Corporate governance is concerned with mechanisms by which capital suppliers exercise control over corporate insiders and management to protect their investment. Since the work of Jensen and Meckling (1976) and Smith and Warner (1979), the economics of the conflicts between bondholders and shareholders are well understood. These conflicts arise when the firm unexpectedly increases cash flow distributions as dividends or repurchases, when the issuer redistributes assets after debt issuance, or via debt refinancing at a similar or higher priority, among other sources.
For creditors, bond covenants provide the major form of protection from these conflicts. There are well over 40 different types of covenants used in practice, with some bond issues using just a few and others employing as many as a dozen. Popular examples of covenants include negative pledge covenants (which restrict issuing secured debt), restrictions of dividend payments, and asset sale clauses (which often require the issuer to redeem the bonds back upon the sale of certain assets).
In equilibrium, the benefit of adding a particular covenant should be offset by the cost of the covenant (see, e.g., Smith and Warner (1979)). Given the prevalence of covenants in the bond indenture and their theoretical cost-benefit trade off, it is surprising how few attempts there have been in the literature to empirically measure their impact on the dynamics of bond prices. Our paper contributes to this literature by presenting a detailed analysis of the effect of bond covenants on the pricing of bonds and measuring their ex post success at protecting bondholders. It also examines the effectiveness of bond covenants in mitigating the managerial incentives for risk-shifting (Jensen and Meckling (1976)).
In particular, we first address the following empirical questions: (1) Does credit spread dynamics reflect the impact of contractual protection? (2) Do covenants protect bondholders when bad states emerge? In the absence of information frictions, bond covenants are chosen to protect bondholders from the expropriation of shareholder. Since fewer covenants imply more potential for shareholders to expropriate bondholder wealth, bondholders require a higher yield to hold such bonds in equilibrium. Ceteris paribus, the bond yield should be decreasing in the degree of protection provided by the covenant structure of a bond issue.