The Quants: How a New Breed of Math Whizzes Conquered Wall Street and Nearly Destroyed It (35 page)

“You know what’s going on?”

The answer was always the same: “No. You?”

Rumors of corporate collapses were making the rounds. Banks and hedge funds were reeling from their exposure to toxic subprime mortgages. Countrywide Financial, some said, was imploding and looking in desperation for a white knight, such as Warren Buffett’s Berkshire Hathaway or Bank of America. But no one wanted anything to do with the distressed mortgage lender.

Inside his office, Asness again stared grimly at his computer screen. Red numbers washed across it. He didn’t know what to do. His greatest fear was that there was nothing he could do.

Outside, people noticed the closed blinds on the boss’s office. It was unusual and a bit spooky. Asness always had an open-door policy, even if very few people used it. Employees figured Asness couldn’t stand the idea of his employees peeking in through the window to see how the big guy was taking it.

The registration statements for AQR’s initial public offering were ready and waiting to be shipped off to the SEC. Indeed, Asness was set to make the big announcement about his plans later that month, making headlines in all the important papers. But now, the IPO and all the money it would have spun off were getting more distant by the second, a distance measured by the tick-by-tick decline of Absolute Return, as well as a number of other funds at AQR that were getting pounded by the mysterious downturn.

Several blocks away from AQR’s office, Michael Mendelson, head of global trading, was in line at the local Greenwich Subway sandwich shop. He glanced at his BlackBerry, which came equipped with a real-time digital readout of all of AQR’s funds. His jaw dropped. Something bad was happening. Something horrible.

A longtime Goldman Sachs veteran, mastermind of the bank’s elite high-frequency trading outfit, Mendelson was one of the brightest thinkers at AQR, and one of the first people Asness would call when he needed answers about a trade going awry. He knew instantly that something needed to be done fast to stem the bleeding.

He raced back to AQR’s office and huddled with several of AQR’s top traders and researchers, including Jacques Friedman, Ronen Israel, and Lars Neilson. After determining that a huge deleveraging was taking place, directly impacting AQR’s funds, they marched to Asness’s office.

“It’s bad, Cliff,” Mendelson said, stepping into the room. Friedman, Israel, and Neilson followed close behind. “This has the feel of a liquidation.”

“Who is it?” Asness said.

“We’re not sure. Maybe Global Alpha.”

“Oh God, no.”

Since Asness had left Goldman in 1998, Global Alpha had been run by Mark Carhart and Ray Iwanowski, alums, like Asness, of the University of Chicago’s finance program, and Fama protégés. Under their guidance, Global Alpha had expanded its reputation as one of the smartest investing outfits on Wall Street. It never had a losing year through 2005, when it posted an eye-popping return of 40 percent.

But Global Alpha had been slipping, losing money in 2006 and the first half of 2007. The worry: to reverse its fortunes, it was juicing up its leverage. And the more leverage, the more risk. Many feared that Global Alpha and its sister fund, Global Equity Opportunity—a stock-focused fund—were doubling down on bad trades with more and more borrowed money.

“They’re one of the few funds big enough to leave such a big footprint.” Mendelson hunched his shoulders in frustration.

“Have you talked to anyone over there?”

“No,” Mendelson said. “I was going to ask you.”

“I’ll give it a shot.”

There had been no small amount of bad blood between Asness and his old colleagues at Goldman, who were embittered after being left behind when Asness and the others bolted. Asness felt bad about it all, but he hadn’t wanted to alienate the head honchos at Goldman by leading a mass exodus. Leaving behind Carhart and Iwanowski to manage Global Alpha, he’d hoped, was a peace offering to the company that had rolled the dice on him when he was fresh meat out of Chicago. But Carhart and Iwanowski weren’t overjoyed about being the sacrificial lambs.

The tensions had cooled over the years. Global Alpha had developed into an elite trading outfit, with $12 billion in assets and a solid record—except for a severe misstep in 2006—that could go head-to-head with the best hedge funds in the business, including AQR.

Asness put in a few calls to Goldman, but no one was picking up the phone. That made him more worried than ever.

Boaz Weinstein
was relaxed that Monday after the weekend bash at his house in the Southhamptons. But soon after lunch, he started to fret. Something was going wrong with Saba’s quant equities desk, which he’d added to his operation to complement his bond-trading group. The news trickled in around two o’clock when Alan Benson, the trader who ran the desk, sent his second daily email with his desk’s P&L.

Benson’s first email, sent at 10:00
A.M.
, showed early signs of losses. But Weinstein had shrugged it off. Benson’s desk, which managed $2 billion worth of positions in stocks and exchange-traded funds, could be highly volatile. Losses in the morning could easily turn into gains by the afternoon.

The 2:00
P.M
. update showed the losses hadn’t turned to gains. They’d gotten much worse. Benson was down tens of millions. Weinstein stood and walked one floor down to Saba’s trading operation on the second floor. Benson looked tense and was sweating.

“What’s up, Alan?” Weinstein said, outwardly calm as always. But there was tension in his voice caused by the startling sight of millions of dollars going up in smoke, just like the GM trade back in 2005.

“It’s weird,” Benson said. “Stocks that we’re betting against are going up, a lot. Looks like short covering on a really big scale, across a lot of industries.”

In a short sale, an investor borrows a stock and sells it, hoping to purchase it back at a lower price sometime in the future. Say IBM is trading at $100 a share and you expect it to decline to $90. You borrow a hundred shares from another investor through a prime broker and sell them to yet another investor for $10,000. If your crystal ball was correct and IBM does in fact fall to $90, you buy the stock back for $9,000, return the shares to the broker, and pocket the $1,000.

But what if, for instance, IBM starts to shoot higher? You’re on the hook for those shares, and every dollar it goes up is a $100 loss. To minimize your losses, you buy the stock back. That can have the effect of pushing the stock even higher. If hundreds or thousands of short sellers are doing this at once, it turns into what’s known as a short squeeze. That Monday, August 6, was beginning to look like possibly the biggest short squeeze in history.

“Has the feel of a big gorilla getting out of a lot of positions, fast,” Benson added.

“Anything we can do about it?”

“Keep an eye on it. I doubt this will last much longer. The rate this guy is unwinding his trades, it can’t go on for long. If it does …”

“What?”

“We’ll have to start unwinding, too.”

At PDT
that same Monday, Peter Muller was AWOL, visiting a friend just outside Boston. Mike Reed and Amy Wong manned the helm, PDT veterans from the old days when the group was nothing more than a thought experiment, its traders a small band of young math whizzes tinkering with computers like brainy teenagers in a cluttered garage.

PDT was now a global powerhouse, with offices in London and Tokyo and about $6 billion in assets (the amount could change daily depending on how much money Morgan funneled its way). It was a well-oiled machine that did little but print money, day after day. That
week, however, PDT wouldn’t print money—it would destroy it like an industrial shredder.

The unusual behavior of stocks that PDT tracked had begun to slip sometime in mid-July and had gotten worse in the first days of August. The previous Friday, five of the biggest gainers on the Nasdaq were stocks that PDT had sold short, expecting them to decline, and five of the biggest losers were stocks PDT had bought, expecting them to rise. It was Bizarro World for quants. Up was down, down was up. The models were operating in reverse. The Truth wasn’t the Truth anymore. It was the anti-Truth.

The losses were accelerating Monday and were especially bad in the roughly $5 billion quant fundamental book—the one PDT had increased in size after Muller returned in late 2006.

Wong and Reed knew that if the losses got much worse, they would need to start liquidating positions in the fundamental book to bring down PDT’s leverage. Already the week before, the group had started to ease back on Midas’s engine as the market’s haphazard volatility picked up steam.

Midas was one thing. It was a high-frequency book that bought and sold securities at a rapid pace all the time. The fundamental book was different. The securities it held, often small-cap stocks that didn’t trade very often, could be hard to get rid of, especially if a number of other traders who owned them were trying to dump them at the same time. The positions would need to be combed through and unwound piece by piece, block by block of unwanted stock. It would be hard, it would be time-consuming, and it would be very costly.

The market moves PDT and other quant funds started to see early that week defied logic. The fine-tuned models, the bell curves and random walks, the calibrated correlations—all the math and science that had propelled the quants to the pinnacle of Wall Street—couldn’t capture what was happening. It was utter chaos driven by pure human fear, the kind that can’t be captured in a computer model or complex algorithm. The wild, fat-tailed moves discovered by Benoit Mandelbrot in the 1950s seemed to be happening on an hourly basis. Nothing like it had ever been seen before.
This wasn’t supposed to happen!

The quants did their best to contain the damage, but they were like firefighters trying to douse a raging inferno with gasoline—the more they tried to fight the flames by selling, the worse the selling became. The downward-driving force of the deleveraging market seemed unstoppable.

Wong and Reed kept Muller posted on the situation through emails and phone calls. It would be Muller’s decision whether to sell into the falling market to deleverage the fund, and by how much. Volatility in the market was surging, confusing PDT’s risk models. Now Muller needed to decide whether to deleverage the fundamental book, which was taking the brunt of the damage. If the losses in that book continued much longer, PDT had little choice but to start selling. It would mean cutting off branches in the hope of saving the tree.

Quant funds everywhere were scrambling to figure out what was going on. Ken Griffin, on vacation in France, kept in touch with the traders at Citadel’s Chicago headquarters. Renaissance was also taking big hits, as were D. E. Shaw, Barclays Global Investors in San Francisco, J. P. Morgan’s quant powerhouse Highbridge Capital Management, and nearly every other quantitative fund in the world, including far-flung operations in London, Paris, and Tokyo.

Tuesday, the downturn accelerated. AQR booked rooms at the nearby Delamar on Greenwich Harbor, a luxury hotel, so they could be available around the clock for stressed-out sleep-deprived quants. Griffin hopped in his private plane and flew back to Chicago for crisis management, and to tie up loose ends on the Sowood deal.

Authorities had little idea about the massive losses taking place across Wall Street. That Tuesday afternoon, the Federal Reserve said it had decided to leave short-term interest rates alone at 5.25 percent. “Financial markets have been volatile in recent weeks, credit conditions have become tighter for some households and businesses, and the housing correction is ongoing,” the Fed said in its policy statement. “Nevertheless, the economy seems likely to continue to expand at a moderate pace over coming quarters, supported by solid growth in employment and incomes and a robust global economy.”

The crisis was mounting, and Washington’s central bankers were completely out of touch. The losses on Monday and Tuesday were
among the worst ever seen by quant hedge funds, with billions of dollars evaporating into thin air. Wednesday they got far worse.

At the
headquarters of Goldman Sachs Asset Management in downtown New York, everyone was on red alert. One of the largest hedge fund managers in the world, with $30 billion in assets, GSAM was getting hit on all sides. It was seeing big losses in value, growth, small-cap stocks, mid-caps, currencies, commodities,
everything
. Global Alpha, the Global Equity Opportunity fund—every strategy was getting crushed. And like every other quant fund, its captains, Carhart and Iwanowski, had no idea why.

GSAM’s risk models, highly sophisticated measures of volatility, had been spiking for all of July. It was a strange sight, because volatility had been declining for years. And the way GSAM’s risk models worked, the decline in volatility meant that it had needed to take more risk, use more leverage, to make the same amount of money. Other quant funds had followed a similar course. Now volatility wasn’t behaving anymore. Volatility was actually …
volatile
.

Another disturbing trend that Goldman’s quants noticed was a rapid unwinding of the worldwide carry trade. Funds such as Global Alpha, AQR, Citadel, and others had been borrowing low-yielding yen on the cheap and investing it in higher-yielding assets, generating huge profits. The trade had been highly successful for years, helping fuel all kinds of speculative bets, but it depended on one trend remaining in place: cheap yen.

In early August 2007, the yen started to surge. Funds that had borrowed the yen, expecting to repay the loan at a later date, were scrambling to repay as the yen leapt in value against other currencies. That triggered a self-reinforcing feedback loop: As the yen kept rising, more funds were needed to repay the loans, pushing the yen up even more.

At GSAM, the sudden unwind meant a potential catastrophe. Many of its positions—bonds, currencies, even stocks—were based on the yen carry trade.

The collapse of the carry trade and the spike in volatility were potentially disastrous. The first major dislocation in the market, one not seen in years, had happened the previous Friday, August 3. The
dislocation turned into an earthquake on Monday. By Tuesday, the situation was critical, and GSAM had to start selling hard.