In this paper we ask the empirical question are bond covenants priced? Consistent with the Costly Contracting Hypothesis (CCH) developed by Smith and Warner (1979), we find that they are. We document a negative relation between the promised yield on corporate debt issues and the presence of covenants. We also find that loans made to high-growth firms are more likely to include restrictive covenants than loans made to low-growth firms. We show that the inclusion of a covenant varies systematically with macroeconomic factors as well as with supply-side factors, especially the identity of the lending institution. Finally, we show that consistent with the CCH, firms that elect to issue private rather than public debt are smaller, have greater growth opportunities, less long term debt, fewer tangible assets, and include more covenants in their debt agreements. An important byproduct of our analysis is to demonstrate empirically that the decision to include a covenant and the corresponding promised yield are determined simultaneously. Consequently, statistical models that ignore this simultaneity in analyzing the effects of covenants, like single-equation probit models, are misspecified and generate unreliable statistics
