There is a renewed debate on the role of exchange rate policies as an industrial policy tool in both academic and policy circles. Policy practitioners usually examine real exchange rate misalignments to monitor the behavior of this key relative price and, if possible, exploit distortions in the traded and non-traded relative price to promote growth. Anecdotal evidence shows that some countries have pursued very active exchange rate policies to promote the export sector and enhance growth by undervaluing their currencies. The main goal of this paper is to provide a systematic characterization of real exchange rate undervaluations. The long-run real exchange rate equation is estimated using: (a) Johansen time series cointegration estimates, and (b) pooled mean group estimates for non-stationary panel data. The paper constructs a dataset of real undervaluation episodes. It first evaluates whether (and if so, to what extent) economic policies can be used to either cause or sustain real undervaluations. In this context the paper empirically models the likelihood and magnitude of sustaining real exchange rate undervaluations by examining their link to policy instruments (such as exchange rate regimes and capital controls, among other policies) using probit and Tobit models. Finally, it investigates whether foreign exchange intervention can generate persistent real exchange rate deviations from equilibrium. In general, it finds that intervention can lead to greater persistence in the incidence and magnitude of real exchange rate undervaluations.
