(Bloomberg View) -- Quants.
I went to work at a derivatives desk at a big investment bank almost 10 years ago, and at the time there was sort of a standard model for how banks and trading firms acquired quants. (To be clear, I was not a quant.) Basically you'd go to a physics Ph.D. student and say: Look, if you stay in physics, you will work for years on some abstract problem, making only slow progress to some vague impractical goal, and also you will be poor. If you come to a bank, you will solve real problems every day, and you will be rich. And the physicists would scoot off to the banks, where they'd immediately stop doing physics. They'd do a different thing, called "quantitative finance." (Derivatives pricing, quantitative risk modeling, that sort of thing.) It uses some of the mathematical techniques and Brownian-motion intuitions of physics, but the problems and equations are different.
Nishant Kumar and Katia Porzecanski have an article about how trading firms acquire quants today, and it is subtly different. There is some superficial stuff (hackathons!), but overall the way that quant-focused firms like Winton, Two Sigma and Citadel recruit is much more academic than it used to be:
In September, Two Sigma will take over “The Bridge,” a space on the new Cornell Tech campus on Roosevelt Island in New York, where engineers and entrepreneurs will work.
The hedge fund staff will collaborate with Cornell students and professors on machine learning and data science projects, creating a pipeline of academic talent to the firm.
"By having our employees able to sit in the same room, share the same coffee with the academics, we find that there's a fantastic osmosis," says the chief executive officer of Man Group. "It's a bit like CERN where scientists worked together to discover the Higgs particle," says the head of human capital at Winton. The appeal is not finance as a more practical, faster-moving alternative to academia. The appeal is finance as a continuation of academia. The goal is to collapse the differences between finance and science, rather than highlight them.
I suspect that part of the difference is that the vogue in finance is for data science, while classic quantitative finance -- quantitative risk management, derivatives structuring -- is perhaps less popular than it was in 2007. And while "quantitative finance" is a different field from "physics" that happens to use some of the same math and some of the same people, "data science" is "data science," regardless of what specific problems it is applied to. We've talked before about Renaissance Technologies, the wildly successful hedge fund that mostly hires people without finance backgrounds, perhaps because (in my words) investing is "a trivial application of broader data-science principles, best addressed by people who were trained on harder and more interesting applications."
The old model of quantitative finance made finance into a sort of science, and then recruited people from the physical sciences to build its models and run its experiments. But moving from physics to finance meant abandoning physics. The new model of machine-learning-based finance makes finance into just a set of problems in data science. You can run a hedge fund without ever abandoning data science, or the university setting, or your academic colleagues, or your self-conception as a scientist.
Insider trading.
Mathew Martoma, you may recall, is the former SAC Capital Advisors analyst who went to prison for insider trading on the results of some drug trials. That happened way back in 2014. Since then, there have been some developments in insider trading law. The U.S. Court of Appeals for the Second Circuit revived the Supreme Court's "personal benefit" test in its Newman decision: It's not enough for prosecutors to prove that you got a tip from a corporate insider, the court held; they also have to prove that the insider got some benefit from the tip. And then last year, the Supreme Court pared that decision back in its Salman decision, holding essentially that if the insider tips a close friend or family member, then that's enough to satisfy the personal benefit test.
The interesting battles, now, will be about tippers and tippees who aren't close friends. We talked last week about an appeals court decision finding that a golf-buddy relationship was sufficient to satisfy the personal benefit test. Eventually, an investor-relations employee at a company is going to give color on upcoming earnings to a hedge-fund analyst as part of his job, and prosecutors will claim that he got a personal benefit because he occasionally made small talk with the analyst and they were therefore friends, and the cycle will repeat. The basic intuition is that information passed along out of self-interested personal corruption is illegal, while information that is passed along out of corporate investor-relations purposes is not. But this is a hard distinction to police, and anyway prosecutors don't really believe it, because they want insider trading law to guarantee a level playing field for investors.
Meanwhile, though, Mathew Martoma is appealing his case on the theory that the doctors who tipped him off to the drug trial results were just business associates of his, not close friends. "The Second Circuit will have to decide how much of Newman's analysis survives when it moves into a murkier area that mixed together business dealings and some measure of personal friendship," writes Peter Henning.
Will it? I mean, I buy Martoma's argument that his relationships with the doctors were not close friendships, and that casual work friendships should not give rise to insider trading liability. But also ... Martoma ... just ... paid them? Both of the doctors who tipped Martoma had been hired by SAC as "expert network" advisers, and were paid thousands of dollars for the information they provided. One tipster-doctor, Sidney Gilman, got $70,000 for his consulting; the other got $1,500 per hour.
That seems like a pretty clear personal benefit? Martoma argues in his appeal that Gilman "himself testified that he was paid nothing 'in exchange for' the disclosure of the efficacy data on which Martoma allegedly traded," but the prosecutors seem rightly skeptical:
This is a debater's point, and not a good one: while Gilman did not submit a bill for the precise meeting during which he tipped Martoma, the meeting occurred in the context of a lucrative consulting arrangement that was paying him thousands of dollars. And the reason he did not submit a bill for the meeting at issue was that doing so “would [have] be[en] tantamount to confessing that [he] was feeding ... [Martoma] inside information.”
It can't seriously be the law that if you pay someone thousands of dollars for information about a company, and he gives you inside information about the company, it's not insider trading if you weren't also close friends. But the law is so muddled right now that smart people seem to think that that's a plausible argument.
Meanwhile, here are Kevin Haeberle and M. Todd Henderson:
In our recent article, we set out an intermediate market-based approach toward achieving the optimal level of corporate disclosure: Create a transparent market for early-access rights to corporate information. In this market, firms could sell access to information that they will later release publicly. For example, when announcing news to the public at 1:00 p.m., they could offer a well-advertised early peek—say, starting at 11:00 a.m.—to anyone willing to pay the market price for it. So long as firms had to make any selectively released disclosure products with material information available to the public in due time, market supply of early-access products and information-consumer demand for the same could generate improved public disclosure.
Don't hold your breath. What I like about this is that you can sort of divide investors into three categories:
- High-speed traders who have the technology and skill to plausibly believe that they can react faster to news than the rest of the market;
- Low-speed traders who understand that they do not have the technology or skill to react faster to news than the rest of the market, and so do not attempt to trade rapidly in response to news; and
- Idiots.
I realize that this oversimplifies, but still. This plan would allow companies to extract some value from category 1. It wouldn't affect category 2. ("It is unlikely that those merely investing for a long-term, market-wide risk premium would pay for early access to public-company information," so "they could avoid the periods in which speculating pros are battling it out to determine the import of new information for securities prices, and then safely return to the market to complete their non-time-sensitive trading.") And it would at least make it really clear to category 3 that, oh yes, someone else has better information than you.
Snap cracks after pop.
Snap Inc., whose stock popped Thursday and Friday after it priced its initial public offering Wednesday night, cracked yesterday, closing at $23.77. That's below where it opened for public trading on Thursday ($24), though still well above the $17 IPO price.
What has changed? Well, one thing that happened is that research analysts keep coming out against Snap: "By Monday, five of the seven analysts who cover the company had a sell rating on it while two said hold." Presumably all the buy ratings will come in when the underwriter banks -- who are currently in a post-IPO quiet period -- are allowed to release research, but there is no guarantee of that. Banks are free to underwrite stocks that their analysts don't like, and aren't allowed to push their analysts to give good ratings just to please their investment-banking clients. Maybe all the underwriting banks will come out with sell recommendations. I wouldn't ... count on it? They'd have had a tougher time selling the deal if the analysts were bearish. But it's within the realm of theoretical possibility.
Another possibility: When a company first IPOs, it is very hard to borrow shares of its stock, so you can't really sell it short. Over time, investors actually get their hands on shares, brokers figure out where the shares are, a stock-borrow market develops, and people are able to sell short. The absence of shorting can keep up the stock for a while: If you think Snap is underpriced, you can buy it, but if you think it's overpriced, you can't short it. But that eventually ends. Snap's IPO is scheduled to settle today -- that is, today is when the shares are actually supposed to be delivered by the underwriters to the investors -- and perhaps we'll start to see more shorting.
Until then, teens are buying Snap stock.
The London Whale.
I enjoyed this profile of former "London Whale" and current grumpy unemployed man Bruno Iksil:
Cedric Lespiau, the head of credit index and options trading at Societe Generale between 2007 and 2012, says he tried to warn Iksil about the danger. Lespiau says that after he read chatroom messages from brokers and dealers that referred to “the big guy,” he told Iksil that his positions left him vulnerable.
But, Lespiau says, Iksil “was on another planet”, effectively in denial about the potential danger. Iksil says he was not in denial, but that he did not see the point in worrying about it. “I was not a typical hedge fund manager that they could squeeze out,” he says.
A London-based banker who counted Iksil as a client from 2008 to 2012 says some bankers were leaking information about Iksil's trades to cultivate relationships with other clients, such as hedge funds. Iksil says that he heard dealers were sharing his positions, but thought his transparency in the market wasn't something to worry about.
Iksil complains a lot now about the "London Whale" name, but at the time he was ... just totally cool with everyone knowing his positions and trading against him? Of course he was right that he was not "a typical hedge fund manager": He was making long-term hedging trades, and had JPMorgan Chase & Co.'s massive balance sheet behind him. He could last a really long time before he cracked. But that made it worse when he did.
Also:
“The typical traders, they did not like me,” he says. “They did not like the way I thought. I don't like this Liar's Poker thing, it's totally stupid. But there are guys who love that. And I'm not this kind of guy.”
I feel like you see a lot of that sort of thing after blow-ups. Rogue traders, Libor kingpins, guys who lose a billion dollars: Their colleagues rarely describe them as "oh yeah, he was a pretty stereotypical trader bro, really into pranks." It's always "he was quiet and didn't fit in with the other traders." If you're a bank and are worried about operational risk, should you limit your hiring to trader stereotypes?
Meanwhile: "Fed May Take Legal Action Against ‘London Whale.'"
ISDA basis.
Here's a delightful Alphaville guest post by Marcello Minenna about what the credit-default swap market for European sovereign debt says about fears of a euro breakup. The idea is that there are two kinds of European sovereign CDS contracts: ones written using the International Swaps and Derivatives Association's 2003 Credit Derivatives Definitions, which don't protect against currency redenomination, and ones under the 2014 ISDA definitions, which do. If you are worried about euro-area sovereigns dropping out of the euro, you should pay more for the 2014 contract than for the 2003 one. This difference is called the "ISDA basis," and it used to be small, but now it is larger, suggesting that more people are worried.
The key lesson is that when you are trading CDS you are betting on contract interpretation at least as much as you are betting on credit. What matters in a CDS trade is not just the general idea of "credit protection," but the specific workings of the instrument that determine when and how you'll actually get paid. Using Bloomberg data, I see 5-year French CDS trading at around 30 basis points, with the 2003 definitions, or around 60, with the 2014 definitions. If you're buying 2014 France CDS to hedge against the risk of a euro breakup, half of your premium is going toward getting the right contract drafting.
Is there a too-big-to-fail premium?
Not exactly, suggest Bernadette Minton, René Stulz and Alvaro Taboada:
Many argue that large banks receive subsidies from the regulatory safety net, so they should be worth more and their valuation should increase with size. Instead, using a variety of approaches, we find (1) no evidence that large banks are valued more highly, (2) strong cross-sectional evidence that the valuation of large banks falls with size, and (3) strong evidence of a within-bank negative relation between valuation and size for large banks from 1987 to 2006 but not when the post-Dodd-Frank period is included in the sample. The negative relation between bank value and bank size for large banks cannot be systematically explained by differences in ROA or ROE, equity volatility, tail risk, distress risk, and equity discount rates. However, we find that banks with more trading assets are worth less. A 1% increase in trading assets is associated with a Tobin's q lower by 0.2% in regressions with year and bank fixed effects. This relation between bank value and trading assets helps explain the cross-sectional negative relation between large bank valuation and size.
People are worried about bond market liquidity.
"The bond benchmark continues to tip to swaps," worry Lawrence Kreicher, Robert Neil McCauley and Philip Wooldridge in the Bank for International Settlements Quarterly Review:
Over a period of two decades, the benchmark in bond markets, as in money markets before them, has tipped from government rates to private swap rates. Developments since the Great Financial Crisis of 2007-09 slowed this shift but did not stop it.
I guess that's not a bond market liquidity worry, or even a worry really. Meanwhile my Bloomberg Gadfly colleague Lisa Abramowicz is worried that long-term traders and short-term traders have different views on Treasury bonds.
Things happen.
Standard Life-Aberdeen Deal Awkwardly Places Chairman Between Barclays, Lloyds. As Regulations Change, Companies Grapple with Accountant Shortage. State Street to Start Voting Against Companies That Don't Have Women Directors. Desperate businessman commits suicide after losing $300K in investment scam: ‘The house always wins.' Where are the world's dollar deposits coming from? Is the Fed's communication ... too clear? SEC to Decide This Week if Bitcoin Could Become the Next ETF Star. Three common misconceptions about smart contracts. Credit rating agency reform is incomplete. "Borrowing is not the opposite of saving." MIT Technology Review's 10 Breakthrough Technologies of 2017. Here's how a scientist grows lifelike human ears on apple slices. New Peeps-Flavored Oreos Reportedly Turning People's Poop Pink. Scientists make step towards humans guiding robots by telepathy. "Soon we will stare into the void adjacent to one another, but not together, in Conference Room B."
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.
Matt Levine is a Bloomberg View columnist. He was an editor of Dealbreaker, an investment banker at Goldman Sachs, a mergers and acquisitions lawyer at Wachtell, Lipton, Rosen & Katz and a clerk for the U.S. Court of Appeals for the Third Circuit.
To contact the author of this story: Matt Levine at mlevine51@bloomberg.net.
To contact the editor responsible for this story: James Greiff at jgreiff@bloomberg.net.
For more columns from Bloomberg View, visit http://www.bloomberg.com/view.
Essential Business Intelligence, Sharp Market Insights, Practical Personal Finance Advice, Daily Fuel, Gold and Silver Prices and Latest Stories — On NDTV Profit.
