On Behavioral Portfolio Construction
I have intentionally modeled the Helical Portfolio on the pyramidal behavioral framework. This structure is described in the Shefrin and Statman 2000 paper “Behavioral Portfolio Theory”. The BPT theory considers how investors deal with the potentially conflicting goals of both increasing and preserving wealth. I found reading this paper to be a bit of a watershed event in my approach to investing, as it well described how I tended to invest, despite my efforts to frame my portfolios along the lines of more traditional pie charted allocation spaces more consistent with Modern Portfolio Theory (MPT). I have found the behavioral model be …. ‘embraceable’ may be the best description.
So I enjoyed reading this 2004 paper by Gregory Curtis that looks at both theories and considers their strengths and shortcomings.
“Since Markowitz won the Nobel in 1990, we have tended to design our portfolios as though we were all mean variance optimizers, perfect little economic beings who always made the “right” decision in our own interests. The work of Kahneman, Tversky, et al. has blown this cozy little conceit to pieces. True, sometimes we behave like perfect economic beings. But other times we behave like, well, human beings.”
As an individual, and not a market, I find I can only invest with my own perceptions of risk and reward, and not think in terms of correlations. As an individual focused on the single (though broad) sector of healthcare, I already give up much of what MPT tells me I should not. I do so with the hope and expectation that I may have an advantage investing in this area. Curtis compares the shortcomings of the two theories. Mainly that MPT doesn’t often resonate with clients and that BPT may have ingrown issues due to how difficult behavior is to effectively study as well as its lack of optimization. His perspective is that of an investment advisor managing the portfolios of clients. His difficult balance is optimizing the portfolios for those clients, while also appeasing them (keeping them as clients). So he summarized the two models flaws as follows:
“In short, employing MPT techniques when we advise investors does tend to cause financial advisors to propose optimal portfolios, but the likelihood that clients will adopt those portfolios (or stick to them) is low. On the other hand, employing behavioral finance techniques when we advise clients tends to result in recommended portfolios that resonate well with the clients, but which are not likely to be optimal in terms of the relationship between risk and reward.”
He considers MPT a ‘rational’ approach to investing and BPT an arational alternative.
“An “arational” approach is not necessarily “irrational.” Indeed, the “wrong” choices we make as investors may be suboptimal from a purely economic perspective, but those choices often serve deeper emotional needs.”
I think considering the BPT approach to be outside reason is a bit harsh. BPT doesn’t merely serve emotional needs (isn't that a good thing?), but in my opinion also provides a framework that helps focus the goals of an investor. In an effort to achieve optimization, MPT would design the same portfolio for a both a young student as for a wealthy man, and the reality is that their investing goals are likely very different. The former is probably more interested in trying reaching to build wealth, the latter to preserve it. What is 'optimal' for one is not necessarily so for the other, due to position in life (not correlation constants).
Curtis proposes shifting from a BPT to an MPT model with clients over time, to minimize their potenatial chafing at the less familiar MPT framework.
“To make this transition palatable to the family, it is necessary that they gain experience with less comfortable asset classes gradually. If the transition is expected to occur over a five-year period…”
I am not an investment advisor, but to me this seems like a lazy effort to form a ‘one size fits all’ portfolio framework and try to shift clients to it. While there individual goals may indeed differ from seeking optimal returns.
“Modern portfolio theory represents the best learning we have about how capital markets actually operate, while behavioral finance offers the best insights into how investors actually behave. [Agree] But markets don’t care what investors think of as risk, and hence idiosyncratic ideas about risk and what to do about it are bound to harm our long-term investment results. On the other hand, Daniel Kahneman, Amos Tversky, and their followers have demonstrated beyond doubt that we all harbor idiosyncratic ideas and that we tend to act on them, regardless of the costs [not benefits?] to our economic welfare. By combining both MPT and behavioral finance models in our work with family investors, we stand the best chance of designing, implementing, and maintaining portfolios that will prove acceptable to our clients and that will prove productive to our clients’ wealth.”
I worry in reading his conclusion that Curtis is more concerned with serving his ideals than his clients. Since I consider risk a perspective, not a measure, I’d agree that markets don’t care about an individual’s consideration of risk, or about risk at all. Heck, markets don’t ‘care’ period (no personification allowed). But since risk is indeed perceived, I can’t ever consider it abstractly as an ‘idiosyncratic idea’. The word risk itself connotes and emotional state (otherwise we would call it a probability, not risk). And of course we tend to act on our motivations. Shifting clients to a MPT framework may well be the best way to generate 'optimal' fees for an investment advisor, but probably is not the best way to serve the client.
I embrace BPT not because it may be the ‘optimal way to invest’, but because it helps me focus on what I stand to gain, and/or lose as I invest. It better helps me appreciate my goals and frame my efforts for achieving them. It helps me appreciate the biases I may be forming, and given suitable consideration of them hopefully mitigate their deleterious effects. BPT makes me think and better participate in the process of investing, which I enjoy. BPT may not make me an optimal investor, but instead amore fulfilled one. Which is right for you depends on your real goal?