Herbalife Trader Sued for Insider Trading Sues SEC Right Back
One of my little obsessions is the social relationship of insider trading. Insider trading is so dumb, and can send you to jail for so long, and pretty much requires an accomplice. How do you decide whom to trust with something so momentous, and so dumb? Some people -- a lot of people actually -- insider trade with their golf buddies. Others insider trade with strangers using Post-it notes that they eat. Both strategies make a kind of sense to me, though I guess I'm more of a strangers-and-Post-its kind of guy.
Anyway, today is a wonderful day for insider-trading sociology. First there is this recent paper about insider-trading social networks, by Kenneth Ahern of the University of Southern California. Basically Ahern went and looked at all of the insider trading cases brought by the Securities and Exchange Commission and the Justice Department between 2009 and 2013, and classified everyone's relationships. It's sort of sweet:
In the sample, there are just three pairs described as acquaintances, compared to 115 described as friends, and 44 described as close friends. In untabulated statistics, I verify that the distinction between friends and close friends is meaningful. In some cases, the SEC complaint identifies when the insider relationship began. Among the 13 pairs of close friends with available data, 77% met before college and 23% met in college. In comparison, among 21 friendship pairs, 29% met before college, 24% met in college, 29% met in graduate school, and 19% met after they had completed their education. This suggests that close friendships were formed during childhood, as might be expected.
Or there is this strangely wondrous ethnic breakdown of insider trading:
The diagonals tell you that insider traders mostly keep their tipping within their ethnic groups, except Western Europeans, who tip and are tipped promiscuously.
I don't know exactly what you'd conclude from this paper. Many of Ahern's findings seem to be about who gets caught, not who insider trades. For instance:
I find that executives are about three times more likely to share inside information than they are to receive it. In contrast, buy side insiders share information about half as often as they receive it. Information tends to flow from subordinates to bosses, from younger tippers to older tippees, and from children to parents.
Maybe? Or maybe it's easier to catch someone who trades on a tip from a corporate executive than it is to catch someone who trades on thirdhand tips from buy-siders. And Ahern recognizes the difficulty in concluding anything about "insider traders" from a sample of "insider traders who get caught." The most basic of all my conclusions about insider traders -- that they are dumb -- is itself probably an artifact of that data problem.
Nonetheless, I believe it. So some of my fascination with this topic is just, insider traders are adorable, so sweet and so dumb, and they move me on a personal level. I want to know about their childhood friendships and their ethnic backgrounds and their trust in each other and their weird obsession with buying all of the short-dated out-of-the-money call options that there are:
I find that some insiders are oblivious to the regular monitoring of financial markets by regulators. For instance, in one case, a trader’s purchases accounted for 100% of the volume of out-of-the-money call options in the days before a merger announcement, sending up red flags to regulators.
But also, this social stuff actually matters for figuring out what the law is. Insider trading means: X gives information to Y, Y has some obligation to X to keep it secret, but Y trades on it anyway, or passes it to Z who trades on it. Y and Z get in trouble for stealing information from X. There's a straightforward form where X is a company and Y is an executive at the company, which is how most people think of insider trading.
But there are so many other, weirder forms. To the SEC, "inside" information doesn't mean information that you got from a company. It means information that you got in violation of a duty of confidence. Which comes not necessarily from, like, working at a company or signing a nondisclosure agreement. It can come from being friends -- close friends, anyway -- or golf buddies.
Or roommates. We've talked about the Herbalife insider-trading case: Pershing Square had a big short on Herbalife and was finishing up its presentation, when an analyst informed his roommate about the presentation, the roommate informed a friend named Jorgan Peixoto and Peixoto traded half an hour before Pershing's position was announced, cratering the stock. The SEC concluded that the roommate, Filip Szymik, had stolen information from the analyst, and therefore came after Peixoto and the roommate for insider trading. But not the analyst: The theory is not that the analyst violated his duty of confidentiality to Pershing Square, but rather that the roommate violated his duty of confidentiality to the analyst.
Today Peixoto decided to sue the SEC over this theory. Or, not precisely over this theory. The point of Peixoto's lawsuit is that the SEC brought its case in an "administrative proceeding" in front of an SEC-appointed administrative law judge, rather than in a real federal court. It used to be that the SEC could only do that with SEC-regulated people, that is, brokers and stuff. Peixoto is not a broker. But Dodd-Frank allowed the SEC to bring cases against non-regulated people, which has caused some controversy.
But Peixoto's basic point is: The SEC wants to try this case in front of its own judges, because it knows its thesis is a little far-fetched. From his lawsuit:
The SEC could not prove that Mr. Peixoto knew or should have known that Szymik and the Analyst had the type of intimate friendship which gives rise to a duty of confidentiality, or that Szymik breached any such purported duty. Nor could the SEC establish that Mr. Peixoto knew or should have known that whatever information Szymik conveyed to him was confidential.
Similarly, the SEC cannot prove the existence of a duty of trust and confidence -- another required element of insider trading charges. Szymik and the Analyst provided conflicting, and self-serving, testimony as to whether Szymik promised to keep information he learned from the Analyst confidential. Similarly, the SEC has scant, if any, evidence that Szymik and the Analyst shared the type of intimate friendship that gives rise to a duty of confidentiality. And if Szymik was under no legal duty to keep the information confidential, Mr. Peixoto cannot be held liable for insider trading as a matter of law.
He's got a point, no? If you took this case to a real court, a judge might reasonably say: Wait, Szymik owed a duty of confidence to the analyst? And not the analyst to Pershing Square? And Peixoto was supposed to know about that? The SEC's theory is not that Peixoto knew or should have suspected that this information was taken from Pershing Square in breach of a duty of confidentiality, which is a reasonable enough thing to suspect. It's that he should have known that it was taken from a roommate in breach of a duty of roommate confidentiality. He was supposed to know about the code of secrecy that bound the analyst and his roommate, and to know that anything the roommate told him was in breach of that duty.
That's a weird thing for a court to think. But it's more natural for the SEC. For one thing, of course, the SEC is just institutionally biased toward catching insider traders. But also: It lives with the social facts of insider trading every day. To the rest of us, insider trading seems like it ought to be connected to, you know, insider status. But to the SEC, insider trading is all about friendship. So when it brings an insider-trading case based on a betrayal of a friendship, it wants to bring it in its home courts.
I learned of it via this Fortune article, which for some reason doesn't link to the paper.
Ha no I'm kidding it's Steve Cohen. Allegedly!
And, if Z does the trading, Y gets a personal benefit. I say "he" advisedly; something like 10 percent of the people in Ahern's data set are women.
SEC Rule 10b5-2 sets out how duties of confidence can come about. Spouses, siblings, parents and children automatically have those duties, as do people who sign nondisclosure agreements. But also:
Whenever the person communicating the material nonpublic information and the person to whom it is communicated have a history, pattern, or practice of sharing confidences, such that the recipient of the information knows or reasonably should know that the person communicating the material nonpublic information expects that the recipient will maintain its confidentiality;
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
To contact the author on this story:
Matthew S Levine at firstname.lastname@example.org
To contact the editor on this story:
Zara Kessler at email@example.com