Performance Perspectives Blog

Risk and numbers … not always so perfect together

by | Jul 20, 2014

Risk Is Never a Strict Numbers Game

In this weekend’s WSJ there’s an article titled “Risk Is Never a Strict Numbers Game.” Well, you know I had to take a look. It comes from “The Norm Chronicles: Stories of Numbers About Danger and Death,” by Bastland and Spiegalhalter. On the surface, this would appear to be just the kind of book someone involved with investment risk measurement would be interested in (okay, we don’t typically get involved with death statistics, but “danger” is something we have concerns with).

For perhaps obvious reasons, the title reminded me of this quote from Yale University Endowment’s David Swensen: “Quantitative measures of risk for individual portfolios leave much to be desired” (from Pioneering Portfolio Management). It’s been a referenced quotation in our Fundamentals of Investment Performance Measurement for more than a decade.

The authors begin by telling us that “Humans don’t assess risk strictly by numbers.” I found this to be a bit ironic, given the recent news of the widow of a smoker who died of cancer being awarded $23.6 billion. Who hasn’t known that smoking causes cancer? Cigarettes were called “cancer sticks” long before I was born. But, that’s a different matter. My point: people smoke regardless of the numbers: the numbers mean very little.

If we took numbers into consideration,  travelers wouldn’t be afraid to fly, given the very low number of deaths (tell that to Raymond Babbit!). But emotions get in the way, and many suffer from pteromerhanophobia (yes, I had to look that one up).

The article suggests that numbers can help us approach risk, and suggests that “for acute risks, a good measure is the MicroMort, devised by Stanford’s Ronald A. Howard in the 1970s. One MicroMort (1 MM) is equal to one-in-a-million chance of death.” I guess it’s a good thing to know that someone’s gone to the trouble of coming up with a statistic, but it’s unlikely that many will be swayed when they’re told the number of MMs an action might carry.

Contrast smoking and flying: I’m sure smoking carries with it many more MMs than flying, but do new smokers think much about that, or those who fear flying (and how many of those who fear flying are smokers?).

With investments, we struggle with numbers, but we’ve got them … plenty. If there’s one thing we’re certain of, we could use a few more. There’s much debate in our investment society today regarding which ones to show, how to calculate them, whether ex ante should rule (vs. ex post),  and more. When we cover the risk section in our Fundamentals, I end by restating Swensen’s quote, as I think we make his case.

Investment risk measurement is very much a numbers game: reporting the correct numbers and getting them right is our challenge. As a lover of numbers, I enjoy spending a lot of my time with them. But I realize they don’t always tell us everything we need to know. Can we do better? Of course, and we’re trying, but don’t expect anything miraculous anytime soon.