The Effect of the Amount and Configuration of News on Inferences about Firm-Specific Events

We examine how the amount and configuration of firm-specific news events affects inferences about the informativeness of eight types of firm-specific announcements. After establishing that confounding news events are neither infrequent nor random around these announcements, we investigate how the presence of confounding news events affects measures of announcement period market reactions. We find that the residual variation of asset pricing models (that are used to extract systematic components of returns) are largely unaffected by confounding news events. This aggregate result does not, however, extend to inferences about the informativeness of specific announcements. In particular, while conclusions about the statistical significance of market reactions to firm-specific announcements remain largely intact for most of the events that we consider, the magnitudes of those reactions (that is, their economic significance) are reliably smaller in absolute terms once we control for confounding news events.