Predicting spreads in the Nordic high yield bond market : a study of credit pricing in the years 2000-2012
MetadataShow full item record
- Master Thesis 
The main objective of this thesis is to identify and measure explanatory factors of observed credit spreads in the Nordic corporate high yield bond market in the period 2000 – 2012. From literature on credit pricing, we found three sources of risk compensation worth investigating; default risk, liquidity risk, and market risk. Our high yield sample consists of 323 bond issues, whereas 49 defaulted during the period. Our spread analysis is twofold. First, we utilize an extended structural credit risk model based on the classic model of Merton (1974) to estimate fair bond spreads based solely on the expected loss from defaults. Loss given default was attempted to be modeled separately, but no systematic relationship was identified, and a static estimate was used instead. Second, we attempted to explain the part of the observed credit spreads not explained by credit risk using a multivariate OLS-regression. This was done by instrumenting liquidity and market risk. Our main findings are that default risk can explain as much as 65 percent of the observed credit spreads on average. Furthermore, the credit model has significantly lower relative mispricing for bonds involved in a credit event, implying that structural characteristics are good predictors of credit risk. The part of the credit spread not explained by default risk was 178 basis points (bps) on average in absolute terms. Our attempt at explaining the variation in mispricing with liquidity and market risk was less conclusive, but liquidity proved to be significant with a premium of 110 bps for illiquid issuers.