| We examine the ability to extract risk information from the market prices of a firm's securities. We use contingent claim models for firm valuation to construct risk measures from equity prices, debt prices, and a combination of both. We provide empirical evidence on the relative accuracy and forecasting ability of these measures for industrial and financial firms.; We compare a number of methodologies for constructing implied asset volatility estimates for industrial firms. We document that while different methodologies produce different estimates, these differences are not crucial in explaining realized asset volatility, Moody's credit ratings, Altman's Z scores, or default occurrences. Within each test, some estimates outperform others, but no estimate is consistently best. We also show that, while the choice of using equity or debt prices to extract firm risk information appears to be inconsequential, the choice of model parameters is quite important. The manner in which we adjust yield spreads to account for embedded call options, and tax differences between corporate and Treasury securities as well as assumptions about the maturity of debt and debt priority structure have a significant effect on the level and rank ordering of firm risk measures.; Finally, we address the value of market information in the government oversight of U.S. bank holding companies. We construct risk measures obtained from equity prices alone, debt prices alone, and a combination of both. We observe that default risk measures constructed from debt prices generally outperform those constructed from equity prices in both contemporaneous and forecasting models. We further document that models using information from both equity and debt prices improves on the explanatory power of equity-only or debt-only models. Risk measures constructed from both equity and debt prices are more closely related to bank credit ratings, asset-portfolio quality indicators, and overall financial health. In addition, models using both equity and debt price information can better predict material changes in the firm's default probability, and quarter-to-quarter changes in the firm's asset-portfolio quality and overall condition. |
