Europe’s stock return predictability

In this paper we look into the lead-lag relationships between different stock markets. For this we investigate the effect that the lagged excess returns of one country has on the excess returns of another country using different model specifications. First we follow Rapach et al. (2013) and we find similar results, namely that Sweden shows a leading role in-sample, whereas the U.S. shows a leading role both in-sample and out-of-sample. Furthermore, we extend this research by looking more in-depth into Europe. For this we add seven European countries into the analysis. We find that the lagged excess returns of the U.S. have a predictive ability in the Pairwise Granger causality tests, where we only include the lagged excess returns of the country itself. However, U.S.’ lagged excess returns have no predictive ability in the general model specification, where we include the lagged excess returns of all countries. So it seems that the U.S. lost its leading role. Besides that, we find that Finland shows a leading role both in-sample in all the specifications and out-of-sample. However, we find that the U.S. has better out-of-sample predictive performance than Finland. This, in combination with the result that the U.S. does show a predictive ability in the Pairwise Granger tests, indicates that the U.S. still has a leading role in the international stock market.