John Meriwether, Former Salomon Brothers trader, Uses Leverage
October 19, 1998
Archimedes on Wall Street
John Meriwether seemed to have a magic touch. What he really had was nerve-wracking leverage.
By Robert Lenzner
"GIVE ME but one firm spot in which to stand," Archimedes declared, "and I will move the earth." While the ancient Greeks discovered the principle of leverage some 2,200 years ago, it was John Meriwether and his Long-Term Capital Management LP that showed how far leverage could take you in financial speculation.
With $4.8 billion in capital, Meriwether controlled $160 billion in stocks and bonds, plus derivatives with a notional value of $1 trillion. That's 240-for-1. As we will explain, it got even worse than that.
This was sophisticated, but it was gambling. At the time it began to sink, unable to meet margin calls, Long-Term Capital was, for example, long large amounts of the 29-year Treasury bond and short equally large amounts of the 30-year Treasury. The fund got into this bet to capitalize on the tiny spread in yields between the two maturities. Tiny indeed. In June the difference was just five basis points—a 29-year, $1,000 bond yielding 50 cents a year more than a 30-year bond.
It didn't make sense that this yield differential would persist in these very similar securities. Meriwether and his friends apparently figured that the yields would converge. So they sold short the lower-yielding bond and bought the higher-yielding bond.
This is what you call a nondirectional bet. Long-Term Capital wasn't betting that interest rates would rise or that they would fall. It was just betting that the yields would converge. If they did and the bond market went down, Long-Term Capital would make more on its shorts than it would lose on its longs. If the market went up the fund would make more on its longs than it would lose on its shorts. Meriwether was simply betting that the two bonds would eventually carry the same yield.
Without leverage, that bet is scarcely worthwhile: With a discrepancy of 5 basis points, you would make $5,000 on a $1 million trade when the discrepancy is eliminated. That's not the kind of profit that causes people to thrust billions of dollars at you, as very smart investors thrust at Meriwether and his partners (who included Nobel economists Robert Merton and Myron Scholes).
But introduce the Archimedes principle and the picture changes. Suppose you were able to buy $1 million worth of Treasury strips on $10,000 in margin. Now that $5,000 profit isn't just 5% on your money—it is 50% on your money.
That, somewhat simplified, is how Long-Term Capital lost its shirt and nearly crashed world markets in the process. Of course, Meriwether and his pals weren't just playing that single discrepancy on Treasury bonds. They were making all kinds of similar but often much more complex transactions. These involved repos and all kinds of derivatives, but the principle was usually the same: Don't bet on the direction of markets; bet on relationships among different securities.
It was Meriwether's bad luck that instead of converging, the yields on those two Treasury issues drifted farther apart. At the end of September the difference in their yields was not 5 basis points but 15.
That loss was costly but not lethal. But too many of his bets then went bad simultaneously, as investors around the world rushed into safe investments and sold riskier ones. A lot of discrepancies that the partners thought would narrow instead got wider. Long-Term Capital had been betting—among other things—that the yield differential would shrink between similar sovereign bonds of other G-7 nations. Instead of shrinking, the gap yawned much wider.
Here's where this gets scary. In some cases Long-Term Capital hadn't put up a 1% or 2% margin. It hadn't put up a dime; the brokers and bankers were lending 100% of the value of the collateral. Much of the firm's $4.8 billion in capital was simply kept in the bank to meet the occasional margin call.
With many of the trades suddenly under water, the banks made margin calls—just as your broker would if you owned stocks on margin and those stocks dropped in value. The $4.8 billion in capital dwindled very fast; it was down to $1.5 billion when the Fed stepped in. Soon it was $600 million. So why didn't the lenders just sell the collateral the way your broker would sell your stocks if you couldn't pay up?
Ever try selling $1 trillion worth of collateral for quick cash? Yes, modern markets have depth—but not that much depth. "There's an illusion of liquidity in times of volatility," says Henry Kaufman, the bond market guru who runs his own firm.
The mess forced the Federal Reserve to find saviors for a private firm instead of making it pay the full price for its mistakes.
Who's to blame ?
Meriwether and his sidekicks had a bad case of hubris. As Kaufman puts it: "There are two kinds of people who lose money: those who know nothing and those who know everything." With two Nobel prize winners in the house, Long-Term Capital clearly fits the second case.
Wall Street and the banks deserve a good deal of the blame, too—securities firms like Merrill Lynch and CS First Boston; banks like Switzerland's great UBS AG.
Banks and investment houses love these so-called hedge funds. The hedgers borrow huge amounts and are willing and able to pay a premium for borrowing. Maybe it's risky to lend 100% on collateral, but what's the harm if the borrower has billions in cash available to meet margin calls? A New York Stock Exchange member firm can earn five basis points by borrowing money and relending to a hedge fund; that works out to $500,000 a year for every billion. Easy money. Then there are commissions and a chance to share in the action.
Anyhow, the hedge funds aren't doing much that the banks and big investment houses aren't doing themselves. Most large securities firms and banks are to some extent hedge funds. Salomon Brothers, Inc. certainly was before the cautious Sandy Weill took it over. Bankers Trust calls itself a bank but makes its big money off spreads in derivatives.
As usually happens with a good thing, the derivatives game got overdone. History keeps repeating. Banks poured money into risky leveraged buyout loans and commercial realty loans in the 1980s. The banks lent to Mexico and Brazil in the late 1970s without knowing their total debt level.
Had they been forced to mark their portfolios to market—as Long-Term Capital was—most U.S. banks would have been insolvent in the late 1980s and early 1990s. Most investment banks were illiquid in October 1987 when the market collapsed because they had made huge bridge loans to finance risky takeovers in companies whose values were evaporating.
It now has happened again, and banks and other financial institutions will be writing off billions of dollars in bad loans and bad bets. Until it becomes clear how deeply our financial institutions were involved in financing and engaging in this kind of speculation, a cloud will hang over all the financial markets. A very dense cloud—because the officials who supervise banks and brokers have no clear idea what has been going on.
It seemed a lovely game while it lasted. Long-Term Capital made 43% for its investors in 1995 and 41% in 1996. That's after cutting principals like Meriwether in for 20% of the profits.
With returns like that, no wonder the chairman of Merrill Lynch and dozens of others at the firm invested in Long-Term Capital.
But, adjusted for the risks, how good really were those returns? It's the old story: Financial genius is a short memory in a rising market.
Copyright (c) 1998, Forbes Magazine