eThesis - School of Business electronic theses

This thesis studies the effect of straight debt issuance announcements on credit default swap (CDS) spreads. As leverage-increasing capital structure decisions, debt issuances should result in increases of credit risk, and thus in CDS and credit spreads. Leverage is, after all, both theoretically and empirically an important determinant of credit risk.

However, previous literature fails to show a significant effect on credit spreads around debt issuances. When does price adjustment of credit risk occur if not around the announcement? In order to increase understanding on the issue, this study has two main objectives: to examine the debt issuance announcement effect on CDS spreads, and to identify potential debt and company specific determinants on these effects.

The data set used in this study consists of daily spread data of 182 CDSs of 83 companies, and 377 straight debt issuances of these companies. The time period covered spans from January 2008 to December 2013.

The announcement effects are sought after with event study methodology. The observed CDS spread changes are compared with predictions of a normal spread change model built on reduced form models of credit risk. Non-parametric statistical tests are used for assessing the significance of the results.

Additionally, to see whether announcement effects can be explained with debt- and company-specific factors, a linear regression model is built. The main factors studied are company sector and asset-based solvency, and relative issue size and maturity.

The analysis indicates that straight debt issuance announcements have a statistically significant positive effect on CDS spreads. This effect results in an increase of 5.5%, on average, over the selected event window, [-20, 20]. The announcement effects are significantly larger for subordinated CDSs than they are for senior CDSs. The effect occurs gradually over the event window, and suggests a lasting increase in CDSs. Debt issuances are clearly anticipated, but price adjustment in CDS markets seems slow and inefficient. The dynamics of CDS price adjustment are of great interest for future research.

The tested linear regression models performed poorly in explaining the announcement effects, as only a small fraction of the variation in these effects could be explained with the model. However, some implications can be extracted from estimated coefficients. Announcement effects increase with issue size for non-financial companies issuing short-term debt. Additionally, the effects decrease with asset-based solvency for financial companies.