt (2002) for further discussion. Acknowledgments Portions of this paper were presented at the conference Monetary Union: Theory, EMU Experience, and Prospects for Latin America organized by the Oesterreichische Nationalbank, the University of Vienna and the Banco Central de Chile, April 15–16, 2002 in Vienna. Andreas Eckner and Sarah Facey provided excellent research and programming assistance. Comments on a previous draft by Sven Arndt, Heinz Herrmann, Tom Mayer, Weshah Razzak and Joob Swank are gratefully acknowledged. Part of the research for this paper was conducted while Siklos was WLU University Research Professor and Visiting Professor at UTS, Sydney. Financial support from the Social Sciences and Humanities Research Council of Canada and Wilfrid Laurier University is gratefully acknowledged. Previous versions of this paper were presented at the University of Vienna, the first annual Viessmann Conference at WLU, and the Conference on Exchange Rates, Economic Integration and the International Economy at Ryerson University. References Alesina, A., & Grilli, V. (1992). The European Central Bank: Reshaping monetary politics in Europe. In M. Canzoneri, V. Grilli & P. R. Masson (Eds.), Establishing a Central Bank: Issues in Europe and lessons from the U.S. (pp. 49–77). Cambridge: Cambridge University Press.
210
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
Arndt, S. W. (2002). The pros and cons of North American monetary integration. Mimeo. Ballabriga, F., Sebastian, M., & Valles, J. (1999). Interdependence of EC economies: A VAR approach. Mimeo. Bayoumi, T., & Eichengreen, B. (1994). One money or many? Analyzing the prospects for monetary unification in various parts of the world. Princeton, NJ: Princeton Studies in International Finance, No. 76. Bayoumi, I., & Prasad, E. (1996). Currency unions, economic fluctuations and adjustment: Some new empirical evidence. Working Paper No. 96/81. Washington, DC: IMF. Berg, A., & Borensztein, E. (2000). The pros and cons of full dollarization. Working Paper No. 00/50. Washington, DC: IMF. Blundell-Wignall, A., & Gregory, R. G. (1990). Exchange rate policy in advanced commodity-exporting countries: Australia and New Zealand. In V. Argy (Ed.), Choosing an exchange rate regime: The challenge for smaller industrial countries (pp. 224–271). Washington, DC: International Monetary Fund. Buiter, W. (1999). The EMU and the NAMU: What is the case for North American monetary union? Journal of Canadian Public Policy, 25, 3. Calvo, G. A., & Reinhart, C. M. (2002). Fear of floating. Quarterly Journal of Economics, 117(2), 379–408. Corsetti, G., & Pesenti, P. (2002). Self-validating optimum currency areas. Working Paper No. 8783. Cambridge, MA: NBER. Courchene, T. (1999). Alternative North American currency arrangements: A research agenda. Journal of Canadian Public Policy, 25, 3. Courchene, T., & Harris, R. (1999). From fixing to monetary union: Options for North American currency integration. CD Howe Institute Commentary, 127. Courchene, T. J., & Harris, R. G. (2000). North American monetary union: Analytical principles and operational guidelines. North American Journal of Economics and Finance, 1(August), 3–18. Crow, J. (1999). Canadian exchange rate policy. Canadian Public Policy, 25, 3. DeSerres, A., & Lalonde, R. (1994). Symetrie des chocs touchant les régions canadiennes et choix d’un réime de change. Working Paper No. 94–99. Bank of Canada. Dupasquier, C., Lalonde, R., & St-Amant, P. (1997). Optimum currency areas asapplied to Canada and the United States. In Exchange rates and monetary policy (pp. 131–170). Proceedings of a conference held by the Bank of Canada, October 1996. Eichengreen, B. (2001). What problems can dollarization solve? Journal of Policy Modeling, 23(April), 267–277. Eichengreen, B. (2002). When to dollarize. Journal of Money, Credit and Banking, 34(February), 1–24. Eichengreen, B., & Von Hagen, J. (1996). Fiscal policy and monetary union: Is there a tradeoff between federalism and budgetary restrictions? Working Paper No. 5517. Cambridge, MA: NBER. Frankel, J. A. (1999). No single currency regime is right for all countries or at all times. Working Paper No. 7338. Cambridge, MA: NBER. Frankel, J. A., & Rose, A. K. (1998, July). The endogeneity of the optimum currency area criteria. The Economic Journal, 108, 1009–1025. Gramlich, E. M., & Wood, P. R. (2001). Fiscal federalism and European integration: Implications for fiscal and monetary authorities. In M. Braimbridge & P. Whyman (Eds.), Fiscal federalism and European integration. London: Routledge. Forthcoming. Grubel, H. (1999). The case for the Amero. The Fraser Institute, Critical Issues Bulletin. Harris, R. (1999). The NAMU debate. Canadian Public Policy, 25, 3. Hochreiter, E., Korinek, A., & Siklos, P. L. (2002). The potential consequences of alternative exchange rate regimes: A study of three candidate regions. International Journal of Economics and Finance. Forthcoming. Hoover, K. D., & Jordá, O. (2001, July/August). Measuring systematic monetary policy Federal Reserve Bank of St. Louis. Review, 83, 113–138. Kletzer, K., & Von Hagen, J. (2002). Monetary union and fiscal federalism. ZEI. Mimeo. Krugman, P. (1993). Lessons from Massachusetts for EMU. In F. Torres & F. Giavazzi (Eds.), Adjustment and growth in the European monetary union (pp. 241–261). Cambridge: Cambridge University Press. Kuttner, K., & Posen, A. (2001). Beyond bipolar: A three-dimensional assessment of monetary frameworks. International Journal of Finance and Economics, 6(4), 369–387. Lafrance, R., & St-Amant, P. (1999). Optimal currency areas: A review of the recent literature. Working Paper No. 99-16. Bank of Canada.
E. Hochreiter, P.L. Siklos / North American Journal of Economics and Finance 13 (2002) 195–211
211
Laidler, D. (1999). Canada’s exchange rate options. Canadian Public Policy, 25(September), 324–332. Lalonde, R., & St-Amant, P. (1995). Optimum currency areas: The case of Mexico and the United States. Centre for Latin American Studies. Monetary Affairs (July–December), 93–128. Lane, P. R. (2001, August). The new open economy macroeconomics: A survey. Journal of International Economics, 54, 2335–2366. Melitz, J., & Weber, A. (1996). The costs/benefits of a common monetary policy in France and Germany and possible lessons for monetary unions. Working Paper No. 1374. CEPR. Murray, J. (2000, August). Why Canada needs a flexible exchange rate. North American Journal of Economics and Finance, 11, 41–60. Mussa, M. (1979, Autumn). Empirical regularities in the behavior of exchange rates and theories of foreign exchange rate markets. Carnegie-Rochester Conference Series in Public Policy, 11, 9–57. Mussa, M. (2002, July). Argentina and the fund: From triumph to tragedy. Policy Analyses in International Economics, Vol. 67. Washington, DC: Institute for International Economics. Obstfeld, M., & Rogoff, K. (1995). The mirage of fixed exchange rates. Journal of Economic Perspectives, 9(4), 73–96. Rose, A. K., & Engel, C. (2000). Currency unions and international integration. Working Paper No. 7872. Cambridge, MA: NBER. Tornell, A., & Velasco, A. (1995). Fixed versus flexible exchange rates: Which provides more fiscal discipline? Working Paper No. 5108. Cambridge, MA: NBER. Von Hagen, J., & Zhou, J. (2002). The choice of exchange rate regimes: An empirical analysis for transition economies. Working Paper No. B 03 2002. ZEI. Willett, T. (2001). Fear of floating needn’t imply fixed rates: Feasible options for intermediate exchange rate regimes. Working Paper. Claremont Graduate University. Williamson, J. (Ed.). (2002, November). An economic policy agenda for Latin America. Washington, DC: Institute for International Economics. Forthcoming.