w

This paper applies a two-country framework that allows for currency substitution in an environment in which policymakers optimally vary interest rates in light of utility-based objectives, one country pegs the value of its currency to the other nation’s currency, and government revenue is generated via explicit taxes and seigniorage. The analysis illustrates the roles that currency substitution, currency preferences, and efficiency of tax systems play in contributing to the likelihood of a “run” on one nation’s currency. We explore how these factors interact to influence the probability of a currency crisis in the country that fixes its exchange rate.

Recent crises in Mexico, East Asia, Russia, Brazil, and Argentina have spurred a renewed interest in understanding the sources of currency crises. As a result, an already large literature has expanded further within the past few years. Most research on currency crises have examined variants of two families of theoretical models of speculative attacks. One strand of the literature emphasizes, how inconsistencies between a nation’s economic fundamentals and its exchange-rate target can engender a run on its currency. Another strand focuses on the potential role of self-fulfilling anticipations that can induce crises even when underlying fundamentals are consistent with a pegged exchange rate. Thus far, surprisingly little attention has been given to the role of international interdependence as a factor influencing the likelihood of currency crises (one noteworthy exception to this is the World Bank, 2000). This paper develops one approach to addressing this issue. We explore how currency substitution, monetary policy (via settings of interest rates), and fiscal policy (through taxation) can produce an environment in which there is a general unwillingness by foreign and/or domestic residents to hold a nation’s currency, creating a more fertile environment for a potential “run” on its nation’s currency. Because of the importance that pegged exchange rates played in the recent financial crises, our analysis centers on a regime of fixed exchange rates. We thereby examine factors, in addition to those already identified in the literature, that may contribute to an increase in the likelihood of a currency crisis. In addition to the slight attention paid to the role of interdependence as a factor influencing the likelihood of a currency crisis, there have been relatively few recent studies of currency substitution. Most of these use money demand formulations to estimate degrees of currency substitution for nations within selected regions of the world (see Mizen & Pentecost, 1996, for more detailed discussions). As Giovannini and Turtelboom (1992) indicate, these and other approaches to measuring the extent of currency substitution suffer from a number of conceptual and data problems. Nevertheless, because currency substitution affects real money demand, it necessarily influences a nation’s susceptibility to currency crises. Studies of currency substitution in Latin American nations that historically have been prone to such events, such as Canto and Nickelsburg (1987), typically conclude that there is evidence of a significant degree of substitution among national and foreign currencies by residents of these nations. This relationship between currency substitution and currency crises is inherently complex. The degree of substitution among national currencies can influence the probability of a crisis, but it is also likely that currency-market instabilities can affect the willingness of agents to seek to substitute among currencies. In this exploratory analysis, we focus solely on the former linkage. Our purpose, therefore, is to evaluate how agents’ fundamental preferences regarding currency substitution influences the underlying favorability of conditions that contribute to currency crises through its effects on the demand for real money balances. As van Aarle and Budina (1996) and Imrohoroglu (1996) have recently re-emphasized, the demand for real money balances form the tax base for seigniorage revenues. Click (1998) measures seigniorage in a cross-section of 90 countries. He finds that seigniorage, on an average, finances 10.5% of government spending while conventional taxation covers 78.5% of government spending. In general, seigniorage is more important in developing and emerging nations than in developed nations. For instance, Click’s estimates indicate that seigniorage finances 2% of government spending in the US, 2.4% in Germany, and 5.6% in Japan. In contrast, seigniorage finances 6.3% in Thailand, 6.9% in Indonesia, 5.3% in Malaysia, 13.7% in Brazil, 19.0%in Mexico, and 62.0% in Argentina. Hence, even though seigniorage is not as important as conventional taxation as a source of government revenue for most nations, it is nonetheless an important component, particularly for developing nations that have experienced currency crises.Consequently, the extent of currency substitution impinges on the ability of fiscal and monetary authorities to fund public expenditures via inflation taxes. Our analysis highlights the interaction between currency substitution and seigniorage within the context of a two-country setting. A clear implication is that seigniorage can play a fundamental role in shaping the prospects of currency crises. In the model we develop below, residents of two nations that are trading partners may substitute between currencies issued by both governments.Within this framework, we consider howcurrency substitution influences the likelihood of a currency crisis when there is a stochastic increase in a foreign risk premium. In this regard, our conceptual approach mirrors the aims of authors such as Kaminsky (1998), who seek to develop “early-warning signals” of currency crises. In the analysis that follows, we attempt to identify various factors that tend to increase the likelihood of a crisis. We find that one key factor influencing the probability of a crisis is the extent of currency substitution. Other factors also affect the likelihood of a currency crisis, however. Our approach adds to the existing literature on currency crises by delineating the nature of the interactions among currency substitution and additional factors that together can create an environment that is susceptible to an exogenous crisis event. The paper is organized as follows: Section 2 describes the model that is used to analyze these issues; Section 3 derives the optimal policy settings when one nation pegs its exchange rate; Section 4 analyzes factors that influence the likelihood of currency crises; Section 5 concludes with a summary and discussion of the policy implications of the analysis.

In this paper we developed a two-country framework that allows agents to substitute among domestic and foreign currencies.With this framework, we have analyzed the case in which the domestic country is a relatively large country that determines monetary policy settings optimally, while the foreign country is smaller and pegs the value of its currency relative to the domestic currency. Within this example we have examined the role that currency preferences, currency substitution, and tax policies play in contributing to the potential for agents in both nations to dump their holdings of the foreign currency. Each of these factors influences the extent to which currency substitution increases the likelihood of a currency crisis for the foreign country. Our most important conclusion is that increased currency substitution increases the likelihood of a currency crisis in response to a sudden perception that foreign assets are more risky. Our analysis indicates that there are number of important policy considerations for a small country that chooses to peg its exchange rate. In general, changes in optimal interest rate settings by the monetary authority of the large nation determine the spreading of taxes from explicit taxation to seigniorage revenues, creating a backdrop for a potential currency crises as agents substitute the currency of the large nation for that of the pegging nation. Small-country policymakers must pay particular attention to marginal currency preferences and the tax collection efficiency of the large country, lest they be forced to be the buyer of their currency or to abandon the exchange rate regime. Structural parameters also are important. In our analysis, the marginal preference for the small country currency is an exogenous parameter. In fact, this parameter realistically is affected by microeconomic policies, such as the regulatory framework for the nation’s system of intermediaries. Such regulatory policies, therefore, are critical in the face of increasing currency substitution. Future research should expand the type of analysis conducted here by considering the demand for the bank deposits within each nation (as in the models of Miller, 1998a, 1998b; Daniels &VanHoose, 1996, for example). In this way, currency and banking crises may be considered simultaneously in the context of international interdependence. This will allow for an analysis of the undoubtedly important role of reserve requirements as a factor affecting the risk of international crises. A further issue is the likely endogeneity of currency substitution to currency crises. In this paper we have examined a unidirectional causality in which greater currency substitution contributes to a climate favorable to a crisis. In the event that shocks actually induce a crisis, of course, the resulting instabilities undoubtedly affect agents’ willingness to substitute currencies. Broadening our framework to allow for a full bidirectional relationship between currency substitution and crises is another area for future research.