Home › Stories › Cautionary tales

LTCM collapse: when Nobel Prize math met a panic

In 1998, a hedge fund with two Nobel Prize winners on staff lost $4.6 billion in less than four months. The Fed had to call 14 banks into one room to stop the damage spreading.

Story · 7-minute read · Updated 2026-10-21

Who they were

Long-Term Capital Management (LTCM) was a hedge fund started in 1994 by John Meriwether, a famous bond trader from Salomon Brothers.

He built a dream team. Its partners included Myron Scholes and Robert Merton, two economists whose work on pricing options won the Nobel Prize in 1997.

Big investors lined up to get in. The fund's early years were very strong.

What happened

LTCM's main idea sounded safe. It looked for prices that were "out of line" with each other and bet they would come back together.

A classic example: two very similar US government bonds. One trades a tiny bit cheaper. LTCM buys the cheap one and sells the dear one, expecting the gap to close.

The catch? Those gaps are tiny. Each trade made very little on its own.

To make it worth doing, LTCM borrowed. A lot. By most accounts, at the start of 1998 the fund had about $4.7 billion of its own money and had borrowed more than $125 billion. That's roughly 25 dollars borrowed for every 1 dollar owned. Its derivatives positions were far larger still.

With leverage like that, a small move against you is not small.

The moment everyone remembers

On 17 August 1998, Russia devalued its currency and stopped paying some of its debts.

Investors around the world panicked. They all ran to the safest things they could find, like the newest US Treasury bonds. They sold almost everything else.

LTCM's bets needed the gaps between "safe" and "slightly less safe" to shrink. Instead, they blew wide open, all at once, in market after market.

Trades that the models treated as unrelated suddenly moved together. In plain words: the fund had spread its bets across many markets, but in a panic they all became the same bet.

Worse, everyone knew LTCM was in trouble. Other traders could guess its positions. Buyers vanished. Selling to get out would only push prices further against it.

In August alone the fund lost a large share of its money. By September it was close to the edge.

The Federal Reserve Bank of New York feared a fire sale could hurt the whole banking system. On 23 September 1998 it brought together 14 banks and investment firms. They agreed to put in about $3.6 billion to take over the fund and wind it down slowly.

No public money was used. But the Fed's role in arranging it was, and still is, debated.

In total, LTCM lost about $4.6 billion in less than four months. It was closed in early 2000.

What it teaches

1. Leverage turns small moves into big ones

Being "right on average" didn't matter. LTCM's borrowing meant a short-term move against it could wipe it out before it was proven right.

The same math works on any account. With high leverage, a small price move becomes a big share of your money. Read leverage and margin explained, then try the numbers in the margin calculator.

2. "Rare" events happen more than models say

LTCM's models were built from past data. The panic of 1998 was outside that data.

This is why we size every trade so that a loss, even an ugly one, stays small. A fixed slice of risk per trade (we use about 1%) is the plain version of the lesson. See how much to risk per trade.

3. Being able to get out matters as much as being right

When everyone wants out at once, there may be no one on the other side. That's liquidity drying up. A stop order can fill far from where you set it in a market like that (that gap is called slippage).

4. Many bets can secretly be one bet

LTCM thought it was diversified. In a panic, its trades moved together. That's correlation jumping when you least want it.

What it doesn't mean

• It doesn't mean math or models are useless. The same partners were very skilled. The problem was how much they borrowed against being right.
• It doesn't mean small traders face the same risk. Most retail accounts can't borrow anywhere near 25 to 1. US forex rules cap major pairs at 50:1, which is still a lot. The lesson is about how leverage works, not about your account size.
• It doesn't mean "never trade spreads". It means no strategy is safe enough to bet the whole firm on.
• The famous line "markets can stay irrational longer than you can stay solvent" is often attributed to John Maynard Keynes. There's no firm record he said it, but LTCM is the story people tell to explain it.

Timeline

Questions