خانه واحدها

PART I

Faith CHAPTER 1

The House of Quants

Before it was the self-proclaimed largest bookstore on Earth or the Web’s dominant superstore, Amazon.com was an idea floating through the New York City offices of one of the most unusual firms on Wall Street: D. E. Shaw & Co.

A quantitative hedge fund, DESCO, as its employees affectionately called it, was started in 1988 by David E. Shaw, a former Columbia University computer science professor. Along with the founders of other groundbreaking quant houses of that era, like Renaissance Technologies and Tudor Investment Corporation, Shaw pioneered the use of computers and sophisticated mathematical formulas to exploit anomalous patterns in global financial markets. When the price of a stock in Europe was fractionally higher than the price of the same stock in the United States, for example, the computer jockeys turned Wall Street warriors at DESCO would write software to quickly execute trades and exploit the disparity.

The broader financial community knew very little about D. E. Shaw, and its polymath founder wanted to keep it that way. The firm preferred operating far below the radar, deploying private capital from wealthy investors such as billionaire financier Donald Sussman and the Tisch family, and keeping its proprietary trading algorithms out of competitors’ hands. Shaw felt strongly that if DESCO was going to be a firm that pioneered new approaches to investing, the only way to maintain its lead was to keep its insights secret and avoid teaching competitors how to think about these new computer-guided frontiers.

David Shaw came of age in the dawning era of powerful new supercomputers. He earned a PhD in computer science from Stanford in 1980 and then moved to New York to teach in Columbia’s computer science department. Throughout the early eighties, high-tech companies tried to lure him to the private sector. Inventor Danny Hillis, founder of the supercomputer manufacturer Thinking Machines Corporation and later one of Jeff Bezos’s closest friends, almost convinced Shaw to come work for him designing parallel computers. Shaw tentatively accepted the job and then changed his mind, telling Hillis he wanted to do something more lucrative and could always return to the supercomputer field after he got wealthy. Hillis argued that even if Shaw did get rich—which seemed unlikely—he’d never return to computer science. (Shaw did, after he became a billionaire and passed on the day-to-day management of D. E. Shaw to others.) “I was spectacularly wrong on both counts,” Hillis says.

Morgan Stanley finally pried Shaw loose from academia in 1986, adding him to a famed group working on statistical arbitrage software for the new wave of automated trading. But Shaw had an urge to set off on his own. He left Morgan Stanley in 1988, and with a $28 million seed investment from investor Donald Sussman, he set up shop over a Communist bookstore in Manhattan’s West Village.

By design, D. E. Shaw would be a different kind of Wall Street firm. Shaw recruited not financiers but scientists and mathematicians—big brains with unusual backgrounds, lofty academic credentials, and more than a touch of social cluelessness. Bob Gelfond, who joined DESCO after the firm moved to a loft on Park Avenue South, says that “David wanted to see the power of technology and computers applied to finance in a scientific way” and that he “looked up to Goldman Sachs and wanted to build an iconic Wall Street firm.”

In these ways and many others, David Shaw brought an exacting sensibility to the management of his company. He regularly sent out missives instructing employees to spell the firm’s name in a specific manner—with a space between the D. and the E. He also mandated that everyone use a canonical description of the company’s mission: it was to “trade stocks, bonds, futures, options and various other financial instruments”—precisely in that order. Shaw’s rigor extended to more substantive matters as well: any of his computer scientists could suggest trading ideas, but the notions had to pass demanding scientific scrutiny and statistical tests to prove they were valid.

In 1991, D. E. Shaw was growing rapidly, and the company moved to the top floors of a midtown Manhattan skyscraper a block from Times Square. The firm’s striking but sparely decorated offices, designed by the architect Steven Holl, included a two-story lobby with luminescent colors that were projected into slots cut into the expansive white walls. That fall, Shaw hosted a thousand-dollar-a-ticket fund-raiser for presidential candidate Bill Clinton that was attended by the likes of Jacqueline Onassis, among others. Employees were asked to clear out of the office that evening before the event. Jeff Bezos, one of the youngest vice presidents at the firm, left to play volleyball with colleagues, but first he stopped and got his photo taken with the future president.

Bezos was twenty-nine at the time, five foot eight inches tall, already balding and with the pasty, rumpled appearance of a committed workaholic. He had spent seven years on Wall Street and impressed seemingly everyone he encountered with his keen intellect and boundless determination. Upon graduating from Princeton in 1986, Bezos worked for a pair of Columbia professors at a company called Fitel that was developing a private transatlantic computer network for stock traders. Graciela Chichilnisky, one of the cofounders and Bezos’s boss, remembers him as a capable and upbeat employee who worked tirelessly and at different times managed the firm’s operations in London and Tokyo. “He was not concerned about what other people were thinking,” Chichilnisky says. “When you gave him a good solid intellectual issue, he would just chew on it and get it done.”

Bezos moved to the financial firm Bankers Trust in 1988, but by then, frustrated by what he viewed as institutional reluctance at companies to challenge the status quo, he was already looking for an opportunity to start his own business. Between 1989 and 1990 he spent several months working in his spare time on a startup with a young Merrill Lynch employee named Halsey Minor, who would later go on to start the online news network CNET. Their fledgling venture, aimed at sending a customized newsletter to people over their fax machines, collapsed when Merrill Lynch withdrew the promised funding. But Bezos nevertheless made an impression. Minor remembers that Bezos had closely studied several wealthy businessmen and that he particularly admired a man named Frank Meeks, a Virginia entrepreneur who had made a fortune owning Domino’s Pizza franchises. Bezos also revered pioneering computer scientist Alan Kay and often quoted his observation that “point of view is worth 80 IQ points”—a reminder that looking at things in new ways can enhance one’s understanding. “He went to school on everybody,” Minor says. “I don’t think there was anybody Jeff knew that he didn’t walk away from with whatever lessons he could.”

Bezos was ready to leave Wall Street altogether when a headhunter convinced him to meet executives at just one more financial firm, a company with an unusual pedigree. Bezos would later say he found a kind of workplace soul mate in David Shaw—“one of the few people I know who has a fully developed left brain and a fully developed right brain.”1

At DESCO, Bezos displayed many of the idiosyncratic qualities his employees would later observe at Amazon. He was disciplined and precise, constantly recording ideas in a notebook he carried with him, as if they might float out of his mind if he didn’t jot them down. He quickly abandoned old notions and embraced new ones when better options presented themselves. He already exhibited the same boyish excitement and conversation-stopping laugh that the world would later come to know.

Bezos thought analytically about everything, including social situations. Single at the time, he started taking ballroom-dance classes, calculating that it would increase his exposure to what he called n+ women. He later famously admitted to thinking about how to increase his “women flow,”2 a Wall Street corollary to deal flow, the number of new opportunities a banker can access. Jeff Holden, who worked for Bezos first at D. E. Shaw & Co. and later at Amazon, says he was “the most introspective guy I ever met. He was very methodical about everything in his life.”

D. E. Shaw had none of the gratuitous formalities of other Wall Street firms; in outward temperament, at least, it was closer to a Silicon Valley startup. Employees wore jeans or khakis, not suits and ties, and the hierarchy was flat (though key information about trading formulas was tightly held). Bezos seemed to love the idea of the nonstop workday; he kept a rolled-up sleeping bag in his office and some egg-crate foam on his windowsill in case he needed to bunk down for the night. Nicholas Lovejoy, a colleague who would later join him at Amazon, believes the sleeping bag “was as much a prop as it was actually useful.” When they did leave the office, Bezos and his DESCO colleagues often socialized together, playing backgammon or bridge until the early hours of the morning, usually for money.

As the company grew, David Shaw started to think about how to broaden its talent base. He looked beyond math and science geeks to what he called generalists, those who’d recently graduated at the tops of their classes and who showed significant aptitude in particular subjects. The firm also combed through the ranks of Fulbright scholars and dean’s-list students at the best colleges and sent hundreds of unsolicited letters to them introducing the firm and proclaiming, “We approach our recruiting in unapologetically elitist fashion.”

Respondents to the letters who seemed particularly extraordinary and who had high enough grade point averages and aptitude-test scores were flown to New York for a grueling day of interviews. Members of the firm delighted in asking these recruits random questions, such as “How many fax machines are in the United States?” The intent was to see how candidates tried to solve difficult problems. After the interviews, everyone who had participated in the hiring process gathered and expressed one of four opinions about each individual: strong no hire; inclined not to hire; inclined to hire; or strong hire. One holdout could sink an applicant.

Bezos would later take these exact processes, along with the seeds of other Shaw management techniques, to Seattle. Even today, Amazon employees use those categories to vote on prospective new hires.

DESCO’s massive recruitment effort and interview processes were finely tuned to Bezos’s mind-set; they even attracted one person who joined Bezos as his life partner. MacKenzie Tuttle, who graduated from Princeton in 1992 with a degree in English and who studied with author Toni Morrison, joined the hedge fund as an administrative assistant and later went to work directly for Bezos. Lovejoy remembers Bezos hiring a limousine one night and taking several colleagues to a nightclub. “He was treating the whole group but he was clearly focused on MacKenzie,” he says.

MacKenzie later said it was she who targeted Bezos, not the other way around. “My office was next door to his, and all day long I listened to that fabulous laugh,” she told Vogue in 2012. “How could you not fall in love with that laugh?” She began her campaign to win him over by suggesting lunch. The couple got engaged three months after they started dating; they were married three months after that.3 Their wedding, held in 1993 at the Breakers, a resort in West Palm Beach, featured game time for adult guests and a late-night party at the hotel pool. Bob Gelfond and a computer programmer named Tom Karzes attended from D. E. Shaw.

Meanwhile, DESCO was growing rapidly and, in the process, becoming more difficult to manage. Several colleagues from that time recall that D. E. Shaw brought in a consultant who administered the Myers-Briggs personality test to all the members of the executive team. Not surprisingly, everyone tested as an introvert. The least introverted person on the team was Jeff Bezos. At D. E. Shaw in the early 1990s, he counted as the token extrovert.

Bezos was a natural leader at DESCO. By 1993, he was remotely running the firm’s Chicago-based options trading group and then its high-profile entry into the third-market business, an alternative over-the-counter exchange that allowed retail investors to trade equities without the usual commissions collected by the New York Stock Exchange.4 Brian Marsh, a programmer for the firm who would later work at Amazon, says that Bezos was “incredibly charismatic and persuasive about the third-market project. It was easy to see then he was a great leader.” Bezos’s division faced constant challenges, however. The dominant player in the space was one Bernard Madoff (the architect of a massive Ponzi scheme that would unravel in 2008). Madoff’s own third-market division pioneered the business and preserved its market lead. Bezos and his team could see Madoff’s offices in the Lipstick Building on the East Side through their windows high above the city.

While the rest of Wall Street saw D. E. Shaw as a highly secretive hedge fund, the firm viewed itself somewhat differently. In David Shaw’s estimation, the company wasn’t really a hedge fund but a versatile technology laboratory full of innovators and talented engineers who could apply computer science to a variety of different problems.5 Investing was only the first domain where it would apply its skills.

So in 1994, when the opportunity of the Internet began to reveal itself to the few people watching closely, Shaw felt that his company was uniquely positioned to exploit it. And the person he anointed to spearhead the effort was Jeff Bezos.

D. E. Shaw was ideally situated to take advantage of the Internet. Most Shaw employees had, instead of proprietary trading terminals, Sun workstations with Internet access, and they utilized early Internet tools like Gopher, Usenet, e-mail, and Mosaic, one of the first Web browsers. To write documents, they used an academic formatting tool called LaTeX, though Bezos refused to touch the program, claiming it was unnecessarily complicated. D. E. Shaw was also among the very first Wall Street firms to register its URL. Internet records show that Deshaw.com was claimed in 1992. Goldman Sachs took its domain in 1995, and Morgan Stanley a year after that.

Shaw, who used the Internet and its predecessor, ARPANET, during his years as a professor, was passionate about the commercial and social implications of a single global computer network. Bezos had first encountered the Internet in an astrophysics class at Princeton in 1985 but hadn’t thought about its commercial potential until arriving at DESCO. Shaw and Bezos would meet for a few hours each week to brainstorm ideas for this coming technological wave, and then Bezos would take those ideas and investigate their feasibility.6

In early 1994, several prescient business plans emerged from the discussions between Bezos and Shaw and others at D. E. Shaw. One was the concept of a free, advertising-supported e-mail service for consumers—the idea behind Gmail and Yahoo Mail. DESCO would develop that idea into a company called Juno, which went public in 1999 and soon after merged with NetZero, a rival.

Another idea was to create a new kind of financial service that allowed Internet users to trade stocks and bonds online. In 1995 Shaw turned that into a subsidiary called FarSight Financial Services, a precursor to companies like E-Trade. He later sold it to Merrill Lynch.

Shaw and Bezos discussed another idea as well. They called it “the everything store.”

Several executives who worked at DESCO at that time say the idea of the everything store was simple: an Internet company that served as the intermediary between customers and manufacturers and sold nearly every type of product, all over the world. One important element in the early vision was that customers could leave written evaluations of any product, a more egalitarian and credible version of the old Montgomery Ward catalog reviews of its own suppliers. Shaw himself confirmed the Internet-store concept when he told the New York Times Magazine in 1999, “The idea was always that someone would be allowed to make a profit as an intermediary. The key question is: Who will get to be that middleman?”7

Intrigued by Shaw’s conviction about the inevitable importance of the Internet, Bezos started researching its growth. A Texas-based author and publisher named John Quarterman had recently started the Matrix News, a monthly newsletter extolling the Internet and discussing its commercial possibilities. One set of numbers in particular in the February 1994 edition of the newsletter was startling. For the first time, Quarterman broke down the growth of the year-old World Wide Web and pointed out that its simple, friendly interface appealed to a far broader audience than other Internet technologies. In one chart, he showed that the number of bytes—a set of binary digits—transmitted over the Web had increased by a factor of 2,057 between January 1993 and January 1994. Another graphic showed the number of packets—a single unit of data—sent over the Web had jumped by 2,560 in the same span.8

Bezos interpolated from this that Web activity overall had gone up that year by a factor of roughly 2,300—a 230,000 percent increase. “Things just don’t grow that fast,” Bezos later said. “It’s highly unusual, and that started me thinking, What kind of business plan might make sense in the context of that growth?”9 (Bezos also liked to say in speeches during Amazon’s early years that it was the Web’s “2,300 percent” annual growth rate that jolted him out of complacency. Which makes for an interesting historical footnote: Amazon began with a math error.)

Bezos concluded that a true everything store would be impractical—at least at the beginning. He made a list of twenty possible product categories, including computer software, office supplies, apparel, and music. The category that eventually jumped out at him as the best option was books. They were pure commodities; a copy of a book in one store was identical to the same book carried in another, so buyers always knew what they were getting. There were two primary distributors of books at that time, Ingram and Baker and Taylor, so a new retailer wouldn’t have to approach each of the thousands of book publishers individually. And, most important, there were three million books in print worldwide, far more than a Barnes & Noble or a Borders superstore could ever stock.

If he couldn’t build a true everything store right away, he could capture its essence—unlimited selection—in at least one important product category. “With that huge diversity of products you could build a store online that simply could not exist in any other way,” Bezos said. “You could build a true superstore with exhaustive selection, and customers value selection.”10

In his offices on the fortieth floor of 120 West Forty-Fifth Street, Bezos could hardly contain his enthusiasm. With DESCO’s recruiting chief, Charles Ardai, he investigated some of the earliest online bookstore websites, such as Book Stacks Unlimited, located in Cleveland, Ohio, and WordsWorth, in Cambridge, Massachusetts. Ardai still has the record from one purchase they made while testing these early sites. He bought a copy of Isaac Asimov’s Cyberdreams from the website of the Future Fantasy bookstore in Palo Alto, California. The price was $6.04. When the book appeared, two weeks later, Ardai ripped open the cardboard package and showed it to Bezos. It had become badly tattered in transit. No one had yet figured out how to do a good job selling books over the Internet. As Bezos saw it, this was a huge, untapped opportunity.

Bezos knew it would never really be his company if he pursued the venture inside D. E. Shaw. Indeed, the firm initially owned all of Juno and FarSight, and Shaw acted as chairman of both. If Bezos wanted to be a true owner and entrepreneur, with significant equity in his creation and the potential to achieve the same kind of financial rewards that businessmen like pizza magnate Frank Meeks did, he had to leave his lucrative and comfortable home on Wall Street.

What happened next became one of the founding legends of the Internet. That spring, Bezos spoke to David Shaw and told him he planned to leave the company to create an online bookstore. Shaw suggested they take a walk. They wandered in Central Park for two hours, discussing the venture and the entrepreneurial drive. Shaw said he understood Bezos’s impulse and sympathized with it—he had done the same thing when he’d left Morgan Stanley. He also noted that D. E. Shaw was growing quickly and that Bezos already had a great job. He told Bezos that the firm might end up competing with his new venture. The two agreed that Bezos would spend a few days thinking about it.

At the time Bezos was thinking about what to do next, he had recently finished the novel Remains of the Day, by Kazuo Ishiguro, about a butler who wistfully recalls his personal and professional choices during a career in service in wartime Great Britain. So looking back on life’s important junctures was on Bezos’s mind when he came up with what he calls “the regret-minimization framework” to decide the next step to take at this juncture in his career.

“When you are in the thick of things, you can get confused by small stuff,” Bezos said a few years later. “I knew when I was eighty that I would never, for example, think about why I walked away from my 1994 Wall Street bonus right in the middle of the year at the worst possible time. That kind of thing just isn’t something you worry about when you’re eighty years old. At the same time, I knew that I might sincerely regret not having participated in this thing called the Internet that I thought was going to be a revolutionizing event. When I thought about it that way… it was incredibly easy to make the decision.”11

Bezos’s parents, Mike and Jackie, were nearing the end of a three-year stay in Bogotá, Colombia, where Mike was working for Exxon as a petroleum engineer, when they got the phone call. “What do you mean, you are going to sell books over the Internet?” was their first reaction, according to Mike Bezos. They had used the early online service Prodigy to correspond with family members and to organize Jeff and MacKenzie’s engagement party, so it wasn’t naïveté about new technology that unnerved them. Rather, it was seeing their accomplished son leave a well-paying job on Wall Street to pursue an idea that sounded like utter madness. Jackie Bezos suggested to her son that he run his new company at night or on the weekends. “No, things are changing fast,” Bezos told her. “I need to move quickly.”

So Jeff Bezos started planning for his journey. He held a party at his Upper West Side apartment to watch the final episode of Star Trek: The Next Generation. Then he flew out to Santa Cruz, California, to meet two experienced programmers who had been introduced to him by Peter Laventhol, David Shaw’s first employee. Over blueberry pancakes at the Old Sash Mill Café in Santa Cruz, Bezos managed to intrigue one of them, a startup veteran named Shel Kaphan. Bezos “was inflamed by a lot of the same excitement as I was about what was happening with the Internet,” Kaphan says. They looked at office space together in Santa Cruz, but Bezos later learned of a 1992 Supreme Court decision that upheld a previous ruling that merchants did not have to collect sales tax in states where they did not have physical operations. As a result, mail-order businesses typically avoided locating in populous states like California and New York, and so would Bezos.

Back in New York, Bezos informed his colleagues that he was leaving D. E. Shaw. Bezos and Jeff Holden, a recent graduate of the University of Illinois at Urbana-Champaign who had worked for Bezos as an engineer on the third-market project, went out one night for drinks. The two were close. Holden was from Rochester Hills, Michigan, and as a teenager, under the hacker nom de guerre the Nova, he had grown adept at cracking copyright protection on software. He was an avid Rollerblader and a fast talker; he spoke so rapidly that Bezos liked to joke that Holden “taught me to listen faster.”

Now they were sitting across from each other at Virgil’s, a barbecue place on Forty-Fourth Street. Bezos had tentatively decided to call his company Cadabra Inc. but was not committed to the name. Holden filled both sides of a piece of notebook paper with alternatives. The one Bezos liked best on the list was MakeItSo.com, after Captain Picard’s frequent command in Star Trek.

Over beers, Holden told Bezos he wanted to come with him. But Bezos was worried; his contract with D. E. Shaw stipulated that if he left the firm, he couldn’t recruit DESCO employees for at least two years. David Shaw was not someone he wanted to cross. “You’re just out of school, you’ve got debt. And this is risky,” Bezos said. “Stay here. Build up some net worth and I’ll be in touch.”

Later that month, Bezos and MacKenzie packed up the contents of their home and told the movers to just start driving their belongings across the country—they said they would call them on the road the next day with a specific destination. First they flew to Fort Worth, Texas, and borrowed a 1988 Chevy Blazer from Bezos’s father. Then they drove northwest, Bezos sitting in the passenger seat, typing revenue projections into an Excel spreadsheet—numbers that would later prove to be radically inaccurate. They tried to check into a Motel 6 in Shamrock, Texas, but it was booked, so they settled for a road motel called the Rambler.12 When MacKenzie saw the room, she declined to take off her shoes that night. A day later, they stopped at the Grand Canyon and watched the sunrise. He was thirty-one, she was twenty-four, and together they were writing an entrepreneurial origin story that would be imprinted on the collective imagination of millions of Internet users and hopeful startup founders.

More than a year passed before Jeff Holden heard from his friend again. Bezos had settled in Seattle, and he e-mailed Holden a link to a website. They were now calling it Amazon.com. The site was primitive, mostly text and somewhat unimpressive. Holden bought a few books through the site and offered some feedback. Then another year passed, and finally, a few months after Bezos’s do-not-poach agreement with David Shaw expired, Holden’s phone rang.

It was Bezos. “It’s time,” he said. “This is going to work.”