An empirical analysis of exchange rates in target zones

This dissertation consists of an empirical evaluation of exchange rates for two major currencies in the target zone arrangement---France and the Netherlands with Germany as the foreign country.; Chapter 1 provides a background on the evolution of the European Monetary System and presents the theoretical and empirical literature on exchange rate studies. In particular, the focus is on the monetary model of exchange rate determination. The general conclusion is, that in spite of its theoretical appeal the monetary model of exchange rates has performed poorly in the empirical literature of target zones.; Chapter 2 analyses the data and applies the Johansen's cointegration method to determine the presence of a long-run equilibrium in the flexible price monetary model of exchange rates. Using a longer sample period and appropriate estimation techniques, we find evidence of a long-run equilibrium for both France and the Netherlands. When the structural change in the economy is modeled with a dummy variable, the parameter estimate for France conform to the theoretical predictions of the monetary model.; The short-run dynamics of exchange rate are analyzed in a discrete-time frame work in Chapter 3. The monetary model is formulated in the absence of bounds and under the assumption of rational expectations. The simulated expectations explain most part of the variation in exchange rates. We empirically demonstrate the importance of seasonally unadjusted data in formulating rational expectations model. We construct two alternative measures for intra-marginal interventions and propose its use in estimating monetary model of exchange rates in target zone.; Chapter 4 extends the short-run dynamics and presents a model with explicit bands under two scenarios. We estimate a model with a constant band and bands subject to realignments in the presence of intra-marginal interventions. The findings indicate the importance of agent's expectations in determining current exchange rate. Further, we find that the model with jumps performs better than the alternative specifications of no bands and a constant band for both countries. The target zone arrangement is more credible in case of Netherlands than France. Our findings indicate the importance of financial variables in explaining the time varying probability of realignment. We perform tests for nonlinearity and mean reversion. Our results indicate an S-shaped curvature and mean reversion for Netherlands but deliver a mixed message France. Finally we estimate a model or Netherlands with an implicit band and find that the specified bilateral limits of the EMS is not a binding constraint for the Dutch guilder. Instead there is evidence of a target zone operation within narrower confines.