Behavioral Portfolio Management: An Alternative to Modern Portfolio Theory

June 04, 2013

CFA Institute

Seeking to bridge the divide between modern portfolio theory and behavioral finance, is C. Thomas Howard’sBehavioral Portfolio Management.” Howard is professor emeritus at Daniels College of Business, University of Denver, and co-founder of AthenaInvest. Application of his behavioral portfolio management has resulted in Athena’s longest running portfolio, Athena Pure Valuation, generating a return over 11 years of 26.1%. Compare that to the Russell 2000 benchmark, which returned 10.6%. Further, Athena Pure is the top performing portfolio in the country over this time period when compared to the active equity mutual fund universe.

Howard’s paper on behavioral portfolio management was one of the most downloaded papers on the Social Science Research Network. While the paper is worth a read, why not get this story of an alternative point of view from the man himself?

CFA Institute: Could you please give an overview of behavioral portfolio management and what makes it unique?

C. Thomas Howard: Behavioral portfolio management is aimed at building superior portfolios based on the pricing distortions created by investor’s emotional behavior. The core of behavioral portfolio management focuses on the specifics of how to build portfolios based on behavioral factors.

We are now at a point where we can not only talk about the behavioral decision errors made by investors, but also are able to measure the price distortions resulting from these errors. While behavioral portfolio management rejects the basic tenets of modern portfolio theory (MPT), the careful and rigorous statistical analysis of historical data remains.

Instead of using these methods to show that markets are informationally efficient, they are used to identify measurable and persistent price distortions. And many have been found.

So in spite of the fact that behavioral is in the name, behavioral portfolio management’s recommendations are based on thorough statistical analyses. If it cannot be objectively measured and confirmed by large, long time period studies, then it is not used. At my core, I am an empiricist, and so if I do not see it in the data, then I do not believe in it for investing purposes.

So what is your thought about modern portfolio theory? It doesn’t sound as if you think it is so “modern” any longer.

MPT is at that awkward stage in which the evidence is overwhelmingly against it, but many professionals and academics have decided to stick with it anyway. In an ideal world, a model lives or dies based on the empirical evidence. But in the case of MPT, many are choosing to reject the world rather than reject the model in light of the overwhelming evidence against it. In other words, this professional decision is as emotionally driven as those of the typical error prone investor!

In my mind, this is a sad state of affairs. When I received my PhD in the late 1970s, I was excited about MPT, as it provided a concise, logical way to think about what are often chaotic financial markets. This resonated with my quantitative way of thinking, but the empiricist in me become ever more disenchanted with MPT as one study after another cast doubts on its ability to explain real world markets. Finally, a few years ago I rejected MPT entirely and have now moved onto behavioral portfolio management, about which I am as excited as I once was about MPT.

Once you reject MPT and accept behavioral portfolio management, everything changes. Portfolios are constructed to reduce the emotional impact of volatility by dividing the client portfolio into a portion to meet short-term needs and a portion to build long-term wealth. The short-term portfolio is built with little or no volatility. The long-term portion is built by focusing on expected and excess (i.e., alpha) returns. Short-term volatility and correlations shrink to insignificance as the time period lengthens. Sadly, the current infatuation with alternatives and short-term volatility mitigation has us forgetting about returns, the most important driver of long-term wealth.

You made an interesting distinction between short-term and long-term volatility. Would you please explain that distinction?

Rejecting MPT also means rejecting the notion that volatility and risk are synonymous. Risk is the chance of underperformance. For the short-term portfolio, volatility contributes to risk, but for the long-term portfolio, it is relatively unimportant. Much more important are expected and excess returns.

What other aspects of your approach are unique?

Truly active management generates superior returns. Current manager behavior is most important in selecting truly active managers, those who have the best opportunity to outperform. The focus should be on strategy, consistency, and conviction.

Based on numerous studies including my own, we know that past performance is not predictive of future performance. So using it, in all its manifestations, for selecting managers is an emotional decision. It just shows the power of emotions in an industry supposedly as sophisticated as the investment industry that almost everyone uses past performance in selecting managers even though there is no evidence that it is useful. Rather current manager behavior should be the focus.

Based on a long line of research including my own, we know that the best idea stocks of the best active equity mutual fund managers earn superior returns. That is, the top relative holding stocks (itself a manager behavior) of those managers with the best current behavior generate superior returns. Furthermore, these returns are more than likely the result of managers harnessing the price distortions resulting from market emotional mistakes.

Even more surprising, investor behavior is predictive of future market returns and thus can be used to pick the best markets going forward. My research shows that how investors are currently rewarding equity strategies is predictive of future US and international equity returns. In addition, Baker and Wurgler have shown that their Sentiment Index, which is based on objective measures not survey data, is predictive of the small firm effect. Putting these together, we can use investor behavior to make tactical market calls.

So behavioral portfolio management changes everything, with behavioral factors underlying all aspects of portfolio management, as a way to build superior portfolios. Behavioral factors can be used for portfolio construction, manager selection, stock picking, and market timing.