Professor of Law
Columbia Law School
September 26, 2006
Professor Coffee's testimony addresses the following principal topics:
1. Has there been a recent increase in insider trading? No generally recognized measure exists, but the press report that the number of referrals to the SEC by the market surveillance unit of NYSE Regulation Inc. rose significantly in 2006 is certainly plausible evidence of some increase.
2. What could be causing any increase in suspicious trading? Insider trading is usually linked to merger and acquisition activity, but "M&A" levels are down in 2006 in terms of the number of deals (but up in terms of the dollar value). Other factors might loom larger in causing any increase, and these would include:
(1) the rise of the hedge funds;
(2) the increase in multi-player friendly buyouts (which make confidentiality restrictions hard to enforce); and
(3) the growth of global trading and the availability of foreign havens.
3. Who is injured by insider trading? Everyone. It is not just shareholders or the corporation (either the acquirer or the target). As informed traders increase their trading upon asymmetric information, bid/asked spreads are likely to widen on all stocks (thus increasing the cost to investors to trade). Ultimately, insider trading causes the cost of equity capital to rise, and this in turn has a macro-economic effect on GNP, employment, and the economy as a whole. As a result, it is not just investors who suffer, but all who are impacted by (marginally) reduced economic growth.
4. What works to curb insider trading? Worldwide, there is evidence that criminal prosecutions do deter insider trading (and reduce the cost of equity capital). The problem for the United States is that it has reached the point of diminishing returns in its ability to deter by increasing authorized penalty levels. Since the time of Cesare Beccaria in the late 18th Century, most criminologists have agreed that the likelihood of apprehension is a more important variable than the severity of the sanction in the deterrence equation.
Other preventive measures could discourage insider trading at fairly low cost: (1) a pre-transaction public notification requirement requiring insiders to disclose an "intention to trade" in a specific amount prior to trading; (2) mandatory confidentiality restrictions that the corporation would be required to impose in certain transactions (most importantly, PIPE transactions - "Private Investment in Public Equity"); (3) advance scheduling of stock option awards, etc. These are not, however, within the jurisdiction of this committee, and so they will not be discussed in detail.
5. What, if any, current problems hobble the enforcement of the insider trading prohibition? Here, there is one new development. Enforcement of the insider prohibition in the United States has depended on close cooperation between the SEC, the NYSE, and the Department of Justice. In other jurisdictions, prosecutors have not been successful in prosecuting insider trading (Canada and Great Britain are relevant examples where despite strong regulatory agencies, criminal prosecutions are rare to non-existent). The inability to prosecute insider trading criminally in these jurisdictions seems attributable to restrictions on close cooperation between civil and criminal regulators. Today, close cooperation between civil and criminal enforcers in the United States is threatened by two recent federal district court decisions that chill their ability to cooperate: United States v. Scrushy, 366 F. Supp. 2d 1134 (N.D. Ala. 2005); United States v. Stringer, 408 F. Supp. 2d 1083 (D. Ore. 2006). Both decisions, particularly the former, seem ill-considered.
6. What could Congress do? To the extent that a Court bases its decision on due process or the Constitution generally, Congress is without power to overrule it, even prospectively. But the Scrushy decision was rested instead on the Court's claimed "supervisory authority over the manner in which Federal agents exercise their power." Id. at 1137. Congress can address that, and it can clarify that Federal agents may cooperate more fully than that the Scrushy Court would allow. Indeed, Congress had already authorized cooperation between the SEC and U.S. Attorneys in § 21(d)(1) of the Securities Exchange Act of 1934, but that provision can be read narrowly. In short, cooperation could easily be chilled by these recent decisions, and Congress therefore would be well advised to amend Section 21(d) to permit cooperation and to indicate that advance notice need not be given to a deponent in an SEC investigation that a U.S. Attorney may have an interest in his testimony or may have suggested specific questions.
I want to thank the Committee for inviting me to testify, and I will get directly to the point by breaking my testimony into subheadings.
1. Has there been an increase in insider trading? The "new" conventional wisdom is that insider trading has increased with a recent rise in the level of merger and acquisition activity. Such a correlation is far from new. Ivan Boesky achieved a still unequaled level of infamy for his insider trading on "M and A" transactions in the 1980s. Predictably, insider trading should rise and fall with the level of "M and A" activity, because no other category of information is as clearly material. But it is not clear that "M&A" activity has actually risen, as the number of deals this year is down from the corresponding period last year, even if their dollar volume has increased. Some have suggested that this increase is attributable to less active enforcement by the SEC in insider trading cases. I do not endorse the view (if anyone actually takes it) that the SEC is "soft" or "passive" about insider trading, but the SEC's enforcement staff has multiple priorities: accounting irregularities, market timing in mutual funds, backdating of stock options, unregistered sales of securities, etc. And its enforcement resources are necessarily limited. Even more importantly, the SEC can only enforce the law when it has evidence of a violation, and therein lies the rub. Although individuals tend to make poor inside traders, often leaving clumsy, messy footprints, institutions know how to leave a clueless trail.
The best evidence of an increase in insider trading lies in the higher number of referrals made this year to the SEC by the market surveillance unit of New York Stock Exchange Regulation Inc. Still, an increase in regulatory referrals does not imply regulatory passivity. What then has changed that could explain the apparent increase in suspicious trading?
Three possibilities stand out:
a. The Rise of the Hedge Funds. For better or worse, hedge funds are the principal destabilizing force in corporate governance today. Hedge funds are different than mutual funds in two principal respects: (1) they need not diversify (and many do not), and (2) they can sell short. As a practical matter, mutual funds and pension funds do neither; the former must diversify, and latter are largely indexed. Hence, neither is as prepared to make a large firm-specific investment as a hedge fund. That's part of the story, but there is another racier part: hedge funds are unregulated, and their managers are not monitored as closely by compliance officers and counsel. Hence, hedge fund behavior may often resemble the Wild West (remember Clint Eastwood in "The Good, The Bad, and The Ugly"?). But who leaks to a hedge fund? After all, they are not loved by the business community. Here, the answer may be that because they trade in larger increments than more diversified institutional investors, they will also pay more for useful tips. This point has a further implication: as usual, the most promising prosecutorial strategy is to "follow the money."
b. The increase in friendly, multi-player deals. Once, hostile deals dominated the scene, and then relatively few knew of an approaching hostile takeover (basically, only the bidder, its lawyers, investment bankers and the commercial bank financing it). More recently, "friendly" leveraged buy-outs have predominated. With this shift, the number who are aware of the transaction in advance of its announcement increases exponentially: target management, multiple private equity firms who are cooperating in the bid, creditors who must consent, plus the usual cadre of lawyers, investment bankers, etc. With each additional player, the risk that information will leak goes up significantly.
c. Global Markets. Off-shore trading is probably the one strategy for which the SEC does not have a ready response. If suspicious trading occurs though a U.S.-based brokerage account, the SEC can find out the identity of the trader from the U.S. broker quickly and easily. But if the trading originates from an overseas account with a non-U.S. broker, detection becomes more difficult.
What do these factors imply in combination? Although some insider trading will be the result of individuals tipping friends and co-conspirators, the lion's share should logically be the product of institutional activity, particularly in the case of short-selling and overseas trading. Institutions can trade in larger volume and can establish overseas havens for trading purposes.
2. What Steps Might Reduce the Incidence of Insider Trading? Sarbanes-Oxley has already elevated the penalty levels for securities fraud to 25 years and mail and wire fraud to 20 years. Thus, we have reached the point of diminishing returns on this front. Greater funds could be invested in enforcement, but the payoff is uncertain.
So what policy reforms make sense and are feasible? Basically, greater reliance needs to be placed on stronger internal controls. In the case of the corporate executive, the basic strategy should be to reduce the executive's discretion over the timing of corporate disclosures because the executive can manipulate timing in order to trade profitably. For example, if the executive can effectively pick the date of a stock option grant (as he often can today), the executive can exploit either favorable or unfavorable information (or both) to inflate the value of the grant - that is, by releasing negative information before the award to drive the price down, then releasing positive information afterwards (a practice already known as "springloading"). The result is that the stock option is awarded at the base of sharp, but manipulated, vortex in the stock's trading price. This practice may or may not constitute "insider trading, but it is clearly manipulative and in violation of Rule 10b-5 and should be halted.
The simplest means to restrict such practices would be to require public corporations to issue options only on a scheduled basis - that is, on the same date each year or at least on a date announced several months in advance. Presumably, the date chosen in advance would be during a window period following the filing of the corporation's quarterly report on Form 10-Q. This is already the better corporate practice, but today most options are still awarded on an unscheduled basis, and the empirical evidence suggests that much more manipulation has surrounded the use of unscheduled than scheduled options. The cost to the corporation of such a reform is low, and exceptions could be recognized, permitting the award of unscheduled options to non-insiders (for example, to an incoming employee who would not be one of the corporation's most highly compensated employees).
Similarly, reasonable controls could limit the insider's ability to exploit other forms of material nonpublic information. Executives can often exploit material nonpublic information by trading sufficiently in advance of the public release of that information as to make their trading appear unrelated to that release. Here, the better answer may not be thirty year criminal sentences or enormous SEC civil penalties, but a requirement that the executive disclose to the market and the SEC an "intent to trade" some short period of time before the executive (or his affiliates) buys or sells in any substantial quantity. Professor Jesse Fried was the first to advance such a pre-transaction notification requirement, and it could realistically be limited to trading over some cut-off level (say 1,000 shares). Recently, the Canadian Task Force on Modernizing Securities Regulation (on which this author served) recommended that insiders be required to give the market at least seven days notice of their intention to sell. Although seven days may be somewhat longer than necessary, the impact of such a rule should be obvious: if the chief financial officer announces an intent to sell 25% of his holdings, the entire market could frontrun him. The negative price impact that large insider sales announced in advance would likely cause might render senior executives less liquid, in effect locking them into a longer-term holding. Depending on your perspective, this is not a cost, but an added benefit, because it more closely aligns their interests with those of long-term, "buy and hold" shareholders.
These reforms will not be as effective, however, in reducing insider trading by hedge funds and other actively trading institutions. One could imagine strict prophylactic rules so that a buyout or other private equity fund that participated in a buyout or LBO transaction had to surrender all profits made from trading in the subject corporation's stock for a specified period - say, six months - prior to the transaction. However, such a sweeping proposal may not be needed, because it seems unlikely that the buyout or private equity firms are, themselves, trading in anticipation of merger activity. Rather, the more likely scenario is that information is leaked by their staffs and by investment bankers to hedge funds and other active traders. These leaks are, however, not gratuitous; they are predictably in return for some likely quid pro quo. As hedge funds have come to dominate trading, they are often paying above market brokerage commissions, at least in comparison to other institutional investors. The funds paying these above-market commissions expect many things in return: (1) priority in the allocation of hot IPOs, (2) the privileged access to sell-side security analyst research (i.e., the first look); and (3) hints about pending deals. The recent IPO "laddering" investigation shows how incestuously hedge funds and investment banks can cooperate in activities that were clearly over the line.
How then can we best discourage the tipping of inside information to them? Using our earlier proposed remedy for insiders, one obvious analogy would be pre-trading notification to the SEC. This would not require the same elaborate disclosure as does Schedule 13D under the Williams Act, because there would be no need to disclose the investment purpose or source of funds, but only the intent to make a large trade over a specified cutoff level. Nonetheless, I recognize that these proposals fall within the jurisdiction of other committees, and hence I will not elaborate on them further.
3. Can Congress make law enforcement more effective against insider trading? Congress can, of course, always appropriate more money for securities enforcement, or it could earmark funds specifically for the prosecution of insider trading - but this Committee does not need a law professor to tell it what it already knows. Nor is it always wise to limit funds to the enforcement of insider trading, as most forms of securities fraud have the same effect on investors, and different abuses flourish at different times.
Still, Congress can seek to make securities enforcement more efficient. Since at least the 1980s, the enforcement of insider trading has usually begun with warning signals from the market surveillance units of the NYSE and Nasdaq, which can effectively detect suspicious trading. The problem lies in moving from suspicion to proof.
a. Revising the Mens Rea Level. One possible approach to simplify enforcement (which I do not recommend) would be to lower the mens rea level that the Government must prove in a criminal prosecution. Under both Section 24 of the Securities Act of 1933 (15 U.S.C. § 77x) and Section 32(a) of the Securities Exchange Act of 1934 (15 U.S.C. § 78ff(a)), the Government must show that the defendant "willfully" violated an SEC rule or statutory provision (usually Rule 10b-5 or Rule 14e-3). The Second Circuit construed this willfulness requirement in United States v. Peltz, 433 F.2d 48, 54 (2d Cir. 1970), cert. denied, 401 U.S. 955 (1971), to require a showing of "a realization on the defendant's part that he was doing a wrongful act ... with the qualifications ... that the act be wrongful under the securities laws and that the knowingly wrongful act involve a significant risk of effecting the violation that has occurred." Hence, good faith on the defendant's part is a viable defense to a securities fraud charge. Subsequent cases have liberalized this test slightly by finding that the requisite willfulness can be found if the defendant deliberately closed his eyes to facts that he had a duty to see." United States v. Benjamin, 328 F.2d 854, 863 (2d Cir.), cert. denied, 377 U.S. 953 (1964). Modifying the mens rea level strikes me as fundamentally unfair given the ambiguity that often surrounds insider trading.
b. Preserve Parallel Proceedings. The SEC and the Department of Justice ("DOJ") frequently conduct parallel civil and criminal investigations, and the SEC often shares the fruits of its investigation with the DOJ. See United States v. Kordel, 397 U.S. 1 (1970); SEC v. Dresser Industries, 628 F.2d 1368 (D.C. Cir. 1980); United States v. Field, 592 F.2d 638, 696 (2d Cir. 1978); Bass v. United States, 409 F.2d 179, 180 (5th Cir. 1979). In fact, Section 21(d)(1) of the Securities Exchange Act of 1934 expressly authorizes the SEC to "transmit such evidence as may be available concerning such acts or practices as may constitute a violation of any provision of this title or the rules or regulations thereunder to the Attorney General, who may, in his discretion, institute the necessary criminal proceedings under this title." Congress was aware of and has approved this pattern of frequent information sharing. See Senate Committee Report on Foreign Corrupt Practices Act, S. Rep. No. 114, 95th Congress, 1st Session 12 (1977). In my experience, SEC Attorneys are often delegated to work with U.S. Attorneys on criminal cases, and they sometimes become Acting Assistant U.S. Attorneys for the duration of the criminal investigation.
In my judgment, this ability of the civil authorities to aid and assist the criminal enforcer is critical to effective law enforcement in securities fraud cases. I have spent the last two years as a member of a Canadian Task Force (the Task Force on Modernizing Securities Regulation) that has examined the need for changes in Canadian securities law. In particular, the Task Force focused on the inability of Canadian authorities to prosecute successfully criminal insider trading cases. We found that a major obstacle was that under Canadian law the civil regulators could not assist the criminal prosecutor. At least partially as a result of this problem, criminal insider trading cases are almost unknown in Canada, and much evidence suggests that insider trading is unchecked in the Canadian markets. Great Britain has had a similar experience and has largely abandoned criminal enforcement of insider trading. Unfortunately, nothing deters as well as the threat of jail.
Thus, the ability of the SEC to gather and provide information to the U.S. Attorney may be critical to effective enforcement. But that ability is today in jeopardy as a result of the decisions in United States v. Scrushy, 366 F. Supp. 2d 1134 (N.D. Ala. 2005) and United States v. Stringer, 408 F. Supp. 2d 1083 (D. Ore. 2006). In Scrushy, the SEC was investigating the defendant in a civil proceeding, and the U.S. Attorney's office in the Northern District of Alabama contacted the SEC's attorney and suggested both questions that might be asked by the SEC (and questions that should not be asked because it might warn Mr. Scrushy that a criminal investigation was underway). Also, the U.S. Attorney asked the SEC's attorney to move the deposition to a location within its jurisdiction. In so doing, the Scrushy Court found that the Government had set a "perjury trap" for the defendant and so suppressed, on the defendant's motion, the SEC deposition testimony given by Scrushy (which had resulted in a perjury indictment in the Northern District of Alabama). By forcing the Government to abandon its perjury count, the Court may have enabled Mr. Scrushy to escape conviction (in that case).
Why the Court should consider such cooperation to amount to a "perjury trap" is not clear to me. Presumably, the defendant must have known that if he lied at his sworn deposition before the SEC, he was subject to a criminal prosecution for perjury. No special warning should be necessary that you could be prosecuted if you lie. The potential impact of the Scrushy decision is that the defendant might have to be given a special warning if the U.S. Attorney has an advance interest in his testimony, and this might give some defendants a sense of relative immunity when they received no such warning. Put simply, there should be no "right to lie," and if the deponent must be told whenever the U.S. Attorney has an advance interest in the deponent's testimony, the deponent learns that there is no such interest when there is no such warning. I see no reason to give the deponent such "peace of mind" and thereby increase the deponent's possible willingness to lie.
Other courts have disagreed with the Scrushy decision. See United States v. Moses, 2005 U.S. Dist. LEXIS 40187 (N.D. Ga. August 31, 2005). Nonetheless, it can be fairly anticipated that understandably zealous defense counsel will make motions to suppress in any future case where the SEC and the DOJ have discussed a possible line of questions in advance of the deposition or where the U.S. Attorney's interest in the SEC proceeding was not disclosed to the deponent. As a result, cooperation is chilled.
How might Congress respond to this problem? As noted above, Section 21(d)(1) of the Securities Exchange Act of 1934 already authorizes the SEC to pass information it gathers in discovery to the Attorney General. But it says nothing about advance cooperation between the two agencies; nor does it address the suggestion by the U.S. Attorney of lines of questions. Section 21(d)(1) might therefore be amended to add a sentence at the end of this section, stating:
"The Commission may consult with other agencies, including the Department of Justice, at any stage of its investigations and shall be under no obligation to disclose these contacts to any person, except on the order of the Court for good cause shown."
The objection to such a provision will be that the deponent might have asserted its Fifth Amendment privilege if the defendant knew of the U.S. Attorney's interest. Or the defendant might seek a stay of civil discovery. In fact, few defendants take the Fifth Amendment before the SEC, because it can result in an adverse inference being taken and the loss of the SEC action. See Baxter v. Palmigiano, 425 U.S. 308 (1976) (permitting adverse inference to be drawn in a civil proceeding from the assertion of the privilege). Still, an occasional deponent does assert the privilege (and Mr. Quattrone recently did). If the civil defendant believes his indictment is imminent (for example, if he has received a grand jury subpoena), he can also still seek a stay of civil discovery. But there seems little justification to force the U.S. Attorney to tip its hand at an early stage before any grand jury has begun its investigation and where the SEC investigation had its own legitimate concerns. This is not an area where the balance of advantage needs to be tipped even further in the criminal defendant's favor. Finally, in extreme circumstances, the proposed language would allow a defendant in an exceptional case to take his concerns to the Court (such as, for example, where the civil defendant believed that he could show that the SEC's action had no possible basis other than to gather information for the criminal investigation. But that is not how the SEC operates).
c. Civil Penalties. Section 21A of the Securities Exchange Act of 1934, 15 U.S.C. § 78u-1, authorizes the SEC to seek civil penalties for insider trading in an amount that "shall not exceed three times the profit gained or loss avoided as a result of the unlawful purchase, sale or communication." In practice, courts have been imposing far less than treble damages. One way to elevate the penalties in civil cases (where they are not high) would be to put a floor on these damages of not less than double the gain or loss (but still bounded by a ceiling equal to three times the gain or loss). Because far more cases are enforced civilly than criminally, this change could have real impact.
SOME CONCLUDING THOUGHTS
Today, the SEC is an effective enforcer of the insider trading laws against individuals, but less so against institutions. To give the SEC greater leverage, the focus needs to shift from simply increasing penalties (Sarbanes-Oxley already has inflated criminal penalties to unconscionable levels) to mandating preventive controls. Pre-transaction disclosure of an intent to trade, transaction reporting, and a requirement that option grants be scheduled in advance of their award are examples of reasonable measures that could reduce the incidence of insider trading. In PIPE transactions, the SEC should similarly insist that an issuer obtain confidentiality agreements from every offeree solicited. All these proposals share the common perception that we have reached the point of diminishing returns from a policy based primarily on deterrent threats.
Finally, the recent Canadian and British experience is instructive: if civil and criminal regulators cannot cooperate, criminal enforcement will not succeed, and, without it, insider trading will not only persist, but flourish.
See Gretchen Morgenson, "Whispers of Mergers Set Off Bouts of Suspicious Trading," New York Times, August 27, 2006, Section 1, at p. 1.
Id. (reporting that there were 76 such referrals for the first six months of this year, up from 60 for the same period in 2005. In turn, 2005 was up by 63% over 2004.)
This question depends largely on whether the board committee that awarded the options was aware of the material, nonpublic information that the executive knew. See SEC v. Texas Gulf Sulphur, 401 F.2d 833, 856-57 (2d Cir. 1968). At the time of the option grants in Texas Gulf Sulphur, neither the Texas Gulf Sulphur board nor its stock option committee was aware of the material information about TGS's ore discovery, but most of the option recipients were aware. 401 F.2d at 844. The Second Circuit indicated that it would enjoin those options that had not already been voluntarily cancelled. Id. at 856-57. Ultimately, the injunction of the one stock option award that had not been cancelled was vacated on other grounds. See SEC v. Texas Gulf Sulphur, 446 F.2d 1301, 1308-09 (2d Cir. 1971).
See Randall Heron and Erik Lee, "What fraction of stock option grants to top executives have been backdated or manipulated?" (July 14, 2006) at Table 4.
See Jesse Fried, Reducing the Profitability of Corporate Insider Trading Through Pretrading Disclosure, 71 S. Calif. L. Rev. 303 (1998).
For related decisions on the mens rea level, see United State v. Dixon, 536 F.2d 1388, 1397 (2d Cir. 1976); United States v. Schwartz, 464 F.2d 494, 509 (2d Cir. 1972).
See Wall v. United States, 384 F.2d 758, 762 (10th Cir. 1967); see also Matthews, Criminal Prosecutions Under the Federal Securities Laws and Related Statutes, 39 Geo. Wash. L. Rev. 901 (1971).