Factor Investing Explained

At a glance

Main factors: Value, size, momentum, quality, low volatility
Why they persist: Risk compensation, behavioural bias, or constraints
Main caveat: Premia have compressed since publication
Practical use: Know what your strategy is exposed to, whether intentionally or not

Key takeaways

Factors are characteristics that have historically explained cross-sectional differences in returns, such as valuation, past return, and profitability.
Each factor has a proposed explanation: compensation for risk, exploitation of a behavioural bias, or an institutional constraint.
Measured premia have declined since publication, consistent with capital arriving to exploit them, though the underlying causes have not vanished.
Factors experience long underperformance periods, sometimes a decade, which is the main practical obstacle to holding them.
Even discretionary strategies have factor exposures; knowing yours tells you what you are actually being paid for.

The major documented factors

Factor	Definition	Proposed explanation	Notable weakness
Market (beta)	Exposure to broad equity returns	Compensation for bearing market risk	Large drawdowns; unavoidable
Value	Cheap relative to fundamentals	Risk of distress, or over-extrapolation of bad news	Extended underperformance periods
Size	Smaller companies	Illiquidity and risk premium	Weak after adjusting for quality and liquidity
Momentum	Strong recent relative performance	Under-reaction and flow-chasing	Sharp crashes after market bottoms
Quality / profitability	Profitable, stable, low leverage	Mispricing of persistent profitability	Definition varies widely between studies
Low volatility	Lower volatility stocks outperform risk-adjusted	Leverage constraints force demand for high beta	Crowded; interest-rate sensitive
Investment	Companies investing conservatively	Overinvestment destroys value	Overlaps heavily with quality

Premium decay

Research examining anomalies before and after publication consistently finds that measured returns decline substantially once a factor becomes widely known. Estimates of the reduction vary, but the direction is consistent across studies and across factors.

Capital arrives. Funds are launched to harvest the factor, which bids up the cheap side and sells the expensive side, compressing the spread.
Implementation improves. Better data and execution mean more of the theoretical premium is captured, which also means less remains.
Definitions get arbitraged. Simple metrics such as book-to-market become crowded, pushing practitioners toward more complex definitions with correspondingly weaker evidence.
Some decay is illusory. Part of the apparent decline is the original estimate having been inflated by data mining, so the true premium was always smaller.
The underlying causes persist. Behavioural biases and institutional constraints have not disappeared, which is why factors have compressed rather than vanished.

Building a factor portfolio

1
Decide long-only or long-short
Long-only tilts are accessible, cheap, and retain market exposure. Long-short isolates the factor but requires shorting, borrow, and considerably more infrastructure.
2
Choose factor definitions and keep them simple
Simple, widely used definitions have the most evidence behind them. Novel definitions with better backtests are usually fitted.
3
Combine factors rather than concentrating
Value and momentum have historically been negatively correlated, so combining them smooths the return stream considerably more than either alone.
4
Neutralise unintended exposures
A value screen concentrates in specific sectors. Sector caps or explicit neutralisation prevents a factor bet becoming a sector bet.
5
Control turnover
Momentum requires frequent rebalancing; value does not. Cost modelling should be part of the construction rather than applied afterwards.
6
Size for the underperformance period
Assume the factor underperforms for five years or more. Allocate an amount you will still hold at the end of that period.

Knowing your own factor exposures

Every strategy has factor exposures, whether or not they were intended. Regressing your returns against factor returns reveals what you are actually being paid for.

Strategy return =
      alpha
    + b1 x Market
    + b2 x Size
    + b3 x Value
    + b4 x Momentum
    + b5 x Quality
    + error

Example output:
   Market   0.58      Momentum  0.41
   Size     0.12      Quality   0.09
   Value   -0.22      alpha    +0.8% annually (not significant)

Interpretation:
   This "unique" strategy is largely a momentum tilt with
   market exposure and a growth bias.  The same profile is
   available in low-cost funds.  The alpha is statistically
   indistinguishable from zero.

That is valuable information, not a failure.  It tells you
what you are being paid for and what you could replace.

A factor regression on your own strategy returns.

Frequently asked questions

Do factors still work?

Measured premia have declined since publication across most documented factors, which is consistent with capital arriving to exploit them. The underlying causes, behavioural biases and institutional constraints, have not disappeared, so most researchers expect compressed but non-zero premia rather than complete elimination.

Which factor is best?

None consistently. Momentum has the most robust cross-market evidence; quality has performed well with lower turnover; value has the longest history and the most severe recent underperformance. Combining negatively correlated factors produces a smoother result than concentrating in whichever performed best recently.

What is smart beta?

A marketing term for index funds that weight by a factor characteristic rather than by market capitalisation. The underlying idea is sound; the concerns are cost, implementation quality, and whether the specific factor definition used has genuine evidence behind it or was selected for its backtest.

How long should I hold a factor allocation?

Long enough to survive a full underperformance cycle, which historically has meant five to ten years for some factors. Allocations held for shorter periods tend to be bought after strong performance and sold after weak performance, which systematically inverts the intended exposure.

Can factor investing be applied outside equities?

Yes. Momentum and carry appear across currencies, commodities, and bonds, and value has analogues in several asset classes. Cross-asset factor evidence is one of the stronger arguments that these effects reflect genuine behaviour rather than equity-specific data mining.

Test this idea before you trade it

Describe the rules in plain language and AlgoTrader AI turns them into a structured strategy blueprint with a configurable historical backtest, cost assumptions, and exportable code.

Build a backtest

Keep reading

Referenced by

Educational use only. This guide explains how a strategy works. It is not investment advice, not a recommendation, and no result described here is a forecast. Test any approach on historical and out-of-sample data, size positions conservatively, and never risk money you cannot afford to lose.