April 19, 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 SR-NYSE-2002-09
Dear Mr. Katz:
Credit Suisse First Boston ("CSFB") is submitting this letter in response to the request by the Securities and Exchange Commission (the "SEC") for comments regarding rule proposals submitted to the SEC by the National Association of Securities Dealers, Inc. (the "NASD") and the New York Stock Exchange, Inc. (the "NYSE") relating to the NASD and NYSE member firms ("member firms" or "firms") research analysts' conflicts of interest and to the conduct of firms' research analysts (the "Proposed Rules").1
We appreciate the opportunity given to us by the SEC to comment on the Proposed Rules. CSFB is committed to the goal of fair and objective research and we are concerned about the loss of investor trust and confidence in equity research. We are committed to protecting the research analysts from conflicts of interest and educating the investing public to any potential conflicts. It was for this reason that CSFB worked on the creation of and adopted the Securities Industry Association ("SIA") Best Practices for Research. CSFB also adopted restrictions beyond those required by the Best Practices by prohibiting ownership, by CSFB analysts and their immediate family members, of securities in companies on which they provide research coverage. CSFB commends the efforts of the NASD and the NYSE for their hard work in this area, and we also commend the leadership of the SEC in encouraging and supporting the efforts of the NASD and NYSE.
CSFB supports both the objectives and the premise of the Proposed Rules: that disclosure cures potential conflicts. Our suggestions are intended to further their purpose and to offer substantive solutions to issues that we believe can be resolved by further clarification and greater specificity. CSFB has joined the efforts of the SIA and its members in submitting a comprehensive comment letter addressing the Proposed Rules. We submit this letter independently to focus on particular issues that are of concern to this firm and certain issues that are technical in nature and not fully addressed in the SIA's letter.
I. Background and Facts
CSFB is a member firm of both the NASD and the NYSE (collectively the "SROs"). CSFB is a global financial institution that provides a full range of investment banking and other securities services in the United States and throughout the world. CSFB's equity research is widely recognized as among the best in the industry. In the Americas alone we have more than 100 analysts covering some 85 industries and over 1200 companies; in Europe over 100 analysts cover more than 600 companies and 80 emerging European companies; and in Asia we have over 70 analysts covering 400 companies in ten countries. CSFB's research products are circulated throughout the world.
CSFB produces and distributes a tremendous volume of reports and analyses to our customers globally. In the United States, CSFB issues approximately 300 "research reports" on a daily basis during peak earnings reporting season and between 150 and 200 reports outside the earnings reporting season. This estimate is based on a conservative reading of the definition of the term "research report" as used in the Proposed Rules - a broader reading would result in a larger number. On a global basis, CSFB produces approximately 500 reports on a daily basis during peak earnings reporting season and between 250 and 350 reports outside the earnings reporting season. Given the amount of reports and analyses that we distribute on a global basis, we have great interest in the various issues addressed in the Proposed Rules, and we believe that the intent of the Proposed Rules is to allow the continued publication of this information to and for the benefit of investors.
We have expressed our position and made recommendations with respect to the following issues:
II. Discussion of the Issues
A. Disclosure of Expected Compensation
Under the Proposed Rules, CSFB would be required to disclose in its research reports whether either the firm itself or any CSFB affiliate has received compensation from the subject company in the past 12 months or expects to receive compensation within the next three months. We suggest that the disclosure of expected compensation from the subject company be limited to a warning about a possible future engagement of business, or a solicitation thereof, which could result in future compensation from the subject company. This type of warning will inform the investor of the potential conflict of interest without requiring disclosure that, at best, is speculative. We believe that the combined requirements to disclose past compensation from a subject company received by a member within the past twelve months together with a warning about possible future engagements or solicitations is sufficient disclosure and notice to investors of potential conflicts of interest with respect to compensation from the subject company.2
In the U.S. alone, CSFB covers approximately 1,200 companies in Equity Research and 5,000 companies in Investment Banking. The number of companies to which CSFB and its affiliates provide financial services, including brokerage and banking services, is far greater. Outside the United States, the numbers increase significantly. Creating a comprehensive tracking system for all "compensation" received by any CSFB entity from any subject company will be a mammoth task to implement and will cost millions of dollars. Even if CSFB is able to determine that it had not received compensation from a subject company in the past 12 months, CSFB still would be required to determine whether it has any expectation of receiving compensation of any kind in the next three months. This required disclosure will place an even greater technology and systems burden on CSFB because there is no global tracking system for the firm's many business lines (i.e., sales, trading, derivatives, banking, advisory, finance, etc.) that would enable us to determine whether compensation is expected from a subject company. As CSFB's tracking and disclosure obligation under the Proposed Rules would, in terms of time, expense and re-allocation of valuable resources, far exceed the marginal benefit that investors might gain from this forward-looking disclosure, we believe that the better course is to warn the readers of research product that investment banking is soliciting a subject company.
Beyond the problem of tracking and aggregating this information, the two principal challenges with respect to the required disclosure of "expected compensation" are: (1) the required disclosure could reveal significant client confidences and potentially tip the market to pending unannounced transactions; and (2) there are no objective standards as to what constitutes a reasonable expectation.
First, the scope of "compensation" intentionally has been designed to be so broad that, in the view of the SROs, it will not tip the research department (or a reader of their product) to non-public transactions. This suggests that expected compensation information underlying the disclosure (whether a potential fee relating to a major unannounced merger assignment that would move market prices or a potential brokerage commission of $500 for a possible sale of 1,000 shares of stock by an Asian company that would be paid to an Asian affiliate of CSFB) is to be made in such a generic manner as to make the disclosure effectively the same as a disclaimer.3 If investors knew the source of the compensation, they would recognize the significance of the information. However, disclosure of what investors would want to know, such as the member firm's expectation of fees from an unannounced transaction, is appropriately prohibited under existing securities law. Accordingly, meaningful disclosure is not a viable option and clearly would have significant negative collateral consequences. The disclosure requirement contained in the Proposed Rules appears to be designed to have little or no value to investors, yet investors may be able to read between the lines, correctly or incorrectly, depending on the particular disclosure.4
More importantly, in certain instances this disclosure could serve as an accurate tip of information that was not intended to be made public. Rather than make such tips, firms may be discouraged from publishing research that could disclose confidential client relationships. Further, in many instances, the existence of a particular client relationship is itself highly confidential and often subject to contractual restrictions on disclosure. Given the sensitive nature of particular transactions and the potential for speculation about the imminence of a non-public transaction, such confidential relationships are highly desirable and often requested by clients. The Proposed Rules would prevent us from keeping client relationships confidential. Moreover, in particularly sensitive transactions, certain clients may be reluctant to rely on our services, or the services of one of our affiliates, when this would entail disclosure of the client relationship in one of our research reports.
Second, the Proposed Rules do not provide any criteria by which to decide whether a firm "reasonably expects to receive compensation." Is compensation reasonably expected when the member firm is actively soliciting business from the company? When the firm has an engagement letter on file? When the firm has booked a receivable? Or is it sufficient that in the past 12 months the firm has regularly received compensation from brokerage or banking accounts? We suggest that the SROs provide clear objective standards, such as the existence of an accrued receivable and/or a signed engagement letter or contract, to guide the industry with respect to their expectations for compliance with this provision of the Proposed Rules.5
B. Disclosure of 1% positions
Under the Proposed Rules, a member firm is required to disclose in research reports or public appearances the beneficial ownership by itself or by an affiliate of 1% or more of any class of common stock of the subject company. However, the NASD and NYSE rule proposals differ in their statements of how the 1% ownership is to be calculated, and consequently it is unclear whether CSFB should apply the definitions and interpretations set forth under Section 13 of the Securities Exchange Act of 1934 (the "Exchange Act") or apply a new standard. It would be unduly burdensome to require firms to develop a new and separate system to track disclosure of the same type of information for purposes of the Proposed Rules when the existing disclosure standards are adequate to safeguard investors' interests. Lowering the threshold from 5% to 1% alone will require major expenditures in information technology.
In addition, it is unclear whether member firms can aggregate or disaggregate the positions of other affiliates. The Proposed Rules should specifically allow the application of the existing aggregation and disaggregation rules and interpretations under Section 13 of the Exchange Act.
C. Definition of "Research Report"
We believe that the expanded disclosure requirements are appropriate for stand-alone company specific research reports which contain the core investment rationale developed by the research analyst and which all customers typically receive. The proposed definition of "research report," however, is too broad in that it would include many types of communications, such as quantitative-based research and market commentary, that are not subject to the risks of a conflict of interest that motivates the Proposed Rules. Another ill effect of the overbroad definition is that the increased disclosure required on certain types of communications, including industry or sector reports that cover numerous issuers, would greatly expand the size of these reports past the point where they could be easily distributed (either in hard copy or electronically), and they would make the reports virtually unreadable. For example, the Proposed Rules would increase the size of abstracts, updates and weekly/monthly summaries by a multiple of 6 or 7. Our CSFB First Edition U.S. Alert, a daily publication, could increase from 9 pages to 66 pages with the required disclosures. Our Monthly Review and Comment could increase from 252 pages to 1,544 pages under the Proposed Rules. It is not clear that CSFB could reasonably continue (based on economic and technological limitations) to produce these reports or that investors would want to (or be able to, from a technological perspective) receive them even if they could be produced. It is important to note that, with respect to abstracts and other summary materials, these derivative publications are distributed to clients in addition to the full, stand-alone research reports, which would contain all the required disclosure statements.
We do not believe that the expanded disclosure requirements are appropriate for the following types of communications, and we suggest that they be exempted from the definition of "research report":
1. Abstracts, Updates, Weekly and Monthly Summaries, Industry/Market Sector Reports and Portfolio Strategy Pieces
CSFB produces many research publications that summarize, or refer to, traditional stand-alone research reports. Virtually all broker-dealers, including CSFB, also distribute abstracts of research reports via First Call, Multex, and other third party electronic providers. These providers use technologies (such as ASCII text file) that are inadequate to contain complete research reports, let alone all the disclosure statements that would be required under the Proposed Rules. Accordingly, if the SROs imposed the full disclosure requirements contained in the Proposed Rules on research distributed through these providers, the effect would be to make it impossible to distribute research through these providers - such research would then become unavailable to retail investors because this is the primary distribution vehicle for retail investors. We recommend that these abstracts, summaries, industry/sector reports and portfolio strategy pieces be exempted from the definition of "research report" where the underlying materials have been prepared and distributed in accordance with applicable rules.
In the alternative, these types of reports could bear a legend that: (i) states that they are not full reports; (ii) specifies how the full research reports can be obtained; and (iii) states that the full research reports contain valuable information and disclosures. To the extent that it were perceived to be important that the disclosure be available even in respect of these summary reports, it is possible that the summary reports could reference a website or other readily accessible location where the relevant disclosures could be found, similar to the posting requirements for order execution quality set forth in Rules 11Ac1-5 and 11Ac1-6 under the Exchange Act. This would allow member firms to continue the distribution of summary reports in a timely and efficient manner.
2. Quantitative Research/Technical Analysis
Quantitative research products may recommend trading strategies or provide valuations or forecasts based entirely on quantitative or technical analysis. Further, CSFB also makes available various types of valuation software that may implicitly be used to generate trading recommendations in accordance with an investor's self-selected criteria. As these software valuation tools rely upon objective numerical criteria (such as stock prices, indices, interest rates, etc.) and not an analyst's subjective view, we do not believe that there is any meaningful risk that an analyst's conflict of interest bias will be introduced. Accordingly, these types of reports should be exempted from the proposed definition of "research report."
3. General Market Commentary and Trading Strategies
Clients may ask a trader or salesperson to provide, on a daily or intra-day basis in some cases, the trader's or salesperson's views on overall market conditions, conditions of particular segments of the market, or the effect of certain news events. The responses to these questions are only that person's opinion on any of these broad topics. There is no opportunity for these types of communications to be tainted by the kind of bias that the Proposed Rules seek to address. We therefore recommend that communications providing market commentary be excluded from the definition of "research reports." In this respect, we note that the NASD has proposed to grant an exception from its pre-approval rules for such communications with institutional investors, although the NYSE has not taken a formal position in this regard. These inconsistent views by the SROs have caused confusion and concern in the securities industry. We suggest that market commentary be covered by the current rules and regulations governing correspondence.
D. Disclosures in Connection with "Public Appearances"
The definition of "public appearance" in the Proposed Rules is without practical limitation and is broad enough to include open conference calls, webcast calls, or appearances by research analysts on a panel. Under the Proposed Rules, a research analyst at any such routine event would have to provide all required disclosure for every company that is mentioned at any of these events. We do not believe this was the intended result of the definition.
If this was the intended result of the definition, then prior to speaking at any such event, an analyst would have to check CSFB's worldwide compensation (and expected compensation) from every subject company that might be mentioned at the event. Any such requirement, even if it were possible to gather the relevant information, would be quite time consuming. Further, the time required for the analyst to provide the audience with the required disclosures would likely take much of the time that might otherwise be given to substance. This would obviously reduce the informative character of the event and, we suspect, be ignored by listeners.
The end result of such requirement would be that analysts would limit their participation in these public events, or if they consented to appear, would agree to discuss only a few selected companies so as to avoid violating the disclosure rules. This would have the effect of decreasing the amount of available information to investors. Accordingly, we believe that any final version of the Proposed Rules should exclude from the definition of "public appearances" events such as webcasts, conference calls, panel appearances and other such appearances where a large number of companies (that may or may not be pre-determined) might be discussed. As an alternative, the Proposed Rules could require that a research analyst state that his employing firm receives compensation from numerous issuers, likely including issuers that will be discussed in the panel, and that further information about any potential conflicts regarding companies covered by the firm's research department is available on the firm's website.
E. The Price Chart Requirement
Under the Proposed Rules, a member must include a price chart in any research report concerning an equity security on which the member has assigned any rating for at least one year. The price chart must indicate the dates on which the member assigned or changed each rating or price target and depict each rating and price target assigned or changed on those dates. However, certain electronically transmitted research reports do not technologically support the inclusion of such price charts. For example, First Call and other applications that accept ASCII text files (used particularly in the retail channel) are incapable of supporting price charts. Therefore, we recommend that the Proposed Rules allow the price charts to be substituted with tables whenever tables are more practical.
F. Delay of Effective Date
CSFB requests a reasonable delay of the effective date of provisions of the Proposed Rules that require substantial information technology and systems development and enhancement.
* * *
We appreciate the opportunity to submit this comment letter regarding the Proposed Rules. We would be pleased to respond to any inquiries regarding this letter or our views generally on the Proposed Rules. Please contact Mary L. Whalen at (202) 354-2801 or Raymond J. Dorado at (212) 325-7258.
Very truly yours,
Credit Suisse First Boston
|1||SEC Release No. 34-45526 (March 8, 2002) (the "Release").|
|2||We believe that the following standardized disclosure could be made part of a research report when applicable: "Prospective investors should be aware that [Research Firm] has received fees over the last twelve months in return for services provided to XYZ Corp. Investors should assume the company is the subject of ongoing solicitation by the [Research Firm] for future business which may result in the payment of future fees to [Research Firm].|
|3||The NASD and the NYSE assert in the release accompanying the Proposed Rules that the Proposed Rules change would not alert the Research Department or the investing public to non-public transactions because of the limited substance of the disclosure and the suggested broad interpretation of "compensation." (See 67 Fed. Register at 11534). We do not understand the SROs' logic in requiring firms to make disclosure (and imposing the very large costs of gathering the needed information), but at the same time asserting that the disclosure does not raise legal issues because the information cannot be linked to a particular service of the member firm. If the information is not meaningful to the investor's assessment of a potential conflict of interest, why go to the very substantial and wasteful expense of tracking it?|
|4||A disclosure stating that "the member firm has not received compensation during the last twelve months but expects to receive compensation in the next three months" may suggest that the subject company is involved in an unannounced transaction.|
|5||Additionally, we request that the SEC publicly provide interpretive guidance and support their view that such forward-looking compensation disclosure would not constitute tipping in violation of Section 10(b) of the Securities Exchange Act of 1934, as amended.|