Since the collapse of the Bretton Woods system the volatility of the real exchange rate has increased substantially, and it casts a doubt of validity of purchasing power parity (PPP). Empirical literature, however, provides the evidence that PPP seems...
Since the collapse of the Bretton Woods system the volatility of the real exchange rate has increased substantially, and it casts a doubt of validity of purchasing power parity (PPP). Empirical literature, however, provides the evidence that PPP seems not to hold in the post-Bretton Woods period. One way to examine a reversion of the real exchange rate to a certain level is using Dickey-Fuller test for an investigation fo the nonstationarity of the real exchange rate, and another is using the variance ratio test to examine the size in the variance of the long-run changes in the real exchange rate. These methods together present that the real exchange rate deviates from a PPP level for a long period of time and reverts to a component that drives long-run deviations.
This paper sets up a model that takes advantage of the previous empirical result. We posit that the dollar/yen real exchange rate is composed of a permanent component and a transitory component. The permanent component represents the long-run deviations of the real exchange rate from a PPP level and the transitory component represents the temporary deviations from the permanent component. The volatility of the real exchange rate after 1972 is not the same throughout the sample period up to 1994, so that heteroskedasticity is considered in modeling the movements of the real exchange rate. Markov-switching mechanism applied to the variance of the shock in the transitory component helps improve the estimation. We found that the dollar/yen real exchange rate deviates from a PPP level and temporarily does from the permanent component as well. In other words, the dollar/yen real exchange rates show over-shooting from the permanent component.