Conditional Hedging and Portfolio Performance

Simple conditional currency-hedging rules often increase risk-adjusted portfolio returns and are thus of interest to investors. Several researchers have reported successful application of a “forward hedge rule” (FHR) in which one hedges whenever the foreign currency trades at a forward premium. An alternative strategy, a “real-interest-rate hedge rule” (RIR), is based on hedging when the domestic real interest rate exceeds the foreign rate. As an extension, we propose a combination of these rules—a “real forward hedge rule” (RFHR). We evaluate the performance of the rules for various currency, stock, and bond portfolios from the developed countries. In tests of risk-adjusted returns for 1976–1997, the RFHR significantly outperformed standard benchmarks and often beat the FHR and the RIR. Moreover, results of a simple dominance test for rolling 5- and 10-year periods suggest that the RFHR consistently improves portfolio performance for a U.S. investor. Managers of international equity and bond portfolios face added complications from currency fluctuations, so currency hedging is often used to minimize risk. Simple conditional currency-hedging rules may decrease risk, but they also frequently increase risk-adjusted portfolio returns (the Sharpe ratio). We tested the efficacy of a “real forward hedge rule” applied to equally weighted portfolios of currencies, stocks, bonds, and stocks plus bonds from the G–5 countries (France, Germany, Japan, the United Kingdom, and the United States) over a 21-year period. The rule we proposed outperformed both conventional benchmarks and two other conditional hedging rules. These results suggest that U.S. managers could benefit from using this rule.Conditional hedging rules make hedging decisions on the basis of the information available to individuals at the investment decision point. Several researchers have reported successful application of a forward hedge rule (FHR), in which one hedges whenever the foreign currency trades at a forward premium. This strategy is based on an assumption that exchange rates follow roughly a random walk and that managers are rewarded by hedging whenever they can capture the forward premium. Other researchers have reported evidence supporting a real-interest-rate hedge rule (RIR) in which one hedges when the domestic real interest rate exceeds the foreign one. This rule is based on the supposition that purchasing power parity is a better predictor of the subsequent exchange rate than is the forward rate.Recognizing that the RIR might help improve results obtained from applying the FHR (because the RIR links inflation expectations to exchange rates and helps explain the reasons for differences in interest rates), we propose a combination of these rules in a real forward hedge rule (RFHR). The RFHR invokes hedging only when (1) the foreign currency is at a forward premium and (2) the domestic real interest rate exceeds the foreign rate. This combined rule hedged less often than the FHR or RIR (about 34 percent versus 50 percent of the time), but it proved to be correct more frequently (59 percent of the time versus 57–58 percent) during the period of our study.We tested the RFHR against standard unconditional benchmarks (never hedged, always half-hedged, and always fully hedged) and against the FHR and RIR for equally weighted portfolios of G–5 currencies, stocks, bonds, and stocks plus bonds. Monthly observations for 1976–1997 were drawn from Datastream, and hedging transaction costs of 5 bps a month were assumed. We adopted the viewpoint of a U.S. investor throughout.With one exception (fully hedged bonds), the RFHR led to significantly higher Sharpe ratios than did the unconditional benchmarks. Also, although the differences in Sharpe ratios between the RFHR and the FHR or RIR generally were not statistically significant, the RFHR showed clear superiority in a simple dominance test. In this test, we computed Sharpe ratios for rolling 5- and 10-year periods of monthly observations and counted the number of times each strategy had the highest Sharpe ratio. That is, for the 204 overlapping 5-year periods for equity (or bond or bond plus equity) portfolios, the dominance test counted the number of 5-year periods in which application of each of the six strategies resulted in the best Sharpe ratio. The test found that for all but the rolling 5-year bond portfolios (in which the FHR, the RFHR, and the fully hedged strategies were roughly equally successful), use of the RFHR strategy led to the highest Sharpe ratio most often. In fact, in many cases, the number of periods in which the RFHR dominated was greater than that of any three competing approaches combined. Although the dominance test is not a statistical test, it does indicate a high level of consistency in performance.One can conclude from this study that application by U.S. investors of the RFHR to international portfolios often leads to better results. The combined rule is less successful with bonds, which may not benefit from the added information from the RIR because bonds are more highly correlated with interest rates. Use of the RFHR with equity portfolios and with stock plus bond portfolios, however, has the potential to provide strong benefits.