I have written before about the impressive economic insights offered by England’s John Kay, leading Oxford University economist, former Director of the Oxford Business School, and currently a columnist for The Financial Times. Let me relay to you the essence of his thoughts regarding the break-down of economic models in predicting extreme outcomes, such as the financial crisis of September 2008.
“If the water in your glass turns to wine, you should consider more prosaic explanations before announcing a miracle. If your coin comes up heads 10 times in a row – a one in a thousand probability – it may be your lucky day. But the more likely reason is that the coin is biased, or the person who flips the penny or reports the result is cheating. The source of most extreme outcomes is not the fulfillment of possible but improbable predictions within models, but events that are outside the scope of these models.” John Kay, ‘Don’t blame luck when your models misfire’, Financial Times, March 2, 2011
Some sixty years ago, a French economist, Maurice Allais described what would become known as the Allais Paradox. This paradox is based on the discovery that most individuals treat very high probabilities quite differently from certainties. According to John Kay, such individuals are right to make such a sharp distinction. Why? Because there are no 99 percent probabilities in the real world.
Very low and very high probabilities are artifices of models. The probability that any model perfectly describes the world is much less than one. Once the probabilities delivered by the model are compounded with the unknown, but large, probability of model failure, the reassurance provided by any model simply disappears. This is especially true when models are relied upon to predict extreme events.
Armed with this insight, John Kay lays into the observable failure of ‘value at risk’ modelers to learn from past experience. For example, the European Union is ploughing ahead with its Solvency II directive for insurers – which is explicitly modeled on the failed Basel II agreements for monitoring bank solvency. Solvency II requires that businesses develop models that show the probability of imminent collapse as being below 0.5 percent.
In John Kay’s judgment, insurance companies do not fail for the reasons described in such models. They fail because of events that were unanticipated or ignored. They fail because underwriters misunderstand the risk characteristics of their policies, as at AIG or because of fraud, as at Equitable Life Fundings.
Huge standard deviations – or multiple sigma – events simply do not occur in real life. Economists, financial analysts and statisticians, who purport to provide objective means of controlling for such risks are frauds, much of a piece with the practitioners of alchemy and quack medicine. Such fraudsters incite gullible outsiders to believe that they are dealing with professionals blessed with genuine expertise. For this reason, they wreak havok on an unsuspecting world.
“We will succeed in managing financial risk better only when we come to recognize the limitations of formal modeling. Control of risk is almost entirely a matter of management competence, well-crafted incentives, robust structures and systems, and simplicity and transparency of design.” John Kay, ibid.