- Fund Regulation
- Retirement Security
- Trading & Markets
- Fund Governance
- ICI Comment Letters
September 10, 1997
Ms. Joan Conley
Office of the Corporate Secretary
NASD Regulation, Inc.
1735 K Street, N.W.
Washington, D.C. 20006-1500
Re: Institute Report on Shareholder Assessment of Bond Mutual Fund Risk Ratings in Supplemental Sales Literature
Dear Ms. Conley:
In March 1997, NASD Regulation, Inc. (NASDR) President Mary Schapiro observed, "We are here to serve investors, so it’s logical that we want to hear from them about our proposed rules."1 In this same spirit, through its ongoing program of research with mutual fund shareholders, the Investment Company Institute2 in recent years has sought to provide the regulatory and policymaking community with empirical data concerning fund investors and how they are likely to be affected by various regulatory decisions.3 We agree with Ms. Schapiro that the rulemaking process functions best when it takes account of "an array of constituencies, investors foremost among them."4
On this same basis, in March 1997, the Institute announced its plans to conduct and to provide NASDR the results of survey research devoted to risk ratings of bond mutual funds, as such ratings might be provided to investors in supplemental sales literature.5 In particular, the ICI’s survey was intended to assess how well mutual fund shareholders understand the nature of such ratings, how they are likely to use ratings in their investment decisions, and what expectations they have about the rating process and rating services. We are pleased to submit the results of the ICI’s research, which are set forth in the enclosed report. See Shareholder Assessment of Bond Fund Risk Ratings: A Report to NASD Regulation, Inc. (Sept. 1997) (the "Report"). The results of the ICI’s survey are summarized below.
I. Overview of Research Methodology and Findings
Response Analysis Corporation, an independent research firm, was engaged by the Institute to survey a randomly selected, representative sampling of mutual fund shareholders6 on bond fund risk ratings. The survey questionnaire was designed to explore, among other things, how risk enters into investment decisionmaking, what impressions investors may gain of bond fund risk ratings when used in sales literature, how they may view such ratings relative to other types of sources of risk information on a fund, how they are likely to use such ratings, and what assumptions they likely will make about the ratings.
Because NASDR does not currently permit the use of risk ratings to sell bond funds, it was necessary to develop simulated sales literature on an existing bond fund that had been assigned a risk rating by a rating service, which in this case was the service’s lowest risk rating.7 The sales literature prominently displayed the fund’s rating and provided an explanation of the rating system together with a variety of other disclosures about the fund and its risks. Preliminarily, each respondent was directed to review the sales literature as well as a copy of the fund’s prospectus. Every effort was made to replicate the likely experience of fund investors with risk ratings, were NASDR to authorize them for use in bond fund sales literature.
The findings of the survey, as detailed below, strongly substantiate concerns voiced about permitting the use of such risk ratings, including those concerns expressed in comments to NASDR by the ICI8 and others. Specifically, the Report indicates the following:
- Risk ratings have a powerful appeal. If they were available, investors would rely on them for their investment decisions. After only a brief exposure to risk ratings, investors are highly confident of their ability to use them. Those most likely to rely on ratings, and most confident in using them, are investors with the least understanding of bond fund investing.
- Investors would regard risk ratings, if they were available, as some of the most important information on a fund’s risk. Investors likely will consult narrative disclosures and other risk information less frequently.
- Investors do not understand the nature and limitations of risk ratings. Only four in ten of all shareholders surveyed were able to interpret the rating correctly. Even among those shareholders with a high level of understanding of bond fund investing, only slightly more interpreted the rating correctly.
- Risk ratings appear to influence investor selection of bond funds in ways neither intended by the ratings nor beneficial to many investors—e.g., by biasing investors toward the "lowest" risk funds, even where these funds may be less appropriate to an individual’s investment time horizon or investment goals.
- When presented with a risk rating, the overwhelming majority of investors have expectations about the rating process—including the availability of ratings, the qualities of a rating system, and the accountability of rating services and rating methodologies—that vary considerably from the rating schemes under consideration by NASDR.
II. Risk Ratings Have Powerful Appeal. Investors Will Rely Heavily upon Them, to the Exclusion of Other Risk Information
The survey provides strong empirical evidence that investors will use and rely on risk ratings if they are available, particularly those investors with the poorest understanding of bond fund investing.9
Following their brief initial exposure to sales literature reflecting a fund’s low risk rating, 80 percent of those surveyed said they would be either very confident or somewhat confident in using risk ratings. Nearly three-quarters of those surveyed thought a rating would simplify investment decisions. Two-thirds said they would pay close attention to a rating in assessing a fund’s risk, while 56 percent said they would rely on a rating. Among investors with a low understanding of bond fund investing, 74 percent said that they would pay close attention to, and 63 percent indicated that they would rely upon, risk ratings. By contrast, less than half of those investors with a high understanding of bond funds indicated they would rely on a rating.
Consistent with the high degree of reliance investors likely will place on risk ratings are survey results indicating that investors would regard ratings as especially important information, likely to displace narrative disclosure and other risk information they currently consider.
Among shareholders who actively consider risk information on a fund, over forty percent indicated they would consider the risk rating the first or second most important item of risk information available to them.10 This proportion was fairly constant across survey participants, irrespective of the level of their understanding of bond fund investing—including those who typically review only one or two items. For example, of the respondents who typically evaluate only one item of information about risk, slightly more than a third placed bond fund risk ratings ahead of that single item. Based on survey data concerning the number of items of risk information shareholders typically consider, it is likely that some current sources of risk information may be consulted less frequently if risk ratings were available. One example is the written description of a fund’s risk, the fifth most commonly identified item of risk information.
III. Investors Do Not Understand the Limitations of Risk Ratings
The survey was conducted in such fashion as to provide assurance that participants actually had reviewed the explanation and disclosure accompanying the rating in fund sales literature, together with the fund’s prospectus. In these controlled circumstances, however, it is apparent that even seasoned investors will widely misunderstand the nature and limitations of a risk rating.11
Less than half of all those surveyed (41 percent) properly identified relative interest rate sensitivity as a factor about the fund that might be determined from its "R-1" or "extremely low" risk rating.12 Among investors with the best understanding of bond fund investing, only 44 percent made this correct identification; over one-third of this group thought the rating reflected the fund’s past performance, while over 40 percent believed it to indicate how the fund’s return would be impacted by a down stock market. A quarter of all those surveyed took the rating to indicate whether they could lose money investing in the fund.
IV. Risk Ratings Will Cause Investors to Make Inappropriate Investment Decisions
The survey demonstrates that risk ratings, if they were available, would influence investment decisions significantly—in many cases to the detriment of the investor.
The survey assembled a variety of information on participants as individual investors, including their risk tolerance, investment time horizon, and investment goals. A substantial proportion of investors with the highest risk tolerance, as well as those with the longest investment time horizon or with long-term growth as their investment goal13 indicated they would buy a bond fund with a low risk rating. This included approximately 40 percent of all investors having a horizon in excess of five years or having growth as their primary strategy, and almost 30 percent of those with substantial or above-average risk tolerance. The results suggest the likelihood of broad misunderstanding of the limitations of a fund’s risk rating as translated into an individual’s investment program. The results also suggest a strong and inappropriate bias toward bond funds with low risk ratings, indicating that investors would perceive such a rating per se to be a favorable fund attribute and an indicator of the fund’s suitability for them. If funds are sold on this basis, the ratings accordingly can be expected to influence investor selection of bond funds in ways that are not intended by the rating nor most beneficial for many investors.
The results of the survey also indicate that many investors may believe rated funds might be superior to unrated funds. Thirty-four percent of the respondents reported that a rated fund was better for themselves personally, and 22 percent felt that rated funds would be better than unrated funds for the majority of investors. Together, these survey results indicate that many investors will regard a fund rated low-risk to possess uniquely favorable characteristics, making it generally more suitable for investment than any other bond fund, whether rated or unrated. This perceived reality of the bond fund marketplace may leave fund sponsors with little choice other than to seek for their funds, and to market them on the basis of, the "most favorable" rating.
V. Investor Expectations about Risk Rating Systems Differ Sharply from the Systems under Consideration by NASDR
The rating systems under consideration by NASDR contemplate that risk ratings would be provided only to those bond funds that purchase them, would be made available to the public on a selective basis, and would be assigned according to rating services’ different proprietary methods that take into account both quantitative and qualitative factors. For these reasons, bond funds could not be compared readily with one another based on a rating. Moreover, it is not anticipated that either the services providing such ratings or the rating systems themselves would be subject to government regulation or independent review. Nor is it anticipated that rating services would be required to accept expert liability for their ratings.
In virtually all of these particulars, the actual bond fund risk rating systems being urged on NASDR are sharply at odds with what investors naturally will expect or assume when provided a risk rating in fund sales literature.14 The substantial majorities reported in the survey on these points are striking. Thus, over 80 percent of those surveyed said they expect that a consistent method would be used to rate all bond funds, that funds could be compared with one another on this basis, and that a fund’s rating would be available even if it were rated high risk. Over 70 percent expect that a risk rating would be available for every fund, that the ratings would be based only on statistical data and not opinion, and that the rating methods would be reviewed by independent third parties. Almost two-thirds assume that rating services and rating systems would be regulated by a government agency. A quarter of those surveyed anticipate that a rating service would be held liable if its rating were incorrect and investors lost money.
These findings are a further demonstration that investors do not understand the nature and limitations of risk ratings, and suggest that the high level of confidence investors have about using them is based upon a series of erroneous assumptions. In addition, the findings call into question whether any disclosures accompanying the ratings would be effective to avoid investor confusion—if for no other reason than the sheer magnitude and variety of the disclosures that would be required for this purpose. The effectiveness of any such detailed disclosure is questionable at best, considering that investors did not even understand the relatively straightforward disclosures contained in the sales literature used for the survey.
The findings of the survey represent the perspectives of the mutual fund investing public on a regulatory issue of great significance before NASDR. These findings provide compelling evidence that NASDR should not reverse its long-standing position and permit the use of bond fund risk ratings in sales material. If it does so in the spirit that "more disclosure" is "better disclosure," investors will not be helped—instead, a great many of them will be harmed.
* * *
The Institute appreciates the opportunity to submit the results of our research focusing on investor use and understanding of bond fund risk ratings. If you have any questions regarding these items or related issues, please contact the undersigned at 202/326-5810, Craig Tyle at 202/326-5815, or Amy Lancellotta at 202/326-5824.
Paul Schott Stevens
Senior Vice President and
cc:NASDR Board of Directors
Mary L. Schapiro, President
NASD Regulation, Inc.
Elise B. Walter, Executive Vice President
NASD Regulation, Inc.
R. Clark Hooper, Senior Vice President
NASD Regulation, Inc.
John M. Ramsey, Deputy General Counsel
NASD Regulation Inc.
Joseph E. Price, Counsel
NASD Regulation, Inc.
1 NASD Regulatory and Compliance Alert, Vol. 11, No. 1 (March 1997) at 3.
2 The Investment Company Institute ("Institute" or "ICI") is the national association of the American investment company industry. Its membership includes 6,642 open-end investment companies ("mutual funds"), 443 closed-end investment companies, and 11 sponsors of unit investment trusts. Its mutual fund members have assets of about $4.206 trillion, accounting for approximately 95% of total industry assets, and have over 59 million individual shareholders.
3 In published research reports, the Institute has devoted extensive attention to the characteristics and behavior of fund investors as well as risk and other disclosure issues of importance to them. See, e.g., Understanding Shareholders’ Use of Information and Advisers (Spring 1997), Mutual Fund Shareholders: People Behind the Growth (Spring 1996), The Profile Prospectus: An Assessment by Mutual Fund Shareholders (Spring 1996), Shareholder Assessment of Risk Disclosure Methods (Spring 1996), and Profiles of First-time Mutual Fund Buyers (Fall 1994).
4 See supra note 1.
5 Keynote Address, Matthew P. Fink, President, Investment Company Institute, at the Annual FBA/ICI Education Foundation Mutual Funds and Investment Management Conference, Palm Springs, California, March 17, 1997.
6 Seventy-seven percent of survey participants owned equity funds, 38% owned balanced or mixed-income funds, 34% owned money market funds, and 30% owned bond funds. Almost 60% of respondents had never owned bond funds at the time of the survey and 12% were previous owners of bond funds.
7 Two versions of the sales literature were prepared. One version used an alpha-numeric designation ("R-1") and the other used a brief descriptive phrase ("extremely low"). A copy of the sales literature can be found on pp. 4-5 of the Report.
8 Letter from Paul Schott Stevens, Senior Vice President and General Counsel, to Joan Conley, Office of the Corporate Secretary, dated February 24, 1997, responding to NASDR’s request for comment on the use of bond fund risk ratings in supplemental sales literature. See NASD Notice to Members 96-84 (December 1996).
9 Based on their responses to certain items, survey participants were categorized based on their level of understanding (i.e., low, moderate, or high) of bond fund investing.
10 Respondents who indicated that they reviewed risk before making a purchase decision were asked to describe the importance they would attach to the risk rating relative to other items of risk information, such as annual total return of the fund for each of the last ten years, a written description of the fund’s risks, the 1-, 5- and 10-year total returns of the fund, the Morningstar ranking for the fund, and the risks of the types of securities held by the fund.
11 This possibility was noted in comments of the Consumer Federation of America: "To the degree that [risk ratings] further the notion that there is a single, reliable measurement of investment risk, such a rating could actually diminish investor understanding of this important concept." Submission of Consumer Federation of America (February 24, 1997) (responding to NASD Notice to Members 96-84, December 1996, soliciting comments on the use of bond fund risk ratings in supplemental sales literature).
12 Respondents were presented with a list of eight possible applications of the risk rating and asked to indicate those that represent appropriate uses for the ratings. These applications were: (1) the effect an increase in interest rates would have on the fund’s performance relative to other rated funds; (2) the past performance of the fund; (3) the effect of the stock market on the performance of the fund; (4) whether the fund is likely to produce above-average gains; (5) the likelihood of losing money in the fund; (6) the quality of the fund’s management; (7) the appropriateness of the fund for the respondent relative to funds with no ratings; and (8) the fund’s appropriateness for most investors relative to funds with no risk ratings.
It is noteworthy that respondents displayed this lack of understanding despite the fact that the sales literature clearly stated in two places that the risk rating reflected the sensitivity of the fund’s net asset value and total return to changes in interest rates. One piece of the simulated sales literature stated, "The risk of the Fund has been rated as ‘R-1’ by the XYZ Corporation, a nationally recognized organization that rates the sensitivity of net asset values and returns of bond funds to changes in interest rates and other market conditions," and "The ‘R-1’ risk rating indicates the XYZ Corporation views the Fund’s net asset value and total return as having minimal sensitivity to fluctuations in interest rates and market conditions." The other piece, which used a narrative rating rather than an alphanumeric one, used identical language.
13 This month’s issue of Consumer Reports advises investors considering a bond fund investment to consider carefully their own investment needs and time horizons before selecting a fund. See "Who needs a bond fund?" Consumer Reports (Sept. 1997).
14 Survey participants were asked to indicate the likelihood that: (1) the ratings would be available for all bond funds; (2) all ratings (including "high risk" ratings) would be made publicly available; (3) the rating agencies would be subject to government regulation; (4) the rating methodologies would be subject to third-party review; (5) the rating agencies would be liable for inaccurate ratings; (6) the ratings would allow investors to compare different funds; (7) the ratings would be calculated solely from statistical data; and (8) the same methodology would be applied to all funds.