Friday, November 3, 2023

Sell the News

The Financial Times recently reported that

A group of veteran US financial journalists is teaming up with investors to launch a trading firm that is designed to trade on market-moving news unearthed by its own investigative reporting.

The business, founded by investor Nathaniel Brooks Horwitz and writer Sam Koppelman, would comprise two entities: a trading fund and a group of analysts and journalists producing stories based on publicly available material…

The fund would place trades before articles were published, and then publish its research and trading thesis….

I saw a lot of online commentary asking why this isn’t just a model for insider trading, and even though Matt Levine went through some of the issues here and here, I am moved to do something I rarely do and delve into insider trading law to explain the matter further.  For a lot of readers, this is probably nothing new, but hopefully this will be helpful for some of you.

So, the first thing to make clear is that the rules for what counts as insider trading in the U.S. are bizarre and arcane.  And the reason for that is, with a few exceptions like the “Eddie Murphy” provisions of Dodd-Frank, there are no statutory prohibitions on insider trading.  What the statutes prohibit is fraud, mainly through Section 10(b) of the Securities Exchange Act.  And, beginning in the 1960s, courts and the SEC began to interpret Section 10(b) fraud to include insider trading.  Except that meant they had to define the contours of what kind of trading is and is not permitted, and those definitions came about through meandering and contradictory common law rulemaking.  The caselaw is meandering and contradictory because people have very different instincts about what should be illegal and what should not be illegal.  As one article amusingly summed it up:

Manifesting the extent to which even authorities on the subject are unable to articulate a compelling legal theory of what insider trading is and why the conduct it encompasses should be declared unlawful, a large body of case law and commentary, for instance, variously portrays insider trading doctrine as based on principles drawn from or analogous to the law of fraud, breach of fiduciary duty, agency, theft, conversion, embezzlement, trusts, property, contracts, corporations, confidential relationships, unjust enrichment, lying, trade secrets, and corruption.

Andrew W. Marrero, Insider Trading: Inside the Quagmire, 17 Berkeley Bus. L.J. 234 (2020).

In general, there are those who believe insider trading should try to level the playing field, by giving all traders equal access to information, or at least equal opportunity to attain access, and there are those who believe that equal access is impossible – ordinary retail traders will never match the resources of professional firms – and what actually protects ordinary traders is market efficiency, which only comes about from informed trading that sets the price appropriately for everybody.  So the caselaw tends to contain rhetoric that switches back and forth between extolling the virtues of a level playing field versus extolling the virtues of informed market prices.  I also think some of the instincts here are driven by specific distributional concerns – i.e., the print shop employees of the world usually have less access to information than the white-shoe M&A lawyers of the world – and so when prohibitions on insider trading are very narrow, the poor stay poor and the rich get rich.

Anyway, you end up drawing a lot of distinctions that do not, from the outside, appear to make a lot of moral sense.

So, back to this new fund model.  The company is called Hunterbrook, and financial journalists will be tasked with writing articles about publicly traded companies.  The plan, quite explicitly, is for the journalists to rely solely on public information.  I.e., carefully read SEC filings and news reports and use big data calculations and perhaps access obscure but not secret information (Matt Levine suggested FOIA requests).  Before publication, the fund will decide whether to place a trade – long, or short, securities, but also commodities and currencies.  And then, the article will run, and the hope of course is that the article’s insights will move markets, which will then permit the traders to profit from their position. 

That model is similar, but not identical, to those of activist short sellers – they too ferret out information and publish reports, but less formalized as journalism, and only for shorting purposes; this model wants to have some kind of regular news arm attached, and will go long as well as short.  It is, under current law, legal.  The entity is generating its own in-house information – including information about which stories it will run – and trading on the information that it generates.

This is very different than, say, the R. Foster Winans case, where a columnist for the Wall Street Journal tipped a broker about the companies he planned to feature in Heard on the Street. Because in that case, the information – which columns would run – did not belong to Winans, but to the Wall Street Journal, and Winans misappropriated it.  (No, David Carpenter was not Winans’s “roommate,” but this was 1986, so.)

But the Hunterbrook model envisions that the trader is the same entity that owns the information.

That raises the question whether, in the Winans case, the Wall Street Journal could have traded on its own knowledge of upcoming columns – or sold that information to a third party.  And the answer to that is, it depends.  Remember, the Hunterbrook model is that all the information used in the articles is public.  If that were true of the Heard on the Street columns, then yes, the Wall Street Journal could have traded on advance knowledge of its own publication plans.  But Wall Street Journal articles are typically based on inside sources.  Trading on that information, whether by the WSJ or anyone else, depends on an analysis of those sources and their relationship to the WSJ.

Let’s say the sources were revealing inside information about someone else – like, say, information about their employers, or clients, or people they do business with.  There might be all kinds of laws that those sources broke by revealing information to a newspaper (trade secret laws, employment contracts and NDAs, etc), but for the law of insider trading, the only issue is whether that source revealed information to the journal wrongfully, and wrongfully is defined in a particularly convoluted manner

First, the source must be bound by some kind of duty to keep the information confidential – that duty can come from law, or contract, or just a personal relationship where there is an expectation of privacy.  And second, the source must be revealing the information to the WSJ for an improper reason.  For a long time, improper reasons meant the source/tipper expected to “personally benefit” from the tip.  And that usually meant, the tipper was paid – like, someone paid them off for their information.  Or the tipper expected to receive confidential tips in return, that he could trade on, and there was a regular practice of people going back and forth tipping each other.  Sometimes, it meant that the tipper expected to “gift” the information to a close friend, in lieu of monetary compensation.  “Happy birthday, Mom, I didn’t have a chance to buy flowers, but Amazon is acquiring Whole Foods.  Buy yourself something nice!”  It might even mean the source expected a job offer – “Look how valuable I am to you, I can tell you that Amazon is buying Whole Foods!”  What it did not mean was, say, whistleblowing.

So, on first order analysis, if the WSJ’s sources were whistleblowing – and not expecting any personal benefit by talking to the paper, they were not being paid for their information – then the information would not have been wrongfully revealed, and the WSJ would be free to trade on it.

But we are not done.

The “personal benefit” test is difficult to apply; it sent government prosecutors searching for any quid pro quo, including steak dinners or theater tickets or plumbing services or other kinds of trivial rewards.  Eventually, then, in United States v. Martoma, the Second Circuit declared that it would be sufficient if the government could show the source tipped someone in the expectation that the recipient of the information would trade – i.e., instead of hunting for a personal benefit to the source, we’d look to see if the source was trying to benefit someone else – like the Mom’s birthday hypothetical above, but now extended to all relationships, not just especially close ones.

Additionally, in United States v. Blaszczak I, the Second Circuit looked at a brand new securities fraud statute – not Section 10(b) of the Exchange Act, but Section 1348 of Sarbanes Oxley – and held that the fraud prohibitions in Section 1348 do not require a showing that the tipper personally benefitted.  I, personally, never understood that logic – the “personal benefit” requirement was, in a roundabout way, intended to identify when tipping inside information constituted deceptive conduct, and since Section 1348 prohibits fraud, just like Section 10(b), you’d think they’d be read the same.  But no matter, because in Blaszczak I, the Second Circuit held that instead of showing a personal benefit, it would be sufficient to show that the information was “misappropriated” for one’s own use – including to give to another to trade on.  In the end I don’t see a whole lot of daylight between that standard and Martoma, so, fine, they are roughly the same. (Then, the Second Circuit decided United States v. Blaszczak II and two judges on a 3-judge panel suggested they might want to go back to a personal benefit test even for Section 1348 fraud, so who knows what that even means now).

Under this analysis, the question would be, did the WSJ’s source provide information intending that the newspaper would trade on it?

Again, the answer would probably be no, and so – on first blush – the information was not provided wrongfully, and the WSJ would be free to trade.

But we still are not done.

Because if the source gave this information to the WSJ as, say, a whistleblower – not for the purpose of having the WSJ trade – it’s possible a court would say that if the WSJ traded, then the WSJ violated its own duty of confidentiality to the source, in a manner of speaking, and misused the information for its own benefit. 

That would be sort of a weird argument, because the source never intended confidentiality – the source intended publication! – but courts tend to punish based on gut instincts of fairness and it’s not at all difficult to imagine a court holding that the WSJ misappropriated information that was given to it for a specific purpose, and, hence, fraud.

But now let’s translate all of this to the Hunterbrook context.  If the journalism arm is attached to a trading arm, then for sure any source who gives the journalist information knows it will be used for trading.  And we’re back to that first analysis: It’s wrongful to give someone confidential information, derived from an employer or a client or whatever, so that they can trade.  And maybe our source could thread the needle – “I knew they would trade but that’s not why I told them; I told them to expose the bad stuff!!” – but I wouldn’t want to be that source’s lawyer, is what I’m saying.

Anyway, all of this means that Hunterbrook will not be talking to confidential sources, but instead believes it has a viable business model by synthesizing and digesting public information, which it can use to earn a trading advantage and move markets.  Note, for whatever reason, Hunterbrook does not thinking trading alone will do it – it does not trust that the market will eventually catch up to Hunterbrook’s own wisdom, or, will do so on a fast enough time frame for Hunterbrook to profit.  No, Hunterbrook has to trade, and then move things along by publishing what it discovers. 

And I don’t know if they can pull it off or they can’t, but if they do manage to regularly publish breaking news stories culled entirely from public information, which have the effect of moving markets and allowing Hunterbrook to earn outsized returns – what do you imagine will happen when securities fraud plaintiffs bring follow-on complaints against the companies targeted with Hunterbrook’s negative information?  Do you think the obvious evidence that this “public” information was nonetheless not incorporated into market prices will make courts any more likely to accept that transaction causation and loss causation have been properly alleged?  (Reader, it will not.)

Ann Lipton | Permalink


Post a comment