Feb 12

Fama & French anomalies may be either risk factors or price inefficiencies

by David Harper, CFA, FRM, CIPM

FRM | CFA |

The CXO Advisory Group helpfully summarizes research by Eugene Fama and Kenneth French ("Dissecting Anomalies"). FRM candidates encounter the Fama-French three factor model in the context of market-neutral long/short equity strategies. Where, according to Jaeger, the goal of a market-neutral strategy is to eliminate (insulate) exposure to broad market risk factors (e.g., beta neutral, currency neutral) and instead exploit price inefficiencies. The Fama-French factor model is like the CAPM but with two additional factors. Its risk factors (blue backgrounds below) are the equity market premium , size (capitalization), and value (book-to-market):

famafrench

Using regression (and sorts), Fama and French analyzed several "anomalies" over micro/small/large capitalization companies from 1963 to 2005. They regressed "anomaly variables" against stock returns. The anomaly variable include the classic Fama-French size and value factors (above) but also several other popular factors:

  • Size (market capitalization. the SMB factor above): draws most of its explanatory power from microcaps! This implies compensation (i.e., higher risk/return) for really small companies but not so much for others
  • Value (book-to-market equity; the HML factor above): holds up pretty well. That is, higher returns for higher book-to-market.
  • Momentum (trailing 10-month cumulative stock return): their findings justify momentum as the "premium anomaly...stocks with low returns over the last year tend to have low returns for the next few months and stocks with high past returns tend to have high future returns"
  • Net stock issuance (change in shares outstanding = shares issued - shares repurchased): "larger net issues of stock are associated with lower future returns, and the marginal relation between issues and returns is much the same for microcaps, small stocks, and big stocks."
  • Accruals (change in operating working capital per split-adjusted share): not as strong as momentum or issuance but "we can infer that higher positive accruals are indeed associated with lower future returns"
  • Asset growth and Profitability (adjusted net income): mixed, stronger form small/micro caps, but "profitability and asset growth tend to be persistent...so the positive relation between average returns and profitability and the negative relation between asset growth and average returns are in line with the valuation equation"

The interesting part is they draw a difference between risk factors and inefficient markets.

  • Under CAPM and the Fama-french factor model, higher returns are (theoretically) compensation for higher risk. Higher risk refers to greater exposure to common risk factors; e.g., a small stocks is riskier, a low book/market stock is riskier
  • But statistically significant anomalies are not necessarily risk factors: anomalies may indicate either risk factors or price inefficiencies. For examples, regressions may validate that positive accruals tend to imply lower future returns, but they don't tell us if that's a risk factor or a price inefficiency. Therefore, they don't validate or invalidate efficient market hypothesis (EMH)

From their conclusion:

"researchers commonly interpret the average returns associated with anomaly variables as evidence of market inefficiency.  The valuation equation says, however, that controlling for the book-to-market ratio, proxies for expected net cashflows will identify differences in expected returns whether they are due to irrational pricing or rational risks.  Thus, evidence that variables that predict future cashflows also predict returns does not, by itself, help us determine how much variation in expected returns is caused by risk and how much is caused by mispricing." - SSRN-Dissecting Anomalies by Eugene Fama, Kenneth French

Comments

  1. Be the first to leave a comment!

Leave a Comment