24 Feb 2011  Working Papers

Issuer Quality and Corporate Bond Returns

Executive Summary — In research that could help regulators and policymakers tell if credit markets are becoming overheated, HBS professor Robin Greenwood and doctoral candidate Samuel G. Hanson suggest that measures of credit quality are just as important to monitor as the more traditional reviews of credit quantity. They also find that time-varying investor beliefs such as over-optimism, or tastes such as a heightened tolerance for risk, can contribute to fluctuations in credit quantity. Key concepts include:

  • The researchers use measures of the quality of corporate debt issuers to forecast excess returns on corporate bonds. When issuer quality is low, corporate bonds subsequently underperform Treasuries. The work uncovers a striking degree of predictability and often forecasts significantly negative excess returns.
  • The 2004-2007 credit boom and collapse was not unique, but rather part of a recurring historical pattern in which investors grant cheap credit to low quality borrowers during credit booms, and experience low returns when those borrowers ultimately default or spreads widen.
  • The research results provide guidance on how we can tell if credit markets are becoming overheated, and suggests that looking at credit quantities alone may not be enough-policymakers might also want to consider the credit quality of debt market financing.

 

Author Abstract

Changes in the pricing of credit risk disproportionately affect the debt financing costs faced by low credit quality firms. As a result, time-series variation in the average quality of debt issuers may be useful for forecasting excess corporate bond returns. We show that when issuance comes disproportionately from lower quality borrowers, future excess returns on high yield and investment grade bonds are low and often significantly negative. The degree of predictability is large in both economic and statistical terms, with univariate R2 statistics as high as 30% at a 3-year horizon. The results are difficult to reconcile with integrated-markets models in which the rationally determined price of risk fluctuates in a countercyclical fashion. The results can be partially explained by models in which shocks to intermediary capital or agency problems drive variation in required returns. Finally, we consider models in which investor over-extrapolation plays a role and find some evidence in favor of these models.

Paper Information