We propose a non-parametric procedure for estimating systemic co-jumps and independent idiosyncratic jumps for 35 stock markets, and study news associated with these jumps as reported in Factiva and Bloomberg from 1988 to 2014. Our results suggest that it is important to distinguish between systemic co-jumps and idiosyncratic jumps. We find both types of jumps have important implications for home-bias investors, while idiosyncratic jumps have economically significant impact on portfolio weights for emerging markets. Our news analysis suggests systemic jumps are typically caused by currency crises, sectoral failure, liquidity issues, and deteriorating economic climate, while idiosyncratic jumps are usually caused by political unrest, currency instability, and large firm effects on small economies. In fact, many of the idiosyncratic jumps share the same origin
although different stock markets experienced the impact differently at different times.