Is there a Purchasing Power Parity (PPP) Puzzle? New Evidence from a Nonlinear Asymmetric Panel Unit Root Test

Abstract

This study re-examines the validity of the purchasing power parity (PPP) hypothesis for 24 OECD countries. The econometric methodology implemented not only allows for asymmetric nonlinear mean reversion within a panel context, but also corrects for the cross-sectional dependence bias frequently encountered in panel data. This feature is important because a test that ignores the presence of asymmetry and cross-sectional dependence when they are in fact present in the data would lead to misleading results. We obtain relatively stronger evidence in favor of the PPP hypothesis when compared to the other alternative panel unit root tests. However, on the whole, this support is still weak even after allowing for asymmetric nonlinear mean reversion in the real exchange rate series. Therefore, to reconcile the data with the theory further methods should be developed.