While my commentary last week dealt with corporate governance rating and executive pay issues, the paper Charles Calomiris and I released on Tuesday applies conflicts of interest for all types of ratings, including credit ratings. The framework we lay out revolves simply around thinking of not two, but three parties associated with ratings: the rating agency, the institutional investor intermediary, and the ultimate investor. When characterized this way, it is the principal-agent problem between the institutional investor intermediary and the ultimate investor – not that between the rating agency and the institutional investor intermediary – that is important. This distinction has important implications for policy.
Excerpting from our policy recommendations (see that attached paper for more details), proposed credit rating agency reforms that would try to empower institutional investors more (by having them pay rating agency fees, or by having them participate in rating agencies’ corporate governance or modeling) would be counterproductive and unlikely to eliminate ratings inflation or improve ratings quality. Increasing the number of NRSROs is laudable, but if the conflict between investors and clients is the problem, increased competition will have no effect. Furthermore, proposals to eliminate NRSRO status entirely are problematic. First, this is unrealistic as a short-term reform, given the extreme dependence of regulation (e.g., the Basel II standards) on NRSRO credit ratings). Furthermore, the elimination of NRSRO status in and of itself likely would not solve the problem of low-quality and inflated credit ratings. The apparent entrenchment of corporate governance ratings shows that eliminating the use of credit ratings for regulatory purposes would not overcome the problem of low-quality ratings, although it likely would remove much of the incentive for ratings inflation.
Ratings reform of all types must make it profitable for rating agencies to issue high-quality, non-inflated ratings, notwithstanding the demand for low-quality or inflated ratings by institutional investors. This can only be accomplished through the following two regulatory interventions: (1) objectification of the meaning of ratings, and (2) linking the fees earned (or penalties paid) by rating agencies to objective measures of their performance.
The objectification of credit ratings could be achieved by requiring all NRSROs to formally and transparently link letter grades to specific numerical estimates of the probability of default and the expected loss given default. Once they have done so, then regulators can specify regulatory limits and capital requirements that are linked to estimated probabilities of default and losses given default (which have concrete meaning), rather than vaguely defined letter grades. Then, for example, if an NRSRO’s ratings for a particular product (say, CDOs) were found to be persistently inflated over a sufficiently long period of time then that NRSRO would face a penalty, like a “clawback” of the fees the agency has already earned on that product or losing its NRSRO status for a brief period of time.
Corporate Governance Ratings
In the case of corporate governance ratings, the pernicious demand for low-quality ratings does not lead to ratings inflation, but rather to noisy signals that avoid identifying either strong or weak companies. Thus, the penalty function to properly incentivize corporate governance rating agencies would focus on penalizing ratings inaccuracy, not just ratings inflation.
While, in theory, the idea of objectifying and penalizing low-quality corporate governance ratings has appeal, in practice, there are significant obstacles to developing a means of doing so. First, the academic literature on corporate governance is far from achieving consensus on a feasible approach for measuring accuracy. Thus, a penalty structure that would reward good ratings and punish bad ones seems not to be immediately feasible. Unlike credit ratings (which measure the objective fact of default and loss), corporate governance translates into long-term performance, which is difficult to measure. Furthermore, the literature on predictors of default has been an active area of research for decades, while corporate governance quality is a relatively new field.
Second, since corporate governance rating agencies are not acting as NRSROs and have no formal regulatory function in the financial system, the corporate governance rating industry is not obliged to meet any degree of accuracy under current law. Thus, before we could consider penalizing corporate governance rating agencies we would have to license them.
In our view, to the extent that the Securities and Exchange Commission relies increasingly on corporate governance ratings as a means by which institutional investors can meet their fiduciary duties to vote shares in a manner reflecting the best interests of the ultimate investors (the individual claimants behind the pension and mutual funds, as well as insurance companies and banks), the corporate governance rating agencies that sell information to meet those fiduciary duties should have a regulatory obligation that requires them to maintain some degree of accuracy. Hence, to the extent that the industry’s role as a corporate governance regulator becomes formalized, appropriate penalty structures – such as the objectively applied clawback provisions envisioned above – should be developed. If in the fullness of time the industry cannot demonstrate a convincing value relationship between corporate governance and firm performance, the ultimate penalty function – preventing meaningless corporate governance rating-based proxy advice from being used to meet institutional investor fiduciary duties – may have to be imposed.
It cannot be denied that corporate governance rating agencies play a significant role in proxy recommendations. Those powers give corporate governance rating agencies significant ability to extract rents from the firms that they rate, and currently governance rating agencies are not subject to any disclosure requirements or business practice standards regarding these conflicts of interest. Thus, under current laws and regulations, abuses of power by corporate governance rating agencies due to conflicts of interest are not observed, defined, or prevented. While such holdup games exerted against large companies are well-known (like ACORN’s infamous actions to extract money from merging banks in exchange for agreeing not to oppose their mergers) historically these sorts of activities have largely escaped regulatory or legal interventions and when intervention has been attempted (for instance, the sunshine provisions for community activist organizations proposed under Gramm-Leach-Bliley), entrenched institutions (such as ACORN) fight hard to suppress reform.
At a minimum, therefore, the SEC should immediately require disclosure of all potential conflicts, and should develop standards that would prohibit egregious conflicts of interest between rating agencies and the firms that they rate. Furthermore, in the interest of mitigating rent seeking behavior, institutional investors should have to disclose all of their points of contact or alliance with the rating agencies, and these conflicts of interest should also be the subject of SEC standards.
In summary, real meaningful policy alternatives to the seeming conundrums of ratings industry conflicts of interest do exist. Those alternatives, however, require seeing properly what gives rise to the conflicts of interest among all the industry participants, including ratings agencies, institutional investor intermediaries, and the ultimate investors, and structuring incentive compatible industry arrangements to lead to the desired policy objectives.
† Hermann Moyse, Jr./Louisiana Bankers Association Professor of Finance, Louisiana State University, Senior Fellow at the Wharton School, and Partner, Empiris LLC. Contact information: email@example.com; (202) 683-8909 office. Copyright Joseph R. Mason, 2009. All rights reserved. Past commentaries and testimony are blogged on http://www.rgemonitor.com/financemarkets-monitor/bio/626/joseph_mason.