The Five Factors Explained
The five-factor model published by Fama and French in 2015 extends their earlier three-factor framework with two additions drawn from corporate fundamentals. This article takes each factor in turn: what it measures, how the portfolio behind it is built, the economic reasoning offered for it, and the objections it faces. The construction details matter more than they might appear, because most disagreements about whether a factor is real turn out to be disagreements about how it was built.
1. Market (Mkt-Rf)
The market factor is the return of a broad value-weighted portfolio of stocks minus the risk-free rate, usually a short-term Treasury bill. It is the oldest and least controversial factor, and for most diversified equity portfolios it explains the large majority of return variation. Everything else in the model is an attempt to account for what the market factor leaves unexplained.
Subtracting the risk-free rate matters conceptually. The quantity being explained is the reward for taking equity risk rather than holding a safe asset, not the total return.
2. Size (SMB, Small Minus Big)
SMB is the return on portfolios of small-capitalisation stocks minus the return on portfolios of large-capitalisation stocks, with companies split by market equity. The original justification was that smaller companies are more fragile, less liquid, and less researched, and should therefore compensate investors more.
Size is also the factor whose standalone evidence has weakened most since publication. Its more durable role in the model is as a control: value and profitability effects behave differently across the size spectrum, and the other factors are constructed within size buckets so those interactions do not contaminate the estimates. The dedicated article on the size effect covers this in detail.
3. Value (HML, High Minus Low)
HML is the return on high book-to-market portfolios minus low book-to-market portfolios. High book-to-market means the market values the company at little relative to its accounting book value, which is the standard definition of a value stock in this literature.
Two objections recur. The first is that book value has become a poor proxy for fundamental value as intangible assets have grown, since accounting rules expense most research and brand spending rather than capitalising it. The second is specific to the five-factor model: once profitability and investment are included, HML becomes partly redundant, and Fama and French acknowledged this directly in the 2015 paper.
4. Profitability (RMW, Robust Minus Weak)
RMW is the return on portfolios of highly profitable companies minus weakly profitable ones, using operating profitability scaled by book equity.
The reasoning follows from valuation theory rather than from pure data mining. If two companies trade at the same book-to-market ratio but one generates substantially more profit from the same book equity, the market must be applying a higher discount rate to the profitable one, and a higher discount rate is a higher expected return. Framed that way, profitability is not a separate anomaly so much as a necessary companion to value: valuation ratios are only interpretable alongside the earnings they are measured against.
Novy-Marx (2013) made the influential empirical case, showing that gross profitability had explanatory power comparable to book-to-market and that the two signals worked well together precisely because they tend to select opposite kinds of companies.
5. Investment (CMA, Conservative Minus Aggressive)
CMA is the return on portfolios of companies growing their assets slowly minus those growing them rapidly, measured by year-over-year growth in total assets. Conservative investors of capital have historically outperformed aggressive ones.
Several explanations compete. Managers with easy access to capital may fund projects that destroy value. Companies tend to issue equity and expand when their shares are richly priced, which mechanically links high asset growth to subsequent disappointment. And rapid growth may itself signal that a firm is approaching the limits of its profitable opportunities.
How the sorts are actually built: Fama and French form portfolios on two dimensions at once, splitting by size and then by the second variable within each size group, and take the average of the resulting spreads. This is why the factors are relatively independent of one another, and why a naive single-sort replication will not reproduce the published series.
What the extension did and did not settle
The five-factor model explains more of the cross-section of average returns than the three-factor version, which was the point. It did not resolve the deeper question of whether these premiums represent compensation for risk or persistent mispricing, and it introduced a new awkwardness by rendering the value factor partly redundant, which sits uncomfortably with two decades of work treating value as fundamental.
Momentum remains excluded despite strong empirical support, a decision Fama and French defend on the grounds that it lacks a convincing risk story. Whether a model should include an effect it cannot explain is a live methodological disagreement rather than a technical detail.
Fama, E. F., & French, K. R. (2015). "A five-factor asset pricing model." Journal of Financial Economics, 116(1), 1-22.
Novy-Marx, R. (2013). "The other side of value: The gross profitability premium." Journal of Financial Economics, 108(1), 1-28.