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

by Roger Lowenstein

The models didn’t order them to trade; they provided a contextual argument for the human computers to consider. They simplified a complicated world. Maybe the yield on two-year Treasury notes was a bit closer than it ordinarily was to the yield on ten-year bonds; or maybe the spread between the two was unusually narrow, compared with a similar spread for some other country’s paper. The models condensed the markets into a pointed inquiry. As one of the group said, “Given the state of things around the world—the shape of yield curves, volatilities, interest rates—are the financial markets making statements that are inconsistent with each other?” This is how they talked, and this is how they thought. Every price was a “statement”; if two statements were in conflict, there might be an opportunity for arbitrage.