Models.Behaving.Badly

Disturbed, disillusioned and ashamed: Those aren’t emotions you expect a Wall Street quant to express when asked why taxpayers were obliged to bail out wealthy bankers.Unless, of course, the quant is Emanuel Derman, a particle physicist and former head of quantitative finance at Goldman Sachs Group Inc. (GS)“I am ashamed at the hypocrisies of the system,” Derman writes in “Models.Behaving.Badly,” an erudite yet pleasantly readable exploration of why financial models failed during the U.S. mortgage meltdown and why modelers must learn to use them more wisely.“We were told not to expect reward without risk, gain without the possibility of loss,” he says in disgust. “Now we have been forced to accept crony capitalism, private profits and socialized losses, and corporate welfare.”Unlike many quants, Derman says he wasn’t surprised that models failed in 2007, as events predicted to happen “once in 10,000 years happened every day for three days,” as one strategist at Lehman Brothers Holdings Inc. (LEHMQ) put it. The breakdown, Derman argues, flows from a misunderstanding of the difference between models and theories.Theories are attempts to uncover the hidden principles underpinning the world around us, as Albert Einstein did with his theory of relativity. Models, says Derman, are metaphors -- analogies that describe one thing relative to another. You can’t have a grand unified theory of securities, he says, because markets are driven by ephemeral human opinions.As the coy punctuation in the book’s title suggests, Derman is interested in three things: models, models that behave, and models that behave badly. Designing a model based on fickle human behavior is like “trying to force the ugly stepsister’s foot into Cinderella’s pretty glass slipper,” he says. “It doesn’t fit without cutting off some essential parts.”That doesn’t mean models are a waste of time, he says. It just means that we must use them with humility and common sense.Derman, a professor at Columbia University, is known for his memoir, “My Life as a Quant.” In “Models.Behav- ing.Badly,” he takes us on an intellectual journey that compares the glories of theoretical physics to the messiness of economics.He begins with a notoriously flawed model -- the apartheid of his boyhood South Africa -- and then, quietly and unpretentiously, guides the reader through topics ranging from Baruch Spinoza’s analysis of human passions to Paul Dirac’s discovery of the positron as a hole in a sea of electrons.This material isn’t always easy to digest, especially if your facility with algebra has faded. Yet Derman tugs the reader along with limpid writing and curios such as the origin of the word “magnet” (from lodestones found in ancient Magnesia) and the Tetragrammaton (the name of the name of God, the four Hebrew letters usually transliterated as YHWH, or Yahweh).The book’s fulcrum comes in two chapters called “The Sublime” and “The Inadequate.” These compare the history of quantum electrodynamics -- complete with diagrams illustrating the achievements of James Clerk Maxwell -- to the development of the efficient-market hypothesis. The elegant success of the former becomes a foil for the failings of the latter, which Derman calls “a model of a hypothetical world rather than a correct hypothesis about the one we inhabit.”The Efficient Market Model, as he dubs it, is based on the notion that stock prices move like pollen particles drifting through the air. Call it a random walk, diffusion or Brownian motion; it amounts to a metaphor, not a fact.Economists concocted the model, Derman says, because of a simple reality: You can’t consistently predict what the market will do tomorrow based on what you know today. So they asserted that it’s impossible to beat the market because current prices already reflect all available information.“Converting their failed attempts at systematic stock price prediction into a fundamental postulate of their field was a fiendishly clever jujitsu response on the part of economists,” he writes.Derman has distilled a lifetime of reading, research and thinking into these pages, and I read the book twice to see how he pulled the threads together without losing the reader.His down-to-earth examples help. He uses the Law of One Price, for instance, to estimate the value of an eight-room penthouse on Park Avenue in New York. The Black-Scholes Model for valuing stock options is, in turn, likened to making fruit salad: If you know the price of the fruit that goes into the salad (shares in, say, Tutti Frutti Bop Inc.), you know what the salad (an option on those shares) is worth.Equally appealing is Derman’s empathy for his fellow human beings: “One has to treat people as responsible for their actions,” he says, “and yet also recognize that they can’t help what they do.”One wonders if he was thinking about his old friends at Goldman as he wrote that.