Pricing default risk : the good, the bad, and the anomaly

While empirical literature has documented a negative relation between default risk and stock returns, the theory suggests that default risk should be positively priced. We provide an explanation for this "default anomaly", by calculating monthly probabilities of default (PDs) for a large sample of firms and decomposing them into systematic and idiosyncratic components. The systematic part, measured as the PD sensitivity to aggregate default risk, is positively related to stock returns. Our results show that riskier stocks underperform because they have on average lower exposures to aggregate default risk.