Evaluating currency market efficiency: are cointegration tests appropriate?

Abstract

This paper investigates the claim that the common finding of cointegration between spot and lagged forward exchange rates reflects the existence of covered interest arbitrage and not, as is generally accepted, long-run market efficiency. Breuer and Wohar's (1996) methodology is employed to match spot and one-month forward rates correctly for three major currencies; the Deutschmark, Sterling and the Yen, relative to the US dollar. Bi-variate analysis shows that spot and lagged forward rates are cointegrated with the vector (1,-1), a necessary condition for market efficiency. However, at variance with theory, in a tri-variate VECM estimation, the spot rate, lagged forward rate and lagged interest rate differential are shown to be cointegrated with the vector (1,-1,1) for the Mark and Sterling. The ‘cointegration’ paradox is explained by investigating the relative magnitudes of the forecast error and the interest rate differential. It is demonstrated that it is impossible to distinguish between the influence of covered interest arbitrage and the existence of market efficiency using cointegration-based tests.