Contractual Complexity in Debt Agreements : The Case of EBITDA

The definition of EBITDA is among the most important parts of a credit agreement. This concept matters to borrowers and creditors because it frequently determines whether a borrower is in breach of its covenants in the loan, and it matters to regulators because it determines the amount of leverage a loan entails. While credit analysts and debt lawyers have commented on the differences in the definition of EBITDA, existing research on debt agreements has almost entirely ignored this variation and the consequences it has for understanding how debt agreements operate. We use supervised learning of income definitions in thousands of credit agreements to show that there is, indeed, massive variation in the definition of EBITDA and that a substantial proportion of these agreements inflate EBITDA by adding back income. In further analysis we show that expansive EBITDA definitions are more common among private equity borrowers and that banks appear to have allowed more permissive EBITDA definitions for non-private equity borrowers in the wake of the Federal Reserve's restrictive guidance on lending leverage. We show that there is a negative relationship between the permissiveness of EBITDA definitions and the amount of covenant slack in loans and that more permissive definitions are associated with higher loan spreads. Finally, we demonstrate that the best predictors of the content of income definitions are the past credit agreements of the borrower