You are here

Evolutionary Portfolio Theory: An Alternative to the CAPM

Posted by
Swiss Finance Institute
Friday, November 17, 2017 - 09:00

Is using the Capital Asset Pricing Model (CAPM) inappropriate or even unethical? What’s the alternative? All seats were sold out in Zurich in September when SFI Professor Thorsten Hens (University of Zurich) brought Swiss Chartered Financial Analysts (CFAs) up to date on the Evolutionary Portfolio Theory approach: The widely used Capital Asset Pricing Model (CAPM) states that while you can minimize investment-specific risk (the α) through diversification, you get more return if you take more non-diversifiable, systematic risk, the β. The model might therefore lead to inappropriate risk-seeking in order to maximize returns. Furthermore, the model assumes that all capital market participants have homogenous expectations, when in reality this is not the case. It has not been possible to prove that the conclusions of the simplistic CAPM approach are right, and empirical studies have even described many anomalies that are simply incompatible with the CAPM - many sources of α seem to exist. However, as Professor Hens has shown, when the assumptions underlying the model are weakened (e.g., heterogeneous expectations can be assumed), the CAPM still offers a fairly good approach to asset pricing: alphas become alternative betas. But the CAPM`s focus is still too narrow when it comes to judging the benefits of different investment strategies since - within the CAPM - all strategies are based on mean-variance portfolio optimization, whereas even simple investment strategies like 1/N have been shown to perform better.  That is where Evolutionary Portfolio Theory (EPT) offers a better alternative. EPT is derived from the biological principle that only the fittest survive, and applies this law to financial markets. The fittest has the highest γ, i.e. the highest growth of wealth. EPT does not build on assumptions like portfolio optimization, but rather on a stochastic model structure, taking into account dynamic interaction of any type of heterogeneous behavior. While EPT does not focus on asset pricing, it is very useful for strategy development. Professor Hens concluded his presentation by showing the implications of EPT for the development of investment strategies. According to this view, any candidate investment strategy has to pass the following three tests that can be performed based on EPT: 

1) Back test: How would the investment strategy have performed in the past?
2) Impact test: What happens if many people use the strategy?
3) Reflexivity test: How will the other investors react to the strategy?

Or in Thorsten Hens’s words: «Go beyond the α and β, and check your γ!» If you want more details, watch the full seminar, take a look at Professor Hens’s slides (with registration), or read here on page 8 what CFA Switzerland writes about his presentation. If you are interested in the methodology of EPT, we recommend you to take a look at this research paper. More readings for practitioners from Thorsten Hens, regarding his behavioral and evolutionary finance research:

  • «Is there Swissness in Investment Behavior?» (SFI Practitioner Roundup)—read here
  • «Bewusster anlegen mit Evolutionary Finance» (Finanz und Wirtschaft)—read here Swiss Finance Institute and CFA Society Switzerland have collaborated closely in the education of finance practitioners since 2013.

Would you like to learn more about asset or portfolio management? We offer various courses:
AM Products and Investment Strategies, Portfolio Management Simulation, CAS in Asset Management or Managing International AM