There is strong evidence that the stock market efficiently assimilates information in establishing stock prices. This is a benefit to ordinary investors, who buy or sell “at the market.” When the market is well-supplied with information, stock prices are at their most accurate, and these investors are assured of getting the best prices a liquid market could produce. However, both Regulation Fair Disclosure (FD) in 2000 and the Global Settlement with the largest investment banking firms in 2003 reduced the amount of information flowing to the market, thus adversely affecting the quality of stock prices. In these efforts to ensure that investors have equivalent information when they trade, and that they are not misled by supposedly biased recommendations of analysts, the Securities and Exchange Commission (SEC) has impaired the market’s role in price discovery. The victims of this intervention are the very investors the SEC is supposed to protect.
Our principal securities laws, the Securities Act of 1933 and the Securities Exchange Act of 1934, are based on the belief that investors can make their own decisions about buying or selling shares of public companies. Accordingly, the SEC is charged with ensuring that companies make accurate disclosures of their financial condition and business prospects to help investors make their investment decisions.
Nevertheless, much academic work suggests that no one--not even market professionals--can beat the market over time, which calls into question the value of the fundamental analysis of companies that extensive disclosure is supposed to support. In A Random Walk Down Wall Street, Princeton economics professor Burton Malkiel writes:
[O]ne has to be impressed with the substantial volume of evidence suggesting that stock prices display a remarkable degree of efficiency. Information contained in past prices or any publicly available fundamental information is rapidly assimilated into market prices. Prices adjust so well to reflect important information that a randomly selected and passively managed portfolio of stocks performs as well or better than portfolios selected by experts.
In other words, if an investor--without doing any analysis at all--were to buy a portfolio of randomly selected stocks, his portfolio would perform as well as one selected by an investment professional.
How can this be? The reason is that the market at any given moment seems to incorporate all the relevant information that is publicly available about a company, so everything that is known and relevant is incorporated into the price of every stock. From the point it has reached at any given moment, a stock may move up or down depending on the information that subsequently becomes available about it. No one lacking important nonpublic information about a stock can know which direction the stock will take in the future, because that direction depends on information that has not yet reached the market. Accordingly, then, forecasting which way the price of an individual stock will move from where it is at any given moment is completely unpredictable--a random walk--hence the title of Malkiel’s book. The “random walk” idea, also known as the “efficient market theory,” has been known for at least thirty years. The first edition of Malkiel’s book was published in 1973. Numerous tests have shown that randomly selected portfolios--some of them selected by throwing darts at the stock pages of a newspaper--occasionally outperform even the broad market indexes.
The success of this theory is important because most people who assemble their own portfolios buy and sell shares of public companies “at the market”--that is, they ask their brokers to buy or sell shares at whatever the current price may be when the order is placed. These investors, then, benefit when the maximum amount of information about public companies is in the market at any given time. The more information available in the market, the more likely it is that the price of a stock reflects the most informed consensus estimate of the company’s value at the moment it is purchased or sold.
What does this mean for policy? For the SEC, it means that investors are best protected when the market has as much information as possible. It almost does not matter where the information comes from. Market participants will give it appropriate weight--whether it is the opinion of a sell-side analyst, a question about the effectiveness of a pharmaceutical, or a warning about earnings from a chief financial officer.
But the SEC continues to pursue policies that restrict the amount of information that gets to the securities markets, thus denying individual investors the best prices the market could produce at any given time. The agency--dominated by lawyers, not economists--does this because it looks at the market the way lawyers do: as a series of individual transactions rather than as a mechanism for assimilating information and producing the most accurate price at any given moment.
Ensuring that maximum information gets into the market each day would seem like an easy assignment for an agency whose principal role is to promote disclosure, but apparently not. In issuing Regulation FD in 2000, and negotiating the “Global Settlement” with the major investment banking firms in 2003, the SEC adopted policies that were not only inconsistent with its mandate to foster disclosure, but also inconsistent with the interests of the investors the agency is supposed to protect. An observer might well ask, paraphrasing manager Casey Stengel’s remark as he watched his Mets play baseball, “Does anyone here know how to play this game?”
Failing to See the Forest for the Trees
In December 2006, the Wall Street Journal reported:
The Securities and Exchange Commission is looking into whether laws are being broken somewhere in the transfer of information between Congress and Wall Street. It’s not illegal for lawmakers to disclose information not publicly known about the workings of Congress, even if it could affect stock prices. . . . But one question the SEC is trying to resolve is whether the passing of market-sensitive information by lobbyists to investors could violate insider-trading law.
What is going on here? The hedge funds appear to be doing basic research to find out whether legislation that might change the value of companies will or will not be enacted. If they were to use this information in purchasing or selling stock, it would affect the price of the stock, making it a more accurate reflection of its true value. Is there something wrong with this? As the SEC has been developing the concepts underlying insider-trading law and Rule 10b-5, there might well be. Over the years, the SEC has been pursuing the idea that it is inherently unfair when one party to a securities trade has more information than another. The prime example of this is Regulation FD, initially proposed under the chairmanship of Arthur Levitt in 2000, which attempted to ensure that companies do not provide material information to analysts unless that information is made public at the same time.
The SEC’s rationale for this regulation was succinctly described in the release that accompanied its adoption:
[W]e have become increasingly concerned about the selective disclosure of material information by issuers. As reflected in recent publicized reports, many issuers are disclosing important nonpublic information, such as advance warnings of earnings results, to securities analysts or selected institutional investors or both, before making full disclosure of the same information to the general public. Where this has happened, those who were privy to the information beforehand were able to make a profit or avoid a loss at the expense of those kept in the dark.
Thus, the SEC is concerned that some people in possession of certain information are trading with others who do not have it. The people who do not have it have been kept in the dark, victimized by those with information. Here, in a nutshell, is the problem: the SEC is concerned about the “justice” of individual transactions between buyers and sellers. This approach sees the market as nothing more than a series of individual transactions. The sum total of those transactions has no independent value. However, the burden of the argument in A Random Walk Down Wall Street, not to mention many academic papers, suggests that this idea is wrong.
The Supreme Court Weighs In
Indeed, when the Supreme Court looked at this issue, it came to a view quite different from that of the SEC. The most important case in this area is Dirks v. SEC, decided in 1983. Dirks, an officer of a broker-dealer, learned from a former employee of a company that the company was engaged in fraudulent practices. He investigated and shared his findings with others, who then sold their shares, although Dirks himself did not trade in the stock. When the fraudulent practices finally became public, the SEC charged Dirks with aiding and abetting violations of the securities laws by passing nonpublic information to those who subsequently sold their shares. The SEC’s charge was upheld at the district and circuit court levels but was reversed by the Supreme Court, which held that there is no obligation on the part of a tippee like Dirks to disclose or abstain from disclosing material nonpublic information unless he has a fiduciary duty to do so. In the case of a corporate insider, that fiduciary obligation runs to the company or its shareholders, but an ordinary tippee such as Dirks has no duty to anyone, and thus, said the court, may trade on the information or not, or disclose it to others or not, without violating securities laws. The only time that a tippee assumes a fiduciary obligation not to trade is when he knows or has reason to know that he received the information improperly, as a result of the tipster’s breach of his own fiduciary duty. The Court’s broader principle is that the person who has nonpublic information is under no duty to disclose it to others before trading on it.
The Supreme Court’s rationale for its Dirks decision is instructive:
Imposing a duty to disclose or abstain, solely because a person knowingly receives material nonpublic information from an insider and trades on it could have an inhibiting influence on the role of market analysts. . . . It is commonplace for analysts to "ferret out and analyze information" . . . and this often is done by meeting with and questioning corporate officers and others who are insiders. . . . It is the nature of this type of information, and indeed of the markets themselves, that such information cannot be made simultaneously available to all the corporation’s stockholders or the public generally.
This statement is a rejection, almost twenty years in advance, of the SEC’s basis for Regulation FD, which expressly forbade companies from providing material nonpublic information to analysts unless they release it simultaneously to the market in general. The Supreme Court clearly saw analysts--and the information they learned through contacts with corporate officers--as contributing something valuable to the securities markets. And the corporate officers who supplied the information were seen as fulfilling an informative role, not violating any fiduciary duty.
What, then, was the SEC’s theory in Dirks? Why did the agency charge Dirks, even though he never traded in the stock and seemed to have done a public service by investigating and exposing a fraud? As described by the Supreme Court: “In effect, the SEC’s theory of tippee liability . . . appears rooted in the idea that the antifraud provisions [of the securities laws, particularly rule 10b-5] require equal information among all traders.” This is exactly the same rationale that the SEC used in adopting Regulation FD almost twenty years later. In other words, despite the Supreme Court’s view that analysts assist the market by “ferret[ing] out and analyz[ing] information,” the SEC has continued to believe--and to act on the belief--that it should suppress this activity in order to equalize the information possessed by each party to a securities trade. Thus, after Regulation FD, it is not surprising that the SEC might be looking for a way to prevent hedge funds and others from trading on information they receive from Capitol Hill.
But there is also an important idea implicit in the Court’s statement that “it is the nature of this type of information” that it “cannot be made simultaneously available to all the corporation’s stockholders or the public generally.” The Court’s point was that the information would have no value to the analyst, and hence would not be collected, if it had to be made public before it could be used. In other words, the Court recognized that incentives are necessary to ensure that information gets into the market. In effect, the Supreme Court in 1983 was drawing a fine line between two competing values. On one side was the notion that corporate insiders should be punished for violating their fiduciary obligations to the company or its shareholders by disclosing inside information improperly. But on the other side was a desire to ensure--through appropriate incentives--that important information gets into the market and affects the market price of shares for the benefit of all investors.
The SEC Gets What Its Policy Implies . . .
The SEC’s approach is quite different. In its analysis, the Supreme Court did not focus on the transaction between the tippee and the investor he trades with. The Court’s objective in tying liability to fiduciary duty was apparently to punish the faithless insider, not to compensate the uninformed buyer or seller--although the penalty may have that effect. Similarly, the Court’s objective in confining liability narrowly to those with a fiduciary duty was clearly to leave as large a channel as possible for information to reach the market. But as we have seen from the SEC’s commentary about Regulation FD, the agency focuses on the inequality of information between the informed and the uninformed traders at the moment of the trade. This approach naturally devalues incentives to supply information to the market, and in the end prefers a market in which both sides of a trade are equally ignorant rather than a market in which one trader profits at the expense of another.
The same approach infused the Global Settlement. This omnibus agreement with the ten largest investment banking firms in the United States arose out of two factors: the great preponderance of “buy” recommendations by sell-side analysts (that is, those at brokerage or investment banking firms) in comparison to their “sell” or neutral recommendations, and an investigation by then-New York attorney general Eliot Spitzer which found instances in which analysts were praising companies that they described quite differently in private communications. The SEC’s hypothesis was that the optimistic forecasts and falsely favorable recommendations were efforts by analysts to assist the firm’s investment bankers in attracting or holding the business of public companies and to assist its securities salesmen in peddling stock. The Global Settlement was intended to break these links. If the SEC’s hypothesis was correct, doing so would reduce the value of analysts to investment banking firms and thus the number of analysts and their production of information. Although the Global Settlement was made with the ten largest firms, it resulted in a set of best practices that the SEC has required all brokerages to adopt.
Why was it important to stop analysts from supporting the sales efforts of investment bankers with overly optimistic recommendations? What the SEC had in mind was to protect unsophisticated investors from being misled by biased recommendations. But reducing the value of analysts would reduce the role of the analyst channel in providing information to the market. Moreover, and more important, the SEC could have punished the analysts who were found to be issuing false recommendations rather than assuming that all sell-side analysts were biased because of their relationships to securities firms. Even so, anyone who has read analysts’ reports--both before and after the Global Settlement--knows that they contain a large amount of useful information, even if the recommendation is often to buy. There were certainly analysts who deserved punishment in the period prior to the Global Settlement, but it was only the SEC’s willingness to restructure an entire industry on the basis of a theory that accounts for the settlement. Here again, the SEC chose to sacrifice a fully informed market in order to shield a few individual investors who might be misled by an excessively optimistic analyst’s report.
. . . But Investors Don’t Get What They Need
The results of Regulation FD and the Global Settlement have been what the SEC seems to have wanted, but they have not been good for the securities markets or for investors.
Regulation FD has contributed to a decline in the number of analysts by making it more difficult and less rewarding for them to do their jobs. Companies now make the material information they provide about themselves available to the market and to analysts simultaneously--if they make it available at all. Complying with Regulation FD, companies are afraid to meet separately with analysts or to answer their questions in detail, lest they inadvertently provide material information that is not available generally to the public. As a result, analysts do not have the opportunity to test their theories or “ferret out and analyze information,” as the Supreme Court put it in Dirks. Because they cannot add value in this way, they receive lower compensation, there are fewer of them, and their contribution to the market’s total information is reduced. This result has been particularly visible in the coverage of small firms, where the remaining usefulness of analysts has the least value to the securities firms that employ them.
A 2004 academic study supports this view, finding that “the adoption of Reg FD caused a significant reallocation of information-producing resources, resulting in a welfare loss for small firms, which now face higher costs of capital. The loss of the ‘selective disclosure’ channel for information flows could not be compensated for via other information transmission channels. This effect was more pronounced for firms communicating complex information and . . . for those losing analyst coverage.” The loss of analyst coverage has been so pronounced that in June 2005 Reuters and NASDAQ launched a venture intended to increase analyst coverage, especially of small companies. The press release for the venture noted that “approximately 50 percent of all publicly held companies have two or fewer analysts and approximately 35 percent of all public companies have no analyst coverage. . . . Since January 2002, 691 companies have lost analyst coverage representing over 17 percent of the entire universe of companies with analyst coverage.”
Another academic study in 2005 records a decline in the quality of earnings forecasts following the adoption of Regulation FD:
First, earnings forecasts became less accurate post-FD at the levels of both the individual analyst and the consensus; this effect is significantly larger for early forecasts than for late forecasts, and for smaller companies than for larger companies. Second, the dispersion in earnings forecasts across individual analysts following a company increases post-FD . . . and increases with the passage of time following FD’s adoption. . . . Our findings suggest that there has been a reduction in both selective guidance and the quality of analyst forecasts post-FD.
In other words, important information in establishing a stock’s price is not getting into the market as effectively as it did before Regulation FD, inevitably decreasing the quality of stock prices.
The Global Settlement has also contributed to this problem. Whether the objectives of the Global Settlement have been met is subject to some question. Academic work has in some cases shown that independent analysts have performed better in their forecasts than analysts connected with securities firms. Other studies, however, have found no substantial reduction in the optimistic quality of analyst recommendations at the ten sanctioned firms after the Global Settlement, but have discovered substantial declines in the number of analysts at the firms that agreed to the settlement--a 14 percent decline relative to the number employed in 2000 and a 20 percent decline relative to the number in 2001. This stands to reason, since it had become less advantageous and profitable for the firms to employ analysts. With the decline in the number of analysts came a decline in the number of companies covered by analysts. Anecdotal information indicates that there have also been substantial declines in the number of analysts at small securities firms, further contributing to the decline in analyst coverage for small public companies. The absence of analyst coverage clearly reduces the amount of useful information about companies that the market previously used to establish prices. These prices are still being established, of course, but they are likely to be less accurate reflections of the real value of companies. Thus, ordinary investors who buy or sell these stocks are less likely to get the best price the market can offer when they purchase “at the market.”
What the SEC Should Do
There is a great deal of evidence that suggests that securities markets are highly efficient in processing information--so much so that the SEC should make use of this resource in the course of discharging its responsibility to protect investors. Regulation FD and the Global Settlement were both based on attempts to protect investors by denying them information that the SEC regarded as harmful. Regulation FD viewed the acquisition and dissemination of information by analysts as harmful because it created inequalities of information between buyers and sellers. In the case of the Global Settlement, allegedly biased stock recommendations by analysts were considered harmful because of their potential to mislead investors.
But Regulation FD and the Global Settlement have had adverse consequences. The former may have reduced the incidence of informed traders transacting with uninformed traders, but at the cost of substantially reducing analyst coverage and the amount of useful information in the market. The Global Settlement has also reduced analyst coverage--especially of small companies--and the number of analysts at sell-side firms without appreciable changes in analyst recommendations. This is another informational loss for the securities markets.
Under these circumstances, the SEC should consider repealing Regulation FD and abandoning the Global Settlement. Both market interventions have done more harm than good. Taking these steps would enhance the amount of information getting to the securities markets, improve the quality of share prices, and thus provide ordinary investors with the protection inherent in a well-informed market.
Peter J. Wallison is a senior fellow at AEI. Research assistant Daniel Geary and editorial assistant Evan Sparks worked with Mr. Wallison to edit and produce this Financial Services Outlook.
1. Burton G. Malkiel, A Random Walk Down Wall Street (New York: W. W. Norton, 1999), 268.
2. Brody Mullins and Kara Scannell, “Hedge Funds Hire Lobbyists to Gather Tips in Washington,” Wall Street Journal, December 8, 2006.
3. 17 U.S.C. § 240.10b-5 (1951).
4. U.S. Securities and Exchange Commission (SEC), Final Rule: Selective Disclosure and Insider Trading, release nos. 33-7881, 34-43154, IC-24599, file no. S7-31-99, August 15, 2000, 2 (emphasis added), available at www.sec.gov/rules/final/33-7881.htm (accessed February 1, 2007).
5. Dirks v. SEC, 463 U.S. 646 (1983).
6. Ibid. at 658–59.
7. Ibid. at 657.
8. SEC, Office of the New York State Attorney General, North American Securities Administrators Association, NASD, and New York Stock Exchange, “Ten of Nation’s Top Investment Firms Settle Enforcement Actions Involving Conflicts of Interest Between Research and Investment Banking,” news release, April 28, 2003, available at www.sec.gov/news/
press/2003-54.htm (accessed February 1, 2007).
9. Armando Gomes, Gary Gorton, and Leonardo Madureira, “SEC Regulation Fair Disclosure, Information, and the Cost of Capital” (working paper, Olin School of Business, Washington University, Saint Louis, 2004), 1, available at www.olin.wustl.edu/workingpapers/pdf/2005-09-007.pdf (accessed February 1, 2007).
10. NASDAQ and Reuters, “NASDAQ and Reuters Launch New Venture to Help Companies Obtain Independent Analyst Coverage,” news release, June 7, 2005, available at www.nasdaq.com/newsroom/news/pr2005/ne_section05_056.stm (accessed February 1, 2007).
11. Anup Agrawal, Sahiba Chadha, and Mark A. Chen, “Who Is Afraid of Reg FD? The Behavior and Performance of Sell-Side Analysts Following the SEC’s Fair Disclosure Rules” (working paper, American Finance Association, Washington, DC, June 2005), 1.
12. See Brad M. Barber, Reuven Lehavy, and Brett Trueman, “Comparing the Stock Recommendation Performance of Investment Banks and Independent Research Firms” (working paper, Anderson School of Management, University of California,
Los Angeles, September 2005), available at www.anderson.ucla.edu/documents/areas/fac/accounting/research.pdf (accessed February 1, 2007).
13. Leslie Boni, “Analyzing Analysts after the Global Settlement” (conference presentation, Brookings-Nomura Seminar, Brookings Institution, Washington, DC, September 28, 2005), available at www.tcf.or.jp/data/20050928_Leslie_Boni.pdf (accessed February 1, 2007).