Reality is wrong. Dreams are for real.
by Raša Karapandža
Eugene F. Fama and Lars P, Hansen of the University of Chicago and Robert J. Shiller of Yale University received the prize “for their empirical analysis of asset prices,” the Nobel committee said today.
Why am I writing this post? I am writing it because in one of my papers (joint work with Jose M. Marin) we show that two main contributions of Eugene Fama are at least to some extent wrong. I believe that the two main contributions of Eugene Fama are:
- The introduction of the concept of market efficiency is often attributed to Fama. Some conceptual work on that field was done long before Fama, however Fama is really responsible i making it a main stream Finance. Investopedia describes Fama’s definition of market efficiency as: “It is not possible for an investor to outperform the market because all available information is already built into all stock prices.”
- In the joint work with Kenneth R. French, Fama creates a multi-factor model of stock returns – where not only market beta of an asset is what drives stock returns like in the Capital Asset Pricing Model (CAPM) that brought a Nobel prize to William Sharpe in 1990. They show that the addition of two factors that they create can help to resolve the mystery that small stock give higher returns than large stocks (and the same thing for B/M). They show that by demonstrating that their three factor model (nowadays known as Fama-French model) is better in pricing 25 portfolios sorted on size and B/M value.
In our paper titled “The Rate of Market Efficiency” we introduce a new concept of market efficiency. The concept that Fama introduced is a dichotomous concept – markets are either efficient or not. Any engineer would find it even silly to discuss if some engine is efficient or not. The question that an engineer would ask is: “How much is the engine efficient? ” In our paper we do exactly that, we introduce the concept of the Rate of Market Efficiency – the rate which tells us how much are markets efficient, in percents, like engineers would do.
Moreover we show that the second contribution of Fama carries its problems. We show that, indeed, introduction of two factors helps in better pricing of 25 portfolios sorted on size and B/M. However, we also show that this comes at a cost of the worse performance in pricing infinitely many other, randomly created portfolios.
Do not get me wrong here, I have a big respect for Eguene Fama. He has really changed the world of modern Finance. The best demonstration of the respect I hold for him is that his portrait from above is one of very few people on the painting that is on the wall of my office. And that is not a coincidence. That painting was built to order and is dedicated to finance.