The Trader's Inherent Vice
Malt Liquidity 121
I must admit that I take a particular delight in hearing about rogue traders, which, to be fair, is a much more enthralling term than the mundane reality deserves. But inherently, every good trader does want to (or should want to) bet more.
I’ve highlighted the peculiar risk/reward that comes with institutional trading before
Take, for example, the institutional trading activities that led to the Volcker rule — prior to it, traders at banks were regularly (implicitly) allowed to bypass leverage limits so long as they made money. The upside? Make a huge trade, collect a massive cut as a bonus. The downside? No big deal! Get fired for violating risk policy, and maybe take gardening leave and find a new job. How about the bank though? The upside is correlated — maybe you’re making money on the trade that you wouldn’t have made otherwise, and have to pay some of it off as a bonus — but on the downside, your risk limits have been violated, which might lead to severely unexpected losses, and you’ll get fined/sanctioned for mismanaging consumer deposits.
but I’ve never really waxed poetic on the topic of rogue trading as I’m wont to do at a bar. I’ve certainly mentioned Kerviel before (and have found out IRL that SocGen people really don’t like the mention of his name), but I haven’t shown Bruno Iksil the same love,
In February 2012, hedge fund insiders such as Boaz Weinstein of Saba Capital Management[10][11] became aware that the market in credit default swaps was possibly being affected by aggressive trading activities. The source of the unusual activity turned out to be Bruno Iksil, a trader for JPMorgan Chase & Co. Market-moving trades by the bank's Chief Investment Office had first been uncovered in June 2011 by Dan Alderson, a reporter at trade journal Creditflux, which reported on anomalies in CDX HY index tranche pricing dynamics caused by Iksil's trading activity.[12] The same journal reported on further tranche trading activity by the JP Morgan unit two months later. By 2012, heavy opposing bets to his positions had been made by traders, including another branch of JPMorgan, who purchased the derivatives that JPMorgan was selling in high volume.[13][14] JPMorgan denied the first news reports, with Chairman and CEO Jamie Dimon calling it a "tempest in a teapot."[15][16] Major losses of $2 billion were reported by the firm in May 2012 in relation to these trades. By this point, the issue was being investigated by the Federal Reserve, SEC, and FBI.[17]
who was given the iconic title of “the London Whale” over this incident. A personal albeit little-known favorite of mine is the oil futures drunk trading incident:
On 29 June 2009, after spending a weekend drinking heavily, Perkins returned to London from a company-sponsored golf weekend and continued drinking at around midday. Around this time, he made some trades, which he said were for a client. According to Perkins, later that day he suffered an alcohol-induced blackout. Beginning at 1:22 a.m. (UTC+1) on 30 June 2009, while still blacked-out, Perkins traded 7 million bbl (1.1 million m3), worth $500 million (£340 million); at times this represented 69% of the volume of oil then being traded[1][2] and ten times the average trade volume. He made his last trade at 3:41 a.m., approximately two and a half hours later.[3]
…These unauthorised trades caused the price of Brent Crude oil to rise by over $1.50 a barrel (equivalent to $2.05 in 2022) within a short period of time, a trend generally associated with major geopolitical events, before dropping rapidly. As a result of the trading, PVM Oil Futures suffered losses of almost $10 million and Perkins was dismissed
My bemusement stems from the fact that the rogue trader operates as kind of a fourth-wall break (though these don’t come from Margot Robbie in a bathtub) — that behind this glossy veneer of math and risk management and trillions of dollars sloshing around, it’s not not gambling, ya know? On a baseline level, we all have to admit that there’s a “sexy” side to speculating (on top of the shitloads of adrenaline that comes with it) as evidenced by the high-flying reputation that trader used to carry pre-2008, so thank god speculation is also necessary for the growth of civilization.
Alas, human traders don’t really exist in the same capacity anymore. I certainly know that human discretion and click-trading is still an ongoing institutional practice, but the general de-emphasis of trading in favor of more predictable profit in wealth management as a whole has largely curtailed the availability of trading jobs which naturally tilts an employee’s personal risk tolerance downwards when it comes to bypassing limits. With so much being automated nowadays, trading failures are probably more inclined to look like Power Peg,
At Knight, some new trading software contained a flaw that became apparent only after the software was activated when the New York Stock Exchange (NYSE) opened that day. The errant software sent Knight on a buying spree, snapping up 150 different stocks at a total cost of around $7 billion, all in the first hour of trading.
Under stock exchange rules, Knight would have been required to pay for those shares three days later. However, there was no way it could pay, since the trades were unintentional and had no source of funds behind them. The only alternatives were to try to have the trades canceled, or to sell the newly acquired shares the same day…Once it was clear that the trades would stand, Knight had no choice but to sell off the stocks it had bought. Just as the morning’s buying rampage had driven up the price of those shares, a massive sale into the market would likely have forced down the price, possibly to a point so low that Knight would not have been able to cover the losses.
which is possibly the most famous trading failure story in HFT. Certainly these stories are not made public as often anymore.
And yet! We have an absolute blast of a story coming out of none other than Two Sigma, a storied firm that helped push forward the current regime of quants and code. What does an aspiring rogue trader do when all the trades are made by models and code? Well, you change the models:
A researcher at Two Sigma Investments adjusted the hedge fund’s investing models without authorization, the firm has told clients, leading to losses in some funds, big gains in others and fresh regulatory scrutiny.
The researcher, Jian Wu, a senior vice president at New York-based Two Sigma, was trying to boost his compensation, Two Sigma has told clients, without identifying Wu. He made changes over the past year that resulted in a total of $620 million in unexpected gains and losses, according to people close to the matter and investor letters.
First off, a VP in finance does not connote the same level of authority as, say, a political VP. It’s a highly compensated senior title, sure, but Goldman Sachs had over 10k VPs a decade ago. So when I see this title combined with what little I do know about building and scaling software systems and models, I’m shocked that someone could alter a model actually affecting PnL at this scale and have it go unnoticed in a firm with over a thousand employees. After all, your name is attached to every alteration and push to the overarching model, and these generally have to be reviewed unless maybe you have top-tier branch access, which a senior VP simply wouldn’t. But it gets weirder:
Wu’s changes led to gains of $450 million in total for some Two Sigma funds—including those in which the firm’s own executives and employees invest, as well as those available to clients. But they also led to a total of $170 million in losses for other funds compared with how they otherwise would have fared—losses largely borne by clients. Two Sigma has made them whole….
Two Sigma’s top executives this summer became aware of Wu’s changes because they resulted in higher than expected correlations between some of the firm’s trading models. The trail pointed to Wu, who made the changes in two stages over the past year.
Usually rogue traders are caught because they’re pushing around markets in a ham-fisted manner and/or losing gobsmacking amounts of money, where the losses become too large to hide. This is a very quant-age way to notice rogue trading strategies — they found that some strategies performed a little too similar to others which probably threw off a different risk metric (as too much correlation across assets and strategies would change how a lot of overall firm risk metrics might compute.) It’s particularly amusing because a quant trading firm is the last place in finance where “look, we made money at the end of the day” is an argument that will save you. Even after making the investors whole, they made $280 million extra (noninclusive of whatever fine they’ll have to pay)! The change arguably worked! And yet, that’s not how the firm works, internal control issues aside.
People familiar with the situation said Wu was trying to improve the firm’s performance, which would have benefited his career and potential pay.
Of course, the more things change, the more things stay the same. You can put up whatever exquisite altruistic facade you want in your corporate communications, but people work in finance to make money, and they always want to make more of it until they leave to play philosopher, politician, philanthropist, or whatever you call what Ray Dalio does at Burning Man. I can certainly see why it would be maddening to observe a certain strategy working in one part of the firm that makes your clients money and be banned from using it in a part of the firm that makes you money directly because of “the correlation.” Such is the curse of statistics — they apply at scale and in the long run, but not in a single instance. Some trade that violates every internal risk metric might be an absolute banger, but the risk metrics are supposed to keep the firm solvent and preserve the provable edge over the long run rather than maximize the immediate profit. And if you’re a quant trading firm, this is all your employees are hired to think about.
There’s two sides to one of my sayings: being right isn’t necessarily being profitable, but being profitable isn’t necessarily a sign that you were right. I guess Wu found out the hard way.