Wednesday, June 2, 2010

Diebold et al., The Known, the Unknown, and the Unknowable in Financial Risk Management

In 1995 Ralph Gomory wrote an essay in Scientific American that read, in part, “We are all taught what is known, but we rarely learn about what is not known, and we almost never learn about the unknowable. That bias can lead to misconceptions about the world around us.”

The Known, the Unknown, and the Unknowable in Financial Risk Management: Measurement and Theory Advancing Practice (Princeton University Press, 2010), a collection of fifteen papers edited by Wharton professors Francis X. Diebold, Neil A. Doherty, and Richard J. Herring, applies the KuU conceptual framework to the financial world. It is a bold book, tackling both theory and practice and spanning the worlds of (among others) banking, insurance, real estate, and investment. It is also utterly engrossing.

There is no way I can do justice to the wealth of material in this book. It incorporates ideas from behavioral finance, statistics, and epistemology; it addresses a range of questions from portfolio management and crisis management (and, no, portfolio management should not be a case study in crisis management) to corporate governance. For the “idea junkie” it’s a real high.

Let me simply touch on a few salient points.

In their introduction the editors suggest that although K situations are often relevant, u and U are of equal or greater relevance in financial risk management. But how can we shed light on the unknown and how can we even begin to approach the unknowable? The editors propose, defining the framework of the book, that we need better data and better theory—and that the two are mutually reinforcing. As we expand our knowledge u can slowly be brought under the umbrella of K. Even U can sometimes be “tamed.” The authors write, “To the extent that U represents a failure of imagination . . . the collection and analysis of data regarding near misses—disasters that were narrowly averted—may provide a window into the domain of U and alternative outcomes.” (p. 7)

Richard Zeckhauser advocates investing wisely in the unknown and unknowable (and, he adds, the unique). He proposes that as long as an investor has a complementary skill—for instance, unusual judgment à la Warren Buffett—he can reap outsized returns because competition is often limited and prices way out of line. And if the investor doesn’t have that skill, he can “ride along in a sidecar pulled by a powerful motorcycle.” (p. 314) He cautions, however, against jumping on the quant bandwagon. The quant world is currently overpopulated, which means that quants may become the new shoeshine boys. That is, “when your math whiz finance Ph.D. tells you that he and his peers have been hired to work in the XYZ field, the spectacular returns in XYZ field have probably vanished forever.” (p. 321)

A Mandelbrot and Taleb thesis: “Large moves beget large moves; markets keep in memory the volatility of past deviations. A subtle concept, fractal memory provides an intrinsic way of modeling both the clustering of large events and the phenomenon of regime switching, which refers to phases when markets move from low to high volatility.” (p. 55)

And, finally, a geek note from the fascinating paper by Colacito and Engle on the term structure of risk. It’s obviously not new since it dates to a 1998 paper, but it was new to me. Anyone who trades volatility should note that “the common practice of converting 1-day volatilities to T-day estimates by scaling by [the square root of T] is inappropriate and produces overestimates of the variability of long-horizon volatility.” (p. 66)

Although this book is most obviously addressed to risk managers and regulators, I think it should be read by every intellectually curious person with skin in the financial game. If the investor or trader doesn’t come away with at least one or two ideas of practical importance to his financial life, he is a “sleepreader.”

No comments:

Post a Comment