Examining the Balassa-Samuelson hypothesis, with an emphasis on the relative abundance of skilled and unskilled labor: A Markov-Switching approach

The purpose of this study is to investigate the Balassa-Samuelson hypothesis, with an emphasis on the ratio of skilled to unskilled labour. In order to, the Balassa-Samuelson hypothesis was tested using the non-linear Markov-switching method during the period 1973-2016, for Iran economy. Using the Akaike information criterion, MSMAH(2)-AR(3) model was selected as the optimal model, to examine the effect of productivity on the real effective exchange rate. The results showed that the effect of productivity on the real exchange rate was split into two regimes, during the studied period, so that the effect of productivity on the real exchange rate was equal to -0.59 in the regime zero and it was equal to -0.84 in the regime one. So the regime zero is a regime in which the impact of productivity on the real exchange rate is low and the regime one is a regime in which the impact of productivity on the real exchange rate is high. In general, it is observed that the effect of productivity on the real effective exchange rate is negative and significant in both regimes for Iran economy with relative abundance of unskilled labour and the Balassa-Samuelson hypothesis does not hold in any of the two regimes.