«Do Currency Regimes Matter in the 21st Century? An Overview Hiroshi Fujiki and Akira Otani This paper selectively reviews the recent literature on ...»
19. As for the reasoning of this argument, Folkerts-Landau and Garber (1992) state, “Financial systems with a limited extent of securitization have in practice a small number of large universal banks in the market for wholesale funds. Wholesale payments and securities transactions are cleared internally in these organizations.
The risk of nonsettlement is low due to the lack of significant exposure to non-bank financial institutions and an increased ability to work out unexpected problems quickly among the small number of players. Hence, although the clearing banks ultimately clear on the books of the central bank, there is little need for the central bank to provide intra-day credit or stand ready to act as lender-of-last-resort to the clearinghouse to ensure the payments settlements.”
20. Parti and Schinasi (1999) also insist that the absence of the central LLR and supervising arrangement may undermine the existence of EMU in the face of area-wide financial crises.
21. Buiter (2000a) also states that LLR actions should be left at the national level, subject to ECB oversight and coordination, since the capital of the ECB is limited and it is not backed, either formally or informally, by the deep pockets of a ministry of finance.
22. Optimum currency area criteria suggest the necessity of fiscal transfer within monetary union. If one region suffers from high unemployment due to a region-specific negative shock, fiscal transfers from low-unemployment regions to high-unemployment regions can smooth out the shock without adjusting nominal exchange rates.
56 MONETARY AND ECONOMIC STUDIES (SPECIAL EDITION)/DECEMBER 2002 Do Currency Regimes Matter in the 21st Century? An Overview Many economists have conducted empirical analyses regarding the necessity of central fiscal policy to guard against asymmetric shocks. For the United States, von Hagen (1992) estimated that 47 cents of net federal transfers would be made in response to a US$1 difference in the level or change in state income compared to U.S. average income. However, for economies other than the United States, such as Canada, many research findings have shown different estimates of the transfer in a range from more than 10 to more than 50 cents (see Kletzer and von Hagen  for a recent review). These studies indicate that fiscal transfer may be significant in some existing monetary unions, but it is difficult to answer the question of how important it is in practice for the stabilization of the regional economies.
Another reason for the need of central coordination of fiscal policy within the EMU could be the possibility of the dynamic inconsistency problem. One may well wonder whether fiscal authorities would expect the EMU to accept discretionary fiscal policies that are slightly inconsistent with the Maastricht Treaty, because a single monetary policy cannot emphasize a particular economic situation of member economies.23 Von Hagen et al. (2001) argue that the EU surveillance of public finance should focus on the content of consolidation efforts, as well as the ceiling on the gross government debt to GDP ratio and general government deficit to GDP ratio to achieve prudent fiscal policies in monetary union.
c. Democratic control (accountability) of the ECB Some economists are concerned about issues relating to democratic control of the ECB, such as its independence and accountability. First, concerning the independence issue, Feldstein (1997) argues that pressure from governments on the ECB would bias monetary policy.24 The decision-making body of the ECB, the Governing Council, consists of six Executive Board Members, appointed by the European Council, and the central bank governors of the EMU member economies. Monetary policy is decided by simple majority in the Governing Council. Therefore, the representatives on the ECB Governing Council might reflect their national attitudes and face political pressure to represent what domestic governments perceive to be their national interests.25 Second, concerning the accountability issues, Buiter (2000a) states that the U.K. arrangements, with an operationally independent central bank pursuing a politically mandated set of objectives, are superior to the current EMU arrangements in a democratic economy. Regarding the procedural openness and transparency of monetary policy, he also suggests that the individual voting records of the members of the Governing Council and its minutes should be in the public domain.
Hämäläinen (2001) opposes these criticisms. First, the Maastricht Treaty prohibits the ECB and NCBs from taking instructions from any external bodies; therefore, the
23. Von Hagen et al. (2001) show the empirical evidence that the pressure of the Maastricht Treaty might have resulted in some short-lived and revenue-based consolidation efforts during the recession years of the early 1990s.
The policy implication of their evidence could be that the pressure from the Maastricht Treaty should not be underestimated.
24. Feldstein (1997) suggests that the future average rate of inflation would rise and that the net economic effect of EMU would be negative, based on the above argument.
25. The future expansion of EMU means an increase in membership of the Governing Council. If Feldstein (1997) is right, in our interpretation, the future expansion of EMU members might also increase the risk that the ECB will receive greater pressure from governments and that its monetary policy will be biased.
B. Expansion of EMU in the Future The discussion so far shows that the introduction of a common currency is not a panacea, and unsolved problems remain with regard to EMU. Thus, it is understandable that some EU economies had still not joined the euro at the time this paper was written.
However, the euro area has its own frontier: with East European, Mediterranean, and African economies. Currently, many East European transition economies are applying for EU accession, and trade and financial links, as well as political dialog, between these economies and the EU have been deepening.26 Noyer (2000) expects the future expansion of the euro area, stating, “The more a process of regional integration moves beyond a free trade area toward a single market, or even an economic union, the stronger the need for intra-area exchange rate stability, and eventually, irrevocably fixed exchange rates.” Are there any pitfalls during the accession process?
To analyze this question, we should take care to distinguish EU membership from EMU membership. EMU candidates must meet both the exchange rate criterion27 and the inflation rate criterion28 in addition to the fiscal deficit criterion. Buiter and Grafe (2001) suppose that the productivity growth differential between the traded and the non-traded sectors is larger in the accession economies than EMU member economies. This means that the relative price of non-traded goods to traded goods is higher in the accession economies and their overall inflation rate will be higher at a given exchange rate—the Balassa-Samuelson effect. Then, the introduction of the euro in accession economies would lead to a period of declining prices and large costs of adjustment to meet the Maastricht ceiling under sticky prices. Buiter and Grafe (2001) suggest that the inflation rate criterion should be specified more concretely in terms of traded good prices to resolve this conflict. Noyer (2001) responded to such a concern on the consistency of nominal and real convergence during the accession
26. Regarding the choice of exchange rate regimes in the accession economies, at the beginning of the 1990s, conventional fixed pegs (intermediate based on Fischer ) were the most common type among those economies. As of the year 2000, except for Hungary and Slovenia, accession economies employed either hard pegs (a currency board in Estonia, Lithuania, Bulgaria, and Latvia) or a float (the Czech Republic, Poland, Romania, and the Slovak Republic). Thus, Begg et al. (2001) report that after the liberalization of short-term capital flows in the accession economies, the bipolar view applies to these economies. See von Hagen and Zhou (2002a) for an empirical explanation of the choice of rate regimes that seems to be consistent with the openness criterion of the optimum currency area in transition economies. Von Hagen and Zhou (2002b) compare de facto exchange rate regime classification based on methods by Levy-Yeyati and Sturzenegger (2000) and official classification by the IMF. Von Hagen and Zhou (2002c) show that the exchange rate regime choice affects capital controls, while feedback effects from capital controls on the choice of exchange rate regime are absent. We omit the discussion on two small economies negotiating entry into the EU (Cyprus and Malta) as well as Turkey.
27. The exchange rate criterion requires EMU candidates to stabilize their exchange rates within a ±15 percent band for two years before joining EMU without capital or exchange controls.
28. The inflation criterion requires that the inflation rate must not exceed the average of the three lowest-performing economies by more than 1.5 percentage points.
58 MONETARY AND ECONOMIC STUDIES (SPECIAL EDITION)/DECEMBER 2002 Do Currency Regimes Matter in the 21st Century? An Overview process as follows. First, most empirical studies estimate the Balassa-Samuelson effect within a range of 1 to 2 percentage points. Second, the Maastricht inflation criterion, which will not be revised to take into account any possible Balassa-Samuelson effect, should not be seen as an immediate requirement for these countries, but rather as a medium-term objective for central banks.
IV. The Experience of the Americas In the Americas, following the large currency crises, such as those in Argentina (2002), Brazil (1999), and Mexico (1994), large economies have moved toward floating exchange rate regimes. Canada, which has sometimes been advised by economists to join the U.S. currency union, remains under a floating exchange rate regime.
However, several small economies are officially dollarized, such as Panama, and some economies have opened the way to unilateral dollarization, such as Ecuador or El Salvador.
In this section, following Edwards and Magendzo (2001), we define “dollarization” as a policy proposal that emerging market nations should give up their national currencies and adopt advanced nations’ currency as legal tender. We identify the dollarization proposal as one of two poles.29 We will first discuss the pros and cons of dollarization relative to clean floating in detail in the context of Latin American economies. Note that the current choice of a floating exchange rate in many American economies does not necessarily mean that there is no possibility of a regional currency area in the Americas in the future. Thus, we will discuss this issue in the latter part of this section.30 A. The Pros and Cons of Unilateral Dollarization
1. The case for dollarization Calvo (2000) points out compelling reasons for emerging market economies to avoid exchange rate flexibility. Dollarization may be costly, but it may put emerging markets on the first stage of the track leading toward monetary and financial stability. He observes that the heart of the problem may lie in imperfect information, inexperience in handling sudden large capital inflows, and shaky political equilibrium, especially in Latin America.
Calvo (2000) summarizes Mundell’s condition regarding the choice of exchange rate regime. Consider a simple model, y = αe + g + u and m = y + v. Here, y, e, g, and
29. Broader definition of dollarization would include the holdings by residents of a significant share of their assets in the form of foreign currency-denominated assets and the use of foreign currency-denominated assets for transaction (see Baliño et al. ). It is well known that in high-inflation economies, citizens abandon the local currencies and choose to use the dollar as the medium of exchange, and this reduces the base for inflation tax and introduces a parallel currency. Needless to say, the allocation of seigniorage between the United States and a dollarized economy depends on the nature of the legal framework proposed in the dollarized economies.
30. This paper does not attempt to list desirable monetary policy strategies for all economies in the Americas.
Mishkin and Savastano (2001) regard hard pegs and constrained discretion based on inflation targeting as two promising strategies. They conclude that the choice between the two depends on political and institutional factors which constrain monetary policy.
59 m denote the logs of output, the nominal exchange rate, the shift parameter of an external factor such as U.S. demand, and money supply. The first equation is an open-economy IS curve, and the second equation is an LM curve. u and v show random shocks, and α is a positive constant. The existence of a fixed exchange rate means e is constant and m is endogenous. The floating exchange rate means m is constant and e is market determined. Hence, under the fixed exchange rate regime, var( y ) = var(u + g ), and var(e) = 0, while under the float, var(y ) = var(v ), and var(e) = (1/α 2)var(u + g + v ). If we worry about the var(y ), a fixed exchange rate is better if var(v) is larger than var(u), abstracting momentarily from g.
In practice, policymakers do not know the size of var(v ) and var(u). It would be better to have a discretionary exchange rate policy depending on the shock, but for most emerging market economies it would be impossible. More seriously, if most bank lending is dollarized, an unexpected devaluation of the nominal exchange rate could lead to debt deflation, hence it may make more sense to pay attention to var(e ), rather than var(y ). In addition, if the markets of emerging economies are subject to asymmetric information in financial markets, it makes sense to offset var(v ) by just pegging the nominal exchange rate.31 By means of dollarization, an economy’s monetary policy obtains credibility, lowers the information cost, and moderates relative price changes compared with floating.