Strategies · Long-term investing
Factor investing: five factor funds against the S&P 500, tested
What value, size, momentum, quality and low volatility are, the rules of an equal mix of five factor funds, and how it did against SPY from 2013 to 2026 after costs.
On this page
A factor is a measurable trait of a stock that has gone with higher returns, or lower risk, across many stocks and many years. Cheap stocks against expensive ones. Small companies against large. Recent winners against recent losers. Factor investing means holding more of the stocks with those traits than a plain index fund does, by rule and at low cost.
The research behind factors covers most of a century. The funds that sell them are much younger. This lesson explains the five factors most often sold, states a simple rule for holding five factor funds in equal parts, and tests it against SPY, the S&P 500 fund, from August 2013 to October 2026. Over those 13 years the mix earned less than the index. Why that happened, and why it does not settle the question, is the useful part.
What a factor is
Take every US stock, sort them by some trait, and compare the average return of the top group with the bottom group. If the difference is large, lasts for decades and appears in other countries too, the trait is called a factor.
| Factor | The trait | First documented |
|---|---|---|
| Value | A low price for each dollar of book value or earnings | Fama and French (1992, 1993) |
| Size | A small market value | Banz (1981); Fama and French (1993) |
| Momentum | A high return over the past 6 to 12 months | Jegadeesh and Titman (1993) |
| Quality | High and stable profits, low debt | Novy-Marx (2013), among others |
| Low volatility | Small price swings, or a low beta | Frazzini and Pedersen (2014), others |
There are two kinds of explanation, and they matter for what to expect. One says a factor is pay for risk: small and cheap companies fail more often in recessions, so investors demand a higher return to hold them. The other says a factor is a mistake that other investors keep making, such as paying too much for exciting stocks. A risk premium should last but will hurt in bad times. A mistake can be corrected once enough money chases it.
A third possibility is that the factor was never there. Harvey, Liu and Zhu counted more than 300 factors published by 2012 and argued that most would not pass a fair statistical test, because researchers had tried so many sorts on the same data. That is overfitting at the scale of a whole field. The five in the table are the ones that survived the most checks, which is not the same as a promise.
Why a fund is not the paper
The academic factor is usually a long-short portfolio: buy the cheapest third of stocks, sell short the most expensive third, with no net money in the market. A factor fund is long only. It holds the cheap stocks and nothing else, so about nine-tenths of what it does each month is simply the stock market going up or down. The factor is a tilt on top.
The fund also has an index provider's exact definition, a rebalancing schedule and a fee. MTUM follows an MSCI momentum index that is rebuilt twice a year, so for months at a time it can hold yesterday's winners. USMV builds the least volatile portfolio it can under limits on sector weights. These are reasonable designs, and they are not the same thing as the series in the papers.
The rules
An educational outline of the simplest multi-factor portfolio, using five funds with different traits:
- Hold five funds: MTUM (momentum), QUAL (quality), USMV (minimum volatility), IWM (small companies, the Russell 2000) and VTV (large-company value).
- Put 20% of the money in each.
- On the last trading day of every December, work out each fund's share of the total. Sell what is above 20% and buy what is below, so each is back at 20%.
- Do nothing between Decembers.
There is no judgment in it: no choosing which factor is "due", and no switching when one has a bad year. That is deliberate. Picking the factor that did best lately is itself a momentum bet made with far less evidence.
A worked example of the reset
Suppose $10,000 went in at the end of 2016, $2,000 a fund. In 2017 the five funds' prices rose 35.9% (MTUM), 20.0% (QUAL), 16.7% (USMV), 13.1% (IWM) and 14.3% (VTV). At the end of December the holdings were worth $2,718, $2,400, $2,334, $2,262 and $2,286: $12,000 in all.
One-fifth of $12,000 is $2,400. The reset sells $318 of MTUM and buys $66 of USMV, $138 of IWM and $114 of VTV. QUAL is already there. A total of $636 changed hands, about 5% of the portfolio, and at 0.10% a trade that cost about 64 cents.
What the backtest shows
The test, stated in full so it can be checked:
- Data. Month-end closing prices of the five funds and of SPY from August 2013, the first full month of the youngest fund (QUAL), to October 2026: 159 months, a little over 13 years.
- Trading. The first purchase and every December reset happen at that month's closing prices.
- Costs. 0.10% of the amount on every buy and every sell. The funds' own fees are already inside their prices: 0.15% a year for MTUM, QUAL and USMV, 0.19% for IWM and 0.03% for VTV (issuers' pages, November 2026), against about 0.09% for SPY.
- Left out. Dividends and taxes. The funds' dividend yields differ: VTV has paid clearly more than SPY and MTUM less, so the comparison is a little unkind to the mix. The difference is a fraction of a percentage point a year, much smaller than the gap in the table.
| Measure | Five factor funds | SPY |
|---|---|---|
| Total return | 280.0% | 361.8% |
| Return per year (CAGR) | 10.7% | 12.3% |
| Deepest fall from a peak | −23.3% | −24.8% |
| Longest time below a peak | 25 months | 23 months |
| Trading costs paid | 0.2% in 13 years | none |
One dollar became $3.80 in the mix and $4.62 in SPY. The mix gave up 1.6 percentage points a year and fell almost as far in its worst stretch, from December 2021 to September 2022.
Costs are not the reason. The resets moved about 30% of the portfolio in total over 13 years, and all the trading together cost 0.2% of the starting amount.
The five funds one at a time
| Fund | Return per year | Deepest fall | When |
|---|---|---|---|
| MTUM, momentum | 13.7% | −31.2% | October 2021 – Sept. 2022 |
| SPY | 12.3% | −24.8% | December 2021 – Sept. 2022 |
| QUAL, quality | 12.2% | −28.6% | December 2021 – Sept. 2022 |
| VTV, value | 9.5% | −25.7% | December 2019 – March 2020 |
| USMV, low vol. | 9.1% | −19.6% | January – March 2020 |
| IWM, small | 8.1% | −33.8% | August 2018 – March 2020 |
Only momentum finished ahead of the index. Quality matched it. Value, low volatility and small companies each trailed by three to four points a year. The low-volatility fund did what its name says, with the smallest worst fall, and paid for it in return.
The two halves of the period
| Period | Five factor funds | SPY |
|---|---|---|
| August 2013 – December 2019 | 11.3% a year | 11.3% a year |
| December 2019 – October 2026 | 10.1% a year | 13.3% a year |
Through 2019 the mix and the index were level. The whole shortfall came afterwards, in the years when a handful of very large technology companies drove the S&P 500 (Market concentration: what mega caps mean for S&P 500 investors). SPY weights companies by size, so it held more and more of them. Small-company, value and low-volatility funds held little or none.
Other ways to run it
| Reset | Return per year | Deepest fall |
|---|---|---|
| Every December | 10.7% | −23.3% |
| Every quarter | 10.8% | −23.4% |
| Every month | 10.7% | −23.4% |
| Never (buy and hold) | 10.8% | −24.0% |
The reset schedule made no difference worth the name. That is reassuring about the test, and a reminder of what it is: five diversified US stock funds that move together most of the time.
Where it fails: the losing periods
Measured against the index, the mix lost in most years after 2018.
| Year | Five factor funds | SPY |
|---|---|---|
| 2019 | 25.6% | 28.8% |
| 2020 | 12.9% | 16.2% |
| 2021 | 18.9% | 27.0% |
| 2022 | −15.7% | −19.5% |
| 2023 | 13.3% | 24.3% |
| 2024 | 18.0% | 23.3% |
| 2025 | 12.5% | 16.4% |
| 2026, to October | 13.5% | 10.8% |
- 2023 was the worst. The mix trailed by 11 points. MTUM gained 7.5% while SPY gained 24.3%: the momentum index had been rebuilt after the falls of 2022 and held that year's defensive winners when the market turned back to technology.
- 2020 split the factors. MTUM rose 28.4% and VTV fell 0.7% in the same year. A single factor can be far from the market for a long time.
- 2022 was the one clear win, by 3.8 points, almost all of it from VTV (−4.6%) and USMV (−10.9%).
- Small companies had the longest drought. IWM fell 33.8% from August 2018 to March 2020, then spent 40 months below its June 2021 peak, until November 2024.
Other ways it fails or misleads:
- Long droughts are normal for factors. In the academic data, value trailed growth for more than a decade after 2007. A factor investor needs the patience to hold something that lags the headline index for years, which is exactly when most people stop (Why most active funds underperform: costs, arithmetic and luck makes the same point about funds).
- The test period is short and one-sided. Thirteen years that ended with large growth companies far ahead is the worst case for a mix that tilts away from them. A period that included 2000 to 2007, when small and value stocks led, would look very different. The data available here cannot show it, because three of the five funds did not exist.
- Crowding. If a factor is a mistake other investors make, money flowing into factor funds should shrink it. Published factors have tended to earn less after publication than before.
- Taxes. Momentum funds trade a lot inside the fund. In a US taxable account the yearly reset also realizes gains.
Compared with its neighbours
| Approach | What it holds | In this test | What it asks of the investor |
|---|---|---|---|
| Index fund (SPY) | Every company, by size | 12.3% a year | Nothing |
| Five factor funds, equal parts | Five tilts, 20% each | 10.7% a year | Staying put while it lags |
| One factor fund | One tilt | 8.1% to 13.7% a year | A view on which factor, and more patience |
| Growth and value funds | The market split in two by valuation | Not tested here | See Growth vs value investing: styles, P/E and the long record |
How Finabase does it
Finabase does not build factor portfolios and does not rate factors. Three tools show the pieces.
- Comparison. The link above opens MTUM with SPY, QUAL and USMV drawn from the same starting value, which is the fastest way to see how far a factor fund has drifted from the index. The chart covers two years by default (Compare stocks side by side on finabase.io).
- Screener. The screener builder filters and orders stocks by P/E ratio, market cap, dividend yield and one-year return, the raw material of value, size and momentum sorts. It screens on today's values; it does not test what such a screen returned in the past (Use the Finabase stock screener and screener builder on the web).
- Backtest and Strategy. The Backtest tool trades indicator signals on one symbol, with a fixed $10,000 per signal and no trading cost. The Strategy tool can split an amount across several funds by weight, but it replays two years and also charges no cost. Neither can run the 13-year, five-fund test on this page.
Frequently asked questions
Does factor investing beat the market?
Not in this test. Five factor funds in equal parts returned 10.7% a year from 2013 to 2026 against 12.3% for SPY. The academic record over much longer periods is more favourable, but it is measured on long-short portfolios before costs, not on funds.
Which factor has done best?
Among these five funds since 2013, momentum: MTUM returned 13.7% a year. Value, low volatility and small companies all trailed the index. Rankings like this have changed from decade to decade, which is the reason for holding several.
Is smart beta the same as factor investing?
Mostly, yes. "Smart beta" is a marketing name for index funds that weight stocks by something other than size, usually one or more factors. The funds in this lesson are smart beta funds.
Do factor funds cost more than index funds?
A little. The three iShares factor funds here charge 0.15% a year and SPY about 0.09%. The fee difference is small beside the difference in what the funds hold.
Summary
Factors are traits, such as cheapness, small size and recent strength, that have gone with higher returns in long academic records. Funds make them easy to hold, but a long-only fund is mostly the stock market with a tilt, defined by an index provider and rebuilt on a schedule. An equal mix of five factor funds, reset each December, matched SPY from 2013 to 2019 and then fell behind, ending at 10.7% a year against 12.3% with almost the same worst fall. Only momentum beat the index. The period is short and favoured the largest growth companies, so the test neither confirms nor refutes the factors. What it does show is the cost of a tilt: years of trailing a simpler fund, with no date at which the payoff is due.
Check your understanding
4 questions on this lesson. Each answer comes with the reason, and nothing is scored.
A long-only value fund holds the cheapest third of large US stocks. In a month when the whole stock market falls 8%, what should its holder expect?
- A gain, because value is a separate source of return
- A fall of roughly the same size, a little more or less
- No change, because factors are unrelated to the market
Show the answer
B. A fall of roughly the same size, a little more or less A long-only factor fund is mostly the stock market with a tilt on top. The academic factor buys some stocks and sells others short, which removes the market; the fund does not. In 2022 the five-fund mix fell 23.3% from its peak and SPY 24.8%.
The five-fund mix matched SPY from 2013 to 2019 and trailed it by about three points a year from 2019 to 2026. What is the sound reading?
- Factors stopped working in 2019
- The result depends heavily on one stretch, when a few very large growth companies led the index, so 13 years cannot settle the question
- The December reset was the wrong schedule
Show the answer
B. The result depends heavily on one stretch, when a few very large growth companies led the index, so 13 years cannot settle the question Splitting the period shows the whole gap came from one stretch with one cause. The reset schedule made no measurable difference (10.7% to 10.8% a year for every schedule tried). A test this short shows how the funds behaved, not whether the factors exist.
Why does the rule hold all five funds at fixed weights and not move money to the factor that did best last year?
- Because switching is not allowed in a fund
- Because chasing the recent winner is its own momentum bet, made on a handful of yearly observations
- Because the five funds always move in opposite directions
Show the answer
B. Because chasing the recent winner is its own momentum bet, made on a handful of yearly observations Factor rankings changed from year to year in the test: VTV fell 0.7% in 2020 when MTUM rose 28.4%, then VTV fell least in 2022. A rule with no judgment can be tested; a habit of switching to last year’s winner cannot, and has little evidence behind it.
Researchers have published more than 300 factors. Why is that a reason for caution and not confidence?
- With so many sorts tried on the same history, some will look strong by chance alone
- More factors mean more diversification
- Only the most recent factors are reliable
Show the answer
A. With so many sorts tried on the same history, some will look strong by chance alone Trying many rules on one set of data and reporting the ones that worked is overfitting. Harvey, Liu and Zhu argued that most published factors would fail a test that allowed for how many had been tried. The few that held up across periods and countries are the ones funds sell.
Sources and further reading
- Eugene F. Fama and Kenneth R. French, “Common risk factors in the returns on stocks and bonds,” Journal of Financial Economics 33 (1993)
- Rolf W. Banz, “The relationship between return and market value of common stocks,” Journal of Financial Economics 9 (1981)
- Narasimhan Jegadeesh and Sheridan Titman, “Returns to buying winners and selling losers,” Journal of Finance 48 (1993)
- Robert Novy-Marx, “The other side of value: the gross profitability premium,” Journal of Financial Economics 108 (2013)
- R. David McLean and Jeffrey Pontiff, “Does academic research destroy stock return predictability?,” Journal of Finance 71 (2016)
- Andrea Frazzini and Lasse Heje Pedersen, “Betting against beta,” Journal of Financial Economics 111 (2014)
- Campbell R. Harvey, Yan Liu and Heqing Zhu, “… and the Cross-Section of Expected Returns,” Review of Financial Studies 29 (2016)
- iShares — MSCI USA Momentum Factor ETF (MTUM), fund page: index, expense ratio 0.15% (read November 2026)
- iShares — MSCI USA Min Vol Factor ETF (USMV), fund page: expense ratio 0.15% (read November 2026)
- Finabase chart data — MTUM, QUAL, USMV, IWM, VTV and SPY month-end closes, August 2013 to October 2026; portfolio and results computed by Finabase
