 |
In this paper, I analyze in more detail credit risk premia embedded in sovereign CDS spreads. In particular, I explicitly take into account ``default event risk premia'', which are risk premia related to the timing of default events. These premia have been investigated in the corporate credit risk literature, but did not receive much attention in the sovereign credit risk literature.
I propose a novel model for the term-structure of sovereign credit risk in which sovereign defaults can be triggered by shocks in either a common or country-specific factor. Both factors are modeled as self-exciting processes, allowing the model to capture apparent features in the data such as the high degree of commonality of sovereign credit risk and the clustering of credit shocks over time and across countries.
The model allows for a natural decomposition of CDS spreads in two dimensions: First, I can decompose CDS spreads in country-specific and systemic risk components. I find a similar decomposition across rating classes in which approximately 65% of CDS spreads can be attributed to country-specific risk and 35% of CDS spreads can be attributed to systemic risk. Second, I can decompose CDS spreads into risk premia and a default risk component. I find that the default event risk premium is heavily priced in CDS spreads and is more important for lower credit ratings. For example, the default event risk premium accounts on average for 22% of (5-year) CDS spreads of A-rated countries and up to 52% for B-rated countries. In the term-structure dimension, I find that default event risk is more important for shorter maturities.
