Speech by SEC Commissioner:
Statement at Open Meeting on Nationally Recognized Statistical Rating Organizations
Commissioner Troy A. Paredes
U.S. Securities and Exchange Commission
December 3, 2008
Thank you, Mr. Chairman.
Congress enacted the Credit Rating Agency Reform Act of 2006 (“Rating Agency Act”) with an eye towards improving the quality of credit ratings. Specifically, its purpose is to “improve ratings quality for the protection of investors and in the public interest by fostering accountability, transparency, and competition in the credit rating industry.”
The rules we consider adopting today take a meaningful step forward in advancing these goals, and I am pleased to support them.
These rules will enhance transparency, thus promoting accountability, through additional disclosures of useful information and will address conflicts of interest with the potential to influence NRSROs’ credit ratings. In short, these rules help ensure the integrity of credit ratings and foster greater confidence in them.
Rarely, if ever, are the benefits of a rule change achieved without some attendant cost and the need to make tradeoffs. By way of illustration, one rule for adoption will require an NRSRO to document its rationale for materially deviating from the rating a quantitative model implies. I support this rule, but am mindful that credit rating agencies may respond in unintended ways. For instance, an NRSRO may more closely adhere to a quantitative model than is warranted. Alternatively, there is the prospect that an NRSRO may adjust its ratings process to reduce the likelihood of a recordable material difference. More to the point, even as we prepare to adopt this rule change, it is worth recognizing that in some cases it may be entirely appropriate for an NRSRO to deviate from the credit rating implied by a quantitative model.
Today we also are voting on whether to require an NRSRO to disclose publicly, with a six-month lag, the ratings history of a random sample of 10% of the ratings issued for each class of credit rating for which the NRSRO issues a threshold number of ratings paid for by the obligor, issuer, underwriter, or sponsor. In other words, this 10%/six-month disclosure requirement applies to the so-called “issuer-pay” model of credit ratings.
Among other things, this rule change responds to concerns about conflicts of interest stemming from the issuer-pay model. The public disclosure of credit rating track records promotes accountability of an issuer-pay NRSRO by shining additional light on its performance.
Before us is a proposal similar to the 10%/six-month disclosure requirement. The proposal requires NRSROs to disclose publicly, with a 12-month lag, ratings histories for 100% of its issuer-pay credit ratings issued on or after a certain date. We seek comment concerning whether an NRSRO using a so-called “subscriber-pay” model also should be subject to a public disclosure requirement of its credit rating track record.
I support this proposal and look forward to reading the comments.
A feature of the actions the Commission is poised to take is that the issuer-pay model and the subscriber-pay model are differentiated.
One stated purpose of the Rating Agency Act is to foster competition. In short, competition is thought to lead to a higher quality product.
NRSROs using the subscriber-pay model are a key source of competition. Accordingly, as the Commission considers rules to improve the performance of the credit rating industry, the Commission must carefully balance the benefits of regulation against the cost of undercutting competition if the subscriber-pay model is overly burdened. Put differently, the Commission must be mindful that regulatory demands could prove counterproductive — stifling competition by erecting entry barriers or otherwise imposing burdens on subscriber-pay NRSROs that compromise the subscriber-pay business model. Given that subscribers pay for ratings, requiring subscriber-pay NRSROs to disclose their ratings publicly may have just this effect.
Further, it is important to state the obvious — namely, subscriber-pay NRSROs compete for subscribers. Such competitive pressures — including the importance of developing and preserving their reputations for accuracy — can incentivize subscriber-pay NRSROs to generate quality ratings that subscribers demand.
To be sure, there are differences between the issuer-pay and the subscriber-pay models. However, it is not enough to appreciate these differences, because there is another question — that is, do the differences between the models argue for different regulatory treatment? The comments will be constructive in assessing this question.
Finally, in an effort to spur competition, we are voting on a proposal to require an arranger of a structured finance product to disclose to NRSROs the information provided to the hired NRSRO to determine the credit rating. This proposal is designed to increase the number of NRSROs rating a particular product via unsolicited ratings. Unsolicited ratings are a means of accountability and a way for newer entrants to establish themselves as providers of quality ratings. I support this proposal and look forward to reading the comments.
Like my colleagues, I want to thank the staff for its hard work and dedication preparing these rules. I commend the Division of Trading and Markets, especially Randall Roy, as well as the Office of Economic Analysis and the Office of the General Counsel, for their time and commitment.