This dissertation consists of three essays linking the business models of rating agencies to the rating decisions these agencies make as market intermediaries between buyers and sellers.
The first study examines the link between a rating agency‟s primary revenue source and its rating decisions. Theoretically, rating payments could influence rating agency decisions or be counterbalanced by reputational rewards for rating accuracy. I explore this relationship in U.S. corporate credit ratings, where some agencies are primarily paid by bond issuers (sellers) and others by investors (buyers). Analysis of a balanced panel of 338 companies rated between 2005 and 2009 reveals that agencies produce differing ratings consistent with the preferences of their paying customers. Changes in buyer-paid ratings are more frequent and generally precede corresponding seller-paid rating changes. Seller-paid ratings are slower to incorporate negative information, particularly for rated firms in the financial services sector and firms with ratings above a critical grading cutoff.
The second study complements the first by estimating the gap between the rating information disclosed by sellers and the information sought by buyers, again using evidence from U.S. corporate credit ratings. While seller willingness to pay for an additional rating is highly concentrated among a subset of relatively high-quality firms, buyers demonstrate more uniform interest in additional ratings for firms at all quality levels. This finding highlights an information gap among high-risk firms that is not a major focus of existing regulation.
The third study focuses on rating decisions by government rating agencies, an alternative rating model to those examined in the first two studies. The empirical setting is Canadian film classification where the existence of multiple regional regulators has been justified by claims of variation in community standards. I find significant and increasing consistency in the regulatory decisions of these agencies, suggesting institutional isomorphism that brings into question the persistence of the parallel regional structure.
Overall, these studies provide new empirical insight into the relevance of rating agency heterogeneity to firm strategy and policy. The findings may also be relevant to a variety of other settings involving information disclosure such as environmental impact and corporate social responsibility.
| Identifer | oai:union.ndltd.org:TORONTO/oai:tspace.library.utoronto.ca:1807/31934 |
| Date | 11 January 2012 |
| Creators | Seaborn, Paul |
| Contributors | Silverman, Brian S. |
| Source Sets | University of Toronto |
| Language | en_ca |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0016 seconds