Risk register · entry
Q4 · Where models dieLTCM
Two Nobel laureates, 25-to-1 leverage, and a model that met reality.
The world stops matching the model. Regime change and leverage turn a small error fatal.
Why this quadrant
The trades were small, well modelled spreads that looked close to riskless one at a time. What the model did not carry was how alike they were. When Russia defaulted in August 1998 the positions stopped being thousands of separate bets and became one, and at 25 to 1 leverage the fund had four months of capital, not four years.
The record
- Founded 1994 by John Meriwether; initial capital ~$1.01 billion by Feb 24, 1994certain
- Partners Myron Scholes and Robert Merton won the 1997 Nobel Prize in Economicscertain
- Annual returns: 21% (1995), 43% (1996), 41% (1997)certain
- Early 1998 equity ~$4.7-5 billion vs. borrowed funds ~$124.5-125 billion (~25-to-1 leverage, one source says ~30-to-1)likely
- Off-balance-sheet derivatives notional ~$1.25 trillionlikely
- Russia devalued the ruble and defaulted on 281 billion roubles (~$13.5 billion) of debt on August 17, 1998certain
- Equity fell from ~$2.3 billion (Sept 1) to ~$400-600 million (Sept 22-25, 1998)likely
- Total losses ~$4.6 billion over about four months in 1998certain
- Buffett/Goldman/AIG offered ~$250 million for the fund's positions on Sept 23, 1998, rejected within the hourlikely
- Fed-brokered bailout: 14 banks injected $3.625 billion (sometimes rounded to $3.5-3.6 billion) for 90% of the fund, Sept 23, 1998certain
- Fund dissolved in early 2000; rescuing banks recovered their investmentlikely
- Model treated the 1998 loss event as roughly a 10-standard-deviation occurrence under a Gaussian distributionuncertain
Sources
The newsletter
One risk story a week, taken apart the way this one was: what was known, what was ignored, and which quadrant it really belonged to.
No spam, one email a week.