Factor Investing: Value, Momentum, Quality, and Size

Decades of data say that stocks sharing certain measurable traits have earned different returns from the market as a whole. That finding is real, well-documented, and routinely oversold — including by people who can't tell you which paper their favourite factor came from.

Last verified 2026-07-28

A factor is a shared trait, not a stock pick

A factor is a measurable characteristic that groups of stocks share and that has historically been associated with different returns. Value is the tendency of cheap-looking stocks — low price relative to book value or earnings — to have outperformed expensive ones. Size is the small-cap premium over large caps.

The important word is 'groups'. Factor investing isn't about identifying a good company; it's about holding hundreds of stocks that share a statistical property, and expecting the property rather than any individual holding to do the work.

These premiums are also long-horizon and deeply unreliable in between. Value underperformed growth for well over a decade in the 2010s. A tilt that works on twenty-year data can test your patience for fifteen of them, and most people who abandon one do so at the worst moment.

Who actually published what

This gets stated wrongly so often it's worth pinning down. Fama and French's 1993 three-factor model is Market, Size and Value. Their 2015 five-factor model adds Profitability and Investment.

Momentum is not in either one. It comes from Carhart's 1997 four-factor model, built on top of the Fama-French three — and Fama and French deliberately left it out of their own later model despite having the chance to include it. Low Volatility comes from a different strand of research again.

This isn't pedantry about citations. Someone selling you a 'Fama-French factor fund' built on momentum and low volatility is describing their product with the credibility of research that does not cover it. Knowing which finding supports which claim is how you tell a well-grounded product from a well-marketed one.

The premium is gross; what reaches you isn't

Every factor premium you read about in a paper is measured before costs. Implementing it involves a fund that charges a fee, trades more than a total-market index, and pays spreads on less liquid stocks.

Those costs come off the top, and factor funds charge meaningfully more than broad index funds — often several times more. When the honest expected premium is a point or two, a fee difference of half a point is not a rounding error; it's a large fraction of the whole thesis.

Momentum is the sharpest case, because the strategy requires continuous turnover by construction: yesterday's winners have to be sold as they stop winning. The gross premium in the research is among the strongest documented, and it's also the one most eroded by the cost of capturing it.