Dissenting Statement at Open Meeting Regarding Final Rules on Nationally Recognized Statistical Rating Organizations

Commissioner Michael S. Piwowar

Aug. 27, 2014

Thank you, Chair White.

Let me start with a simple observation.  When investors place too much weight on the value of credit ratings and too little on their own due diligence, the negative consequences can be momentous.  This statement seems rather obvious as we look back on the devastating effects of the recent financial crisis, which has been attributed in part to investor overreliance on credit ratings for structured finance products that, in hindsight, used faulty assumptions and led to a large number of downgrades.  As regulators, we must ask ourselves what we can do to address the problem of investor overreliance on credit ratings in our financial markets. 

I am pleased that the SEC has undertaken, and continues to undertake, efforts to discourage investors from unduly relying on credit ratings.  As I already mentioned in my statement regarding final rules on asset-backed securities,[1] I am especially happy that the Commission is removing references to credit ratings from our rules, as required by Section 939A of the Dodd-Frank Wall Street Reform and Consumer Protection Act[2] ("Dodd-Frank").[3]  The removal of credit rating references should encourage investors to view ratings as only one of many estimates of credit risk. 

In contradiction to our reference removal efforts are Dodd-Frank provisions that, by seeking to "impose discipline" on credit rating agencies, could have the perverse effect of entrenching undue reliance on credit ratings.  The rulemaking we are considering today purportedly will enhance the regulatory structure for nationally recognized statistical rating organizations ("NRSROs").  I regret that is not the case.

Some of the requirements that will be imposed under the rules and rule amendments being adopted today follow thoughtful staff recommendations to implement specific Dodd-Frank mandates.  Make no mistake, I question whether many of these reforms will have the intended – or really any – practical, beneficial impact for investors.  Nonetheless, if all of the rules had been carefully designed, so that the rule text closely adhered to the statutory language and minimized any potential detrimental effects on investor protection, competition, efficiency, and capital formation, I could have considered supporting them. 

Unfortunately, the rules before us also reflect discretionary choices by a majority of the Commission that go well beyond the prescriptive Dodd-Frank mandates.  These discretionary choices presume to make the rulemaking "better" and "tougher," but there is no evidence that will be the case.  Instead, they are simply unnecessary.     

Statutory Mandates

Let me give two examples of provisions in the final rules that are required by Dodd-Frank and yet, in my view, have questionable value.  First, the statute directs that the Commission require each NRSRO to have policies and procedures to apply rating symbols consistently for all instruments for which the symbol is used.[4]  The final rules implement that mandate through language that faithfully tracks the statutory text.[5]  I agree with the concern that credit ratings may be confusing and even downright misleading if they are not applied consistently.  But academic research indicates that trying to achieve perfectly comparable rating scales is not only impractical – it is impossible.  Despite any efforts by the credit rating agencies to maintain ratings comparability, the risk profiles of distinct asset classes are significantly different and thus result in varied performance of the instruments.[6]  In addition, scholars have pointed out that one potential unintended consequence of standardized ratings is that the credit rating agencies might become more cautious in their assessments, thereby reducing the amount of useful information they provide to investors.[7]

Furthermore, there is always the possibility that in an effort to avoid universal rating requirements, an NRSRO could simply create very similar, but distinctly different, symbols for each type of instrument – e.g., the familiar AAA through D scale for corporate securities, the same scale with a subscript "M" added for municipal securities, and the same scale with a subscript "S" for structured products – which would be entirely consistent with the statute and our rule.[8]  The other rules before us today, and those that precede them, already provide transparency regarding ratings performance and methodologies.  So, even without imposing universal rating requirements, other rules could achieve much of what such a system would be designed to do, without unnecessary additional burdens and uncertainties. 

Second, pursuant to Exchange Act Section 15E(c)(3)(B)(iii), final Rule 17g-3(b)(2) requires that the chief executive officer (or a similar individual) of each NRSRO annually sign a statement regarding the effectiveness of the entity´s internal control structure.[9]  These certifications will accomplish nothing other than to create potential personal liability.  I recognize that some people believe putting a member of management at an NRSRO "on the hook" will result in more so-called "accurate" ratings.  But a CEO certification will do more to line the pocketbooks of plaintiffs´ lawyers than it will ever do to make credit ratings more informative and useful or to protect the investors that utilize them.    

While I disagree that universal ratings and CEO certifications will achieve any of the goals they are intended to serve, at least they each are limited to targeting a specific Dodd-Frank mandate.  That cannot be said for all components of today´s rulemaking.

Discretionary Choices

With respect to additional obligations being imposed on NRSROs based on judgments about what is dictated by the "spirit" of Dodd-Frank, two provisions are particularly troublesome.  Rule 17g-5 is being amended to mandate the separation of ratings determinations from NRSRO sales and marketing activities.  It is important to look at the prohibition in the context of the existing provisions of Rule 17g-5(c), which themselves limit certain conflicts of interest at an NRSRO.  The final rule will have eight – count them eight – prohibitions regarding an NRSRO´s behavior and structure.  As a general matter, I see little reason to adopt further restrictions on credit rating agencies´ sales and marketing activities at this time given the multitude of protections already in place. 

But, notably, the rule does not stop with prohibiting sales and marketing employees from participating in ratings determinations.  It includes language that prohibits an NRSRO from having an employee involved in the ratings determination process from being "influenced by" sales and marketing considerations.  This new rule text sets an impossible standard for compliance and has no limiting principle.  Any NRSRO that grows its business in any asset class could be suspected of having the ratings determination process "influenced by" sales and marketing considerations.  It is not just management at the NRSRO whose motives could be questioned; every NRSRO employee, including those involved in ratings determinations, has an interest in the success of the enterprise.  Could the efforts of an NRSRO to increase, or even maintain, its market share and presence, and therefore support its workforce, be seen as "influencing" employee conclusions about ratings?  Had we posed focused questions about these issues at the proposing stage, perhaps the public could have suggested better ways to achieve the same objectives.  Unfortunately, because the majority of the Commission supports moving forward without a re-proposal, we will never know. 

Second, and just as concerning, is the rule the Commission is adopting that will require NRSROs to establish, maintain, enforce, and document an internal control structure.  The release purports to allow firms to tailor their internal control structures to their particular circumstances.  And yet, the rule being adopted lays out twelve specific, nonexclusive factors that NRSROs must "consider" when establishing an internal control structure, another three that apply to the maintenance of the structure, and two that relate to enforcement of the structure.  That is seventeen too many factors to actually create any true flexibility.  I have made clear during the process leading up to today´s vote that I was not precluding the possibility that it might be appropriate for us to later propose (or re-propose) to supplement our rules to prescribe specific standards for an NRSRO´s internal control structure. 

However, we should not be doing so today; we must be judicious about the need for – and timing of – rulemakings.  We should allow NRSROs sufficient time to comply with the comprehensive set of rules we are adopting today, together with the statutes and rules that have previously transformed the industry.[10]  Then, our expert examination staff can evaluate the implementation of internal controls and identify items that might be further enhanced via guidance or rulemaking.[11]              

Other Observations

Finally, there is one element of today´s rulemaking that deserves special attention for a different reason.  It is not only something we should absolutely be doing here, but it is so important that it is appropriate in countless other regulatory contexts.  Rule 17g-5(f), as being adopted, allows the Commission to individually exempt small NRSROs from the prohibition on sales and marketing activities,[12] on the basis that it may be difficult, or even impossible, for those NRSROs to maintain a strict separation between functions.  I wholeheartedly support including this provision in the rule.  It acknowledges that one-size regulation should not fit all, and should serve as a model for the Commission.[13]  We must always be cognizant that complex requirements may disproportionately impact small entities and in many circumstances unnecessarily create barriers to entry and impose negative effects on competition. 

Before I close, I want to join my colleagues in acknowledging the hard work of the Commission staff.  This rulemaking covers an area of our financial markets that is full of nuanced issues, which are challenging in and of themselves.  Layering upon that the discretionary decisions of a divided Commission has put you in an unenviable position.  I may not support the recommendation being put forward today, but that in no way should detract from the tireless efforts that culminated with this vote. 

Thank you.  I have no questions.