International stock return predictability: statistical evidence and economic significance

The predictability of stock returns in ten countries is assessed taking into account recently developed out-of-sample statistical tests and risk-adjusted metrics. Predictive variables include both valuation ratios and interest rate variables. Out-of-sample predictive power is found to be greatest for the short-term and long-term interest rate variables. Given the importance of trading profitability in assessing market efficiency, we show that such statistical predictive power is economically meaningless across countries and investment horizons. All in all, no common pattern of stock return predictability emerges across countries, be it on statistical or economic grounds.