The authors provide evidence that lagged U.S. returns significantly predict returns in numerous non-U.S. industrialized countries. But the reverse is not true. Lagged non-U.S. returns display limited predictive ability with respect to U.S. returns. The authors also show that return shocks arising in the United States are only fully reflected in equity prices outside of the United States with a lag. These results hold for both in-sample and out-of-sample tests.

This research is relevant to practitioners for several reasons. First, hedgers can use the predictive ability of U.S. returns to anticipate movements in non-U.S. markets. As such, hedgers can make better forecasts that lead to more efficient risk management decisions. In turn, better risk management decisions could potentially enhance profitability by better mitigating losses.

Second, the authors’ results indicate that international investors may be missing a vital input in their analyses if such investors merely focus on domestic factors when evaluating investments in foreign markets. As the authors show, the predictive ability of U.S. returns can be substantially better than the foreign country’s own economic variables.

Third, the predictive ability of U.S. returns shows that investors cannot simply apply U.S.-based asset pricing models to other countries. Instead, investors must use an international asset pricing model that explicitly incorporates the leading role of the United States.

The authors first estimate regression models of a country’s excess return on its lagged interest rate and dividend yield. They perform this regression for 11 industrialized countries using monthly data from 1980 to 2010. They find that interest rates exhibit stronger predictive ability across countries than dividend yields.

Next, the authors perform Granger causality tests, which is a common test used for studying lead–lag relationships in portfolios of U.S. stocks. The regressions are augmented to include two additional independent variables: the country’s own lagged return and another country’s lagged return. In this step, the authors find that U.S. returns are useful for predicting returns in 9 of the 10 non-U.S. countries and that lagged U.S. returns have an important impact on non-U.S. returns.

Then, they estimate a news-diffusion model, which allows them to examine how return shocks arising in one country affect returns in another country. The results of this analysis suggest that information frictions explain a substantial portion of the predictive power of lagged U.S. returns.

Finally, the authors examine the out-of-sample predictive power of lagged U.S. returns. They find significant out-of-sample results that tend to be concentrated during NBER-dated business cycle recessions and were particularly large during the recent global financial crisis and Great Recession.

From an empirical standpoint, the authors provide a compelling argument for adding U.S. lagged returns to international asset pricing models. The challenge for practitioners is how best to incorporate this variable into their analyses. In addition, it would be interesting to see whether the predictive ability of U.S. lagged returns will persist after its discovery. Otherwise, the result of this research may not be of any practical significance.