This paper shows that the systematic risk (or "beta") of individual stocks increases by an economically and statistically significant amount on days of firm-specific news announcements, and reverts to its average level two to five days later. We employ intra-daily data and recent advances in econometric theory to obtain daily firm-level estimates of beta for all constituents of the S&P 500 index over the period 1995-2006, and estimate the behavior of beta around the dates of over 22,000 quarterly earnings announcements. We find that the increase in beta is larger for more liquid and more visible stocks, and for announcements with greater information content and higher ex-ante uncertainty. We also find important differences in the behavior of beta across different industries. Our analysis reveals that changes in beta around news announcements are mostly driven by an increase in the covariance of announcing firms with other firms in the market. We provide a simple model of investorsíexpectations formation that helps explain our empirical findings: changes in beta can be generated by investors learning about the profitability of a given firm by using information on other firms.
