Measuring the probability of financial covenant violation in private debt contracts☆

Abstract

We measure the probability that a borrower will violate financial covenants in private debt contracts. We analyze hand-coded data and specify standard covenant definitions using Compustat data that minimize measurement error for all individual Dealscan covenants. We use these definitions to create a measure of aggregate probability of violation, which can be used across all covenants in a loan or among covenant subsets of interest. We provide evidence that our aggregate probability measure is superior to alternatives used in prior literature.