This study examines whether and how the properties of corporate bond ratings change following Moody׳s and S&P׳s adoptions of the issuer-pay business model in the early 1970s. Regulators and debt market observers have criticized the issuer-pay model for creating an independence problem. However, the issuer-pay model allows for economic bonding between rating agencies and issuers through explicit contractual arrangements, which should improve the flow of nonpublic information. Using a difference-in-difference research design, I find that more optimistic ratings by issuer-pay rating agencies predict greater future profitability, differences between the ratings of issuer-pay and investor-pay rating agencies are associated with narrower secondary bond market bid-ask spreads, and that issuer-pay rating agencies become relatively more accurate and timely predictors of default compared to investor-pay agencies after the adoption of issuer-pay. These results reinterpret the recent findings of optimistic ratings by Jiang et al. (2012) as consistent with more informative bond ratings.

This paper examines whether firms that deviate from an empirically modeled (“expected”) credit rating engage in earnings management activities, as measured by abnormal accruals and real activities earnings management. We find evidence that firms use income-increasing (-decreasing) earnings management activities when they are below (above) their expected ratings. We then test whether such actions are successful in helping these firms move toward their expected credit ratings. The results suggest that firms below or above their expected credit ratings may be able to move toward expected ratings through the use of directional earnings management.

We investigate whether increased investor demand for financial information arising from higher market uncertainty leads to greater media coverage of earnings announcements. We also investigate whether greater coverage during times of higher uncertainty further destabilizes financial markets because of greater attention-based trading or, alternatively, improves trading and pricing by lowering investor acquisition and interpretation costs. When uncertainty is higher, we find evidence of greater media coverage of earnings announcements and that the greater coverage leads to improvements in investor informedness, information asymmetry, and intraperiod price timeliness, and greater trade by both retail and institutional investors. In contrast to the media serving an expanded role in improving capital markets during more uncertain times, we fail to find that changes in firm-initiated disclosures lead to similar improvements and find that less frequent analyst forecast revisions exacerbate problems in capital markets during earnings announcements.

We provide the first evidence of systematic bias among an emerging type of credit rating agency that relies on subscriptions from institutional clients as its primary source of revenue. Using data from Egan-Jones Ratings Company (EJR), a representative subscriber-paid rating agency, we show that EJR issues more optimistically biased credit ratings, less timely downgrades, and less accurate ratings for firms held by more EJR clients. Our evidence is consistent with EJR optimistically biasing its ratings to bolster subscriber revenue, which allows institutional clients to invest in riskier bonds with higher expected returns. Taken together, our findings suggest that the emergence of subscriber-paid rating agencies as an alternative to more traditional issuer-paid agencies is unlikely to resolve problems arising from conflicts of interest but rather alter the nature of these conflicts in the ratings process.

We propose a new measure of readability, the Bog Index, which captures the plain English attributes of disclosure (e.g., active voice, fewer hidden verbs, etc.). We validate this measure using a series of controlled experiments and an archival-based regulatory intervention to prospectus filing readability. We also demonstrate the importance of understanding the underlying drivers of quantity-based measures of readability. In particular, we caution researchers that a vast amount of the variation in Form 10-K file size over time is driven by the inclusion of content unrelated to the underlying text in the 10-K (e.g., HTML, XML, PDFs).