In this paper we focus on the credit spread component of corporate bond returns, enabling us to find new predictors that were previously unknown to the literature. Moreover, by re-examining previously documented predictors, we are able to dismiss several of them as irrelevant for credit spread returns and to explain inconsistent findings between investment grade and high yield corporate bonds. In total, we find four factors with significant in-sample and out-of-sample predictability of both investment grade and high yield excess returns over Treasury. Two variables come from the existing literature: past equity return and past corporate bond return. Evidence for the other two variables is new: change in implied equity volatility and the Halloween indicator.
Read the research paper at SSRN