Why Risk Managers Should Be Spymasters
When he left physics, John Breit had the choice of a job in naval intelligence or on Wall Street. "My wife said I couldn't be a spy. She hates capitalism, but said to go to Wall Street," he told me recently. "And then I ended up running a spy network."
Calvin Trillin once attributed the financial collapse to the influx of smart people on Wall Street. The physicists, computer scientists and mathematicians displaced the slow-thinking country club types. With their incomprehensibly complex models, the smart guys' hubris brought our economy low.
Mr. Breit was part of that initial incursion. A Ph.D. in physics from Columbia, he was doing postdoctoral work when he realized that he could never be as good as his contemporary Edward Witten, who went on to pioneer string theory.
So in 1986, he joined Wall Street, moving not to the trading floor, like many of his fellow rocket scientists, but into risk management. In 1990, he took his skills to Merrill Lynch, rising to become the firm's head of market risk oversight. The physicist came to understand the limits of mathematical models. He learned that his job was really psychologist, confessor and detective. He became the financial version of a counterintelligence officer, searching for the missed clues and hidden dangers in the firm's trading strategies.
Mr. Breit is retired now, studying ancient Greek in his spare time and volunteering as an adviser for New York's pension funds. He comes across as George Smiley if he were a Southerner — gracious, reluctant to talk about himself, with iconoclastic opinions just below the surface. I've been talking to him periodically over the years about how giant financial institutions should manage the aggressive traders slinging giant sums around the world in ever more complex transactions.
After the Senate issued a report last month on JPMorgan Chase's multibillion-dollar "London Whale" trading loss, an incident where the mathematical modeling went seriously wrong, I reached out to him again.
That debacle encapsulates much of what is wrong about how banks manage their risk and how the regulators oversee those efforts. At JPMorgan Chase, the risk models hid — and were used to hide — risks from the traders and top executives. Too many measures and too many numbers undid the risk managers. But ultimately, they failed because of human frailties; the risk managers lost sight of their mission and tried to protect the traders and their trades. As in all spy debacles, the counterintelligence officers got co-opted.
Early in his career, Mr. Breit figured out that models for markets aren't like those for physics. They don't come from nature. It was necessary to know the math, if only so that he couldn't be intimidated by the quantitative analysts....MORE
