Abstract :
In this study, we specify a set of sufficient parity conditions for real interest rates to be equalized internationally, what we call real interest parity (RIP). Using multivariate unit root tests, which have significantly greater power than univariate alternatives, we demonstrate that these sufficient conditions are not satisfied for five industrialized nations over the period of 1960-1996. We then examine each parity condition individually to shed some light on the source of the rejection of RIP. Our results suggest that no single violation can explain the failure of RIP in all cases. It does appear, however, that the Fisher relation is the least likely to violate the RIP equilibrium, whereas uncovered interest parity (UIP) appears to be the most commonly violated. This result is consistent with a nonstationary risk premium in the foreign exchange market.
