Parity reversion in real exchange rates during the post-Bretton Woods period

A common view among recent studies on purchasing power parity is that the post-Bretton Woods period is far too short to reveal any significant parity reversion in individual series of real exchange rates. Is this really so? The answer, this study shows, depends very much on the statistical test being used. Two efficient univariate unit-root tests are applied to uncover parity reversion. These tests require much shorter sample sizes than conventional tests to attain the same statistical power. Empirical results show that parity reversion can be unveiled over the modern float if an efficient unit-root test is applied. (C) 1998 Elsevier Science Ltd. All rights reserved.