El-Erian, When Markets Collide

Mohamed El-Erian’s When Markets Collide: Investment Strategies for the Age of Global Economic Change (McGraw-Hill, 2008) achieved best seller status and won prestigious awards despite mixed reader reaction. I’m not going to weigh in on the merits of the book because, if I did, it would be “on the one hand, on the other hand, on the third hand” dithering best reserved for economists. Rather, I’m going to take a different tack, focusing on one point that caught my attention when I first read the book. El-Erian’s prose may be turgid, but I think he highlights a potentially fertile area for increased investment returns.

Rather than diversifying simply by using asset classes—a method that has seen some deterioration in results of late, he suggests starting with “the risk factors that give rise to investment returns over time.” He continues: “The ideal situation is to come up with a small set (three to five) of distinct (and ideally orthogonal) risk factors that command a risk premium. The next step is to assess the stability of the factors and how they can be best captured through the use of tradable instruments.” (p. 233)

I’m certainly no match for the experts at Harvard Management Company or PIMCO, and even they have found it hard to execute this game plan. But they set a high bar for themselves because they want their solutions to be quantifiable and optimizable (and we all know that they demand much more mathematical and statistical rigor than standard backtesting programs provide). I assume they want to build on the work of such academics as Fama and French—for instance, their 1992 paper “Common Risk Factors in the Returns on Stocks and Bonds,” which is not exactly bedtime reading.

So let’s try to dumb down the idea a bit. Where can the independent trader or investor earn risk premia? Some argue that using two systems, one trend following and the other mean reverting, is an answer, but I’m not convinced that this is the brightest idea. The trader earns risk premia on the one hand and loses it on the other. Think back to Einhorn’s argument against pairs trading; though not exactly analogous, the general point carries over. So let’s move on. One fairly sophisticated possibility is to look at the signal-to-noise ratio of an instrument. (Refer back to my review of Drobny’s Inside the House of Money for a little more insight into this trading hypothesis.) Another possibility is to exploit volatility risk; there are many strategies appropriate here, for the most part involving the use of options. Overnight risk is another area worthy of exploration, possibly as part of a multi-timeframe trading program. The list is bounded only by one’s imagination.

Whatever the hypothesis, the trader or investor must backtest it to determine whether, if we believe that history repeats itself or at least rhymes, it has any merit and, if so, on what markets it’s most successful. A strategy that works on currencies, for example, may provide no edge when applied to stock indices. And, of course, it always bears repeating that it is nothing short of suicidal to take on risk that’s too large for the size of the portfolio.

Ideally, the trader or investor will end up with three to five distinct, uncorrelated strategies that, when combined, will outperform any single strategy as well as any appropriate benchmarks. Some money managers claim to do this, but I’m certainly not touting their wares. I’m a Connecticut Yankee, a do-it-yourselfer by inclination. What I’m suggesting is that experienced traders or investors seek to develop a few trading strategies that fall well within their psychological comfort zone and their risk profile and that, when pursued simultaneously, have the potential to outperform the standard benchmarks and the more finely tuned benchmarks. A little outperformance here, a little outperformance there, it all adds up.