Does the Tail Wag the Dog?: The Effect of Credit Default Swaps on Credit Risk (Digest Summary)

Credit default swaps (CDSs) are insurance-type contracts that buyers can use as protection against default by a debtor. For example, creditors can buy CDS protection to hedge the credit risk of a borrower. On the one hand, the hedge may increase the supply of credit to borrowers who may, in turn, undertake more risky projects. On the other hand, increased credit availability may increase borrowers’ financial flexibility and resilience in financial distress. Also, a creditor who hedges using a CDS may be tougher during credit negotiations with a borrower, increasing the probability of bankruptcy because the creditor’s losses from potential default are reduced. Overall, the availability of CDS trading may affect not only the relationship between the creditor and borrower but also the borrowers’ credit risk. The authors examine whether a firm’s credit risk increases after CDS trading in its name commences.

The authors demonstrate that the credit risk of a reference firm increases significantly after the initiation of CDS trading. They arrive at the following conclusions: (1) distressed firms that have an active CDS market are more likely to file for bankruptcy compared with firms that have no CDS market; (2) firms with “no-restructuring” CDS contracts are more likely to file for bankruptcy compared with firms with CDS contracts that include restructuring; and (3) the number of creditors increases after CDS trading begins, which exacerbates creditor coordination failure in the resolution of financial distress.

Although the authors’ findings show that firms are more vulnerable to bankruptcy once CDSs are traded on them, this finding does not imply that CDS trading reduces social welfare because the benefits of higher firm leverage (resulting from the ability of creditors to hedge risk) may outweigh the greater bankruptcy costs.

The authors obtain North American CDS transaction data, including contract specifications, price, volume, and settlement terms, from two sources: CreditTrade and GFI Group. The sample period is from June 1997 to April 2009 and includes 901 US firms with CDS trading over the period.

In their baseline analysis, the authors use information about the first day of CDS trading and compare changes in firm default risk at the onset of CDS trading. Bankruptcy data from New Generation Research are linked to the CDS sample to identify bankrupt firms that had CDS trading prior to their bankruptcy filings.

The authors conduct an event study to determine the effects of the introduction of CDS trading on credit ratings as well as investigate the channels and mechanisms through which CDS trading affects credit risk.

At the firm level, the downgrade frequency is 2.6% higher for CDS firms than for non-CDS firms, after the authors adjust for size and industry type. The likelihood of bankruptcy in the overall sample is 0.14%, but the marginal effect of CDS trading on the likelihood of bankruptcy is 0.33%.

The authors use instrumental variable techniques to control for the possibility that firms selected for CDS trading are firms whose credit quality was expected to deteriorate. Their findings remain robust after they control for this possible simultaneity bias.

CDS contracts are used as a way to trade and/or hedge the credit risk of a reference firm. But as the authors show, the risk of bankruptcy for firms that are linked to CDS trading is more than twice that of firms that are not linked. This result heightens the credit risk that thus far has probably not been in the mind of hedgers, potentially leading to a situation in which market participants are underhedging their exposure to borrower default.