Structural Models of Corporate Bond Pricing: An Empirical Analysis

Abstract

This paper empirically tests five structural models of corporate bond pricing: Those of Merton (1974), Geske (1977), Leland and Toft (1996), Longstaff and Schwartz (1995), and Collin-Dufresne and Goldstein (2001). We implement the models using a sample of 182 bond prices from firms with simple capital structures during the period 1986-1997. The conventional wisdom is that structural models do not generate spreads as high as those seen in the bond market, and true to expectations we find that the predicted spreads in our implementation of the Merton model are too low. The compound option approach of Geske comes much closer to the spreads observed in the market, on average, but still underpredicts spreads. In contrast, the Leland and Toft model substantially overestimates credit risk on most bonds, and especially so for high coupon bonds. The Longstaff and Schwartz model modifies Merton to incorporate a stochastic interest rate and a correlation between interest rates and firm value. While the correlation and the level of interest rates have little effect, higher interest rate volatility leads to higher predicted spreads. However, this and other features of this model result in spreads that are often too high for risky bonds and too low for safe bonds. The target leverage ratio model of Collin-Dufresne and Goldstein helps to raise the spreads on the bonds that were considered very safe by the Longstaff and Schwartz model, but overall tends toward overestimation of credit risk. We conclude that structural models do not systematically underpredict spreads, as the previous literature implies, but accuracy is a problem. Moreover, some of the simplifications made to date lead to overestimation of credit risk on the riskier bonds while scarcely affecting the spreads of the safest bonds.