U.S. Securities & Exchange Commission
SEC Seal
Home | Previous Page
U.S. Securities and Exchange Commission

Speech by SEC Commissioner:
Opening Statement at SEC Open Meeting — NRSROs


Commissioner Elisse B. Walter

U.S. Securities and Exchange Commission

Washington, D.C.
September 17, 2009

I, too, would like to thank the staff for all your hard work on the NRSRO releases, and I strongly support the matters before us today. Your tireless efforts have created a thoughtful and well-balanced approach to improve disclosure of information regarding credit ratings, address conflicts of interest that arise in the credit rating business, and open up what I hope will be a meaningful and constructive dialogue on NRSRO accountability.

Before I say anything else, I would like to join Commissioner Casey in recognizing our Chairman's leadership on this issue. She has been calling for stronger regulation of credit rating agencies since the early 1990's, and I am confident that with her expert guidance, we will do everything we can to increase transparency and accountability at NRSROs and address concerns about the integrity of their credit rating procedures and methodologies.

As we consider the wide range of regulatory and policy issues in addressing NRSROs, I think it is critical that we once again take a look back and remember how the recent financial crisis started. It began with deteriorating mortgage-lending practices, particularly in the subprime area. There was an almost complete collapse of mortgage underwriting standards, which was most obvious in the notorious no-down payment loans and "no-doc" loans in which borrowers were not required to disclose income or assets, and even their employment was not verified. Through the process of securitization, underwriters packaged these risky loans into products that carried top credit ratings issued by the leading credit rating agencies. While advertised as a way to diversify and reduce risk, these financial instruments were often poorly understood and far riskier than their credit ratings suggested. Investors and our markets have paid a heavy price for this failure.

To address that failure and improve our oversight of credit rating agencies, we adopted several measures this past February that are designed to increase transparency and accountability with respect to NRSROs' credit rating procedures and rating methodologies. At the same time, the Commission published other proposals. As you know, we are here today to take action on those matters and to consider issuing further proposals.

Given the scope of the Commission's docket today, I will confine my remarks to just a few items.

Final Rule Amendments and Proposed Rule Amendments under the Credit Rating Agency Reform Act of 2006

I support adopting the final rule amendments and new rule proposals to impose additional disclosure and conflicts of interest requirements on NRSROs in order to address concerns about the integrity of the credit rating procedures and methodologies. In particular, I fully support the requirement to have a broader disclosure of credit ratings history information. This would expand our current requirement that certain NRSROs disclose 10% of their ratings history for issuer-paid ratings to also disclose 100% of all NRSRO credit ratings initially determined on or after June 26, 2007, regardless of the business model under which they are determined. The lag time has been adjusted in a bifurcated fashion, to 12 months for issuer-paid credit ratings and 24 months for subscriber-paid ratings, in order to respond to commenters' concerns regarding the potentially disproportionate negative effects on the subscriber-paid business model. While I would prefer a uniform 12-month delay, I still believe that this comprehensive disclosure of ratings histories for all outstanding credit ratings will allow for more granular ratings-by-ratings comparisons across all NRSROs. It will, in my current view, generate a sufficient volume of raw data that can be used to develop independent statistical analyses of the overall performance of an NRSRO's credit ratings in total and within classes and subclasses of credit ratings — especially for issuer-paid ratings, which apparently account for over 98% of the current credit ratings issued by NRSROs.

I'd also like to comment briefly on the proposal to impose additional requirements on NRSROs to, among other things, make publicly available a consolidated report containing information about revenues of the NRSRO attributable to persons paying the NRSRO for the issuance or maintenance of a credit rating. Detailed revenues and fees currently are not public information. A recent working paper by Professors Milbourn and Becker (of Washington University and Harvard Business School, respectively) commented that "[p]erhaps the largest potential cost to providing honest and accurate ratings is the potential forgone revenue from unhappy issuers," even in the absence of a single instance of a price premium being paid for a favorable rating, because "amounts of future business could quite possibly be related to current ratings."1 Yet, as the two authors state, "[t]his is a speculative argument, since detailed revenues and fees are not public information."2 This proposal will at least begin to close that informational gap. It would provide users of credit ratings with valuable information about the potential risk of undue influence, enabling users of credit ratings to more effectively evaluate the integrity of an NRSRO's credit ratings and determine for themselves whether the NRSRO is effectively managing its conflicts of interests.

References to Nationally Recognized Statistical Rating Organization Ratings in Commission Rules and Forms

I also am happy to support the recommendation to remove references to NRSRO ratings from certain rules and forms under the Exchange and Investment Company Acts. Those references in certain rules, such as Regulation ATS and its related forms, appear to have had limited utility, and their removal should contribute to the Commission's efforts to reduce undue reliance on those ratings by market participants.

I am hopeful that we will further this effort by obtaining additional information from the companion release that will allow for additional input on rules and forms under several statutes. I am not convinced that we should remove all references, and I am particularly interested in hearing whether the references in those provisions have led to undue reliance. For example, since Rules 101 and 102 of Regulation M are directed at distribution participants, issuers, and selling security holders, I have serious questions about whether they pose any danger of undue reliance on NRSRO ratings by investors. I am also interested in whether alternative subjective standards for credit risk and liquidity risk are effective replacements, for instance, in the Net Capital Rule, instead of the current bright-line test using NRSRO ratings. I look forward to reading the comments on these issues.

Credit Ratings and Rating Shopping Disclosure

Moving to the issuer disclosure proposals, I believe that our proposal to require credit ratings disclosure by registrants would help investors better understand what credit ratings mean and, in particular, highlight for investors the fact that similar ratings can have very different meanings, depending on what type of product the investor is evaluating. The proposal also would require additional disclosure to inform investors about potential conflicts of interest that could affect the credit rating. I believe that the disclosures required by this proposal and the proposed rules for NRSROs are complementary and together should arm the investor with critical information needed to help them judge for themselves how a conflict of interest impacts a particular rating.

To bring additional sunlight to the credit ratings process, we are also proposing disclosure of preliminary credit ratings in certain circumstances. Today, investors are completely in the dark as to whether a registrant engaged in rating shopping. Without a regulatory mandate that this information be provided, investors may not have access to information to help them assess the possibility of rating inflation due to ratings shopping.

Now, before I move to the concept release on Securities Act Rule 436(g), the final piece in our package of amendments, I'd like to take a few moments to remind everyone just how important a role MD&A plays in connection with the matters we are considering today. In case any of you have forgotten just how important MD&A is, let me remind you of the functions that MD&A should serve. It should "give the investor an opportunity to look at the company through the eyes of management by providing both a short and long-term analysis of the business of the company."3 And, as the Commission stated in 1981, "[i]t is the responsibility of management to identify and address those key variables and other qualitative and quantitative factors which are peculiar to and necessary for an understanding and evaluation of the individual company."4

I cannot stress enough how important I believe a robust discussion in the MD&A section is to an investor's evaluation of his or her investment.

Rule 436(g)

Finally, our concept release. I am extremely interested in the responses to the questions we have raised regarding the potential rescission of Rule 436(g). If you look at the issue of liability on its face, it is difficult to understand why NRSROs are treated differently than lawyers, auditors, and other experts. If this were a new issue, I would strongly advocate for an even playing field for NRSROs and other expert contributors to registration statements. And, as an aside, although this question is not before us today, given modern realities, I would also even the liability playing field between the primary offering and secondary trading markets.

With respect to credit rating agencies, in 1981, the Commission emphasized that the liability concession made to NRSROs rested in part on "the fact that rating organizations are already subject to substantial liability under the antifraud provisions of the federal securities laws so that they now must adhere to the highest professional standards in determining a security rating."5 Recent events lead me to question whether that predicate remains true today.

But, in terms of NRSRO liability, I recognize that we are not writing on a clean slate. We must examine very carefully the impact of subjecting those credit rating agencies to a stricter standard of liability, not only the impact on them, but also the impact on our markets.

Therefore, I believe that our issuance of this concept release is the right approach for us to move forward in our consideration of this issue. I look forward to hearing from credit rating agencies, investors, and all interested parties to better understand the pragmatic impact of our taking further action in this area.

Thank you.



Modified: 09/18/2009