The effect of liquidity and solvency risk on the inclusion of bond covenants

Studies have analyzed the impact of firm and issue characteristics but not liquidity and solvency components of financial distress on the use of bond covenants. Using a comprehensive database of corporate bonds from 2001 to 2012, we find that firm liquidity, measured by standardized Lambda, has a negative statistical and economic impact on the inclusion of all categories and sub-categories of restrictive bond covenants. Developed from financial statement information by Emery and Lyons (1991), Lambda is designed as a coverage ratio that, under certain distribution assumptions, maps into the probability of a firm being unable to pay its short-term bills. The strongest solvency proxy is the 10-year credit default swap (CDS) spread which is significant across the categories and sub-categories for investment and payment covenants, weakly significant for the subordinated debt sub-category of the subsequent financing covenant, but strongly significant for the control poison put sub-category of event covenants. This evidence supports a model that uses SLambda as a proxy for liquidity risk and the 10-year CDS spread as a proxy for solvency risk. The liquidity/covenant relationship is dampened when firms have access to commercial paper funding or bank loans. However, during the recent financial crisis liquidity event this liquidity/covenant relationship was enhanced especially for firms which were dependent on commercial paper during this time when the commercial paper market was deteriorating.