Testing the presence and the absence of purchasing power parity: results for fixed and flexible regimes

Recent empirical studies concerning purchasing power parity (PPP), using the concept of cointegraton, have tested the null hypothesis of the absence orthe presence of PPP. The disadvantage of using either of these approaches is that the conclusion is ambiguous when not rejecting the null: the result could be due to either the null hypothesis being true or by the low power of the test and lack of information in the data. Therefore, there is substantial information to be gained by applying both approaches within the same study. This study uses the ordinary DF test with PPP as the alternative hypothesis and a recently suggested test by Kwiatkowski et al. (1992) with PPP as the null. The tests are applied to four major industrial countries real exchange rates, each split into four different samples, including fixed as well as flexible exchange rate regimes. The absence of PPP cannot be rejected while the presence can be rejected.