Currency Hedging over Long Horizons

This paper reexamines the widely-held wisdom that the currency exposure of international investments should be entirely hedged. It finds that the previously documented ability of hedges to reduce portfolio return variance holds at short horizons, but not at long horizons. At horizons of several years, complete hedging not only does not lower return variance, it actually increases the return variance of many portfolios. Hedge ratios chosen to minimize long-run return variance are not only low, they also have no perceptible impact on return variance. The paper reports and explores these results, their apparent causes, and investigates their implications for hedging practice.