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

Résumé de When Genius Failed

1. 1

   When Geniuses Gambled with the World

   

   00:00

   00:00

   Four Nobel laureates, legendary Wall Street traders, and the most sophisticated mathematical models money could buy-what could possibly go wrong? In August 1998, a Connecticut hedge fund with positions worth over $1 trillion nearly brought the global financial system to its knees. This wasn't some reckless startup run by cowboys. Long-Term Capital Management was the most intellectually decorated investment firm ever assembled, a collection of financial geniuses who genuinely believed they had conquered risk itself. Their spectacular implosion became the defining cautionary tale of modern finance, yet its lessons remain hauntingly unlearned. John Meriwether built his reputation at Salomon Brothers on a deceptively simple philosophy: ride your losses until they turn into gains. The insight came from classic arbitrage-finding two nearly identical securities trading at different prices, buying the cheap one, shorting the expensive one, and waiting for the gap to close. It didn't matter whether markets rose or fell; all that mattered was convergence. In 1979, when securities dealer J.F. Eckstein faced bankruptcy from arbitrage trades gone wrong, Meriwether convinced Salomon to absorb his positions. Though initially hemorrhaging money, the prices eventually converged exactly as predicted, earning substantial profits and cementing Meriwether's mystique. This philosophy would eventually threaten the entire global economy.

2. 2

   Building the Cathedral of Leverage and Recruiting Genius

   Commencer l'écoute

   Meriwether assembled an elite team of mathematicians at Salomon, including Larry Hilibrand, who earned $23 million in one year from his 15% profit share. This extraordinary compensation bred resentment that would haunt them later. After scandal forced Meriwether's resignation, he immediately began planning his own hedge fund. Meriwether designed Long-Term Capital Management around "relative value" trades-betting on spreads between pairs of securities that typically move together. The critical difference was leverage: LTCM planned to borrow twenty to thirty times its capital or more, multiplying both potential returns and catastrophic risks. His fundraising terms were audacious: 25% of profits plus 2% of assets annually, with investors locked in for three years. He recruited Harvard's Robert Merton and Myron Scholes, co-creator of the Black-Scholes formula. The stunning recruitment of Federal Reserve vice chairman David Mullins sealed their mystique. Despite awkward salesmanship-one investor found Rosenfeld nearly comatose-their Salomon track record prevailed. By February 1994, Long-Term opened with $1.25 billion, the largest hedge fund launch ever.

3. 3

   The Culture of Certainty and Early Triumphs

   Commencer l'écoute

   Long-Term refused standard margin requirements, insisting their capital made collateral unnecessary. Banks that resisted lost their business. As Goldman's Jon Corzine admitted, "You had no choice if you were going to do business with them." Banks rationalized this because Long-Term settled daily-only sudden, catastrophic losses would threaten collateral, which seemed mathematically improbable. The partners operated with uncanny closeness, as if sharing one mind. Larry Hilibrand embodied the firm's essence-cool, self-confident, following models religiously. Meriwether maintained mysterious authority despite his bashful demeanor, speaking in fragments and rarely contributing until meetings ended. Long-Term specialized in exploiting tiny price discrepancies with massive leverage. Their first major success-on-the-run/off-the-run bond trades-netted $15 million without using capital. They preferred reaping sure nickels to gambling on uncertain dollars, multiplying small margins thousands of times. In 1994, the fund earned 28% while most bond investors lost money. By spring 1996, Long-Term controlled $140 billion in assets-thirty times its capital-rivaling Wall Street's oldest institutions. Mathematical precision gave them complete confidence.

4. Chapitre 4

   The Fatal Flaw in the Models

   Long-Term treated money management as quantifiable science, not judgment-requiring art. Merton and Scholes calculated precise probabilities - the fund would lose 5% in twelve of every hundred years. This precision was seductive but fundamentally flawed. The Black-Scholes formula assumed market movements followed a bell curve, with constant volatility and prices flowing smoothly in "continuous time." Reality differed sharply. Eugene Fama discovered stock prices showed far more extreme movements than normal distributions predicted. Five-standard-deviation events that should occur once every 7,000 years actually happened every three to four years. Financial markets displayed "fat tails." Why? Because markets have memories. Unlike coin flips, market trends continue as traders expect them to. After three bad days, some traders are forced to sell while others panic, creating momentum that defies randomness. The human element causing these runs was entirely absent from Merton's idealized markets. Long-Term had a serious structural flaw: unlike banks with independent risk managers, the partners monitored themselves. There was no watchdog to challenge their assumptions.

   Chapitre 5

   Chasing Returns into Dangerous Territory

   By 1997, Long-Term's success had spawned imitators. Wall Street banks established arbitrage desks, eroding profitable spreads. As Rosenfeld complained, "We'd go to put on a trade, but when we started to nibble, the opportunity would vanish." With traditional opportunities dwindling, Meriwether pushed into unfamiliar territory-commercial mortgage-backed securities, Brazilian and Russian bonds, Danish mortgages, even insurance operations. Most dramatically, they ventured into equities. Haghani bet $2.3 billion on Royal Dutch/Shell convergence-ten times larger than Goldman's similar trade and practically illiquid. Hilibrand moved aggressively into merger arbitrage, buying virtually every announced deal rather than selectively choosing based on deep knowledge. This sparked internal debate, with Scholes and Merton arguing they lacked expertise in this specialized field. By circumventing Federal Reserve regulations through derivative swaps, Long-Term built massive equity positions with extreme leverage while avoiding regulatory disclosures. The partners fundamentally believed markets would become more rational over time. As one former Salomon trader observed, "The MIT types always want to short volatility. Academics think these models are the Holy Grail."

   Chapitre 6

   The Perfect Storm and System-Wide Panic

   The collapse began in spring 1998. As traders liquidated positions, volatility surged, forcing more liquidations in a vicious cycle. Russia raised interest rates dramatically, and Long-Term suffered its worst month ever, losing 6.7%. By June, credit spreads widened across every market where Long-Term operated. Jim McEntee, the one partner relying on intuition, repeatedly urged reducing risks but was dismissed as an "old-fashioned gambler." Then Russia declared a debt moratorium on August 17, shattering the assumption that nuclear powers wouldn't default. By Friday, August 21, Long-Term lost $553 million-15% of its capital-in a single day. Their models had calculated it was unlikely to lose more than $35 million daily. Since April, they'd lost over a third of their equity. Peter Fisher from the Federal Reserve visited and had an epiphany: Long-Term's seemingly diverse trades were actually correlated-"the same spread trade everywhere in the world." During crises, all correlations go to one. By Monday, September 21, Long-Term lost another $553 million, leaving less than a billion dollars against over $100 billion in assets. Its leverage exceeded 100-to-1-a mere 1% additional loss would wipe it out completely.

   Chapitre 7

   The Lessons We Refuse to Learn

   Fisher convened an emergency meeting at the Fed that evening. Twenty-five Wall Street leaders crammed into the boardroom. Warren Buffett faxed an offer to buy Long-Term for $250 million-a fund worth $4.7 billion at year's start. The final terms: banks invested $3.6 billion for 90% equity. The partners' $1.9 billion stake vanished in five weeks. Larry Hilibrand, in tears, initially refused to sign, then relented. The press dubbed them "The Brightest and the Brokest." Despite the rescue, Long-Term kept plummeting, losing $750 million in two weeks. Only after Greenspan cut rates did spreads narrow. Meriwether immediately planned a comeback, portraying themselves as brilliant rationalists undone by an unreasoning market. They blamed "liquidity shortages" and a "perfect storm"-despite similar crises recurring regularly throughout financial history. Ten years later, September 2008 brought Fannie Mae and Freddie Mac bailouts, Lehman's collapse, and AIG's rescue. The parallels: excessive leverage, incomprehensible instruments, and blind faith in backward-looking models. Just as LTCM's models failed to predict losing $553 million in a day, credit agencies never imagined housing prices falling 30%. Markets remain inherently unstable due to panic. Risk models based on past behavior can't account for fear, greed, and herd mentality. The solution isn't sophisticated mathematics-it's recognizing its limits, restricting leverage, and maintaining genuine oversight. But these lessons seem destined for endless repetition. The question isn't whether the next LTCM will emerge-it's when.

   Continuer la lecture