On biases in the measurement of foreign exchange risk premiums

The hypothesis that the forward rate is an unbiased predictor of the future spot rate has been consistently rejected in recent empirical studies. This paper examines several sources of measurement error and misspecification that might induce biases in such studies. Although previous inferences are shown to be robust to a failure to construct true returns and to omitted variable bias arising from conditional heteroskedasticity in spot rates, we show that the parameters were not stable over the 1975-89 sample period. Estimation that allows for endogenous regime shifts in the parameters demonstrates that deviations from unbiasedness were more severe in the 1980s. (JEL F31).