Capital Flows and Exchange Rates An empirical Analysis

This paper investigates the empirical relationship between capital flows and nominal ex-change rates for five major countries. It is well known that no theory based on current account or interest rates has ever been shown to work empirically at short to medium horizons. Recent international finance theory, however, suggests that currencies are influenced by capital flows as much as by current account balances and log-term interest rates. Using unrestricted VAR's we document the following: a) Incorporating net cross-border equity flows into linear exchange rate models can improve their in-sample performance. Using net cross-border bond flows, however, has no such effect; b) Positive shocks to home equity returns (relative to foreign markets) are associated with short-run home currency appreciation and equity inflow. Positive shocks to home interest rates (relative to foreign countries) cause currency movements that are not consistent with uncovered interest rate parity (UIP); c) An equity-augmented linear model supports exchange rate predictability and outperforms a random walk in several cases. Such superior forecast performance, however, depends on the exchange rate and the forecast horizon.