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

Roger Lowenstein, When Genius Failed: The Rise and Fall of Long-Term Capital Management, Random House, 2000.

The Professors and the Absent-Minded Lenders

Imagine a drama that revolves around a cast of brainy, quirky characters, is set against a backdrop of phenomenal wealth, and has a plot that threatens to bring the entire financial system to its knees. Then go out and read Roger Lowenstein's When Genius Failed: The Rise and Fall of Long-Term Capital Management.

The book chronicles the history of a hedge fund, designated the "Dream Team" by Business Week in a 1994 cover story, and the nightmare it engendered when it almost collapsed in September 1998.

It's a tale of John Meriwether and his professors, the best mathematical minds from academia, who were fabulously successful as Salomon Brothers' legendary bond arbitrage group in the 1980's. When he started his own firm in 1993, Meriwether added former Federal Reserve Vice Chairman David Mullins, and two soon-to-be Nobel Laureates in economics, Myron Scholes and Robert Merton.

Based on his reputation, Meriwether raised $1.25 billion to start. Investors were so eager to get a peek at the inner workings of the mysterious hedge fund that they financed 100 percent of LTCM's positions.

Long-Term enjoyed enormous success at first, returning 20 percent to investors in 1994 and more than 40 percent in 1995 and 1996 by making big bets on small discrepancies in related markets. It bought securities that were cheap and sold securities that were expensive, based on a mathematical model of historical relationships that assumed markets become more efficient over time; the reduction in uncertainty was bound to narrow the spread between risky and risk-free assets.

That model broke down in August 1998, when Russia defaulted on some of its debt. Investors fled assets with any cred it risks and sought safety in risk-free sovereign debt, especially U.S. Treasuries.

"In every arbitrage, [LTCM] owned the riskier asset; in every country, the least safe bond," Lowenstein writes. "It had made that one same bet hundreds of times, and now that bet was losing"

That August, the fund lost $1.9 billion, 45 percent of its capital. After a $553 million loss in September, the finn was supporting $100 billion of assets with just $1 billion of capital.

Meriwether & Co. still believed in their model. They thought if they could raise more capital, the markets would start acting rationally, and their losses would be reversed. That didn't happen, forcing the Fed to step in. It cajoled Wall Street's titans to pony up $3.65 billion to save LTCM, their firms, and the financial system.

Would a forced liquidation of LTCM have caused the entire financial system to collapse? Would the fourteen Wall Street firms have decided that it was in their self-interest to rescue the fund without the interference of the central bank?

We'll never know. And because of that, history will repeat itself, Lowenstein suggests: "Permitting such losses to occur is what deters most other people and institutions from taking imprudent risks."

In additon to Long-Term's partners and Wall Street's banks, Lowenstein offers

up an unlikely villain: Fed Chairman Alan Greenspan. He chastises Greenspan for his cavalier attitude on increased regulation and disclosure of derivatives, even as he was condoning the Fed's participation in the rescue. "Head in the sand before a crisis, intervention after the fact" is Lowenstein's characterization of the Fed's "two-headed policy."

Whether you agree with Lowenstein's premise that the role of government should be prevention, not intervention, or you question government's legitimate role in either, you will probably come to the same inescapable conclusion as Lowenstein. At some point, a large institution will have to be allowed to fail to discourage risky lending in the future.

Caroline Baum is a columnist with Bloomberg News in New York.