He's ready.

Photographer: Scott Eells/Bloomberg.

Steve Cohen Will Be Back in 2018. That's Good.

Matt Levine is a Bloomberg View columnist. He was an editor of Dealbreaker, an investment banker at Goldman Sachs, a mergers and acquisitions lawyer at Wachtell, Lipton, Rosen & Katz and a clerk for the U.S. Court of Appeals for the Third Circuit.
Read More.
a | A

Here's a simple model of Steve Cohen:

  1. He is really really really really good at making money in the stock market.
  2. He generously shared those skills and used them to make money for pension funds, university endowments and other investors, in exchange for a modest 3 percent management fee and a 50 percent cut of the profits.
  3. Now, due to circumstances, he has stopped doing that, and is making money only for himself and his employees.
  4. That is somewhere between a misallocation of resources and a national tragedy.

In this cold dark time of year, when hedge funds are constantly being criticized, losing money and shutting down, it is heartening to remember that there's at least one hedge-fund manager with a multi-decade record of consistent outperformance, and frustrating to remember that no one can invest with him. Come back, Steve Cohen! An embattled industry needs you.

You could have an alternate model, which goes something like: "Steve Cohen is just good at insider trading and should stop." But I think events of the past few years have -- well, not disproved that model, necessarily, but at least shown that Cohen is good at trading even without the insider part. In 2013, the last year he managed his hedge fund SAC Capital Advisors, it was up 20.1 percent (versus 6.5 percent for the average hedge fund). The next year, he closed SAC and shifted to a family office, Point72, which ended the year up ... a lot? Something on the order of 30 percent gross, though that's before fees and expenses, versus 3.3 percent (net) for the average hedge fund. Through October 2015, the latest performance numbers I could find, Point72 was up about 12 percent, versus basically flat for the average hedge fund. That all happened after the intensive investigations by the Securities and Exchange Commission and the Department of Justice were announced, and some of it happened after Cohen started hiring a fleet of former prosecutors and FBI agents, paying bonuses for compliance tips, banning instant messaging and generally turning his firm into a compliance operation that does the occasional stock trade. If people at SAC/Point72 were insider trading in the last three years, they had to be incredibly sneaky about it.

Of course, SAC had a fairly astounding two decades before those investigations, but you could always doubt the legitimacy of those results, and plenty of people -- not least at the SEC and DOJ -- did. You can interpret Cohen's last few years in the wilderness as a carefully proctored test of his ability to make money without cheating.

And he passed! Today Cohen settled the SEC's case against him for "failing to supervise" SAC appropriately. Cohen will be banned from managing outside money until the end of 2017, but starting on New Year's Day 2018, he's back in business.  (If he wants to be.) But "Before Cohen can handle outside money again, an independent consultant will ensure there are legally sufficient policies, procedures, and supervision mechanisms in place to detect and deter any insider trading"; it seems that the consultant is likely to be one Bart Schwartz, a former prosecutor who previously did compliance monitoring at SAC as part of its 2013 guilty plea agreement to insider trading.

It strikes me as a reasonable compromise in a case that both sides might feel nervous about taking to trial. On Cohen's side, I mean, I like to emphasize that four SAC entities and seven former SAC employees have been convicted of insider trading, including two named Richard Lee.  Cohen has fancy lawyers, and I am sure that they would have been able to make sophisticated arguments at trial, but at the end of the day if seven of your employees insider traded then it is hard to argue that you did a great job of supervising them. Also, the actual SEC allegations against him, as contained in today's settlement agreement, are both less wide-ranging and more troubling than that simple counting exercise suggests. The SEC case only touches on one of those convicted insider traders, Mathew Martoma, but it is full of suggestive evidence that Cohen at best failed to look into stuff that looked like insider trading, and at worst was complicit in Martoma's insider trading.

On the SEC's side, though, the legal landscape for insider trading has shifted a lot since it first brought its case against Cohen in 2013.  In the 2014 Newman decision, a federal appeals court ruled that to prove illegal insider trading prosecutors have to show that the person providing the inside information received a "personal benefit" for doing so, and that the person trading on it knew about that benefit. That's a big deal for insider trading law, but it is a particularly big deal for insider trading law at hedge funds, where analysts are employed to build relationships with knowledgeable people in their industries, including at the companies they cover. And it is an even bigger deal for insider trading law at hedge funds like SAC Capital, a decentralized operation made up of "groups of individual funds that are managed separately and compete against one another." Under Newman, it seems virtually impossible that the manager of a fund like that could commit insider trading.

Of course, Cohen was charged with failure to supervise, not insider trading. But similar principles apply. In 2013, you could read messages from Cohen to his subordinates saying things like "i think mat [Martoma] is the closest to it" or "seems like mat [Martoma] has a lot of good relationships in this arena" and interpret them as suspicious. In 2013, it was easy to argue that phrases like "closest to it" and "good relationships" were red flags, and when combined with the fact that Martoma was actually getting material nonpublic information from a doctor involved in confidential drug trials (and paying that doctor), meant that Cohen knew Martoma was up to no good. But in 2016, the law of insider trading is much more tolerant of hedge-fund analysts who build relationships with people who are close to situations. It's only illegal now if the people providing the information are benefiting, and if the analysts know about it -- a common situation in sloppy amateur insider trading, but a rare one in high finance. Those relationships are no longer the red flags that they were in 2013, and a hedge fund analyst's "good relationships" are again a sign of diligence rather than one of illegality.  

Also, like, what I said at the beginning. What is the SEC's job? One part of its mandate is to stop and deter insider trading, sure. But it is also supposed to "protect investors" and "facilitate capital formation," and denying investors access to a wildly successful fund manager fits a bit awkwardly with that mission. He's spent the last three years being washed clean of the taint of insider trading, and will spend the next two years being rinsed under SEC supervision. After that, though, if investors want to give him their money, and if he wants to take it, then let no one stand in their way.

The most promising part of this settlement for the SEC seems to me to be the monitor that it is installing. The independent consultant will:

(a) conduct a review of the Cohen Entity’s compliance with the federal securities laws and issue a report at least every six months to the Cohen Entity and the staff of the Commission describing the scope and results of the IC’s review; and

(b) In connection with each report described in paragraph 71(a) above, recommend any additional policies and procedures which, on the basis of the review, the IC believes are reasonably designed to ensure the Cohen Entity complies with the federal securities laws (the “Recommendations”).

And then Cohen will just have to go implement whatever recommendations the consultant comes up with.  I think it is fair to say that prosecutors, and the SEC enforcement staff, take rather a stricter view of insider trading law than, you know, the law does. As a strict matter of federal criminal law, it's not illegal to trade on insider tips unless the insider received a personal benefit, and as a practical matter that makes it very difficult to punish hedge funds for aggressively seeking "edge" from corporate insiders. But as a matter of best practices recommended by a monitor? There are no rules. Remember, Cohen's monitor will be a former prosecutor chosen with the SEC staff's agreement ; it seems unlikely that he will be an advocate for pushing the envelope and staying just the right side of the criminal law.

For the next two years, Cohen's family office will be regulated less like a hedge fund, which can fight back if it feels like the SEC is wrong on the law, and more like a bank, dependent on unreviewed regulatory goodwill to survive. The SEC can try to shape how Cohen's firm interacts with companies, how it rewards analysts for getting "edge," and generally how it monitors behavior that the courts might not think is illegal but that the SEC dislikes. It's an opportunity for Cohen/Point72/New SAC to become "a leader in compliance," and for the SEC to have a hand in shaping what exactly that means -- and to push it beyond what the Newman decision requires of hedge funds. And then, when Cohen goes back on the market in 2018, perhaps other hedge funds will follow his lead on insider-trading compliance, and the SEC will achieve through Cohen what it can't quite get through the courts.

  1. SAC did underperform the S&P 500 index that year, which I feel like is relevant for an equities-focused hedge fund, but it is long/short and has other strategies so it's not a totally fair comparison. In any case, I am hiding the S&P reference down here so that I don't enrage purists like my Bloomberg colleague Tracy Alloway, who criticized Showtime's "Billions" because it portrays "a hedge fund manager opting to benchmark his performance against the Standard & Poor’s 500 Index." I gather that the hedge fund manager in "Billions" is not exactly based on Cohen but, you know.

  2. But I think SAC isn't. SAC's 2013 plea agreement provided that the "each of the SAC Entity Defendants agrees to cease operating as an investment adviser and to not accept any additional funds from third-party investors." The SEC settlement doesn't on its face seem to reverse that, so I guess those defendants (S.A.C. Capital Advisors, L.P.; S.A.C. Capital Advisors, LLC; CR Intrinsic Investors, LLC; and Sigma Capital Management, LLC) are out of luck. Fortunately it is cheap to get a new entity, and the plea agreement didn't cover Cohen himself.

  3. And two named S.A.C. Capital Advisors. (L.P. and LLC that is.) The list of entities is S.A.C. Capital Advisors, L.P.; S.A.C. Capital Advisors, LLC; CR Intrinsic Investors, LLC; and Sigma Capital Management, LLC; who pleaded guilty in 2013. The list of humans is, according to Bloomberg, "Noah Freeman, Donald Longueuil, Wesley Wang, Richard Choo-Beng Lee, Jon Horvath and Richard Lee," all of whom pleaded guilty, and Mathew Martoma, who was convicted at trial and sentenced to nine years in prison. (An eighth former employee, Michael Steinberg, was convicted at trial, but his conviction was reversed and the charges dropped.)

  4. For instance:

    43. In March and April 2008, two analysts at CR Intrinsic (the “Analysts”) repeatedly sent emails to Cohen advocating against the Elan and Wyeth positions and suggesting trading strategies to hedge them.

    44. The Analysts exchanged a number of emails and instant messages with Cohen about whether Martoma’s advice on Elan and Wyeth was sound. On March 28, 2008, one of the Analysts told Cohen in an instant message that he did not think anyone could yet know the Phase II Trial data, because the trial was not over yet. Cohen responded by saying he would follow Martoma’s and another hedge fund manager’s advice, because “they are closer to it than you.” Later in the same message, the Analyst asked Cohen: “[I] don't know if [the other hedge fund manager] or mat [Martoma] will answer, but do you think they know something or do they have a very strong feeling.” Cohen replied: “[T]ough one . . . i think mat [Martoma] is the closest to it.” The Analyst responded: “[T]he question that I would ask is if it[’]s possible to know the data yet – i could be wrong, but i don't think it is yet.”

    45. A few days later, on April 6, 2008, Cohen exchanged instant messages with the other Analyst about the Drug. Cohen remarked that it “seems like mat [Martoma] has a lot of good relationships in this arena.” 

    Those relationships turned out to be with doctors involved in Elan's drug trials who were illegally leaking information to Martoma. The e-mails and IMs are not proof that Cohen knew that. But they are ... Cohen probably wouldn't want to read them to a jury.

  5. In rereading what I wrote about SAC's November 2013 guilty plea, I came across this paragraph:

    In March, I suppose it was possible to believe that SAC's settlement with the SEC might put an end to its insider trading troubles. Now, though, even though this plea deal puts an end to SACno one quite believes that its troubles are over. The government is having too much fun to stop this case just because it has recovered X dollars, for any X.

    But in fact that came very near to being the high-water mark for SAC trouble, though it took the Newman decision a year later to really turn the tide.

  6. In fact, the SEC's original lawsuit involved failure to supervise, not just Martoma, but also Steinberg; the SEC dropped his trading from the case against Cohen when his criminal case was dismissed. 

  7. I mean, he can object, but the system is stacked against him:

    Cohen agrees that within 60 days following the receipt by a Cohen Entity of the Recommendations, the Cohen Entity shall adopt all Recommendations of the IC; provided, however, that within 30 days of the receipt of the Recommendations, Cohen shall in writing (i) notify the IC and the staff of the Commission of any Recommendations that he considers to be unnecessary, inappropriate, or unduly burdensome, and (ii) propose an alternative policy, procedure, or system designed to achieve the same objective or purpose. As to any Recommendation on which Cohen and the IC do not agree, such parties shall attempt in good faith to reach an agreement within 30 days after Cohen serves the written notice and proposal described above. In the event that Cohen and the IC are unable to agree to an alternative proposal, Cohen will ensure that the Cohen Entity abides by the determinations of the IC by no later than the 75th day following the receipt of the Recommendations.

  8. I base this on things like the Justice Department's unsuccessful appeal of the Newman decision, or the SEC's unsuccessful argument to one of its own administrative judges that he shouldn't apply that decision in an SEC administrative action.

  9. It doesn't have to be a former prosecutor, but it has to be Bart Schwartz (a former prosecutor) or "another IC not unacceptable to the Commission staff."

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

To contact the author of this story:
Matt Levine at mlevine51@bloomberg.net

To contact the editor responsible for this story:
Tobin Harshaw at tharshaw@bloomberg.net