April 30, 2002
Securities and Exchange Commission
450 Fifth Street, N.W.
Washington, DC 20549
Re: Proposed Rules Relating to Research Analyst Conflicts of Interest SEC
File Nos. SR-NASD-2002-21 and SR NYSE-2002-09
Dear Ladies and Gentlemen:
These are my individual comments on the rules proposed by the NASD and the NYSE (the "SROs") concerning the publication of views by securities analysts (the "SRO proposals"). The proposals were published in the Federal Register on March 14, 2002.
My comments do not necessarily represent the views of my firm or its clients.
In summary, I believe the Commission should require the SROs to modify their proposals. As they now stand, the proposals would:
I. Adverse Competitive Effects
The Commission should evaluate the SRO proposals against the background of certain undisputed facts about competition in the securities industry.
First, financial services firms fiercely compete not only for customers but also for registered representatives ("brokers"). An important factor in the competition for brokers is the extent and quality of research support a firm is able to offer to its brokers.
Second, one of the primary means by which analysts compete is by reference to the timeliness of their reports. In no business like the securities business is yesterday's news worth less. In fact, many of the newer formats for research output have been designed in the expectation that research will be distributed electronically, by email or website, for the purpose of giving customers timely and short-term trading recommendations. Anything that handicaps a competitor from issuing timely advice is a help to other competitors and a distortion of competitive conditions.
Third, analysts compete in a fishbowl. Contrary to the implications of some of the SRO proposals, the success or failure of analysts at picking stocks or estimating earnings is immediately apparent with each trading day or each earnings reporting period. Analysts are ranked regularly by a number of investor polls and services. The results of these rankings are readily available to all investors and play an important role in an analyst's reputation and compensation.
Fourth, securities firms compete not only among themselves but also with other providers of financial services such as registered and unregistered investment advisers, insurance companies and banks. They also compete with providers of financial information such as the electronic and printed media. Some of these competitors are subject to less regulation than securities firms or no regulation at all.
Finally, securities firms compete on a global basis with foreign providers of financial services.
Unfortunately, the SROs have ignored the competitive consequences of their proposals. Some of these consequences are as follows:
1. Competitive Advantages for Non-Member Firms. The SRO proposals apply only to members and member firms of the NYSE and NASD. This leaves unaffected other providers of financial services such as registered and unregistered investment advisers, banks and insurance companies as well as other competitors such as newsletter writers and the electronic and printed financial press. Such firms could publish during the proposed "blackout periods" and make public appearances without providing the disclosures proposed to be required of member firms.
2. Competitive Advantages for Large Firms. The SRO proposals apply equally to large and small firms. But while large firms may be able to afford the legal and compliance reviews and written policies contemplated by the SRO proposals, this is emphatically not the case with small firms.
3. Competitive Advantages for Institutional Investors. During the various "blackout periods" proposed by the SROs, institutional customers will be able to receive oral updates from analysts. Retail investors will have to wait.
4. Competitive Advantages for Foreign Firms. Non-U.S. firms that are not member organizations of the SROs will be able to publish their research on the Internet and otherwise for the benefit of U.S. customers, particularly U.S. institutional customers, at a time when U.S. analysts are subject to blackout periods.
These competitive consequences of the SRO proposals would, on the merits, constitute sufficient justification for the Commission to disapprove the proposals. Even if the Commission does not come to this conclusion, however, I do not believe it can justify approving the proposals in view of the SROs' failure even to consider their competitive consequences.
Item 4 of Form 19b-4 requires an SRO that submits a rule change to the Commission to "specify the particular categories of persons and kinds of businesses on which any burden [on competition] will be imposed [by the proposed rule change] and the ways in which the proposed rule change will affect them." It also requires the SRO, if it believes that any impact on competition will not be significant, to explain why this will be the case. Also, the SROs' statement regarding competition "should be sufficiently detailed and specific to support a Commission finding that the proposed rule change does not impose any unnecessary or inappropriate burden on competition" (emphasis added).
Rather than provide the foregoing specific information on the competitive impact of their proposals, the SROs in their submissions to the Commission make the conclusory statement that the proposals "will not result in any burden on competition that is not necessary or appropriate" in view of the purposes of the 1934 Act. This statement is not responsive to the Commission's own requirements and cannot serve as a basis for the Commission's own findings regarding the competitive impact of the proposals.
II. Opportunities for Short Sellers
Several of the SRO proposals would play into the hands of short sellers by barring any research reports by analysts for stated periods of time.
For example, the proposals would bar an analyst from publishing for a period of 40 days following his or her firm's participation as a manager or co-manager in an issuer's IPO and for 10 days following such participation in a follow-on offering.
The SRO proposal would apply to the managers and co-managers of the IPO and would not be linked to the delivery of a final prospectus. It would therefore constitute a total bar on the dissemination of information by the firms most likely to know best the strengths and weaknesses of the IPO issuer. It is difficult to see how investors would be served by such a bar, particularly since other firms could begin to publish immediately after the IPO (if they delivered a prospectus or provided a hyperlink to a prospectus from electronic research). If research published by other firms contained negative assertions about the IPO stock, the firms best able to respond would be unable to do so. Investors and issues would be at the mercy of critics of the IPO for up to 40 days. The only persons who could profit from this scenario would be short sellers.
The same distortions arise in the case of the 10-day blackout period following add-on offerings. There is currently no such blackout period for reporting issuers because the issuer's 1934 Act reports provide a reservoir of public information against which the public can judge the accuracy of research reports. A blackout period would be an arbitrary bar on publication by analysts employed by the firm or firms who presumptively know best the issuer's strengths and weaknesses. Again, the blackout period would not serve anyone's interest except short sellers.
The SRO proposals would ban transactions by analysts for five days preceding and 30 days following a research report or certain rating or estimate changes. This proposal will not raise issues for firms that prohibit their analysts from buying or selling stocks that they cover. Many firms still permit such transactions, however, and for those firms the SRO proposals really amount to a 30-day blackout following an analyst's transaction. Again, an analyst that might be the leading analyst on a given stock would be unable for 30 days to respond to negative views expressed by other firms or persons.
The only persons who could profit from the blackout periods proposed by the SROs are short sellers, who can be expected to time their criticisms of an issuer's accounting or business model for those times when the leading analyst is unable to comment.
The traditional way of dealing with conflicts of interest in the securities industry is to disclose them so that investors can decide for themselves what weight to give to a recommendation. A flat bar on publication not only deprives the investor of this opportunity but constitutes a raw form of prior restraint inconsistent with First Amendment principles.1
III. Integrated Disclosure System
Much of the SEC's development over the past 30 years of the integrated disclosure system under the 1933 Act and 1934 Act has been based on the "efficient market theory," under which an important role is assigned to the securities analyst.
The Commission has often stated that its integrated disclosure policy is premised on the analyst's role as a disseminator of information. Various eligibility tests for the Commission's abbreviated disclosure forms rely on criteria such as "float," which is recognized as an indicator of the likelihood that a company's securities are covered by one or more analysts.
As the Commission put it in a 1998 description of the relationship among issuers, analysts and investors:
Investors acquire useful information regarding companies from sources other than Commission mandated disclosure. One such source is analysts' research reports. As the Commission has long acknowledge and the Supreme Court [has] recognized ..., analysts fulfill an important function by keeping investors informed. They digest information from Exchange Act reports and other sources, actively pursue new company information, put all of it into context, and act as conduits in the flow of information by publishing reports explaining the effects of this information to investors.
With the efficient market theory in mind, the Commission has steadily over the past 20 years reduced the frequency and duration of the occasions on which its rules prevent the analyst from performing his or her function. The "research rules" (Rules 137 through 139 under the 1933 Act) have been liberalized; the "blackout periods" following a public offering have been reduced from 90 days in the case of an IPO to 25 days and to zero days in the case of public offerings by reporting companies.
Given the importance of the analyst's contribution to the efficiency of the nation's securities markets, regulators should hesitate to impose additional regulation, especially where such regulation could hamper or delay investor access to what the analyst has to say.
It would be especially remarkable if the Commission would now approve SRO rule proposals that would move in the opposite direction of Commission policy for more than 20 years, i.e., by imposing 40-day and 10-day blackout periods where Commission rules now impose none. And especially where this reversal was based on nothing more than anecdotal evidence about the need to reduce the incidence of analyst conflicts or "restore investor confidence."
Finally, no reduction of market efficiency occurs without benefit to someone. In the case where analysts are subject to unpredictable blackout periods, short sellers will be tempted to put out adverse commentary on a security at a time when they know the leading sell-side "bulls" on the stock to be barred from publishing.
IV. The SRO Proposals Raise Substantial First Amendment Questions2
The First Amendment to the U.S. Constitution generally places strict limits on government attempts to regulate speech. "Where a government restricts the speech of a private person, the state action may be sustained only if the government can show that the regulation is a precisely drawn means of serving a compelling state interest." Consolidated Edison Co. v. Public Service Comm'n of New York, 447 U.S. 530, 540 (1980).
While for several decades during the mid-1900s commercial speech was excluded from First Amendment protection, Valentine v. Chrestensen, 316 U.S. 52, 54 (1942), the Supreme Court eventually recognized the value of, and protection for, even purely commercial speech in Virginia State Board of Pharmacy v. Virginia Citizens Consumer Council, Inc., 425 U.S. 748 (1976). The reasons identified by the Court in Virginia State Board for protecting commercial speech sound virtually identical to its reasons for protecting political and other noncommercial speech:
There is, of course, an alternative to this highly paternalistic approach [of allowing a government to keep its citizens in ignorance]. That alternative is to assume that this information is not in itself harmful, that people will perceive their own best interests if only they are well enough informed, and that the best means to that end is to open the channels of communication rather than to close them. . . . It is precisely this kind of choice, between the dangers of suppressing information, and the dangers of its misuse if it is freely available, that the First Amendment makes for us.
Id. at 770. The Court in Virginia State Board further explained that a "particular consumer's interest in the free flow of commercial information . . . may be as keen, if not keener by far, than his interest in the day's most urgent political debate." Id. at 763. The Court has likewise recognized that commercial speakers themselves have a cognizable First Amendment interest, writing that so long as the sale of a product is lawful, the industry producing and selling that product "has a protected interest in communicating information about its products and adult customers have an interest in receiving that information. Lorillard Tobacco Co. v. Reilly, -- U.S. --, 121 S. Ct. 2404, 2430 (2001).
Since 1980, the analytical framework for reviewing restrictions on commercial speech has been provided by a test expressed in Central Hudson Gas & Electric Corp. v. Public Service Commission of New York, 447 U.S. 557 (1980). In its original formulation, that test contained the following four elements: (1) if the communication is neither misleading nor related to unlawful activity, (2) the government must assert a substantial interest, (3) the restriction must directly advance the governmental interest involved and (4) the restriction must be no more extensive than necessary. Central Hudson, 447 U.S. at 564.
Despite a brief dilution of that test in the Court's 1986 decision in Posadas de Puerto Rico Associates v. Tourism Co. of Puerto Rico, 478 U.S. 328 (1986), the Court has since retreated from Posadas and applied the Central Hudson test with substantial vigor. For example, in Edenfield v. Fane, 507 U.S. 761, 771 (1993), the Court required the government under the third element of the Central Hudson test to "demonstrate that the harms it recites are real and that its restriction will in fact alleviate them to a material degree" (emphasis added). Other decisions have likewise rigorously applied Central Hudson's third element. See, e.g., Rubin v. Coors Brewing Co., 514 U.S. 476, 486-88 (1995); Florida Bar v. Went for It, Inc., 515 U.S. 618, 626 (1995); Ibanez v. Florida Dep't of Bus. & Prof'l Regulation, 512 U.S. 136, 142-48 (1994).
The Supreme Court has similarly strengthened the fourth element of the Central Hudson test, holding that a restriction fails if there are "obvious less-burdensome alternatives to the restriction on commercial speech." City of Cincinnati v. Discovery Network, Inc., 507 U.S. 410, 417 n.13 (1993). That clarification of the fourth element of the Central Hudson test has become a significant independent ground under which restrictions on commercial speech are invalidated. See, e.g., Rubin, 514 U.S. at 490-91.
Other recent Supreme Court opinions have continued to emphasize the renewed stringency of the Central Hudson test and have confirmed the demise of the briefly lenient approach used in Posadas. Thus, in 44 Liquormart, Inc. v. Rhode Island, 517 U.S. 484 (1996), Justice Stevens, joined by Justices Kennedy, Thomas, and Ginsburg, flatly stated that "Posadas erroneously performed the First Amendment analysis" and concluded that government "does not have the broad discretion to suppress truthful, nonmisleading information for paternalistic purposes that the Posadas majority was willing to tolerate." 517 U.S. at 509-10. And writing separately, commercial-speech skeptic Justice O'Connor, joined by Chief Justice Rehnquist (who authored Posadas) and two other Justices, concluded that:
Since Posadas, however, this Court has examined more searchingly the State's professed goal, and the speech restriction put into place to further it, before accepting a State's claim that the speech restriction satisfies First Amendment scrutiny. . . . [W]e declined to accept at face value the proffered justification for the State's regulation . . . . The closer look that we have required since Posadas comports better with the purpose of the analysis set out in Central Hudson, by requiring the State to show that the speech restriction directly advances its interest and is narrowly tailored.
Id. at 531-32 (citations omitted).
Likewise, in Greater New Orleans Broadcasting Ass'n v. United States, 527 U.S. 173 (1999), the Court was virtually unanimous in describing 44 Liquormart's indication that the Posadas decision is dead. The Court described the several opinions in 44 Liquormart as "conclud[ing] that our precedent both preceding and following Posadas had applied the Central Hudson test more strictly." 527 U.S. at 182. And Justice Thomas, concurring in the judgment, would have gone farther and applied strict scrutiny rather than the Central Hudson test. Id. at 197.
The Supreme Court's most recent decisions have continued to apply Central Hudson vigorously and to strike down restrictions on commercial speech. See, e.g., Lorillard Tobacco Co., -- U.S. at --, 121 S. Ct. at 2421. And they have begun to recognize, as well, that even greater protection may be warranted for commercial speech. Thus, in Greater New Orleans, the majority acknowledged Justice Thomas's concurring views and raised the possibility that Central Hudson should be replaced with a "more stringent" test, recognizing that various legal thinkers, including "reasonable" judges, have adopted that position. Id. at 1930. And in Lorillard Tobacco, the Court responded to an argument in favor of strict scrutiny for commercial speech by recognizing that "several Members of the Court have expressed doubts about the Central Hudson analysis and whether it should apply in particular cases. -- U.S. at --, 121 S. Ct at 2421. In both those cases, however, the restrictions at issue failed even the Central Hudson test, and thus the Court declined to decide whether a stricter First Amendment standard was required. However, in United States v. United Foods, Inc, -- U.S. --, 121 S. Ct. 2334 (2001), the Court seemed to go a step further. The Court in United Foods again recognized the criticisms of Central Hudson but once again was able to strike down the compulsion even assuming some lesser protection of commercial speech. Id. at --, 121 S. Ct. at 2337-38. However, the union dues and state bar association cases it ultimately relied upon did not involve a differential standard for commercial speech and the Court never applied the Central Hudson test.
Are the SROs "State Actors" for First Amendment Purposes?
The SROs invariably take the position that they are private organizations and not "state actors" for federal constitutional purposes. See, e.g., Desiderio v. NASD and SEC, 191 F.3d 198 (2d Cir. 1999). It is hard to ignore, however, the manner in which the SRO proposals arose. The accompanying press releases made it clear that they had been "initiated" by the SEC. Industry sources state that SEC staff members were intimately involved in the SROs' deliberations and off-the-record consultations with industry groups. The SEC chairman and representatives of Congressional committees were present at the press conferences at which the nearly-identical SRO proposals were unveiled. Representatives of the SEC made public comments shortly thereafter about allowing only a short comment period and anticipating a rapid phase-in of the SRO proposals.
Whatever the status of the NYSE or NASD when acting as regulators of their markets or as administrators of arbitration and disciplinary schemes, the SRO proposals would operate "not as a private compact among brokers and dealers but as federal law." Blount v. SEC, 61 F.3d 938 (D.C. Cir. 1995). As in Blount, which found the Municipal Securities Rulemaking Board to be a state actor for purposes of a First Amendment challenge to its Rule G-37, a violation of the rule exposes the violator to monetary sanctions and possible suspension or expulsion from the securities industry. The SRO proposals should therefore be tested against the standards of the First Amendment.
Status of Research Under the First Amendment
Securities analysts publish information, recommendations and opinions on a wide range of companies, focusing on political and economic conditions and trends, technological innovation, competitive conditions, the quality of management and the quality of financial reporting. It is certainly more difficult to characterize their activity as "commercial speech" than the activity at issue in many of the commercial speech cases cited above.
It is not necessary to characterize analyst activity as pure speech, however, if the SRO proposals do not meet the Supreme Court's test for the regulation of commercial speech as set forth in Central Hudson.
The key questions for the SRO proposals under Central Hudson are likely to be:
- whether the proposals directly advance the SRO or governmental interest being asserted, keeping in mind the Supreme Court's admonition in Edenfield, supra, that the government is required to "demonstrate that the harms it recites are real and that its restriction will in fact alleviate them to a material degree," and
- whether the restrictions proposed are no more restrictive than necessary, keeping in mind the Court's admonition in City of Cincinnati, supra, that a restriction fails if there are "obvious less burdensome alternatives to the restriction on commercial speech."
I submit that neither the SROs nor the Commission has offered any evidence that the SRO proposals can be justified under the foregoing tests. In particular, there has been no showing that disclosure of potential conflicts is not sufficient to maintain public confidence in analyst reports. The proposals therefore fail the third and fourth prongs of the Central Hudson test.
The SRO proposals should be the subject of a public hearing convened by the Commission. I would be pleased to offer further comments at such a hearing. At a minimum, they should be republished in amended form for further public comment.
Very truly yours,
cc: Harvey L. Pitt, Chairman
Isaac C. Hunt, Jr., Commissioner
Cynthia A. Glassman, Commissioner
Annette L. Nazareth, Director, Division of Market Regulation
Alan L. Beller, Director, Division of Corporation Finance
Edward A. Kwalwasser, Group Executive Vice President, NYSE
Thomas Selman, Senior Vice President, NASD
|1||The proposals' exception for "significant" news or events with prior legal or compliance approval is not clearly applicable to a need to respond to third-party comments on a stock, and the need to involve legal or compliance personnel practically guarantees delay and a narrow application of the exception.|
|2||The description of case law in this section of the letter is based on a white paper, "The First Amendment and the Securities Laws," prepared earlier this year under the auspices of The Federalist Society. The writer of this letter was a member of the working group that prepared the paper, which is available at http://www.fed-soc.org/Publications/White%20Papers/SEC-1stAmendment.PDF.|