Forward market efficiency and foreign exchange rate determination.

Abstract

This dissertation studies the simple efficiency hypothesis, which states that the forward exchange rate is an unbiased and efficient predictor of the future spot exchange rate. This hypothesis has been extensively tested, and is overwhelmingly rejected. However, researchers are unable to determine the exact cause of rejections since it is the result of joint assumptions of risk neutrality and rational expectations. An interest rate differential model is developed assuming that the spot rate follows a random walk process and the covered interest rate parity condition holds. In this model, the spot rate is equal to the sum of the lagged forward rate and the interest rate differential, and a random error term. This model shows that the interest rate differential term belongs to the relationship between the spot and lagged forward rates, but it is not accounted for by the simple efficiency hypothesis. Therefore, the simple efficiency hypothesis is rejected because the interest rate differential term belongs to the spot and forward relationship rather than because of assumptions of risk neutrality or rational expectations. Empirical evidence supports this model using the exchange rates of the United Kingdom, Canada, Germany, Japan, and Switzerland versus the United States. Furthermore, the time series properties of interest rate differentials are sensitive to changes in monetary and fiscal policies. The interest rate differential is non-stationary, and the spot and forward rates are not cointegrated for samples between 1982 to 1993, which is a strong indication that the simple efficiency hypothesis is rejected because of interest rate differentials.