BALASSA-SAMUELSON HYPOTHESIS: A TEST FOR THE TURKISH ECONOMY

The Balassa-Samuelson (B-S) hypothesis relies on the productivity differential between tradable and non-tradable sectors to explain deviations in purchasing power parity. According to the B-S hypothesis, because productivity growth in tradable sectors is higher than in non-tradable sectors, real wages increase in tradable sectors. On the other hand, because the prices of tradable goods are determined in the world market, tradable prices will not increase. With an assumption of perfect labor mobility within a country, increases in wages in tradable sectors will be reflected in non-tradable sectors as well. However, an increase in wages in non-tradable sectors is not accompanied by an increase in productivity. As a result, the prices of non-tradable goods will increase, leading to an increase in the overall price level and the appreciation of the real exchange rate for the domestic economy. Thus, within this framework, the relative productivity differences in tradable vis-à-vis non-tradable sectors between two countries will determine the long-run changes of the real exchange rate. The Turkish Central Bank in recent years has emphasized the importance of the Balassa-Samuelson hypothesis for Turkey (Inflation Outlook II, 2006). The purpose of this study then, given the increased consideration of the Balassa-Samuelson effect by the Central Bank is to test the hypothesis between Turkey and the Euro-27 area, which includes the majority of Turkey’s main trading partners, for the post-financial liberalization era, using time series methods. Given the dataset and econometric techniques used, the results do not support the B-S hypothesis.