Testing the BalassA-Samuelson hypothesis in two different groups of countries: OECD and Latin America

This paper studies the Balassa-Samuelson hypothesis (BSH) in the context of two areas with strong differences in economic development, twelve OECD countries and twelve Latin American economies, taking the USA as the benchmark. Applying panel cointegration techniques, we find that while the first stage of the hypothesis, which links productivities and prices, is satisfied in each group of countries, the second stage, which relates relative sector prices with the real exchange rate, only holds in the Latin American area. The failure of the latter in the OECD countries as a whole is reflected in departures from PPP in the tradable sectors.