Stifel, Nicolaus & Company, Incorporated
One Financial Plaza
501 N. Broadway
Saint Louis
63102

March 10, 2003

Mr. Jonathan G. Katz
Secretary
United States Securities and Exchange Commission
450 Fifth Street, N.W.
Washington, D.C. 20549-0609

Re: File No. NYSE-2002-49
File No. NASD-2002-154
Release No. 34-47110 (December 31, 2002), 68 Fed. Reg. 826 (January 7, 2003) (Notice of Filing of Proposed Rule Changes by the New York Stock Exchange, Inc. Relating to Exchange Rules 344 ("Supervising Analysts"), 345A ("Continuing Education for Registered Persons"), 351 ("Reporting Requirements") and 472 ("Communications with the Public") and by the National Association of Securities Dealers, Inc. Relating to Research Analyst Conflicts of Interest)

Dear Mr. Katz:

Stifel, Nicolaus & Company, Incorporated is a full-service broker-dealer that provides investment banking and research services, primarily to middle market and emerging growth companies throughout the United States and to the investors in those companies. We appreciate this opportunity to comment on the proposed amendments submitted by the New York Stock Exchange, Inc. (the "NYSE") and the National Association of Securities Dealers, Inc. (the "NASD") to NYSE Rule 472 and NASD Rule 2711 (the "Proposed Amendments"), regarding research analysts' conflicts of interest. We note that we are joined in our comments by the firms of B.C. Ziegler and Company, City Securities Corporation, Sterne, Agee & Leach, Inc., and SWS Group, Inc.

We applaud the efforts of securities regulators to re-establish confidence in the United States equity securities markets. We believe several of the provisions of the Proposed Amendments, such as the proposal to restrict the issuance of research by a firm that has managed or co-managed a securities offering by a company in the window before and following expiration of any lock-up agreement1 and the proposal to require research analysts to register and pass qualification examinations,2 will help strengthen public trust and confidence in the research process. Other provisions of the NYSE and NASD proposals (which we refer to as the "detail-based proposals") may have unintended consequences and may, in fact, weaken rather than strengthen investor protection and the capital formation process. In particular, we are concerned with the following detail-based proposals:

Restrictions on Issuance of Research Reports - NASD Rule 2711(c)(4) and NYSE Rule 472(b)(4)

The NYSE and NASD propose to prohibit a research analyst from issuing a research report or making a public appearance concerning a subject company if the analyst engaged in communications with the subject company "in furtherance of obtaining investment banking business" prior to the time the subject company designated the firm, pursuant to a written agreement, as an underwriter of an initial public offering ("IPO").

Restrictions on Research Analyst Compensation - NASD Rule 2711(d)(2) and NYSE Rule 472(h)(2)

The NYSE and NASD propose to prohibit consideration of any contribution by an analyst to investment banking in determining an analyst's compensation. The NYSE and NASD also propose to change the present principles for determining analyst compensation by substituting a compensation committee's collective judgement and understanding of a firm's organization and a specific analyst's contributions to the firm with specifically mandated factors.

Extension of Trading Restrictions - NASD Rule 2711(a)(5) and NYSE Rule 472.40

Both the NYSE and the NASD propose to expand existing trading restrictions to apply to research supervisors, research directors and members of committees who oversee research analyst independence and recommendations.

Definition of Research Report - NASD Rule 2711(a)(8) and NYSE Rule 472.10(2)

The NYSE and the NASD propose to expand the definition of "research report" by removing the requirement that a communication include a recommendation to be considered a research report. This revision could expand the definition of "research report" to include general market commentary, technical analysis, sales literature and scholarly articles.

As more fully discussed below, we respectively submit that the investor protection concerns which these provisions attempt to address would be more effectively met through additional principles-based rules and more rigorous disclosure requirements. A principles-based system would also minimize the potential for unintended consequences, several of which could be detrimental to investors and the capital raising process.

We firmly support the underlying objectives of these detail-based proposals, which we believe is to restore investor confidence in the objectivity of research. Certain well-publicized and questionable practices involving research analysts adversely impacted investor confidence in the objectivity and credibility of equity research in recent years. However, preexisting rules specifically prohibited certain of these practices, in particular NASD and NYSE rules forbidding securities industry professionals from making exaggerated, unwarranted or misleading statements in public statements.3 Preexisting rules also more generally required that all communications with the public be based on the fundamental principles articulated in the Securities Exchange Act4 and the standards embodied in existing Securities and Exchange Commission ("SEC") and NYSE and NASD rules-standards of market integrity, investor protection, and standards of commercial honor and just and equitable principles of trade.5 The fundamental principles and standards, which have governed the securities industry for years, are clearly strengthened with the adoption of NYSE Rule 472 and NASD Rule 2711 in May 2002,6 and Regulation AC just last month.7

We are concerned that the detail-based proposals discussed in this letter relating to analyst conflicts of interest are both overly specific and yet vague. Instead of amending the existing rules to make them more detailed, we would recommend that regulation of research analyst conflicts of interest occur within a framework of the technical rules already in effect, and additional principles-based rules.

In order to restore investor confidence, SEC Chairman William Donaldson stated in his first speech as Chairman that companies generally must "make a conscious decision that ethics and integrity should be at the heart of every business decision."8 In the Chairman's view, "[c]ompanies must look beyond rules, regulations and laws and look to the principles upon which sound business is based." To achieve this goal, "those at the helm will have to make a conscious decision that ethics and integrity should be at the heart of every business decision" and, "we cannot legislate it or promulgate rules to enforce it."9 Although Chairman Donaldson was not speaking about analyst conflicts of interest, he could have been; ethics and integrity, after all, are the principles according to which securities professionals should always conduct business.

In addition, we draw the attention of the NYSE and NASD to a Consultation Paper recently released by the Financial Services Authority in the United Kingdom ("FSA") addressing conflicts of interest in the issuance of investment research. In that paper, the FSA expressly rejects a detailed rule-based approach, stating that it does not believe "that standards can be fixed through `black letter rules'." Rather, the FSA proposes to identify conduct and practices that are likely to interfere with an analyst's objectivity and that each firm should avoid. The FSA proposes to direct senior management of investment banks to implement controls with regard to, among other things, supervision of research analysts, analysts' involvement in marketing activity, analysts' compensation, and susceptibility to pressure from subject companies. The FSA also recommends improved disclosure of the spread of an investment bank's recommendations in particular markets or sectors as compared with recommendations for particular clients, and of the specific analyst's track record generally and in specific sectors. FSA expressly acknowledges that controls and disclosure may vary from firm to firm, but emphasizes that "[i]f conflicts of interest are not properly managed and controlled, investors will not be adequately protected and confidence in financial markets will be undermined..."10

We believe that management of potential conflicts of interest requires constant exercise of judgement and that direct responsibility for managing these potential conflicts of interest must rest with the senior management of a firm. After all, the integrity of the securities markets is directly dependent on the integrity of securities firms, and the integrity of each securities firm is the foremost responsibility of firm senior management. Like the FSA, we do not believe that detailed, prescriptive rules, which, by their very nature are designed to supplant individual judgement, are in the best interests of investors and the markets, particularly when securities firms come in many sizes and have very diverse organizational structures.

In the same spirit that lead to the implementation of Chinese Wall procedures over a decade ago, regulators should now require that every integrated securities firm recognize the need for increased diligence and that every firm establish procedures that ensure analysts' activities and relationships result in the issuance of objective, independent research in accordance with existing rules relating to just and equitable principles of trade. We encourage the NYSE and NASD to adopt this approach in lieu of the additional detail-based proposals. And we believe that responsibility for strict adherence to these procedures should rest explicitly with the senior management of each firm.

Our remaining comments will address:

  • A proposal for a set of principles-based rules to supplement the existing rules and standards of market integrity, investor protection and commercial honor and just and equitable principles of trade.

  • A critique of the detail-based proposals and discussion of why principles-based rules will more effectively address the concerns at which several detail-based proposals are addressed.

A. Principles-Based Rules

As stated above, we believe that analyst conflicts of interest are well addressed by the existing rules and principles supplemented by the addition of NYSE Rule 472 and NASD Rule 2711 as presently in effect, and Regulation AC. Any additional rules, however, should reinforce compliance with fair and ethical principles of business and trade.

We believe a principles-based approach would serve as a stronger, more effective force to remedy any residual problems with respect to analyst conflicts of interest. A principles-based approach also would be flexible enough to apply to future advancements in the securities markets and changes in the financial services industry. In addition, a principles-based approach is less susceptible to clever manipulation in which firms meet the letter of the law but not the underlying principle. Under a principles-based system, member firms would be motivated to establish compliance procedures designed to ensure compliance with the spirit, as well as the letter, of the law. Ultimately, we believe this will provide the strongest, most durable foundation for ensuring objectivity of research, and the most effective path for reestablishing investors' faith in the securities markets.

We propose that the NYSE and NASD replace the detail-based proposals with a directive that their respective members adopt and implement scrupulous policies and procedures according to the following principles:

  • Any research produced by sell-side research analysts must have as its paramount objective the investors' interest, as opposed to the interest of the subject company or industry.

  • No securities firm or associated person should ever promise or otherwise represent that research about any company, industry or industry sector will be favorable or unfavorable.

  • Research analysts should not report to or in any way directly or indirectly be answerable to any division, including but by no means limited to investment banking, of a firm other than the department of research and its head.

  • Research analyst compensation should not be related to a specific investment banking transaction or transactions. Individual analysts' compensation should be based on the utility of an individual analyst's research.

  • Firms and associated persons must disclose all material information about current or potential relationships with any issuer, recognizing that certain disclosure can be withheld when it would result in prematurely disclosing a material prospective transaction.

  • Firms must make public periodically the track record of each of their analysts based on an objective evaluation conducted either through a peer review process or through an independent evaluator (such as StarMine.com).

  • Each analyst (or any person with material input on an analyst's opinions) must disclose his or her intent to buy or sell a covered security within a reasonable time before the date on which the transaction will take place.

  • All required disclosure should be made in the same manner that a firm disseminates research.

The use of these additional guidelines and principles expanding on existing principles, would, we believe, achieve the sought after result of ensuring the integrity and objectivity of research and recommendations by research analysts. This approach also would permit each securities firm the flexibility to determine how best to achieve this result within the framework of its own business model.

B. Comments on Detail-Based Proposals That Would Be Better Addressed Through Principles-Based Rules or Through Enhanced Disclosure

A principles-based approach to the concerns underlying the detail-based proposals discussed below, coupled with improved disclosure, would in our view more successfully achieve the stated goal of the Proposed Amendments - restoration of integrity to the public equity markets.11

Restrictions on Issuance of Research Reports - NASD Rule 2711(c)(4) and NYSE Rule 472(b)(4)

The NYSE and NASD propose to prohibit a research analyst from issuing a research report or making a public appearance concerning a subject company if the analyst engaged in communications with the subject company "in furtherance of obtaining investment banking business" prior to the time the subject company designated the member pursuant to a written agreement as an underwriter of an initial public offering ("IPO").

In our opinion, the "in furtherance of obtaining investment banking business" provision of the rule casts far too broad a net. This provision could potentially have a significant adverse effect on investor protection and capital formation. We are concerned that this provision will discourage analysts from visiting (or firms even prohibiting visits) or otherwise communicating with private companies, and will result in the reduction in the breadth and quality of available research. This fear will not only stifle research of companies prior to IPOs, but will impair the quality of information available to the public about newly public companies, because analysts were not able to gather baseline information about companies prior to their IPOs. This provision also could have a particularly crippling effect on smaller firms that do not have the resources to hire two sets of analysts (one to research companies prior to IPOs and another to issue research reports on companies once they have completed their IPOs).

The NYSE and NASD both propose to permit due diligence meetings, but the NYSE's statement of purpose regarding such meetings suggests the carve out may be limited to due diligence in anticipation of an underwriting. The NYSE stated in the proposing release that, "[r]ecognizing the need for critical financial analysis of a subject company during the period an issuer is preparing to engage in a securities offering with the public, the rule allows research analysts to participate in due diligence communications." We request that the NYSE and NASD clarify that the exception is not so limited.

This detail-based proposal appears intended to prevent the use of analysts to promote their firms' investment banking capabilities, and to prevent analysts marketing companies that are the subject of an IPO through the guise of research. But we believe that existing rules address this. NASD Rule 2711(e) and NYSE Rule 472(g) already prohibit promises of positive research, and NASD Rule 2210(d)(1)(B) and NYSE Rule 472.40 forbid securities industry professionals from making exaggerated, unwarranted or misleading statements in public statements. To the extent that the NASD and NYSE believe that the existing rules should be strengthened, a principles-based rule addressing these concerns would reinforce the existing rules, while recognizing that each firm must decide how best to prevent these prohibited activities.

Restrictions on Research Analyst Compensation - NASD Rule 2711(d)(2) and NYSE Rule 472(h)(1) and (h)(2)

NYSE Rule 472 and NASD Rule 2711 as currently in effect already prohibit a firm from compensating or rewarding an analyst based on any specific investment banking transaction. The NYSE and NASD now propose to restrict analyst compensation even further, by prohibiting any consideration of the analysts' contribution to investment banking department performance in determining an analyst's compensation. The NYSE proposes to permit a firm to compensate the analyst based on the firm's (rather than analysts') overall performance, including the performance of the investment banking department. The NYSE and NASD both further propose to require that compensation committees consider three specific factors when reviewing research analyst's compensation: (i) an analyst's individual performance; (ii) the correlation between the analyst's recommendations and stock price performance, and (iii) overall ratings from clients, sales force and peers other than investment banking and other independent rating services.

We support the principle that analyst compensation should not be based on the success or failure of specific investment banking transactions. And we believe that the present provisions of NYSE Rule 472 and NASD Rule 2711 very effectively address this principle. The new detail-based proposals are unworkable, however, given the diversity of structure and size among securities firms in the United States. These proposals disregard the contributions that unbiased research analysts can make to a firm's investment banking business, in understanding industries and public companies generally, and in evaluating whether specific private companies are sufficiently robust to perform successfully as public companies. These contributions may be unrelated to specific transactions, but are very important to the integrity both of a firm and to the securities markets.

If the NYSE and NASD believe that additional rules regarding compensation are necessary, we propose the adoption of principles-based rules that require disclosure that the analyst's compensation may be based in part on investment banking revenue. Furthermore, the disclosure should include the analyst's performance track record, including the performance spread between companies that are investment banking clients and other companies. We believe that this approach would permit individual firms and their compensation committees to make compensation decisions for individual analysts based on a firm's structure and business model, and the analyst's utility to investors and overall contribution to all areas of the firm.

At a minimum, any compensation rules adopted should permit a firm to pay its research analysts based on: (1) the firm's overall profitability, (2) the profitability of the firm's capital markets division, (3) the profitability of the firm's investment banking department, and (4) the profitability of specific industry and/or product groups within the firm's investment banking department.

Extension of Trading Restrictions - NASD Rule 2711(a)(5) and NYSE Rule 472.40

Both the NYSE and the NASD propose to expand existing trading restrictions to apply to research supervisors, research directors and members of committees who oversee research analyst independence and recommendations. This proposal would discourage qualified people from serving in these positions, despite the importance of their presence in efforts to strengthen research integrity.

We believe that this proposal as presently drafted would result in what would effectively be a prohibition of ownership of individual equity securities by research analysts and others with oversight into analysts' research. This would be an onerous result. In effect, this proposal could prevent senior management and others with seasoned judgement from serving on the committees that oversee or evaluate research, with the concomitant result of diminished quality of research. Also, this provision would be particularly problematic for small firms where senior personnel of necessity serve multiple functions, including reviewing research.

Ownership of securities by analysts, after all, is not inherently bad. Indeed, what better recommendation by an analyst than that he or she has invested personal funds in a company that the analyst covers? Rather, an analyst who owns a covered equity security should not be able to trade ahead of firm clients who also own the security.

Therefore, we propose a principles-based rule that enhances disclosure by requiring every analyst, and any other person with material input into the analyst's recommendations, to disclose personal ownership of any covered security and to disclose any intention to engage in a transaction that will result in a change in that ownership interest. This disclosure would be made in the same way that the firm disseminates research. The result, we believe would reinforce the goals of the NYSE and NASD and result in increased information for investors and the markets generally.

* * * * * * * *

Several of the detail-based proposals also raise investor protection concerns due to the practical effect of proposed technical changes. The most significant of these in our view is the proposal to expand the definition of "research report" discussed below:

Definition of Research Report - NASD Rule 2711(a)(8) and NYSE Rule 472.10(2)

The NYSE and NASD propose to expand the definition of "research report" by removing the requirement that a communication include a recommendation in order be considered a research report. This revision arguably sweeps communications such as general market commentary, technical analysis, sales literature and scholarly articles into the definition of "research reports." If experts are unwilling to produce technical market analysis or write editorials or scholarly articles because they may fall under the definition of research reports under NASD Rule 2711 or NYSE 472, the effect of the amendment may be to reduce further the amount and quality of information available to investors.

The proposed change appears to reverse the decision of the NYSE and NASD in May of 2002 to include the recommendation requirement in response to comments similar to the ones made above. The NYSE's statement of purpose relating to the proposed amendments indicates the change was precipitated by a perceived need to conform the definition to the one required by the Sarbanes-Oxley Act of 2002. We encourage the NASD and NYSE to clarify, through written interpretation, those communications (such as those described above) that are not intended to fall within the definition of research report. As discussed above, rules already exist that prohibit the publication of any statements known to be false.

C. Discrepancies Between NASD Rule 2711 and NYSE Rule 472

Finally, there are a number of instances in which the proposed amendments to NASD 2711 and NYSE Rule 472 are inconsistent and compliance with the different provisions would require different, and perhaps inconsistent, actions by firms. We understand that the Securities Industry Association is submitting a letter to you that highlights these discrepancies. We urge the NASD and NYSE to reconcile these inconsistencies in any final rulemaking.

Conclusion

In conclusion, we appreciate the efforts of the securities regulators in recent months to implement rules and reemphasize standards of integrity and ethics that will re-establish confidence in the United States equity markets. The adoption by the NYSE, NASD and SEC of the rules now in effect was a very important step in that direction. And the NYSE and NASD proposals to limit issuance of research in the 15 day windows before and after expiration of lock-up agreements and to require research analysts and their supervisors to register and pass qualification requirements should enhance these efforts further. But we encourage the NYSE and NASD to pause before they finalize the Proposed Amendments and risk weakening rather than strengthening investor protection and the capital formation process. Chairman Donaldson's call for honesty, integrity, transparency and accountability sets the stage for rulemaking that is principles based, rather than exclusively prescriptive. We believe that in the areas discussed in this letter, principles-based rules would be a more effective approach to ensure that the research practices of securities firms and securities professionals always adhere to the highest standards of fair and ethical principles of business and trade.

To the extent that the NYSE and NASD retain the Proposed Amendments as currently proposed, we concur with the recommendations made in the comment letter submitted to you by the firms Adams, Harkness & Hill, Inc., AG Edwards, Keefe, Bruyette & Woods, Inc., Pacific Growth Equities, LLC, RBC Capital Markets, Stephens, Inc., Stifel, Nicolaus & Company, Incorporated and William Blair & Company.

We appreciate this opportunity to comment. If the NYSE, NASD or SEC staff would like to discuss these comments further, please feel free to contact Tom Prince, General Counsel of Stifel, Nicolaus & Company (314-342-4068).

Sincerely yours,

Stifel, Nicolaus & Company, Incorporated

cc: William H. Donaldson, Chairman
Cynthia A. Glassman, Commissioner
Harvey J. Goldschmid, Commissioner
Paul S. Atkins, Commissioner
Roel C. Campos, Commissioner

____________________________
1 Proposed NYSE Rule 472(f)(4) and Proposed NASD Rule 2711(f)(3).
2 Proposed NYSE Rule 344 and Proposed NASD Rule 1050.
3 NASD Rule 2210(d)(1)(B); NYSE Rule 472.40.
4 Section 2 of the Securities Exchange Act.
5 E.g., NASD Conduct Rules 2210(a), 2310, 3010(a) and NYSE Rules 401, 476(a)(6).
6 See Securities Exchange Act Release No. 45908 (May 10, 2002) 67 Fed. Reg. 34968 (May 16, 2002).
7 See Securities Act Release No. 8193 (Feb. 20, 2003), 68 Fed. Reg. 9482 (Feb. 27, 2003).
8 William H. Donaldson, Remarks at the Practicing Law Institute - SEC Speaks (Feb. 28, 2003).
9 Id.
10 Conflicts of Interest: Investment Research and Issues of Securities, Consultation Paper 171, Financial Services Authority (United Kingdom) (February 2002).
11 See Securities Exchange Act Release No. 47110 (Dec. 31, 2002) 68 Fed. Reg. 826, 832 (Jan. 7, 2003).