The empirical tests for structural models of credit risk have been unsuccessful. Rather strict estimation or calibration reveals that the predicted credit spread is far below the observed ones (Jones et al., 1984), the structural variables explain very little of the credit spread variation (Huang and Huang, 2003), or the pricing error is very large for corporate bonds (Eom et al., 2004). More flexible regression analysis, although confirming the validity of the cross-sectional or long-run factors in predicting the bond spread, suggests that the explaining power of default risk factors for credit spread is still very small (Collin-Dufresne et al., 2001), the temporal changes of bond spread is not directly related to expected default loss (Elton et al., 2001), or the forecasting power of long-run volatility cannot be reconciled with the classical Merton (1974) model (Campbell and Taksler, 2003). These negative findings are robust to the extensions of stochastic interest rates (Longstaff and Schwartz, 1995), endogenously determined default boundaries (Leland, 1994; Leland and Toft, 1996), strategic defaults (Anderson et al., 1996; Mella-Barral and Perraudin, 1997), and mean-reverting leverage ratios (Collin-Dufresne and Goldstein, 2001).