Skip to main content

Everybody's Jumping On The Insider Trading Bandwagon

  • Author:
  • Updated:

It's a tribute to Steve Cohen's prescience/power/something that, on what is otherwise not a great day for SAC on the insider trading front, the rest of the news in the world is all pretty much "everybody is insider trading everything all of the time," so, like, leave Steve alone! Today brings some old-fashioned mustache-twirling insider trading - here you can find the SEC's charges against Delta Petroleum's CEO, who apparently tipped his friends about an upcoming buyout and other news; those friends then sent each other emails saying things like "our mutual friend who will go unnamed WINK WINK WINK tells me that DPTR has good news coming MASSIVE NECK-STRAINING WINK" - but the real action is in thesetwoJournal articles on what you might characterize as pervasive insider-trading-lite.

Who insider trades? Insiders, for one. This article about how executives have suspiciously good luck trading for their own account is perhaps too suspicious, as a lot of it is anecdotal or cherry-picked and it conflates 10b5-1 and discretionary trading a bit. Rule 10b5-1 plans, in which executives who do not have material nonpublic information set up automatic future sales (mostly) to top-tick the stock and/or pay for their kids' college tuition, may have good or bad or indifferent results but you mostly can't get mad at the executives if their 10b5-1 robots have suspiciously good timing; the robots really are mostly robots. On the other hand they're not entirely robots and might be ripe for reforming; I'm like 75% on board with Ronald Barusch's suggestions (I am not as troubled as he is by secret adoption of 10b5-1 plans during clean windows) but the bigger conceptual hole is that, as the Journal notes:

Executives can generally cancel a trading plan at any time. This includes times when they possess private information and when the cancellation of the plan's prescribed trades could benefit them financially.

By, like, put-call parity,1 setting up a [buy | sell] at time T0 when you have no inside information, and then at time T1 cancelling it when you have [bad | good] inside information (or not cancelling it when you have [good | bad] inside information), is the same as just trading on inside information. If I were a shady shady CEO - and, let's face it, I would be - I'd set up a 10b5-1 plan that involved selling all my shares the day before every quarterly earnings announcement. Then I'd just cancel it the day before that, as long as earnings weren't going to be a disaster. (And then set up a new one just after earnings, when I was in a clean window again.) Then, the one time that earnings were going to be a disaster - and, let's face it, that time would come - I'd let it sell all my shares. That's about as shady an insider-trading scheme as you can have, and also it's legal.2

More broadly though: you can only put a 10b5-1 plan in place when you are clean of material nonpublic information, but that's just material nonpublic information. As Barusch notes "executives understand the business better than any member of the general public or analyst ever could and even if they do not have material nonpublic information, their trades will generally be more profitable." That is - entirely right, but - a weird thing to say! "Material nonpublic information" is just information that (1) the general public doesn't have but that (2) "would have been viewed by the reasonable investor as having significantly altered the 'total mix' of information made available." A naive person might think that altering your trades from "less profitable" to "more profitable" would be significant to the reasonable investor.

And I guess it is? You know who else insider trades? Outsiders! Specifically, and amusingly, the ones who bother meeting with management. This article, about how investors like to meet with executives, and how the executives are getting sick of it, but how the investors who do lots of meetings perform better than the ones who don't, is full of riches.3 But let's focus on the following quote, which is based on a somewhat silly study - of one company! - but still:

While most investors met with firm executives only once during the six-year period, several hedge fund managers and large stakeholders scored meetings several times a year.

Those who had more frequent meetings tended to make better investments, Mr. Soltes found, increasing their positions ahead of rises in the share price and paring back as shares fell.

So I mean first of all, if you (1) are in the sort of business where you sometimes meet with companies in which you invest but (2) only do so once every six years, should you expect to do well on that investment? I mean if you're an algo robot, skip the meetings, whatever, but why are you meeting with companies sometimes if you're not bothering to meet with them a lot? What value does a once-in-six-years management meeting add?

But more generally, duh, right? How could meeting with management a lot - and generally being in the discipline of meeting with managements a lot, and knowing how to interpret what they say and don't say and their body language, and befriending them enough to get them to say more and not say less and speak more eloquently with their bodies, etc. - not make you better at investing in their companies? This is the mosaic theory of investing, I guess, where you can bop around finding out a little information here and a little information there and stitching it all together into a buy or a sell. On this theory if the CEO tips you that "things are pretty good and we feel confident about the future" by how he crosses his legs at your frequent one-on-ones, that's fine, but if he tips you that "we are being acquired by Tracinda tomorrow" by calling you up and telling you that, that's not fine.4

You can sort of have two views of the interaction of investing and information. In one view the way you make money on stocks is:

  • Sit around being uninformed.
  • One day get a sure thing.
  • Put all your money on the sure thing.
  • Profit.

In the other it's like:

  • Go around looking for things that alter your total mix and give you a slight comparative advantage.
  • Put all your money in some sort of Kelly-criterion way in a diversified pool of things where you have a comparative advantage.
  • Profit a little more than you lose, in expectation.

The SEC view of insider trading is kind of the former, as witnessed by the name "Operation Perfect Hedge," which, gaah. But: that's also lots of people's view of insider trading! Like, including the insider traders! If you find out that a drug company is going to announce a bad result from a clinical trial, and the next day liquidate hundreds of millions of stock and swing to a huge short position, yeah, sure, whatever, Perfect Hedge.

And fine those are bad things whatever. But I feel like if I were the SEC or DOJ lawyer charged with bringing Mathew Martoma or that Delta Petroleum guy to justice, I'd be pretty depressed by today's Journal articles, no? What am I doing with my life, in this alternate universe?5 I'm sort of sanding down the very highest hills of the playing field, punishing people who blatantly trade on super-material information and profit at the expense of other, less connected investors without that information. But the field still tilts! The people with access - executives or just favored shareholders - will do better than the people without access, though worse than SAC usually. If your faith in the integrity of the capital markets depends on everyone having access to the same information as everyone else,6 then I feel like your faith should be weaker today. Despite that SAC Wells notice.

Executives' Good Luck in Trading Own Stock [WSJ]
Investors Demand CEO Face Time [WSJ]
Dealpolitik: Reining in Executives’ ‘Luck’ in Trading Their Company’s Stock [Deal Journal / Ronald Barusch]


2.The advanced move is to also have a cancellable 10b5-1 plan that buys tons of options just before you announce you're being acquired. This is much harder - acquisitions don't come around at a predictable time every quarter, and once you can predict one you can't really set up a 10b5-1 plan - but also less necessary. As the CEO you're sort of naturally long both stock and career-and/or-golden-parachute effects on being acquired. Selling before you announce disastrous earnings is much more important, and much easier.

Also for the love of all that is holy I am not your lawyer and don't try this at home or in your office or elsewhere. Rule 10b5-1 requires that you enter a plan "in good faith and not as part of a plan or scheme to evade the prohibitions of this section," and after you've cancelled your sell-everything order a few times you're kind of clearly not that. This is more reductio-ad-absurdum than sound financial planning advice.

3.Like, I feel like you could spend all day pondering this question: If you are a CEO and you spend 20% of your time meeting with investors, is that too much, too little, or just right? How do you measure? When they conflict, is your job "maximizing the profits of the company" or "maximizing the economic well-being of your shareholders"? Henry Blodget a while back put up some FRED charts showing wages at 44% of GDP and profits at 11% of GDP. So if your time is divided among employees and shareholders the way your money is then 20% to shareholders seems right. Of course that is a dumb way to think! Also like, customers and suppliers and stuff. Still though: if a theme of recent years is, like, capital demanding more attention, in general, then this is a nice example of it. Don't just give us our dividends, make us feel special about getting our dividends.

4.I have recommended this before, but I will recommend it again here, because it is good:

I have tried to get securities lawyers and prosecutors to define the exact moment at which information crosses from legal to illegal. When the inference and analysis of what will likely happen becomes an unfair advantage of positively knowing what will happen.

From what I have learned, reality very rarely hands you a smoking gun; there is seldom a moment anymore, as with Dennis Levine, of a tit-for-tit exchange of state secrets. That's a prosecutor's construct. …

You can see the one-dimensional ways the prosecutor will construct [the Martoma case]. Dr Gilman, who will go to jail if he doesn't frame reality from the prosecutor's view, will say he gave up proprietary information, which he probably did, even though he probably didn't think he was, or probably thought he was controlling the information game. Martoma, on his part, will believe and maintain, until prosecutors flatten him, that if he got two mosaic tiles from Dr Gilman, the telling detail in the picture involved at least three or four.

5.In which I am not a blogger. Hmm.

6.Which I feel like it maybe shouldn't? That's one viable conclusion, though the alternative of "put everyone in jail" is another. (Is it?)