Money Stuff

Hedge-Fund Pay and Trader Lies

Also "21st century Glass-Steagall," blockchain Ponzis, proxy fights and embezzlement.

Hedge-fund pay.

Institutional Investor's Alpha's annual "Rich List of the World's Top-Earning Hedge Fund Managers" is out today, and as usual the best-paid hedge-fund manager in 2016 was Warren Buffett, who made about $13.7 billion for the year. He earned it: His fund -- Berkshire Hathaway Inc. -- returned 23 percent in 2016, versus about a 12 percent total return for the S&P 500 index. 

Actually wait no, Alpha's list starts with James Simons of Renaissance Technologies, who made only $1.6 billion. (He also had a good year.) Buffett is for some reason not on the list. Perhaps that is because he no longer manages a hedge fund? (Berkshire is a public company, and Buffett got out of the hedge fund business in 1969.) But Simons doesn't either; he retired at the end of 2010. Perhaps it is because Buffett's $13.7 billion in 2016 earnings wasn't actually "pay" at all; it's just appreciation on the Berkshire stock that he already owned? But that is also true for Simons, and for most of Alpha's list, which includes earnings on the managers' own capital in their funds as part of their "pay."

We talk about this every year. (Here are the versions for 2013, 2014 and 2015.) The "Rich List," as its title suggests, ranks rich hedge-fund managers. (And rich former hedge-fund managers still closely associated with their funds, like Simons, or like David Shaw, ranked number 9 this year, who founded D.E. Shaw Group but "is no longer actively involved in their day-to-day operations.") If you are very rich, and you run a hedge fund, and you keep billions of dollars of your own money in the fund (as one does), and it has positive returns -- even small positive returns -- then you will make a lot of money. But that is not compensation for your hedge-fund-managing, good or bad; it's return on capital. The best-paid job in the world, year after year, is being a billionaire.

But every year the discussion of the list suggests otherwise. "Even the lowest-ranking manager on Alpha magazine’s expanded top-50 list made more money in 2016 than any big United States bank executive," reports the New York Times, incorrectly. Lloyd Blankfein owns 2.7 million shares of Goldman Sachs Group Inc., which were up about $158 million in 2016; add $7 million of dividends and his $22 million in official compensation and you get about $187 million, putting him easily into the top 25. Jamie Dimon's numbers at JPMorgan Chase & Co. -- 10.4 million shares (including options), almost $210 million in capital gain, $19 million in dividends and $27 million in reported compensation -- put him at about $256 million; the cutoff for the top 10 hedge-fund managers is $410 million. Of course we don't usually think about a bank chief executive officer's stock price appreciation as compensation, but if a hedge-fund manager's returns on his own capital count as pay, then so should the bank CEO's.

Anyway the big picture is that pay, or "pay," was down this year, in line with performance generally being down. Only two managers -- Simons and Ray Dalio of Bridgewater Associates -- cracked a billion dollars, versus five last year. The cutoff for the top 50 was just $30 million dollars. "Some of the best-known names in the industry — including William A. Ackman, John A. Paulson and Edward S. Lampert — failed to make the list." A small percentage return on a lot of money is a lot of money, but a negative return on a lot of money is still negative.


A thing that apparently happens a lot, or at least used to happen a lot, is that bond traders lie to their customers about the price that they paid for bonds, in order to induce the customers to pay more for the bonds. "Ugh, it is killing me to sell you these bonds at 77, I paid 76.99 for them, I'm barely making any money," the bond trader will say, even though she really bought the bonds for like 74 and is making tons of money. You can tell that this happens a lot because the traders keep getting arrested, or sued by the Securities and Exchange Commission, or at least fined by the Financial Industry Regulatory Authority.

It happens so much, in fact, that the customers are catching on. Yesterday the SEC brought civil charges against the former heads of Nomura's commercial mortgage-backed securities desk, James Im and Kee Chan. On one trade, Chan allegedly told a customer ("Trader B") that he had bid 51-05 for some bonds and ended up buying them at 52, and that he would sell them to the customer for 52 plus an additional spread to compensate him for the work. (He had actually paid 51-01.) The customer agreed, but grumbled:

Expressing a concern that Investment Manager B would now be overpaying for the bond by paying Nomura "on top" of 52, Trader B responded: "cool ... i bet the 51-05 was already good ... u want some of it? Feel like we got smoked on this one." Showing again the importance to their negotiations of the bid price information misrepresented by Chan, Trader B went on to say that his partner wanted to modify the trade, stating to Chan: "between u and me basically [Trader B's partner] is certain u made up that u bid 51-05 already in order to get us to bid."

They were "certain" that Chan "made up" his bid! They knew that he was lying, but they paid him anyway. The SEC cites this as evidence that Chan's lies were material to the customer, but it sounds a bit like the reverse: It sounds like Chan's customers grudgingly accepted that they were in a used-car-dealing world in which dealers lie about their profit margins, and there wasn't much they could do about it. "im trying to get paid a scrap," Chan told a counterparty in another trade, sounding like a stereotype of a used car dealer. The customers may not have taken this sort of hyperbole all that seriously. They traded mortgage bonds all day; they perhaps had an inkling that some of the traders were lying.

On the other hand, they did ask, and Chan put them off by doing this:

A short while later, Chan sent an email to Trader B purportedly forwarding an earlier internal Nomura email from Chan that seemed to corroborate Chan's 51-05 bid. The email forwarded by Chan was a fake. In fact, Chan fabricated the phony email from an actual email that he had sent internally earlier in the day, altering the bid price from 51-01, which appeared in the actual email, to 51-05, the price that Chan told Trader B he had bid for the bond.

Where have I seen that before? Jesse Litvak, the former Jefferies LLC managing director who was the first person prosecuted for this sort of thing, was convicted of lying to customers and sentenced to two years in prison, but then had his conviction reversed on appeal and sent back for a new trial where he could argue to a jury that everyone lies to customers so it's no big deal. So he did, and he mostly convinced the jury -- except that they were bothered that he had altered a chat transcript to make his lies to a customer more convincing, and convicted him of fraud for that. He was re-sentenced to the same two years in prison. Meanwhile, Chan settled with the SEC for about $214,000 in fines and disgorgement (and a three-year industry ban), without admitting or denying the charges; Im is fighting the charges.

Incidentally, I say "bond traders lie to their customers," but all of the cases so far are about mortgage-backed securities. (Residential mortgages for Litvak, and most of the others so far; commercial mortgages here.) Corporate bond traders don't lie to their customers about how much they paid for bonds, or at least they don't get in trouble for it. There's a reason for that, and it's not the innate honesty of corporate bond traders. It's that corporate bond trades in the U.S. are generally reported to Finra's TRACE system, so their prices are publicly available. If a dealer buys a corporate bond for 50 and then tells a customer that she paid 52, the customer can just look on TRACE to see that she's lying. So you don't see a lot of lying, or successful lying anyway.

Meanwhile in commercial mortgage-backed securities, anything goes. The SEC's press release says that "Im and Chan operated under cover of an opaque CMBS secondary market to gain illegal trading profits," but the word "opaque" here pretty much just means "not TRACE eligible." If bond dealers were required to publicly report the prices of CMBS trades, then they'd have no opportunity to lie about them. You get the sense that that's what the SEC wants. But actually doing the notice-and-comment rulemaking for post-trade transparency in the CMBS market would be a complicated endeavor for the SEC. It's easier to do the rulemaking by enforcement, to mandate honesty by suing a few traders instead of mandating transparency with a rule.

Break up the banks!

The Glass-Steagall Act of 1933 was a law forbidding commercial banks from doing investment banking, and vice versa. It was more or less dead by the end of the 1990s, with big universal banks combining banking and securities businesses. After the financial crisis of 2008, many people argued -- on sketchy evidence -- that the repeal of Glass-Steagall contributed to the crisis, and that we needed some sort of "21st Century Glass-Steagall Act" to break up the universal banks and make them safer and less crash-prone. And since Donald Trump likes to say words that people like to hear, he has echoed those suggestions:

Trump, Treasury Secretary Steven Mnuchin and economic adviser Gary Cohn have all endorsed a “21st century” version of Glass-Steagall without explaining what they mean. In the interview this month, Trump told Bloomberg News that bank breakups were under consideration. “I’m looking at that right now as we speak,” he said.

But, while Trump likes to say popular words, he does not seem to care about their meaning, and so Max Abelson reports that no one thinks that his "21st century" approach to Glass-Steagall will involve breaking up the banks, or separating commercial and investment banking, or otherwise doing anything related to the Glass-Steagall Act. Instead it will just be about bank deregulation:

Tim Pawlenty, who runs the Financial Services Roundtable, a Wall Street lobbying group, doesn’t think the administration is really talking about Glass-Steagall when it refers to it.

“Following the president’s remarks on the topic, Gary Cohn clarified the administration’s view of a modern-day Glass-Steagall is a two-tiered approach to regulation in which smaller banks would receive some regulatory relief.” 

Hey great yes.

Executives at three of the six biggest banks were on the same page, saying they’d be happy if Trump loosened rules for smaller lenders and sold it as a new and improved Glass-Steagall.

Sure, right, improved.

On the same day Trump made his breakup threat, Sean Spicer, his spokesman, said a 21st century Glass-Steagall would allow smaller banks to “spend less time complying with unnecessary requirements.”

Uh-huh. Repealing Glass-Steagall was a deregulatory move, but bringing back a new Glass-Steagall will also be deregulatory. Meanwhile larger banks will also get regulatory relief in the form of a rollback of the Volcker Rule. It will all be Glass-Steagall! Glass-Steagall could mean taking principal trading activities out of commercial banks, or putting them back in, or whatever. "Glass-Steagall" means doing a thing about banks, and declaring it good. 

Blockchain blockchain blockchain.

I am a big fan of Ethereum, the smart-contract blockchain platform that has allowed a lot of innovation in ... well, Ponzi schemes? Think about it. The criteria for an early-adopter smart-contract application are pretty much:

  1. It benefits from the trustless automatic-execution properties of smart contracts on a distributed blockchain, meaning basically that it's a project where you can't trust anyone else; and
  2. It is purely financial: A smart contract in the shipping industry requires a boat, while your early-adopter contracts should really just be about cash flows.

A Ponzi scheme is perfect, and so, when we first talked about Ethereum, I went to its app store and quickly found a Ponzi scheme, a chain-letter scheme, and two pyramid schemes. (Later we talked about the Distributed Autonomous Organization on Ethereum, which was not a Ponzi scheme, but which did get hilariously hacked.) These schemes were all pretty open about what they were! One innovation of the blockchain is that you can just say that your Ponzi scheme is a Ponzi scheme, and people will buy in anyway, because, you know, innovation. And because at a certain point the sort of paranoid skepticism that brings you to trustless solutions like the blockchain circles back into naivety. 

The latest Ethereum Ponzi scheme is called PonzICO, and it is especially open about being a Ponzi scheme, but it comes with a charming white paper (on Medium or, much better, as a PDF):

The author proposed a continuous, tokenless ICO that dispenses with all the fanciful illusions pervading the space. He demonstrated the soundness of his approach with selective history, a hand-wavy graph, tables, self-referential citations, and vague commitments to show his smart contract code eventually. He anticipated no critiques to this approach, and fully commits to ignoring critics who emerge on social media by labeling them trolls.

"As a double bonus, if everyone stays irrational it might greatly enrich the author, providing future funding for blockchain performance art." The great art project of our age is to entirely collapse the distinctions between "fraud" and "performance art," so that one day mortgage-bond traders will be able to say "wait, no, I wasn't lying about bond prices to increase my bonus, I was performing a metafictional narrative about bond-price negotiations in order to problematize the underlying foundations of bond trading in late capitalism." 

Elsewhere: "Fed's Kashkari: blockchain has more potential than bitcoin itself." I mean, for comedy, sure.

A proxy fight.

I have no idea what is going on at Cypress Semiconductor Corp., but it's more fun that way. Just cast your eyes over this proxy filing from T.J. Rodgers, Cypress's former chief executive officer, who was pushed out in 2016 and who is now running a proxy fight against the board. "The Board’s lawyers have 'spun' my efforts to get at the truth on behalf of all Cypress stockholders as a 'personal vendetta' against executive chairman Ray Bingham," writes Rodgers. Then check out this press release from the board:

On April 24, 2016, during a dinner with two independent Board members and Cypress' outside counsel, Rodgers was asked to resign from his position as CEO. During that dinner:

  • Rodgers got out of his chair, leaned across the table and yelled at the Independent Directors:
  • "this is war"
  • "in a matter of weeks I will be back and you will be out"
  • "you Judas"

I don't know if this is true or what they are fighting about or who should win or anything. My point is only that all proxy fights should be like this. Like if you are going to mount an insurgent campaign against a corporate board of directors, you should first have to have dinner with the directors and yell at them to their faces, and then the proxy statements should contain conflicting accounts of what you said. 

An embezzlement.

Here is a story about a former Bank of America Corp. wealth manager who possesses the glorious name Palestine "Pam" Ace, but who nonetheless allegedly turned to crime:

Prosecutors claim Pam Ace from 2010 to 2015 approved 75 donations to basketball and educational programs and other organizations providing support to children with HIV/AIDS before insisting that about half of the money "be returned in order to ensure that Bank of America would continue to fund the organization," according to the statement.

Ace’s husband then allegedly threatened the recipients with "public humiliation" to extract the return of as much of the money as possible.

According to prosecutors, the money came out of Bank of America's marketing budget, and came back to Ace and her co-conspirators in the form of personal checks (what?) or "to a Bank of America account, to which the defendants had access." Honestly if I were a nonprofit extorted in this way I would mostly be ... confused? I have only an amateur interest in bribery, but my understanding is that the way it works is that you first negotiate the bribe, then give out your employer's money. Giving out Bank of America money to nonprofits and then demanding bribes seems kind of backwards.

People are worried that people aren't worried enough.

"Should We Fear the Stock Market’s Lack of Fear?" asks the Wall Street Journal's Streetwise column. "Fear Not the Fear of Complacency in Markets" responds the Wall Street Journal's Heard on the Street column. I am glad we have settled that, I think.

People are worried about unicorns.

Well Uber Technologies Inc. keeps being worrying. The latest bad news is that a federal judge has enjoined Anthony Levandowski -- an engineer whom Uber poached from Alphabet Inc. and who allegedly stole a bunch of secrets on his way out -- from working on Uber's self-driving cars. Things could always be worse, though: "Despite the judge’s ruling on Mr. Levandowski, Uber also had cause for celebration because its self-driving research program was not shut down, which would have been a more serious blow."

Things happen.

Debt Island: How $74 Billion in Bonds Bankrupted Puerto Rico. Researchers Identify Clue Connecting Ransomware Assault to Group Tied to North Korea. Hackers prime second classified US cyber weapon. A Whistle-Blower Tells of Health Insurers Bilking Medicare. Rich Retirees are Hoarding Cash Out of Fear. ValueAct’s Jeffrey Ubben Hands Reins to Protégé Mason Morfit. Goldman Sachs Sees Big Potential for Fintech in Brazil. The New Bond Workouts. For Many Companies, a Good Cyber Chief Is Hard to Find. Ferrari vending machine. Fyre Festival. Avocado toast. Dog ratings. Man rompers.

If you'd like to get Money Stuff in handy email form, right in your inbox, please subscribe at this link. Thanks! 

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

    To contact the author of this story:
    Matt Levine at

    To contact the editor responsible for this story:
    James Greiff at

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