Facing up to the polysemy of purchasing power parity: New international evidence

Abstract The purchasing power parity (PPP) hypothesizes the real exchange rate to be a stationary process. Since the test results are model-dependent, the hypothesis has long been debated. This paper makes a new contribution to the PPP debate by exploring why PPP is more robust to model specification in some countries than others. To this end, we adopt a variety of the state-of-the-art unit root tests, which are categorized into four versions of PPP, according to whether the exchange rate process is level (trend) stationary with temporary (permanent) structural break(s). We then investigate why one or more versions of PPP hold for a specific country. By using the real effective exchange rates of 23 OECD countries plus the euro area, we find that the size of GDP, the membership of the euro area, and the government debt-to-GDP ratio have significant effects on the validity of the PPP hypothesis.