Sell-side research isn’t inside information

By Felix Salmon
May 20, 2012

Ted Parmigiani, a research analyst, was fired by Lehman Brothers in 2005, after he upgraded a company called Amkor from Sell to Neutral. He discussed his upgrade on the firm-wide squawk box at 10am on June 1, and it seems that he was hoping that his upgrade would move the market. He was disappointed: while Amkor did rise that day, it rose before his 10am call, rather than after it. Convinced that the early-morning move was attributable to Lehman insiders and their clients buying Amkor with foreknowledge of his upgrade, Parmigiani started a series of complaints.

First, Parmigiani complained to Lehman superiors; that just got him fired. So he filed a wrongful-termination suit, which got settled out of court. And with his suit behind him, Parmigiani kept on complaining about what had happened. First he went to the SEC, which looked into his allegations and found no evidence of wrongdoing. After striking out with the SEC, Parmigiani complained to Senator Charles Grassley, and also to Preet Bharara, United States attorney for the Southern District of New York. Again, nothing. Finally, with nowhere left to go, and despite the fact that Lehman went spectacularly bust back in 2008, Parmigiani went to Gretchen Morgenson, some seven years after the events in question happened. And this time, he got results, if by “results” you mean a big 3,000-word article in the New York Times.

I’ll let Morgenson explain what Lehman is supposed to have done wrong:

Mr. Parmigiani says traders there were routinely advised of changes in analysts’ company ratings before those changes were made public. That way, Lehman could profit on subsequent market moves. Here is how he describes it: First, research officials tipped off the traders; then Lehman’s proprietary trading desk, which cast bets with the firm’s own money, positioned itself accordingly. Lehman salespeople also alerted favored hedge funds. Only later, he says, were ratings changes made public.

I have to admit that I’m a bit confused with what exactly Morgenson means by the phrase “made public”. I, for one, never had access to Lehman Brothers squawk-box calls, and I’ve never seen sell-side upgrades or downgrades filed with the SEC or be made subject to Reg FD.

Now it’s true that the SEC does consider the news that an investment bank is going to change its rating on a stock to be material nonpublic information, and therefore subject to insider-trading rules. But it’s a very weird, grey-area sort of inside information. Let’s say Lehman upgrades Amkor: is that inside information about Lehman, or is it inside information about Amkor? The news would be very unlikely to affect Lehman stock, so it’s not really the former. But at the same time, no Amkor insiders are involved at all, so it’s hard to see how it can be the latter.

There can’t and shouldn’t be any rule against analysts doing independent analysis on companies, based on publicly-available information. Similarly, once that analysis has been done, the analyst can do what she likes with her analysis. She can trade the stock, she can write it up, she can talk to hedge funds about what she thinks, she can sell it to clients, she can make it public. Or, she can do all of the above, in any order she likes.

The tricky thing happens when that analyst gets a job at an investment bank, and the individual’s analysis mutates into being the bank’s analysis. There’s a kind of invisible special sauce which a bank pours onto its analysts’ reports: if Joe Schmo puts out a “buy” rating on XYZ Corp on Seeking Alpha, few people will notice or care. But if Mr Schmo gets a job at Goldman Sachs, then that rating can move XYZ shares. And at that point, the SEC starts getting interested.

Now the SEC does not say that banks need to make their research reports fully public. In its complaint against Goldman Sachs “huddles”, for instance, the SEC has no problem with news of upgrades and downgrades being “disseminated broadly to all clients of the firm”. There’s a huge difference, of course, between the set of Goldman Sachs clients and the set of public-company investors, but as far as the SEC is concerned, it seems that if all of Goldman’s clients have a certain piece of information, then that piece of information can be considered to be public information.

Similarly, Morgenson and Parmigiani seem to think that once a call has been made on the Lehman Brothers squawk box, that too counts as public information. If something’s available — in theory — to any Lehman client, then it’s public. If it’s only available to select Lehman clients, then it’s inside information.

Here’s Morgenson:

While insider trading commonly involves nonpublic corporate information, advance warning on research changes can also yield quick, illicit gains. The S.E.C. said as much in a rare research case it filed in 2007 involving an executive at the Swiss banking giant UBS. In that case, eight individuals and three hedge funds were charged with profiting on tips about coming analyst ratings changes — “valuable and material, nonpublic information,” the S.E.C. said. One executive went to jail, and others settled with the S.E.C.

This is disingenuous. The case in question involved individuals, not the bank as a whole: a UBS executive director, Mitchel Guttenberg, was personally tipping off his friends, in return for under-the-table payoffs, about upcoming UBS upgrades and downgrades. In that sense, the case is really about misappropriation rather than about insider trading. As John Carney explains, using the example of Foster Winans:

An employee’s undisclosed and self-serving use of information belonging to his employer to trade securities is a securities fraud. The employee is “misappropriating” the information from his employer and using it for his own personal gain…

If the Wall Street Journal had authorized Winans to leak to his roommate and his stockbroker, no violation would have occurred. In that case, there would be no breach of duty to his employer, which means there’s no misappropriation.

In other words, as far as I can tell, no one has ever been successfully prosecuted for the crime that Morgenson and Parmigiani are so upset about here — the crime of giving information about ratings actions to some clients before other clients. The Goldman huddle case, in particular, seems particularly thin: the SEC is basically shocked — shocked! — that before an analyst upgraded a company, that analyst was bullish on the company in question. Whereas the rest of us would probably be shocked if an analyst wasn’t bullish before upgrading a company. And in the end, the Goldman huddle case was more about internal controls than it was about inside information.

The fact is that if no corporate insiders are involved, there’s really no good reason at all to prosecute or criminalize anybody using information they’ve obtained to act in the markets. The markets should be a game of people using independently-obtained information and analysis to make their own determinations about what various securities are worth — that’s the best way to maximize the amount of information reflected in the price of those securities.

As the Facebook IPO demonstrated, pricing securities, especially stocks, is always more of an art than a science. Facebook’s underwriters thought they had a pretty good handle on where the demand was for Facebook’s stock, but they were wrong: all those investors were deliberately upsizing their orders just because they knew they wouldn’t get all the shares they were asking for. And because they also reckoned they could lock in a nice profit by selling some of their shares at the very first tick. By the end of the day, Facebook’s banks were being forced to buy back stock, in significant quantities, at the same price they’d sold it.

So let’s let brokerages’ clients trade what they like, so long as they’re not trading on genuinely inside information from the company in question. If we’re going to be serious about the Volcker Rule, and prevent the brokerages from trading for their own account, the least we can do is let them monetize their analysts’ research as best they can.


Comments are closed.