Let it float: new empirical evidence on de facto exchange rate regimes and growth in Latin America

Abstract

This paper reassesses the evidence presented in Levy-Yeyati and Sturzenegger (LYS) (2003) on the relation between exchange rate regimes and economic growth. We use their de facto classification as well as their database, in order to gain robustness and efficiency in the results. We run System GMM estimations. Additionally, we focus on Latin American countries for the period 1974-2004. Differently to LYS, our evidence indicates that exchange rate regimes are not significant to explain economic growth, both in a worldwide sample of countries and particularly in Latin America. However, in this region flexible regimes appear to have more advantages in terms of the role of the determinants of economic growth in relation to the other exchange regimes.