What is risk management all about anyway? We use the words as though everybody understands what we are talking about. But life is not that simple.

Risk means more things can happen than will happen—which is a fancy way of saying we do not know what is going to happen. If we do not know what is going to happen, some surprises will be damaging, but others will make us wealthier than we had anticipated. For anyone wrestling with the task of managing risk, it’s important to understand both the potential losses and the potential gains. As we’ll see, if more risk managers had thoroughly assessed the former, the current financial crisis might never have come to pass.

In today’s world, mathematical analysis has become the fashionable method for managing risks. This has been most obvious on Wall Street, where the so-called Quants, many of them with PhDs in math or theoretical physics, used their arcane modeling skills to conjure profitable trading models and new products. Non-Quants were often afraid of these new techniques, but liked the profits they produced and shied away from debating whether or not they were reliable.

In fact, math can only take us part way toward managing risk. What it cannot manage is what is at the heart of risk: the unknown. With a mathematical approach alone, the risk manager risks never correctly defining the problem and never asking the right questions. Let me say the same thing from another perspective: risk management really means attempting to optimize outcomes under conditions in which the full range of outcomes might be unknown. Numbers will be a frail reed in such an environment.

How, then, can we come to grips with risk management when mathematics fails us? I begin the answer to this question with an anecdote from my experience managing other peoples’ money back in the late 1960s. The story is interesting in its own right, but it also provides a profound insight into the essence of risk management.

A married couple came to see us at our offices one day, scruffily dressed and shy in manner. The man, John, opened the discussion by showing us their portfolio, which was worth a couple of hundred thousand dollars and consisted of only three stocks: AT&T, U.S. Steel, and Thiokol. The whole portfolio was on margin, but each position showed enormous profits. We were nonplussed and asked John why he needed us. Then he told us their story. He had been a reporter for the Brooklyn Eagle, but the paper had folded and he was now unemployed. At the time that he was fired, the couple had $15,000 in the bank, just about the amount they spent in the course of a year to meet their basic needs (don’t forget this episode took place over 40 years ago). John’s wife was a school teacher, so they had some money coming in—but nowhere near enough to cover their basic living expenses.

Unlike many people who hoard small accumulations of assets for fear of losing the little they have, John decided to go into the stock market and shoot the moon. “If I lose,” he figured, “I’ll just be broke a year earlier than I would be otherwise. If it works, though, we can be comfortable for the rest of our lives.”

That logic is what made John a master risk manager. The key question in decision making is, “What happens if I am wrong?” or, as in John’s case, “What happens if I lose?” In this instance, he could see that he and his wife were in such bad shape that they had little to lose but much to gain by taking this highly concentrated and leveraged risk in the stock market. (It did work. The problem John wanted my firm to help with was a different, though not uncommon one. John’s wife, who had been a firm companion while the great experiment was under way, now had high anxieties about the risk of losing their windfall. But John, transfixed by success, was psychologically incapable of unwinding his fantastic success. That was the role they wanted us to play.)

John was unaware of the hallowed history of his line of analysis, which dated back nearly four centuries to one of English literature’s most glorious achievements, Shakespeare’s Hamlet and Hamlet’s most famous soliloquy, “To be, or not to be.” There is a powerful connection between the investment decision by an unemployed newspaper reporter and the contemplation of suicide by the Prince of Denmark (which is what he meant by the phrase, “To be, or not to be”).

Hamlet always deliberated at length before acting on his impulses. In this crucial scene in the play, he is considering what might happen to him if he does commit suicide: “To die: to sleep; no more; and by a sleep to say we end the heartaches and the thousand natural shocks that flesh is heir to.” An end to heartache. That is the outcome Hamlet hopes he will achieve. But then he asks a critical question: can we be certain of “no more”? Hardly. We can be certain of nothing that happens after death.

That being the case, suicide is not a riskless act. Suppose there is more to come after death? “To die, to sleep; to sleep: perchance to dream: ay, there’s the rub; for in that sleep of death what dreams may come . . . There’s the respect that makes calamity of so long life.” Hamlet prefers to suffer the immense pain of life rather than face the unknowable consequences of suicide: “to grunt and sweat under a weary life” because “the undiscovered country from whose bourne no traveler returns, puzzles the will, and makes us rather bear those ills we have than fly to others that we know not of.” Hamlet left a perfect model for my client John to follow: what happens if I am wrong in taking this step? In John’s case, the consequences of being wrong were minor compared with the transformation he could achieve if his decision turned out to be the right thing to do. Hamlet’s choice of whether to be or not to be was precisely the opposite, because not acting was the better selection. But both John and Hamlet asked the right question.

Another example of the power of this line of analysis is Blaise Pascal, the great 17th century French mathematician who first articulated the use of probability in making estimates about the future. Despite the tremendous importance of Pascal’s work in the field of probability, his more fundamental contribution to risk management is similar to Hamlet’s, although I doubt that Pascal ever read any Shakespeare—or had even heard of Shakespeare. Like Hamlet, however, Pascal applied common sense, rather than higher mathematics, in managing the risk he faced.

Pascal was a deeply religious man, although he was also a skilled gambler—which is what led him to the study of probability in the first place. One day, in the silence of his room, Pascal asked himself a tantalizing question: “If I had to make a bet on whether God is or God is not, how should I respond?” He immediately recognized that “reason cannot answer.” Nevertheless, this was clearly a question of overwhelming importance, even though he would have to wait until the end of eternity to find out how his bet on God’s existence had worked out.

Pascal finally turned to an indirect approach to this puzzle, in which reason would help him decide which way he should bet on whether God is or God is not. He could choose what kind of a life he wanted to live: a life of virtue, restraint, and unselfishness, or a life of lust and self-indulgence. Suppose he chose a life of virtue and abstinence and then found out at the end of time that God is not? He would have passed up some goodies unnecessarily, but it would not have been such a bad life. If, on the other hand, he chose a life of lust, deceit, and self-gratification, and then found that God is, he would be in big trouble. Therefore, Pascal decided he must bet that God is and choose the virtuous life, even though he had no way of knowing whether the bet would pay off. One thing he did know: the consequences of being wrong by betting that God is not and leading a life of sin could have intolerable consequences.

The heart of the matter is defining the choices. Shoot the moon or leave the $15,000 in the bank? To be, or not to be? God is or God is not? John, Hamlet, and Pascal all provide the same framework for risk management: whichever road you choose—whether you play it safe or take a risk—what are the consequences of being wrong?

In our own time, let us suppose that some of the people who took out subprime mortgages, or some of the lenders, had stopped for a moment to ask themselves, “What if we are wrong about the outlook for housing prices? I am confident they will keep rising, but nothing is certain in this life. I better think through the consequences if I am wrong.” The consequence of being wrong about the trend of housing prices would be disaster, as the declining value of the mortgaged homes would soon wipe out the borrowers’ paper-thin equities. This tragedy is precisely what did happen when housing prices began falling in autumn 2006. If buyers or lenders had only taken the trouble to ask the question about being wrong, the history of the world economy could have been entirely different from the chaos that has overtaken us.

Thinking through what might happen if things go wrong is also a useful approach in routine business decision making. In 1994, Avinash Dixit and Robert Pindyck wrote a book called Investment Under Uncertainty, which emphasizes the inherent irreversibility of most business-investment decisions. Dixit and Pindyck suggest a method for dealing with the risks posed by irreversibility: wait before acting. Waiting is valuable because the passage of time can produce additional information. This new information may have value that would not have been available if an irreversible decision had been made earlier. Thus, Dixit and Pindyck’s case has its roots in the deliberations of Hamlet and Pascal, because taking the risk of not waiting can have negative consequences. Waiting to make an irreversible decision may lead to a more rational choice than not waiting. My client John, Hamlet, and Pascal, each in his own way, laid out the fundamental principle of risk management by defining the right problem and asking the right question. A purely mathematical approach would make sense only if the risk manager were indifferent to the possible outcomes, in which case, there would be no point to the exercise. There is always uncertainty. Nothing is 100 percent sure. While a 95 percent probability is statistically significant, that still leaves us in the dark about the remaining 5 percent; we may decide to accept that uncertainly and bet on the 95 percent sure thing, but there is still a possibility of being wrong. The crucial question to ask is, “What would be the consequence if that 5 percent chance comes to pass?”

Back to top