The Balassa-Samuelson Effect in Slovenia: an Olley-Pakes Total Factor Productivity Approach

This paper introduces Olley-Pakes productivity estimates into the well-known Balassa-Samuelson framework. The Balassa-Samuelson theorem as developed by Balassa (1964) and Samuelson (1964) states that productivity differentials between the tradable and non-tradable sector are the main drivers for an appreciating real exchange rate. Previous empirical studies on the Balassa-Samuelson effect implemented productivity measures which suffer from econometric issues, most notably simultaneity. I correct for simultaneity by using the Olley-Pakes method of productivity estimation. Firm level data provided by the Slovenian Statistical Office allowed me to reliably estimate productivity differentials. I find evidence of a present bias where productivity differentials are typically overestimated if simultaneity has not been accounted for. Following this result I was able to find evidence of a present Balassa-Samuelson effect in levels. However, due to non-stationarity I implemented first differences where I was not able to find cointegration. Consequently, I conclude that the Balassa-Samuelson effect is not as prevalent when econometrically sound productivity measures are implemented.