Securities Industry Association

Federal Regulation Committee

March 10, 2003

Margaret H. McFarland
Deputy Secretary, United States Securities and Exchange Commission
450 Fifth Street NW
Washington, DC 20549-0609

Re: Proposed Rule Changes of New York Stock Exchange and National Association of Securities Dealers Relating to Research Analyst Conflicts of Interest, File Nos. SR-NYSE-2002-49, SR-NASD-2002-154

Dear Ms. McFarland:

Thank you for giving the Federal Regulation Committee (the "Committee") of the Securities Industry Association ("SIA")1 the opportunity to comment on the above-referenced proposals to further strengthen regulatory protections to ensure the objectivity of securities analysts employed by U.S. broker-dealers. The New York Stock Exchange ("NYSE") and the National Association of Securities Dealers ("NASD") (collectively the "SROs) adopted sweeping new rules in this area last year in an effort to answer questions that had been raised about conflicts of interest that might impair analyst performance. Subject to addressing a number of concerns that we detail below, we believe that these proposed amendments will further strengthen public trust and confidence in broker-dealer research.

As noted throughout our letter, there are a considerable number of inconsistencies and ambiguities in the two sets of SRO rules. This suggests the merit, either now or at some later date, of replacing the SRO rules with a single Securities and Exchange Commission ("SEC") rule. Such a rule would be a stronger signal of a national commitment to addressing investor concerns about research independence, and would make a single regulator accountable for the efficacy of its rule. It would also avoid the complications and unintended consequences that arise when two sets of regulators try to apply two sets of similar but distinct rules to the same subject. The SEC would strengthen its own leadership, both domestically and globally, if it adopted one comprehensive set of rules on this subject.

Overview and Summary.

The proposed rules, while in some respects constructive additions to the rules adopted last year, contain provisions that if adopted will unintentionally weaken investor protections, and other provisions that are unnecessarily problematic or vague. We want to highlight two concerns with provisions in the rules that may prove detrimental to investor confidence. These are:

  • ambiguity as to whether the "due diligence" that analysts are permitted to undertake regarding investment banking candidates encompasses vetting of underwriting candidates prior to the analyst's firm receiving an investment banking mandate (this ambiguity threatens a key protection for the public markets) (see pages 5-6 below), and

  • imposing a virtual ban on research supervisors from owning equity securities, which will discourage the most qualified people from wanting to oversee research, including members of research independence committees that state and federal regulators, including the NASD and NYSE, sought as part of a settlement of major enforcement matters. (see pages 6-8 below).

In addition, we have a number of significant concerns with drafting issues and apparent contradictions in the proposals. Some of the problems that we have identified could frustrate the rules' purpose by making them so vague or confusing as to be largely ineffectual. These concerns include:

  • The NYSE proposes to expand the definition of the scope of the persons covered so broadly that its rules would potentially apply to a huge range of personnel outside firms' research departments. This creates enormous complications, most of which we doubt the NYSE intended, as well as putting the NYSE proposal in conflict with the NASD proposal (see pages 9-10 below);

  • The proposed bar on analysts' solicitation of investment banking business, coupled with the uncertain scope of the "due diligence" exemption, is so vague that it will inhibit normal analyst communications with private companies in the industry that the analyst follows. This will inevitably weaken analysts' understanding of the industries that they cover, thereby harming investors who use that research. The proposed bar is also unworkable insofar as it relies on a triggering event, the signing of a letter of intent, that does not reflect common practice in the underwriting process (see pages 5-6 and 10-14 below); and

  • The approach to analysts' public appearances , in the case of the NYSE proposal, inappropriately attempts to indirectly regulate news media editorial discretion, and in the case of both proposals injects unnecessary confusion into the rules on this topic. (see pages 17-20 below).

To respond to these concerns, we have attached as Appendix A to this letter a summary of specific recommendations for changes to the proposed rules.

We also are concerned that the proposed amendments come at a time when other significant changes in regulation of analysts are newly in place or in the offing. These changes have emanated from several quarters, and have unfolded in an incremental and somewhat uncoordinated fashion. So far, we have seen (i) the first round of important SRO rules followed by written guidance on those rules; (ii) the SEC's Regulation AC; (iii) the settlement of a state enforcement action against a major firm, followed by an initiative by various state treasurers to require broker-dealers and fund managers that do business with various state entities, particularly state pension funds, to comply with the principles of that settlement; and (iv) enactment of the Sarbanes-Oxley Act, which contains provisions that require additional regulatory steps. Looking ahead, further changes will be necessary before the end of July to implement additional requirements of the Sarbanes-Oxley Act that are not addressed by these proposed amendments. Regulators have stated that they intend to consider additional regulatory changes as a result of the announcement last December of an impending settlement of enforcement actions brought by a consortium of state and federal regulators against a number of broker-dealers concerning analyst conflicts of interest. Rather than adding to this series of incremental regulatory changes, each of which requires changes to procedures, staff training and other implementation costs, it would be more efficient and less confusing if the SROs and SEC could consolidate all impending regulatory changes into one package, with a single set of consistent final rules and effective dates.2

Finally, in adopting any amendments to their existing rules, the SROs should clarify how these rules impact their existing guidance in this area. They should also be prepared to offer additional guidance on issues in the existing rules that have not been sufficiently harmonized. We urge the SROs to address these, as well as any additional issues that may be identified once the amendments are adopted.

I. Comments on The Rule Proposals.

We have divided our comments below between those that go to core questions of investor confidence, and other significant questions about ambiguities, inconsistencies and other important drafting issues that could undermine the rules' purpose if not addressed.

A. Investor Protection Concerns.

Rule 2711(c)(4) and 472(b)(4) and Rule 2711(d)(2) and 472(h)(2): Role of Due Diligence. As currently drafted, the proposed restriction on analysts accompanying investment bankers on "pitches" is written so broadly that it might restrict not only conduct that might be detrimental to investors' interests, but also conduct that clearly benefits investors. The proposed rule recognizes the important investor-protection role that analysts play in performing "due diligence" to ensure that the underwriters as well as investors fully understand the financial condition and business operations of the company and how that company is positioned within its particular industry. However, the rule should clearly state that such "due diligence" includes reviews by analysts of the issuer both prior to and after receiving an investment banking mandate to determine if the transaction is in the interests of investors. Analysts play an important role in investor protection by helping their firms to make informed decisions at an early stage about investment banking transactions, especially decisions to help take a company public.3 A ban on contacts between analysts and investment banking candidates that are intended to assist the firm's commitment committee runs counter to investors' interests.

Likewise, the ban in proposed Rule 2711(d)(2) and 472(h)(2) on consideration by a compensation review committee of "contributions to the . . . [firm's] investment banking business" should not preclude considering such contributions to the extent that they benefit investors, such as in screening potential candidates for initial public offerings ("IPOs"). Investors benefit when the caliber of IPO offerings is enhanced as a result of analysts ably performing this role. Consistent with the proposed rule on analyst involvement in certain investment banking "pitches", firms should be permitted to reward analysts for helping to ensure that only the most deserving potential issuers are able to access the public markets.

Rule 2711(a)(5) and 472.40's Extension of Trading Restrictions. We are very concerned by the proposal to extend the trading restrictions from analysts to directors of research, supervisory analysts, members of committees or other individuals who oversee analyst independence and recommendations (collectively, "Senior Research Management"). Unlike other aspects of the SROs' research rules, this proposal has been advanced with no explanation, apart from brief conclusory remarks, as to why the SROs think that such an expansion of the trading restrictions is necessary. In its current form this proposal will discourage the most qualified people from wanting to serve in these positions, each of which is critical to strengthening the integrity of research. Because at some firms hundreds of analysts, covering thousands of securities, can report to Senior Research Management, extending the existing trading blackouts, 30-day look-back restrictions on prior trading, and restrictions on private fund holdings to these personnel will make it virtually impossible for Senior Research Management, as well as their household members, to own any investment instruments other than registered diversified mutual funds in their personal portfolios. In contrast, under the current SRO rules, in most instances, research analysts are only restricted with regard to issuers that they cover.4

A de facto ban on owning equity securities is harsh, and is likely to discourage the best people from serving on committees that oversee research independence. We note that establishment of such committees is a prominent feature of proposed settlements of enforcement actions involving analyst conflicts, as described in public statements by the SEC, NYSE, NASD and state regulators, and in "investment protection principles" that have been subsequently adopted by certain States, public employee service providers and asset managers. The goal of the enforcement staffs of these agencies seem to be at odds with this provision of the proposed rule.

The purpose of the rules is weakened by extending the full panoply of trading restrictions to Senior Research Management. Instead, we recommend that the SROs consider a requirement that firms set up an internal written pre-transaction approval and/or post-trade monitoring process reasonably designed to ensure that such transactions do not create a conflict of interest between the professional responsibilities and any personal trading activities of Senior Research Management. Additionally, the SROs could require an internal system to require disclosure of personal holdings to research compliance and senior management5 so that they can make judgments about whether potential conflicts require recusal from matters such as approving rating changes.6

In any event, managed accounts not controlled by the account owner should be excepted from the trading restrictions, for both analysts and supervisors. This would eliminate the de facto ban on owning equity securities. NYSE Rule 472(e)(4)(v) currently seems to exempt such accounts from the trading restrictions for research analysts, but NASD Rule 2711(g)(5) does not. These rules should be harmonized to exempt managed accounts from trading restrictions, since they do not pose the risks that those restrictions are intended to address. If necessary, these accounts could be made "blind," so that the owner is unaware of the securities held in the account.

Finally, whether or not these suggested changes are made, at a minimum the rule should provide a reasonable phase-in period to enable Senior Research Management to restructure their personal portfolios.

B. Critical Drafting and Interpretive Concerns.

Below we highlight a number of important and problematic issues and ambiguities in the rule, and suggest drafting and interpretive solutions. Regardless of whether our particular suggestions are accepted, the final rule must eliminate these uncertainties so that research analysts and broker-dealers will understand what the rule requires of them and not have to parse the differences between the rules. The rules will be far less effective if broker-dealers have to guess at their meaning. Moreover, it is important to investors that the SRO rules are clear and consistent so that the protections that an investor receives are not dependent on whether the broker-dealer offering the investor research happens to be a member of one SRO or another.

NYSE Rule 472.40 and NASD Rule 2711(a)(5): definition of research analyst. The NYSE proposes to expand its definition of the "associated persons" covered by the analyst rules to, among other things, include anyone "making recommendations or offering opinions in public appearances." This change contradicts not only the NASD rule, which focuses on persons directly or indirectly responsible for the "preparation of the substance of a research report," but also the Joint Interpretive Memorandum, which states the rules "do not include every registered person who may express an opinion on an equity security. Thus for example, the terms exclude registered representatives who recommend securities to their customers, . . . [or] investment advisers" except to the extent these individuals prepare the substance of research reports."7

The NYSE proposal disconnects its rule from both equity securities and research reports. This is a serious mistake. By divorcing the rule from the preparation of research reports, the proposal potentially sweeps in a wide range of traders, sales personnel and other employees who work outside the research department. This leads to a wide range of complications which the NYSE could not have intended (e.g., an observation at a cocktail party about general market or economic conditions leads to trading restrictions on the speaker and his or her household members). It also seems unlikely that the NYSE wanted the proposed compensation review committee to review and approve not just the compensation of research analysts, but also the compensation of any person outside the research department who makes a recommendation or offers an opinion at a public appearance.

The NYSE proposal goes far beyond the concerns that motivated both regulatory and legislative action in this area.8 We recommend that the NYSE rule be made consistent with the NASD's definition, or that this element of the definition should be tied to the term "subject company's equity securities" like the other elements of the definition.

Rule 2711(c)(4) and 472(b)(4): Ban on IPO "Pitches." As noted at pages 5-6 above, as proposed this rule could be harmful to investors' interests. The SROs should clarify that they do not intend for these rules to bar analysts from communicating with private companies for the purpose of screening potential IPO candidates. In addition to that broad concern, this rule needs to be clarified in several respects in order to be effective and to avoid impairing analysts' ability to effectively serve investors. We suggest simplifying the rule by replacing the "letter of intent or other written agreement" limitation, and the confusing language about communications "in furtherance of obtaining investment banking business," with a simple bar on publishing research for a defined period of time if an analyst solicited an IPO offering mandate.

"letter of intent or other written agreement." The signing of a letter of intent or similar document is not a realistic measurement of the point in time at which a broker-dealer has begun to act as an underwriter in an IPO. It is not common practice to sign a written agreement evidencing the mandate other than the underwriting agreement itself - a document executed long after the firm has begun acting in an underwriting capacity.9 The SEC has recognized this in other rules. Rather than using the signing of a letter of intent as the trigger date before which the restrictions on analyst contact with the issuer apply, it would be better to simply prohibit analysts from communicating with issuers for the purpose of soliciting IPO business. There is no need to frame this prohibition around the signing of a letter of intent. The rule need only state that the analysts are prohibited from issuing research reports or making public appearances for some defined time if the analyst has engaged in any communication with the subject company "for the purpose of soliciting an IPO mandate." If an analyst is aware that a mandate has already been awarded to his or her firm, then any subsequent communication will, by definition, be for some purpose other than soliciting the mandate.

"in furtherance of obtaining investment banking business...." It is unclear whether the term "investment banking business" is intended to only pick up initial public offering transactions or other investment banking transactions, particularly private merger and acquisition deals. In view of the proposed rule's express ban on analysts' providing coverage if they helped to solicit an IPO underwriting mandate, it is puzzling that the SROs chose the term "investment banking services," which is defined much more broadly than that under their rules.10 The reasoning is not apparent for barring research coverage simply because the firm asked an analyst to meet with a private client on a nonpublic investment banking transaction that was unrelated to a subsequent IPO that might not have occurred until many years later. For example, if an analyst is involved in a solicitation of a venture capital transaction a decade prior to the broker-dealer becoming an underwriter for the issuer's IPO, the research restriction seemingly would apply even though the analyst had no involvement in obtaining that business. This is an arbitrary result that is unrelated to the true purpose of the rule as we understand it, to ban analysts from pitching IPOs of issuers that they will subsequently cover.

The phrase "in furtherance of" also poses fundamental problems that would reduce the quality of research received by investors. This term might pick up communications that are not made with a specific relationship in mind, but that may fortuitously or inadvertently lead to a later investment banking relationship. An analyst's participation in an industry conference, or even an analyst's casual encounter with a prospective investment banking client of the firm might be viewed as being "in furtherance of obtaining investment banking business." Firms might be loathe to permit analysts to engage in any activity that a regulator, with hindsight, might view as acting "in furtherance of investment banking activities" because separate and apart from their efforts, the firm might, now or in the future, pursue an investment banking relationship with the company. This concern, coupled with the concern noted at pages 5-6 above about the uncertain scope of the "due diligence" exemption, will impede analysts' ability to actively communicate with any private company in an industry that the analyst covers. Without the ability to speak to such companies, the analyst's insight into the industry will surely suffer, and the quality of research offered to investors will deteriorate.

The alternative language that we propose, barring analysts from communicating with issuers "for the purpose of soliciting an IPO mandate," would address these concerns. Wording such as this should connote that the rule does not apply to all "communications . . .in furtherance" of investment banking business (which may be made by the analyst in the ordinary course of visiting companies in the industry he or she covers, outside the presence of investment banking personnel, but which may unintentionally happen to further investment banking objectives), but communications between the analyst and the company that are related to an ongoing investment banking pitch for IPO business. Alternatively, a discussion in the adopting release explaining that "in furtherance" is limited in this manner would be helpful.

Some of these concerns would be mitigated if the rule contained a time limit on the ban on coverage. An analyst's involvement in communications about a private investment banking transaction, or situation where the analyst says or does something that could be interpreted as a solicitation for IPO business, should not result in the firm forever losing the ability to provide research coverage for an IPO.11 We suggest that the rule should limit the bar on research to situations where the analyst was involved in a solicitation of an IPO underwriting engagement during the period 180 days prior to the filing of the IPO. We believe that such a 180 period is consistent with the view taken (and already being applied) by the NASD in its proposed amendments to Rule 2710, in which acquisitions of equity securities of an issuer (or receipt of other items of value) by an underwriter prior to the 180-day period preceding the filing date of the IPO are not considered to be underwriting compensation. An analyst visit to a company prior to the 180-day period preceding the IPO filing date should, similarly, not be viewed as having been made for the purpose of soliciting an IPO mandate. Moreover, we suggest that the bar on publishing research should not continue in perpetuity, but should be limited to a fixed time period after the IPO.12

Retroactive and Residual Effect. This rule should not apply retroactively to activities that were not covered prior to its adoption. In other words, it should not prevent analysts from initiating coverage or force analysts to drop coverage due to activities that were not proscribed when they occurred. If the rule is applied retroactively, research coverage by many U.S. broker-dealers would come to a halt while firms try to reconstruct whether or not the rule bars coverage due to activities that were permitted at the time.

Subject to one exception, the restriction on research coverage likewise should not have a residual effect on an analyst when he or she moves to a new firm. If an analyst participates in a solicitation of IPO business, and then moves to a new firm, he or she should not be barred from providing research coverage due to his former firm obtaining an IPO mandate from that company. We would, however, support applying the restriction in a case where the analyst participates in a solicitation for IPO business, and then moves to a different firm which receives that company's IPO business within 180 days of hiring the analyst.

"Use or promise of research" vs. "Pitch Investment Banking Services." We have a concern about a conflict between the NYSE and NASD Statements of Purpose regarding analyst participation in soliciting investment banking business. The NYSE Statement of Purpose indicates that the provision is intended to "prevent the use or promise of research as an influence or a sales and marketing tool with prospective investment banking clients of the member organization, and would cause subject companies to choose a prospective investment banking firm based on the merits of its underwriting capabilities, rather than its research coverage."13 The NASD Statement of Purpose indicates that the purpose of the provision is to "prevent analysts from attending "bake-off" meetings or otherwise communicating where the intention is to pitch the member's investment banking services."14

The language in the NYSE statement contradicts prior guidance provided in the NYSE/NASD Joint Interpretive Memorandum, in which the SROs emphasized that the rules "do not prevent a member from agreeing to provide research as long as a part of its investment banking agreement with a subject company, so long as there is not promise of favorable research."15 The Joint Interpretive Memorandum appropriately recognizes that the focus of the rules should not be on preventing investment bankers from promising research coverage, a matter in which a prospective issuer has a legitimate interest as long as there is no promise of positive coverage. The NASD statement of purpose on this point is better focused on analyst-issuer communications and the intent of the analyst's communication. The NASD's statement is also more faithful to the actual text of the rules and the proposed amendments, which do not prohibit promises of research coverage.

The NASD's approach also better accommodates the interest of the investing public in general, and shareholders and management of an issuer in particular, in knowing that the issuer's stock will receive sufficient coverage (whether favorable or not) in the aftermarket. The integrity and credibility of an investment bank's research product, and not just the strength of its underwriting abilities, is an important and appropriate factor to an issuer in selecting an underwriter.

By suggesting that investment banks should be allowed to solicit IPO business solely on the basis of their underwriting capability, the NYSE weakens the ability of firms to compete for underwriting business based factors such as its reputation for research integrity. There is no conflict of interest if an investment banker highlights, for example, that the firm has the most respected analyst in the relevant industry. The proposed rule clearly prohibits the analyst from being a party to the conversation, and also prohibits any express or implied promise of favorable research. Extending this restriction so that investment banks can compete for IPO mandates based solely "on the merits of their underwriting capabilities" will needlessly detract from the ability of small and mid-sized broker-dealers to compete for IPO business.

We urge the SROs to resolve these contradictory statements by confirming the NASD's stated rationale.

Rule 2711 (a)(4) and 472.50: Public Appearances. The proposed changes to the rules concerning public appearances contain several significant flaws that are likely to greatly impede useful analyst communications with the public.

Regulation of Media Outlet Editorial Discretion. We support the NASD's statement in its proposing release that it is modifying its current guidance so that the rule concerning disclosures required in public appearances does not obligate an analyst to boycott further communications with a media outlet if the outlet does not print or broadcast disclosure information specified in the rule, as long as the analyst has made the required disclosures to the media outlet. The NYSE proposal goes in the opposite direction, by continuing to require such a boycott, and by expressly extending the obligation to the print media. We agree with the NASD that the regulation should be focused on having the analyst make the required disclosures. To further require that the analyst cut off communications with a media outlet that exercised its editorial discretion not to carry some or all of the disclosures amounts to inappropriate indirect regulation of the news media.

Discrepancy Concerning "Industry" Recommendations and Opinions. The NYSE, without any discussion of the change, currently proposes to amend the definition of "public appearance" to create an inconsistency with the corresponding NASD rule. The NYSE seeks to expand the term "public appearance" in NYSE Rule 472.50 to cover communications in which an associated person makes a recommendation or offers an opinion "concerning any equity security and/or industries." (emphasis added). The definition of "public appearance" in NASD Rule 2711(a)(4) lacks that new language.

If it is permitted to remain, this discrepancy will place analysts associated with NYSE-regulated firms at an unfair disadvantage compared with analysts not associated with such firms. While this is surely not the NYSE's intention, read literally this language would compel an analyst who gives an "opinion" concerning an industry as a whole to make disclosures concerning every company in that industry. Where a firm has an affiliate engaged in asset management, NYSE Rule 472(k)(1), as amended, could require the analyst to read off the names of potentially hundreds of stocks that the affiliate has in accounts it manages.

This language creates further complications when coupled with proposed new NYSE Rule 472(l) and 472.40, which expose anyone at the firm, not just research analysts, to these disclosure obligations if they discuss specific securities or industries in a public appearance, or are deemed to be "making recommendations or offering opinions" in a public appearance. In addition, analysts and other employees swept up in the proposed rules may well have different understandings about the scope of an "industry" and whether certain issuers fall within that scope. As a result, they may be uncertain as to what disclosures they are obligated to make

In the Joint Memorandum the NASD and NYSE stated that they would not consider "research reports" to include "reports that recommend increasing or decreasing holdings in particular industries or sectors but that do not contain recommendations or ratings for individual securities." It would be odd to mandate disclosure of the type proposed by the NYSE in the context of a public appearance, where an analyst generally has less time to communicate and is more subject to the whims of the media's editorial discretion, than in the context of a written research report prepared by the firm.

Opinion Concerning an Equity Security. The SROs should also use the opportunity of these rule amendments to replace the phrase "opinion concerning an equity security" in the definition of "public appearance." The inclusion of that phrase makes this definition ambiguous and at variance with the rules' approach in other areas, as well as the SEC's definition of "public appearance" in Regulation AC. This phrase raises questions about what exactly is an "opinion," as opposed to a statement of fact, and the difference between an opinion related to an issuer and an opinion "concerning an equity security."16 In contrast, in the definition of "research report," current SRO rules require an analyst to make disclosure of potential conflicts of interest only when the report "provides information reasonably sufficient to base an investment decision." The SEC's definition of "public appearance" in Regulation AC contains that same limitation. We recommend that the ambiguities discussed above be removed by deleting "opinion concerning an equity security" in the definition of "public appearance" and replacing it with the language from the definition of "research report," so that a "public appearance" would cover events at which "a research analyst makes a recommendation or provides information reasonably sufficient upon which to base an investment decision on an equity security." This change would make the SROs' definition consistent with their definition of "research report" and the SEC's definition of "public appearance" in Regulation AC.

NASD Rule 2711(f)(4) and NYSE Rule 472(f)(5): Withdrawal of Coverage. This provision is a sensible extension of current rules, and we support it. However, there are two respects in which the proposal needs to be clarified.

Definition of "Withdrawal." Exactly what constitutes "withdrawal" is unclear. The NASD rule uses the term "discontinue" and "withdrawal" interchangeably, but the two terms do not necessarily mean the same thing. A related issue is how a firm is to determine when its obligation to provide notice begins. Suppose that a firm offers coverage of a company only sporadically, or is reassessing whether to continue offering research on that issuer. Is there a point where its inaction constitutes a decision to "withdraw," even though it has not made a decision whether or not to abandon coverage? Suppose that a firm issues quarterly research reports on an issuer, but for some reason skips one quarterly report, or temporarily withholds research as it assesses, based on significant events impacting the company, whether to continue offering research on the company. These situations should not constitute a "withdrawal" of coverage, but the proposed rules do not make that clear.

We recommend that the SROs offer a definition of "withdrawal" as a decision by the firm or its head of research that the firm's applicable research service will no longer include coverage of a particular issuer's equity securities.

"made in the same manner." Read literally, the proposed rule amendments would require that a termination report be issued on paper, even if the report has been offered only electronically for many years, if coverage was initiated via a printed report decades ago. We assume this literal reading is not the SROs' intent, and that notice of withdrawal should be via the medium or mediums in which the penultimate report was circulated. We also understand the rules not to require a substantive final report, but simply a statement that the firm is withdrawing its research, repeating the last rating on the security, together with a statement that investors should not continue to rely on that rating. Confirmation of these points in the final rule amendments would be helpful.

NASD Rules 1050, 1120, and NYSE Rules 344 and 345A, Qualification of Analysts. We support requiring a qualification examination and continuing education requirement for securities analysts and supervisory analysts. The following accommodations and clarifications should be made:

    (i) There should be a delayed effective date for analysts to prepare for and take the qualifying examination. The NYSE indicates that it plans to do this. However, the SROs may also want to impose the requirement in stages, so that they are not inundated by thousands of securities analysts trying to take the examination at the same time. For example, they could require that analysts with social security numbers ending in 0-3 must be qualified by x date, those with social security numbers ending in 4-6 must be qualified 60 days later, etc.

    (ii) The SROs should give comity for the qualification examination to the CFA Level One qualification. In addition, we recommend that comity should be granted for personnel who, on the date the rules become effective, have been principally responsible for the preparation of the substance of research reports for three or more years.

    (iii) The SROs should clarify that these requirements, like the rest of the SRO analyst rules, do not extend to fixed income analysts.

    (iv) The NYSE proposal would require firms to provide continuing education to both analysts and supervisory analysts. In contrast, the NASD proposal mandates both firm and regulatory elements of continuing education, but only for analysts. We recommend that this discrepancy be resolved in favor of the NYSE's proposal, permitting both analysts and supervisory analysts to complete their continuing education requirements through their firms. Because analysts who are employed at firms that engage in investment banking or asset management have obligations different from than those at firms that do not, firms can tailor continuing education programs to their analysts' needs better and more easily than the SROs.

    (v) As noted above, under the current SRO rules research analysts generally include those persons who are responsible for the preparation of research reports as well as individuals who report directly or indirectly to research analysts in connection with such preparation. The SRO proposals however appear to require only those persons who are directly responsible for the preparation of research reports to register with the applicable SRO and be required to pass the qualification exam. We request clarification that the registration and qualification examination apply only to the analysts primarily responsible for the content of the research report.17

NASD Rule 2711(f)(3) and NYSE Rule 472(f)(4): Lock-Ups While we support the concept of restricting the issuance of research around the expiration of lock-up agreements, the inclusion of waivers as one of the conditions triggering a freeze on research creates significant problems that the SROs do not appear to have recognized. Co-managing underwriters often have no knowledge of lock-up waivers granted by the lead manager, and therefore will never be certain whether they can publish research during the 180-day lockup period. Lead managers may also find this troubling. Since they do not have the ability to control the publication of co-manager's research, they will not have any certainty of whether an early lock-up waiver will end up causing a violation by an analyst for a co-manager who unwittingly publishes research within the 15-day blackout window.

At a minimum, the application of this restriction to lock-up waivers will raise difficult compliance issues and pose a trap for the unwary. The practical effect may be to dissuade issuance of lock-up waivers prior to their normal expiration time. This is undesirable and an unnecessary interference with private contractual commitments. A lock-up waiver is often a commercially prudent step.18 Partial lock-up waivers are sometimes granted to investors due to special circumstances, such as the need by the investor to meet charitable commitments, to exercise obligations under family trusts, or because of personal hardship.

While the SROs' reasons for wanting to restrict research around the time of lockup expirations are understandable, extending these restrictions to lockup waivers creates unnecessary problems. The SROs should avoid an approach that will have the practical effect of automatically disfavoring waivers. An alternative approach that would address the SROs concerns with fewer complications would be to simply bar firms or their analysts from issuing research reports for the purpose, in whole or in part, of affecting the price of the issuers' securities for the benefit of a selling shareholder.

NASD Rule 2711(a)(8) and NYSE Rule 472.10(2). Definition of "Research Report." SIA understands that the Sarbanes-Oxley Act arguably compels the SROs to remove the recommendation element from their definitions of the term "research report" because the Act's definition of the term omits specific reference to a "recommendation." 19 However, it is far from clear that this alteration in the wording of the definition was intended by Congress to create a substantive change in the scope of the term as used in the current SRO definition. Unfortunately, the legislative history is silent on this point. However, the structure of the Act suggests no intention to require a substantive change by regulators in the scope of the term "research report." The phrase "equity analysis [that is] sufficient to base an investment decision"20 used by the Act appears to be on its face tantamount to a recommendation. Moreover, the context in which the term "research report" is used in the Act strongly implies that Congress viewed the term "recommendation" as being synonymous with "equity analysis sufficient to base an investment decision," suggesting that Congress dropped the term "recommendation" simply because it was redundant. The Act directs the Commission or SROs to adopt rules "reasonably designed to address conflicts of interest" that can arise when securities analysts recommend equity securities in research reports and public appearances . . . ."21

For this reason, while we do not oppose the deletion of the term "recommendation," we urge the SROs not to construe it as a substantive change compelled by the Sarbanes-Oxley Act. We note that if the SROs were to view the revised definition as a substantive change, especially if they were to view the revision as sweeping in communications that are factual and objective in nature and that do not contain a subjective view or conclusion, a vast range of communications that are not currently subject to the rule would be swept in, with consequences that are too numerous and complex to enumerate here.

NASD Rule 2711(d)(2) and NYSE Rule 472(h)(2): Compensation Review Committee.

As noted at page 6 above, the broad proscription against considering any "contributions to the member's investment banking business" could run counter to investors' interests because it would eliminate incentives for analysts to play their critical role in screening out clients that are not suitable for entry into the public markets. This provision is also unclear about exactly what the role of this committee is supposed to be. The rule states at the outset that the committee's role is to review and approve analysts' compensation. However, it goes on to state that the committee shall not consider contributions to investment banking in "determining the research analyst's compensation." Determining compensation is a very different function than reviewing and approving compensation. We recommend replacing the word "determining" with the word "reviewing," or simply deleting this sentence.

In addition, due to the expanded definition of the term "research report," the compensation committee could be faced with the enormous burden of having to review the compensation of a much wider range or personnel, including many both within and outside the research department who may play a secondary role in preparing the report (e.g., collecting data, fact-checking, etc.), but who do not determine a research report's substantive content or conclusions. In order to keep the committee's responsibilities better focused on the core concern, we recommend that, as with the Commission's recently-adopted Regulation AC, the compensation committee's responsibilities should be focused on analysts who are "principally responsible" for producing research reports.

NYSE Rule 472(l): Other Communications Activities. As proposed, this rule would require a non-analyst discussing an equity security in a public appearance to make all of the public appearance disclosures required by analysts, plus an additional disclosure as to whether his firm makes a market in the security. It is incongruous to require personnel who are not analysts to make a disclosure that analysts do not have to make in their public appearances. Moreover, the SROs should be cautious about expanding the required disclosures in public appearances too far, since each layer of disclosure required to be recited risks making the overall disclosures seem more like boilerplate to listeners or viewers. We recommend that the NYSE tailor this provision to address this concern.

The new requirement that all external communications activities, even those that contain no discussion of a security or industry, must be pre-approved could be quite burdensome, especially in areas like corporate communications, personnel recruiting, or governmental affairs. This seems quite remote from any concern about analyst conflicts of interest, and we recommend that the NYSE reconsider this requirement.

II. Effective dates.

At least one more round of rule changes besides these proposed rules will be necessary under the terms of the Sarbanes-Oxley Act. Regulators have also stated that they may propose rules drawing from some of the terms of the recently-announced settlement of enforcement actions against several major firms relating to analyst conflicts of interest. We urge the SEC to withhold approval of these proposed SRO amendments for a fairly brief time, so that it can coordinate the implementation of all applicable SRO or Commission requirements so that they can all be phased in as one package, rather than imposing the needless burden on firms of making systems changes and staff training in response to one set of regulatory changes, only to have to go through it again for a separate set of changes on the same or related issues a few months later.

We also note a potential discrepancy between the SROs regarding their plans for implementing these rule changes. The NYSE would require compliance with all of its changes except the qualification/ continuing education within 60 days of their approval by the SEC. The NASD has said that it will announce the effective date of its proposed changes in a notice to members no later than 60 days after SEC approval, and that the effective date will be at least 30 days after the notice. We urge the two SROs to coordinate their effective dates and ensure that firms have adequate time to implement new rules.

III. Other Issues.

A. Status of Current Interpretive Guidance.

The written guidance on the current SRO rules works off of a definition of "research report" that will become outmoded under the proposed amendments (and under the Sarbanes-Oxley Act). For example, the guidance clarifies the treatment of "reports that recommend increasing or decreasing holdings in particular industries or sectors but that do not contain recommendations or ratings for individual securities." Is this guidance still valid in light of the amended definition of research report?22 As part of implementation of the amendments, the SROs should issue new written guidance on the issues addressed in the current guidance, or indicate the extent to which that guidance is still valid.

One of our concerns is that the proposed amendments could be viewed as overriding provisions of the Joint Interpretive Memorandum that exempt some categories of quantitative and technical analysis from the application of the current rules. We note that the Joint Interpretive Memorandum does not sufficiently addresses all of our concerns on this point. That guidance excludes from the definition of "research report," inter alia, (i) reports discussing broad-based indices that do not recommend or rate individual securities, (ii) technical analysis concerning the demand and supply for a sector, index or industry based on trading volume and price, or (iii) statistical summaries of multiple companies' financial data that do not contain any narrative discussion or analysis of individual companies' data.

While this is very helpful guidance, it does not cover all reports that are purely technical in nature and that do not contain any subjective judgments about an individual company. For example, a report that relies entirely on one or more quantitative models or indices to formulate recommendations on transactions in individual securities would not appear to be covered by this exemption, even though it does not pose any concerns about research conflicts impairing the recommendation's objectivity. We urge the SROs, in reaffirming their existing guidance or issuing new guidance, to consider modifications to the exemption for technical and quantitative analysis to better address these concerns.

B. Quantitative Research and Technical Analysis

The current SRO trading restriction rules have a disproportionate impact on certain analysts. Specifically, for analysts who are principally responsible for the preparation of research based on quantitative models or technical analysis the current SRO trading restriction rules pose significant difficulties. Quantitative research analysts and analyst that prepare technical analysis currently can be subject to trading restrictions so onerous that they are tantamount to a bar on owning equity securities. That is because the universe of stocks that their firm's quantitative research model or technical analysis covers or rates can run into the hundreds or thousands. While the Joint Interpretive Memorandum exempted some of this type of research, a great many quantitative and technical research reports are nevertheless subject to the current rules, as noted at page 29 above.

This current de facto bar on quantitative or technical analysts' owning equity securities is particularly harsh when these reports are not the source of concerns about analyst conflicts, and these types of research do not have the same potential for conflict of interest as does traditional, sector-based research attributed to individual research analysts. Unlike traditional research reports on specific issuers, these reports do not make subjective judgments about the economic prospects of individual companies, but instead are based on objective criteria, such as quantitative mathematical models, or comparison to publicly-known measures of stock performance such as price-earnings ratios or the S&P 500. Additionally, the securities mentioned in such research reports may turn over completely from one report to the next.

Similar to our recommendations on page 8 above with regard to personal trading by Senior Research Management, the SROs should provide guidance to allow firms to impose trade pre-approval and disclosure requirements for these analysts. We urge the SROs, in reaffirming their existing guidance or issuing new guidance, to consider modifications to the current SRO trading restrictions as they apply to technical and quantitative analysis to better address these concerns.

C. Addressing Discrepancies in Current Rules.

In conjunction with adopting amendments to the rules governing research, we urge the SROs to resolve inconsistencies in their current rules. While these discrepancies are not an urgent problem, the differences are a concern for all firms that want to be meticulous in following the rules. When firms are subject to two sets of rules on the same issue, the rules should be as consistent with each other as possible. Differences that should be resolved include inconsistencies regarding:

  • Disclosures of analysts' financial interests in a company that is the subject of a research report or public appearance. While NYSE Rule 472(k)(1)(i)(b) requires firms and analysts to disclose simply whether a financial interest exists, NASD Rule 2711(h)(1)(A) also requires disclosure of the nature of that interest (e.g., whether it is an option, right, warrant, future, long or short position);

  • Disclosures in research reports relating to firm compensation. While NYSE Rule 472(k)(1)(ii)(a) requires specific disclosure of public offerings of equity securities within the past 12 months, NASD Rule 2711(h)(2)(ii)(a) requires disclosure of public offerings generally--not just equity offerings--within the past 12 months;

  • Treatment of managed accounts not controlled by the account owner.. As noted above, NYSE Rule 472(e)(4)(v) appears to exempt such accounts from the trading restrictions for research analysts, while NASD Rule 2711(g)(5) does not. Since these are accounts in which by definition the research analyst does not control trading decisions, we think the NYSE exemption is appropriate, especially to mitigate the effect that the proposed extension of the trading restrictions would have on the ability to attract highly qualified personnel to oversee research independence;

  • NYSE Rule 472(e)(3) states that associated persons and their household members cannot trade contrary to the firm's most current recommendation, while NASD Rule 2711(g)(3) states that a research analyst or household member cannot trade inconsistent with the analyst's most recent recommendations. A ban on trading inconsistent with one's own recommendations is highly appropriate, but it is difficult to see why the ban should extend to recommendations that an analyst had no part in forming. While the Joint Interpretive Guidance supports the position that trade restrictions should be limited to the analyst's own recommendations, we understand that in a recent survey of its members the NYSE reverted to the language in its rule. To avoid further confusion, we recommend that the NYSE formally adopt the NASD's approach on this point; and

  • NYSE Rule 351(f) and NASD Rule 2711(i) require members to provide annual certifications concerning compliance with the respective SRO research rules. The SROs should clarify that firms must submit the required certification only with its designated examining authority, rather than to both SROs. Short of this, the SROs should have consistent filing deadlines, rather than the current December 31 deadline for NASD certification and April 1 filing deadline for the NYSE certification.


Thank you for giving SIA this opportunity to comment on the NASD and NYSE proposed amendments regarding research analyst conflicts. As detailed above, the proposed SRO rule amendments contain numerous inconsistencies between each other and between themselves and the SEC's Regulation AC. The rules also contain significant ambiguities that will make compliance difficult and will lead to firms taking divergent approaches. Rather than continue down the road of regulating this issue of national importance through conflicting and confusing rules issued by multiple regulators, investors and the capital markets of the United States would be better served if the SEC took full command of research regulation, replacing the SRO rules with a comprehensive SEC rule.

Nevertheless, we believe that with the modifications that we suggest these proposals can help to improve public trust and confidence in research analysts. If you have any questions on any aspect of this letter, please contact George R. Kramer, staff adviser to the Committee, at 202-296-9410. or by e-mail to


Robert C. Dinerstein, Chairman
SIA Federal Regulation Committee

Cc: Chairman William H. Donaldson, U.S. Securities and Exchange Commission
Commissioner Paul S. Atkins, U.S. Securities and Exchange Commission
Commissioner Roel C. Campos, U.S. Securities and Exchange Commission
Commissioner Cynthia A. Glassman, U.S. Securities and Exchange Commission
Commissioner Harvey J. Goldschmid, U.S. Securities and Exchange Commission
Robert R. Glauber, Chairman and Chief Executive Officer, National Association of Securities Dealers, Inc.
Mary L. Schapiro, President, NASD Regulation, Inc.
Elisse B. Walter, Chief Operating Officer and Executive Vice President, NASD Regulation, Inc.
Thomas Selman, Senior Vice President, NASD Regulation, Inc.
Richard Grasso, Chairman and Chief Executive Officer, New York Stock Exchange, Inc.
Edward A. Kwalwasser, Group Executive Vice President, New York Stock Exchange, Inc.
Donald van Weezel, Vice President, Regulatory Affairs, New York Stock Exchange, Inc.
Annette Nazareth, Director, Division of Market Regulation, U.S. Securities and Exchange Commission
Alan Beller, Director, Division of Corporation Finance, U.S. Securities and Exchange Commission
Stephen Cutler, Director, Division of Enforcement, U.S. Securities and Exchange Commission
Giovanni P. Prezioso, General Counsel, U.S. Securities and Exchange Commission
Larry L. Bergmann, Senior Associate Director, Division of Market Regulation, U.S. Securities and Exchange Commission
James A. Brigagliano, Assistant Director, Trading Practices, Division of Market Regulation, U.S. Securities and Exchange Commission

Appendix A

Recommended Changes to Amended NYSE Rule 472 and NASD Rule 2711.

  1. In proposed NASD Rule 2711(c)(4) and NYSE Rule 472(b)(4) add a definition of the term "due diligence" that includes, at a minimum, pre-deal vetting by research analysts.

  2. In proposed NASD Rule 2711(a)(5) and NYSE Rule 472.40, in place of the proposed extension of trading restrictions to research management, adopt a rule that firms (i) set up an internal written pre-transaction approval and/or post-trade monitoring process for research management designed to ensure that such transactions do not create a conflict of interest between their professional responsibilities and personal trading activities; and (ii) establish a system to require disclosure of research managers' personal holdings to research compliance and senior management so that they can make a judgment about whether a conflict could affect decision-making so as to require recusal. Such a recusal decision should be permitted to take into consideration whether recusal would leave no one to perform a necessary research management function.

  3. Add a provision to the proposed rules for research analysts who are principally responsible for the preparation of research based on firms' quantitative or technical models, that would exempt them from the rules' requirements if their firms adopt systems and procedures to monitor their trading and positions for conflicts of interest comparable to what we propose for research management. (Similar to the treatment of such analysts that the Commission took in its adopting release for Regulation AC).

  4. The NYSE should make its definition of the "associated persons" covered by the rules consistent with the NASD's definition of its term "research analyst."

  5. The proposed bar on publishing research if an analyst solicits investment banking activity in advance of an IPO should be revised as follows: an analyst should be barred from publishing research on an issuer, or discussing an issuer in a public appearance, if the analyst had any communication with the issuer for the purpose of soliciting an IPO mandate within 180 days prior to filing an IPO in which the analyst's firm acted and lead or co-managing underwriter. This bar would last 12 months from the date of the IPO. This provision would not apply retroactively to communications made prior to the effective date of the rule, and it would not apply to an analyst who moves to a new firm, provided that the firm does not receive an underwriting mandate from the issuer within 180 days of hiring the analyst.

  6. The SROs should resolve various inconsistencies noted in our comment letter between their rules and in their proposing statements regarding public appearances and the "use or promise of research."

  7. The SROs' proposed requirements on withdrawal of research should be revised to define the term "withdrawal" as a decision by the firm or its head of research that the firm's applicable research service will no longer include coverage of a particular issuer's equity securities.

  8. The proposed requirements for qualification of analysts should be revised to give comity to the CFA Level One examination, and for personnel who have been primarily responsible for preparing research reports for three or more years on the date the amended rules become effective. A discrepancy between the NYSE and NASD rules regarding whether both firm and regulatory elements of continuing education should be imposed should be resolved by requiring both analysts and supervisory analysts to complete their continuing education requirements through their firms. The SROs should also clarify that the registration and qualification examination apply only to the analysts who are primarily responsible for the content of the research report.

  9. Proposed NASD Rule 2711(f)(3) and NYSE Rule 472(f)(4) should simplify the proposed restriction on research reports around the expiration of lock-up agreements so that it simply bars a firm or a research analyst from issuing a research report for the purpose, in whole or in part, of affecting the price of the issuer's securities for the benefit of a selling shareholder.

  10. The proposed compensation review committee specified in proposed NASD Rule 2711(d)(2) and NYSE Rule 472(h)(2) should only review and approve the compensation of analysts who are principally responsible for producing a research report.

1 The Securities Industry Association brings together the shared interests of more than 600 securities firms to accomplish common goals. SIA member-firms (including investment banks, broker-dealers, and mutual fund companies) are active in all U.S. and foreign markets and in all phases of corporate and public finance. Collectively they employ more than 495,000 individuals, representing 97 percent of total employment in securities brokers and dealers. The U.S. securities industry manages the accounts of nearly 93-million investors directly and indirectly through corporate, thrift, and pension plans. In 2001, the industry generated $280 billion in U.S. revenue and $383 billion in global revenues. (More information about SIA is available on its home page: SIA's Federal Regulation Committee, consisting of over two dozen chief legal officers of securities firms, is SIA's principal legal and regulatory committee.
2 The Committee does not currently have a position on whether the ultimate terms of the global settlement announced last December with a small number of the securities industry's major firms should be applied on an industry-wide basis.
3 The NASD/NYSE Joint Interpretive Memorandum recognized the importance of this role, stating that "these provisions are not intended to prevent a member's investment banking department from obtaining a research analyst's view of a prospective investment banking client before committing to undertake an investment banking transaction." NASD NTM 02-39, July 2002, at 370 (hereafter, the "Joint Interpretive Memorandum").
4 See page 30-31 below for a discussion of the impact the current SRO rules have on certain analysts who are principally responsible for the preparation of research based on quantitative models or technical analysis.
5 Currently firms are required to track employees' personal trading, but are not required to compel employees to disclose all personal equity holdings.
6 Such a recusal decision should be permitted to take into consideration whether recusal would leave no one to perform a necessary research management function.
7 Joint Interpretive Memorandum at 366. The NYSE definition is also inconsistent with recently-adopted SEC Regulation AC, which applies more narrowly to "any natural person who is primarily responsible for the preparation of the content of a research report." Regulation AC, 17 C.F.R. Sec. 244.500.
8 The proposed NYSE definition also goes much further than the definition of "securities analyst" in the Sarbanes-Oxley Act, which, similarly to the NASD, defines the term to mean "any associated person of a registered broker or dealer that is principally responsible for, and any associated person who reports directly or indirectly to a securities analyst in connection with, the preparation of the substance of a research report, whether or not any such person has the job title of `securities analyst.'" Securities Exchange Act Sec. 15D(c)(1).
9 The SEC has recognized this in rules such as Securities Act Rule 139 and 100(b) of Regulation M, which offer more precise and realistic definitions of when a firm is a prospective underwriter.
10 "'Investment banking services' include, without limitation, acting as an underwriter in an offering for the issuer, acting as a financial adviser in a merger or acquisition; providing venture capital, equity lines of credit, PIPES [private investment, public equity transactions] or similar investments, or serving as a placement agent for the issuer." NYSE Rule 472.40, NASD Rule 2711(a)(2).
11 As written, the rule could be read to apply in situations where the broker-dealer is not even an underwriter of the IPO, so long as one of its research analysts communicated with the subject company about a nonpublic investment banking transaction prior to the IPO.
12 In a different context concerning potential conflicts in soliciting underwritings, the Municipal Securities Rulemaking Board's G-37"pay-to-play" rule imposes a 2-year ban on participating in municipal market transactions under certain circumstances. Any ban on analyst research should be no longer than that. We suggest that a one-year ban should suffice to address conflict concerns in this context.
13 See 68 Fed. Register 826, at 833 (January 7, 2003) (emphasis added).
14 Id. at 835. (emphasis added).
15 Joint Interpretive Memorandum, NTM 02-39, at 370 (emphasis in original).
16 To illustrate, media sources frequently contact analysts to obtain factual information on companies and industries as well as the analyst's views on the likely effect of current events on the issuer or its businesses. Public appearances - particularly in the press - frequently involve brief commentary that might be deemed an "opinion" but certainly is not information on which a person could base an investment decision. Indeed, such commentary often is not material to an investor and/or provides no view on any possible impact to the issuer's stock price.
17The SEC recently decided that under Regulation Analyst Certification, the certification requirements apply only to the analyst or analysts primarily responsible for the content of a research report while junior analysts are not required to certify. SEC Release Nos. 33-8193; 34-47384 (February 20, 2003).
18 Waivers can help to maintain liquidity and avoid market disruptions by allowing stock to be sold into the market in a more gradual and controlled manner than might be the case if all the shares were sold immediately upon expiration of the lock up.
19 Exchange Act Sec. 15D(c)(2).
20 Id.
21 Id., Sec. 15D(a). (emphasis added).
22 As noted at pages 15-17 above, the NYSE and NASD seem to be at variance as to whether the guidance that they gave in the Joint Memorandum permitting firms to promise research coverage - but not favorable coverage -- to prospective underwriting clients is still valid.