Fama-French Three-Factor Model
The Fama-French Three-Factor Model is an asset pricing model developed by Eugene Fama and Kenneth French (1992, 1993) that extends the Capital Asset Pricing Model (CAPM) by adding two additional systematic risk factors—Small Minus Big (SMB) market capitalization and High Minus Low (HML) book-to-market ratio—to the market excess return factor, explaining a substantially larger fraction of cross-sectional equity return variation than CAPM alone.
Key takeaways
- SMB (Small Minus Big) captures the size premium: small-cap stocks have historically earned higher returns than large-cap stocks after adjusting for market beta.
- HML (High Minus Low) captures the value premium: high book-to-market (cheap) stocks have outperformed low book-to-market (expensive/growth) stocks on average.
- The three-factor model substantially reduces CAPM's alpha anomalies: many funds that appeared to generate alpha under CAPM are loading positively on SMB and HML factors.
- The model is used as the primary benchmark for hedge fund and mutual fund alpha measurement—true alpha requires outperforming after controlling for all three factors.
- Fama and French later extended to a five-factor model (2015) by adding profitability (RMW) and investment (CMA) factors, further reducing unexplained return variation.
Explanation
The Fama-French model emerged from systematic empirical research documenting that the CAPM's single market factor was insufficient to explain the cross-section of equity returns. Banz (1981) demonstrated a size premium—small-cap stocks earned higher average returns than large-cap stocks after adjusting for beta. Basu (1977) and Stattman (1980) documented a value premium—stocks with high book-to-market ratios (cheap stocks) earned higher returns than low book-to-market stocks (expensive/growth stocks). Fama and French unified these findings into a coherent three-factor framework that has become the standard tool for performance attribution in academic and applied finance.
The mathematical specification of the model is: R_i - R_f = α_i + β_MKT(R_M - R_f) + β_SMB × SMB + β_HML × HML + ε_i. The factors are defined as portfolio returns: SMB is the monthly return of a portfolio long small-cap stocks and short large-cap stocks, constructed to be sector-neutral. HML is the monthly return of a portfolio long high book-to-market stocks (value) and short low book-to-market stocks (growth), also sector-neutralized. Historical data for these factor returns is maintained by Ken French on his Dartmouth website, providing freely available empirical data for academic and practitioner research.
The interpretation of factor loadings in the three-factor model is economically intuitive. A portfolio with a large positive β_SMB has high exposure to small-cap stocks and should earn a premium for that exposure. A portfolio with a high positive β_HML tilts toward cheap, value-oriented stocks. If a fund manager claims to generate alpha (positive α_i) but the fund actually has positive SMB and HML loadings, much of the apparent alpha is actually compensation for systematic factor risk—not genuine stock selection skill. The three-factor model thus deflates many claimed alpha generators by attributing their performance to documented risk premia.
The economic interpretation of why SMB and HML carry risk premia is contested between risk-based and behavioral explanations. Fama and French's own preferred explanation is risk-based: small stocks and value stocks are exposed to systematic distress risk—they perform especially poorly during economic downturns and financial crises, precisely when investors value diversification most. Investors require higher expected returns to bear this undiversifiable systematic risk. Behavioral economists offer an alternative: the value premium reflects investor tendencies to extrapolate past earnings growth into the future, systematically overvaluing glamour growth stocks and undervaluing out-of-favor value stocks. Both explanations may partially correct.
For hedge fund due diligence, factor regression using the Fama-French model is a standard analytical step. A fund reporting 15% annual gross returns may have most of that performance explained by positive market beta, a small-cap tilt (positive SMB), and a value tilt (positive HML)—all of which investors could cheaply replicate with passive factor ETFs. The residual alpha after accounting for all three factors represents the true idiosyncratic performance that justifies the fund's management fees. If α_i is statistically indistinguishable from zero, the fund is charging active management fees for passive factor exposure.
Formula
R_i - R_f = α_i + β_{MKT}(R_M - R_f) + β_{SMB} × SMB + β_{HML} × HML + ε_i
Example
A quantitative equity fund reports a 3-year annual gross return of 14%, versus the S&P 500's 10% return over the same period. On a CAPM basis, with beta of 0.9, the expected return is 10% × 0.9 = 9%; apparent alpha = 14% - 9% = 5%. Running the three-factor regression: β_MKT = 0.9, β_SMB = 0.6 (significant small-cap tilt), β_HML = 0.5 (significant value tilt). Three-factor expected return = 9% + 0.6 × 3% (SMB premium) + 0.5 × 4% (HML premium) = 9% + 1.8% + 2.0% = 12.8%. Three-factor alpha = 14% - 12.8% = 1.2%, not statistically significant given the standard error. Conclusion: the fund's outperformance is fully explained by documented factor exposures—it earns no true alpha after controlling for size and value tilts.
Related terms
Alpha Basis Beta Cap Capital Asset Pricing Model Diversification Equity Equity Risk Premium Factor Model Hedge Fund Market Capitalization Omega Ratio