16 Nov 1998

How Salesmanship and Brainpower
Failed to Save Long-Term Capital


Even with the market tremors of the preceding weeks, no one foresaw the earthquake about to rock Greenwich, Conn., one summer morning.

It was Aug. 21, a sultry Friday, and nearly half the partners at Long-Term Capital Management LP were out of the office. Outside the fund's glass-and-granite headquarters, a fountain languidly streamed over a copper osprey clawing its prey.

Inside, associates logged on to their computers and saw something deeply disturbing: U.S. Treasurys were skyrocketing, throwing their relationship to other securities out of whack. The Dow Jones Industrial Average was swooning -- by noon, down 283 points. The European bond market was in shambles.

LTCM's biggest bets were blowing up, and no one could do anything about it.

By 11 a.m., the fund had lost $150 million in a wager on the prices of two telecommunications stocks involved in a takeover. Then, a single bet tied to the U.S. bond market lost $100 million. Another $100 million evaporated in a similar trade in Britain. By day's end, LTCM had hemorrhaged half a billion dollars. Its equity had sunk to $3.1 billion -- down a third for the year.

Partners scrambled out of their offices and onto the trading floor as associates stared at their screens in disbelief. Making frantic phone calls around the globe, they reached John Meriwether, the fund's founder, at a dinner in Beijing. He boarded the next plane to the U.S. Eric Rosenfeld, a top lieutenant, called in from Sun Valley, Idaho, where he was settling in for a vacation. He left his wife and children behind and made an all-night trip back to Greenwich.

The brass assembled at headquarters at 7 a.m. that Sunday. One after another, LTCM's partners, calling in from Tokyo and London, reported that their markets had dried up. There were no buyers, no sellers. It was all but impossible to maneuver out of large trading bets. They had seen nothing like it.

The carnage that weekend set off events unprecedented in the world of high finance, culminating with a $3.625 billion bailout funded by a consortium of 14 Wall Street banks and engineered by the Federal Reserve.

LTCM lost more than 90% of its assets by the time it was bailed out, and the markets were roiled for weeks. Longer term, it forced many of the world's most sophisticated institutional investors to redefine the ways they manage risk and triggered calls for tougher regulation of hedge funds, those freewheeling investment pools that cater to the wealthy.

In an industry populated by sharp money managers, LTCM had the most renowned of all -- including Nobel Prize winners Robert Merton and Myron Scholes. But in the end, it wasn't all rocket science. It was about smart marketing -- appealing to a wealthy clientele who wanted to be able to say their money was being managed by a passel of Ph.D.s. And it was about massive borrowing, up to $50 for every dollar invested. LTCM was, ultimately, like a supermarket -- a high-volume, low-margin business, trying to eke out small profits from thousands of individual transactions.

"Myron once told me they are sucking up nickels from all over the world," says Merton Miller, a University of Chicago business professor and himself a Nobel Prize winner in economics. "But because they are so leveraged, that amounts to a lot of money."

All of which helps to explain how so many geniuses, sometimes overcoming divisions within their ranks, got it so wrong. And all the while, vanity, greed and a cult of personality blinded some of the world's most reputable financial institutions, from Wall Street stalwarts to Swiss banks, to the pitfalls inherent in such a strategy.

Not that everyone ignored the risks. In his Nobel Prize address last year, Mr. Scholes himself predicted "there will be failures" among firms that make the types of transactions that his fund favored. That shouldn't provoke fear, he said, for while the past was "the age of innocence," the techniques of the future heralded "the age of excitement."

A Start at Salomon

The seeds of Long-Term Capital were sown in 1984 at Salomon Brothers, where Mr. Meriwether ran one of the firm's most profitable bond divisions. By then, the explosive growth of global bond markets and the increasing volatility of interest rates brought a new level of complexity to Wall Street. Mr. Meriwether, who a decade earlier had dropped out of the University of Chicago's Ph.D. program in business administration, cherry-picked the best academic minds he could find.

One of his first recruits was Mr. Rosenfeld, then a Harvard Business School professor. Mr. Meriwether called him, looking to hire one of his bright students. But Mr. Rosenfeld, mired in grading final exams, volunteered that he "would like to try out." Ten days later, he left Harvard for Salomon.

Mr. Rosenfeld, in turn, recruited another Harvard economics professor, and then hired a finance professor from the University of California at Berkeley. Mr. Meriwether went after bigger guns, hiring Harvard's Mr. Merton as a consultant. And Mr. Meriwether persuaded Mr. Scholes, a Stanford University luminary, to join Salomon.

Mr. Meriwether took pains to maintain an academic pipeline. He invited promising scholars to present papers at Salomon. At meetings of the American Finance Association, he mingled with newly minted Ph.D.s. The promises of riches at Salomon, compared to the relatively paltry wages of academia, made it easy to recruit.

Mr. Meriwether's bond group operated as a firm within a firm. It made millions by using rigorous mathematical methods to bet on changes in interest rates on bonds of different maturities. Secrecy was sacrosanct. Even senior Salomon executives had no idea how the group's positions were faring before the end of quarterly reporting periods. And Mr. Meriwether quietly managed to get his group a fat 15% cut of its profit, rather than depend on evaluations of his department. When colleagues found out, there was a firestorm of protest.

Mr. Meriwether weathered the infighting. But he couldn't survive Salomon's 1991 bond scandal. Then a vice chairman, Mr. Meriwether was one of three Salomon officials who resigned after the firm disclosed a series of improper bids at U.S. Treasury note auctions.

As memories of the scandal faded, his old colleagues pushed for Salomon to rehire Mr. Meriwether. They were rebuffed. So six of them left Salomon to join Mr. Meriwether.

By early 1993, the group was convening regularly to brainstorm about where to re-create their trading shop. Members considered joining an investment bank or offering their services to an insurance company or bank. Then they went on the road.

Messrs. Meriwether and Rosenfeld traveled to Omaha, Neb., and, over a steak dinner, tried to persuade billionaire investor Warren Buffett to invest with them. With his support, raising more money would have been a breeze. But Mr. Buffett wasn't interested.

They went to Zurich for a meeting with senior executives at UBS, the big Swiss bank. Mr. Meriwether threw out a smattering of ideas. Would UBS set up a unit where Mr. Meriwether and his troops could trade? Or would the bank finance a fund set up by Mr. Meriwether? UBS executives, put off by the lack of a clear plan, took a pass.

But Mr. Meriwether caught a break on his next call, just across the street. At Bank Julius Baer & Co., senior executive Raymond Baer, who knew the partners from his days at Salomon, told Mr. Meriwether he could count the bank in if he started a hedge fund.

Heartened by the pledge, Mr. Meriwether laid the groundwork for LTCM. Among those who joined: David Mullins Jr., a former Harvard professor who was serving as the Fed's vice chairman.

At one point, LTCM counted 25 Ph.D.s on its payroll. The fund was "probably the best academic finance department in the world," says William Sharpe, a 1990 Nobel economics laureate and a fellow faculty member of Mr. Scholes at Stanford.

By late 1993, LTCM was born. The partners located in Greenwich -- near where many of them lived. Together, they pledged to put more than $100 million of their own money on the line.

Refining the Sales Pitch

The partners went on a spirited marketing campaign, canvassing the globe for clients. At each stop, they extolled the virtues of their ability to scour global bond markets for opportunities. When asked about the downside, they would presciently tell investors of one of the biggest, and rarest, risks of all: a sudden flight to supersafe securities (like Treasury bonds) that could demolish their bets on riskier transactions.

LTCM usually brought along salesmen from Merrill Lynch & Co. who had been hired as sales agents for the fund. And they often featured Mr. Scholes in the road shows. It made sense: Mr. Scholes, along with the late Fischer Black, is credited with discovering how to price options and provide a way to manage risk for Wall Street -- the breakthrough for which he would win the Nobel.

Their timing was excellent. It was 1994, with the Dow Jones Industrial Average still under 4000. Investors were looking for opportunities not tied directly to the stock market. Globalization of investments still was in its infancy.

Even so, Mr. Scholes stumbled in some early pitches. In one, he and other LTCM representatives met with executives at Conseco Capital Management, the investing arm of the big insurer. They sat around a long table, spelling out their strategy. The fund, Mr. Scholes said, would leverage its capital to take advantage of pricing "anomalies" in global markets.

"You're not adding any value," interjected Andrew Chow, the Conseco vice president in charge of derivatives. He added: "I don't think there are that many pure anomalies that can occur."

Mr. Scholes gazed at Mr. Chow. Then, sitting back in his leather-padded chair, he dressed Mr. Chow down, deriding him for believing he knew "all the answers." According to Maxwell Bublitz, Conseco Capital's president, Mr. Scholes added, "As long as there continue to be people like you, we'll make money."

Conseco executives were dumbfounded. "It was very condescending. We booted them out," recalls Mr. Bublitz. LTCM's pitch consisted of, "Our resumes are better than yours -- wouldn't you like to invest?"

In the parking lot on the way out, Mr. Scholes told Chad Schultz, one of the Merrill representatives, that the blitz of marketing meetings had exhausted him. "I'm sorry I lost my cool," he said.

Later that day, Mr. Schultz called Conseco to apologize, explaining to Mr. Bublitz that he told Mr. Scholes after the meeting: "When you're trying to raise money from people, it's not a great idea to insult them." Mr. Schultz declines to discuss the incident, as does an LTCM spokesman.

Outside of the U.S., the partners became dispirited because they weren't well known. In Europe, Mr. Meriwether and his lieutenants were reduced to pulling out old Salomon earnings releases. And it was hard to explain their strategies in terms that institutional finance officers could understand. So Merrill Lynch schooled the partners in the language of investors, and the pitch became more compelling. LTCM would be "market neutral," that is, uncorrelated to stock, bond or currency markets. And the firm's trades would be spread around the world, making the investments more diversified.

Still, Mr. Meriwether wanted some stiff restrictions. He insisted that investors not be allowed to withdraw money for at least three years. (Most hedge funds allow investors to withdraw annually, and in some cases quarterly.) The fund also required a $10 million minimum -- one of the highest in the industry.

Then there were the sky-high fees: an annual management charge of 2% of assets, plus 25% of profits, compared with 1% and 20%, respectively, for most of the industry. Still, the partners told investors they had put their money where their mouths were. Indeed, so confident were the partners about the fund that Mr. Rosenfeld, among others, put his children's money into LTCM.

Later, the partners would joke that their kids were richer than they -- a reference to the fact that many of them had invested in trusts created in their children's names.

Joining the A-Team

LTCM's star power impressed investors like Terry Sullivan, president of Paragon Advisors Inc., a Shaker Heights, Ohio, firm. In February 1994, he headed to Greenwich, where the partners were dressed in khakis and golf shirts. The exception was Mr. Mullins, who was wearing a dark business suit. During the pitch, Mr. Mullins explained that because he once was the Fed's vice chairman, he was in the "heads" of Fed members and could "figure out" what they would do, Mr. Sullivan recalls, referring to notes he took at the meeting.

The partners were reassuring. LTCM would seek trading opportunities that were "uncorrelated" to the broader securities markets, Mr. Sullivan says. They assured Mr. Sullivan they typically would use a ratio of investment capital to assets of only 20 to 1. (The fund's leverage ratio ultimately soared to 50 to 1.) They told Mr. Sullivan they would use leverage to boost returns from a "very low-risk strategy" and bring the fund's risk to a "standard deviation of the stock market."

Mr. Sullivan was wowed. With an M.B.A. of his own, he had studied works by Messrs. Merton and Scholes at the University of Pittsburgh. Mr. Sullivan invested $10 million for clients and felt lucky to have hooked up with the "A-Team."

The Nobel and Fed connections were essential. "Meriwether was very good at marketing them," says Roy Smith, a former Goldman, Sachs & Co. partner who is now a professor at New York University. "Investing in this thing was done on the basis of networking, wanting to do the cool thing and trusting the superstars."

The pitch started to click better overseas. Dresdner Bank AG, after repeated visits by LTCM partners, signed up. The Bank of Italy's foreign-exchange office invested $100 million, and later lent $150 million. International banker Edmund Safra got in through a joint venture between his Republic New York Corp. and Safra Republic Holdings SA, a private banking affiliate. His Swiss bankers enthusiastically pitched the fund to European millionaires.

Bank Julius Baer was one of several private Swiss banks to invest. Julius Baer also invested through Creinvest, a publicly traded fund of funds.

In the U.S., PaineWebber Group Inc. invested $100 million of its capital; its chairman, Donald Marron, chipped in $10 million of his own cash. Even St. John's University, in Jamaica, N.Y., invested $10 million after members of its investment and finance committee told the Rev. Donald Harrington, St. John's president, and its board that Mr. Meriwether was "brilliant." (A university spokeswoman declined to comment.)

But LTCM realized its biggest client would have to be Wall Street. The most important hook: Merrill Lynch, with its huge sales force and unmatched access.

In early 1993, Mr. Meriwether called then-Merrill Chairman Daniel Tully and laid out his plans to form his fund. Mr. Tully was intrigued. Soon, Merrill agreed to help sell the fund to investors.

It was a wild success. Merrill ended up raising more than $1.5 billion of the fund's assets and put in $15 million of its own capital in the portfolio -- the amount of its placement fee. Merrill later extended $1.4 billion in financing and swap agreements to the fund.

On Wall Street, the word got out that LTCM was the place to be. It got little press -- partners cooperated for just one major story, a Business Week cover piece in 1994 -- but wealthy investors learned about it soon enough.

Wall Street soon discovered that LTCM was a tough customer. It was hard to make much money trading with the fund because the experienced partners could negotiate on bond transactions that lack posted market prices. That's why, despite its extensive relationship with the fund, Merrill generated revenue of only $25 million from LTCM annually from 1995 through 1998, analysts say. "They were an account with a sharp pencil," says Herbert Allison Jr., Merrill's president. "They weren't going to give something away."

The fund needed a powerful partner to process its many trades. Tapping a relationship with Vincent Mattone, an executive vice president at Bear Stearns Cos. who had worked at Salomon, LTCM signed up Bear Stearns as clearing agent at what an executive calls "skimpy" rates. The fund sent much of its trading through Bear Stearns, making LTCM its largest hedge-fund client. Bear Stearns also handled futures, risk arbitrage and mortgage trading with the fund, collecting $25 million in annual revenue from the account. (A Bear Stearns spokeswoman declined to comment on the firm's relationship with LTCM.)

Mr. Mattone also urged Bear Stearns Chief Executive James Cayne to personally invest. Mr. Cayne, without having met Mr. Meriwether, put up $13 million.

The Risks Get Bigger

The fund formally began trading in February 1994, during one of the biggest bond-market routs in years. Many bondholders were desperate to unload. As one of the few buyers, the fund found ample disparities between price and value. That year, LTCM returned 19.9%, after fees.

The partners had worked hard to woo foreign investors like the Bank of Italy's foreign-exchange office. The idea was that they would serve as the firm's eyes and ears in far-flung places, sniffing out cheap securities and helping the fund execute trades in unfamiliar markets. Meanwhile, LTCM scoured the world for securities whose prices in relation to each other were out of whack, then bet they would eventually return to historical norms -- and generate a big profit.

Consider this trade: In 1994, the firm noticed that 29 1/2-year U.S. Treasury bonds seemed cheap in relation to 30-year Treasurys. The values of the two bonds, the partners figured, would converge over time. So they bought $2 billion of the 29 1/2-year Treasurys, and sold short $2 billion of 30-year bonds. Six months later, they took a $25 million profit on $12 million of capital used to execute the trade.

With such successes, the fund's returns hit 42.8% in 1995, then 40.8% in 1996, after fees. That far outpaced hedge funds' average performance of 16% and 17%, respectively. By mid-1996, the original partners' stake had tripled to more than $300 million.

But soon, Mr. Meriwether's traders were no longer the only big players in the bond-arbitrage game. Heightened competition squeezed the margins. So LTCM quietly began investing beyond its core strategy. One area was stock-takeover arbitrage, where it bought the stocks of companies slated for a takeover, and took short, or sold, positions in the acquiring companies.

Even by LTCM's standards, this was risky business. On one try, the fund lost about $100 million on a bet on the proposed takeover of the former MCI Communications Corp. by British Telecommunications PLC. LTCM eventually made the money back when MCI was bought by WorldCom Inc.

As LTCM became the biggest player in markets for some highly illiquid securities, some of its partners, including Messrs. Scholes, Merton and Mullins, began raising concerns. At Tuesday morning risk-management meetings, Mr. Scholes pressed his associates on the pitfalls of relatively simple transactions, such as straight bets on a currency. "What informational advantage do we have over other traders?" he asked. "What if you had to get out?" The three also questioned whether the fund was setting aside enough collateral on trades involving such exotic securities as Danish mortgage bonds.

And as word leaked out about the fund's changes in strategy, some investors got worried, too.

Mr. Sullivan, president of the Ohio advisory firm, says a trader tipped him off to LTCM's position in a planned merger of Staples Inc. and Office Depot Inc. "I wasn't comfortable," he says. "They didn't tell people what was going on. They were changing their approach."

He cashed out a $10 million stake for his clients. "We dodged a bazooka," he says. "It was dumb luck."

A Swiss Stake

Despite some naysayers, lenders and investors were falling over themselves to do business with LTCM. But by late 1995, the fund was closed to new equity investment. That didn't stop some investors who, working through middlemen in London and Zurich, paid a premium to buy existing equity in the fund from people seeking to cash out early.

Even the staid Swiss banks that had turned away Mr. Meriwether were scrambling to get a piece of the action. And nowhere was the demand greater than the Union Bank of Switzerland. (The bank merged earlier this year with Swiss Bank Corp. to form UBS AG; it declined to comment on its dealings with LTCM.)

Back in 1994, when Long-Term Capital came to UBS for credit, the bank's credit group recommended that the request be declined because the hedge fund had such a limited track record and was borrowing so heavily elsewhere. But UBS executives kept pressing the credit group and got approval to make the financing available anyway.

By 1996, as LTCM was making huge profits, UBS wanted a piece of the action. Hans Peter Bauer, a UBS executive, asked Ron Tannenbaum, a UBS salesman with a close relationship with Mr. Meriwether from their days at Salomon, to approach the fund. Yet each time Mr. Tannenbaum broached the subject of an equity investment, he came away empty-handed. The fund was performing so well that adding new equity investors would have diluted the stakes of the original investors -- including LTCM's partners.

But then Mr. Scholes devised a tantalizing offer: UBS would get to buy more than $1 billion of stock in the fund. Meanwhile, the fund's partners would pay UBS $320 million for a seven-year option to acquire $800 million of stock in the hedge fund from the bank at a fixed price.

That transaction also reduced the partners' personal tax liabilities. Here's how: If the partners had simply borrowed money to buy larger stakes, a portion of their investment returns and fees would be taxable as ordinary income at a rate of 39.6%. But with the UBS deal, they didn't own the additional fund stakes outright, only the option to buy the shares at a fixed price in seven years. The gains on those shares eventually would be taxable at the 20% rate for long-term capital gains, fund specialists say.

UBS executives were so ecstatic, they included the trade in company slide shows as one of the most successful deals of the year. And the LTCM stake was considered such a prize that various UBS divisions sought to claim part of it for their own books. In fact, UBS's trading desk began soliciting bids for the position it held. In the end, the bank's treasury department won out, paying a 5% premium for the LTCM equity position.

The Losses Begin

By 1997, LTCM was losing momentum. The bond-arbitrage landscape was looking played out, and the firm's computer models weren't spotting the kind of out-of-kilter prices that spelled big profits. In 1997, the fund returned just 17.1%, after fees.

Yet LTCM was sitting on more than $7 billion in capital. Worried about the appearance of collecting high management fees at a time of slim investment opportunities, partners returned $2.7 billion, effectively cashing out investors who had gotten into the fund after 1994. Some investors were upset because the partners weren't reducing their own equity. What they didn't know was that partners had decided to boost their stakes.

Meanwhile, some partners questioned the fund's "directional" trades -- simple bets on the direction of a market. Mr. Scholes, in particular, was far more comfortable with transactions based on intricate mathematical models. So he was skeptical about the fund's decision to simultaneously buy the Norwegian currency, the krone, and sell the German mark. He argued that the transactions lacked a "date with destiny," or a time when two securities would converge and there would be a high probability that the bet would pay off. The currency trade did lose money, and is being sold. He also questioned LTCM's decision to bet that German interest rates would rise on the belief that the European Central Bank would hike rates in early 1999. Instead, economists say, it now appears that rates in Europe will be headed lower. And LTCM is dismantling the trade. An LTCM spokesman denies that Mr. Scholes objected to either trade, or that there were significant divisions over investment strategy.

While the debates over strategy heated up within LTCM, to most outsiders, the first hint of trouble in the portfolio came in June 1998. The fund was getting slammed in bond markets around the world, triggering a 10.1% loss -- its largest-ever one-month loss.

The fund scaled back emerging-markets positions and unloaded more easily traded instruments, such as bond futures. By early July, LTCM's woes were starting to worry some investors and lenders, and partners put in calls to a few of them.

But it took time for many investors and lenders to grasp the depth of the problem because they were privy only to snippets of the fund's business -- the pieces they were involved with. LTCM often would execute just one side of a trade with one bank, and do the other with another bank. Neither firm would have a clear idea about the position LTCM was taking.

The fund "never wanted to discuss what the whole picture was," says Sanford Weill, co-chairman of Citigroup Inc., whose Salomon Smith Barney unit was a big lender to the fund.

Even Merrill Lynch didn't have a handle on the fund's total holdings. "We really only saw that part of the portfolio that we did business with," says David Komansky, Merrill chairman and chief executive. "A large part of their story was their considerable expertise, reputation and brainpower." The fund, he noted, "never failed on any trade."

Back at LTCM headquarters, the partners didn't appear especially concerned. In a meeting with Andrew Siciliano, a UBS executive, Mr. Meriwether seemed unfazed. He said he had been expecting that the string of heady returns wouldn't continue uninterrupted. In fact, he said, the fund was loading up again on its core bond-market bets even as it was scaling back trades in emerging markets.

And as the markets calmed in early August, the partners went ahead with their travel plans. Then the real carnage began.

'A Shock to Us'

In mid-August, Russia abruptly defaulted on part of its debt and let the ruble fall, triggering a flight by investors from all types of risk into safe investments. That devastated some of LTCM's bets, leading to the huge losses of Aug. 21.

Two days later -- and just hours after the LTCM partners met to assess the damage -- Mr. Rosenfeld made an overture to Warren Buffett. Would the billionaire be interested in a chunk of the fund's risk-arbitrage positions, now available at fire-sale prices? Mr. Buffett said he was "pretty intrigued" but told Mr. Rosenfeld the fund would be better off approaching a Wall Street securities firm.

The next day, the partners launched a capital-raising scramble. With rumors swirling, they knew it was important to line up a big investor before the fund disclosed its August results. They had a week.

Mr. Meriwether coolly called Mr. Allison, Merrill's president. "We've lost more money, but I see real opportunity out there," Mr. Meriwether told Mr. Allison, according to Wall Street executives. The fund could "capitalize" on depressed market prices if only it had more cash, Mr. Meriwether said. Could Merrill invest between $300 million and $500 million -- and quickly?

It was quintessential Meriwether, believing his bets would work out if he just had the chips to wager with. This time, Merrill didn't bite. (Merrill declined to comment on discussions between its executives and Mr. Meriwether.)

Mr. Rosenfeld called PaineWebber. No takers there. Mr. Meriwether met with Stanley Druckenmiller, George Soros's chief investment strategist at the rival hedge-fund firm. How about $500 million? Mr. Druckenmiller decided the markets were too tumultuous. LTCM also approached Julian Robertson of Tiger Management, another hedge-fund outfit. But he passed as well, as did the Ziff brothers, New York-based private investors.

On Aug. 26, around 10 p.m., Mr. Rosenfeld again approached Mr. Buffett, telling him the situation was "more serious." Joined by Mr. Meriwether on the phone this time, they asked if they could send somebody out to Omaha.

Mr. Buffett didn't hold out much hope, but the next morning, he picked up LTCM partner Lawrence Hilibrand at the airport. For four hours, Mr. Hilibrand painted a bare-bones picture of LTCM's investments. Again, Mr. Buffett wasn't interested, saying it was hard to get his "hands around" the portfolio. "You are looking for an investor, and I am not an investor in other people's funds," he told Mr. Hilibrand.

Mr. Scholes called Mr. Sharpe, his old colleague from the Stanford faculty. What followed was a cat-and-mouse game between two Nobel winners: Mr. Scholes was after more money from a wealthy family Mr. Sharpe was advising; Mr. Sharpe wanted a clear picture of how much trouble the firm was in. Neither got what he wanted.

Mr. Scholes admitted that the August losses would be steep, recalls Mr. Sharpe, but insisted the opportunities were the best he had seen in years. "We knew that we didn't know enough to decide whether we should invest," says Mr. Sharpe. "It takes you back to Watergate: What did they know, and when did they know it?"

Meanwhile, LTCM employees were becoming restive. They could see money streaming out more quickly than it was coming in. To calm nerves, the partners borrowed $38 million from the fund to meet operating expenses, including paying nonpartners' salaries until year end.

Equally anxious were LTCM's lenders -- and particularly its clearing agent, Bear Stearns. In fact, Bear Stearns was so worried the fund would be unable to back up its obligations that it dispatched a team to sit on the financing desk in Greenwich. There, Bear Stearns executives listened in on phone calls that the fund made with some of its lenders.

Some lenders were slow to recognize the fund's problems, for while the markets were suffering, they were scrambling to stem losses of their own. But on Sept. 2, LTCM disclosed that the value of the fund's holdings had dropped by 44%, or $1.8 billion, in August. "Losses of this magnitude are a shock to us as they surely are to you," Mr. Meriwether wrote to investors. Then he asked them for more money, at discounted fees.

Marlon Pease, director of finance at the University of Pittsburgh, says he flew to LTCM's headquarters; his university had invested $5 million of its endowment in LTCM. In Greenwich, he recalls, Messrs. Merton and Scholes were "upbeat" and reported confidently that they "were looking to raise about $1 billion by the end of September, and another billion after that." The markets, he recalls them saying, "were the most remarkable opportunity that they'd seen in their lifetimes." Mr. Pease declined to invest more, and he returned home without fully understanding how dire the situation was. "They gave us some information, but in retrospect, it wasn't as candid as it could or should have been," he says.

Calling the Fed

On Sept. 17, Mr. Meriwether and his top lieutenants met with Goldman Co-Chairman Jon Corzine and his associates. The agenda: to see if Goldman and Long-Term Capital could work together as partners. "Would you be willing to accept risk controls?" Mr. Corzine pressed the group. "And oversight?"

No problem, they said. If a Goldman-led investor group could raise at least $2 billion in capital, Mr. Meriwether said, the fund would agree to stepped-up controls and supervision. Investors could also have 50% of his management company, which collects fees and a cut of the profits. Goldman executives started contacting potential investors, including Michael Dell of Dell Computer Corp.

The next day, Mr. Corzine called New York Federal Reserve Bank President William McDonough. He told him that LTCM "is weak, but we're trying to raise money." Peter Fisher, the No. 2 at the New York Fed, was dispatched to LTCM that Sunday.

But Goldman had trouble enticing investors, and LTCM's portfolio fell to about $1.5 billion. So Peter Kraus, a Goldman banker, approached Mr. Buffett, who was headed to the Alaskan wilds for a sightseeing trip with Microsoft Corp. CEO Bill Gates. "The only way I'll do this is if we jointly buy the portfolio and you take over the portfolio company," Mr. Buffett told Mr. Kraus. Mr. Buffett made clear: He wanted Goldman -- not Mr. Meriwether and his team -- running the portfolio.

All the while, potential investors lined up by Goldman piled into LTCM's offices to pore over the books. One after another, they delivered the same message: This thing is scary.

Fears of Default

Mr. Corzine called the Fed to say Goldman couldn't put together a rescue package anytime soon. So a new plan had to be hatched quickly. LTCM faced continued margin calls, or demands for more collateral on loans. More importantly, the markets were so fragile that Wall Street executives feared that a default by the fund would ricochet, forcing securities firms to close out some of LTCM's positions at fire-sale prices. That wave of selling would cause heavy losses on trading desks that had placed similar bets.

Bear Stearns, the fund's clearing agent, then set a new condition: Fall below $500 million, and it would stop processing trades. LTCM tapped a $500 million revolving loan from Chase Manhattan Corp. to keep Bear Stearns off its back.

Over breakfast at the Fed on Sept. 22, Wall Street executives explored other ways to avert the crisis. One idea: Set up a consortium of securities firms and commercial banks to inject equity into the fund. But LTCM would have to agree to an oversight committee.

By Tuesday afternoon, Mr. Buffett told his bankers at Goldman that he would be willing to put up a least $3 billion for LTCM's portfolio. Goldman was in for $300 million; its bankers told Mr. Buffett that they would enlist Maurice Greenberg at American International Group Inc. to chip in $700 million, bringing the total bid to $4 billion.

At 6 p.m., the Fed put out a call to more than a dozen top executives whose firms had lent money to Long-Term Capital. They should come to an 8 p.m. meeting. After they gathered in the 10th-floor boardroom at the Fed's fortress-like headquarters near Wall Street, Mr. Fisher took the floor. He warned that the systemic market risk posed by LTCM going into default was "very real."

Merrill's Mr. Allison, reading from a handwritten sheet, spelled out the terms. "We think we need $4 billion to assure the fund can withstand any concerted attack by others against its positions." Sixteen firms were asked to pitch in $250 million.

Immediately, the complaints started. How do we know everyone has the same exposure? Why should everyone put up the same amount?

Mr. Corzine urged his colleagues to focus on the big picture. "There's no way we can look at this book right now and say who has the most to gain and who has the most to lose," he argued. The four firms leading the bailout -- Goldman, Merrill, J.P. Morgan & Co. and UBS -- said they were each in for $250 million. Other firms, hearing the proposal for the first time, couldn't commit. They broke up at 11 p.m. with plans to reconvene Wednesday morning.

That day, the 10 a.m. meeting was delayed as word spread that Goldman had found a mystery bidder, believed to be Mr. Buffett. Mr. McDonough, the New York Fed chief, called Mr. Buffett to ask him if he was serious. He was.

But when Mr. Buffett's bankers reached Mr. Meriwether, he turned down the offer, explaining that, on such short notice, he couldn't get the approval of his investors to accept the bid. "Our lawyers said we can't do this," he said.

The broader group reconvened at 1 p.m. that day. The mood was tense. Many of the executives -- surrounded by paintings of former New York Fed presidents -- didn't take their jackets off. Some didn't touch their water glasses, covered with white paper tops.

In a five-hour meeting, executives hotly debated whether an LTCM collapse would put the system at risk. Then it came time to put the bids in. They got an immediate jolt: Mr. Cayne, the Bear Stearns chief, wasn't going to play. He didn't say why.

Some executives were incensed. Did Bear, the fund's clearing agent, know something they didn't about the portfolio? During a break, Mr. Komansky buttonholed Mr. Cayne, who said Bear Stearns already had taken sufficient risk clearing LTCM's trades.

Back in the meeting room, Mr. Komansky announced that he had discussed the matter with Mr. Cayne, and that he was convinced his decision didn't suggest there were broader problems with the portfolio. Mr. Cayne followed up, telling the other executives that Bear Stearns's clearing risk "was a helluva lot more than $250 to $300 million."

Other firms balked at $250 million. Lehman Brothers executives decided they would offer $100 million, though they argued the firm's exposure wasn't that high. "We eat here, too," Richard Fuld, Lehman's chairman, told the group. Mr. Komansky said the consortium should accept the smaller contribution, noting that Lehman was facing a rough patch after recent credit-rating agency scrutiny.

It was close to 6 p.m., and there still wasn't enough money. So about a dozen firms upped their antes to $300 million. The deal was sealed. The executives broke into applause.

As they filed out, they were left to ponder whether all this was necessary, and whether a collapse would really have jolted the global financial system. "It was a very large unknown," Merrill's Mr. Allison says. "It wasn't worth a jump into the abyss to find out how deep it was."

An Efficient Market

In the weeks since the rescue, there has been more frustration and fallout.

LTCM's new owners -- the consortium that orchestrated the bailout -- now oversee all trading and can veto decisions made by the partners. The fund suffered additional losses in September, though the bleeding has been stanched, and LTCM registered $100 million in profits amid the recent rebound in the markets.

The fund has cut its work force by 20%, or 33 traders, research analysts and back-office employees. The consortium also is considering ousting several of the fund's partners. And a big investor may still end up buying the fund's assets. Meanwhile, several congressmen are calling for more oversight of hedge funds, and federal regulators are likely to seek greater disclosure of hedge-fund investments.

As for the 16 partners, their stakes, which early this year were valued at $1.6 billion, are now worth $30 million. Four partners, including Mr. Hilibrand, are also on the hook for personal loans totaling nearly $43 million. So vexed was Mr. Hilibrand that he and his lawyer pleaded that the consortium use some of the bailout money to help partners pay off their loans. But Mr. Corzine, among others, held firm, maintaining that "our money isn't there to bail out the individual shareholders."

The failure of LTCM has reverberated in the executive suites of Wall Street. Since the rescue, UBS Chairman Mathis Cabiallavetta resigned, as did three other UBS executives, amid disclosures that the Swiss bank had losses of $680 million from its dealings with the fund.

Through it all, LTCM's rise and fall proved what some of those economics professors who stayed on campus had been saying all along: The market is brutally efficient. "Most of academic finance is teaching that you can't earn 40% a year without some risk of losing a lot of money," says Mr. Sharpe, the former Stanford colleague of Mr. Scholes. "In some sense, what happened is nicely consistent with what we teach."