April 11, 2002

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

Re: National Association of Securities Dealers and New York Stock Exchange Proposals Relating to Research Analyst Conflicts of Interest, File Nos. SR-NASD-2002-21 and SRNYSE-2002-09

Dear Mr. Katz:

Thank you for giving the Securities Industry Association ("SIA")1 the opportunity to comment on the above-referenced proposals to strengthen regulatory oversight of securities analysts employed by U.S. broker-dealers. We are disturbed by the perceived reduction of public confidence in the U.S. equity markets due to the potential for conflicts of interest between broker-dealers' research and investment banking functions. Strengthening investor trust and confidence is critical to our markets, and we believe that the proposals can be, with some needed adjustments, an important step toward that end. For its own part, SIA last year adopted best practices to deal with this issue, recognizing that taking steps to disclose such conflicts is an important goal of the securities industry. Although we believe that best practices can often be effective in raising standards in the securities industry, we support the efforts of the NASD-Regulation, Inc. ("NASD") and the NewYork Stock Exchange ("NYSE") to adopt regulations in this area.

We commend the National Association of Securities Dealers ("NASD") and the New York Stock Exchange ("NYSE") for their hard work in seeking to rebuild public trust and confidence in this area. We also commend Congressman Michael Oxley, Chairman of the House Committee on Financial Services, Congressman Richard Baker, Chairman of the Subcommittee on Capital Markets, Insurance, and Government Sponsored Entities, and Chairman Harvey L. Pitt, his fellow commissioners and staff of the United States Securities and Exchange Commission ("SEC") for their leadership in seeking improvements in this area.

While we support the objectives of the rule proposals generally (many of which seek to codify our Best Practices), we believe that the current proposals need to be significantly revised in several important respects to provide investors with better and more timely access to information contained in research reports. In particular, several of the proposals would not provide sufficiently meaningful disclosure to investors to justify the considerable cost involved in implementing them or would potentially undermine public policy regarding disclosure of material non-public information. We have suggestions for revising these proposals to provide equally useful disclosure at lower costs and without the same potential for "tipping" analysts or recipients of research.

The most important issues are (i) clarifying the definition of research, (ii) disclosing compensation (especially prospective compensation) received by broker-dealers and their affiliates in a way that is meaningful but does not tip analysts or research recipients about nonpublic transactions, (iii) disclosing ownership positions held by broker-dealers and their affiliates in subject companies on a time frame like that used in the proposals regarding distribution of ratings and time charts, (iv) reconsidering the partial ban on research by managing and co-managing underwriters, (v) clarifying the role of legal and compliance personnel in monitoring communications between research and investment banking and the types of communications that should be monitored, (vi) developing more consistency between the two proposals, and (vii) adopting a realistic time frame for implementation. Other definitional and interpretive issues will be dealt with in an appendix to this letter.

We agree with the premise that, as a general principle, disclosure is a better approach to managing real or apparent conflicts of interest than prescriptive prohibitions. which can be both overly complex and subject to evasion. It is important to caution that disclosure can be counterproductive if filled with information not meaningful to readers. Successful regulatory disclosure requirements must find the right balance between not enough and too much. We make a number of suggestions below to strike a better balance without diminishing the value of the disclosure to investors.

I. Overview and Summary.

Notwithstanding our overall support for the proposals' goals, certain aspects of the proposed rules in their current form pose serious concerns. A number of these proposals are likely to produce information that will be of dubious value to most investors. The burden and cost of gathering some of this less relevant information is very high. Some proposals will also reduce the flow of research to investors. We think all of these concerns can be ameliorated to a considerable degree by implementing the suggestions that we offer below.

We are particularly concerned with the breadth of two proposals. First, the requirement that firms track and disclose whether any form of compensation from an investment banking client or other subject company is received by any affiliate of the firm worldwide in its current form fails to address the NASD's and NYSE's legitimate objectives, while posing significant systems costs for broker-dealers and dangers of tipping selected market participants about non-public transactions. It is highly questionable whether this proposal can survive any sort of objective cost-benefit analysis. It also appears to fall short of current statutory obligations not to impose unnecessary burdens on efficiency, competition and capital formation, given that it appears on its face to have virtually no beneficial value for most investors but will impose significant costs on both broker-dealers and issuers. Second, we are also very concerned that the requirement to report ownership positions held in the subject company during time periods inconsistent with time periods used for other aspects of these proposed rules will create significant unnecessary costs.

Other proposals will also impose significant costs, especially on smaller broker-dealers. One almost certain byproduct of the rules will be to raise costs in such a way that will make it difficult for all but the largest securities firms to provide research services.

There are also a number of significant drafting ambiguities and interpretive issues which, if left unresolved, will significantly impair research. We point out some of these issues in Appendix A to this letter.

The discussion below includes the following key recommended changes to the proposals, which would significantly improve the utility of the disclosures to investors, while reducing unnecessary costs and uncertainties for broker-dealers:

In light of our concerns over some specific proposals, we strongly recommend that the proposals, if adopted, be phased in over a staggered period. Some of the provisions, while they would involve substantial work and expense to hire and train new legal and compliance personnel as well as education programs for research analysts, investment banking personnel and other employees, could be implemented in fairly short order. However, those proposals that require developing new disclosure systems will require significantly more time to create, test and implement. We suggest the following implementation schedule:

II. Principal Substantive Concerns.

A. NASD Rule 2711(a)(8) and NYSE Rule 472.10(2) -- Definitions of "Research Report."

Summary of SIA Position: The term "research report" as defined in the proposals is overbroad, and will unnecessarily impede the flow of information to investors. In particular, the NASD and NYSE should better clarify that the proposed definitions are not intended to sweep in the following documents, to the extent that they mention specific companies:

The proposals also raise issues of speed, space and breadth of distribution. By imposing a range of new regulatory restrictions on various electronic communications that can be deemed "research reports," they will have the practical effect of slowing down, in a way that technology cannot overcome, the pre-market opening distribution process that many investors want. They will also impede the use of electronic delivery devices that many investors like, such as e-mail alerts, to the extent that those devices are not capable of carrying the full set of disclosures mandated by the rule.

These issues can be addressed in part by narrowing the definition of "research report" or offering written guidance on what it is intended to cover. The rules should also permit the use of electronic delivery media that are ancillary to complete research reports, such as First Call notes, to transmit recommendations, if they contain certain disclaimers and point to another source, such as via a reference or hyperlink to a web site, where the complete required disclosure can be found. Such an "access equals delivery' approach, coupled with our suggestions below for prominent disclosure of potential conflicts due to compensation, and disclosure of holdings by the broker-dealer or its affiliates exceeding one per cent of the subject company, would be effective approaches that would not generate significant unintended negative consequences on the speed, accessibility or cost of research.

Recommendations: We suggest that the NASD and NYSE issue guidance clarifying that their proposed rules are not intended to apply to the types of reports mentioned above. In addition, the NYSE and NASD should work together to harmonize their definitions of research. The uniform definition should either follow the NASD approach, or as an alternative way of addressing our concerns about the scope of the term, the definition of "research report" should be limited to

An analysis disseminated by a member regarding a company, other than general market commentary or analysis based on objective quantitative criteria applied to a range of companies, that:

(a) provides information reasonably sufficient upon which to base a decision regarding an equity investment in that company;

(b) has been prepared by an employee hired by the member's equity research department to work within the member's research department in the normal course of his or her duties;

(c) includes a recommendation for a securities transaction, or a price target;

(d) is held out to customers as being the product of an independent equity research analyst; and

(e) discusses a common equity security.

We also recommend that the NASD and NYSE create an exception to the disclosure provisions of the new rules to permit distribution of short electronic "research reports" where it is not practicable to include the full range of mandated disclosures, as long as the communication includes a reference or hyperlink to a web site or other easily accessible location where the full required disclosures can be found.


Discrepancies in Definitions. As an initial matter, we are concerned that the NASD and NYSE definitions of "research report" are substantially different from each other. Definition .10 of the NYSE's proposed amendments to Rule 472 defines "research report" as reports that "are generally defined as, but not limited to, an analysis of equity securities of individual companies, or industries, which provide information reasonably sufficient upon which to base an investment decision." (emphasis added). The NASD's proposed Rule 2711(a)(8) defines the term to mean "a written or electronic communication that the member has distributed or will distribute with reasonable regularity to its customers or the general public, which presents an opinion or recommendation concerning an equity security."

While below we raise questions about the scope and meaning of the NASD's definition, we think it is much preferable to the divergent NYSE definition. We are particularly concerned with the NYSE's expansive phrase "but not limited to." This calls into question whether the requirements of amended Rule 472 go beyond the source of the regulators' concern, equity research. The NYSE's definition also is not clearly limited to written or electronic communications, or communications distributed with reasonable regularity.

Scope of the Term "Research Reports." The NASD offered guidance in its proposing release that the term "research report" is not intended to apply to reports by mutual fund portfolio managers. A comprehensive regulatory framework already governs the conduct of mutual fund portfolio managers, making this exclusion appropriate. We urge the NYSE to make a similar clarification. We also recommend that the NASD and NYSE clarify that reports to clients by other types of investment advisers, who are also subject to a comprehensive regulatory scheme and wide-ranging disclosure requirements, are similarly outside the scope of the term "research report."

The proposed definitions are ambiguous with regard to several other categories of information that could be deemed "research." We would like to highlight three of these: quantitative analyses of securities portfolios, market commentary, and information provided by traders or marketing groups.

These concerns can be addressed, at least in part, if NASD-Regulation issues guidance clarifying that these types of reports are not intended to be covered, or alternatively, adopts the definition of "research report" that we propose above.

1. Quantitative and Technical Analyses. SIA recommends that the NASD and NYSE not apply the proposed rules to quantitative, technical or relative analyses of securities or portfolios of securities. Such analyses contain recommendations derived from quantitative models based on objective publicly-known measures of stock performance, or by comparing the relative performance of a security to a benchmark (such as a comparison between a number of equity securities and an index such as the S&P 500). These reports may recommend transactions in specific issuers based on this quantitative or technical analysis, or they may recommend trading strategies, quantify value or offer forecasts for entire sectors based on a portfolio analysis of volatility and return of indices for these sectors.

Since these reports are based on objective criteria, rather than on any subjective judgments about the economic prospects of particular companies, it is hard to see any purported benefit to investors from having "conflict" information in such a report. Attaching price charts to these types of reports for each security mentioned would be extremely time-consuming and expensive, and would present readers with a bulky morass of information of little practical value, since the reports could potentially cover dozens or hundreds of securities, and the securities mentioned could be entirely different from one report to the next. Moreover, the enormous potential range of securities covered would effectively, but unfairly, bar analysts who prepare this type of research from owning any equity securities.

Applying these rules to items such as quantitative or technical analyses, which are most commonly shared with institutional investors, also goes in the opposite direction from other recent NASD initiatives. For example, the NASD issued a no-action letter in 1999 that permitted electronic communications to institutional accounts to be sent without pre-review.2 This position was based on the premise that such materials warrant less regulatory scrutiny and because of their time-sensitive nature would be hindered if pre-review were required. In contrast, requiring price charts and other items required under the proposed new rules would significantly delay issuance of these types of materials.

2. Market Commentary. The proposed rules are also not appropriate as applied to market commentary. A number of firms put out frequent reports on overall market conditions, discussing matters such as factors contributing to intraday market index movements, or the potential macroeconomic impact of current news developments such as an interest rate change by the Federal Reserve Board, new information from the Labor Department on unemployment statistics, an international political crisis, etc. While these reports may comment on specific issuers, these reports have nothing to do with concerns about internal conflicts biasing analyst coverage of specific companies. The requirements of the proposed amendments in this context would make little sense. They merely create an additional expense and procedural hurdle that might delay issuance of these time-sensitive reports, and insert disclosures that do not advance the purpose of these rules. As noted above, this seems contrary to efforts that the NASD has recently made to avoid hindering timely issuance of these reports.

3. Information Provided by Trading, Sales and Marketing Personnel. Many firms provide information through traders, sales personnel or marketing groups via media such as e-mail or web sites. These communications may reiterate recommendations made by research analysts, and also may have shorter-term trading recommendations or strategies. No one takes the personnel making these communications to be independent research analysts, and it does not seem sensible to subject them, for example, to the personal trading restrictions of the proposed rules. Communications between brokers or traders and their clients that refer to recommendations made by research analysts and contain other commentary or information not inconsistent with the analyst recommendation should not themselves be deemed "research reports." Sales literature, which is subject to comprehensive regulation under other rules, should likewise not be regulated as a "research report" simply because it refers to an analyst recommendation. The fundamental issue is reliance by the public on equity research that it has assumed to be independent. A recommendation or commentary that is not held out or "branded" as independent research by an equity analyst should not be covered by the rules.

Impact on Timely Research Distribution. The proposed definitions of "research report" will have a significant adverse effect on the ability of securities analysts to communicate their views in a timely manner. As a practical matter, the proposed rules appear to restrict delivery devices that are not amenable to the full set of required disclosures (e.g., e-mail alerts to wireless devices). The need to ensure that the full panoply of disclosures is attached also puts a significant practical limit the ability of an analyst to quickly issue any type of electronic advice in the face of a sudden market development. The anticipated benefits of newly proposed individual disclosure requirements should be weighed against the practical impact they would have on the subsequent achievable speed and breadth of research report distribution.

It is also not clear how or whether the definition is meant to apply to recommendations distributed by electronic research vendors such as First Call. Many investors rely on these products. These services are especially important for retail investors, since they afford retail investors access to research that might otherwise only be readily available to institutional investors. A broker-dealer supplying research to such vendors does not control how that research is repackaged. For example, First Call notes are currently comprised of ASCII text files with limited line space and no graphical capability. We do not know if First Call or other electronic research vendors will be willing to upgrade the technical specifications of their products in response to any new rules, how quickly they could make such an upgrade, or what impact that might have on the speed, accessibility or cost to subscribers.

B. NASD Rule 2711(h)(2)(ii) and NYSE Rule 472(k)(1)(ii) - disclosure of trailing and future compensation.

Summary of SIA Position: This proposal in its current form is highly problematic public policy, especially its forward-looking provision. As written, the proposal will provide no real benefit to investors, while hurting all broker-dealers that perform investment banking services (as well as other broker-dealers that publish research reports), and in many instances the companies that use them (especially smaller companies) and shareholders in those companies. Instead of this approach, we recommend that the rules require broker-dealers to have clear and prominent disclosures at the front of all equity research reports that indicate that conflicts of interest arising from compensation may exist. We also recommend that the rules preserve the existing NYSE requirement to disclose if the U.S. member firm preparing the research report has acted as managing or co-managing underwriter for an offering of the issuer's securities, but consistent with the time period for issuer compensation disclosure, or makes a market in the issuer's securities.

Recommendation. We propose that research reports be required to contain disclosure of any investment banking compensation received by the broker-dealer or its affiliates from publicly announced transactions within the past 12 months (or even a longer period), as well as retaining the requirement currently in NYSE Rule 472 (which the NYSE proposal would remove) that firms disclose where they have acted as managing or co-managing underwriter of a public offering of securities for the subject company within the same time period and the requirement that firms disclose whether they are currently market makers in the subject company's securities. In addition, the NASD and NYSE should require concise and prominent standardized disclosure concerning non-public transactions generally and their potential for conflict. E.g.:

"We have received the following public compensation, such as advisory fees, investment banking fees and brokerage commissions, from [the Subject Company] during the past 12 months in connection with transactions that have been publicly disclosed. [describe]. We also may have received compensation, such as advisory fees, investment banking fees and brokerage commissions, from [the Subject Company] during the past 12 months in connection with transactions that have not been publicly disclosed. Because of the nature of the business in which we are engaged, readers of this report should assume that we may receive compensation from [the Subject Company] in the future. Investors should not rely solely on this report in evaluating whether or not to buy or sell the securities of [the Subject Company]."

Strong cautionary disclosure makes particularly good sense regarding possible future compensation that the firm "reasonably expects" to receive within the following three months. Designing and examining these systems for compliance with this vague injunction is likely to be frustrating for both the firms and self-regulatory organization ("SRO") examiners. Firms are likely to take a cautious approach toward this arm of the proposed rule, and so will make disclosure in almost all instances. Our suggested approach would more clearly caution investors about the potential conflicts of future compensation than would the statement suggested in the proposal to the effect that "this firm or one or more of its affiliates has received compensation from Firm X, or reasonably expects to receive compensation from Firm X within three months following publication of this research report."

In the event that the regulators do not accept our primary recommendation, we have two other suggestions that might be alternative ways of addressing some of our concerns.

1. The required disclosure of expected future compensation from the subject company would be easier to understand and apply if the NASD and NYSE clarified how a firm could base its "reasonable expectation" about such future compensation. For example, this provision would be easier to apply if the NASD and NYSE clarified that a firm could base its judgment that it reasonably expects to receive compensation within the next three months on the fact that the firm engages in transactions for which it receives compensation with companies that are in business lines similar to the subject company. Without such guidance, compliance with this provision will be sheer guesswork, and there will be great variations between how different firms apply this requirement.

2. The NASD and NYSE should clarify, with regard to this provision as well as with regard to the proposals generally, that the term "affiliate" is intended only to include corporate entities, not individuals. We gather from the discussion in the release that this is their intent. If the intent is to include individuals, such as officers, directors and employees of the broker-dealer and its corporate affiliates, then the proposal is unworkably complex and expensive. Firms would have to develop systems to track a myriad of forms of compensation (e.g., dividend checks, rebates on consumer purchases, insurance claim payments, pensions, severance pay or IRA rollovers from prior employment) received by thousands or tens of thousands of employees, including both the broker-dealer as well as affiliates, to determine whether any such payments were made by a subject company.

Discussion. A requirement that firms disclose situations where they have received compensation for non-public transactions would create the prospect of tipping analysts and/or recipients of their research. This would exacerbate the conflicts that engendered the proposals in the first place. Moreover, this sort of tipping of select market participants (analysts and recipients of their research) would raise the same social and economic issues that drove Congress, the SEC, the SROs and broker-dealers to develop rigorous laws, regulations and procedures against insider trading together with prophylactic protection for material nonpublic information within a broker-dealer concerning material nonpublic information about corporate transactions.3

To avoid this problem, the NASD and NYSE opt instead to make the disclosure as broad as possible. Thus, for example, custodial or other pension management services to the subject company by an asset management affiliate of the broker-dealer, payroll administration services to the subject company by a data services affiliate of the broker-dealer, corporate credit cards used by the subject company and sponsored by an affiliate, or a rebate received from the subject company for office equipment purchased from it by an affiliate of the broker-dealer would all have to be centrally tracked by the broker-dealer and would trigger disclosure under this provision. Unfortunately, disclosure this broad provides virtually no useful information to the intended audience - readers of research reports seeking to evaluate analysts' forecasts and recommendations.

By requiring disclosure of compensation even if the compensation was generated by a non-public transaction, the NASD and NYSE chose to place this proposal on the horns of a dilemma: either it calls for a data field that is narrow enough to risk disclosing material nonpublic information about corporate transactions to select individuals in contravention of well-established public policy, or it calls for a data field so broad, encompassing all compensation received by all affiliates, that disclosure is, on its face, of no value. The NASD and NYSE effectively opt for the second choice, at least for firms with affiliates in diverse businesses. We will not challenge whether this is the better of two undesirable self-imposed choices. What we do question is whether this particular proposal is sound public policy.

Balanced against the fact that the disclosure is, by design, largely of no practical use, it will be extremely expensive for firms to implement systems to track this information. Firms that perform most investment banking business in the U.S. tend to be affiliates of many other large entities, located both domestically and internationally, that perform a wide range of services, not all of which are necessarily even financial in nature. Tracking on a real-time basis every form of compensation received by every entity from the issuer will be enormously expensive. This cost will translate into less timely research coverage. This requirement will also have anticompetitive effects, since it is a cost that will not be borne by other providers of research, such as a buy-side institution, financial periodical, or foreign broker-dealer, all of whom will be able to communicate recommendations of their analysts without the costs and burdens imposed by this requirement.

While the implementation costs will be especially onerous for the broker-dealers affiliated with large financial services affiliates, this proposal will work a different sort of hardship on small broker-dealers with few affiliates, and on the smaller companies that use them. In that situation, the tipping problem will continue to be serious.4 Small and mid-size companies might thereby find it more difficult to do any sort of investment banking transaction with a local broker-dealer that might best understand the company and the market for its securities.

The NASD, NYSE and SEC are all under similar statutory obligations not to take actions that unjustifiably impair competition, market efficiency or capital formation.5 This provision seems to impinge on these federal statutory restraints. Out of well-founded concern that they not impose a requirement that would tip market participants about nonpublic transactions, the NASD and NYSE have devised a rule that manifestly does not provide useful information to investors about potential conflicts of interest, and that imposes substantial new costs on broker-dealers, and indirectly on issuers and shareholders.

Our suggested proposal would tell investors about any public investment banking compensation received, as well as provide investors with clear, plain English warnings that analyst conflicts exist in sell-side research. In contrast, the NASD and NYSE proposals would contain a much less specific statement, not pointing to investment banking compensation and easily dismissed by readers, that a broker-dealer or one of its affiliates has in some fashion received, or reasonably expects to receive, some sort of compensation. We believe that our proposal better serves investors.

C. NASD Rule 2711(h)(1)(B) and NYSE Rule 472(k)(1)(i)(c) -- disclosure of 1% positions.

Summary of SIA Position. This proposal has the unfortunate effect of exacerbating the potential for impairing analyst objectivity, since it may put an analyst on notice of firm positions of which the analyst might not otherwise have been aware. Also, while the releases notes that NASD and NYSE intend to use sections 13(d) and 13(g) of the Securities Exchange Act of 1934 ("Exchange Act") as standards for defining "beneficial ownership," they do not apply the same percentage threshold or timing provisions. It should not be especially difficult for firms to track beneficial ownership positions at a one per cent level rather than a five per cent level for securities subject to the 13(d) and 13(g) reporting requirements, if they can rely on timing requirements similar to those used for disclosure of proprietary positions under the Exchange Act. However, from a systems perspective compliance with this proposal will be very expensive and difficult if the reporting must be based on a time period much narrower than that used for Exchange Act beneficial reporting obligations or if securities not subject to the 13(d) requirements must be included.

Recommendation: We recommend that this provision be modified to make it more consistent with other aspects of the proposed rules, by modifying the timing periods for reporting positions to match those for the proposed disclosure of ratings distributions and price charts.

Discussion. One of the principal concerns underlying both our Best Practices as well as the NASD and NYSE proposals is concern that securities analysts may be swayed in their recommendations by knowledge of their firms' economic interests in the securities issuer. These proposals may have the unintended consequence of exacerbating that concern. Other than securities positions reported under the 13D and 13G disclosure systems, a securities analyst will generally not have any reason to know whether his or her firm has any proprietary position in the subject company's securities. This provision may potentially be a new source of concern about impaired analyst objectivity. While we do not have a perfect answer for this shortcoming, the proposal would be more workable if the following concerns are addressed.

Timing Periods. The more the proposals diverge from existing reporting regimes under the federal securities laws, the more expensive and complex they will be to implement. Imposing a short time period also does not necessarily give investors a more accurate view of the firm's current position in the issuer, since research reports often have a shelf-life of many weeks. We recommend that the NASD and NYSE require disclosure of firm and affiliate positions as of the end of the most recent quarter, or the second most recent calendar quarter if the publication date is less than 15 calendar days after the most recent calendar quarter. This is the same time period that the NASD and NYSE propose for capturing data for disclosure of ratings distribution and the proposed price charts. This will make implementation easier, without sacrificing information helpful to investors.

Aggregation and Disaggregation of Positions. Consistent with the statement of the NASD and NYSE that they intend to use Exchange Act concepts to define beneficial ownership for purposes of this rule, they should permit firms to determine for themselves whether to aggregate or disaggregate positions of other affiliates, consistent with existing or future SEC guidance in this complex area. However, firms should be required to explain in their research report disclosure which approach they are following.

D. NASD Rule 2711(f) and NYSE Rule 472(f) -- imposition of quiet periods

Summary of SIA Position: This proposal has the effect of reducing the information available to investors at a time when they are most likely to want it, and will likely result in greater information disparities between large institutional investors and retail investors. We believe that to the extent possible, investors and markets are best served when all research reports are available simultaneously in the mix of information available to investors. This proposal moves in the opposite direction. Below we offer several suggestions to reduce these concerns.

Recommendations: We believe that this proposal will be inadvertently unfair to retail investors, and should be dropped for that reason. Instead, the NASD and NYSE should rely on the 25-day restriction on distribution of research that is as a practical matter implicit in Securities Act Rule 174.

Short of this, we believe the proposed rules would apply far too broadly and should be narrowed both in terms of the type of offerings and the type of issuers that trigger the restrictions. The proposals will particularly disadvantage retail investors, who unlike many institutional investors will not have other means of obtaining current research on the subject company. The proposals are also inconsistent with SEC policy encouraging greater investor access to research. As discussed below, Securities Act Rules 138 and 139 provide exceptions for research on seasoned issuers between the time that a member becomes mandated for a securities offering and completion of the offering. We believe that the reasons for granting these exceptions in the period prior to completion of an offering are equally compelling in the period following an offering. To create symmetry with Rule 138, the 10-day quiet period should not apply to equity research issued in the period following a non-convertible debt offering by a seasoned issuer. To create symmetry with Rule 139, the quiet periods should not apply to research on seasoned issuers contained in a publication that is distributed with reasonable regularity in the normal course of business.

In addition, the rules should (i) carve out research following secondary offerings of shelf-eligible issuers, (ii) exempt quantitative analyses, analyses of securities "baskets" and indices and market commentary, (iii) exclude transactions that are not "distributions" under Regulation M from the restriction on research concerning secondary offerings,6 (iv) not extend restrictions on research beyond applicable prospectus delivery periods, and (v) exempt offerings by S-3 and F-3 issuers and their equivalent under multijurisdictional offering systems entered into between the SEC and other nations.

Discussion: As proposed, this provision has numerous fundamental problems.

Discrimination Against Retail Investors. This proposal will operate to the potential disadvantage of retail investors. Institutional investors will still have other means, such as through foreign broker-dealers involved in the offering or buy-side research, to obtain current research on the subject company. While institutional investors will have a number of ways to access high-quality current research, the rule will deprive retail investors of the views of the most informed analyst. This will exacerbate any perceived or actual advantage of institutional investors in obtaining information about a company.

Inconsistent with SEC Policy. The proposal is inconsistent with the approach of providing greater investor access to research supported by the SEC. The federal securities laws effectively, if unintentionally, discourage an underwriter in a securities offering from publishing and distributing research on the subject company for fear that the issuance of the report could be deemed a "prospectus" or an "offer" in violation of Section 5.7 The SEC has recognized the excessively adverse impact that the federal securities laws have in this regard, and sought in a long series of pronouncements, most notably Exchange Act Rule 139, to address conflicts in a manner least likely to interfere with the dissemination of information to investors. These pronouncements recognize the importance of an uninterrupted flow of research analysis to the secondary market, particularly as it relates to seasoned issuers, even if a member firm is a managing or co-managing underwriter of the offering.

Following its 1969 study, "Disclosure to Investors - A Reappraisal of Administrative Policies Under the '33 and '34 Act" (the "Wheat Report"), the SEC began to adopt a series of rules designed to minimize the number and scope of restrictions on research.8 While recognizing the possibility that research issued during the offering process could unduly influence investors, the Commission has placed equal or greater value on ensuring continuous investor access to research information. For example, in proposing to eliminate most restrictions on research of large issuers during the secondary offering process, the SEC justified the change by claiming that such restrictions were not necessary due to the likelihood of multiple analysts covering the company and the belief that investors would be relatively informed about the company already.9 Moreover, in connection with extending the same proposal to seasoned issuers, the SEC stated its belief that "where the issuer has been reporting under the '34 Act for more than one year, investors have public disclosure to refer to in weighing the contents of a focused research report."10 Even unseasoned issuers of a certain size were thought to have significant market and analyst attention to warrant allowing research during the offering process under certain circumstances.11

It makes little sense to ban research in the period following the secondary offering, when selling efforts have ended and investors with a stake in the company are seeking current information and analysis. In fact, there is currently no regulation banning the publication of research following an IPO or secondary offering; rather, there is a requirement that a copy of the final prospectus accompany the research.

Thus, the prevailing SEC view of research has been to recognize its tremendous value to investment decision-making and the ability of investors to consider the relative value of this or any other piece of information in the greater context of available data, filings and opinions. In that vein, Rule 139 affirmatively permits managing and co-managing underwriters to publish research by S-3 eligible issuers (i.e., generally only the largest corporate issuers) throughout the offering process.12 The proposed rules undercut Rule 139's objectives regarding the availability of research from managing and co-managing underwriters. We do not believe that the concern over analyst conflicts, while important, justifies changing this approach.

Reduction in Flow of Information. The proposal will bar the securities analysts who may potentially have the deepest understanding of the company from conveying their insights to investors at a critical moment. In the case of an IPO, investors and markets have the greatest need for analysis about an issuer just after the offering is completed. This proposal will reduce the supply of that information. The proposal will also deprive investors of information in other contexts. For example, the proposed rules would affect research on indices, baskets or sectors that include the issuer, unless these reports are excluded from the definition of "research report" as we propose on pages 9-10 above. This type of research could become markedly less useful by causing firms to winnow down the dimensions of the analysis as numerous issuers cannot be mentioned. Comprehensive surveys of industry sectors by equities analysts would also be less useful, due to the inability to mention various companies within the sector due to the blackout restrictions.

International Implications. The concerns about reducing the mix of information and disparate treatment of retail and institutional investors are heightened in connection with issuers that raise capital globally. Currently, U.S. law and regulation already impose more restrictions on the use of research in securities offerings than exist in many other markets. This proposal takes these restrictions much farther by prohibiting any research by the managing or co-managing underwriters. In contrast, foreign broker-dealers are already able to publish research on a more current basis than U.S. broker-dealers. As a result of the proposal, for example, a U.S broker-dealer acting as managing or co-managing underwriter of the U.S. tranche of a global initial public offering by a foreign private issuer would be barred from publishing research for at least15 days longer than under current law, while overseas underwriters of other tranches of the offering would face few or no restrictions on offering research immediately. In the increasingly common situation where a company raises capital globally, this proposal puts U.S. retail investors at a greater competitive disadvantage with institutional investors who have access to research by the managing or co-managing underwriter of non-U.S. tranches of such a global offering at a time when the U.S. lead underwriters are unable to offer such research.13 It also harms the overall quality of information available to all investors by reducing the amount of research that is simultaneously available to the overall mix of information.

Application to Follow-On Offerings. The rules' application to non-IPOs is also troubling in many respects. A shelf block trade may be executed in a matter of hours. In the case of a registered shelf takedown by selling shareholders, the issuer may have no involvement at all in the offering. Since these types of offerings, as well as a "dribble out" offering by a shelf-eligible issuer can occur repeatedly, the requirement would be extremely cumbersome and disruptive to research coverage of the issuer. This concern would be reduced if the proposals exempted research permitted under Exchange Act Rule 139, and if the NASD proposal, like the NYSE proposal, exempted secondary offerings that are not considered "distributions" under Regulation M.

Length of Blackout Period. The length of the blackout period for research following an IPO is unnecessary and disadvantageous to retail investors. Under Securities Act Rule 174, the obligation of the underwriter and other dealers under Section 4(3) of the Securities Act of 1933 to deliver a prospectus expires 25 days after the offering date if the securities are to be listed on a registered national securities exchange or authorized for inclusion on NASDAQ. The blackout period for research should not be any longer than the period during which a prospectus delivery requirement exists.

E. Inconsistencies between the NASD and NYSE Proposals.

As a general policy matter, we believe that all rules of self-regulatory organizations governing the conduct of their broker-dealer members should be identical, or at least as consistent as possible. Unnecessary divergences between regulatory requirements drive up compliance costs without providing any additional investor protection.

While the NASD and NYSE worked together to develop their proposals, and the proposals are quite similar, there are some distinct differences that should be harmonized prior to adoption. In particular, as discussed on pages 8-9 above, the two proposals have very different definitions of the term "research report," on which the entire apparatus of the proposals rests. In Appendix A we cite five other examples of differences between the two proposals that should be harmonized. Unless the two rules are consistent, broker-dealers that provide research services, who are generally subject to the rules of both the NASD and NYSE, will be subject to two divergent regulatory standards and the unnecessary costs and uncertainties attendant to meeting those standards.

F. NASD Rule 2711(b)(3) and (c)(2)(C) and NYSE Rule 472(b)(2) and (3)(ii) -- Role of Supervisors in Intermediating Communications Between Research and Investment Banking or the Subject Company.

Summary of SIA Position. We strongly agree with the goal of these proposals of ensuring that equity research analysts' views or recommendations are not swayed by the business interests of the firm's investment bankers, or subject to the approval of the subject company. We also agree with the NASD and NYSE that this should be balanced with the desirability of having investment banking staff and corporate officials available to review draft research reports for factual accuracy and to identify any potential conflict of interest. However, as currently framed the proposal to have legal and compliance personnel intermediate all communications between research and investment banking concerning proposed research reports, and certain communications between research and the subject company, raises two difficulties.

1. The rule could for some purposes turn legal and compliance personnel into supervisors of research analysts, a role for which they are not suited.

2. Smaller broker-dealers are likely to find the proposal inordinately costly. The practical result may be to drive smaller firms out of the business of providing research, concentrating public research in larger firms. This in turn may reduce research coverage, especially of smaller companies.

Recommendation. NYSE Rule 472(a)(2) currently requires that research reports be prepared or approved by a supervisory analyst acceptable to the NYSE. This approach rests accountability for the content of research reports squarely with a supervisor who is in a position to evaluate the bases for the report's conclusions. The proposal may dilute the nature of this responsibility by making legal and compliance staff responsible for "authorizing" any changes that a research analyst wants to make to a rating or price target following review by the subject company. This seems to be based on an unrealistic assumption that a lawyer or compliance professional can make an informed judgment about whether a proposed change to a rating or price target is warranted. We recommend that the NASD and NYSE clarify that this authorization should come from a supervisory analyst, but should be documented and retained by legal or compliance personnel.

We also recommend that prior to approval by the SEC or final adoption, the regulators examine the impact that this proposal may have on smaller broker-dealers and their coverage of smaller companies.

Discussion. The proposals would have legal or compliance staff act as intermediaries for written or verbal communications between research and investment banking staff or certain written communications with the subject company concerning draft research reports, and require that legal and compliance staff give written authorization for any changes following communications about a draft research report with the subject company.14 We support the NASD's and NYSE's objective of ensuring that no pressure is applied on a research analyst by either investment banking or the subject company to change his or her recommendation.

However, it is not clear from the proposal how legal or compliance staff are expected to determine whether or not to "authorize" changes in ratings or price targets following communications between research analysts and corporate officials. This is asking a legal or compliance officer to play a role for which he or she is not well suited. A compliance professional or lawyer is unlikely to be able to independently assess the credibility of a claim by a research analyst that a recommendation was changed as a result of information given by the subject company. Senior research management, however, is suited to play this role. The purpose of requiring legal and compliance intermediation is, we assume, to ensure that a research recommendation is made in good faith and without improper interference. Legal and compliance personnel can perform this function by screening communications for overt signs of pressure and ensuring that parties with the right expertise are brought in where appropriate.

The proposal is likely to pose significant costs on smaller broker-dealers that perform research functions. A firm that employs a small number of research analysts, faced with the need to hire and train new personnel to monitor a wide range of communications involving its research staff, may conclude that its research function is no longer justified. The impact will be most pronounced for smaller broker-dealers that have "niche" research practices that focus on smaller segments of the market. We recommend that the regulators examine the extent to which this proposal will diminish research coverage of smaller companies, and that they consider that effect in deciding whether and how to proceed with this proposal.


Thank you for giving SIA this opportunity to comment on the NASD and NYSE proposals regarding research analyst conflicts. We believe that the proposals will substantially improve public trust and confidence in research analysts, with the suggestions that we offer above, as well as by addressing the requests for guidance and clarification contained in Appendix A to this letter. If you have any questions on any aspect of this letter, please contact either me or George Kramer, Vice President and Associate General Counsel, at 202-296-9410, or by e-mail to gkramer@sia.com or skaswell@sia.com.


Stuart J. Kaswell
Senior Vice President and
General Counsel


Appendix A: Interpretive and Definitional Questions

Cc: Chairman Harvey L. Pitt, U.S. Securities and Exchange Commission
Commissioner Isaac C. Hunt, Jr., U.S. Securities and Exchange Commission
Commissioner Cynthia A. Glassman, U.S. Securities and Exchange Commission
The Honorable Michael G. Oxley, Chairman, House Financial Services Committee
The Honorable Richard Baker, Chairman, Subcommittee on Capital Markets, Insurance, and Government Sponsored Entities
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
David M. Becker, General Counsel, U.S. Securities and Exchange Commission
James A. Brigagliano, Assistant Director, Trading Practices, Division of Market Regulation, U.S. Securities and Exchange Commission

1 SIA brings together the shared interests of nearly 700 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. The U.S. securities industry manages the accounts of nearly 80 million investors directly and indirectly through corporate, thrift and pension plans. The industry generates over $350 billion of revenue and employs approximately 760,000 individuals. (More information about SIA is available on our web site: http://www.sia.com).
2 See letter from Alden S. Adkins, Senior Vice President and General Counsel, NASD Regulation, Inc., to Yoon-Young Lee, Wilmer, Cutler & Pickering, December 17, 1999 (available at www.nasdr.com/2910/2210_07.asp). The position would be subsumed in the changes to the NASD public communication rules published by the SEC at the end of December 2001 that would grant relaxed treatment to all institutional and certain other correspondence. See Rel. No. 34-45181, File No. SR-NASD-00-12, 66 Fed. Register 67586 (December 31, 2001).
3 See Sections 14(e) and 15(f) of the Securities Exchange Act of 1934 ("Exchange Act") (establishing liability for disclosure of material non-public information about corporate transactions), Exchange Act Sections 15(f) and 21A(b) (requiring that broker-dealers maintain internal controls to prevent misuse of such information, and treble-damage liability for failure to do so), U.S. v. O'Hagan, 521 U.S. 642 (1997) (liability under Exchange Act Section 10(b) for misappropriating material nonpublic information about corporate transactions), NASD NTM 91-45 and NYSE Information Memo 91-22 (June 28, 1991) (establishing minimum information wall requirements within broker-dealers).
4 For example, suppose that Acme Inc., a Wilshire 5000 public company, decides to explore a possible merger, and hires Alpha Investments, a "home town" broker-dealer that has relatively few outside lines of business outside investment banking, research on local companies, market-making and retail brokerage. Disclosure in a research report that Alpha expects, for the first time, to receive compensation from Acme might very well tip recipients of the research that a major corporate event involving Acme is underway. Alternatively, Alpha could suspend its research on Acme, but that could be just as likely to tip clients that a major corporate event is underway.
5 Exchange Act 3(f) requires the SEC to consider, as part of its weighing of the public interest in reviewing a proposed rule of s self-regulatory organization, "whether the action will promote efficiency, competition and capital formation." The NASD and NYSE's rulemaking authorities are under similar obligations. Section 15A(6) of the Exchange Act requires that any rule of the NASD "remove impediments to and perfect the mechanism of a free and open market . . . and, in general, to protect investors and the public interest." Exchange Act Section 6(b)(8) requires that the NYSE's rules "not impose any burden on competition not necessary or appropriate in furtherance of the purposes of this title."
6 The NYSE proposal contains such an exemption, and we believe that the NASD proposal should do the same.
7 The issuance of the report could also be deemed an inducement to purchase in violation of Rule 101 of Regulation M.
8 Rules 137-139. Securities Act Release No. 5101 (November 19, 1970).
9 The Regulation of Securities Offerings ("Aircraft Carrier"), Securities Act Release No. 33-7606A (11/13/98).
10 Id.
11 Id.
12 Rule 139 also permits publication of research on non-S-3 issuers provided that the research must (i) be in a publication that is regularly distributed to investors, (ii) contain similar coverage with respect to a substantial number of companies in the issuer's industry or contains a comprehensive list of securities currently recommended by the issuer, (iii) give no greater prominence to the subject company in the publication than to securities of other issuers, and (iv) not contain any recommendation on the issuer more favorable than in the last research report issued prior to the broker-dealer's participation in the distribution.
13 It may be possible in some situations that both retail and institutional investors could find such research over the Internet. However, institutional investors would be more likely to know where to look for overseas research.
14 The NASD noted in the release accompanying its proposal that the proposed rule change "would not restrict or impose any conditions on any communication between a research department and an investment banking department that does not concern a proposed research report." 67 Fed. Register at 11533.

Appendix A.

Interpretive and Definitional Questions.


"household." It is not hard to think of situations where it would be difficult for an analyst or his or her employer to monitor or restrict other household members. For example, an analyst who just graduated from business school might share an apartment with roommates who move in and out every few months, or an analyst with young children might hire a live-in childcare or au pair. It is not clear how the analyst or the firm could restrict or monitor the share ownership of such household members. It may also be difficult to try to require financially independent adult relatives residing in the house, such as parents or siblings to abide by these restrictions. This definition should be scaled back to household members who are financially dependent on the analyst.

"research analyst." It should be clarified that, in addition to tightening the definition of "research report" as discussed on pages 7-8 of our letter, this term should be limited to members of the equity research department. It should not sweep in personnel such as a salesman sending an e-mail to a client concerning a security. Other NASD and NYSE rules already cover recommendations to customers.

Additionally, we note that the SEC had expressed concern that earlier drafts of the NASD proposal might inadvertently encompass the normal activities of investment advisers although they were not principally responsible for the preparation of research reports.1 To address such concerns, the NASD added language to its release to reflect that mutual fund portfolio managers would generally not be deemed to be "research analysts," even if the portfolio manager is an associated person of a broker-dealer. We believe that the intent of the NASD and NYSE is to exclude all investment advisers who are not principally involved in the preparation of research reports from the definition of the term "research analyst," not just those who advise mutual funds. Therefore, the final NASD and NYSE proposals should clarify and confirm all investment advisers not principally involved in the preparation of research reports are excluded from coverage, not just mutual fund portfolio managers.

A definition is needed for "affiliate" regarding disclosure of 1 % positions and compensation. Presumably this is intended to only include corporate affiliates where the broker-dealer or a parent entity has a controlling interest. We assume that the intent is not to include individuals, such as officers, directors and employees of the broker-dealer or its affiliates, because doing so would make the proposal unworkably complex and expensive, as discussed at page 15 of our letter.

Intermediation Between Research and the Subject Company.

The proposal should clarify that firms would only be required to retain copies of reports that actually are changed following submission to the subject company.

As noted on page 28 of our letter, we question whether written authorization from legal or compliance for ratings and price target changes following consultation with the issuer is appropriate. We suggest changing the wording from "written authorization from" legal or compliance personnel to "documentation that legal or compliance staff have reviewed the justification."

Prohibition on Promises of Favorable Research.

The NASD and NYSE should clarify that it should not be improper under this rule for a broker-dealer to obtain the analyst's view of a company, or to consider whether the view is unfavorable, before committing to an investment banking transaction. Moreover, an issuer has a legitimate interest in knowing ahead of time if a potential underwriter has a favorable opinion. The problem is that the bar applies not only to directly or indirectly offering favorable research as consideration for the receipt of business or compensation, but also offering such research as an "inducement" for such business or compensation. This could be interpreted as limiting the broker-dealer's ability to seek its own analyst's view of a potential investment banking client, as well as forbidding any response to an inquiry from the potential client about the firm's general opinion of the company, for fear of indirectly "inducing" the transaction through a promise of favorable research. We suggest deleting the word "inducement" to make it clear that the rules are not intended to have such a broad sweep. We strongly support this proposal in all other respects.

This provision should also extend to promises of specific ratings and price targets to an issuer's affiliate, or any such promise to an issuer as inducement for business or compensation from an affiliate of the issuer.

Restrictions on Personal Trading.

As drafted, these requirements are likely to amount to a complete bar on analyst ownership of securities that he or she covers. Some firms have taken this action on their own, but there are strong policy reasons to let firms make this decision for themselves. In addition to the clarifications suggested below, the most important change to preserve the practical ability of analysts to own securities that they cover is to limit the scope of the restriction so that it only restricts purchases or sales of the securities of a subject company within 30 days before, or 5 days after, issuance of a change in the analyst's recommendation or price target.

The NASD and NYSE should clarify whether the ban on pre-IPO issues bars all private equity investments, or simply prohibits coverage if and when the private company goes public. Similarly, clarification would be helpful as to whether an analyst can satisfy proposed NASD Rule 2711(g)(1) or proposed NYSE Rule 472(e)(1) by disposing of a private equity investment prior to the IPO.

The NASD and NYSE also need to address grandfathering issues (e.g., an analyst who holds prohibited private equity investment when rule becomes effective). In addition, the NASD and NYSE need to address situations where receipt by an analyst of pre-IPO shares is involuntary - for example, shares received through inheritance, or as merger consideration when a company that constitutes a permitted investment merges into a company that constitutes a prohibited pre-IPO investment. In all of these cases, it should be recognized that disgorgement of a pre-IPO investment is not always possible. Private holdings are often subject to narrow contractual restrictions on transfer that make disgorgement impracticable. We suggest that the rules be modified to permit an analyst to hold the securities after the company goes public and the analyst commences research coverage, provided that disgorgement is impracticable and the analyst submits to a long-term lockup.

Firms will need guidance on the meaning of the term "specific news or a significant event" that triggers an analyst's ability to sell within 30 days of a research report. For example, the standard used for filing Form 8-K or a press release would be one approach that would make this provision easier to apply.2

The exception provided for sales when an analyst changes firms or picks up an issuer should be expanded to apply to purchases to cover a short position.

Clarification is needed on how the exceptions permitting analysts to purchase or sell a security within 30 days of issuing a research report work with the prohibition on transactions made that are inconsistent with the analyst's most recent recommendation. For example, is it the intention of the NASD and NYSE that an analyst, after changing firms and selling securities that he owns in a company he begins to cover, is then barred from issuing a positive report on the company?

Guidance is needed on the "significant personal financial circumstances" provision. In its accompanying release, the NASD cites an unforeseen medical emergency as its only illustration of a significant personal financial circumstance. While we agree with the NASD's statement that "[r] eliance on this provision should be rare," we hope that it will be more flexibly applied than that. Other unusual personal expenses, whether or not they are "unforeseen," should be permitted under this exception. Examples might include paying adoption expenses, expenses related to recovery from a natural disaster, terrorist attack or act of war, or funeral expenses for a family member. Obviously an exhaustive list is not feasible, and some reasonable flexibility should be permitted.

More clarity is needed on the prohibition of pre-IPO investments by research analysts in companies "principally engaged in the same types of business as companies that the research analyst follows." E.g., are "pharmaceuticals and "health care" the same types of business? Semiconductors and software producers? Textile manufacturers and clothing retailers? The same issue is raised under the NASD's proposed Rule 2711 (g)(5)(B)(ii) and Rule 472(e)(4)(v) of the NYSE's proposed amendment. What if the nature of a company changes as technology or business model evolves? Real world examples of such evolution include an internet or networking company evolving into a telecommunications firm, or a mining and manufacturing firm evolving into an office supplier.

Public appearances.

This proposal may greatly limit public appearances by sell-side analysts because of the extensive checking that would have to be done before the appearance and the consumption of airtime making all the required disclosures in a public appearance. Analysts making public appearances may have to require assurances that they will be told in advance about any company that may be mentioned so that they can perform this due diligence. This may preclude analysts from participating in many types of appearances, such as where audience or viewers ask questions, or panel appearances that discuss a range of companies. The only feasible format may be one-on-one interview where only discussion is about companies agreed upon in advance.

It would also be helpful if the NASD and NYSE could clarify that analyst participation in open conference or web cast calls with the issuer would not be a "public appearance" triggering the disclosure obligations of the rule. If analysts were required to make the required disclosures in this context, the constant interruption as different analysts prefaced their questions with the disclosures would be disruptive and make the presentation less useful to investors listening to the call.

Distribution of ratings.

The proposals require member firms to disclose ratings distributions in categories labeled "buy," "hold" and "sell." While we have no objection to the use of three categories in presenting this information, we do not believe that members should be required to define the three categories using these labels. Not all member firms rate their stocks "buy," "hold" and "sell." These terms carry meanings that are not consistent with the ratings definitions used by some members. The proposals should retain the three category requirement but should be modified to allow members to use terms other than "buy," "hold" and "sell" to the extent necessary to accurately reflect the ratings definitions used by a member in its rating system.

The NASD and NYSE should clarify whether ratings distributions should be calculated including foreign affiliates or companies outside the U.S.

Price chart.

The NASD and NYSE should clarify how the price chart rule will apply to multiple rating systems (e.g., short, intermediate and long-term ratings)? They should also indicate whether relevant benchmarks, such as market or sector information, can be included in the price chart.

Unless industry and multi-company reports are excluded from definition of "research report," as discussed at pages 9-11 of our letter, those reports will be overwhelmed by inclusion of multiple charts. We suggest that if these sorts of reports are included in the definition, it should be permissible in those cases to refer to other recent reports for charts, or alternatively, to a web site.

A requirement to begin tracking changes in price targets for the prior three years on a price chart will be problematic for many firms. Many firms do not currently keep data on each target price move over the last three years, or can only reconstruct it with difficulty. We understand that services that keep track of ratings and other estimates generally do not keep track of target price changes. This proposal will need to be implemented over a period of time.

Application of Rule 2210.

It is unclear if underwriting and director footnotes required under NASD Rule 2210 would still be still applicable even though both conflicts are addressed more comprehensively in the proposed rules.

Discrepancies Between the NASD and NYSE Proposals.

1. As discussed at page 8 of our letter, the NASD and NYSE have markedly divergent definitions of the key term "research report." Harmonizing these definitions is critically important.

2. NASD Rule 2711(g)(2)(A) permits a research analyst account to "sell all of the securities held by them" (emphasis added) within 30 days after the analyst begins following the issuer for the company. By contrast, NYSE Rule 472(e)(4)(iii) and (iv) permits sales of less than all of the securities held by an analyst or household member in this situation. We cannot see a reason for the NASD's restricting this provision to sales that entirely close out the account of an analyst or household member.

3. NASD Rule 2711(h)(5)(A) contains no limitation on the time period over which a research report must disclose the historical data on the firm's use of "buy," "hold/neutral" and "sell" ratings. Presumably this is an oversight, since historical data going back many years would be of little use to investors. NYSE Rule 472(k)(2)(v) contains a three-year period. While this is better than no time limit at all, even this may skew investors' understanding of the firm's recent tendencies toward various rating categories. For example, a firm might have had a higher percentage of "buy" ratings during a strong macroeconomic growth period, but a lower percentage during a current recessionary period. Data going back a shorter period, such as six months or one year would be more indicative of the firm's proclivities than data going back three years, which would be more likely to capture ratings that were made in a completely different economic cycle. In addition, if the data covers a shorter period, the necessary implementation time for firms to construct this data also will be shorter.

4. NYSE Rule 472(b) seems to indirectly indicate that compliance, research and investment banking should be housed in separate departments. NASD Rule 2711(b) uses different wording, but also seems to imply that firms' research and investment banking functions, although not necessarily compliance functions, should be housed in separate departments. SIA prefers the NASD formulation, since it does not suggest a separate organizational structure for legal or compliance personnel. We do not think that it is necessary or appropriate for either proposal to mandate a particular organizational structure. As long as firms comply with the regulatory requirements it should be immaterial what organizational structure is used.

5. The NYSE proposal appears to restrict analyst trading in all securities covered by the firm's research department, not just those covered by the analyst. This should be conformed to the NASD restriction on trading in securities covered by the analyst.

6. Section .110, Supplementary Material, of the NYSE proposal limits the 10-day quiet period for secondary offerings to securities distributions subject to Exchange Act Regulation M. The NASD proposal has no comparable exemption. As we discuss at page 26 of our letter, we believe that this exception appropriately limits the application of the 10-day quiet period to true distributions, as opposed to sales of smaller magnitude that have more in common with secondary trades than primary distributions. However, the exemption has little utility unless the NASD adopts a parallel exemption.

1 See speech of Paul F. Roye, Director, SEC Division of Investment Management, before the IA Compliance Summit (April 8, 2002). Available at: http://www.sec.gov/news/speech/spch549.htm.
2 The term "specific news or a significant event" is also used in the provisions on quiet periods, and the same guidance is needed with regard to the term as it is used in those provisions.