Levine on Wall Street: Grading Analysts and Activists

The Fed makes fun of Deutsche Bank, Eliot Spitzer ranks stock analysts, Bill Ackman hates Herbalife, Argentina hates U.S. courts, and Marty Lipton hates activists.

Deutsche Bank is terrible: New York Fed.

Here is the story of how the New York Fed in December sent Deutsche Bank an enormously long letter (excerpted here) saying in effect "You are garbage, everything you do is wrong, we hate you, okay, carry on, just thought you should know." I mean really the letter says:

We have concluded that the regulatory reports provided by DB are of low quality, inaccurate and unreliable. The size and breadth of errors strongly suggest that the firm’s entire U.S. regulatory reporting structure requires wide-ranging remedial action. We also concluded that no progress was made in remediating prior supervisory concerns as the firm’s U.S. operations regulatory reporting process continues to be fragmented and suffers from weak or inadequate internal controls.

And then, like, tells them to fix it? After just saying that they never fix anything? You'd sort of think that the consequences of (1) terrible reports and (2) ignoring regulators' complaining letters would be somewhat stronger than another complaining letter sent privately to Deutsche Bank executives. Like, at least you'd think the Fed would escalate by publicly releasing the let -- ohhhhhh, right. Okay.

As for the substance of the letter, it is boring stuff, and some of it is dumb formality ("With regard to reports filed with the FRBNY, the firm has not developed a formal Accountability Policy and governance structure that clearly articulates roles and responsibilities for all areas responsible for the integrity of regulatory reports"), but I have plenty sympathy with the Fed here. This, for instance, isn't critically bad, but it is a really good thing for a regulator to be angry about:

Regulatory reporting information is compiled from a variety of applications and data sources, these data sources cannot be reconciled to report control totals. Additionally, multiple manual adjustments (in excess of 800 and totaling approximately $337 billion) are required to prepare the FR Y- 9C. These adjustments require substantial resources, lack transparency and do not allow for sufficient time to effectively review and analyze the data, prior to filing with the FRBNY.

It's not about preparing the FR Y-9C. It's about: Do you know what's going on in your organization? Can you find out quickly? If you need to make 800 manual adjustments to fill out a form, that doesn't bode well for how you manage risk.

Stock analysts are great: Eliot Spitzer.

Eliot Spitzer is now backing -- as an investor, board member, and guy going on CNBC to talk about -- an Israeli start-up called TipRanks that ranks stock analysts (and also bloggers) on the accuracy of their recommendations. This feels a little strange -- don't buy a stock just because an analyst with a good past performance recommends it, come on! -- but it also tells a useful story. If you are a government lawyer, as Spitzer was when he reached the Global Analyst Settlement, what you care about is conflict of interest: The best evidence you can have that an analyst is bad is an e-mail from him saying "hahahaha I don't believe this recommendation." If you are a private businessperson, as Spitzer is now, what you care about is quantifiable performance: The best evidence you can have that an analyst is good is a track record of Buy (Sell) recommendations on stocks that go up (down). Those things are not necessarily related: A terrible and corrupt analyst might always recommend stocks that he hates, but have a good track record anyway because he hates the wrong stocks. ("I have a bad gut instinct about this coin flip but am going to lie about it.") If you're buying stocks, would you want to follow that terrible and corrupt analyst's picks? Probably, right? Very much related is Cathy O'Neil on quantitative versus lawyerly views of wrongdoing.

Herbalife is terrible: Bill Ackman.

If you're interested in the Herbalife saga, obviously you should read Dan McCrum's reaction to Bill Ackman's presentation yesterday. He's impressed, saying that it "has fundamentally shifted the debate" by focusing on how "a pyramid scheme targeting the very poor can work," showing that Herbalife's "Nutrition Clubs are a way to both sustain and mask the structure of the pyramid designed to exploit the poor, and one at the heart of the company’s strategy." It's pretty persuasive but you run into the problem of: All multilevel marketing schemes are pyramidal, and they all encourage people to buy the product and recruit others. The question is not, "is this a way to sustain and mask the structure of a pyramid," but rather, "does it successfully mask the pyramid?"

American judges are terrible: Argentina.

Basically ever since the Supreme Court rejected Argentina's appeals, Argentina has been arguing, energetically, in the press and elsewhere, that it can't reach a settlement with holdout bondholders in time to avoid default on an interest payment due next week. So it's been pursuing a parallel strategy of (1) taking out ads calling New York federal judge Thomas Griesa corrupt and biased and evil and (2) politely asking Judge Griesa for an extension of the deadline to work something out with the holdouts. How is that going? Exactly as you'd expect:

U.S. District Judge Thomas Griesa in Manhattan today rejected arguments by a lawyer for Argentina who warned of “a very imminent risk of default” unless there’s a temporary suspension of his order that defaulted bondholders must be paid at the same time as holders of its restructured debt.

“I expect the parties and their lawyers to work really continuously,” Griesa told attorneys in his packed courtroom. “There is not a lot of time.”

On the other hand, "Judge Griesa said he would decide at a later time whether interest payments can be made to holders of Argentine bonds issued under Argentine, U.K. and Japanese law"; most people seem to think that there's a pretty good argument that they can be but, of course, if you were Judge Griesa you'd be pretty mad at this point, right? Here is Argentina's latest. Joe Cotterill has more here and here, and Anna Gelpern here.

Activists are terrible: Marty Lipton.

There's a long-running debate about whether activist hedge funds are good or bad for companies. Law professor Lucian Bebchuk is basically the leader of the side that thinks they're good, and poison-pill inventor 1 Marty Lipton is basically the leader of the side that thinks they're bad. A while back Bebchuk, with Alon Brav and Wei Jiang, wrote a paper measuring what they called the long-term effects of hedge fund activism, and found that companies targeted by activists outperformed their peers -- on operating-performance measures like Tobin's Q and return on assets, and on stock-price abnormal returns (capital asset pricing model alphas) -- over 3 and 5-year periods. It's pretty standard -- and thus, by its nature, uncertain and model-driven and debatable -- econometric stuff. If you just sort of reject econometrics, you might not believe in things like statistical significance (how significant is it really?) or peer group comparisons (how did you choose the peers?) or measuring Tobin's Q (maybe it doesn't capture real long-term value) or CAPM alphas (because "the number of treatments, estimations and assumptions going into producing their results are mind-numbing"). And so you might ignore their paper and choose to instead believe your "practical experience" and gut instinct that activist hedge funds are bad.

Here is a pretty amazing paper from the Institute for Governance of Private and Public Organizations doing just that. Here is Marty Lipton calling it an "academically rigorous, as well as common sense, refutation of Bebchuk's claims."

Insider trading is dumb: Me.

If you work in investor relations, and you prepare a press release with good or bad news about your client, don't buy or sell the client's stock before issuing the press release. Or do, I don't know, the world is complicated, but don't do it if your "illicit profits and avoided losses from insider trading in both companies totaled $11,776." Is that really worth it?

Things happen.

Evercore is expanding. Troll trolls. Michael Lewis claims another victim. Accountant runs triathlons. You can read the New Yorker for free now. 1 in 25 New Yorkers is a millionaire.

Corrects Judge Griesa's first name in fourth item.
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
  1. And my former boss.

To contact the author on this story:
Matthew S Levine at

To contact the editor on this story:
Toby Harshaw at

Before it's here, it's on the Bloomberg Terminal.