The Quants: How a New Breed of Math Whizzes Conquered Wall Street and Nearly Destroyed It (28 page)

Over the next decade, Taleb, wealthier than he’d ever dreamed he’d become, bounced from firm to firm, at the same time earning a Ph.D. from the University of Paris Dauphine, writing a textbook on option trading, and working as a pit trader at the Chicago Mercantile Exchange. In 1999, he started teaching graduate courses in finance at NYU, at the same time launching a hedge fund called Empirica for its focus on empirical knowledge.

By the time of Chriss’s wedding, Taleb had gained a reputation as a gadfly of the quants, constantly questioning their ability to beat the market. Taleb didn’t believe in the Truth. He certainly didn’t believe it could be quantified.

Due in part to his experience on Black Monday, Taleb believed that markets tended to make moves that were much more extreme than had been factored into quantitative models. As a teacher of financial engineering at NYU, he was well aware of the proliferation of models that attempted to take account of extreme moves—the “jump diffusion” model that allowed for sudden leaps in price; the tongue-twisting “generalized autoregressive conditional heteroscedasticity model,” or GARCH, which doesn’t look at prices as a coin flip but rather takes into account the immediate past, and allows for feedback processes that can result in sudden jumps that create fat tails (Brownian motion with a kick); and a number of others. Taleb argued that no matter what
model the quants used—even those that factored in Mandelbrot’s Lévy fat-tailed processes—the volatility in market events could be so extreme and unpredictable that no model could capture it.

The conversation at the table was cordial at first. Then people started noticing Taleb getting agitated. His voice was rising, and he was pounding the table. “It is impossible,” he shouted at Muller. “You will be wiped out, I swear it!”

“I don’t think so,” Muller said. The normally calm and collected Muller was sweating, his face flushed. “We’ve proven we can beat the market year after year.”

“There is no free lunch,” Taleb boomed in his thick Levant accent, his forefinger wagging in Muller’s face. “If ten thousand people flip a coin, after ten flips the odds are there will be someone who has turned up heads every time. People will hail this man as a genius, with a natural ability to flip heads. Some idiots will actually give him money. This is exactly what happened to LTCM. But it’s obvious that LTCM didn’t know shit about risk control. They were all charlatans.”

Muller knew when he was being insulted. LTCM? Hardly. PDT could
never
melt down. Taleb didn’t know what he was talking about.

At the end of the day, he didn’t care about Taleb. He knew he had alpha. He knew the Truth, or a respectable slice of it. But he still didn’t want to trade every day. There was more to life than making money, and he’d already proven he could do that in spades. He became more serious about his music and about poker.

In 2004, Muller pocketed $98,000 in a tournament on the World Poker Tour, often bringing his golden retriever to the table as a tail-wagging good-luck charm. When he won the Wall Street Poker Night challenge in March 2006, beating Cliff Asness in the final round, he didn’t collect any money, but he did gain a healthy dose of bragging rights over his fellow poker-playing quants.

Once or twice a month, Muller, Weinstein, Asness, and Chriss, among other top-tier quants and hedge fund managers, would meet in ritzy New York hotels for private poker games. The buy-in was $10,000, but the pots were often much higher.

The money was chump change to all of them. It was all about the game: who knew when to raise, when to fold, when to bluff like there
was no tomorrow. Asness loved playing, and he hated it. He couldn’t stand folding, taking the small, incremental losses so essential to success at poker. He was too competitive, too aggressive. But he knew that the only way to win was to fold until he had a hand he could really believe in, until the odds shifted in his favor. But it seemed like he never got that hand.

Muller, however, had mastered the art of knowing exactly when to fold, when to raise, when to go all in. He never lost his cool, even when he was down. He knew it was only a matter of time before he was back on top. The quant poker games lasted late into the night, at times stretching into the following morning.

In 2006, Muller took the PDTers on a ski trip to an exclusive ski resort out west, flying the gang on a NetJets private plane. His treat. It would be one of the last few such trips they would make in years. A credit crisis brewing on Wall Street would put an end to such carefree jaunts. But that was a worry for another day.

Muller, meanwhile, was getting restless. Playing endless rounds of poker, hiking exotic trails in Hawaii, kayaking in Peru, shooting off on private jets to the Caribbean, dating models—it was all fun, but something was missing: trading, making millions in the blink of an eye, watching the winnings tick up like a rocket. He had to admit it. He missed it.

Muller decided he wanted back in. He had a steady girlfriend once again and was thinking about settling down. Plus PDT’s returns weren’t what they used to be. It had put up just single-digit gains in 2006 as a flood of copycats plowed into stat arb strategies, making it harder to discover untapped opportunities. Morgan’s higher-ups wanted more. Muller said he could deliver.

A power struggle over control of PDT ensued. Shakil Ahmed, who’d been running PDT for the past seven years, quit the firm, outraged that Morgan would hand over the reins to their absentee leader. He soon took a job as head of quant strategies and electronic trading at Citigroup. Vikram Pandit, his former boss, had recently taken charge at Citi after Chuck Prince left in disgrace amid massive subprime losses. Pandit was quick to hire Ahmed, long considered one of the secret geniuses behind PDT.

Back at Morgan, Muller was on top again at his old trading outfit. He had bold plans to expand the operation and increase its profits. Part of his plans included juicing returns by taking on bigger positions. One portfolio at PDT with the capacity to take more risk was the quant fundamental book, longer-term trades based on a stock’s value, momentum—AQR’s bread and butter—or other metrics used to judge whether a stock will go up or down. Such positions were typically held for several weeks or months, rather than the superfast Midas trades that usually lasted a day or less.

“They skewed the book much more toward quant fundamental,” said a onetime PDTer. “They basically turned a large part of PDT into AQR.” The size of the book grew from about $2 billion to more than $5 billion, according to traders familiar with the position.

Ken Griffin, who ran strategies similar to PDT’s, wasn’t overjoyed by Muller’s return. He was overheard telling Muller that he was sorry to hear he’d come back—a typical double-edged dig by Griffin. Muller took it as a compliment. He was eager to get back in the mix of things, eager to start making money. Big money.

He wouldn’t have much time to enjoy himself. Just months after he returned, Muller would face the biggest test of his career: a brutal meltdown that nearly destroyed PDT.

ASNESS

On November 13, 1998, shares of a little-known company called
Theglobe.com
Inc. debuted on the Nasdaq Stock Market at $9 apiece. Founders of the Web-based social networking site were expecting a strong reaction.

The frenzy that greeted the IPO defied all expectations and common sense. The stock surged like a freight train, hitting $97 at one point that day.
Theglobe.com
, formed by Cornell students Stephan Paternot and Todd Krizelman, was, for a brief moment, the most successful IPO in history.

A few days earlier, EarthWeb Inc., perhaps feeling the force of gravity, merely tripled in its IPO. Investors gobbled up EarthWeb’s shares despite the following warning in its prospectus: “The company
anticipates that it will continue to incur net losses for the foreseeable future.”

A few months before the dot-com IPO frenzy began, LTCM had collapsed. Alan Greenspan and the Federal Reserve swept in, orchestrating a bailout. Greenspan also slashed interest rates to salve the wounds to the financial system left by LTCM’s implosion and flood the system with liquidity. The easy money added fuel to the smoldering Internet fires, which were soon raging and pushing the tech-laden Nasdaq to all-time highs on an almost daily basis.

While minting instant millionaires among dot-com pioneers, this series of unlikely events proved a disaster for AQR, which had started trading in August 1998. Asness’s strategy involved investing in cheap companies with a low price-to-book-value ratio, while betting against companies his models deemed expensive. In 1999, this was the worst possible strategy in the world. Expensive stocks—dot-com babies with no earnings and lots of hot air—surged insanely. Cheap stocks, sleepy financial companies such as Bank of America, and steady-as-she-goes automakers such as Ford and GM were stuck, left in the blazing wake of their futuristic New Economy betters.

AQR and its Goldman golden boys were hammered mercilessly, losing 35 percent in their first twenty months. In August 1999, in the middle of the free fall, Asness married Laurel Fraser, whom he’d met at Goldman, where she was an administrative assistant in the bond division. As AQR’s fortunes plummeted, he complained bitterly to her about the insanity of the market.
What is wrong with these people? They are so monumentally stupid. Their stupidity is killing me
.

Asness believed his strategy worked because people made mistakes about value and momentum. Eventually they wised up, pushing markets back into equilibrium—the Truth was restored. He made money on the gap between their irrationality and the time it took them to wise up.

Now investors were acting far more stupidly and self-destructively than he could possibly have imagined.

“I thought you made money because people make mistakes,” his wife chided him. “But when the mistakes are too big, your strategy
doesn’t work. You want this Goldilocks story of just the right irrationality.”

Asness realized she was right. His Chicago School training about efficient markets had blinded him to the wilder side of human behavior. It was a lesson he’d remember in the future: People could act far more irrationally than he’d realized, and he had better be ready for it. Of course, it’s impossible to prepare for every kind of irrationality, and it’s always the kind you don’t see coming that gets you in the end.

By early 2000, AQR was on life support. It was a matter of months before it would have been forced to shut its doors. Asness and company had coughed up $600 million of its $1 billion seed capital, in part due to investors pulling out of the fund. Only a few loyal investors remained. It was a brutally humbling experience for Goldman’s wonderquant.

Adding to the misery for AQR’s leading lights was Goldman’s highly lucrative initial public offering. It was too easy for Asness to do the math in his head. By leaving when he had, he’d missed out on a fortune. His hedge fund was on the brink of disaster. Worthless dotcom stocks were sucking in obscene sums. The world had gone mad.

His response? Like any good academic, he wrote a paper.

“Bubble Logic:
Or, How to Learn to Stop Worrying and Love the Bull” is a quant’s cri de coeur, Asness’s protest against the insanity of prices ascribed to dot-com stocks such as Theglobe.

The stock market’s price-to-earnings ratio hit 44 in June 2000—more than doubling in just five years and triple the long-term average. The title’s nod to Stanley Kubrick’s black-humor satire of the Cold War,
Dr. Strangelove, or, How I Learned to Stop Worrying and Love the Bomb
, gives a clue to Asness’s dark mood as he banged away at “Bubble Logic” late at night from the confines of AQR’s offices near Rockefeller Center (the fund later moved to Greenwich, Connecticut). The bubble was about as welcome to Asness as an atomic bomb. AQR has “had our assets handed to us,” he writes in the introduction. “Have pity on a partially gored bear.”

“Bubble Logic” began by making a rather startling argument: the
market of early 2000 was not like the market of the past. Of course, that was exactly what the dot-com cheerleaders were claiming. The economy was different. Inflation was low. Productivity had surged thanks to new advances in technology, such as laptops, cell phones, and the Internet. Stocks should be given higher values in such an environment, because companies would spit out more cash.