When Genius Failed: The Rise and Fall of Long Term Capital Management

Soybeans suddenly seemed quaint; money was the hot commodity now. Futures exchanges devised new contracts in financial goods such as Treasury bills and bonds and Japanese yen, and everywhere there were new instruments, new options, new bonds to trade, just when professional portfolio managers were waking up and wanting to trade them. By the end of the 1970s, firms such as Salomon were slicing and dicing bonds in ways that Homer had never dreamed of: blending mortgages together, for instance, and distilling them into bite-sized, easily chewable securities.

The other big change was the computer. As late as the end of the 1960s, whenever traders wanted to price a bond, they would look it up in a thick blue book. In 1969, Salomon hired a mathematician, Martin Leibowitz, who got Salomon’s first computer. Leibowitz became the most popular mathematician in history, or so it seemed when the bond market was hot and Salomon’s traders, who no longer had time to page through the blue book, crowded around him to get bond prices that they now needed on the double. By the early 1970s, traders had their own crude handheld calculators, which subtly quickened the rhythm of the bond markets.

Meriwether, who joined Salomon on the financing desk, known as the Repo Department, got there just as the bond world was turning topsy-turvy. Once predictable and relatively low risk, the bond world was pulsating with change and opportunity, especially for younger, sharp-eyed analysts. Meriwether, who didn’t know a soul when he arrived in New York, rented a room at a Manhattan athletic club and soon discovered that bonds were made for him. Bonds have a particular appeal to mathematical types because so much of what determines their value is readily quantifiable. Essentially, two factors dictate a bond’s price. One can be gleaned from the coupon on the bond itself. If you can lend money at 10 percent today, you would pay a premium for a bond that yielded 12 percent. How much of a premium? That would depend on the maturity of the bond, the timing of the payments, your outlook (if you have one) for interest rates in the future, plus all manner of wrinkles devised by clever issuers, such as whether the bond is callable, convertible into equity, and so forth.

The other factor is the risk of default. In most cases, that is not strictly quantifiable, nor is it very great. Still, it exists. General Electric is a good risk, but not as good as Uncle Sam. Hewlett-Packard is somewhat riskier than GE; Amazon.com, riskier still. Therefore, bond investors demand a higher interest rate when they lend to Amazon as compared with GE, or to Bolivia as compared with France. Deciding how much higher is the heart of bond trading, but the point is that bonds trade on a mathematical spread. The riskier the bond, the wider the spread—that is, the greater the difference between the yield on it and the yield on (virtually risk free) Treasurys. Generally, though not always, the spread also increases with time—that is, investors demand a slightly higher yield on a two-year note than on a thirty-day bill because the uncertainty is greater.

These rules are the catechism of bond trading; they ordain a vast matrix of yields and spreads on debt securities throughout the world. They are as intricate and immutable as the rules of a great religion, and it is no wonder that Meriwether, who kept rosary beads and prayer cards in his briefcase, found them satisfying. Eager to learn, he peppered his bosses with questions like a divinity student. Sensing his promise, the suits at Salomon put him to trading government agency bonds. Soon after, New York City nearly defaulted, and the spreads on various agency bonds soared. Meriwether reckoned that the market had goofed—surely, not every government entity was about to go bust—and he bought all the bonds he could. Spreads did contract, and Meriwether’s trades made millions.3

The Arbitrage Group, which he formed in 1977, marked a subtle but important shift in Salomon’s evolution. It was also the model that Long-Term Capital was to replicate, brick for brick, in the 1990s—a laboratory in which Meriwether would become accustomed to, and comfortable with, taking big risks. Although Salomon had always traded bonds, its primary focus had been the relatively safer business of buying and selling bonds for customers. But the Arbitrage Group, led by Meriwether, became a principal, risking Salomon’s own capital. Because the field was new, Meriwether had few competitors, and the pickings were rich. As in the Eckstein trade, he often bet that a spread—say, between a futures contract and the underlying bond, or between two bonds—would converge. He could also bet on spreads to widen, but convergence was his dominant theme. The people on the other side of his trades might be insurers, banks, or speculators; Meriwether wouldn’t know, and usually he wouldn’t care. Occasionally, these other investors might get scared and withdraw their capital, causing spreads to widen further and causing Meriwether to lose money, at least temporarily. But if he had the capital to stay the course, he’d be rewarded in the long run, or so his experience seemed to prove. Eventually, spreads always came in; that was the lesson he had learned from the Eckstein affair, and it was a lesson he would count on, years later, at Long-Term Capital. But there was a different lesson, equally valuable, that Meriwether might have drawn from the Eckstein business, had his success not come so fast: while a losing trade may well turn around eventually (assuming, of course, that it was properly conceived to begin with), the turn could arrive too late to do the trader any good—meaning, of course, that he might go broke in the interim.

By the early 1980s, Meriwether was one of Salomon’s bright young stars. His shyness and implacable poker face played perfectly to his skill as a trader. William Mcintosh, the Salomon partner who had interviewed him, said, “John has a steel-trap mind. You have no clue to what he’s thinking.” Meriwether’s former colleague, the writer Michael Lewis, echoed this assessment of Meriwether in Liar’s Poker:

He wore the same blank half-tense expression when he won as he did when he lost. He had, I think, a profound ability to control the two emotions that commonly destroy traders—fear and greed—and it made him as noble as a man who pursues his self-interest so fiercely can be.4

It was a pity that the book emphasized a supposed incident in which Meriwether allegedly dared Gutfreund to play a single hand of poker for $10 million, not merely because the story seems apocryphal, but because it canonized Meriwether for a recklessness that wasn’t his.5 Meriwether was the priest of the calculated gamble. He was cautious to a fault; he gave away nothing of himself. His background, his family, his entire past were as much of a blank to colleagues as if, one said, he had “drawn a line in the sand.” He was so intensely private that even when the Long-Term Capital affair was front-page news, a New York Times writer, after trying to determine if Meriwether had any siblings, settled for citing the inaccurate opinion of friends who thought him an only child.6 Such reticence was a perfect attribute for a trader, but it was not enough. What Meriwether lacked, he must have sensed, was an edge—some special forte like the one he had developed on the links in high school, something that would distinguish Salomon from every other bond trader.

His solution was deceptively simple: Why not hire traders who were smarter? Traders who would treat markets as an intellectual discipline, as opposed to the folkloric, unscientific Neanderthals who traded from their bellies? Academia was teeming with nerdy mathematicians who had been publishing unintelligible dissertations on markets for years. Wall Street had started to hire them, but only for research, where they’d be out of harm’s way. On Wall Street, the eggheads were stigmatized as “quants,” unfit for the man’s game of trading. Craig Coats, Jr., head of government-bond trading at Salomon, was a type typical of trading floors: tall, likable, handsome, bound to get along with clients. Sure, he had been a goof-off in college, but he had played forward on the basketball team, and he had trading in his heart. It was just this element of passion that Meriwether wanted to eliminate; he preferred the cool discipline of scholars, with their rigorous and highly quantitative approach to markets.

Most Wall Street executives were mystified by the academic world, but Meriwether, a math teacher with an M.B.A. from Chicago, was comfortable with it. That would be his edge. In 1983, Meriwether called Eric Rosenfeld, a sweet-natured MIT-trained Harvard Business School assistant professor, to see if Rosenfeld could recommend any of his students. The son of a modestly successful Concord, Massachusetts, money manager, Rosenfeld was a computer freak who had already been using quantitative methods to make investments. At Harvard, he was struggling.7 Laconic and dry, Rosenfeld was compellingly bright, but he was less than commanding in a classroom. At a distance, he looked like a thin, bespectacled mouse. The students were tough on him; “they beat the shit out of him,” according to a future colleague. Rosenfeld, who was grading exams when Meriwether called and was making, as he recalled, roughly $30,000 a year, instantly offered to audition for Salomon himself. Ten days later, he was hired.8

Meriwether didn’t stop there. After Rosenfeld, he hired Victor J. Haghani, an Iranian American with a master’s degree in finance from the London School of Economics; Gregory Hawkins, an Arkansan who had helped run Bill Clinton’s campaign for state attorney general and had then gotten a Ph.D. in financial economics from MIT; and William Krasker, an intense, mathematically minded economist with a Ph.D. from—once again—MIT and a colleague of Rosenfeld at Harvard. Probably the nerdiest, and surely the smartest, was Lawrence Hilibrand, who had two degrees from MIT. Hilibrand was hired by Salomon’s research department, the traditional home of quants, but Meriwether quickly moved him into the Arbitrage Group, which, of course, was the heart of the future Long-Term Capital.

The eggheads immediately took to Wall Street. They downloaded into their computers all of the past bond prices they could get their hands on. They distilled the bonds’ historical relationships, and they modeled how these prices should behave in the future. And then, when a market price somewhere, somehow got out of line, the computer models told them.

The models didn’t order them to trade; they provided a contextual argument for the human computers to consider. They simplified a complicated world. Maybe the yield on two-year Treasury notes was a bit closer than it ordinarily was to the yield on ten-year bonds; or maybe the spread between the two was unusually narrow, compared with a similar spread for some other country’s paper. The models condensed the markets into a pointed inquiry. As one of the group said, “Given the state of things around the world—the shape of yield curves, volatilities, interest rates—are the financial markets making statements that are inconsistent with each other?” This is how they talked, and this is how they thought. Every price was a “statement”; if two statements were in conflict, there might be an opportunity for arbitrage.

The whole experiment would surely have failed, except for two happy circumstances. First, the professors were smart. They stuck to their knitting, and opportunities were plentiful, especially in newer markets such as derivatives. The professors spoke of opportunities as inefficiencies; in a perfectly efficient market, in which all prices were correct, no one would have anything to trade. Since the markets they traded in were still evolving, though, prices were often incorrect and there were opportunities aplenty. Moreover, the professors brought to the job an abiding credo, learned from academia, that over time, all markets tend to get more efficient.

In particular, they believed, spreads between riskier and less risky bonds would tend to narrow. This followed logically because spreads reflect, in part, the uncertainty that is attached to chancier assets. Over time, if markets did become more efficient, such riskier bonds would be less volatile and therefore more certain-seeming, and so the premium demanded by investors would tend to shrink. In the early 1980s, for instance, the spreads on swaps—a type of derivative trade, of which more later—were 2 percentage points. “They looked at this and said, ‘It can’t be right; there can’t be that much risk,’” a junior member of Arbitrage recalled. “They said, ‘There is going to be a secular trend toward a more efficient market.’”

And swap spreads did tighten—to 1 percentage point and eventually to a quarter point. All of Wall Street did this trade, including the Salomon government desk, run by the increasingly wary Coats. The difference was that Meriwether’s Arbitrage Group did it in very big dollars. If a trade went against them, the arbitrageurs, especially the ever-confident Hilibrand, merely redoubled the bet. Backed by their models, they felt more certain than others did—almost invincible. Given enough time, given enough capital, the young geniuses from academe felt they could do no wrong, and Meriwether, who regularly journeyed to academic conferences to recruit such talent, began to believe that the geniuses were right.

That was the second happy circumstance: the professors had a protector who shielded them from company politics and got them the capital to trade. But for Meriwether, the experiment couldn’t have worked; the professors were simply too out of place. Hilibrand, an engineer’s son from Cherry Hill, New Jersey, was like an academic version of Al Gore; socially awkward, he answered the simplest-seeming questions with wooden and technical—albeit mathematically precise—replies. Once, a trader not in the Arbitrage Group tried to talk Hilibrand out of buying and selling a certain pair of securities. Hilibrand replied, as if conducting a tutorial, “But they are priced so egregiously.” His colleague, accustomed to the profane banter of the trading floor, shot back, “I was thinking the same thing—‘egregiously’!” Surrounded by unruly traders, the arbitrageurs were quiet intellectuals. Krasker, the cautious professor who built many of the group’s models, had all the charisma of a tabletop. Rosenfeld had a wry sense of humor, but in a firm in which many of the partners hadn’t gone to college, much less graduate school at MIT, he was shy and taciturn.

Meriwether had the particular genius to bring this group to Wall Street—a move that Salomon’s competitors would later imitate. “He took a bunch of guys who in the corporate world were considered freaks,” noted Jay Higgins, then an investment banker at Salomon. “Those guys would be playing with their slide rules at Bell Labs if it wasn’t for John, and they knew it.”9

The professors were brilliant at reducing a trade to pluses and minuses; they could strip a ham sandwich to its component risks; but they could barely carry on a normal conversation. Meriwether created a safe, self-contained place for them to develop their skills; he adoringly made Arbitrage into a world apart. Because of Meriwether, the traders fraternized with one another, and they didn’t feel the need to fraternize with anyone else.

Meriwether would say, “We’re playing golf on Sunday,” and he didn’t have to add, “I’d like you to be there.” The traders who hadn’t played golf before, such as Hilibrand and Rosenfeld, quickly learned. Meriwether also developed a passion for horses and acquired some thoroughbreds; naturally, he took his traders to the track, too. He even shepherded the gang and their spouses to Antigua every year. He didn’t want them just during trading hours, he wanted all of them, all the time. He nurtured his traders, all the while building a protective fence around the group as sturdy as the red board fence in Rosemoor.

Typical of Meriwether, he made gambling an intimate part of the group’s shared life. The arbitrageurs devised elaborate betting pools over golf weekends; they bet on horses; they took day trips to Atlantic City together. They bet on elections. They bet on anything that aroused their passion for odds. When they talked sports, it wasn’t about the game; it was about the point spread.

Meriwether loved for his traders to play liar’s poker, a game that involves making poker hands from the serial numbers on dollar bills. He liked to test his traders; he thought the game honed their instincts, and he would get churlish and threaten to quit when they played poorly. It started as fun, but then it got serious; the traders would play for hours, occasionally for stakes in the tens of thousands of dollars. Rosenfeld kept an envelope stuffed with hundreds of single bills in his desk. Then, when it seemed that certain bills were cropping up too often, they did away with bills and got a computer to generate random lists of numbers. The Arbitrage boys seemed addicted to gambling: “You could never go out to dinner with J.M.’s guys without playing liar’s poker to see who would pick up the check,” Gerald Rosenfeld, Salomon’s chief financial officer, recalled. Meriwether was a good player, and so was Eric Rosenfeld (no relation), who had an inscrutable poker face. The straight-arrow Hilibrand was a bit too literal. He was incapable of lying and for a long time never bluffed; mustachioed and eerily intelligent, he had a detachment that was almost extrahuman. Once, when asked whether it was awkward to have a wife who worked in mortgages (which Hilibrand traded), he answered flatly, “Well, I never talk to my wife about business.”

The Arbitrage Group, about twelve in all, became incredibly close. They sat in a double row of desks in the middle of Salomon’s raucous trading floor, which was the model for the investment bank in Tom Wolfe’s The Bonfire of the Vanities. Randy Hiller, a mortgage trader in Arbitrage, found its cliquish aspect overbearing and left. Another defector was treated like a traitor; Meriwether vengefully ordered the crew not to even golf with him. But very few traders left, and those who remained all but worshiped Meriwether. They spoke of him in hushed tones, as of a Moses who had brought their tribe to Palestine. Meriwether didn’t exactly return the praise, but he gave them something more worthwhile. His interest and curiosity stimulated the professors; it challenged them and made them better. And he rewarded them with heartfelt loyalty. He never screamed, but it wouldn’t have mattered if he had. To the traders, the two initials “J.M.”—for that was his unfailing sobriquet—were as powerful as any two letters could be.

Though he had a private office upstairs, Meriwether usually sat on the trading floor, at a tiny desk squeezed in with the others. He would chain-smoke while doing Eurodollar trades, and supervise the professors by asking probing questions. Somehow, he sheathed great ambition in an affecting modesty. He liked to say that he never hired anyone who wasn’t smarter than he was. He didn’t talk about himself, but no one noticed because he was genuinely interested in what the others were doing. He didn’t build the models, but he grasped what the models were saying. And he trusted the models because his guys had built them. One time, a trader named Andy who was losing money on a mortgage trade asked for permission to double up, and J.M. gave it rather offhandedly. “Don’t you want to know more about this trade?” Andy asked. Meriwether’s trusting reply deeply affected the trader. J.M. said, “My trade was when I hired you.”

Meriwether had married Mimi Murray, a serious equestrian from California, in 1981, and the two of them lived in a modest two-bedroom apartment on York Avenue on the Upper East Side. They wanted children, according to a colleague, but remained childless.

Aside from Mimi, J.M.’s family was Salomon. He didn’t leave his desk even for lunch; in fact, his noontime was as routinized as the professors’ models. Salomon did a china-service lunch, and for a long time, every day, a waiter would waft over to Meriwether bearing a bologna sandwich on white bread, two apples, and a Tab hidden under a silver dome. J.M. would eat one of the apples and randomly offer the other to one of the troops as a sort of token. The rest of the gang might order Chinese food, and if any sauce leaked onto his desk, J.M., his precious territory violated, would scowl and say, “Look, I guess I’m going to have to give up my desk and go back to my office and work there.”

A misfit among Wall Street’s Waspish bankers, J.M. identified more with the parochial school boys he had grown up with than with the rich executives whose number he had joined. Unlike other financiers in the roaring eighties, who were fast becoming trendy habitués of the social pages, Meriwether disdained attention (he purged his picture from Salomon’s annual report) and refused to dine on any food that smacked of French. When in Tokyo, he went to McDonald’s. Ever an outsider, he molded his group into a tribe of outsiders as cohesive, loyal, and protective as the world he had left in Rosemoor. His cohorts were known by schoolboy nicknames such as Vic, the Sheik, E.R., and Hawk.

Although J.M. knew his markets, his reputation as a trader was overwrought. His real skill was in shaping people, which he did in singularly understated style. He was awkward when speaking to a group; his words came out in uneven bunches, leaving others to piece together their meaning.10 But his confidence in his troops was written on his face, and it worked on their spirits like a tonic. Combined with the traders’ uncommon self-confidence, Meriwether’s faith in them was a potent but potentially combustible mix. It inflated their already supreme self-assurance. Moreover, J.M.’s willingness to bankroll Hilibrand and the others with Salomon’s capital dangerously conditioned the troops to think that they would always have access to more.

As Arbitrage made more money, the group’s turf inevitably expanded. Meriwether, eclipsing rivals such as Coats, gained command over all bond trading, including government bonds, mortgages, high-yield corporate bonds, European bonds, and Japanese warrants. It seemed logical, for the group to apply its models in new and greener pastures. But others in Salomon began to seethe. J.M. would send one of his boys—Hilibrand or Victor Haghani—to Salomon’s London office or its Tokyo office, and the emissary would declare, “This trade is very good, but you should be ten times bigger in it.” Not two times, but ten times! As if they couldn’t fail. Hilibrand and Haghani were in their twenties, and they might be talking to guys twice their age. Then they started to say, “Don’t do this trade; we’re better at this than anyone else, so we’ll do all of this trade on the arbitrage desk.”

Hilibrand was particularly annoying. He was formal and polite, but he struck old hands as condescending, infuriating them with his mathematical certitude. One time, he tried to persuade some commodity traders that they should bet on oil prices following a pattern similar to that of bond prices. The traders listened dubiously while Hilibrand bobbed his head back and forth. Suddenly he raised a hand and sonorously declaimed, “Consider the following hypothesis.” It was as if he were delivering an edict from on high, to be etched in stone.

Traders had an anxious life; they’d spend the day shouting into a phone, hollering across the room, and nervously eyeballing a computer screen. The Arbitrage Group, right in the middle of this controlled pandemonium, seemed to be a mysterious, privileged subculture. Half the time, the boys were discussing trades in obscure, esoteric language, as if in a seminar; the other half, they were laughing and playing liar’s poker. In their cheap suits and with their leisurely mien, they could seemingly cherry-pick the best trades while everyone else worked at a frenetic pace.

The group was extremely private; it seemed to have adopted J.M.’s innate secretiveness as a protective coloring. Though any trader is well advised to be discreet, the professors’ refusal to share any information with their Salomon colleagues fueled the resentment felt by Coats and others. Though Arbitrage soaked up all of the valuable tidbits that passed through a premier bond-trading floor, it set up its own private research arm and strictly forbade others in Salomon to learn about its trades. One time, the rival Prudential-Bache hired away a Salomon mortgage trader, which was considered a coup. “What was the first thing he wanted?” a then-Pru-Bache manager laughingly remembered. “Analytics? Better computer system or software? No. He wanted locks on the filing cabinet. It reflected their mentality!” Driven by fanatical loyalty to Meriwether, the Arbitrage Group nurtured an us-against-them clannishness that would leave the future Long-Term dangerously remote from the rest of Wall Street. Hilibrand became so obsessed with his privacy that he even refused to let Salomon Brothers take his picture.11