Long-Term Capital Management

What happens when you try to manage a complex system with tools designed for merely complicated problems.

Chez Guevara — Dining in style at the Disaster Café™ →Go uptown {{{3}}} Extract from the JC’s financial disasters roll of honour Scandal Date Where Loss Reason Firings Jail-Time? LTCM 1998 US $3.6 bn bailout orchestrated by the Fed Black-Scholes option pricing model attacked them from nowhere Everyone out of a job, Master of the Universe-sized reputations (John Meriwether, Myron Scholes, Robert C. Merton) cut to ribbons No, though fourteen bank CEOs were locked in a room by the Federal Reserve for a weekend so that is something. edit template Index: Click ᐅ to expand: Disaster Café Tell me more Sign up for our newsletter — or just get in touch: for ½ a weekly 🍺 you get to consult JC. Ask about it here.

See also: Hubris. and Charles Perrow’s super book, Normal Accidents, which wasn’t really about the financial markets but sure could have been.

LTCM was a hedge fund, founded in 1993, and stuffed to the gunwhales with splendid brainboxes and Nobel prize-winners, including at least one of the team who “solved” the problem of how to accurately price options with the Black-Scholes option pricing model. LTCM used their braininess, and the Black-Scholes model, to engage in leveraged arbitrage, ultimately doing the world the large favour of testing the Black-Scholes model to destruction.

Destroy it they did, alas, in the process destroying their fund, and nearly taking the entire financial system with them.

But at least we now know that the Black Scholes model is only reliable when you don’t really need it, in times of relative market calm, so everyone has learned only to use it to manage positions that do not have any tail risk, not even once in a gazilion years, anymore.

Then those “ten-sigma” events — like, ooooh, say the correlation of a Russian government default with a spike in the price of all other G20 Treasury securities, just to pick something at random — that should, in the world of normal distributions, happen only once in every 1 x 10²⁴ times — say, every hundred million years or so — but, since investment decisions are not, even remotely independent events, happened once — and only needed to happen once, to blow Long Term Capital Management and much of the market to smithereens — just four years after they started trading.

See also

• Systems analysis and complexity
• Normal distribution

A good collection of the JC’s book club books relate.

• Normal Accidents
• Inventing Money
• When Genius Failed
• Models.Behaving.Badly