The 2008 financial crisis spurred an examination of credit ratings and rating agencies. A key dimension of the crisis, though probably not the central cause, has been the mis-assessment by credit rating agencies of the riskiness of various securities, such as tranches of mortgage-backed instruments. The use of ratings has been hard-wired into our regulatory system for a variety of purposes, including determining the capital adequacy of financial institutions and assessing whether specific securities are suitable choices for investors and asset managers who are responsible for certain investment portfolios. When credit ratings began to appear unreliable, investors throughout the economy changed their assessment of the risks (and values) of various assets. Substantial revisions in the levels and perceived reliability of assessments by rating agencies (and investors as a whole) led to dramatic re-evaluations of large categories of assets and a considerable increase in systemic risk. One of the responses to the problems with rating agencies was the provisions in the Dodd-Frank Act that directed the Securities and Exchange Commission (SEC) and banking regulators to reduce reliance on ratings for a broad array of regulatory objectives.
The Dodd-Frank restriction on regulators’ use of ratings means that this source of systemic risk will not be hard-wired into our regulatory system, but significant systemic risk could still remain. Despite reduced reliance upon ratings for regulatory purposes, many investors may continue to utilize ratings as a major input to their evaluation of assets. Consequently, dramatic revaluation by the rating agencies still could send shock waves through the economy.
Even if there were no rating agencies, in the event that most institutional investors adopted a common approach, there would be considerable adverse systemic difficulties if that approach proved incorrect. To the extent that substantial economies to scale exist in information production, it is not surprising that a small number of financial intermediaries would play a central role in the financial sector. But if only a few intermediaries have a central role, systemic risk from widespread similarities in investor approaches could be hard to avoid. The systemic nature of the risk reflects not only the role of ratings in regulation, but also that a common point of view often dominates fundamental risk assessment. At its most basic level, a lack of diversity of opinions can lead to systemic risk.
The Dodd-Frank Act not only instructs regulators to substantially reduce reliance on credit rating agencies, it also calls for regulators to supervise those agencies more tightly. The Act created the SEC’s “Office of Credit Ratings,” which administers a variety of rules that affect Nationally Recognized Statistical Rating Organizations (NRSROs), which include the major credit rating agencies. The Act requires that office to perform annual examinations of each NRSRO.
These two concepts represent very different approaches to the regulation of the agencies. The value of tight supervision of the rating agencies and relatively uniform standards would be greater if the ratings were to continue to be used for regulatory purposes. There is arguably less need and value for the regulator to supervise the rating agencies, or for ratings to have a common meaning across agencies, if the ratings are not going to be used for regulation. Reducing reliance on ratings for regulation is consistent with treating the rating agencies simply as private firms. If credit rating agencies have no role in regulation, then they can be allowed greater flexibility to establish their own norms, so they can compete freely with each other.
The costs of the financial crisis were so dramatic that the desire to use multiple regulatory approaches to important issues is understandable. Nonetheless, the combination of the two different approaches of the Dodd-Frank Act towards credit rating agencies appears paradoxical. Different approaches to regulation may be complements or substitutes. Complementary approaches would strengthen each other, but it does not appear that these two approaches are complementary. Instead, these approaches are substitutes, because the value of tighter supervision is greater when regulators are relying on rating for regulatory objectives than when reliance is reduced. The two approaches offer alternative methods of reaching the same goal, reducing risk stemming from systemic credit rating errors. The total costs of using two alternative methods of regulation are often the sum of their individual costs, but the total benefits may be far less than the sum of their individual benefits, because the benefits of the two approaches overlap substantially.