Measuring credit spreads: Evidence from Australian Eurobonds

Recent theoretical models including the closed-form valuation model of Longstaff and Schwartz (1995) predict that credit spreads are driven by both an asset and interest rate factor. In empirical studies the credit spread may be expressed as either the difference between, or ratio of, the risky bond to a riskless bond. Using a daily sample of non-callable Australian dollar denominated Eurobonds it is found, consistent with theory, that changes in credit spreads are negatively related to both changes in the return on All Ordinaries stock Index and changes in the Government bond yield. Interestingly, the ratio measure - termed a relative credit spread - tends to be statistically more significant than the alternate measure based upon the difference - termed an actual credit spread. However, it is shown that this result is spurious and due to the way in which relative credit spreads are constructed. Noting Duffee's (1998) warning against using callable bonds, the use of only non-callable Eurobonds provides a cleaner result when compared with tests conducted by Longstaff and Schwartz (1995).