Measuring Risk Information

Abstract

We develop a measure of how information events impact investors' perceptions of firms' riskiness. We derive this measure from an option-pricing model where investors anticipate an announcement containing information on the mean and variance of firms' future prices. We apply the measure to firms' earnings announcements and show it has many desirable properties: it predicts firms' return volatilities, risk-factor exposures, implied costs of capital, the timing of heightened volatility, and deterioration in fundamental performance, and outperforms textual-based proxies. Together, our study offers an approach for studying risk information conveyed by information events that is simple to implement and broadly applicable.