The Yale Law Journal

March 2016

Returning to Common-Law Principles of Insider Trading After United States v. Newman

Richard A. Epstein

abstract. Spurred on by the recent Second Circuit decision in United States v. Newman, this Feature examines the proper scope of the prohibition against insider trading under the securities laws. It argues that in some instances the law does not reach far enough, while in other instances the law reaches too far. On the first point, it is a mistake to require the government to show that a tippee who receives inside information supplies any kind of benefit to the insider before the tippee is subject to criminal prosecution. The simple status of the tippee as donee or bad-faith purchaser of improperly released information should suffice.

On the second point, the prohibition against fraud and manipulation contained in Rule 10b-5 should cover only those activities actionable under common-law theories dealing with misrepresentation, nondisclosure, and breach of fiduciary duty. In no way does the language or structure of the provision mandate a level playing field in which all players are entitled to have equal access to all nonpublic information. Accordingly, it is highly doubtful that Rule 10b-5 should apply to so-called misappropriation cases in which individuals improperly use confidential information for their own purposes, as in United States v. O’Hagan. Nor is it wise to create civil liability under Regulation Fair Disclosure (Regulation FD), which may well retard the production of useful information by requiring that it be shared simultaneously with all players. In both Regulation FD and misappropriation cases, private sanctions that regulate the uneven flow of information should suffice to control any abuses, and these sanctions should include the imposition of constructive trust, based on a restitution theory of unjust enrichment, against all tippees who know that they have received misappropriated information. It is much more difficult to decide whether to invoke criminal prosecutions for misappropriation of firm information against analysts who receive, directly or indirectly, information from insiders who disclose that information consistent with company policies intended to lift overall share levels. There is no reason for that question to be decided in a misappropriation context differently from how it is decided in other contexts, most notably that of trade secrets, where the legal response is itself divided.

author. Laurence A. Tisch Professor of Law, New York University School of Law; Peter and Kirsten Bedford Senior Fellow, Hoover Institution; and James Parker Hall Distinguished Service Professor Emeritus and Senior Lecturer, University of Chicago. Robert Miller, Ed Rock, and Joe Grundfest supplied valuable comments on an earlier draft of this paper. The paper also received a thorough vetting at the Law & Economics Workshop at NYU on October 28, 2015. My thanks to the participants for their trenchant criticisms, which I have tried to answer in the revised version of the paper. I would also like to thank Rachel Cohn, Madeline Lansky, and Krista Perry, University of Chicago Law School, Class of 2016; and Julia Haines, University of Chicago Law School, Class of 2017, for their usual excellent research assistance.


On December 10, 2014, the Second Circuit handed Preet Bharara, the hugely ambitious United States Attorney for the Southern District of New York, one of his rare defeats in securities fraud litigation. In United States v. Newman,1 the Second Circuit unanimously reversed, with prejudice, the insider trading convictions for securities fraud and conspiracy to commit securities fraud of two analysts, Todd Newman and Anthony Chiasson. The actions were brought pursuant to Section 10(b) of the Exchange Act and Rule 10b-5, which are set out in the margin.2 In the words of Judge Parker: “The Government alleged that a cohort of analysts at various hedge funds and investment firms obtained material, nonpublic information from employees of publicly traded technology companies, shared it amongst each other, and subsequently passed this information to the portfolio managers at their respective companies.”3 According to the indictment, these analysts then passed this information on to Newman and Chiasson, who “willfully” participated in the scheme by trading on this information in the course of their own business.4 The taint arose because this behavior was inconsistent with Regulation Fair Disclosure (Regulation FD), which requires that the information must be disclosed simultaneously to all outsiders if it is disclosed to any.5

The unexpected outcome in this case has fueled an effort to reexamine the fundamental principles governing insider trading, which are still in flux even today. In order to reexamine these principles, this Feature proceeds as follows.

Part I gives an account of the various factual and legal issues that were arrayed in Newman to set the stage for a more comprehensive reexamination of the fundamental principles of insider trading.

Part II examines—in light of Newman—the two major forms of insider trading liability, the so-called classical theory and the more modern misappropriation theory, first as they apply to insiders and then as they apply to tippees (those persons who traded on the tipped information). Part II argues two points. First, it argues that contractual solutions work better than regulatory solutions to constrain various forms of misrepresentation, concealment, and nondisclosure that arise in connection with insider trading. Second, it argues that the standard doctrines of the constructive trust do better than the so-called personal-benefit test of Dirks in identifying which tippees should be subject to liability for receiving information; this is the case where a constructive trust theory undoes the unjust enrichment that takes place if tippees are allowed to use that information for their own benefit.

Part III extends the analysis of the misappropriation theory of securities violation to critique Regulation FD. Regulation FD places an unfortunate straitjacket on how various firms do business with analysts of their stock. The insiders owe no fiduciary duties to analysts. But they do owe such duties to their shareholders, and the firms’ officers and directors should be allowed to authorize their employees to make selective disclosures of inside information so long as these officers and directors have concluded in good faith that the release of that information will increase overall firm value.

Part IV applies the conclusions reached in the earlier three parts to examine more closely the role that the personal-benefit test and information flows have in dealing with insider trading. On the former, the Second Circuit incorrectly stiffened the government’s burden on the personal-benefit test relative to what it was in Dirks. On the latter, the government’s inability to trace the flow of information from the insiders to the defendant tippees moots the former error—at least on the evidence accepted in the Second Circuit—and justifies the outcome, but not the reasoning, in Newman.

Part V then applies this general analysis to other recent cases, both before and after Newman, to further examine the contours of the misappropriation theory. In general, the cases are correct to downplay the personal-benefit prong of the test. These cases also illustrate the vast gulf that exists between the disclosure of information in the ordinary course of business, for which no liability should be imposed either on the insider or any tippees, and the clandestine release of information, which in virtually all cases should result in criminal liability.

A short conclusion then urges a return of the law of securities fraud to its traditional contours, which should limit criminal prosecutions to insiders and their tippees who make deliberate use of information that they know was limited to use for firm purposes. The most appropriate goal of insider trading laws is not to advance some ad hoc theory of fairness, which typically shrinks the size of the pie without offering any coherent account of how that reduced stock of wealth should be divided. Instead, insider trading laws should work to increase capital market efficiency, which often requires the Securities and Exchange Commission (SEC) to shrink its oversight role.

I. newman and chiasson in the dock

In May 2013, Todd Newman and Anthony Chiasson were both sentenced for securities fraud and conspiracy to commit securities fraud after a six-week trial before Judge Sullivan.6 Newman was a portfolio manager at Diamondback Capital Management, LLC, and Chiasson was a portfolio manager at Level Global Investors, L.P.7 The two men were charged with trading on inside information that originated inside two companies, Dell and NVIDIA.8 Neither man received the information directly from parties inside the companies, but only through a set of intermediaries.9

With respect to Dell, Newman’s inside source of the information was Rob Ray, an employee in Dell’s investor relations department. Ray tipped some information about anticipated earnings reports to Sandy Goyal, an analyst at Neuberger Berman. That information was in turn relayed to Jesse Tortora, an analyst at Diamondback, who in turn gave that information about Dell first to Newman and then to Spyridon Adondakis at Level Global, who in turn passed the information on to Chiasson. Newman was thus three persons removed from the original source and Chiasson was four persons removed. The identical information was also given to other analysts from different companies.

The chain of communication with NVIDIA started with Chris Choi, who worked in NVIDIA’s finance unit, who tipped information about NVIDIA’s earnings projections to Hyung Lim, an executive at another technology company, whom Choi knew from church. Thereafter, Lim shared the information with Danny Kuo, an employee at Whittier Trust, who in turn circulated it to a group of analysts including both Tortora and Adondakis, who in turn gave the information respectively to Newman and Chiasson, so that in both cases the chain contained three links.

In dealing with these two criminal indictments, the district court held that the government could make its case by showing that the defendants “knew that the material, nonpublic information had been disclosed by the insider in breach of a duty of trust and confidence” owed to his employer.10 At no point did the district court instruct the jury that the corporate insiders who had provided the inside information to the defendants must have done so in exchange for some personal benefit to them, which was required for a conviction under the Supreme Court’s decision in Dirks.11Defendants were convicted on all counts. Newman had to pay about $1,738,000 in fines and forfeitures and was sentenced to fifty-four months in prison followed by one year of supervised release.12 Chiasson had to pay up to seven million dollars in fines and forfeitures and was sentenced to seventy-eight months in prison followed by one year of supervised release.13

Two questions were posed on appeal. The first was whether the evidence introduced into the record could support the proposition that the insider had received a personal benefit on the strength of his personal and social ties.14 The second was whether the chain of causation that linked the defendants as remote tippees of the leaked information was tight enough to support the claim that the two defendants “knew that they were trading on information obtained from insiders.”15 Regarding the personal-benefit issue, the Second Circuit reached two conclusions. The first was that the personal benefit to the insider standard required showing more than some social friendship or connection.16 The second was that, in light of the weakness of the evidence, the failure of the District Court to instruct on the personal-benefit question should not be disregarded as “harmless” error.17 On the chain-of-causation issue, the court found that the government presented “absolutely no testimony or any other evidence” on the critical scienter requirement—that is, whether Newman and Chiasson knew that they received forbidden information.18 One conclusion from that observation is that the original insiders may have been guilty, but the recipients of the information were not.

In my view, the Second Circuit’s decision to dismiss the complaint with prejudice was correct under current law. However, for purposes of this Feature, the outcome of the case is less important than the legal framework used to decide it; there are doctrinal and institutional issues raised by Newman, both under current law and as a matter of first principle. Under current law, the two key elements involved in this case should not be considered of equal importance in the general theory of insider trading. In a first-best world, the requirement of a “personal benefit” as derived from Dirks should be regarded as a red herring and removed from securities cases altogether—which is a real advantage to the government. Similarly, the requirement not to trade on inside information should not be limited only to individuals who are subject to fiduciary duties.19 But more importantly, the remainder of Newman deals with the knowing use of nonpublic and material information, which is hard to analyze given the complex paths over which that information traveled before resulting in trades in the two companies’ shares. As will become clear, the defendants in Newman should be acquitted under current law. The sole ground needed for reaching that conclusion is that the Second Circuit found that there was no evidence in the record to establish that the defendants knew they were trading on unauthorized information released in violation of the fiduciary duties of insiders.

Normatively, it is also appropriate to ask whether Regulation FD should impose a uniform obligation that requires that all information be released simultaneously to all persons. If that regulation is dismissed, then the scope of securities laws will be necessarily narrowed because the definition of inside information will be narrowed. This narrowed scope would allow parties autonomy over any information that is released in accordance with firm policy, without fear of SEC enforcement.20 It follows, therefore, that the Regulation has no direct effect on enforcement actions by ambitious United States Attorneys like Preet Bharara, but the Regulation nonetheless reinforces the belief that an extensive reading of the securities law works in general for the public interest. In the remainder of this piece, I shall deal in Part II with matters of first principle and then turn in Part III to the larger question of the extent to which the government, through the Department of Justice, should engage in criminal enforcement of the securities fraud law.

II. back to first principles

A. Contract Versus Regulation

As a matter of first principle, I am in general deeply suspicious of any government-imposed insider trading prohibitions and think that they do little to improve the overall condition of trading in American securities markets beyond what private agreements can achieve.21 The best way to examine this question is to start with yet another variation of the “single owner” theorem.22 Start with one person who owns any particular asset that is divided among multiple players through a succession of contracts, such that when the dust settles, the network of contractual arrangements binds each person to everyone else in that common business venture, regardless of its form. At this point the notion of “externality” disappears because all of the harms that come to anyone locked within this system are borne by the original owner when he parts seriatim with portions of his initial shares in the new company. Hence any increase that he gets by giving one party an advantage over another redounds both to his benefit and harm. The benefit comes from the one side, but the harm comes from having to reduce the amount charged to the other purchasers or to the interest that he retains. In the relatively easy cases, the parties stand in symmetrical relationships to each other so that each gets a fixed fraction of the pie. On that austere assumption, the original owner will always opt to pick the solution that maximizes the size of the whole pie in order to maximize the size of his own slice. This situation can arise with various land-use transactions, when parcels are sold off subject to a network of reciprocal covenants and easements that bind all to all, regardless of their respective times of acquisition.23

The same logic applies to the formation of a corporation, as Dennis Carlton and Daniel Fischel argued many years ago, if the original charter contains key governance provisions that govern all subsequent purchasers.24 In this context, the question is whether—and if so, in what form—these parties would opt to include some prohibition on insider trading, knowing that they will have to internalize the insider trading rules. It would be a grievous mistake to assume that they would choose to impose no restrictions whatsoever. More precisely, they will impose these restrictions whenever they think that the benefits to the firm’s shareholders on net outweigh their costs, for that maneuver will allow the parties to maximize the revenues that they can get from an original public offering. It hardly follows, however, that the optimal set of restrictions generated by this approach would look anything like the current prohibitions on insider trading, most especially the criminal sanctions that were at issue in Newman. The point here is not that investment markets can thrive in the constant presence of fraud and manipulation. The point is quite the opposite. The risks of fraud and manipulation are so deadly to the market that private firms have every incentive to seek out the optimal solution to insider trading, whether by directors, officers, or ordinary employees, wholly apart from any government sanctions, in order to preserve the value of their shares. It is of course difficult if not impossible today to demonstrate the truth of this proposition by empirical evidence, given that every firm now works in the shadow of the insider trading laws and thus has to address these issues in the current externally regulated setting. But as will become clear later in this Feature, the scope of the private prohibitions on insider trading is often quite extensive, and goes well beyond what the law requires. Reputation in general is a powerful determinant of firm value, and it can be eroded by regulation that undermines the operation of traditional informal bonding devices.25 That proposition applies to insider trading as much as it does to any other government activity.

Evidently, to state the problem in this general form is not to solve it, because of the complexity of any disclosure regime. One source of the difficulty is clear enough. Any public disclosure goes not only to shareholders of a particular firm, but also to competitors of the firm, who can take advantage of that information in planning their own behavior. Thus, in SEC v. Texas Gulf Sulphur Co.,26 a group of insiders purchased shares of the company knowing that the firm had discovered valuable copper deposits near Timmins, Ontario, which the firm desired to keep confidential until it completed the acquisition of nearby lands.27 Secrecy benefited the shareholders, because at this point it would have been foolish to publicly disclose these discoveries, which would allow competitors to hone in on the same territory and thereby drive up the acquisition cost of nearby land. Similarly, it would have been equally improvident for the insiders to dole out that information piecemeal to their friends and business associates, knowing that any actions that these people took to acquire nearby properties would necessarily work to the detriment of the shareholders.28 In contrast to these two scenarios, allowing the insiders to trade on that information could have driven up share prices in ways that reflected the value of that information, without disclosing its source. Insider trading could thus have led to more accurate pricing that in turn would have reduced sharp fluctuations in share values when the information did come to light.29

This bald proposition is, and should be, contentious, and the same answer might not work for all firms in all cases. But this is the juncture where the contractual approach shows its strength. By announcing in advance that insiders may trade on nonpublic information, that declaration allows insiders to initiate price movements useful to the general public without having to link them to the Timmins site. In response, it could be argued that the only source of market inaccuracy in these cases is a short delay in the correction of stock prices, a cost that is worth bearing to protect against various sources of insider abuse.30 But at the same time, the social cost of the delay may be great even if the time until correction is short. This outcome is possible whenever other parties in related businesses make major decisions to the detriment of the firm right after the prompt disclosure of the information. Thus in Texas Gulf Sulphur, nearby landowners could raise the prices that they charge for mineral leases. It is hard to know in the abstract what the right answer is, and it could well be that contractual disclosure norms in the absence of the current SEC prohibitions would evolve if the risk of unjust insider enrichment were not offset by some gain to the firm at large. It is therefore possible that consensual arrangements would reach the same position that the law requires today—namely, that the insider must live with the choice to either disclose or abstain from trading.31 Under government regulation, the disclosures have to be full, but it is possible that in some situations it may be wiser for insiders to disclose that they have either bought or sold, without explaining why. Or, perhaps, some limits could be placed on the number of shares that various key figures are allowed to purchase, or the number of options they may be allowed to acquire.

These permutations could set up a yellow flag to others without disclosing the information in question and risking the flaws of Regulation FD. In one sense, Texas Gulf Sulphur is the exceptional situation because it involves buying on good news, not selling on bad news. Even in the latter situation, the firm might be able to limit shareholder losses by inducing price signals, but it is harder to think of scenarios in which it seems clearly wise to abandon the disclose-or-abstain position of modern law.

There is also a third scenario in which it appears that the securities laws do impose excessive liability on insiders. This is the dilemma that corporate insiders face when they are asked whether the firm is in play, as in Basic Inc. v. Levinson,which held that insiders could be sued under the insider trading laws when they falsely denied that the company was engaged in merger negotiations even when they did not trade in the stock.32 At this point, the risk of self-aggrandizement is gone, so the actions in question could be justified as the only way to secure the confidentiality needed to increase the odds that the deal could go through. Quite simply, if the information becomes public, the stock price of the target moves upward, which will in turn make the deal less attractive to the acquiring corporation.

In these circumstances, it seems appropriate to allow the directors to determine in good faith whether it was proper to deny the existence of transactions. In practice, firms are able to take one effective countermeasure, which is to announce in advance a uniform policy of never commenting on possible takeover transactions, which in this context at least allows them to avoid potential liability under the fraud-on-the-market theory, which is still very much in an unhappy state of flux.33 But even in the cases where that theory is allowed to operate, it surely cannot make sense to have a regime in which the insiders have to compensate in full all those shareholders who sold in ignorance of the information, without being able to recoup the gains from the outsiders who were fortunate enough to gain from the delay in the release of the information. Whatever the social losses from the delayed release of the pricing information, it is far smaller than the potential liability under the securities law.

Whatever the ideal solution, however, I see no comparative advantage in having the SEC decide once and for all what the ideal strategy is in cases of this sort, especially as a criminal matter. More specifically, there is little reason to credit the view that insider trading, if subject only to contractual limitations, should be regarded as a threat to the integrity of the U.S. securities market. That proposition would be true if the patterns of trading by insiders were not disclosed in advance. But once the key corporate documents reveal the relevant information about insider trades, the market has more information about what will happen rather than less. The common SEC position on this point, as follows, ignores the issue of advance disclosure:

It is the trading that takes place when those privileged with confidential information about important events use the special advantage of that knowledge to reap profits or avoid losses on the stock market, to the detriment of the source of the information and to the typical investors who buy or sell their stock without the advantage of “inside” information.34

This passage sets out the conventional rationale for the prohibition on insider trading. Its great vice is that it frames the issue in terms of dealing with the gains to the insiders relative to the losses to other parties. By casting the issue in this fashion, the SEC necessarily expands its role and that of the Department of Justice until they can become perpetual censors of all that goes on in the day-to-day operation of markets, without clearly explaining how it is that ordinary investors, many of whom are clients of the firms charged with criminal offenses, are themselves helped by the government action. The SEC pronouncement makes it appear as though the central calculation concerns the distribution of benefits and losses to various players; the one social objective of the insider trading rules, however, is to improve the pricing of shares and thus the long-term market effectiveness, which redounds ex ante to the benefit of all market participants. More concretely, there is no reason to worry about any “detriment to the source of the information” who is in a position to take care of himself by contract. Nor is there any reason to worry about the position of the public at large, all of whose members can organize their trading strategies with full knowledge of the permissible activities of the insiders by looking to the corporate policy on insider trades.

The counterstrategies are legion. One strategy that is available to small investors is to adopt a buy-and-hold strategy in which they keep, at low administrative costs, a balanced portfolio. The portfolio allows them to share in favorable price movements generated by insiders (and the cost of sharing unfavorable movements as well) without having specific knowledge of those events. Another approach is for typical investors to buy shares in a mutual fund that is run by managers who are familiar with the intricacies of the marketplace, and thus can fend for their shareholders. A large information problem is thus displaced by a much smaller agency cost problem. Finally, these shareholders are entitled to all the protections against various forms of insider trading that a firm imposes on its insiders in order to induce others to invest in capital markets. The SEC does not have to mount a charge to protect typical investors who in modern capital markets have cheaper and more effective ways to protect themselves. The narrower focus on controlling traditional forms of fraud offers a far higher rate of return on public administrative dollars than the SEC’s preferred method.

B. The Classical and Misappropriation Theories of Insider Trading

The shakiness of the basic SEC position is revealed by a closer examination of today’s common typology that distinguishes between the classical form of insider trading and its misappropriation variation. Both theories received their canonical formulation in the 1997 decision of United States v. O’Hagan35:

Under the “traditional” or “classical theory” of insider trading liability, § 10(b) and Rule 10b-5 are violated when a corporate insider trades in the securities of his corporation on the basis of material, nonpublic information. Trading on such information qualifies as a “deceptive device” under § 10(b), we have affirmed, because “a relationship of trust and confidence [exists] between the shareholders of a corporation and those insiders who have obtained confidential information by reason of their position with that corporation.” That relationship, we recognized, “gives rise to a duty to disclose [or to abstain from trading] because of the ‘necessity of preventing a corporate insider from . . . tak[ing] unfair advantage of . . . uninformed . . . stockholders.’” The classical theory applies not only to officers, directors, and other permanent insiders of a corporation, but also to attorneys, accountants, consultants, and others who temporarily become fiduciaries of a corporation.

The “misappropriation theory” holds that a person commits fraud “in connection with” a securities transaction, and thereby violates § 10(b) and Rule 10b-5, when he misappropriates confidential information for securities trading purposes, in breach of a duty owed to the source of the information. Under this theory, a fiduciary’s undisclosed, self-serving use of a principal’s information to purchase or sell securities, in breach of a duty of loyalty and confidentiality, defrauds the principal of the exclusive use of that information. In lieu of premising liability on a fiduciary relationship between company insider and purchaser or seller of the company’s stock, the misappropriation theory premises liability on a fiduciary-turned-trader’s deception of those who entrusted him with access to confidential information.36

This oft-quoted passage bristles with conceptual difficulties. I shall look at them first in connection with the immediate parties to the transaction, and then extend them to deal with the vexing question of the tippees, the subject of the prosecution in Newman.

1. Insiders

The difficulty with the so-called classical theory is that it does not take into account the notion that the fiduciary duties in question sound in contract, not in regulatory fiat.37 So long as there are appropriate disclosures in advance as to the rules of the game, there is no deception and hence no manipulation within the meaning of the securities law. Nor is it appropriate to say that the insiders have taken advantage of “uninformed shareholders,” because the level of knowledge that the shareholders acquire is not externally fixed for all time but rather depends heavily on the rules of the game in which trading takes place. Thus, all parties have to acquire some information before they decide to trade, and how they acquire that information greatly depends on the known legal environment in which the market operates. Where the insiders announce that they will trade, that information will be incorporated into the market as individuals change their pattern of trading, or, more commonly, hire other people to do the trading for them (perhaps by using brokers or investing in mutual funds with professional management). The breach of fiduciary duty should be confined to those cases where the behavior of insiders is contrary to their stated positions. Otherwise, there is no “unfair advantage” at all. In general, a comprehensive theory of insider trading has no place for the classical theory about the misuse of material nonpublic information, unless this potential ground for liability is waivable by the corporation’s shareholders, which under current law it decidedly is not.

The position of insiders under the misappropriation theory is quite different. In these cases, the party who has taken advantage of the inside information is someone who receives that information as part and parcel of his duties on behalf of the firm. At this point, the use of this information by its recipient does not cause any harm to the market at large. Indeed, by trading on accurate information, the new trader improves the accuracy of market pricing, making the overall market even more efficient. But the correct focus of the misappropriation theory has nothing to do with outsiders to the firm. It has to do with the damage that the recipient of the information does when he trades against the interests of his two principals—the firm for which he works and the client who has retained the firm. In this connection, the misappropriation of information not only causes short-term dislocations, but it also reduces the frequency of deals by making it harder to organize them in secrecy, as in the Basic situation.38 But that huge risk is no mystery. It is all too well known to principals everywhere who understand the “agency cost” risk, given that the agent’s incentives are never perfectly aligned with those of the principal.39 It should come therefore as no surprise that, in many cases, we observe even today sharp limitations on the use of information that are privately imposed to ensure that employees who receive valuable inside information from their firms’ clients do not use that information to hurt either their firm or its clients.40

In this instance the need for public enforcement is much reduced. But it need not be eliminated. In the first instance, it may well be that the firm in question needs to rely on the SEC to turn over information that the agency holds in its system, which allows the firm to learn of all trades made by those persons with whom it has trusted information. If so, then its contracts could specify that the firm will rely on the SEC to do its detective work. Indeed, it could go further and indicate that it clearly supports criminal prosecution for the abuses of information. Yet again, it is dangerous to conclude a priori that a firm would choose to turn to the SEC for all or part of its enforcement business. It could instead rely on its own internal reporting requirements—for example, turning over brokerage statements and tax returns, and relying on exchange data—to gain the needed information. Or it could find some other private firm, which specializes in this line of compliance work, to whom it could delegate its inspection and monitoring work. Indeed, that work could be done by the exchanges themselves, which could make clear what practices must be followed in order to be listed. Individual firms often require that their employees make available to them all their own private financial records and those of their family members as well.41 In general, therefore, in an unregulated setting the first line of defense is likely to be private. Subject to some complications addressed later,42 criminal law may well be invoked in cases of misappropriation of what are in essence firm trade secrets, just as it may be invoked in any other case of employee misappropriation of corporate assets. The theory of criminal liability for this trade secret information is similar to that applicable to the embezzlement of corporate funds.43 There is of course no reason why the government criminal prosecution requires the cooperation of the corporation in this, any more than in other cases of misappropriation or theft. The government can decide to prosecute for the misappropriation even if the corporation does not, although the government may in practice be less willing to do so.

The basic points of this analysis are well illustrated by the facts in O’Hagan, in which the Supreme Court accepted the misappropriation theory of securities fraud.44 O’Hagan was a lawyer for the Minneapolis firm Dorsey & Whitney.45 O’Hagan knew that Dorsey & Whitney’s client, a British company, Grand Metropolitan (“Grand Met”), was planning a tender offer for Pillsbury.46 Both Dorsey & Whitney and Grand Met took steps to keep Grand Met’s proposed tender offer under wraps.47 O’Hagan did no work on the deal but acquired both stock and options in Pillsbury during the period that Dorsey & Whitney represented Grand Met.48 After Dorsey & Whitney withdrew from its representation, Grand Met made public its tender offer, which drove up the price of both the shares and options, which O’Hagan then sold, reaping a profit in excess of $4.3 million.49 O’Hagan knew that the transaction was confidential.50

Wholly apart from any of the finer points of securities laws, it is clear that Grand Met supplied this information to its lawyers and investment advisors so they could use it only for Grand Met’s benefit in the tender offer. It was equally clear that Grand Met had to invest considerable resources to determine that the tender offer made sense. Grand Met also knew that the price of its tender offer would have to rise if other individuals found out about its interest in Pillsbury, which is why it insisted that all parties who worked on the transaction would not appropriate that information for their own benefit. This was not a close case. Wholly apart from the securities law, it is quite inconceivable that any employment agreement would allow a company agent to bid up the price of a target against his principal on the strength of the information that the agent acquired from its client. The standard duty of loyalty requires that such information not be used in ways that hurt the client. It is an open question whether the early purchases of Pillsbury stock hurt the public at large. But the answer to that question does not matter to the employer or client. The harm to a trading partner in breach of contract is always actionable regardless. Left unexplained is why the SEC has to get into the middle of this fight by setting employee standards.

The general lesson on insider misappropriation is this: any firm that uses inside information to trade against its own customers will not last long in the marketplace, as potential clients will move elsewhere for their business. Here is the proof: all law firms and all investment banks have elaborate rules in place that limit the ability of their partners and associates to trade on information they acquire in the ordinary course of business, many of which right now go beyond the SEC requirements.51 At this level reputational constraints are so powerful that any lawyer, banker, or accountant caught using confidential information for his own benefit would be signing his professional death warrant. The problem of the misuse of confidential information of course goes beyond the securities context, so the security-specific rules are often supplemented by legal constraints on the overall practice of law, which imposes, most notably, duties of confidentiality for lawyer-client communications52 and work product privilege.53 It is unlikely that the SEC is the best actor to catch the odd case that might slip through these two types of sanctions, given the strong private incentives to make sure that errors of this sort do not happen.

For example, in one recent account, Goldman Sachs took prophylactic steps “to block bankers and other employees from trading individual stocks and debt securities in their own personal accounts, or investing in certain hedge funds.”54 For a firm whose practice extends to all market groups, the broad rule is likely needed to assure Goldman that its traders will not trade against Goldman itself with firm information, and it allows Goldman to reassure its clients that their information will not go astray when entrusted to Goldman’s employees in a sensitive deal. For other firms with different business profiles, perhaps a less restrictive rule on individual trading might do. But whether or not this is true, the SEC does not have individualized information that allows it to make more sensible determinations than those now made by market-driven actors. If anything, the SEC’s broad discretion gives it too much power to decide which perceived violations of the law to chase after and which to ignore, in ways that could lead to invidious favoritism of some parties over others. Since the private-law response is so powerful, why go through tortured reasoning to determine the scope of liability under Rule 10b-5 for insider trading? And is it wise to impose criminal sanctions if the firm has no desire to do so?

It should be evident that the concerns with insider trading are most acute with financial service firms, banks, and law firms, which constantly acquire information from all sources as a routine part of their businesses. But virtually every firm has an insider trading policy that is calibrated to the risks that it faces. For example, the Pitney Bowes insider trading policy55 takes a two-tier approach in which more stringent preclearance obligations are imposed on senior officials, called “restricted persons,” before entering into any planned transaction in Pitney Bowes securities. NetLogic Microsystems, Inc. has a general prohibition against insider trading, coupled with an injunction directing employees who have questions about the policy to contact the chief financial officer.56

Against this backdrop, O’Hagan represents a situation in which the application of the securities laws is redundant in light of the specific contractual prohibitions on the use of this information. These rules, moreover, form part of a larger set of institutional arrangements intended to protect confidential information. Thus, the misuse of that information is also covered by the usual rules of corporate law that impose duties of loyalty and care on directors and key officers. In the absence of any explicit waiver, the duty of loyalty surely applies to O’Hagan’s actions as an agent of the company.57 Ironically, the securities case law ties itself into knots in order to come up with the right answer. Its first move is to distinguish between the “classical” form of insider trading, like that in Texas Gulf Sulphur, where insiders trade firm stock on inside information, from cases where the stock traded is that of another company (in O’Hagan, the target of Dorsey & Whitney’s own client). At this point, the applicable theory is that the case involves the conversion of inside information to purposes that are prohibited by the owner of the information.

The problem here is a very old one when dealing with tangible objects. Under Roman law, it was held that any knowingly unauthorized use of a chattel constituted a form of theft, which was then a delictual offense—a rough cross between a civil and criminal sanction—that allowed the injured party to recover multiple damages for the actual loss.58 The rule had its inevitable ambiguity, for it is not crystal clear that entering a borrowed horse in the steeplechase is outside the scope of the original loan. But given the extra risk, that conclusion seems clear enough. The unauthorized use of information should be treated exactly as the unauthorized use of chattel. Whether or not there is an explicit policy, using the information against the principal is a virtual per se violation of the employee’s contractual duties. It is therefore odd that securities law has to go into flights of conceptual fancy before it concludes, quite simply, that “[t]he undisclosed misappropriation of [confidential] information, in violation of a fiduciary duty . . . constitutes fraud akin to embezzlement—‘“the fraudulent appropriation to one’s own use of the money or goods entrusted to one’s care by another.”’”59 The exact same principles that apply to the misappropriation of a chattel apply to the misappropriation of a trade secret. There is no need to reinvent the wheel on issues that have already been resolved.

Therein lies the correct starting point: treat the law of fiduciary duties as the baseline for Rule 10b-5. The hard question here is whether the breach of these contractual duties of loyalty should be regarded as serious enough to merit criminal prosecution. It is easy enough to imagine situations where that might be the case. One of the most serious difficulties in the law of insider trading is that any given bit of information is of equal value to all comers, regardless of the income that they derive from the firm. The point is important because it indicates that private sanctions, such as dismissal or demotion, will not hit all employees equally. To take a highly stylized example, assume that certain information is worth $100,000. A low-level assistant in the mailroom will think that the loss of $30,000 in salary is a small price to pay for using the illicit information. A trader that makes $1 million may well take the opposite view, at least on these stylized facts. Yet it does not follow that criminal sanctions are unnecessary in these contexts because the same information could easily be worth more to the high-placed insider who has greater capital to invest. But by the same token, the low-level employee could connect with outsiders who have capital in order to increase his earnings.

The implications here are hard to sort out. The first lesson is that prevention and monitoring are critical across the board, which is why the Goldman board takes the position it does. The second lesson is that, in any given case, ordinary criminal sanctions for theft can be imposed on all insiders, whether rich or poor, who take advantage of trade secret information. In this regard, the problem here is no different from that involving embezzlement of a constant sum of money, which is always a larger temptation for a low-income employee who has less to lose than a high-level one. But there is no reason to have a special SEC regulation to deal with these situations after the fact. The usual rules of criminal law on misappropriation should be able to cover the case every bit as well.

In other cases, it is somewhat less clear that the behavior is serious enough to merit criminal prosecution. For example, R. Foster Winans was convicted of insider trading for taking information that he received for the Wall Street Journal’s “Heard on the Street” column and supplying it before publication to third parties who then traded on that information in anticipation of the column’s effect on the market.60 The breach of fiduciary duty seems clear. At this point, criminal liability seems to follow for the knowing misappropriation. What is less clear is how the matter would be resolved if approval of the Wall Street Journal were somehow needed for the government to commence its prosecution. Would the Wall Street Journal have been prepared to write a contract with Winans stipulating that if the information were used to facilitate insider trading by others, Winans would be subject to criminal sanctions? If so, then the case is easy. But perhaps the Wall Street Journal would have considered it sufficient deterrence to dismiss Winans, demote him, or dock his pay, knowing that strong reputational sanctions would prevent him from getting another job in the industry.

As ever, the mixture of remedies is hard to determine a priori. Indeed, part of the difficulty lies in the changes in social perceptions of the misappropriation of trade secrets. Winans’s actions took place in 1983,61 at a time when trade secret misappropriation had “been the near-exclusive province of state civil law, usually in the form of the Uniform Trade Secrets Act . . . .”62 But the legal landscape changed radically with the passage of the Economic Espionage Act of 1996,63 which imposed federal criminal liability for the misappropriation of trade secrets. In dealing with this background set of expectations, the argument for criminal sanctions was weaker when Winans’s case was decided than it is today. In 1983, it seems unlikely that the various parties who supplied the column with sensitive information would have demanded that criminal sanctions be imposed on Winans. The problem here is an old one.

The solution to any enforcement problem always seems easy to those government agencies that think underdeterrence is the only game in town. The choice of remedy becomes vastly more complicated when one acknowledges that there are two forms of error, and that overdeterrence—especially criminal sanctions in particular cases—may cause negative externalities. So if we think of the insider trading laws as a cocked gun in the cupboard of private parties, the distress call to the government for criminal sanctions is unlikely to go out very often. Private parties are likely to prefer working within a framework that combines a private set of ex ante precautions, ex post firm sanctions, and a large reputational hit. Yet even though the general legal climate has moved toward criminal sanctions, it is less likely that private firms, such as Goldman Sachs, would move toward that solution if left to their own devices.

2. Tippees

The next part of this inquiry addresses the systematic treatment of tippees—i.e., third parties who receive information from an admitted insider. The problem in question can arise with either of the two canonical forms of insider trading, the classical and misappropriation theories, assuming these to be otherwise viable. To understand the correct approach, it is critical to understand that the parallels to the law of chattels, conversion, and trust work as well as they do in the case where the insiders themselves use the information in question.

On this question, the seminal Supreme Court decision in Dirks v. SEC lays down an underinclusive rule that insider liability extends to any outsiders who “have entered into a special confidential relationship in the conduct of the business of the enterprise and are given access to information solely for corporate purposes.”64 But the Court in Dirks then qualified this proposition by invoking its earlier decision in Chiarella v. United States65to say that “there can be no duty to disclose where the person who has traded on inside information ‘was not [the corporation’s] agent, . . . was not a fiduciary, [or] was not a person in whom the sellers [of the securities] had placed their trust and confidence.’”66 This includes the printer in Chiarella who misappropriated information that he had acquired while working as a “markup man” preparing documents about a pending corporate takeover. The argument in Dirks and Chiarella was that to extend the duty beyond fiduciaries would necessarily result in “recognizing a general duty between all participants in market transactions to forgo actions based on material, nonpublic information.”67 That refusal to extend the prohibition against misappropriation beyond fiduciaries is all too favorable to the tippee because it ignores the important intermediate case where the third party, like Chiarella or O’Hagan, receives information that he knows he should not use for his personal gain. In dealing with this issue, there is much to be said in favor of Rule 10b5-2, which gives a broad account of which individuals should be subject to a duty of trust and confidence, by stressing both actual agreement on the one hand and shared expectations from a course of dealing on the other.68

The strength of this particular rule against misappropriation has deep roots outside securities markets. In other contexts, the gains from that misappropriation should be treated as being held in a constructive trust, with a duty to turn such gains over to the proper holder of the information.69 The term “constructive” in this situation is there for a purpose. An instructive parallel is the term “constructive notice.”70 It is quite clear that parties who take property with actual notice that they cannot receive it have engaged in misappropriation. It is equally clear that in many cases a party has enough information to know that there is a serious risk that the party who claims to own the property does not. Just that happens when A claims to have title to sell land on which B is now living.71 The same notion applies when the purchaser of property can learn of the legal state of the title by an inspection of public records.72 At this point, the potential buyer is under a duty to inquire further to discover the true state of the title. Once again, the use of the term “constructive” concedes that there is no actual knowledge, but that the duty is nonetheless imposed so that the potential buyer cannot turn a blind eye to a risk of which he is or should be aware.

The same approach applies to the constructive trust. This relationship is routinely imposed on third parties who receive chattels or land that they know to be, or of which they have constructive knowledge is, owned by someone else. There is no implication that they have voluntarily assumed any fiduciary duties to the true owner of the property. Quite the opposite: it is well known that they have no such intentions at all. But the obligation of a fiduciary is to preserve the asset value for the beneficiary. That same duty should be imposed under a theory of unjust enrichment against any party who is in possession of stolen information that he knows is not his. Hence the constructive trust is imposed to force him to act as if he were a trustee, which means that he must make restitution of the monies received (and any gains derived from their use) to their rightful owner. Indeed, generally the duty is so strong that the constructive trustee is faced with the following no-win alternatives. If he takes the money or other property and invests it in a risky venture, he loses either way. If the investment goes up in value, he pays over the full amount. If it goes down in value, then he must pay back the original sum with interest.

These arguments extend to the transfer of information to tippees that they know or should have known to be illegally taken. The tippees are treated as though they are trustees and thus have to turn over all their winnings to the true owner of the property, a rule that applies with full force to the defendant printer in Chiarella. This rule does not require, as Dirks intimates, that all players be on an equal footing in securities markets. Quite the opposite: it only deals with people who receive illegal disclosures of information. In a world devoid of Regulation FD, the imposition of a constructive trust for misappropriated information does not require a firm to make disclosures either to all or to none, but leaves that decision in private hands. The constructive trust language that is appropriate for Chiarella no more upsets the market for the sale and use of information in securities markets than the parallel duties upset the markets for the sale and use of land, chattels, or any other property that can be illicitly converted. Dirks thus runs sharply counter to the private law, and should be rejected on this point as all too favorable to the tippee who trades improperly on inside information. Whether its rejection creates the case for criminal sanctions is a separate question to be analyzed on the same grounds as above.73

Dirks also misfires on the question of whether proof of a violation of insider trading prohibitions would require that the tippee of information supply some return benefit to the party who supplied the tip. Once again, the rules that govern information are similar to those that govern other forms of property, such that the donee who takes with knowledge is again subject to the trust whether or not he supplied some nonpecuniary benefit to the tipper. In contrast, Dirks stands for the following canonical proposition:

Whether disclosure is a breach of duty therefore depends in large part on the purpose of the disclosure . . . . Thus, the test is whether the insider personally will benefit, directly or indirectly, from his disclosure. Absent some personal gain, there has been no breach of duty to stockholders. And absent a breach by the insider, there is no derivative breach.74

Stated at that level of unguarded generality, the proposition must be wrong; the test of criminal liability is too restrictive. Here once again, analogies to ordinary fiduciary duties of trustees and directors, so useful in dealing with the conversion analogies, help clarify the situation. Under the standard rules of trust, any person who receives property, including shares of stock, will be subject to the trust unless he is a bona fide purchaser for the value of the legal interest in the property in question.75 The point of these requirements is to impress the (constructive) trust on two classes of individuals who receive a trust property from the trustees. The first are purchasers with knowledge that the trustee does not have the authority to sell. They are co-conspirators and not innocent purchasers. The second are the donees of the property, who will have to surrender it back to the trust for the simple reason that no person is allowed to make gifts to his friends of property that is owned by another. “Be just before you were generous” was the way the point was put to me many years ago by Yale’s late bankruptcy professor, J. William Moore. The question of “derivative liability” is quite beside the point.

The question then arises as to whether the analysis ought to change when what passes between parties is not property but information. The answer seems to be that it should not. In the first place, some information, such as a trade secret, is regarded as property,76 so that a straightforward application of the rule that the beneficiary takes priority over the donee covers the case. But in some instances the information may not qualify as a trade secret, yet here too the equities apply between the parties.77 Consider the example where an insider in Texas Gulf Sulphur discloses information about the copper strike to a friend down on his luck. The disclosure is not quid pro quo; it is not in payment for some antecedent debt; it is not an effort to curry new business. It is just a gift, for old times’ sake, of information that both parties know should be used solely for the advantage of the corporation. How could that not be an improper form of trading on inside information, especially if both sides keep the transaction secret from all corporate officials and do not share the information more widely? At this point, any general rule that exonerates the donee or the insider would be a bizarre affront to the traditional duty of loyalty.

Stated as a general proposition then, the rule in Dirks makes no sense. Yet put into its peculiar factual context, the result makes a good deal of sense. As Justice Powell noted in Dirks, the government’s case turned on “extraordinary facts,”78 insofar as the recipient of the information, Ray Dirks, who learned it from former Equity employee Ronald Secrist, made the disclosures in 1973 to expose widespread fraud at Equity Funding of America.79 The information generated was widely shared and discussed among potential analysts, some of whom sold stock on the strength of the rumors before they became public. As is common in these cases, Dirks did not just come up with this information in a void, but relied on Secrist for the key information. In such cases, a better approach carves out an exception to the general rule that disclosures to donees normally constitute a breach of fiduciary duty. The obvious point of distinction is that Dirks did not trade on the information, let alone trade for his own benefit, but used it to expose fraud to great public benefit. In general, the value of any inside information for trading purposes varies inversely with the number of people who obtain that information. Dirks supplied key information from which someone had to benefit and someone had to lose as the shares of Equity Funding fell back to their proper value. All Dirks did was hasten the removal of market error from which he obtained no advantage. Does it make sense that for this conduct he received a criminal prosecution, even one in which the SEC only censured him but did not ask for jail time?80 Nothing in Dirks undermines the general proposition that donees should not be allowed to trade on inside information obtained from an insider in breach of fiduciary duty. But it does point to the necessity of creating principled, if limited, exceptions to cover those donees who supply social benefits, especially when they do not trade on the information they have gathered to their own personal advantage. Someone has to benefit from the sudden disclosure of this potent information, and the law should not care unduly about that party’s identity.

A similar analysis applies to the second explanation that Justice Powell gave for adding in the personal-benefit rule, which gets to the heart of Regulation FD (which still lay sixteen years in the future). Justice Powell observed that some special provision had to be made to protect market letters and other devices used to communicate information to the firm. “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 of the corporation’s stockholders or the public generally.”81 The best way to defend that conclusion is to note the flexible nature of fiduciary duties under the business judgment rule, which should apply here. The analysts are not beneficiaries to whom insiders owe a duty of loyalty. Nor do these releases exhibit any hint of self-dealing that gives rise to a rejection of the ordinary business judgment rule in favor of the stricter fair-value rule, with its stringent procedural and substantive components.82

Under the ordinary business judgment rule, then, it should be perfectly legal for the proper officials within the corporation to instruct their key employees and analysts to share information with various groups, even if that information cannot be, or is not, supplied “simultaneously” to all shareholders. To be sure, any given instance of disclosure might make it impossible to release all the information to the entire public at one time. But why should the business judgment rule preclude the directors and officers of the corporation from authorizing those selective disclosures? One good reason for allowing them is that the partial release of information may spur interest in the stock, which could on average lead to an increase in share prices overall—which behind a veil of ignorance is a development that current shareholders should welcome. It is a far cry from trading shares in O’Hagan on the strength of confidential information in competition with the client that supplied it. Nor is it correct in these cases to analyze any one meeting with, or disclosure from, an insider in isolation. It thus makes perfectly good sense for a firm to entertain one group of analysts on one occasion, and a second one later on, or to have different representatives of any given firm meet with different analysts. It also makes sense for other firms to engage in similar practices with their own preferred clientele. There is with all forms of information a tradeoff between the slower but more even distribution of information and a more rapid and asymmetrical release. Exactly how public calls and private meetings should be coordinated is hard to say in the abstract. But that is exactly the reason why Regulation FD goes too far, as discussed further in Part III. It assumes that a single paradigm should apply to all firms in all settings, without any concrete knowledge of their distinct circumstances. To look only at ex post parity of recipients is to ignore all dynamic features of the market, including those that result in more rapid, accurate repricing of financial assets.

The result of the overall analysis should now be clear. Historically, the law on insider trading rested on the assumption that any insider who trades on material nonpublic information has acted in breach of his fiduciary duty to the corporation and its shareholders. But that conclusion rests on a command from the SEC, and not only on any duty that the corporate insiders have assumed. The statement therefore is overbroad unless and until it is made clear that the corporation has imposed, as it may well do, such duties that limit how insiders may use nonpublic information. In contrast, the misappropriation theory is the later comer to the law, having received the Supreme Court’s blessing only in O’Hagan in 1997. Yet here the pedigree for liability is far stronger insofar as there is never any doubt that an employee who uses confidential information to trade either against the firm or its clients is in breach of explicit and extensive contractual duties, all of which are intended to protect the firm’s trade secrets. But the irony is this: the public at large has nothing to do with the misappropriation theory. The losses there are solely private, such that the first line of defense against breach is private as well. It is certainly appropriate in this case to consider criminal liability for employees that act in breach of their duties. But the source of concern is how theft of trade secrets undermines efforts of firms to collect information about potential market moves. Ironically, trading conducted in violation of fiduciary duties helps improve share prices to the public at large, albeit at too great a social cost.

III. a critique of regulation fd

At this point, the correct inquiry is whether Regulation FD could survive examination as a matter of first principle. I put aside here the long dispute over whether the SEC is entitled to deference in setting rules, in either criminal or civil proceedings,83 in order to show why Regulation FD is at war with the basic assumptions of the statute it is said to interpret. The first point is that Regulation FD flies in the face of Dirks, which stated the exact opposite conclusion with respect to communications between analysts and insiders. The SEC is well aware of this point because, as it states in Regulation FD, “[t]he regulation now includes an express provision in the text stating that a failure to make a disclosure required solely by Regulation FD will not result in a violation of Rule 10b-5.”84 Yet at the same time it notes that “[i]ssuer selective disclosure bears a close resemblance in this regard to ordinary ‘tipping’ and insider trading.”85

Second, the broad reach of Regulation FD rests on a dubious statutory balancing act. Indeed, it is far from clear where the SEC’s authority to issue this regulation comes from, given that the SEC gives no explanation for its express reliance on multiple sections of various statutes, nor does it hint at which provisions specifically cover this rule.86 It is not possible to extract Regulation FD from Section 10(b), given that the section’s focus is on “any manipulative or deceptive device or contrivance.”87 The SEC has substantial regulatory authority in figuring out how to deal with these fraud-related risks. One example of such deceptive devices is a wash trade, where parties stage fake transactions, in which no risk is created or shifted, to deceive other individuals about the market price of the traded security or the level of liquidity in the market. The former happens, for example, by a public trade at a high price, and a secret repurchase of the shares for the same price shortly thereafter. The latter takes place when a party buys and sells the same shares under different names in order to create a false impression of high market liquidity.88 But the SEC does not have the same authority to take practices that are not deceptive and manipulative and treat them as though they were. In particular, most firms would most often want their key employees to speak to analysts in ways that Regulation FD prohibits.89

The defense of Regulation FD therefore must derive from the view that imperfections in security markets are everywhere, so that any deviation from the model of equality will necessarily work some kind of systematic fraud. One key SEC argument in Regulation FD reads:

We believe that the practice of selective disclosure leads to a loss of investor confidence in the integrity of our capital markets. Investors who see a security’s price change dramatically and only later are given access to the information responsible for that move rightly question whether they are on a level playing field with market insiders.90

And further:

The vast majority of these commenters consisted of individual investors, who urged—almost uniformly—that we adopt Regulation FD. Individual investors expressed frustration with the practice of selective disclosure, believing that it places them at a severe disadvantage in the market.91

On this view, Regulation FD is necessary to maintain confidence in the securities markets. It is this view that drove the government’s all-out prosecution in Newman. As the government warned:“The consequences for investor confidence are plain: individuals will perceive that cozy relationships between insiders and the most sophisticated traders allow exploitation of nonpublic information for personal gain.”92

These reasons repudiate one sound warning against level playing fields in Dirks, although the conflict between the rule and the earlier case is never made explicit.93 Indeed, the rule appears to be inconsistent with Chiarella. Chiarella rejected an earlier Second Circuit decision that had contended that ‘‘the federal securities laws have created a system providing equal access to information necessary for reasoned and intelligent investment decisions because [material nonpublic] information gives certain buyers or sellers an unfair advantage over less informed buyers and sellers.’’94 Chiarella further held that “not every instance of financial unfairness constitutes fraudulent activity under § 10(b).”95

This last point is consistent with the view that no dynamic market is perfectly competitive. Indeed, innovation depends on astute individuals finding ways to take advantage of gaps in markets, and it is through their effort to obtain extra returns that the system starts to hum. There are always entrepreneurial individuals who invest resources in an effort to locate new bits of information that will give them a leg up, which translates into higher rates of return for greater amounts of work. The more people who seek to exploit this information, the better markets will work. As stated in the Newman amicus brief authored by Professors Stephen Bainbridge, Todd Henderson, and Jonathan Macey, the information these entrepreneurs “obtain and pass on to their clients enables more accurate pricing in capital markets and helps to assure that capital will ultimately be allocated to the highest value users.”96 That flexibility could prove especially important to smallcap and midcap firms, which, while publicly traded, are not normally followed by a cadre of analysts. The prospect of some type of exclusive arrangement might increase analyst interest in following these firms’ stock. On balance, more information in a world where some firms—through their analysts—have the inside track may well prove better than the alternative world where no investor has an incentive to follow these firms’ stocks at all. Once again, the question here is hard to answer in the abstract. But there seems little doubt that the right answer might well differ from case to case. If something like Regulation FD works to stimulate interest in their firm, firms will adopt it voluntarily. If it does not, they will tend to employ other strategies. So long as one-size-need-not-fit-all, Regulation FD does not have a sensible role to play.

The SEC tries to additionally defend Regulation FD by stating that the rule guards against conflicts of interest that may otherwise encourage “analysts [to] predominantly issue ‘buy’ recommendations on covered issuers, because they fear losing their access to selectively disclosed information.”97 Under Regulation FD, the standard practice is to open all calls from management to anyone who wants to listen in. People may listen without speaking and management may at any time decline to answer any question. At this point, people can make their own decisions and recommendations without fearing that they will be cut out entirely from all information about the firm. But by the same token, it should be clear that all questioners will be more guarded in their questions, knowing that they might publicly reveal some of the firm’s private information about either itself or the industry. It therefore could make perfectly good sense to have some candid discussions in private in addition to those which take place in public. Once again, the variety of situations makes it highly unlikely that a one-size-fits-all approach works for all different parts of the complex securities market.

Nor should Regulation FD be justified as an independent backstop to Rule 10b-5. In Regulation FD, the SEC treats the two provisions as complements.98 But Regulation FD is better understood as Rule 10b-5’s antithesis. As Professor Aldave wrote: “The Chiarella-Dirks emphasis on fiduciary duties reflects the Court’s determination that the meaning of ‘fraud’ in Rule 10b-5 is essentially the same as the meaning of ‘fraud’ at common law.”99 Regulation FD cuts in the opposite direction because it does not require proof that one person made deliberate false statements to another that were relied on to the second party’s detriment.100

It is also worth noting the positive consequences for the scope of litigation if Regulation FD were overturned, leaving the classical theory of insider trading intact. The directors and officers of a firm could explicitly authorize the selective release of firm information so that when designated insiders act in accordance with that authorization, any claim for securities fraud against them or their firm would basically be over. These types of lawsuits would disappear and the resulting greater clarity of the law should help to improve information flows in capital markets. Regulation FD therefore tends to push in the wrong direction by increasing government oversight over securities markets in areas where a light hand would better serve the public interest.

Not only is Regulation FD antithetical to Rule 10b-5, it also gives an unduly broad reading to the term “fair” in legal discourse. The basic ambiguity in the use of the term is as follows. In the common-law context, in contrast to Regulation FD, the notion of fairness was clearly tethered to traditional theories of liability that involved the use of either force or fraud. Thus, the tort of unfair competition was tied to the use of either force or disparagement in order to prevent current or future customers from trading with the plaintiff. The paradigmatic cases were as follows. The first involves one schoolmaster shooting at the students of a rival school in order to drive them away from their current teacher. There was no use of force against the rival schoolmaster, but in Keeble v. Hickeringill,101 Judge Holt allowed the action for interference (by force) of advantageous relationships, even in the absence of contract, and that position was followed in Tarleton v. M’Gawley.102 The same limitations were also recognized in A.L.A. Schechter Poultry Corp. v. United States,103 which struck down, at least for the moment, the New Deal’s competition codes, when Chief Justice Hughes reverted to the common definition of unfair competition:

“Unfair competition,” as known to the common law, is a limited concept. Primarily, and strictly, it relates to the palming off of one’s goods as those of a rival trader. In recent years, its scope has been extended. It has been held to apply to misappropriation as well as misrepresentation, to the selling of another’s goods as one’s own—to misappropriation of what equitably belongs to a competitor. Unfairness in competition has been predicated of acts which lie outside the ordinary course of business and are tainted by fraud or coercion or conduct otherwise prohibited by law. But it is evident that in its widest range, “unfair competition,” as it has been understood in the law, does not reach the objectives of the codes which are authorized by the National Industrial Recovery Act.104

Schechter, of course, did not last. The codes of fair competition that it rejected under the National Industrial Recovery Act105 quickly took hold in other progressive, New Deal legislation. The new list included “unfair labor practices” under the National Labor Relations Act,106 wage and hours violations under the Fair Labor Standards Act of 1938,107 and various forms of discrimination that ran afoul of the Fair Housing Act of 1968.108 The meaning of the term “unfairness” in these progressive statutes is at complete loggerheads with its common-law meaning. No one is concerned with the prohibition on the private use of force and fraud in any of these cases. In each of these cases, there is not a question of means, but rather a perceived end-state that counts as fair or just, and it is the duty of the government to implement that new goal through a comprehensive global policy that uses state coercion and subsidies in endless permutations. No longer is there an effort to remove obstacles to efficient voluntary markets. Rather, the goal is to displace those voluntary markets to achieve some distributional outcome, which always requires an enormous expansion of the notion of unlawful conduct.

The text of the Exchange Act does not read like these other statutes. It reads like an antifraud statute, which is keen to cover not only obvious forms of lying but also subtler practices that could achieve the same end.109 Thus, it is a clear securities law violation to engage in “channel stuffing”—sales in one period that are recorded as income and are subject to an unstated obligation to repurchase the items sold in the next period.110 The private transaction is an artifice that is intended to mislead people as to the underlying activity of the firm by ignoring the unstated liabilities that should be properly recorded on the balance sheet. Nothing of that sort is involved in the selective disclosure of information to some analysts but not others. The insiders owe fiduciary duties of equal treatment only to their shareholders, not to their analysts. So long as the directors and officers who make selective disclosures as a matter of practice give notice to the rest of the world of that practice, any notion of concealment is eliminated from the case.

At this point in the analysis, it becomes clear that Regulation FD, along with other aggressive enforcement of the securities laws, marks a major departure from the proper objectives of antifraud regulation. It is equally clear that the newer system of securities regulation is a change for the worse. The costs of regulation are heaviest on smaller firms, whose entry into the public marketplace is retarded by the full range of securities regulation over every aspect of the business. The regulation of market transactions is generally a negative-sum proposition, even before the steep administrative costs of enforcing Regulation FD are taken into account. It is imperative to condemn this shift of emphasis from controlling fraud to mandating disclosure in the broadest possible terms. Locally, Regulation FD illustrates the complications that come from implementing the SEC’s new imperatives. Globally, it illustrates the train of abuses that follow from the aggressive implementation of the progressive definitions of “fair” and “unfair” behavior. It is vital that we forget neither.

This critique of Regulation FD helps explain the proper mix between government regulation and private contract. There is no reason why the rules of the game have to be set by the SEC for all corporations on the familiar, if dangerous, one-size-fits-all model. It is quite sufficient that firms can issue, with appropriate advance notice, a general disclosure that indicates their pattern of business, as per the basic argument developed earlier.111 Indeed, it is quite clear that Justice Powell’s simple explanation no longer represents current policy, now that the SEC has prohibited the practice of “selective disclosure.”112

It is important, therefore, to stress that cases of asymmetrical information need not involve some form of financial unfairness. The party who gets the extra information has often put in greater effort to acquire it. And the parties who lack information have the opportunity and motivation to acquire it as quickly as possible. Indeed, in many cases the optimal strategy for the small investor is to ally himself with some large public firm by buying shares in a mutual fund that has the resources to thrive in dynamic markets with asymmetrical information.

To be sure, there are powerful instincts today on behalf of protecting the small investor who chooses to trade on his own account. The efficiency losses of that protectionist strategy seem clear, so it is fair to ask exactly from where the benefits come. In this sense, there is no instinct to protect poor or ignorant people, because few individuals of either type are active as individual players in the securities market. Rather, the more modest objective is for the SEC to protect that small sliver of individuals who wish to manage their own portfolios with complex trading strategies that often do not work well at all.113 But the SEC is the wrong institution to attack this problem, for financial education on such matters as index funds and portfolio diversification is better provided for by private firms operating independent of the SEC.

Whatever the sentiment for this view, this rationale should be resisted for the same reason that we should resist imitating the worst features of the Robinson-Patman Act, whose major mission was to protect small businesses that were losing market share to the more efficient chain stores.114 These distributional objectives are murky at best. In general, open entry can preserve competitive pricing. Accordingly, it is a mistake to try to redesign the Indianapolis speedway to accommodate go-karts, which is what the parity principle tries to do. The better strategy is to let the go-karts be hopelessly outclassed, after which the savvy small investor places his money in a real racecar. Unwisely, the SEC took the opposite tack, which was to slow down the entire process by harping on the supposedly favorable distributional consequences that come from sacrificing the efficiency gains obtainable from the freer flow of information.

IV. back to newman: of personal benefits and information flows

The previous analysis now makes it possible to revisit Newman and examine how in principle it should deal with both the personal-benefit and information-flow issues that form the core of the case.

A. Personal Benefit

In Newman, the personal-benefit rule operated as an essential cog of the basic legal framework. That role is not entirely unwelcome in Dirks’s second-best universe in which the personal-benefit rule functions as an oblique check on excessive SEC power by creating a zone of legality that should be routinely allowed under the business judgment rule. Most analyst disclosures are not made for cash or other equivalents, so they fall outside the scope of SEC regulation. But doctrinal inaccuracy does exact an intellectual toll by forcing a new inquiry into just how tangible or fixed a personal benefit has to be to meet the requirements of this new rule.

In this context, cash or benefits in kind will count as personal benefits. So too will quid pro quo introductions to potential clients or business partners, or easing the path toward regulatory approval. But in Newman,the SEC insisted that a diffuse set of social interactions was able to fill the hole created by the personal-benefit rule.115 These ostensible benefits include social friendship, cooperation as members of the same church or club, career advice, and examination preparation. If the personal-benefit rule means anything, the Second Circuit concluded, these soft benefits cannot count because they are virtually always present in a clubby industry that relies on high levels of informal interaction.116 In support of that position is the proposition that forms of mutual and reciprocal assistance are part of a social network among analysts wholly apart from the application of securities law, so that the SEC’s expansive view of personal benefits is a far cry from the specific benefit made as part of a quid pro quo. Judge Parker thus had a point when he said that if any of these facts established the requisite benefit, then “practically anything would qualify.”117

To which the answer is, perhaps, not so fast. In its forceful brief in the petition for rehearing, the government took the view that these social interactions were far out of the general social norm. In its view,

Ray [the insider at Dell] “desperately” wanted to be an analyst—a more lucrative job than his job at Dell—and looked to Goyal [one of the tippees] for career advice and help in securing such a position. To maintain the stream of valuable inside information from Ray, Goyal spoke with Ray more often and longer than he otherwise would have, typically at night and on weekends.118

The clear implication is that this particular friendship went beyond simple reciprocity and thus counted as an implied quid pro quo: you get us information, and I will help you get a better job. But even here the inferences are difficult to draw because in fact no job offer came out of the arrangements, and of course, the defendants offered a very different interpretation of the evidence:

The government never proved that Ray provided Goyal with material nonpublic information about Dell in order to get career advice. The prosecutors’ decision not to call Ray as a witness spoke volumes: Ray had proffered that he never connected Goyal’s career advice to the Dell information in his own mind, and Goyal’s advice “did not influence the manner in which [Ray] performed his duties at Dell.” Goyal testified that he gave Ray career advice for “one, one and a half years” before Ray started providing any information about Dell. And Goyal confirmed that Ray did not once link the Dell information to Goyal’s career advice in their conversations.119

It is clear that appellate courts are not in a position to resolve these sorts of detailed factual conflicts. But they are supposed to decide whether the government has presented enough evidence to command a retrial of the issue of personal benefit, which was not raised in the court below. That is a difficult call. But if the case had been properly pleaded in the first place, it is again an open question of what inferences should be drawn. In a civil case there might be enough to go to a jury, but in a criminal case the matter surely is a lot closer.

Yet the larger question remains: who cares? Why sift through all the fine nuances of this dispute on a question that in principle should have no relevance to the outcome of the case? In principle, the Supreme Court should overrule the personal-benefit prong of the insider trading offense. In practice, however, it should take this step only if it substitutes in its place the more nuanced account of fiduciary duty. In turn, that objective can be achieved only if the Court rejects the SEC’s selective disclosure prohibition by shielding authorized disclosures to the analysts following the stock under the business judgment rule, which has sadly fallen by the wayside in all these cases. It is to that issue that I now turn.

B. The Information Transfer to the Defendants

The court in Newman noted that“the Supreme Court held that a tippee may be found liable ‘only when the insider has breached his fiduciary duty . . . and the tippee knows or should know that there has been a breach.’’’120 The court did address the information transfer that disclosed “those companies’ earnings numbers before they were publicly released” that started with the insiders and made its way to the two defendants, and from them to their traders.121

In dealing with this last issue, it is important to look at the transaction from the vantage point of both the transferor and transferee. Starting with the transferor side, it bears restatement that under the rule stated in Dirks, the authorized and selective release of this information should not count as a breach of fiduciary duty to shareholders, given that it was consistent with improving the overall position of the firm. At this point, the information should be treated as part of the public domain, which means that its use can no longer trigger any potential liability on the part of any downstream parties who incorporate that information into their own decision making. That bright-line rule clears the air and allows for the rapid dissemination of the relevant information. Prosecutorial discretion is kept to a minimum.

Within this alternative conceptual framework, the next question is whether the disclosure of that information was authorized, and on this point, the parties in Newman again clashed. The government took the position that Dell and NVIDIA each had formal policies that prohibited the disclosure of the information in question.122 If that were all there were to the issue, then it would be necessary to chase down the ebb and flow of the information once it left the hands of the insiders to see how it influenced the behavior of Newman and Chiasson. But there is more to the case than this simple scenario suggests, because the defendants claimed that there were systematic deviations between the official and day-to-day policies:

Dell and NVIDIA routinely leaked this information to analysts. The evidence of leaks is significant because it shows that insiders provided this type of information without any personal benefits—not even the government argues that the leaks were motivated by self-dealing. More importantly, it shows that Newman would have had no basis to believe that the information he received was fraudulently disclosed.123

The hard question is why the gap between official and actual policy? The answer seems to be that it would be suicidal for any corporation to adopt an explicit policy that allowed for informal contacts with analysts, especially after Regulation FD went on the books. Yet the constant interaction with the analyst community is so important that it now takes place informally, in an intellectual gray market. The uncertain legal status of these interchanges then renders the whole matter ripe for controversy in litigation. But the baleful consequences of forcing firms to adopt these shadowy disclosure practices extend beyond that. Covert practices are clumsy practices, breeding unnecessary inequity and confusion of their own. It is impossible to set out precise guidelines about how and when the dissemination of information should take place, lest those actions count as an open admission of illegal conduct. The consequence is that the program is done less well than it ought to be if the entire matter were left to the private choice of the company’s officers and directors.

Thus, as a matter of first principle, the best strategy for the law is to avoid both the personal-benefit issue and the knowledge and information question by making it clear that corporations may, through their directors and officers, allow firm employees to engage in the selective release of information to the firm’s general client list, and the problem goes away.

The current legal doctrine blocks that approach. Unfortunately, the SEC’s prohibition on selective disclosure makes any organized release of information improper. Therefore, under the current approach, it becomes critical to ask the extent to which the information is material and nonpublic. It should be evident that the dispute in Newman is miles removed from any hypothetical case where one insider gives valuable information to some preferred donee, even out of the goodness of his heart. In that setting, the information is surely nonpublic and is likely to be material as well, given that no one else shares it. But the situation is different here. The government’s allegations, as summarized by the Second Circuit, were that “a cohort of analysts at various hedge funds obtained material, nonpublic information from employees of publicly traded technology companies, shared it amongst each other, and subsequently passed this information to the portfolio managers at their respective companies.”124

This brief passage scarcely does justice to the tangled web of interactions that took place among the various parties over the several years covered by the investigation. A “cohort” implies a large number of parties who worked at multiple firms, not just these two individuals. But just how many is unclear. In this scenario, there is no hard-edged line between public and nonpublic information. Indeed, it might be a stretch to say that this information is “nonpublic” if a largish number of professionals were able to put it to quick use, so that the market fully took it into account in setting the share price. The situation was certainly leagues away from Texas Gulf Sulphur,where the group of insiders who traded on inside information was tiny relative to the enormous increase in value that came with one key fact—the discovery of major new finds of copper.

There is also a question of whether this information counts as “material.” The value of information varies inversely with the number of people who share it, and the rapidity with which that information is factored into overall market valuations. Both of these points suggest that the information in question, coming as it did in uncertain dribs and dabs, could not have been critical in shaping overall price movements. That conclusion is only fortified because it is equally clear that all of the analysts and traders had acquired large amounts of information from other sources about the financial condition of Dell and NVIDIA, and thus independent of anything contained in the information supplied to these analysts. The abundance of other information further diluted the significance of the information that had been transferred to this cohort of analysts.

The point is important because it shows how difficult it is to make judgments in isolation about what kinds of information are “material.” Everything depends on context. If information about the earnings reports was the only information available, it might be able to move stock prices substantially. But in this case its likely effect was small. Indeed, in addressing this instance, the Second Circuit noted that it had far less than a one percent influence on all the relevant numbers.125

Needless to say, the case is even more complex than this because the defendants point out, fairly it seems, that some of the information was qualitative, and some of it was unreliable, if not downright erroneous, a fact that was known (even if only imperfectly) by the parties who received it. So in its petition for rehearing, the government offers this version of the facts: “Newman and Chiasson made $4 million and $68 million in profits for their respective funds by trading on secret earnings numbers that they encouraged their analysts to collect from Dell, Inc. and NVIDIA Corporation over multiple successive quarters.”126 The government’s view of causation seems to assume a push-pull connection between the receipt of information and gains from trade. That connection in turn makes conviction inexorable, even if the underlying reality suggests a complex network of information rivulets that combine, divide, and combine again.

In litigation, the role of inside information takes on a different appearance when looked at from the side of the transferee. Did the various defendants know that the information in question was properly released, or did they think that it had all come from a tainted source? The Second Circuit was emphatic in its insistence that the swirling mass of information came from so many sources that the defendants did not know or have any reason to suspect that it was tainted at the point of its release.127 It was very clear that the parties that received this information did not segregate it out from information they received from other sources, if only because it is not possible during the course of quick conversations to verify the pedigree of each separate bit of information that is included in the analysis. Indeed, each new layer of parties introduces an added layer of complexity. So why go through this exhaustive trial?

The uneasiness with the government’s condemnation of inside information is that it does not seek to disaggregate any individual transfer of information from the larger whole. It is quite possible that just one, or very few, communications were prearranged from start to finish, while others were not, and this one episode from many did generate substantial benefits. After all, the parties in all the chains were familiar with each other and had worked together on multiple occasions.

To get evidence on particular transactions requires a massive inquiry. But why should the government invest so much in this problem if the aggregate impact is likely to be modest relative to the kind of serious abuses that take place in cases like O’Hagan? From any sensible point of view, the information that was released should no longer be regarded as either nonpublic or material. It was just one piece in a large mosaic. As with all information mosaics, there is always a causation question of which bits of information obtained from what sources influenced any decision to trade. It therefore does seem plausible that the government could not carry its general burden of proof that the defendants in this criminal trial had sufficient mens rea on the information in question.

So what should have been done procedurally, under the current framework, if the knowledge issue had not been properly resolved in the original trial? There are conflicting impulses, and the final judgment is not free from doubt. On the one hand, if the record contains a jumble of facts, then the questions of transfer of information (and personal benefit) generally resist summary judgment. The government’s effort to resist final disposition on appeal is indeed fortified by the settled rule in the Second Circuit, which states: “Although sufficiency review is de novo, we will uphold the judgments of conviction if ‘any rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt.’”128 On that standard, it is hard to explain why the jury could not in some instances draw an inference of knowledge from the tightness with which the information was passed. It should, one could argue, be the province of the jury to decide whether any given communication and trade had the requisite effect, at which point a remand, not the Second Circuit’s dismissal with prejudice, again seems to be the proper result.

But once again, the prospect of retrial would have carried unwelcome complications. This is a criminal case, for which the proper approach might be to call for a higher threshold before the particular issue gets to the jury, at which point the mass of evidence cutting against the government’s position could matter in the outcome of the case. The analogous issue has come up before. In Anderson v. Liberty Lobby, Inc.,129 the plaintiff, a public figure, sued the defendant for defamation, under the applicable rule from New York Times Co. v. Sullivan,130 which required that the plaintiff establish malice by clear and convincing evidence.131The Court held in Anderson that when considering whether to grant summary judgment, “a trial judge must bear in mind the actual quantum and quality of proof necessary to support liability under New York Times.132The standard of proof in a criminal case is still higher, and that should be reflected in the overall calculus as well, which again counsels in favor of a higher standard of review for criminal convictions. The case is quite different from a civil action demanding forfeiture of illicit gains, without jail time or penalty. So again, within the existing framework of the law, the case is, sadly, a toss-up.

The saga of Newman has now come to an end, as the U.S. Supreme Court denied the government’s petition for certiorari on October 5, 2015,133 as well it should have. In its brief, the United States spent most of its firepower attacking the relaxation of the personal-benefit standard in Newman, and relatively little dealing with the information point. Two sentences from the government’s briefing show the equivocation. First, the petition for certiorari asserted, “The Second Circuit also stated that ‘the Government presented absolutely no testimony or any other evidence’ that respondents knew, or consciously avoided knowing, that they were trading on information in exchange for which the insiders ‘received any benefit.’”134 But whether respondents knew that the insiders obtained a personal benefit is likewise bound up with the legal question of what constitutes a personal benefit in the first place. Second, in its reply brief, the government acknowledged in a backhanded fashion that the grant of certiorari on the knowledge question would not change the outcome of the case.135 Finally, in his remarks at New York University Law School, Mr. Bharara spoke only of the personal-benefit prong of the case and ignored the issue of knowledge.136 There is a real cost in taking this limited view, for it gives Newman greater significance than it deserves. Just recently, Mr. Bharara announced that he was dropping several other prosecutions.137 In my view, he has a good chance of getting the personal-benefit prong of Newman reversed in a case that presents clean evidence that the recipients knew they had received inside information. His correct strategy therefore should be to minimize the importance of Newman by noting that most other prosecutions do not present fatal weaknesses on that issue. This point gains strength by looking at the response to Newman and the issues the personal-benefit prong raises in other cases.

V. newman and the misappropriation theory in other cases

The conceptual difficulties that complicated the analysis in Newman are also evident in other cases that deal with this issue, to which some brief attention should be given.138 In this regard, I start with two cases that should be easy wins for the government, and then turn to a third case that presents more difficult challenges.

The first case is United States v. McGee,139which turned exclusively on the personal-benefit prong of the insider-trading test. Sometime between 1999 and 2001, Timothy McGeebefriended Christopher Maguire, who was an insider at the Philadelphia Consolidated Holding Corporation (PHLY), at sessions of Alcoholics Anonymous, and became Maguire’s informal mentor. Years later, the two met again when Maguire mentioned to McGee that a looming sale of PHLY had led to his relapse. “McGee borrowed approximately $226,000 at 6.875% interest to partially finance the purchase of 10,750 PHLY shares. Shortly after the public announcement of PHLY’s sale, McGee sold his shares, resulting in a $292,128 profit.”140 Thereafter, he sought to escape conviction by insisting that he did not have a tight enough relationship with Maguire to satisfy the personal-benefit prong of the test.141 The Third Circuit gave the right answer to, as it were, the wrong question when it stated that their past relationships were sufficiently close.142 The case is a perfect example of why the personal-benefit test is irrelevant. The only point that matters is whether the firm authorized the release of the information for its own benefit, which it manifestly did not. Since McGee knew the exact state of affairs, his trades were patently illegal and the criminal conviction was justified wholly without regard to the details of his past relationships.

The second of the easy post-Newman cases is United States v. Salman,143 which unfolded as follows. Maher Kara, a new member of Citibank’s health care group, leaked information concerning the activities of companies that worked in cancer and pain management to his brother, Mounir “Michael” Kara. Michael in turn shared that information with his future brother-in-law, Bassam Salman, who traded on the information and shared the profits with Michael.144 The case is an easy one for conviction. The release of the information was unauthorized, was known to be unauthorized,145 and supplied no benefit whatsoever to Citibank. The situation is far removed from the general release of information to analysts in the ordinary course of business in Newman. The defense that Salman offered was that his connection to Maher and Michael was not close enough to satisfy the personal-benefit prong in Newman.146 Jed Rakoff, a Senior District Judge within the Second Circuit, sitting by designation, took the occasion to say that if Newman required more beyond “a gift of confidential information to a trading relative or a friend,”147 “we decline to follow it.”148

And so he should, especially since the entire personal-benefit prong of Dirks is a mistake in the first place. The key line of distinction between the two cases is that Salman was in cahoots with Maher and Michael in the collection and unauthorized use of stolen information, which makes the case an easy criminal conviction. The personal-benefit prong is a distraction when the illegal collection and sharing of information is undisputed.

The analysis is a good deal more difficult in SEC v. Cuban.149Mark Cuban, the colorful owner of the Dallas Mavericks, had received an invitation from the CEO of to participate in a new round of funding for the company in which Cuban was already a substantial minority shareholder.150 Cuban strongly opposed that financial plan, and promptly sold all his shares on receiving that information, thereby avoiding some $750,000 in losses when the stock tanked after the refinancing plan became public.151 The factual dispute in question was over the conditions that the CEO attached to the sharing of the information with Cuban. The government’s position was that Cuban agreed not to disclose the plan to anyone else and not to sell his shares.152 The Fifth Circuit held that the misappropriation theory applied if he had agreed not to sell his shares.153 On that factual dispute, Cuban prevailed when the case was retried.154

The entire matter is ticklish. It makes perfectly good sense for Cuban to agree to be silent about the potential offering when he receives the information. But it is quite another to ask him to retain the shares after he formed the independent judgment that the proposed financing plan would be a disaster. It therefore is highly unlikely that anyone in his position would agree not to trade on that information. But by the same token, Cuban would have suffered serious losses if he had never received the information before the plan went public. On this view, it may be possible that Cuban would have desired the information so much that he would have made that fatal concession. But if he did, then he would have been worse off than if he had not heard anything from the CEO because he could have sold the stocks in ignorance of the entire transaction.

The case shows just how difficult it is to deal with these matters on an ad hoc basis. The correct response is not to find out what is expected in these cases through a criminal trial. Instead, it is for all companies to articulate a policy in which they decide in advance how insiders should be able to respond to this information, which is very hard to do, to say the least. The difficulty here is in a sense unavoidable. The second prong of Rule 10b5-2(b) points to reasonable expectations as the touchstone of potential liability in those cases where there is no explicit agreement to use the information only for limited purposes. As is always the case, the test works perfectly well in easy cases like McGee,but fitfully at best in cases like Cuban. In the end, it probably matters that Cuban is a criminal prosecution and not a civil case. In line with general principles, the benefit of the doubt should go to the criminal defendant and not to the State.


This Feature sought to reexamine the legal principles governing insider trading in light of the recent Second Circuit decision in United States v. Newman. The emergent picture is complex. The difficulties start with the inability to identify any persuasive rationale for the insider trading prohibition in the first place. With respect to classical insider trading, a corporation should be able to define its own policy on how insiders trade and use that to guide how investors should respond in the marketplace. With respect to misappropriation cases, the wrongs in question are directed toward the corporation that supplied the information and not the public at large. Thus the risks of misappropriation are best met by explicit contractual principles that limit the use of the information by the recipient and his ability to share it with other individuals. In general, contractual restrictions should be sufficient to deal with these instances, which leaves only a small place for the securities law to impose additional sanctions in an area that is best left to private ordering.

The current situation, of course, allows for criminal prosecutions for trading on inside information, and the above analysis offers guidelines as to how that should be done—if it is to be done at all. The central takeaway is that the sole violation that matters is the deliberate use or sharing of information contrary to the wish of the firm that has supplied it in the first place. These unauthorized uses should impose liability on the immediate recipient and any person who takes with knowledge of the illegal release. That prohibition should apply under a constructive trust theory, whether or not the recipient is deemed to have in fact some relationship of trust and confidence with the corporation, and it should apply wholly without regard to whether the party who leaked the information received some return benefit, tangible or intangible. Following these simple principles should vastly improve the overall operation of the securities law, which is now in a sad state of intellectual and administrative disarray.