The real exchange rate and the Balassa–Samuelson hypothesis in SAARC countries: an appraisal

This paper empirically tests the Balassa–Samuelson (B-S) productivity bias hypothesis in seven South Asian countries. We use auto regressive distributed lag (ARDL) modelling to develop a dynamic structure of the B-S hypothesis. Using annual data from Penn World Table Version 7.0 and the ‘bounds test’, we found no evidence of the B-S hypothesis in six sampled countries (Bhutan, India, Maldives, Nepal, Pakistan and Sri Lanka) and the only exception was Bangladesh. We found that a 1% increase in labour productivity in Bangladesh relative to the US will lead to 2.5% appreciation in the real exchange rate of Bangladesh. The speed of adjustment towards equilibrium was found to be high, with short-run disequilibrium correcting by nearly 25% per annum in Bangladesh. The two endogenously determined structural breaks are found to be negative and statistically significant. These results add new insights to the time-series literature on the B-S hypothesis.