Do Policy-Related Shocks Affect Real Exchange Rates? An Empirical Analysis Using Sign Restrictions and a Penalty-Function Approach

We examine the response of real exchange rates to shocks in real exchange rate determinants, a monetary policy shock, and a fiscal policy shock in 30 countries over the period 1970-2008. The country set is divided into 4 groups - European, developed-country, Asian developing-country, and non Asian developing-country groups. We propose and apply a new approach, i.e. we employ a panel Bayesian structural vector error correction model, and we impose sign restrictions with a penalty-function approach to identify the shocks. We find that most of our impulse response analysis is in line with economic theories. Specifically, there is strong evidence that trade liberalization generates a real depreciation and an increase in government spending leads to a real appreciation over the long run. We also find that a contractionary monetary policy shock has only short-run impacts on real exchange rates, corresponding to the long-run neutrality of monetary policy. The responses to a productivity shock are interesting, i.e. productivity growth in traded sectors has no effect on the real exchange rate of the Asian developing-country group, and it leads to a long-run real appreciation in the non Asian developing-country group. In contrast, this shock causes a real depreciation in the European country group over the long run. Variance decomposition suggests that international trade policy contributes the most to real exchange rate movements in most country groups, with the exception of the non Asian developing-country group, for which fiscal policy via government spending seems to be the most important.