November 30, 1998
Mr. Jonathan G. Katz
Secretary
Securities and Exchange Commission
450 Fifth Street, N.W.
Washington, D.C. 20549
Re: Notice of Filing of Proposed Rule Change by the National Association of Securities Dealers, Inc. Relating to Bond Mutual Fund Volatility Ratings
(File No. SR-NASD-97-89, Amendment No. 1)
Dear Mr. Katz:
The Investment Company Institute1 appreciates the opportunity to comment on the proposal of NASD Regulation, Inc. (NASDR) to add a new interpretation to Rule 2210 of the National Association of Securities Dealers, Inc. (NASD) that would permit members and associated persons to include bond mutual fund volatility ratings in supplemental sales literature on an 18 month interim basis.2
As discussed in more detail below, this proposal is of enormous significance to mutual funds and their shareholders. Since the NASD first proposed allowing bond fund "risk" ratings in sales material two years ago, there has been much study, analysis and debate regarding this proposal. Advocates of these ratings have asserted that they should be allowed on the theory that greater information can only help investors.3 The ICI and others, however, have argued that certain types of information can be inherently misleading, and hence harmful. Based in part upon the results of a comprehensive consumer survey, we believe that "risk" ratings fall within this category.
Accordingly, we continue to harbor serious reservations about the use of "risk" ratings in mutual fund sales material. Nevertheless, we generally support NASDR’s proposal to permit such ratings for a trial period. In large part, this is due to the fact that the proposed conditions on the use of these ratings would address many – though not all – of their potential hazards. Consequently, the SEC and NASDR should vigorously resist arguments to weaken these conditions and put investors at risk. In particular, we support the requirements that volatility ratings must be based on objective factors, must be in narrative form, must meet timeliness standards, and must be accompanied by clear, comprehensive disclosures. We also support the requirement that NASDR approve any fund sales literature containing volatility ratings. We recommend, however, that the proposal be revised to prohibit the payment for ratings by the fund being rated.
With respect to other aspects of the proposal, we support the deletion of the requirement that only nationally recognized statistical rating organizations (NRSROs) be allowed to issue volatility ratings. In addition, consistent with the idea of allowing these ratings on a trial basis, we recommend that NASDR commit to undertake a formal evaluation of the results of the trial period, and solicit further comment before considering whether to adopt the rule on a permanent basis. Finally, in response to a request for comment in the Release, we do not believe that the proposal should apply to risk designations produced by mutual funds themselves.
Each of these points is discussed below.
Background
In December 1996, the NASD issued a Notice to Members seeking comment on the use of bond fund risk ratings in supplemental sales literature. The ICI opposed the use of such ratings, as formulated at that time by the rating agencies, because: (1) such ratings would be misleading insofar as they imply any certainty about a fund’s future investment performance; (2) the all-in depiction of risk that such ratings purport to provide would be illusory, since risk is not a unitary concept nor identical for all investors; (3) the ratings would be new, largely untested, not prepared in accordance with a standardized methodology and inherently subjective; (4) the ratings implicitly viewed risk to be synonymous with short-term volatility, a dubious standard for many investors; (5) the payment arrangement for the ratings would present a significant conflict of interest; (6) a fund’s risk rating could become quickly outdated, unbeknownst to those investors who relied on them; and (7) the services issuing the ratings are unregulated, and they disclaim any liability for the accuracy of their ratings.4
Shortly thereafter, the ICI conducted a research survey focusing on shareholders’ assessment of bond fund risk ratings.5 We surveyed mutual fund shareholders to determine how well they understood bond fund risk ratings, how they would use them and their expectations about them. The results of the survey substantially reinforced the ICI’s concerns noted above. They showed, among other things, that: (1) most investors surveyed did not understand risk ratings, yet they would rely upon them when considering the purchase of a bond fund; and (2) many investors would be encouraged to invest in "low risk" rated funds, even though such investments might be inappropriate given their long-term investment goals. The survey also showed that the potential for risk ratings to mislead investors was accentuated by the use of numbers, letters, symbols or stock phrases to designate the ratings.
The ICI’s concerns were shared by organizations both within and outside the mutual fund industry. For example, the Consumer Federation of America expressed many of the same concerns about the potential harm of these types of risk ratings to investors.6 Many commenters on the NASD’s Notice to Members recommended that, if risk ratings were to be permitted, the NASD should adopt appropriate conditions to ensure the integrity of the ratings.7
In light of the significance of this issue, the ICI’s Board of Governors appointed a committee of industry leaders in October 1997 to examine the use of bond fund risk ratings in sales literature and their suitability for investors. The committee met with NASDR board representatives and staff and expressed the view that, if risk ratings were permitted to be used, appropriate investor safeguards needed to be put in place. Examples of such safeguards include the following:
- Risk ratings should only be permitted if they are not paid for by the fund being rated.
- Only risk ratings that are determined on the basis of objective methodologies should be permitted.
- Only risk ratings that are in a narrative format should be permitted. The term "risk" rating should be prohibited.
- Risk ratings should be accompanied by meaningful disclosure of equal prominence to the rating. The specific disclosures that should accompany a rating should include both information about the specific rating (e.g., a description of the rating methodology) and more general cautionary disclosure (e.g., the rating’s prediction of the fund’s likely future performance may be wrong, and the rating evaluation may not be current as of the time an investor purchases shares in the fund).
NASDR’s Proposal
NASDR’s proposed rule change would permit, on an 18-month pilot basis, the use of volatility ratings in bond mutual fund supplemental sales literature provided that such ratings meet certain requirements and are accompanied by certain disclosures. We support the conditions in the proposed rule, which appear to address many of the problems with the initial proposal that we and other commenters had identified. Any diminution in these safeguards would leave investors exposed to potential harm. In fact, the Institute believes NASDR’s proposal should be strengthened, by prohibiting the payment for the ratings by the fund being rated. Our specific comments on the proposal are set forth below.
Objective Factors
The proposed requirement that the ratings be based exclusively on objective, quantifiable factors is one of the most important conditions in the rule. If rating agencies are permitted to base ratings on subjective factors, there will be an increased risk of inconsistent ratings, which would reduce the ratings’ comparability and increase investors’ misunderstanding of ratings. Moreover, competitive pressures on the rating agencies may encourage mutual funds to "shop around" for the most favorable ratings, thereby compromising the integrity of the ratings. As noted in the Release, the proposed requirement that the ratings be based upon objective factors should reduce the potential for ratings shopping.
Use of Symbols, Numbers or Letters; Use of Term "Risk" Ratings
We strongly support the proposal’s prohibition on ratings being designated by use of a single symbol, number or letter, or being described as "risk" ratings. Allowing ratings to be designated by a single symbol, number or letter would increase the likelihood that an individual investor would not evaluate the risk of a bond fund based on his or her investment objectives and risk tolerance, and instead would look to a single symbol, letter or number to make this crucial decision. The same danger is present if these ratings were to be described as "risk" ratings.
Timeliness; Description of Methodology
We support the proposed requirements that a rating be current and that the rating’s issuer disclose to investors its rating methodology through a toll-free number or web site. Ratings can quickly become stale based on changes in a fund’s portfolio and in market conditions; accordingly, the timeliness of a fund volatility rating is crucial. Similarly, investors should have access to the methodology used by a rating agency to help them determine the meaning and usefulness of a particular rating.
Disclosure
We also support the proposal’s disclosure requirements. As we have indicated in the past, volatility ratings are fraught with risk that they will mislead investors as to their meaning and as to their limitations. It is important that any supplemental sales literature explain the meaning and shortcomings of the ratings. It is also important that such disclosure be given equal prominence to the rating to ensure that investors will read it. The proposed requirement that the rating be contained in the text of the mandated disclosure statement should help to accomplish this goal.
Limitation to NRSROs
We support the deletion from earlier versions of the proposal of the requirement that only NRSROs be allowed to issue volatility ratings. The NRSRO designation is used in other SEC rules principally in connection with the issuance of opinions regarding a debt securities issuer’s creditworthiness. The expertise that NRSROs have in issuing credit ratings of debt securities is not at all indicative of their ability to predict the expected volatility of a bond mutual fund. Accordingly, we support the deletion of this requirement.
Payment for Ratings
Notwithstanding our general support of the proposal, we strongly recommend that it be revised to prohibit the payment for the ratings by the fund being rated (or its affiliates). This prohibition, which is one of the conditions on use of fund performance rankings,8 would ensure that the rating agency is independent of the fund being rated and, thus, would avoid the inherent conflicts of interest that this type of payment arrangement presents. Because the rating agency is paid by the fund being rated, there is likely to be pressure on the agency to produce a rating that is more favorable than the facts suggest. This will severely undermine the integrity of the ratings.9 The only justification of which we are aware for permitting payment for ratings is that this is consistent with current business practices of those NRSROs that wish to offer ratings. We do not believe that the desire to avoid adjusting compensation practices outweighs these significant investor protection concerns and accordingly urge that the rule prohibit the use of ratings that have been paid for by the rated fund.10
Scope of Proposal
The proposed rule by its terms applies to bond "mutual funds." We assume that this means that the proposal only applies to open-end investment companies (i.e., mutual funds), and does not apply to other types of investment companies, such as unit investment trusts or closed-end investment companies. There have been instances in the past where there was confusion regarding the applicability of NASD rules to such other investment companies. Therefore, to avoid any confusion, the SEC’s release approving the proposal should expressly state that it does not apply to any investment companies other than open-end investment companies.
Proposed Trial Period
Consistent with the notion of conducting a trial period for the use of risk ratings, it is important that NASDR proceed cautiously after the period’s end, and not automatically extend the rule for an indefinite period or adopt it on a permanent basis. In particular, NASDR should publish the results of the trial period, and solicit further comment before determining whether to adopt a final rule.11
Pre-Approval
Under the proposal, sales literature that includes a volatility rating would have to be filed with NASDR at least 10 days prior to use for approval, and such literature could not be used with the public unless it had been approved or revised to include any changes requested by NASDR. While we do not believe such review and approval is ordinarily necessary for sales material filed with NASDR, we believe it is appropriate in these circumstances. Risk ratings are being allowed only on an interim basis because of the significant concerns that have been raised over their potential to mislead the investing public. Consequently, it is entirely appropriate for NASDR to adopt additional precautions as it continues to evaluate these concerns.
Fund Descriptions of Risk
The Release states that many funds currently provide various descriptions of risk and volatility for their own funds that may involve some of the same processes and considerations that are used by independent rating agencies. The Release solicits comments on whether such fund-produced volatility descriptions should be subject to the proposed rule.
Fund-produced risk or volatility designations should not be subject to the proposed rule for at least two obvious reasons. First, unlike self-assessments made by funds themselves, volatility ratings purport to represent an objective third party’s evaluation of risk. Therefore, investors will likely place greater reliance – indeed, as the Institute’s survey indicated, too much reliance – on them. In fact, many of the conditions included in NASDR’s proposal (e.g., the precautionary disclosures) are designed to address this particular concern.
Second, fund self-assessments are designed for a completely different purpose than third-party volatility ratings. These measures are generally provided to investors as an educational tool to assist them in comparing different types of funds within a single complex (e.g., by illustrating that a long-term bond fund is riskier than a short-term bond fund). Volatility ratings, in contrast, are likely to be used by funds as a marketing tool to distinguish themselves from similar types of funds offered by other complexes (e.g., three long-term bond funds). Thus, fund self-assessments do not present the same potential for abuse as volatility ratings. Accordingly, given the significant differences between volatility ratings and fund self-assessments, it would be unnecessary and inappropriate to subject the latter to the proposed rule.
* * *
We appreciate the opportunity to comment on this important matter. If you have any questions regarding our comments, please do not hesitate to contact me at (202) 326-5815, or Amy Lancellotta at (202) 326-5824.
Sincerely,
Craig S. Tyle
General Counsel
cc: Paul F. Roye
Director
Douglas J. Scheidt
Associate Director and Chief Counsel
Division of Investment Management
Securities and Exchange Commission
Mary L. Schapiro
President
Elise B. Walter
Executive Vice President
R. Clark Hooper
Executive Vice President
John M. Ramsey
Deputy General Counsel
NASD Regulation, Inc.
ENDNOTES
1 The Investment Company Institute is the national association of the American investment company industry. Its membership includes 7,373 open-end investment companies ("mutual funds"), 450 closed-end investment companies, and 9 sponsors of unit investment trusts. Its mutual fund members have assets of about $5.061 trillion, accounting for approximately 95% of total industry assets, and have over 62 million individual shareholders.
2 SEC Release No. 34-40627 (Nov. 2, 1998), 63 Fed. Reg. 60431 (Nov. 9, 1998) (the "Release").
3 Proponents of risk ratings argue that these ratings, like credit ratings, provide investors with useful information. A credit rating, however, differs from a risk rating in several significant respects, including the long history of experience and the very nature of the rating (due to a commonly agreed upon measure of success: the likelihood of timely payment of interest and return of principal).
4 See Letter from Paul Schott Stevens, Senior Vice President and General Counsel, ICI, to Joan Conley, Office of the Corporate Secretary, NASD Regulation, Inc., dated Feb. 24, 1997 (commenting on NASD Notice to Members 96-84, December 1996).
5 See Shareholder Assessment of Bond Fund Risk Ratings: Report to NASD Regulation, Inc., Investment Company Institute, October 1997. See also Letter from Paul Schott Stevens, Senior Vice President and General Counsel, ICI, to Joan Conley, Office of the Corporate Secretary, NASD Regulation, Inc., dated Sept. 10, 1997 (transmitting the preliminary report); Letter from Craig S. Tyle, General Counsel, ICI, to Mary L. Schapiro, President, NASD Regulation, Inc., dated Oct. 29, 1997 (transmitting the final report).
6 See Letter from Barbara Roper, Director of Investor Protection, Consumer Federation of America, to Joan Conley, Office of the Corporate Secretary, NASD Regulation, Inc., dated Feb. 24, 1997.
7 See, e.g., Letter from Martin A. Corry, Director, Federal Affairs, American Association of Retired Persons, to Joan Conley, Office of the Corporate Secretary, NASD Regulation, Inc., dated Feb. 24, 1997.
8 See NASD Interpretive Material 2210-3.
9 Even a senior officer of one of the rating agencies has expressed concern about issuer fees. In a 1995 speech, Moody’s Executive Vice President Thomas J. McGuire noted that when they first considered charging issuers for ratings in the late 1950s, rather than charging subscription fees, the then-president of Moody’s said that they would never do so because "taking money from rated issuers would inevitably erode the credibility and integrity of the rating system." While on balance Mr. McGuire believes that issuer fees are justified, he recognizes the potential dangers that they present. Thomas J. McGuire, Executive Vice President and Director – Corporate Department, Moody’s Investors Service, Speech to the U.S. Securities and Exchange Commission, International Institute for Securities Markets Developments (April 28, 1995).
10 We acknowledge that limiting the scope of the rule to ratings based on objective factors reduces, to some extent, the potential for conflicts arising from payment for ratings (although it does not eliminate it). Consequently, if the SEC determines not to prohibit ratings that have been paid for, it is imperative that it maintain the conditions limiting the use of ratings to those based on objective factors and requiring that the methodologies be publicly available through a toll-free number or web site.
11 In this regard, it is important to keep in mind that evidence gathered during the trial period may not necessarily be indicative of whether bond fund volatility ratings pose significant investor protection concerns. For instance, the initial providers of these ratings likely will be limited to well-established and respected organizations. Moreover, these entities will presumably take special precautions during the trial period, given their interest in the proposal. NASDR needs to take into account the possibility that, when and if these ratings are permitted on a permanent basis, there will be a "race to the bottom" in both the quality and the presentation of the ratings, due to competitive pressures.