Essays on exchange rates and inflation (original) (raw)

and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal Chapter 1 frequent realignments. However, after the realignment of March 1983, France shifted away from its policy of budgetary stimulation. Since then, devaluations became rare, while, after January 1987 until very recently, no realignments occurred.' The perceived success of the system during this period, led to more concrete plans for the realisation of a European Monetary Union (EMU) by the Delors Committee (1989). Important questions were raised, such as the institutional setup and status of a European central bank and the speed at which the transition to an EMU should take place. Other problems of such a transition, which could have been addressed within the context of the literature on exchange rate bands, are the amount of reserves needed by the central banks and the trade-off between exchange rate variability and interest rate variability (see Svensson, 1991a), when the fluctuation margins of the exchange rates are reduced. The progress of monetary integration culminated in the Maastricht Treaty in December 1991, in which, among other things, the time-schedule for the transition to an EMU and the conditions for admission into this system were laid down. By then, already, the progress of monetary unification had weakened. Under the threat of inflation caused by the German unification, the Bundesbank raised interest rates, which had to be followed by the other EMS participants in order to avoid devaluation of their currencies. Furthermore, the criteria for admission to the EMU, as specified in the Maastricht Treaty raised serious doubts about the eligibility of some countries to participate in the EMU. Subsequently, the difficulties in the ratification process of the Treaty (witness the Danish rejection and the French referendum) were obviously too much for a number a central banks to knock down the massive speculations against their currencies. The British pound and the Italian lira, which had suffered from yeats of overvaluation, were forced to leave the system in September 1992. Within a few months time after that, the Spanish peseta, the Portugese escudo and the Irish punt have been devalued, while the French franc and the Danish kroner have been under have speculative pressure for some time. 1.2. Why managing exchange rates? An obvious question, which might have become relevant again with the recent developments in the EMS, is why one would want to have bands on the exchange rate or, more generally, why one would want to manage exchange rates at all. Despite its long history, the debate about the desirability of exchange rate management is still open both for policymakers and academics. ' The only change in this period occurred in January 1990, when the band on the Italian lira was reduced from tóo~o to }2.25"~0. Lttroduction 3 One of the main arguments in favour of flexible exchange rates is that they make it easier for economies to adjust to asymmetric shocks. Exchange rates can move fast in response to disequilibria and adjustment via a change in the exchange rate means that less, maybe more costly, adjustment is needed elsewhere in the economy. However, there is ample evidence that nominal exchange rates do not respond to deviations from purchasing power parity, at least not in the short run. For example, Mussa (1986), documents a positive correlation between nominal and real exchange rates for a broad group of countries. The correlation is stronger for more flexible nominal exchange rate regimes. Broadly speaking, the advocates of managed exchange rates fall into two camps (see Williamson, 1993). One group regards a commitment to a fixed exchange rate as providing a nominal anchor for the economy. This philosophy may have nurtured the attemps to suppress realignments in the EMS after January 1987. The other group regards exchange rate management as a device against destabilising speculation or erratic movements of nominal exchange rates from their equilibrium values. Williamson, therefore, argues that exchange rate management should be directed at targeting the real exchange rate around some fundamental equilibrium level (which is partly determined by purchasing power parity considerations; see also Williamson, 1985) and, in order to prevent speculative runs, providing enough flexibility to shift the target zone if this is required by changes in the equilibrium level. From this point of view, it was the wrong policy to defend the pound Sterling and the halian lira against a realignment, because they had deviated too far from their purchasing power parities. This view, however, denies the role of the EMS as a disciplinary device (cf Giavazzi and Pagano, 1988) on policymakers to keep inflation under control. By hanging their political prestige on avoiding devaluations against the Deutsche Mark, policymakers enhance the credibility of their commitment to the exchange rate band. A credible commitment to targeting one's own currency to a strong currency implies low money growth, low inflation and therefore moderate wage claims. The net result is the absence of a wage-price spiral with its distortionary losses. Therefore, a relevant question in taking a position between the two views is whether inflation differentials, which are a major determinant of the 'fundamental equilibrium exchange rate', are an endogenous phenomenon (a self-fulfilling wage-price spiral) or whether they are caused by asymmetric exogenous developments or by structural deficiencies in the economic system, such as an underdeveloped tax system. Unfortunately, it is beyond the scope of this thesis to pursue this question in greater detail. 4 Chapter 1 1.3. Modeling exchange rate bands To highlight how chapters 2 to fit within a more general picture, a very short overview is given of the literature on exchange rate bands. For a more complete overview, the reader is referred to Svensson (1992c), who assesses the ability of these models to explain the empirically observed behaviour of exchange rates in bands, and Bertola (1993), who provides a more general theoretical survey of continuous-time interven[ion models. There are several reasons for modeling exchange rate behaviour in a band explicitly, rather than treating the exchange rate as being completely fixed. In the first place, it is not clear for which bandwidth it is still safe to model the exchange rate as being fixed. Of course, the answer depends on the underlying problem that is being investigated. Secondly, by allowing for an explicit role of the band, one might address a number of interesting issues, which otherwise would be very difficult to investigate, concerning, for example, the optimal width of the band, optimal intervention policy inside the band, the transition from a band on the exchange rate to a fixed exchange rate, the relationship between the position in the band and the danger of a realignment, and predictabiliry of exchange rate changes based on past behaviour in the band. Explicit modeling of exchange rate bands originates from the literature on target zones, which were originally proposed as a zones for real exchange rates, in order to avoid that destabilising speculation would cause too large disparities of real exchange rates from their fundamental equilibrium values (Williamson, 1985; Krugman and Miller, 1992). Interventions are supposed to take place only at the boundaries of the target zone. Krugman (1991) was the first to develop a formal model of exchange rate behaviour in a target zone. In his standard target zone model, it is however the nominal exchange rate, rather than the real exchange rate, which is kept in a band by interventions at the boundaries. The main result is that, even without intervention now, the exchange rate is stabilised by the prospect of future interventions at the boundaries. The theoretical implications of the standard model have been worked out in detail. Svensson (1991a) studies interest rate differentials at infinitesimal maturity and shows that there is a trade-off between the instantaneous variabilities of [he exchange rate and the interest rate differential. Thus, if the width of the exchange rate band goes to zero, then the instantaneous interest rate differential variability gces to a positive limít. For narrow bands the asymptotic variability of the interest rate differential increases in the width, while for wide bands the asymptotic variabiliry decreases in the width. Svensson (1991a) also shows that, if the fundamental is a regulated Brownian motion without drift, then the unconditional exchange rate