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

Speech by SEC Commissioner:
Remarks at the Commission Open Meeting


Commissioner Kathleen L. Casey

U.S. Securities and Exchange Commission

Washington, D.C.
December 3, 2008

I would like to join Chairman Cox in commending the Commission staff for its outstanding work on these important releases. Erik Sirri, Dan Gallagher, Bob Colby, Mike Macchiaroli, Tom McGowan, Randall Roy, Joseph Levinson, Carrie O’Brien, Sheila Swartz, and Rose Russo Wells of Trading and Markets all did a wonderful job under considerable pressure. Randall, the indispensable man on the two releases before us this morning, deserves special recognition for his heroic efforts. I also want to thank GC and OEA for their contributions to the releases today.

The Big Three credit rating agencies have failed investors. That is why we are here today. The largest rating agencies awarded their highest ratings to complex debt instruments that were undeserving of investment grade status. And investors and the markets have paid a heavy price. The rating agencies certainly were not the only cause of the current crisis, but they played a role, and it was not insignificant.

This is not the first time that rating agencies have failed in their role, of course. Recall the absence of any timely downgrades of Enron, WorldCom, Parmalat, and many other companies earlier this decade. And I suspect the chairman well recalls the Orange County debacle from the mid-1990s.

But an important difference now is that the Commission has a statute to administer. After nearly a century of self-regulation, the rating agencies became subject to SEC oversight after the enactment of the Credit Rating Agency Reform Act of 2006 (“Rating Agency Act” or “Act”). The rules implementing the Act did not become effective until June 2007. These rules further the Act’s interrelated goals of enhancing accountability, transparency, and competition in the rating industry.

The Commission has actively used its authority from the very beginning. Starting in August 2007, the Commission staff conducted a comprehensive, year-long inspection and examination of the Big Three rating agencies and produced a public report detailing considerable deficiencies. That inspection, in part, informs the releases we are considering today.

One of the defining characteristics of the rating industry has been the virtual absence of market-based competition. Longstanding, deeply entrenched incumbents have not retained market share by virtue of issuing ratings of a superior quality. They simply did not have to.

There are now ten Nationally Recognized Statistical Rating Organizations (“NRSROs”). Just a few years ago, there were only three NRSROs. That is the good news. The bad news is that the Big Three continue to overwhelmingly dominate what has to be one of the most concentrated private industries in the world. One firm issues almost half of the outstanding credit ratings.1 The Big Three issue 98% of total ratings2 and collect more than 90% of industry revenues.3

The Commission needs to be mindful of these facts when it acts in this area. One way for the new entrants in the rating industry to distinguish themselves from the largest firms is by issuing ratings that prove to be more credible and reliable. But we need to create an environment, as envisioned and required by Congress, that is unapologetically pro-competitive, because until there is a market penalty for being wrong, the industry will not serve investors as effectively as it would otherwise.

Today’s releases are designed to increase transparency and accountability, reduce conflicts of interest, and enhance competition. And, as I noted earlier, the staff has done a phenomenal job of crafting a package of reforms that carefully balances difficult legal, practical, and technical issues as well as important cost/benefit interests.

Let me briefly discuss the reforms that I believe are the most significant.

But before I do that, I want to emphasize how important I believe it is that we act in the near future on the other two releases published for public comment earlier this year. The first would enhance disclosure to investors of the different risk characteristics of structured financial products and traditional debt products.

The second would address the oligopoly in the rating industry and the overreliance on NRSRO ratings by removing the regulatory requirements embedded in SEC rules.

NRSRO References in SEC Rules

It is imperative that we remove the regulatory requirements that have served to elevate NRSRO ratings to a status that does not reflect the actual purpose and limitations of credit ratings. This evolution in the use of ratings has been recognized by numerous commentators and market participants, including the rating agencies, who have emphasized what a credit rating is intended, and not intended, to represent.

To remove the NRSRO references from SEC rules will require a reassessment of our longstanding uses of NRSRO ratings. But, in my view, doing so is absolutely essential to the Commission’s efforts to faithfully implement the clear congressional intent of enhancing transparency, accountability, and competition in the rating industry.

The Commission has referred to NRSRO ratings for over three decades. Although it is quite understandable why they were first incorporated into the Exchange Act’s Net Capital Rule and in subsequent rules here and elsewhere, it has become evident over time that there are considerable unintended consequences to the regulatory use of credit ratings. It was never intended to establish and preserve a valuable franchise for the large rating agencies, while simultaneously inoculating them from market competition. Nor was it intended to serve as a substitute for adequate due diligence on the part of investors, managers, directors, and others. Unfortunately, as recent events demonstrate, it appears to have led to such results in too many cases.

Removing NRSRO references from our rules is particularly significant, and any reforms would not be complete without doing so.

Ratings Histories

One of the key reforms we are adopting today will require each NRSRO to make and retain a record for each outstanding credit rating it maintains showing all rating actions (initial rating, upgrades, downgrades, placements on watch for upgrade or downgrade, and withdrawals) and the date of such actions.

In addition, NRSROs must make publicly available, on a six-month delayed basis, a random sample of 10% of the ratings histories for each class of credit rating for which it is registered and has issued 500 or more issuer-paid ratings. This will allow market participants, academics, and others to use the information to conduct analyses comparing NRSRO performance. And requiring the data to be made available in XBRL format ensures that it will be analyzed with greater speed and accuracy, and at less cost.

Importantly, this will greatly facilitate third-party development of ratings performance measurement statistics, and create an environment that will allow those NRSROs issuing more accurate ratings to gain market share. In other words, it will help the rating industry function like any other industry.

The Commission also is proposing that NRSROs make publicly available in an XBRL format, and on a 12-month delayed basis, ratings history information for 100% of their current issuer-paid ratings. This is a significant, pro-competitive reform proposal that would further the objectives of the Act. In my view, it is vital that 100% of the credit ratings are disclosed to investors and market participants. I am pleased we proposed that approach. As the proposing release from June indicated, this will permit maximum comparability of ratings on an obligor-by-obligor or instrument-by-instrument basis. Again, investors should know who is issuing the most useful ratings, and without comparability such determinations are not feasible.

In my view, the Commission properly decided to limit the application of the adopting and proposing rules to only issuer-paid NRSROs. Congress has identified the issuer-paid model as an inherent conflict of interest and specifically encouraged the proliferation of competing business models that are less conflicted.

While I do not believe that the subscriber-based model is necessarily free from the conflicts that may affect issuer-paid ratings, I am more concerned with killing competition in the cradle and sustaining a ratings franchise that has failed investors and our markets.

Making subscriber-paid firms release their ratings for free could very likely cripple these firms as they begin to grow, and equally important, it would almost certainly discourage new entrants into the ratings business. That would be antithetical to the primary purpose of the Rating Agency Act, which is to increase competition in an oligopolistic industry. In any event, we are asking a long list of questions in this area, and I look forward to the comments.

Ratings Performance Measurement Statistics/Enhanced Disclosure of Ratings Methodologies

A related reform in today’s adopting release is the refinement in Commission rules implementing the Act’s requirement that performance measurement statistics be disclosed over short-, mid-, and long-term periods. Specifically, we are requiring disclosure of separate sets of default and transition statistics for different asset classes, displayed over 1-, 3-, and 10-year periods. Also, the statistics now will include upgrades and defaults relative to the initial rating. The purpose of these changes is to make it easier to compare the accuracy of ratings on a class-by-class basis.

Another important reform in the adopting release would require enhanced disclosure of ratings methodologies. NRSROs would have to disclose whether and, if so, how asset verification and qualitative assessments of originators are relied upon in determining structured finance ratings. In addition, NRSROs would need to disclose, for all asset classes, the frequency of its surveillance efforts and how changes to its quantitative and qualitative ratings models are incorporated into the surveillance process.

Re-proposal of Information Disclosure Program

Perhaps the most significant and far-reaching reform before us today would require the disclosure of the information an NRSRO uses to issue a structured finance rating. We are re-proposing this pro-competitive amendment with substantial modifications. In my view, these changes are necessary and advisable, and I commend the Trading and Markets staff for its thoughtful and innovative work on this re-proposal. Although the concept of disclosing the information underlying ratings is not complex, drafting a rule in this context that is workable in practice is another story, as the staff can surely attest.

The re-proposal makes three significant changes. It specifies that the arrangers (i.e., issuer or underwriter), rather than the NRSROs, would be responsible for disclosing the information provided to the hired NRSRO to determine the credit rating. It limits the disclosure to other NRSROs, thus avoiding any Securities Act implications. And it modifies Regulation FD (“Reg FD” or “FD”) because the arranger will be disclosing material non-public information and would need to rely on the exclusions to FD in order to disclose it to NRSROs without simultaneously making a public disclosure of this information.

In order to allow subscriber-paid NRSROs to take advantage of this new information disclosure program, Reg FD must be amended to permit the disclosure of material non-public information to NRSROs irrespective of whether they make their ratings publicly available.

It would also accommodate any NRSROs that access the information under the proposed program but ultimately do not issue a credit rating using this information. In addition, Reg FD’s definition of “credit rating agency” would be replaced by the statutory definition of the term. These are all constructive amendments, and I again applaud the staff for its outstanding work.

The re-proposal would permit competitive analysis by other NRSROs that are not paid by the issuer to rate the product. They could issue unsolicited ratings, and the marketplace could decide whose performance is superior. This is exactly the type of pro-competitive reform contemplated by the Rating Agency Act. I look forward to reviewing the comments. I do not expect the firms with a stranglehold on the industry to applaud this reform, but that should not surprise anyone.

I have several questions.




Modified: 12/11/2008