Factor Investing - Systematic Premia Beyond Beta
By EC Assets Research Team, Systematic Strategies · Published · Updated
Factor Investing: Factor investing systematically harvests return premia associated with specific security characteristics: value, momentum, quality, low-volatility, size, and carry. Replaces stock-picking with rules-based exposure to documented risk premia.
What Factor Investing Is
Factor investing replaces discretionary stock picking with rules-based exposure to characteristics that have been linked to differences in average returns. A factor portfolio ranks securities on a characteristic, such as cheapness or recent performance, and buys the top while selling or underweighting the bottom.
It grew from the failure of the CAPM to explain returns with market beta alone. Fama and French (1993) added size and value to the market factor; Jegadeesh and Titman (1993) documented momentum, which Carhart (1997) added as a fourth factor; Fama and French (2015) extended the model with profitability and investment.
How It Works
Each factor is the return of a long-short portfolio: SMB (small minus big) for size, HML (high minus low book-to-market) for value, RMW (robust minus weak profitability), CMA (conservative minus aggressive investment) and MOM (winners minus losers over the past year, skipping the latest month). A portfolio's exposure to them is measured by regressing its returns on the factors:
$$R_i - R_f = \alpha_i + \beta_i,(R_m - R_f) + s_i,\text{SMB} + h_i,\text{HML} + r_i,\text{RMW} + c_i,\text{CMA} + m_i,\text{MOM} + \varepsilon_i$$
The research factors are built with simple, transparent rules. Fama and French sort stocks into two size groups and three groups on the characteristic, form value-weighted portfolios and take the difference between the top and bottom groups, rebalancing once a year in June; momentum is re-sorted every month. Commercial products use similar ideas with more definitions, constraints and attention to trading costs.
The intercept is the alpha left after accounting for factor exposures; much apparent manager skill turns out to be factor exposure in this regression.
Two explanations compete for why the premia exist. Risk-based accounts see compensation for bad-times exposure: value stocks are often distressed, and momentum crashes when markets rebound. Behavioural accounts see systematic mispricing from overreaction or underreaction. Carry, a related style across asset classes, is covered in the carry trade entry.
Worked Example
Factor returns from Kenneth French's data library, July 1963 to August 2026:
| Factor | Annual premium | Volatility | Sharpe ratio |
|---|---|---|---|
| Market minus cash | 7.22% | 15.44% | 0.47 |
| Size (SMB) | 2.24% | 10.46% | 0.21 |
| Value (HML) | 3.53% | 10.29% | 0.34 |
| Profitability (RMW) | 3.02% | 7.92% | 0.38 |
| Investment (CMA) | 2.92% | 7.18% | 0.41 |
| Momentum (MOM) | 7.16% | 14.57% | 0.49 |
Average monthly long-short returns times 12, over 758 months, before trading costs; the ratio bench compares other return and risk combinations.
The long-run premia are real but come with deep and long drawdowns. Value fell 57.8 percent from December 2006 to September 2020, almost fourteen years, before gaining 22.2 percent in 2021 and 31.7 percent in 2022; by August 2026 it was still below its 2006 peak. Momentum lost 49.4 percent in March to May 2009 as beaten-down stocks rebounded, the pattern Daniel and Moskowitz (2016) describe as momentum crashes.
Combining factors helps. Value and momentum had a correlation of -0.19 over the period, so their drawdowns rarely coincided, and a blend has been smoother than either alone.
When It Applies (and Limitations)
Data mining. Hundreds of factors have been published. Harvey, Liu and Zhu (2016) argue that most fail a sensible statistical threshold, so only factors with economic rationale and evidence across markets and periods deserve capital.
Decay after publication. McLean and Pontiff (2016) find that anomaly returns fall by about a third after publication, as investors trade on them.
Costs and implementation. Long-short factor returns ignore trading costs, shorting fees and capacity. Momentum, with high turnover, is most exposed. Long-only versions capture only part of the premium and keep full market exposure.
Crowding. When many funds hold similar factor portfolios with leverage, losses at one can force sales that hit the others. Khandani and Lo (2011) trace the sudden losses of quantitative equity funds in August 2007 to such an unwind of crowded positions.
Definitions matter. Value measured by book-to-market, earnings or cash flow can give quite different portfolios, and the gap between products targeting the same factor can be large.
Why It Matters for Institutional Investors
Diversified return sources. Factor premia that are weakly correlated with each other and with the market can improve a portfolio's risk-adjusted return, which is the main institutional case rather than any single factor's outperformance.
Understanding managers. Regressing a manager's returns on factors separates cheap, replicable factor exposure from genuine alpha, which affects what fees are worth paying.
Patience as a requirement. Factor drawdowns can last a decade. Investors who cannot hold through them, or whose governance forces exits after a few bad years, tend to realise the losses without the recoveries; maximum drawdown and time underwater are the statistics to plan around.
Hidden exposures. Many portfolios carry factor tilts without intending to, such as a value bias from dividend mandates or a size bias from equal weighting. Measuring them avoids paying active fees for exposures that are cheap to buy.
Systematic implementation. Factor portfolios sit at the core of quantitative investing, where definitions, risk controls and trading costs decide whether a paper premium becomes a realised one.
References
- Fama, E. F., & French, K. R. (1993). Common Risk Factors in the Returns on Stocks and Bonds. Journal of Financial Economics, 33(1), 3-56.
- Fama, E. F., & French, K. R. (2015). A Five-Factor Asset Pricing Model. Journal of Financial Economics, 116(1), 1-22.
- Jegadeesh, N., & Titman, S. (1993). Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency. Journal of Finance, 48(1), 65-91.
- Carhart, M. M. (1997). On Persistence in Mutual Fund Performance. Journal of Finance, 52(1), 57-82.
- Daniel, K., & Moskowitz, T. J. (2016). Momentum Crashes. Journal of Financial Economics, 122(2), 221-247.
- Harvey, C. R., Liu, Y., & Zhu, H. (2016). ... and the Cross-Section of Expected Returns. Review of Financial Studies, 29(1), 5-68.
- Khandani, A. E., & Lo, A. W. (2011). What Happened to the Quants in August 2007? Evidence from Factors and Transactions Data. Journal of Financial Markets, 14(1), 1-46.
- McLean, R. D., & Pontiff, J. (2016). Does Academic Research Destroy Stock Return Predictability? Journal of Finance, 71(1), 5-32.
Frequently asked questions
What is a factor?
A characteristic of securities, such as cheapness, recent performance, profitability or size, that has been associated with differences in average returns. A factor portfolio buys securities ranked high on the characteristic and sells or underweights those ranked low, isolating the return linked to it.
Which factors have the strongest evidence?
The market, value, momentum, profitability and investment have the broadest support across periods and countries. Size has weaker evidence after its discovery. Harvey, Liu and Zhu (2016) warn that many of the hundreds of published factors are likely false discoveries, so the bar for new ones should be high.
Why do factor premia exist?
Two explanations compete. Risk-based views say factors pay for exposure to bad states of the world, such as distress for value or crash risk for momentum. Behavioural views say investors systematically misprice certain stocks, for example by overreacting to bad news or underreacting to new information. Both may contribute.
Can factor timing improve returns?
The evidence is weak. Factor returns are hard to predict, and timing signals add turnover and the risk of exiting just before a recovery. Most institutions hold a diversified, steady factor allocation and rebalance it rather than trying to time individual factors.
Do factor premia survive publication?
Partly. McLean and Pontiff (2016) find that returns on published anomalies fall by about a third after publication, consistent with investors trading them away. The major factors have persisted, but expectations should be lower than the historical record.
Stay informed
Market commentary, firm news and research from EC Assets - direct to your inbox.