Currency Wars_ The Making of the Next Global Crisis - James Rickards [100]
FIGURE 1: A bell curve showing a normal distribution of risk
The problem with the Nobel Prize–winning theories based on the bell curve is that empirical evidence shows they do not correspond to the real world of markets and human behavior. Based on an enormous body of statistical and social science research, it is clear that markets are not efficient, that price movements are not random and risk is not normally distributed.
The academic counterattack on these tenets of financial economics have come from two directions. From the fields of psychology, sociology and biology came a flood of studies showing that investors are irrational after all, at least from the perspective of wealth maximization. From iconoclastic mathematical genius Benoît Mandelbrot came insights that showed future prices are not independent of the past—that the market had a kind of “memory” that could cause it to react or overreact in disruptive ways, giving rise to alternating periods of boom and bust.
Daniel Kahneman and his colleague Amos Tversky demonstrated in a series of simple but brilliantly constructed experiments that individuals were full of irrational biases. The subjects of their experiments were more concerned about avoiding a loss than achieving a gain, even though an economist would say the two outcomes had exactly the same value. This trait, called risk aversion, helps to explain why investors will dump stocks in a panic but be slow to reenter the market once it turns around.
When economists began searching capital markets data for the kinds of irrationality that Kahneman and Tversky had demonstrated, they had no trouble finding it. Among the anomalies discovered were that trends, once set in motion, were more likely to continue than to reverse—the basis of “momentum” investing. It also appeared that small-cap stocks outperform large-cap stocks. Others identified the so-called January effect, which showed that stocks performed better in January than other months. None of these findings are consistent with efficient markets or random price movements.
The debate between the efficient markets theorists and the social scientists would be just another arcane academic struggle but for one critical fact. The theory of efficient markets and its corollaries of random price movements and a bell curve distribution of risk had escaped from the lab and infected the entire trading apparatus of Wall Street and the modern banking system. The application of these flawed theories to actual capital markets activity contributed to the 1987 stock market crash, the 1998 implosion of Long-Term Capital Management and the greatest catastrophe of all—the Panic of 2008. One contagious virus that spread the financial economics disease was known as value at risk, or VaR.
Value at risk is the method Wall Street used to manage risk in the decade leading up to the Panic of 2008 and it is still in widespread use today. It is a way to measure risk in an overall portfolio—certain risky positions are offset against other positions to reduce risk, and VaR claims to measure that offset. For example, a long position in ten-year Treasury notes might be offset by a short position in five-year Treasury notes so that the net risk, according to VaR, is much less than either of the separate risks of the notes. There is no limit to the number of complicated offsetting baskets that can be constructed. The mathematics quickly become daunting, because clear relationships such as longs and shorts in the same bond give way to the multiple relationships of many items in the hedging basket.
Value at risk is the mathematical culmination of fifty years of financial economics. Importantly, it assumes that future relationships between prices will resemble the past. VaR assumes that price fluctuations are random and that risk is embedded in net positions—long minus short—instead of gross positions. VaR carries