“Predicting the Unpredictable”

Sydney M. Williams

Thought of the Day

“Predicting the Unpredictable”

January 11, 2013

Wall Street has always been an El Dorado, a place of almost infinite possibilities where allegedly alchemists deftly spin theorems into gold. It has always attracted the best and the brightest, but also the greediest and most nefarious. Its siren call has beckoned the most creative among us, for self, not public, interests. They come to Wall Street because, as Willy Sutton once said about banks, it’s where the money is. In recent years, as the power of computers have become more intrusive, the pull of the Street’s gravity has yanked mathematicians and scientists from the halls of academia to the immense trading floors of Wall Street banks and hedge funds. And, throughout it all, while humility is in short supply, hubris is not.

Into this realm has arrived yet another book designed to prove that it was not the quantitative geniuses that almost did in Wall Street five years ago. According to this author, it was the misuse of algorithmic models by those who didn’t fully understand what they had. The road to salvation, according to The Physics of Wall Street, requires a more comprehensive use of physics than we have witnessed heretofore, not less. The book has been written by James Owen Weatherall, professor of logic and philosophy of science at the University of California, Irvine. The subtitle of his book is “A Brief History of Predicting the Unpredictable.” Professor Weatherall cites as Exhibit A, the phenomenal success of James Simon’s hedge fund, Renaissance Capital and the fact that they do not hire financial experts. A third of their 200 employees have PhDs in math, physics and statistics. Other firms hire quantitative analysts, but their decisions, bemoans Professor Weatherall, are too often over-ridden by their associates whose qualitative skills take precedence over their quantitative ones. Certainly, though, Mr. Simons’ success cannot be denied.

But there are, to this observer whose quantitative skills lag those of his eleven-year old grandson, at least two risks associated with a growing reliance on sophisticated modeling techniques. One is the very unpredictability of human behavior. And the second is that predictability, in any measure, leads to an excess of confidence by the speculator. Audacity can cause the speculator/investor to forego what advantage he or she may have had by applying unreasonable amounts of leverage to what otherwise might have been a sensible trade.

Behavior economics may be a relatively new science, but its genesis is as old as mankind. It captures, as Adam Smith well understood, the essence of why people respond to different stimuli. The Russian physiologist, Ivan Pavlov, was famous for noting how people (and dogs) can be conditioned to respond in predictable ways. Hercule Poirot, Agatha Christy’s famous diminutive Belgium detective with large mustaches, was always talking of the psychology of victims and suspects. Consumer companies have long understood the need to condition their customers, and certainly successful politicians know how to appeal to both the best and the worst instincts of their constituents. But measuring response is as much an art as it is a science, as every action causes a new and different response. Reducing the study of markets and companies to individual algorithmic equations can save time and provide a backdrop against which conclusions may be compared, but subtleties and a sense of value are not so easily quantifiable.

The second problem is perhaps more dangerous. There is little question that supposed knowledge as to how people and markets should respond to predetermined stimuli provides a level of comfort, which translates into greater confidence and that, in turn, leads to a greater use of leverage. If we assume that a U.S. Treasury Note or Bond is the most riskless investment available to a U.S. taxpayer, anything that deviates from that return involves more risk. And the greater the deviation, the greater the risk. Much of quantitative theory deals with controlling risk. Risks may be quantifiable, in that normalized spreads between one asset class and another, or normalized interest rate differentials are theoretically calculable. But risk may be qualitative, in that they deal with unknowns, such as management, forecasting skills, interest rate movements, business and economic cycles, or behavioral changes. The problem develops when an investor is inoculated with more confidence than he should have because of conviction in his models, a confidence that causes him to increase leverage in search of higher returns.

In his review of Professor Weatherall’s book in last Sunday’s New York Times, Floyd Norris writes that the basic problem that caused the near financial meltdown in 2007-2008 was not that models don’t work – engineering models, after all, have been used for generations in bridge construction – but that they became broadly and unquestionably accepted on the part of banks and regulators. It “allowed an entire system to develop out of models that encouraged a lot of borrowing in order to transform tiny profits into big ones. The result was that rather than losing one bridge we came very close to losing them all.”

Wall Street is better defined as Winston Churchill once described the Soviet Union foreign policy – a puzzle inside a riddle, wrapped up in an enigma. No one has fully mastered Wall Street. But a few have done extraordinarily well, giving aspiration to others. It reminds one of the store that sells lottery tickets, and puts up signs advertising a few past winners but, of course, never acknowledging the thousands of losers – the purpose being, of course, to attract more buyers, which it does.

While Professor Weatherall writes well and is entertaining, no one should believe he has unlocked an understanding of Wall Street. He has not. And Wall Street plays too important a role in all of our lives, no matter what we do, to be treated so trivially. Despite its image of greed and cronyism, it incorporates our investments and those of millions of people with their IRAs, 401Ks and pension plans. The fact that it is indecipherable and unpredictable doesn’t mean it cannot be a force for good. It generally has been. Writers like the Professor add to the mystique that is Wall Street, but should not be taken too seriously.