On Interpreting the Random Walk and Unit Root in Nominal and Real Exchange Rates

The random walk property of exchange rates is regarded as carrying implications for the kinds of shocks that have driven exchange rates and the models appropriate for analyzing their behavior. This paper describes the results of stochastic simulations of Dornbusch's (1976) sticky-price monetary model, calibrated for representative parameter values for the United States. The paper shows that when all shocks are nominal, the model generates time series for real and nominal exchange rates that are statistically indistinguishable from random walks and that cointegration tests can provide misleading information about long-run relationships between the variables in the model.