A. Overview of the FINRA Research Proposal
FINRA has issued for consultation its proposed equity research rules for the Consolidated FINRA Rulebook (the “FINRA Research Proposal”).1 The proposals include significant changes to the current NASD and NYSE research rules that warrant attention and comment. In particular, FINRA’s proposals relating to the “fair distribution” of research raises some very fundamental issues as to whether and how the writing and distribution of research can be profitable. FINRA’s proposals may also increase the possibility of firms being subject to liability when distributing research. In addition, the proposals require firms to revise and expand their compliance policies and procedures relating to research, particularly those governing management of conflicts between investment banking and research departments.
In addition to making substantive changes to the current NASD and NYSE research rules, FINRA’s proposed research rule incorporates many requirements of the current research rules and liberalizes other provisions, including those relating to personal securities transactions by research analysts, and research analyst registration requirements.
Section 1.B of this memorandum raises some philosophical concerns with the FINRA Research Proposal. The remainder of this memorandum is organized to address the FINRA Research Proposals in the order of importance they are likely to have to firms. Thus, Section 2 of this memorandum addresses issues relating to the distribution of research, and discusses the cost and liability implications of the proposals. Section 3 addresses proposals to require firms to establish policies and procedures to identify and manage conflicts between research and investment banking departments. Section 4 discusses amendments to disclosure requirements applicable to research reports. Section 5 discusses the remainder of the FINRA Research Proposal, including those requirements of the current rules that FINRA intends to liberalize.
The comment period on the FINRA Research Proposal expires November 14, 2008.
B. Background and Economics
The background to the FINRA Research Proposal, and of course to any regulatory pronouncement in areas related to research, is the Settlement (as defined in footnote 1) resulting from the allegations that the contents of research reports by analysts and investments banks were improperly motivated by the desire to win investment banking business. Following the Settlement, the SEC and other regulators were criticized for not being aware of (or at least sufficiently focused on) the connection between investment banking and research that resulted in the tainting of research. The resulting rules were intended to decouple the economics of investment banking and research and hopefully lead to “pure” research.
While I think the basic concept of separating research from the influence of investment banking is desirable, that leaves open a fundamental question: “What are the economic benefits to an investment bank of producing research?” If investment banks cannot make a profit, direct or indirect, from producing research, they will not produce it. The profit must come from increased trading volume or other revenues from readers of research who are willing, in some way, to pay for that research.
Thus, the economic question, and I think therefore the regulatory question is this: How do we design a system for the production and distribution of research that is both “fair” and honest, on the one hand, and profitable on the other?
In regards to maximizing fairness, let me propose one approach, albeit a theoretical and I think unworkable one. All investment banks that produce research would have to immediately publish that research on their public websites which anyone can view, thus making the research as widely available as possible. The entire investment community thus sees the research and can act on it at the same time. This seems perfectly “fair.” The problem, of course, is that there is no profit in it.
No one is going to pay the investment bank for research that is required to be publicly available on the Internet.2
One step away from the public distribution on the Web is distribution to all “customers” of the firm. The problem with this approach is that it is not so different from the Web approach: if customers are broadly defined, (i) the value of the research quickly depreciates in proportion to the number of people who know it;3 and (ii) if research is distributed widely enough, it will quickly be recirculated and redistributed through the Internet so that the research might as well be posted on the Web.
If research is not to be made freely available, then it follows that its distribution should be limited in a way that allows the writers of research to make a profit. One solution to this is to allow firms to control the distribution of research, sending research to customers who send them substantial trading business before those who do not. There is, of course, a problem with this approach. The less favored customers will be receiving research that has already been acted upon by the more favored customers.
One way to deal with this issue might be by simply disclosing to the less favored customers that they are receiving “stale” research. If market prices have changed since the research was issued, a reader can take the change into account. If the research is intended for the long term investors, a short delay in its receipt may not matter.
Perhaps that suggestion is not workable.4 Nonetheless, my point is just that no good system of regulating the contents of research can be developed unless the regulators have an understanding of why research is profitable to produce. The regulators arguably missed the problems with research that led to the Settlement because they did not understand the economics of its production and the link between research and investment banking. The solution to that problem is not to destroy the economics of research, but to understand and control them. Then the regulators may attempt to create a framework that is more likely to result in more good research.
Against that background, there follows a substantive discussion of the FINRA Research Proposal.
2. Distribution of Research Reports
(a) Selective Distribution of Research Reports
The FINRA Research Proposal states that FINRA intends to codify an existing NASD interpretation regarding selective, or tiered, distribution of research reports, and to provide additional guidance on this subject. However, as discussed in this Section, the proposed FINRA Rule is significantly broader than the current NASD requirement, and has potentially significant consequences as to the viability of producing research.
NASD-IM-2110-4 (“Trading Ahead of Research Reports”) currently “recommends, but does not require” firms to develop policies and procedures to isolate specific information within research and other relevant departments of the firm “so as to prevent the trading department from utilizing the advance knowledge of the issuance of a research report [emphasis added].”5 Under the proposed FINRA Rule, firms must establish policies and procedures to ensure that a research report is not distributed either to internal trading personnel or a particular customer or class of customers in advance of other customers that are entitled to receive the research report.6 Thus, while firms are currently restricted from sending research to their own trading departments ahead of distribution to customers, the proposed Rule, in addition, prohibits a firm from distributing research to certain customers ahead of others.
Under guidance to the proposed Rule, FINRA provides that firms may provide different research products and services to different categories of customers. Thus, a firm may provide one research product to investors with a long-term investment horizon (“investor research”) and a separate product to investors with a short-term investment strategy (“trading research”). A firm may not, however, send the same product to customers within the same class at different times, so as to favor certain customers within the same class over others.7 Thus, a firm would not be permitted to send a trading recommendation to large institutional trading customers ahead of sending that recommendation to retail trading customers. This potentially undermines the economic rationale for firms to produce research, which is to generate brokerage business from customers since firms will not be able to distinguish between big and small customers.
If it is not rational for investment banks to send research to all customers at the same time, they must do one of three things. First, they may produce and publish research broadly without regard to the fact that there is no benefit to doing so. Second, they may limit the expense of producing research to a level that appeases small investors who are willing to settle for a research product that does not give them any knowledge advantage over any other investors in the markets. Third, they may develop other ways to provide institutional investors with research that attracts business, for example by providing oral research only, or by allowing institutions to buy research away. I do not know what investment banks will do, perhaps it will be some combination of the three, but the regulators should have some theory as to the effect of the FINRA Research Proposal on the quality, quantity and distribution of research. It should not be sufficient for regulators merely to adopt rules, they should anticipate the consequences.
(b) Liability for Research Reports
The FINRA Research Proposal contains two important amendments relating to a firm’s liability for research reports. First, FINRA proposes to require firms to ensure that purported facts in a firm’s own research reports are “based on reliable information.”8 Second, FINRA proposes to require a firm to ensure that any third-party research distributed by the firm is “reliable and objective.”9 Neither of these provisions is contained in the current research rules. Both proposals may increase a firm’s potential liability for research. For example, if certain facts or recommendations contained in a research report turned out to be wrong, could a firm be held liable for distributing unreliable research? Further, the proposal does not specify what steps a firm must take in order to satisfy itself that research is reliable. In particular, in relation to third-party research, the proposal does not state whether firms must carry out additional due diligence beyond the limited supervisory approval currently required for third-party research, which is also included in the proposed Rule.10
Regardless of the merits of having some form of reliability requirement for research, the proposed rules are perverse in that the provision relating to third-party research seem more onerous than the provision governing a firm’s own research. For example, while the provision relating to third-party research applies to the report in its entirety (including both facts and recommendations), the provision relating to a firm’s own research only relates to the “purported facts” contained in the report.
3. Conflicts of Interest – Required Policies and Procedures
The FINRA Research Proposal contains a new, overarching requirement that firms adopt policies and procedures to identify and effectively manage conflicts of interest relating to the preparation, content and distribution of research reports and public appearances by research analysts.11 These policies must be reasonably designed to promote objective and reliable research that reflects the “truly held opinions of research analysts” and prevent the use of research or research analysts to manipulate or condition the market or favor the interests of the firm or certain clients.12
This proposal is similar to a requirement of the Settlement, under which a firm must adopt and implement policies and procedures that are reasonably designed to ensure that its associated persons, including investment banking personnel, cannot influence the contents of a research report or the activities of analysts.13 The FINRA proposal is also intended to obtain compliance with Section 501 of the Sarbanes Oxley Act (“Sarbanes Oxley”), which directed the SEC to address research analysts’ conflicts of interest by shielding research analysts from inappropriate pressures. FINRA notes that while the current NASD and NYSE research rules aim to implement the requirements of Sarbanes Oxley by imposing prescriptive requirements on firms, the FINRA Research Proposal provides firms with greater flexibility by permitting them to adopt policies and procedures tailored to the firm’s size and organizational structure.14
Under the FINRA Research Proposal, firms must adopt policies and procedures addressing conflicts affecting the production of research that at a minimum:
- prohibit prepublication review, clearance or approval of research reports by investment banking personnel, and restrict or prohibit prepublication review by other personnel not directly responsible for the preparation of research, other than legal and compliance personnel. This requirement reflects a prohibition in the Settlement on communications between research and investment banking personnel, but is more restrictive than the current research rules, which permit limited prepublication review of research by investment banking and other non-research personnel for specified purposes.15
- prohibit investment banking personnel from any involvement in the supervision or compensatory evaluation of research analysts.16
- prohibit compensating research analysts based on specific investment banking transactions or contribution to the firm’s investment banking services activities (including public and private offerings),17 and prohibit participation of investment banking personnel in the annual compensation review for research analysts who are primarily responsible for the preparation of a research report.18
- establish information barriers and other safeguards to insulate research analysts from pressure by investment banking personnel “or other persons who might be biased in their judgment or supervision.”19 This is a new requirement, not contained in the current research rules, which mirrors a similar requirement in the Settlement, at least as it relates to the separation between research and investment banking personnel.20
- prevent retaliation against research analysts by investment banking or non-research personnel as a result of the content of a research report that may adversely affect a current or prospective client relationship. The current ban on retaliation only applies to investment banking personnel.
In line with the 2007 Research Proposal, FINRA intends to extend the ban to include all employees, not just investment banking personnel.21
- establish quiet periods of at least 10 days after an initial public offering (an “IPO”) during which a firm must not publish or otherwise distribute research reports, and research analysts must not make public appearances, relating to the issuer if the firm has participated as an underwriter or dealer in the offering.22 The 10 day quiet period is significantly shorter than the 40 day quiet period currently applicable to firms acting as a manager or co-manger of an IPO, and the 25 day quiet period applicable to firms otherwise participating as an underwriter or a dealer in an IPO.23 Further, in line with the 2007 Research Proposal, the FINRA Research Proposal eliminates the current (i) 10-day quiet period after secondary offerings; and (ii) 15-day quiet period before and after the expiration, waiver or termination of a lock-up agreement.24
- prohibit promises of particular research, recommendations or ratings as an inducement for receipt of business or compensation.25
- restrict or limit activities by research analysts that can reasonably be expected to compromise their objectivity, including (as under existing rules) prohibiting participation in the solicitation of investment banking business, road shows and other marketing on behalf of an issuer.26
- prohibit prepublication review by a subject company for purposes other than verification of facts.27
4. Disclosure Requirements
(a) Firm-produced Research
FINRA proposes to maintain most of the disclosure requirements of the current research rules applicable to research produced by a firm, subject to the changes described in this Section.28
First, the proposal expands the requirement to disclose when a firm or its affiliates owns securities of a subject company. Currently, broker-dealers must disclose if the firm or its affiliates own 1 percent or more of the equity securities of the subject company.29 The proposed FINRA Rule requires disclosure if a firm or its affiliates maintain a “significant financial interest” in the equity or debt of the subject company including, at a minimum, if the firm or its affiliates beneficially own 1 percent or more of any class of common equity securities of the subject company.30 Ownership would continue to be determined using the beneficial ownership rules under Section 13(d) of the Exchange Act.31 The inclusion of debt securities in the required disclosure will likely require at least moderate operational fixes.
Second, with respect to disclosure of potential conflicts, the proposal modifies the existing research rules to require a firm to disclose in any research report all conflicts that “reasonably could be expected to influence the objectivity of the research report.”32 The current requirement applies to “actual, material conflicts of interest.”33 The FINRA Research Proposal includes among disclosable conflicts most of those that must be disclosed under the current research rules, including those related to receipt of investment banking and non-investment banking compensation and market making.34 In our view, the more ambiguous and open-ended disclosure requirement, which does not include any materiality standard, is of little benefit beyond being an invitation to write a long CYA disclaimer that is unlikely to be read or have meaning.35
Third, FINRA proposes to retain the current requirement that disclosures be contained in the research report itself, and not refer the reader to disclosures contained on the firm’s website. While the 2007 Research Proposal contained a proposal to permit website disclosure, the FINRA Research Proposal indicates that the SEC insisted on disclosures being contained in the research report itself. 36 This reflects a requirement under the Settlement that disclosures be contained in the research report itself.37 The one exception to this requirement is electronic research reports, where a report may contain a hyperlink to the required disclosures.38
(b) Third-Party Research
While the current research rules set forth certain specified disclosures that are required on third-party research, the FINRA Research Proposal replaces these requirements with a more general requirement to disclose “any material conflict of interest that can reasonably be expected to have influenced the choice of a third party research provider or the subject company of a third-party research report.”39
5. Other Proposals
The FINRA Research Proposal contains a number of proposals that either liberalize the current research rules, or adopt current requirements largely unamended, as described in this Section.
(a) Personal Trading Restrictions
FINRA proposes to eliminate the detailed restrictions and supervisory requirements that currently apply to research analysts’ personal securities transactions under the NASD and NYSE research rules.40 In its place, FINRA proposes to require firms to establish policies and procedures that restrict or limit research analysts’ personal account trading in securities that the research analyst covers, any derivatives of such securities and funds whose performance is materially dependent upon the performance of such securities.41 These policies must ensure that research analysts and others with the ability to influence the content of research reports do not benefit in their personal trading from knowledge of the content or timing of a research report before the intended recipients of the research have had a reasonable opportunity to act on the information in the research. A firm’s policies must also define financial hardship circumstances, if any, in which a firm would permit a research analyst to trade against his or her recommendation.42
(b) Research Analyst Registration
In line with the 2007 Research Proposal, FINRA proposes to limit the research analyst registration requirement to individuals whose primary job is producing research.43 Individuals involved in the production of research, but who have another primary job (e.g., trading) will not be required to register as research analysts. This amendment limits research analyst registration to those individuals who fall within the definition of “Research” personnel under the Settlement.44 However, the other research rules, including those regarding the content of research and conflicts of interest, would apply to all individuals involved in the preparation of research reports, irrespective of whether the individual is required to register as a research analyst.45
(c) Attestation Requirement
Currently, the research rules require a senior officer to attest annually that the firm has in place supervisory policies and procedures reasonably designed to comply with the research rules.46 FINRA proposes to eliminate this requirement. FINRA notes that firms would remain subject to the general obligation to have a supervisory system reasonably designed to achieve compliance with all applicable securities laws and regulations.47
(d) Termination of Coverage
The FINRA Research Proposal adopts the requirement of the current research rules for broker-dealers to provide notice of their intention to terminate research coverage and to make available a final research report on the subject company (1) using dissemination means equivalent to those it ordinarily used for research reports on the company, (2) that is comparable in scope and detail to prior research reports on the company, and (3) that includes a final recommendation or rating. If it is impracticable to produce a final recommendation or rating (e.g., if the research analyst covering the subject company or sector is no longer employed by the broker dealer), the final research report must disclose the broker dealer’s reason for the decision to terminate coverage.48
(e) Public Appearances
The FINRA Research Proposal groups together in a separate provision all of the requirements applicable to public appearances by research analysts.49 The proposed requirements are largely the same as under the current NASD and NYSE research rules, subject to an expanded requirement to disclose if the firm employing the research analyst or its affiliates maintain a significant financial interest in the debt or equity of the subject company, in line with the equivalent proposal for disclosure of financial interests in research reports discussed in Section 4 above.