Terror attacks and stock-market fluctuations : evidence based on a nonparametric causality-in-quantiles test for the G7 countries

Abstract

We use a novel nonparametric causality-in-quantiles test to study the effects of terror attacks on stock-market returns and volatility in G7 countries. We also use the novel test to study the international repercussions of terror attacks. Test results show that terror attacks often have significant effects on returns, whereas the effect on volatility is significant only for Japan and the UK for several quantiles above the median. The effects on returns in many cases become stronger in terms of significance for the upper and lower quantiles of the conditional distribution of stock-market returns. As for international repercussions, we find that terror attacks mainly affect the tails of the conditional distribution of stock-market returns. We find no evidence of a significant cross-border effects of terror attacks on stock-market volatility, where again Japan and the UK are exceptions as far as terror attacks on the US are concerned. Finally, our results continue to hold following various robustness checks involving model structure, lag-lengths and possible omitted variable bias.