This paper explores the empirical relationship between corporate debt pricing and firm characteristics which influence strategic actions of borrowers and lenders. Our analysis of bond yield spreads is based on a large body of corporate finance literature that studies the effect of firm’s characteristics on the outcome of distressed restructuring. The key question we ask is, to what extent are firm-specific factors that influence strategic decisions related to default and ex-post reorganization reflected in ex-ante prices of corporate bonds.
We find that, on average, the threat of strategic default increases corporate debt spreads, even though ex-post there may be efficiency gains to renegotiation. This impact of strategic actions is economically significant for firms with little tangible assets, high equity holders’ bargaining power, and low credit rating. However, for the overall sample the average economic effect of strategic variables is well below transactions costs. Thus, although incorporating strategic behavior may improve cross-sectional performance of debt pricing models, it cannot remedy their inability to predict the general level of spreads.
The pricing of defaultable corporate debt has been the subject of extensive research over many years; yet market yield spreads remain largely unexplained. The conceptually appealing contingent claims models pioneered by Merton (1974) and Black and Cox (1976) were soon found to produce spreads far below empirically observed levels (Jones, Mason, and Rosenfeld, 1984). In such models, the value of the firm’s assets is assumed to follow a random process, and default happens when this value falls below a certain threshold.
While the basic framework has been extended to incorporate many realistic features of bond markets, such as corporate taxes and endogenous default (Leland, 1994), stochastic risk-free rate (Longstaff and Schwartz, 1995), stochastic default boundary (Santa-Clara and Sa´a-Requejo, 1999), and mean-reverting leverage (Collin-Dufresne and Goldstein, 2001), existing structural models are still unable to match the observed cross-section of bond spreads (Eom, Helwege and Huang, 2003).