Factor Investing Explained: Value, Momentum, Quality, and What Persists

Factor investing is the practice of tilting a portfolio toward measurable stock characteristics, such as cheapness, recent price strength, or profitability, that academic research has associated with higher long-run returns. The best-documented factors are value, momentum, quality, size, and low volatility. The evidence behind them is real but weaker than marketing suggests: returns to published factors have historically declined after publication, factors can underperform for a decade, and implementation costs consume part of whatever premium remains.

What is a factor?

A factor is a characteristic, measurable across many stocks, that helps explain differences in their returns. The idea took its modern form when Eugene Fama and Kenneth French showed in the early 1990s that two characteristics, small market capitalization and a low price relative to book value, helped explain stock returns beyond overall market exposure.

Factor investing turns that research into portfolios. Instead of picking individual stocks by story, a factor strategy systematically buys the stocks ranking highest on a characteristic and underweights or avoids those ranking lowest, refreshing the rankings on a schedule. The approach is rules-based by construction, which makes it transparent and testable, and which connects it to the broader discipline of systematic investing.

What are the major factors and what is the evidence?

Value

Value means buying stocks that are cheap relative to fundamentals such as book value, earnings, or cash flow. Fama and French documented the premium in their 1992 and 1993 papers, and it has been observed across countries and asset classes. The standard explanations are risk, since cheap firms are often distressed or cyclical, and mispricing, since investors overpay for exciting growth. Value also endured a deep drawdown from roughly 2017 through 2020, a reminder that the premium arrives irregularly and tests patience.

Momentum

Momentum means buying recent winners, typically stocks with strong returns over the prior 6 to 12 months, and avoiding recent losers. Jegadeesh and Titman documented the effect in the Journal of Finance in 1993, and it has proven one of the most pervasive patterns in asset pricing. Its weaknesses are high turnover, which makes it costly to trade and tax-inefficient in taxable accounts, and rare but severe crashes when market direction reverses sharply, as in 2009.

Quality

Quality favors firms with high profitability, stable earnings, and conservative balance sheets. Novy-Marx's 2013 work on gross profitability and the "quality minus junk" research by Asness, Frazzini, and Pedersen formalized the evidence. Quality premiums have been more modest than value or momentum historically, but they have tended to hold up relatively well in market stress, making quality a common complement to value.

Size and low volatility

The size effect, the tendency of small stocks to outperform, is the most contested of the classic factors; much of the original premium weakens after adjusting for quality and liquidity. Low volatility, the observation that boring stocks have delivered better risk-adjusted returns than their risk would suggest, is better supported, though it behaves partly like a bond substitute and is sensitive to interest rates and valuation.

Do factor premiums survive after they are published?

This is the question honest practitioners take seriously. McLean and Pontiff, in a 2016 Journal of Finance study of 97 published return predictors, found that returns were about 26 percent lower out-of-sample and about 58 percent lower after publication. In plain terms, once a pattern is widely known, arbitrage capital erodes a meaningful share of it, though typically not all of it.

The practical reading is neither cynical nor credulous. The major factors did not vanish after publication, which suggests part of each premium reflects real risk-bearing or durable behavioral patterns rather than a quirk of old data. But an investor should expect post-publication premiums materially smaller than the backtests that made the factors famous, and should discount any newly discovered "factor" accordingly. Hundreds have been proposed in what researchers call the factor zoo, and most fail out-of-sample.

What do factors cost to implement?

A paper premium is not a realized return. Three costs sit between them.

Trading costs scale with turnover, which makes momentum, the highest-turnover major factor, the most expensive to run. Taxes follow the same logic in taxable accounts, where frequent realization of gains can surrender a large share of the pre-tax premium; tax-managed implementation and the use of tax-advantaged accounts matter. Fees for factor funds typically exceed plain index fund fees, and the gap is a direct subtraction from any premium captured.

Design choices also drive results. Two funds both labeled "value" can differ substantially in measurement, sector constraints, and rebalancing rules, and their returns can diverge for years. The label tells you less than the methodology document does.

A hypothetical example: what a tilt actually changes

Consider a hypothetical investor with a $5 million equity allocation who moves $1 million, 20 percent of it, from a broad market index into a multifactor fund tilted toward value, momentum, and quality. Suppose the tilt delivers a net premium of 1 percent per year over the market in a given period. The portfolio-level effect is 20 percent of that, or roughly 0.2 percent per year, about $10,000 on this portfolio.

Now suppose instead the factors underperform the market by 3 percent per year for five years, which is well within historical experience for value in the late 2010s. The same position costs roughly 0.6 percent per year at the portfolio level, and the investor must watch a deliberate choice lag a simple index for half a decade. Both arithmetic results are realistic, the figures are hypothetical, and the second scenario is the one that determines whether an investor can actually hold the strategy.

Should you try to time factors?

The temptation is obvious: if value is cheap relative to history, overweight it now. The evidence is discouraging. Research from AQR's Cliff Asness and others has found that factor timing signals, mostly based on valuation spreads, are weak, slow, and easily swamped by the cost of trading in and out. Timing factors is, in effect, just another factor bet layered on top, with less evidence behind it.

The more defensible posture is strategic: hold modest, diversified tilts across factors with low correlation to one another, such as value plus momentum plus quality, size them so underperformance is tolerable, and rebalance on rules rather than conviction. Diversification across factors smooths the ride precisely because their bad years tend not to coincide.

Key numbers

Frequently asked questions

Is factor investing the same as smart beta?Largely, yes. "Smart beta" is the marketing term for index-like funds with factor tilts; the underlying research and mechanics are the same, and the same scrutiny of methodology, costs, and turnover applies.

Which factor has the strongest evidence?Momentum and value have the longest and broadest records across markets and asset classes, with quality close behind. Size is the most contested of the classic factors.

How long can a factor underperform?A decade is within historical experience; value lagged growth severely from roughly 2017 to 2020 after a weak preceding decade. Position sizes should assume droughts of that length.

Do factors work after taxes?Less well, particularly high-turnover factors like momentum held in taxable accounts. Tax-managed implementations and locating high-turnover strategies in tax-advantaged accounts can preserve more of the premium.

Are new factors worth investing in?Rarely on their own. McLean and Pontiff's evidence on post-publication decay suggests applying a steep discount to any recently discovered pattern, especially one with high turnover or thin economic logic.

Should factor tilts replace my market portfolio?For most investors, no. Tilts are best held as a measured complement to broad market exposure, sized so that several years of factor underperformance does not jeopardize the plan or the investor's resolve.