Currency Risk Premia in Emerging Markets

Abstract

Understanding the dynamics of time-varying currency risk premia is the central issue of current empirical FX research. Since most findings, however, are based on developed market (DM) currencies, this paper contributes in such a way that it explicitly analyzes the specific dynamics of emerging market (EM) currencies. We use the well-documented Harrod-Balassa-Samuelson effect to relax the assumption of long-run purchasing power parity, which cannot be assumed to hold for EM economies. We show that the interest rate differential is the best positive predictor of short-run currency returns, while the real exchange rate (RER) is a better negative predictor over longer horizons. Interestingly, the positive relation between the interest rate differential and future currency returns reverses at a 6-months time horizon for EM currencies. Besides that, we find that - in contrast to DM currencies - high-yield EM currencies do not tend to appreciate over the short-run, and that the predictive power of the RER is the strongest for the lowest 12-months ahead EM currency returns, while it becomes insignificant for the highest 12-months returns.