Sources of Real Exchange Rate Fluctuations: How Important are Nominal Shocks?

This paper investigates empirically and attempts to identify the sources of real exchange rate fluctuations since the collapse of Bretton Woods. The paper's first two sections survey and extend earlier, non-structural empirical work on this subject by Campbell and Clarida (1987), Meese and Rogoff (1988), and Cumby and Huizinga (1990). The paper's main contribution is to build and estimate a three equation open macro model in the spirit of Dornbusch (1976) and Obstfeld (1985) and to identify the model's structural shocks - to demand, supply, and money -using the approach pioneered by Blanchard and Quah (1989). For two of the four countries we study, Germany and Japan, our structural estimates imply that monetary shocks, to money supply as well as to the demand for real money balances, explain a substantial amount of the variance of real exchange rates relative to the dollar. We find that demand shocks, to national saving and investment, explain the majority of the variance in real exchange rate fluctuations, while supply shocks explain very little. The model's estimated short run dynamics are strikingly consistent with the predictions of the simple textbook Mundell-Fleming model.