Statement of The Bond Market Association
SEC Hearing on Credit Rating Agencies
November 21, 2002
The Bond Market Association (the "Association")1 is pleased to submit this statement to the Securities and Exchange Commission (the "SEC" or "Commission") in connection with its November 21, 2002 hearing on credit rating agencies. This hearing is being held in connection with the study required to be conducted by the Commission pursuant to Section 702 of the Sarbanes-Oxley Act of 2002 (the "Act").
Summary of Views
Credit rating agencies play a critically important role in the efficient functioning of the fixed income markets by providing an independent source of information on the credit standing of corporate and other issuers of debt securities. Although the reliability of ratings as a source of information differs among credit market sectors, the current system functions reasonably well, and the Association does not believe that significant additional regulatory oversight of the process by which ratings are assigned is either necessary or desirable. Further, any attempt to do so could imply that the government has "certified" ratings, which could cause investors to over rely on ratings.
The Association does believe, however, that given the importance of the role played by rating agencies in the capital markets, the process by which agencies are designated as nationally recognized statistical rating organizations ("NRSROs") should be made more transparent, and the standards that are applied should be more clearly articulated. Further, we believe that the Commission should consider whether such standards could be applied in a way so as to encourage greater competition by designating as many NRSROs as possible for purposes of rating securities in one or more market sectors. In other words, the NRSRO process should not, in and of itself, be a "barrier to entry."
Impact of Rating Agencies on the Fixed Income Markets
Rating agencies have long played a critical role in the efficient functioning of the national and international fixed income markets. Of the factors that impact the economic value of any particular credit instrument, most - like maturity, yield, call features, and priority vis-à-vis other classes of creditors - can be identified and measured with precision. Issuer credit, however, innately cannot be measured with precision, because many of the factors that relate to issuer or counterparty credit, including the capability and experience of management, the quality of risk controls, and the ability to adapt quickly to changing market demands, among others, require a significant degree of subjective assessment. Rating agencies, by aggregating all of the factors, both objective and subjective, that form a part of credit standing give market participants an additional source of information that can help to confirm market assessments of credit risk. Because ratings provide a reasoned and categorical assessment of the relative creditworthiness of credit instruments and issuers, it is appropriate that SEC rules incorporate references to ratings as benchmarks in the context of specific regulations, such as the broker-dealer net capital rules, that are sensitive to the relative safety of different investments.
At the same time, it is important not to overstate the importance of ratings from a risk management perspective. Sell-side and large buy-side firms that are active in the fixed income markets conduct their own intensive credit analyses for risk management purposes, including the maintenance of adequate capital against proprietary positions, and for purposes of identifying pricing discrepancies in conducting their trading operations. From the standpoint of major participants in the fixed income markets, therefore, ratings do not substitute for the need to carefully monitor the credit of issuers as to which firms have substantial exposure, but instead form one part of the information mix that they use in performing that function.
In general, the current system works well in evaluating the credit of corporate issuers on a current basis.
Ratings issued by the major rating agencies have proved to be a reliable source of information for the fixed income markets. The reputational and commercial interests of the agencies provide a strong motivation to maintain the credibility of ratings. Further, a variety of studies over time have demonstrated a consistent and clear correlation between ratings and probability of default. Studies also show a correlation between credit rating and default rate volatility, or the length of time preceding a default. Less information is available with regard to structured financings or securities other than long-term corporate debt, and so there may not be sufficient information from which to draw firm conclusions about the extent of the correlation between ratings and default probability for these other types of instruments.2 Further, the correlation observed for corporate debt is not perfect, and there is significant heterogeneity within individual ratings categories.3
In any event, there should be no expectation on the part of regulators or market participants that any ratings system will act as a perfect evaluator of credit quality. This is true because of the complexity of evaluating the various factors and reflecting those factors in a single symbolic rating. Further, as noted above, it is important to appreciate that credit ratings represent only one information source available to market participants about credit quality. Sell-side and major buy-side institutions must and do conduct their own analyses. Further, much information is available to the marketplace in the form of research conducted by both securities firms and independent research firms that is disseminated to clients.
In performing their reviews, rating agencies rely partly on publicly available financial and other information and also to a large degree on information that is not publicly disclosed. It is important that rating agencies have the flexibility to look beyond the information that is available in publicly available reports in order to discern risks and trends that may not be readily apparent to the marketplace at large. For this reason, we strongly support the continued exclusion of credit rating agencies from the persons as to which disclosure of material non-public information triggers a public disclosure requirement under Regulation FD.4 Removing this exclusion would provide a substantial disincentive to rating agencies to "dig beneath the numbers" and therefore sharply limit their ability to add value to the marketplace.
At the same time, rating agencies should not and cannot be reasonably charged with uncovering all possible undisclosed risks or liabilities that might impact credit quality. In particular, rating agencies, like other market participants, must be able to rely on the integrity of the audit process in producing financial information that is accurate and complete. Rating agencies, like the regulatory agencies, lack the resources and expertise to conduct an independent audit of all the financial information produced by the corporate issuers they rate and in particular cannot be expected to police in any meaningful way the review conducted and decisions made by accounting professionals. This issue is instead more appropriately addressed by the various reform measures recently adopted by Congress in the Act and related rulemaking by the Commission that are directed to the responsibilities of corporate issuers, officers, and accountants.
Significant additional regulatory oversight of the ratings process is neither necessary nor desirable.
As described above, the present system under which rating agencies operate, while not perfect, works reasonably well. Under the present regime, the SEC staff, through the "no-action letter" process, designates certain agencies as NRSROs, based on a number of criteria, including the requirement that the agencies be "nationally recognized," and criteria pertaining to operational capability and reliability.5 In addition, in practice each rating agency is registered as an investment adviser under the Investment Advisers Act of 1940. Under this approach, the rating agencies effectively are subject to a baseline level of regulatory oversight, and maintenance of the conditions on which the designation is based is enforced by the potential that it can be withdrawn at any time.
Beyond this degree of oversight, the Commission does not attempt to review or mandate the process by which individual ratings are assigned or updated. The Association believes that it would not be appropriate or desirable for it to attempt to do so. First, the complex nature of the ratings process and its application to individual companies does not lend itself to a system of rigorous government oversight. Neither the Commission nor any other government agency is well positioned to evaluate the validity of individual rating decisions based on numerous, and to some extent subjective, factors.
More important, the entire federal scheme of securities regulation is designed deliberately to avoid "merit regulation," or the endorsement or certification of the value or safety of any particular investment. The federal securities laws and regulations reject this approach for at least two reasons. First, it is impossible for any one authority to accurately measure relative investment merits in the context of a constantly changing marketplace.
Second, the Congress, the Commission, and other regulators have made a determination that the markets work most efficiently and provide the maximum common benefit when the merits of individual securities are judged by the countless decisions of individual market participants, subject to legal obligations to fully disclose all material facts and risks. Any attempt by a regulatory authority to mandate or certify in a specific way the process by which ratings decisions are made would necessarily carry with it a governmental imprimatur that investors might rely upon. It is critical that investors understand that ratings are not infallible and that a favorable rating is not a guarantee of safety. A regime that implies a different message would give investors false comfort which, when proven false, would undermine rather than reinforce investor confidence.
In this regard, the recent Senate staff report on the financial oversight of Enron Corp. (the "Staff Report") draws a troubling comparison between the role played by rating agencies in the financial markets and that performed by the Food and Drug Administration (the "FDA") with regard to the marketing of drugs.6 Certification of drug safety is, of course, a unique public safety concern that calls for direct government oversight in order to protect the public health. The issuance of security ratings is one discrete part of the overall process by which market participants evaluate the relative safety of different investments. Further, the process for testing new drugs by its nature is amenable to scientific research; the process for issuing credit ratings may never be entirely objective and is certainly not scientific.
If individual ratings were deemed to be "certified" in the same way that drugs are certified for use by the FDA, the result, as discussed above, would be to mislead the investing public and potentially undermine investor confidence. Also, the statement in the Staff Report that the rating agencies "answer to no authority"7 is simply not accurate. As described above, the rating agencies are registered as investment advisers with the Commission, which also has the authority at any time to withdraw or amend the terms under which they may be treated as NRSROs.
The SEC designation process should be more transparent, and the SEC should consider whether it could be applied in a way so as to promote competition.
Although the Association opposes any substantially more direct or substantive governmental role in the ratings process, we do believe that the current designation process should be made more transparent and understandable to the investing public. Greater transparency would reflect the important role that rating agencies play in the process of capital formation and secondary market trading, as well as the importance of ratings to the Commission's own rules, without creating an inappropriate governmental stamp of approval of individual ratings.
The Commission itself made a proposal to increase transparency in the designation process in 1997 (the "1997 Proposal).8 Under the 1997 Proposal, the Commission for the first time would explicitly define in its rules the term "nationally recognized statistical rating organization" as an organization that has been specifically designated as such by the Commission, after considering the following general factors: (1) national recognition; (2) adequate staffing, financial resources, and organizational structure; (3) use of systematic rating procedures that are designed to provide credible and accurate ratings; (4) the extent of contact with management of issuers; and (5) the existence of procedures to prevent misuse of non-public information. In addition, each agency that is so designated would be required to be registered as an investment adviser under the Investment Advisers Act of 1940.
Under the 1997 Proposal, each rating agency would file an application with the Commission. The initial decision would be made by delegated authority to the staff of the Division of Market Regulation, with a right to appeal an adverse decision directly to the Commission. Organizations that are currently operating under an NRSRO designation would retain this status and would not be required to reapply, although the Commission would conduct reviews to determine that they meet the requirements of the rule.
The Association favors the general thrust of the 1997 Proposal, but would suggest the following modifications:
First, it would be helpful for the Commission to promulgate a specific form of application that specifies the questions to be answered and information items to be furnished by each applicant. This would help to ensure that the application process is standardized and that each applicant is asked to provide the same type and scope of information.
Second, in order to ensure that the process is as transparent as possible, the individual applications and exhibits should be made public by the time that they are acted upon by the Commission. An applicant that is likely to receive an adverse determination would be permitted to withdraw the application without public disclosure if it chose to do so. Upon request by the applicant, the Commission could keep confidential certain of the information that was provided if it were deemed necessary to protect legitimate proprietary interests of the applicant.
Third, the decision by the staff (or, upon appeal, the Commission) should in explicit terms describe why the application is being granted or denied, by reference to the factors contained in the rule and the information that has been furnished by the applicant and should clearly describe any conditions that are imposed in the individual case in order to maintain the designation. It is critical to the integrity of the process that investors and other interested parties be able to understand the Commission's or staff's reasoning in acting on each application.
Fourth, we believe that existing NRSROs should be required to complete the application process, while remaining designated pending final Commission action on each application. This step would be important in order to ensure that all applicants, now or in the future, are being treated fairly and held to the same standards.
Fifth, the rule should include an annual certification requirement relating to continuing satisfaction of the criteria and any other conditions the Commission may have imposed in approving the designation. In assuring that a rating agency continues to meet the standards that have been set, it is not sufficient to hold out the threat of withdrawing the designation based on a subsequent determination that the standards are no longer satisfied or conditions have not been met. Given that most rating agencies are likely to operate as private corporations and not subject to public disclosure requirements, following the initial designation, the Commission likely will not have the realistic ability to carefully audit continuing compliance. An annual certification could help to ensure a comprehensive ongoing review of their compliance by the agencies themselves.
Finally, we believe that the Commission should undertake the designation process with a view to promoting competition by designating additional organizations, consistent with prudent application of the regulatory standards. One striking characteristic of the status quo is that there are at present only three entities designated as NRSROs. Although part of the explanation for this fact undoubtedly lies in market-driven factors like economies of scale, it may also be affected by past reluctance by the staff to countenance new entrants.9 Of course, in cases where agencies propose or are deemed competent to rate debt in only one or more market sectors, the designation could be appropriately limited to those sectors.
Further, in terms of considering new entrants, although the factors identified in the 1997 Proposal are appropriate and relevant to determining whether to designate an entity as an NRSRO, we would suggest that the "nationally recognized" criterion not be established as a pre-condition for the designation if all the other requirements are satisfied (in this event, the more appropriate term for designated agencies might be "designated statistical rating organizations"). An agency that is well established in a foreign jurisdiction, for example, may not be "nationally recognized" in the U.S. and yet still be capable of credibly rating certain classes of instruments. Further, because the NRSRO designation is so important to treatment under various of the Commission's rules, and is therefore important to the marketplace, the absence of the designation could itself effectively preclude an agency from becoming nationally recognized. To require, therefore, that national recognition is a prerequisite to designation could be to enforce a "Catch 22" that will effectively maintain the number of rating agencies at an artificially low number.
We are pleased to have been given the opportunity to participate in the Commission's review and look forward to further dialogue with the staff on this topic.
|1|| The Association represents securities firms and banks that underwrite, distribute, and trade in fixed income securities, both domestically and internationally. Further information regarding the Association, its members, and its activities can be obtained from our public website, www.bondmarkets.com.
|2|| See generally, Credit Ratings and Complementary Sources of Credit Quality Information, Basel Committee on Banking Supervision Working Papers (August 2000), at 126-7.
|3|| Id. at 128. Note that these studies precede the recent default experience of various corporate issuers and so do not necessarily point to a similar correlation during 2001-2002.
|4|| Regulation FD Rule 100(b)(2)(iii).
|5|| See Testimony of Isaac C. Hunt, Jr. Concerning the Role of Credit Rating Agencies in the U.S. Securities Markets, Before the Senate Committee on Governmental Affairs (March 20, 2002), available at www.sec.gov/news/testimony/032002tsih.htm.
|6|| "Financial Oversight of Enron: The SEC and Private-Sector Watchdogs," Report of the Staff to the Senate Committee on Governmental Affairs (October 8, 2002), at 126.
|8|| Securities Exchange Act Release No. 39457 (December 17, 1997).
|9|| See, e.g., Lawrence J. White, Stern School of Business, New York University, "The Credit Rating Industry: An Industrial Organization Analysis" (2001).