Does the Balassa-Samuelson hypothesis still work? Evidence from OECD countries

Abstract

This study tests the Balassa-Samuelson (BS) hypothesis, which explains the real exchange rate with relative productivity differences, for ten OECD countries between 1975 and 2007 by using the Johansen cointegration approach. The study further tests the effect of the terms of trade variable by adding it into a standard BS model. The cointegration analysis confirms the existence of cointegration among real effective exchange rate, relative productivity and terms of trade. The study’s major findings suggest that the BS hypothesis still works well in OECD countries while explaining the long-run movements of the real exchange rates that it holds for eight out of ten OECD countries in the extended BS model. The results also exert the existence of significant cross-country differences about the size the BS effects. These results are important for calibrating the real exchange models, which is crucial for sustainable allocation of resources in international trade and finance.