Great article from Kiran Pande:
Since the 1960s, we’ve been dependent on a model called CAPM (capital asset pricing model) to understand the relationship between risk and return, despite the fact that its measure of risk only explains about 70% of return. This measure, beta, makes the assumption that the entirety of every stock’s return is due to its exposure to the market. Put simply, every stock’s returns will equal a factor of the S&P 500’s returns. Thus, if a stock’s beta is 2.0, it will double the S&P 500’s returns on a bull day and double its losses on a bear day. Obviously, this assumption is wrong almost every day, but the idea is that this factor is explaining most of a stock’s returns.
All returns not explained by beta in the CAPM model are called alpha. This is traditionally accepted as the level of skill and value added by a portfolio’s manager……
There is a whole laundry list of reasons not to use CAPM, beta, and alpha but here are some highlights…
- 70% is not 100%, not even close
- Beta is symmetrical, risk is not… downside risk is rarely the same as upside risk.
- Since the market index used to calculate beta (usually the S&P 500) contains stocks whose returns are supposedly dependent upon beta, these stocks’ returns are somewhat dependent upon themselves.
These counterpoints do not render beta, alpha, and CAPM useless, but we can do much better. The Fama-French Three Factor model is the answer. Rather than a single factor (market performance), the model throws a size factor and a value factor into the mix, replacing much of the nebulous alpha term. With the addition of these factors, Fama and French boast that their model explains as much as 96% of returns with quantifiable measures.