What it means
The oldest model in finance blames everything on the market. A multifactor model argues that returns have several parents, not one, and that measuring each separately explains portfolios far better.
The idea grew from academic work on asset pricing. Researchers noticed that small companies and cheap companies beat the single-factor market model's predictions so persistently that the anomalies had to be risks or rewards of their own.
The famous versions carry their authors' names. Fama and French added size and value factors to the market, later profitability and investment, and their factor data is published openly through Kenneth French's library at Dartmouth's Tuck School.
A factor is a recipe, not a vibe. Each is built from real portfolios, for example the return of small stocks minus big stocks, so every factor has a measurable history and a live daily value that models can regress against.
Practitioners use the models in two directions. Backward, they decompose a fund's past returns to see what the manager actually bet on; forward, they build portfolios that load deliberately on the factors they believe are rewarded.
The discipline cuts both ways. A manager who charges for skill but merely tilts toward value and small companies is exposed by a factor regression, while a genuinely factor-driven return can be had cheaply through systematic funds.
For a business owner reviewing a pension fund or treasury portfolio, the question the model answers is simple. Is this return the market, a known factor anyone can buy, or something the manager truly added?
Momentum and quality joined the family later. Researchers documented that recent winners tend to keep winning for a while, and that profitable, conservatively financed firms earn premiums, so many models now carry five or six factors.
Regressions turn the theory into arithmetic. By fitting a fund's historical returns against factor series, the model estimates loadings that describe the portfolio's character better than any marketing label.
In practice
Real-world examples.
Example
A consultant shows an endowment that its star manager's outperformance vanishes once small-cap and value factors are accounted for, ending the manager's mandate. The board accepts the evidence rather than the narrative.
Example
A quantitative fund builds a portfolio loading equally on value, momentum and quality factors, accepting that each will disappoint in some years. Rebalancing among factors is rules-based, never emotional.
Example
A wealth manager uses a three-factor model to explain to a client why her portfolio lagged in a year when large growth stocks dominated the index. The explanation preserves the relationship through the downturn.
Formula
Calculation
Return = a + b1 x market factor + b2 x size factor + b3 x value factor + error. If a fund returns 9% and its factor loadings explain 8.2% of that, the unexplained slice, alpha, is 9% - 8.2% = 0.8%, and it is usually smaller than the fees.
Here is how the loadings produce the explained slice. Suppose the market premium is 5%, the size premium 2% and the value premium 3%, and the fund has loadings of 1.1, 0.5 and 0.4. The factor-explained excess return is 1.1 x 5% + 0.5 x 2% + 0.4 x 3% = 5.5% + 1.0% + 1.2% = 7.7%, so a fund with an 8.5% excess return has alpha of 8.5% - 7.7% = 0.8%.Case study
Seen in the real world.
In this illustrative fictional case, Sofia, trustee of a family company's pension scheme, runs a factor regression on an expensive active fund. The output shows returns fully explained by the market plus a strong value tilt, both available in index funds for a tenth of the fee. She moves the assets to a cheap multifactor fund, keeps the same exposures, and saves the scheme six figures a year. The lesson she carries forward is that what gets measured honestly gets priced honestly.
Watch out
Common mistakes.
- Believing more factors automatically mean a better model, when each added factor risks fitting noise in the past that never repeats in the future.
- Reading alpha as skill over short windows, when factor luck can masquerade as genius for years before the regression tells the truth.
- Treating factors as guaranteed payoffs, when value and size can underperform for a decade, testing every investor's conviction. Position size and patience decide whether the premium is ever harvested.
Questions
People also ask.
What is a multifactor model in finance?
A model that explains asset returns using several shared drivers, or factors, such as the market, company size and value style, rather than market returns alone. It measures how much each factor contributed. The unexplained remainder is called alpha.
Which multifactor models are best known?
The Fama-French three-factor and five-factor models, built on market, size, value, profitability and investment factors. Their factor histories are published through Kenneth French's Dartmouth data library. Practitioners often add momentum as a sixth.
Why do investors use factor models?
To see what really drives a portfolio, to test whether active fees buy anything beyond cheap factor exposure, and to build deliberate, diversified bets on rewarded risks. They also discipline risk budgeting across the whole portfolio.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%