Beta Bunk
CAPM is a model for valuing stocks, based on the idea that investors demand additional expected return to take on additional risk.
The risk is assessed on a stock or security’s “beta,” a measure of a company’s volatility and correlation with the market as a whole. A company with a share price that tends to rise and fall more than the market will have a high beta and vice versa.
Damning evidence that CAPM doesn’t work came from a 2004 study by Eugene Fama and Kenneth French, which looked at all stocks on the NYSE, the American Stock Exchange and Nasdaq from 1923 to 2003. The study shows CAPM underpredicts the returns to low beta stocks and massively overstates the returns to high beta stocks.
A similar study of the 600 largest US stocks by Jeremy Grantham showed that from 1969 to the end of 2005, the lowest decile of beta stocks – notionally the lowest risk – outperformed by an average 1.5% a year. The highest beta stocks actually underperformed by 2.7% a year.
“Pricing model that is fine in theory but worthless in reality,” by Tony Tassell