REASSESSMENT OF BALASSA-SAMUELSON HYPOTHESIS: EVIDENCE FROM SELECTED EMERGING MARKET ECONOMIES

The widely acknowledged theory rationalizing the existence of long-run deviations is known as the Balassa-Samuelson (BS) hypothesis (Balassa, 1964; Samuelson, 1964). Balassa-Samuelson theory assumes the Law of One Price that holds among traded goods and an increase in productivity of traded goods sector will lead to increase in wages in traded goods sector without a change in the price of goods. The increase in wages of tradable goods sector will demand comparable wages in the non-tradable sector. As a result, relative prices of non-tradable sector rise due to unmatched increase in wages and productivity. Relative prices of non-tradable sector rise and thereby general price level in the economy rises resulting in real exchange rate appreciation. Empirical evidence validates both supply and demand side effects upon relative price movements. BS model assumes that demand side factors do not affect real exchange rate and the real exchange rate is entirely supply side phenomena. The major issues addressed in this thesis are as follows. Firstly, the present study has addressed the inclusion of demand side factor by including terms of trade in the traditional BS framework which is acknowledged as extended BS model to adjust the model to a wide set of market inadequacies and institutional structures. Secondly, the majority of the studies have constructed the tradable and non-tradable sectors by taking primary, secondary and tertiary sectors without looking at the disaggregated level. The bulk of the previous studies treated service sector under non-tradable sector too. Therefore, the present study examines the BS effect by formulating the tradable and non-tradable sectors using 35 industries at disaggregated level data for selected emerging market economies. Third, the majority of researchers in the literature have classified sectors into tradable and non-tradable irrespective of their export performance. The present study has followed exports to total production ratio to estimate tradability of a sector, De Gregorio et al., (1994) and then grouped the tradable and non-tradable sectors. Fourth, while the bulk of studies have used GDP per capita as the proxy for labour productivity, the present study uses a ratio of gross value added to total hours worked by the persons engaged as the measure of labour productivity. Lastly, the majority of studies have examined the BS effect in developed and OECD countries. However, to my knowledge, the examination of the BS effect in the case of emerging economies that to using disaggregated industry level data is scanty. The present study addresses the above issue by analyzing BS effect in selected six emerging market economies. To examine the BS effect, the study uses both benchmark and goods producing approaches. The empirical results from both benchmark approach and goods producing approach confirm that increase in productivity differentials are leading to real exchange rate depreciation and terms of trade are exerting a positive effect on the real exchange rate in the panel of selected emerging market economies. Further decomposing productivity differential, our results confirm that tradable productivity differential is causing depreciation of the real exchange rate whereas nontradable productivity is leading to real exchange rate appreciation with terms of trade incorporated into the model. Policies promoting more tradability of sectors by subsidizing the trade cost, productivity enhancement of sectors, and perfect access to international capital markets may supplement and aid economies in experiencing appreciation