Information Risk and the Cost of Debt Capital

by. Sattar A. Mansi , William F. Maxwell , and Darius P. Miller

Abstract
We test whether forecast dispersion is related to the uncertainty of economic fundamentals by exploiting the unique attributes of the corporate bond market. We find strong evidence that forecast dispersion is positively priced in corporate bond yields. We also provide evidence that other firm specific information measures, such as the level of idiosyncratic risk, analyst following, forecast bias, ownership structure, probability of informed trading (PIN), executive compensation, age, size, and accounting disclosures impact the spread on corporate bonds, but do not subsume the effect of dispersion. Our results are consistent with the hypothesis that the dispersion of analysts’ forecasts represents a forward looking measure of information risk that is priced in financial markets.

Conclusion
In this paper, we examine the impact of forecast dispersion on asset prices to a different claimant, bondholders. We rely on Merton’s (1974) framework which acknowledges that corporate bondholders are short a put option on the firm’s assets. Therefore, if analyst disagreement about future earnings represents a measure of uncertainty about firm value, it should be unambiguous associated with a lower bond price because of its connection with the volatility of the underlying asset. Focusing on corporate bonds provides several potential advantages for examining dispersion and information risk. For example, the bond market has limited short selling and is primarily composed of institutional investors. In addition, bond yields, as a measure of internal rates of return, contain the ex-ante expected bond return. Further, bond pricing models are relatively well-specified.

After controlling for idiosyncratic risk, we find strong evidence that forecast dispersion is
negatively related to credit ratings and positively priced in corporate bond yields. In so doing, we also provide evidence that other firm specific information measures, such as level of analyst
following, ownership structure, probability of informed trading (PIN), executive compensation, age, size, and accounting disclosures also impact the spread on corporate bonds. These results suggest the dispersion of analysts’ forecasts is a forward looking measure of information risk that is priced in financial markets.