Is the Value Premium Back? What AVUV, AVDV and DFSV's Big Year Shows
AVUV returned 27.9% in the year to Oct 31, 2026, vs 12.1% for the S&P 500, yet trails it over three years. What drove the run, and is value still cheap?
Small-cap value funds had a year worth posting about. In the 12 months to October 31, 2026, Avantis U.S. Small Cap Value (AVUV) returned 27.9% at net asset value, against 12.1% for the S&P 500, and Avantis International Small Cap Value (AVDV) returned 21.8%.124 A Portfolio Visualizer table that circulated on X in November put AVDV at 14.59% a year since its 2019 launch, against 10.10% for Vanguard’s total international stock fund (VXUS).
The Fama-French value factor spent nearly 14 years losing more than half its value against growth stocks, so a run like this invites the question in the title. I hold AVGV and other Avantis value funds, which gives me a stake in the answer. I checked the returns on matched dates against each fund’s own benchmark, split the past year into factor exposures, measured how cheap value stocks are today with a century of Ken French’s data, and tested whether that cheapness has ever told investors when to buy.
The short version
Value is having an impressive year, and in developed markets outside the U.S. it has been paying since 2021. AVDV beat its benchmark by 8.1 points a year over the five years to October 2026, and the value factor in those markets gained in every calendar year from 2021 to 2025 and again through September 2026. The U.S. record is thinner. Over three years AVUV still trails the S&P 500 (20.5% against 23.4% a year), and since its 2019 launch it has roughly matched the total U.S. market with a much deeper drawdown. AVUV returned 19.9% over the 12 months to September 2026 even though the profitability screen that sets it and DFSV apart from an index fund had one of its worst stretches since the 1960s. Value stocks are cheap relative to growth on book value and on earnings, though on earnings less so than at the 1999 and 2019 highs. That cheapness has not said when value would pay. Across a century of starting points, small value beat small growth in about three of every four five-year windows, by a similar average margin whether the spread began wide or narrow. I am rebalancing to my target, which after this year means trimming.
The returns, with dates and benchmarks attached
Fund returns depend on the end date, the benchmark and the share class, and most screenshots leave out at least one of the three. These are the issuers’ own net-asset-value figures, all to the same date.
| Fund or index | Year to date | 1 year | 3 years | 5 years |
|---|---|---|---|---|
| AVUV | 22.20% | 27.93% | 20.47% | 11.33% |
| Russell 2000 Value (AVUV benchmark) | 19.90% | 23.49% | 21.37% | 7.53% |
| AVDV | 14.06% | 21.84% | 28.25% | 13.68% |
| MSCI World ex USA Small Cap (AVDV benchmark) | 7.99% | 12.52% | 20.48% | 5.54% |
| QVAL | 33.96% | 42.22% | 24.29% | 14.52% |
| S&P 500 | 11.77% | 12.11% | 23.40% | 12.06% |
Total returns through Oct 31, 2026; 3- and 5-year figures are annualized. Funds at net asset value. Sources: Avantis fund pages for AVUV, AVDV and their benchmarks; Alpha Architect for QVAL; State Street’s SPY page for the S&P 500 index.
The longer record splits by region. AVUV beat its benchmark year to date and over one and five years but trailed it over three, and it trailed the S&P 500 over three and five. AVDV beat its benchmark in every column, by 9.3 points over the past year and by 8.1 points a year over five. Its benchmark holds international small companies of every style, so that lead is mostly the value tilt paying off abroad. QVAL, Alpha Architect’s concentrated quantitative value fund, returned 42.2% over 12 months, 12.92% a year over ten years and 9.95% a year since its October 2014 launch.3
One month can reverse the comparison. Through September 30, AVUV’s one-year return was 19.91% against its benchmark’s 20.77%, and its year-to-date return was 16.60% against 16.96%. Through October 31 it led on both.1 Against the S&P 500 the swing is larger: AVUV led by 4.2 points over the 12 months to September 30 (19.91% against 15.74%) and by 15.8 points over the 12 months to October 31. The later window dropped October 2025, when AVUV fell 1.7% and SPY rose 2.4%, and added October 2026, when AVUV rose 4.8% and SPY fell 0.9% (market prices). A comparison that does not state its end date can be arranged to show either result.
Dimensional’s US Small Cap Value ETF (DFSV) shows 19.19% for the year to September 30, the date of its latest fact sheet.5 That figure is almost entirely the mutual fund DFSVX’s record: the ETF share class began trading on September 28, 2026, for reasons explained in the DFSV section.
Since AVUV launched
AVUV and AVDV both began trading on September 24, 2019. To compare them with the funds people usually set beside them, I used month-end market prices with distributions reinvested over the same 85 months.
| Fund | Annual return | Annual volatility | Worst month-end drawdown |
|---|---|---|---|
| AVUV (Avantis U.S. small value) | 15.39% | 25.4% | −42.4% |
| DFSVX (Dimensional U.S. small value) | 14.33% | 23.4% | −39.0% |
| IWN (Russell 2000 Value index fund) | 10.67% | 22.8% | −35.8% |
| VTI (total U.S. market) | 15.20% | 17.0% | −24.8% |
| SPY (S&P 500) | 15.73% | 16.5% | −23.9% |
| VT (total world market) | 13.04% | 16.0% | −25.5% |
| AVDV (Avantis international small value) | 14.59% | 20.1% | −33.9% |
| DISVX (Dimensional international small value) | 14.03% | 19.5% | −33.7% |
| VSS (international small caps) | 8.39% | 18.7% | −31.8% |
| VXUS (total international market) | 9.93% | 16.2% | −27.8% |
| QVAL (Alpha Architect quantitative value) | 14.60% | 23.1% | −40.3% |
Sep 30, 2019 to Oct 31, 2026, from Yahoo Finance adjusted closes (market prices with distributions reinvested; DFSVX and DISVX are mutual funds priced at NAV). Drawdowns use month-end prices, so intra-month lows were deeper. Source: Summitward calculations.
Since launch, AVUV compounded at 15.4% a year, against 15.2% for the total U.S. market (VTI) and 13.0% for the world market (VT), and it fell 42% from peak to trough at month-ends against 25% for VTI. Of the 74 rolling 12-month windows ending September 2020 through October 2026, AVUV beat VTI in 53%. Its clearest lead is over the Russell 2000 Value index fund, IWN: 15.4% against 10.7% a year. In regressions on the five Fama-French factors plus momentum over October 2019 to September 2026, IWN’s annual intercept was −2.3% (95% interval −4.1% to −0.6%) and AVUV’s was +0.5% (−2.1% to +3.1%). My Avantis history covers why a small-value index fund lags the factors it is built on.
Where the past year’s value premium came from
Ken French’s six portfolios formed on size and book-to-market split the market’s moves by company size.8 His data currently run through September 2026.
| Portfolio | 12 months to Sep 2026 | Jan to Sep 2026 |
|---|---|---|
| Large value | 33.0% | 21.9% |
| Large growth | 10.4% | 8.2% |
| Small value | 18.5% | 15.8% |
| Small growth | 16.8% | 13.2% |
Value-weighted total returns of the Fama-French 2x3 portfolios: value is the cheapest 30% by book-to-market within each size half, growth the most expensive 30%. Source: Kenneth R. French Data Library (202609 vintage); Summitward calculations.
Value beat growth by 22.6 points among large companies and by 1.7 points among small ones. Compounded monthly, the value factor (HML) returned 9.7% over the 12 months. The size factor was close to zero: −1.3% compounded in the three-factor version, +1.5% summed in the five-factor version used in the next section. Small-cap value funds did well mostly because stocks rose and value beat growth; being small added little. Outside the U.S., the developed-market value factor’s monthly returns summed to 20.1% over the same 12 months, about double the U.S. sum of 9.8%. That is consistent with AVDV’s lead over AVUV in the 12 months to September 30 (26.99% against 19.91% at NAV).2 By October 31, after a strong month for U.S. small caps and a weak one abroad, AVUV led again, 27.93% to 21.84%.
Factor exposure accounts for most of AVUV’s 12 months to September
A factor regression measures how a fund’s returns move with the market, size, value, profitability, investment and momentum factors. I estimated each fund’s loadings from October 2019 to September 2025 and multiplied them by the factor returns of the following 12 months, the latest in French’s data. The loadings use no data from that year, but the factor returns are the ones the year delivered, so this is an after-the-fact attribution, not a forecast.
For AVUV, the loadings implied 13.5 points of return above Treasury bills over October 2025 to September 2026, and the fund delivered 15.4 (monthly returns summed; as compounded total returns, 17.4% implied and 19.9% actual). Market exposure supplied 11.5 points and the value loading 5.3. The 1.9-point remainder is less than half of AVUV’s residual volatility of 4.2% a year over the estimation window, so a single year of it is weak evidence about the manager. DFSVX, the mutual fund whose record DFSV now carries, shows the same pattern: 13.4 implied, 14.8 actual. AVDV’s value loading supplied 10.5 of its 22.2 points.
The profitability bar went the other way. Avantis and Dimensional tilt away from unprofitable small companies, and the Russell index fund does not. From October 2025 to September 2026 the profitability factor’s (RMW, robust minus weak) monthly returns summed to −15.6%, lower than all but 2.7% of rolling 12-month windows since its data begin in 1963. The worst on record was −44.5%, in the year to February 2000. That headwind cost AVUV about 4.3 points. On the same arithmetic, AVUV’s higher market, size and value loadings implied about 2.5 points more return than IWN and its profitability loading 4.4 points less, a net gap of −1.9; the monthly differences summed to −0.4. The feature that distinguishes these funds worked against them this year, and they roughly kept pace anyway. The quality-minus-junk guide covers why that tilt is there.
Checking the table from X
The table on X compared AVDV with VXUS, Dimensional’s international small value mutual fund (DISVX) and Vanguard’s FTSE All-World ex-US Small-Cap ETF (VSS). My month-end recalculation for September 2019 to October 2026, in the table above, reproduces it closely: AVDV 14.59%, DISVX 14.03%, VSS 8.39%, and VXUS 9.93% against the table’s 10.10%. Four details change what such a table means.
- The benchmark. VXUS holds international stocks of every size and style, including emerging markets, so beating it shows a small-value tilt paid over these seven years. VSS holds international small caps of every style. AVDV’s own benchmark, MSCI World ex USA Small Cap, leaves out emerging markets. Each comparison answers a different question.
- The end date. AVDV’s year-to-date return was 18.87% at September 30 and 14.06% at October 31, 2026.2
- The share class. DISVX is a mutual fund. Its ETF sibling, DISV, launched in March 2022 and has a different tax record (see the distributions below).
- Price or NAV. Portfolio Visualizer and Yahoo use market prices; issuers report NAV. On November 13, 2026, AVDV’s market price was 0.54% above its NAV, per Avantis.2 For an international fund, part of such a gap reflects foreign markets closing hours before U.S. trading ends.
None of the four overturns the table’s main result. Over the same 85 months AVDV beat VXUS by 4.7 points a year and VSS by 6.2, and over five years it beat its own benchmark, which leaves out emerging markets, by 8.1 points a year.
DFSV’s track record changed on September 25, 2026
Dimensional merged the stand-alone DFSV ETF into its U.S. Small Cap Value Portfolio, the mutual fund behind DFSVX, as a new ETF share class.67 The ticker kept trading, and the ETF class inherited the mutual fund’s record back to March 1993. The fact sheet now shows a 1993 inception date and different calendar-year returns.
| Year | DFSV ETF as reported Dec 2025 | DFSV as reported Sep 2026 | Difference |
|---|---|---|---|
| 2023 | 19.23% | 18.91% | −0.32 |
| 2024 | 7.27% | 9.57% | +2.30 |
| 2025 | 8.51% | 8.38% | −0.13 |
NAV returns. The left column is the ETF’s own record (inception Feb 23, 2022) from Dimensional’s Dec 31, 2025 fact sheet; the right column is the Sep 30, 2026 fact sheet, whose returns before Sep 25, 2026 are those of the fund’s oldest share class. Yahoo Finance prices give 7.1% for the old ETF and 9.6% for DFSVX in 2024.
Any DFSV return that reaches back before late September 2026 now mostly reflects DFSVX. Earlier Summitward guides that ran regressions on “DFSV” used the old ETF’s February 2022 to September 2026 returns, and they say so.
A long wait in the U.S., a shorter one abroad
The U.S. value factor fell 57.8% from December 2006 to September 2020 and was still 32% below that peak in August 2026, as the value traps guide shows. Li, Ng and Swaminathan report that it averaged 5.20% a year from July 1926 to December 2007 and −0.86% a year from January 2008 to March 2025. Their paper is titled “Is the Value Premium Dead?” and concludes that “recent value underperformance reflects cyclical variation rather than a permanent shift.”9 Arnott, Harvey, Kalesnik and Linnainmaa found that value’s 55% drawdown through mid-2020 is accounted for by value stocks becoming cheaper relative to growth stocks; in a separate result, a value measure that counts intangible capital did better than the traditional one.10 Among small companies the drought was as long: French’s small value and small growth portfolios tied at 8.4% a year from February 2004 to December 2025.
Outside the U.S., value’s drawdown was shallower (35% from April 2014 to September 2020) and the recovery has been steadier. French’s developed ex-US value factor returned 9.9% in 2021, 25.3% in 2022, 7.6% in 2023, 12.7% in 2024 and 22.5% in 2025, and 13.7% from January to September 2026; in July 2026 it set a high for a series that begins in 1990. The U.S. factor gained 22.2% in 2021 and 31.7% in 2022, then lost 11.1% and 7.0% over the next two years, and from September 2019, when AVUV and AVDV launched, to September 2026 it gained 2.6% in total.8 Each fund’s record against its home market follows the same split: since launch, AVDV beat VXUS by 4.7 points a year and AVUV beat VTI by 0.2.
Value stocks are still cheap on book value
The value spread measures how much more investors pay for growth stocks than for value stocks: growth’s price-to-book multiple divided by value’s. I built it from the portfolio characteristics French publishes, which go back to 1926. The value spread page updates it each month when his new data post.
- Small caps. The spread was 6.3 when French formed his portfolios in June 2026, down from 8.1 a year earlier. Rolled forward on price changes to September, it was about 6.5, higher than in 82% of months since 1926 and 92% since 1963. Its 2020 peak was 9.4, in December 2020; at the dot-com top it was 7.3. The all-time highs came between 1935 and 1940.
- Large caps. 9.7 in June 2026 and about 10.1 by September, higher than in 97% of months since 1926, above the 9.8 of June 2000 and below the 11.4 of September 2020.
- Holding profitability constant. Comparing cheap and expensive small companies within each of four profitability groups gives 6.45 for June 2026, almost the same as the 6.33 for all small caps, so the small-cap discount is not explained by cheap companies being less profitable, at least on French’s measure of operating profit over book equity. Large caps differ: the same adjustment lowers the June 2026 reading from 9.71 to 8.33. A 14% cut is larger than in all but 6% of years since 1963 (the adjustment usually raises the reading by about 5%), so more of today’s large-cap discount than usual reflects cheap large companies being less profitable.
- On earnings. For all U.S. stocks at December 2025 prices, the earnings-to-price spread was 4.35, higher than in 87% of years since 1950 but well below 6.48 in December 1999 and 6.75 in December 2019. The book-to-market version was 9.90, level with December 1999. Book value leaves out internally built intangibles such as research and brands, which are concentrated in growth companies, so part of the book-value gap is accounting.14
The large-cap numbers show how a big value year and a wide spread can coexist. If you hold June 2025’s large-cap portfolios fixed, the next 12 months of prices (value up 43%, growth up 17%, before dividends) narrowed their spread from 9.5 to 7.7. When French re-formed the portfolios in June 2026, with new book values and with the stocks that had rallied moved out of the cheapest group, it read 9.7 again. A typical June re-sort moves the large-cap reading by about 6%; this one moved it by 25.5%. A value fund that keeps rebalancing into whatever is cheapest now holds something closer to the re-formed portfolio, so the re-formed reading is the better guide to what its investors face, though a fund’s holdings differ from French’s portfolios.
AQR’s measure, which blends five valuation ratios for large stocks, put the U.S. at the 80th percentile of its history, emerging markets at the 62nd and developed international markets at the 42nd, according to Larry Swedroe’s October 23 Morningstar column. The column’s text says the data run through September 2026, and its chart caption says June 30, 2026.11 On that measure international value, the region whose value factor has gained every year since 2021, is the least cheap of the three.
A wide spread has been a weak timing signal
Each June from 1926 to 2021, I recorded the spread and what value then returned over growth, within the same size half, for the next five years.
| Starting spread | Starts | Average edge, next 5 years | Share where value trailed |
|---|---|---|---|
| Small: narrowest third (3.5 to 4.4) | 32 | +5.3% a year | 22% |
| Small: middle third (4.4 to 5.2) | 32 | +6.6% a year | 9% |
| Small: widest third (5.2 to 13.9) | 32 | +4.8% a year | 38% |
| Small: at least 6.33 (June 2026 level) | 18 | +5.7% a year | 39% |
| Large: narrowest third (3.1 to 4.4) | 32 | +0.8% a year | 44% |
| Large: middle third (4.4 to 5.3) | 32 | +2.4% a year | 22% |
| Large: widest third (5.3 to 12.3) | 32 | +3.9% a year | 38% |
| Large: at least 9.71 (June 2026 level) | 5 | +8.9% a year | 20% |
Edge is the annualized total return of French’s value portfolio minus growth in the same size half over the five years from each July, 1926 to 2021 starts. Windows overlap, so the 96 starts contain only 19 independent five-year periods, and the thirds are ranked with hindsight. Source: Kenneth R. French Data Library; Summitward calculations.
For small caps, starting points in the widest third were followed by about the same average edge as the narrowest third and trailed more often. The 18 starts at least as wide as June 2026 averaged 5.7% a year, close to the 5.6% average for all 96. A forecast that used only the data available at each date did worse than the running historical average, with an out-of-sample R² of −1.7% at five years and −14% at ten.
The same starts also show how often value won wherever the spread began. Small value beat small growth in 74 of the 96 overlapping five-year windows, by 5.6% a year on average, and in 77 of 91 ten-year windows. Large value beat large growth in 63 of 96 five-year windows and 65 of 91 ten-year windows, by 2.4% and 2.5% a year on average. These are paper portfolios before fees and trading costs, and small growth has had the lowest return of French’s six size and value portfolios (8.8% a year since July 1926, against 14.2% for small value), so a fund investor would have captured less of the small-cap edge. IWN’s −2.3% intercept above shows what implementation can cost.
Large caps look better in the full sample. Starts in the widest third were followed by a value edge of 3.9% a year against 0.8% for the narrowest third, and the five starts at least as wide as June 2026 averaged 8.9% (a Newey-West t-statistic of 2.2, which overlapping windows tend to overstate, and R² of 8% at ten years). The relationship did not hold up when each forecast was fit only on data available at the time (out-of-sample R² −4.2% at five years and −6.1% at ten), and five overlapping starts are too few to lean on. Li, Ng and Swaminathan find the same for the plain value spread in month-ahead forecasts from December 2000 to December 2023: its out-of-sample R² is negative in all four of their tests, while their implied value premium, built from analysts’ earnings forecasts, scores between 2.11% and 2.85%. The authors conclude that investors should allocate more to value when it is cheap on their measure; their sample ends in 2023, and the paper does not report a current reading.9 Asness, Friedman, Krail and Liew reported in 2000 that spreads forecast value-minus-growth returns in their sample, and in May 2020 Cliff Asness wrote that AQR’s price-to-book spread had reached its 100th percentile at the end of March.1312 The spread says value is priced cheaply. For small caps it has not said when that will pay off, and for large caps the evidence is weak.
What I recommend
If you already hold a value tilt
Rebalance to your target. After a year when AVUV gained 28% and the total market 12%, that may mean trimming. In a taxable account, send new contributions and dividends to the underweight funds before selling anything. For 2026, Avantis estimates no capital gains distributions from AVUV, AVDV or AVES, and Dimensional estimates none from DFSV, DISV or DFEV. Dimensional’s international small value mutual fund, DISVX, expects $1.577 a share, 4.5% of its value.1516 In a taxable account, that distribution is a tax bill this year that the ETF holders will not get.
If this year makes you want to start one
Size it so you could hold it through droughts like the ones above. The U.S. value factor spent nearly 14 years below its 2006 peak, and French’s small value portfolio only tied small growth from February 2004 to December 2025. A tilt is worth holding only if you would keep it through a stretch like that. The factor loadings guide estimates the U.S. value premium at 3.5% a year from July 1963 to August 2026, with a 95% interval of 0.4% to 6.7%: positive over that span, with a wide range for its size.
The fund-level drawdowns above overstate what a measured tilt does to a portfolio. Over AVUV’s and AVDV’s first seven years, a 20% sleeve beside a total-market fund changed the portfolio’s return and worst drawdown by about a point or less.
| Portfolio | Annual return | Worst month-end drawdown | Tracking error |
|---|---|---|---|
| VTI | 15.20% | −24.8% | 0.0% |
| 80% VTI, 20% AVUV | 15.46% | −25.4% | 3.1% |
| AVUV | 15.39% | −42.4% | 15.3% |
| VXUS | 9.93% | −27.8% | 0.0% |
| 80% VXUS, 20% AVDV | 10.89% | −27.4% | 1.4% |
| AVDV | 14.59% | −33.9% | 7.0% |
Sep 30, 2019 to Oct 31, 2026 (85 months), Yahoo Finance adjusted closes; mixes rebalanced monthly, before costs and taxes. Tracking error is the annualized volatility of the return gap against VTI or VXUS. One window, which includes value’s 2020 low but no decade-long drought. Source: Summitward calculations.
The cost of a tilt shows up as tracking error, the years when your portfolio trails the index fund you could have owned instead; for the 20% AVUV mix it was 3.1% a year. AVUV vs. AVDV covers how to split a tilt between the U.S. and other markets.
If you need the money within five years
Do not count on a factor premium. AVUV fell 42% from peak to trough at month-ends between its 2019 launch and October 2026, and over the five years after the widest-spread starting points, small value trailed small growth 38% of the time.
When a tilt is the wrong choice
Skip it if trailing a world index fund for several years would push you to sell, if the extra holdings make your portfolio harder to manage, or if selling an existing taxable position to fund it would realize large gains. The guide to sponsor and tax lock-in risk covers the last case. A total-market fund is a sound portfolio on its own.
Measure your own value tilt
Summitward's Portfolio tab runs Fama-French factor regressions on your holdings and shows how much size, value and profitability exposure you carry.
Open portfolio analysisFrequently asked questions
Is the value premium back?
In developed markets outside the U.S., it has been back since 2021: French’s value factor there gained in every calendar year from 2021 and set a record in July 2026. In the U.S., the 12 months to September 2026 were strong, mostly among large companies (French’s large value portfolio beat large growth by 22.6 points), after losses in 2023 and 2024. One year does not show whether the U.S. premium has returned for good. Value stocks are still cheap relative to growth on book value, and that cheapness has been a poor guide to timing.
Why did AVUV beat the S&P 500 in 2026?
Value beat growth, and the end date matters. AVUV led the S&P 500 by 4.2 points over the 12 months to September 30 and by 15.8 points over the 12 months to October 31, because the later window swapped a weak October 2025 for a strong October 2026. Over the 12 months to September, its value loading added about 5 points (monthly sums) and its profitability tilt subtracted about 4. Over three years to October 2026, AVUV still trailed the S&P 500.
Why does DFSV now show a 1993 inception date?
On September 25, 2026, Dimensional merged the DFSV ETF into the U.S. Small Cap Value Portfolio as an ETF share class, and the class inherited the mutual fund’s record. The ETF’s own history began on February 23, 2022.
Is AVDV better than AVUV?
AVDV led over three years to October 2026 (28.3% against 20.5% a year), and most of that lead built up from October 2023 to October 2025, when AVDV returned 31.2% a year to AVUV’s 16.9% at market prices. Over the 12 months to October 31, 2026, AVUV led, 27.9% to 21.8%. They hold different markets and do different jobs in a portfolio; see AVUV vs. AVDV.
What is the value spread?
The price-to-book multiple of growth stocks divided by that of value stocks. A reading of 6 means growth stocks cost six times as much per dollar of book value. The value spread page tracks it monthly with its history since 1926.
Should I buy small-cap value after a big year?
Buy it for its long-run exposure, at a size you would hold through a decade of lagging. Neither the past year’s return nor the current spread has been a reliable guide to the next five years.
Key takeaways
- AVUV’s lead over the S&P 500 is about one year old; international value has paid since 2021. AVUV returned 27.9% in the 12 months to October 31, 2026, but 20.5% a year over three years, behind the S&P 500’s 23.4%, and since its 2019 launch it roughly matched VTI. AVDV beat its benchmark by 8.1 points a year over five years, and the developed ex-US value factor gained in every calendar year from 2021 and set a record in July 2026.
- Large companies supplied most of the U.S. premium. Over the 12 months to September 2026, French’s large value portfolio beat large growth by 22.6 points; small value beat small growth by 1.7.
- AVUV’s year came despite its profitability tilt. Over the 12 months to September 2026, loadings estimated beforehand, applied to that year’s factor returns, account for 13.5 of its 15.4 points above T-bills (monthly sums), including about −4.3 points from profitability in one of that factor’s worst years since the 1960s.
- DFSV’s record is now DFSVX’s. Since the September 25, 2026 merger, DFSV’s returns before that date are the mutual fund’s.
- Value has usually beaten growth over five years, and the spread has not said when. Small value beat small growth in 74 of 96 overlapping five-year windows starting 1926 to 2021, by 5.6% a year on average before costs, with a similar edge after wide and narrow spreads. The small-cap spread was higher than in an estimated 82% of months since 1926 as of September 2026 (79% of June readings), and a weak large-cap relationship failed out of sample.
- A 20% tilt moved portfolio drawdowns by under a point. From September 2019 to October 2026, a 20% AVUV sleeve beside VTI took the worst month-end drawdown from 24.8% to 25.4%; AVUV alone fell 42.4%.
Related guides
- Value Spread: the monthly reading behind this guide, with its history and method.
- Small-Cap Value: the long-run case and the risks.
- The Rise of Avantis: how AVUV and DFSVX compare with index funds over their full histories.
- Cheap for a Reason?: value traps and the profitability screen.
- Is Your Value ETF an AI Fund?: what value ETFs hold and how they did in the autumn 2026 AI selloff.
- The Profitability Factor: what RMW measures and how Avantis and Dimensional use it.
Sources
- Avantis U.S. Small Cap Value ETF (AVUV), fund page, performance as of Sep 30 and Oct 31, 2026. avantisinvestors.com
- Avantis International Small Cap Value ETF (AVDV), fund page, performance as of Sep 30 and Oct 31, 2026. avantisinvestors.com
- Alpha Architect U.S. Quantitative Value ETF (QVAL), fund page, performance as of Oct 31, 2026. funds.alphaarchitect.com
- State Street SPDR S&P 500 ETF Trust (SPY), fund page, index performance as of Oct 31, 2026. ssga.com
- Dimensional US Small Cap Value ETF fact sheets, Sep 30, 2026 (dimensional.com) and Dec 31, 2025 (copy hosted by IFA).
- Dimensional ETF Trust, prospectus supplement on the reorganization of the Dimensional US Small Cap Value ETF into an ETF share class of the U.S. Small Cap Value Portfolio, June 18, 2026. sec.gov
- Dimensional, “Recently Completed Mergers,” as of Oct 5, 2026. dimensional.com
- Kenneth R. French, Data Library: 6 Portfolios Formed on Size and Book-to-Market (2x3), with and without dividends; 32 Portfolios Formed on Size, Book-to-Market and Operating Profitability; Portfolios Formed on Book-to-Market, Earnings/Price and Cashflow/Price; Fama/French 3 and 5 Factors and Momentum, U.S. and Developed ex US; monthly through Sep 2026. mba.tuck.dartmouth.edu
- Yan Li, David T. Ng and Bhaskaran Swaminathan, “Is the Value Premium Dead? Forecasting Value–Growth Cycles with the Implied Value Premium,” Journal of Financial and Quantitative Analysis 61(5), 2026 (online Feb 2026). HML averages from the introduction; out-of-sample R² from Table 7. doi.org
- Robert D. Arnott, Campbell R. Harvey, Vitali Kalesnik and Juhani T. Linnainmaa, “Reports of Value’s Death May Be Greatly Exaggerated,” Financial Analysts Journal 77(1), 2021. doi.org
- Larry Swedroe, “Valuation Spreads Continue to Favor Value Stocks,” Morningstar, Oct 23, 2026. morningstar.com
- Cliff Asness, “Is (Systematic) Value Investing Dead?” AQR, May 8, 2020. aqr.com
- Clifford S. Asness, Jacques A. Friedman, Robert J. Krail and John M. Liew, “Style Timing: Value versus Growth,” Journal of Portfolio Management 26(3), 2000. doi.org
- Andrea L. Eisfeldt, Edward Kim and Dimitris Papanikolaou, “Intangible Value,” Critical Finance Review 11(2), 2022. doi.org
- Avantis Investors, estimated year-end distributions, as of Sep 30, 2026. avantisinvestors.com
- Dimensional, 2026 capital gain distribution estimates, as of Sep 30, 2026. dimensional.com
Method
Fund returns in the first table are the issuers’ NAV figures as published; the attribution covers October 2025 to September 2026, the latest 12 months in French’s data, while the headline returns run to October 31, 2026. The since-launch table and the factor regressions use Yahoo Finance adjusted closes read Nov 15, 2026, resampled to month-ends. Factor loadings: monthly returns over the one-month Treasury rate, regressed on the Fama-French five factors plus momentum with Newey-West (6 lag) standard errors; U.S. factors for U.S. funds and developed ex-US factors for AVDV. The attribution multiplies loadings estimated October 2019 to September 2025 by each factor’s summed monthly returns from October 2025 to September 2026. Value spreads use French’s June value-weighted book-to-market for each portfolio, rolled forward between Junes on returns without dividends; the forward test uses each June reading and value-minus-growth total returns over the next five or ten years, with an out-of-sample R² that fits each forecast only on outcomes already known at its date. All figures use French’s 202609 data and describe past windows; none is a forecast. The scripts and output are in summitward-research.
Author disclosure
I hold AVGV and other Avantis value funds. Summitward has no business relationship with Avantis, Dimensional, Alpha Architect, Vanguard, BlackRock or any firm mentioned here and receives no compensation from them. Nothing here is investment advice.
More in Investing & Portfolio
Browse all investing & portfolio guidesGet new guides by email
Evidence-based, no jargon. At most two emails a month. Unsubscribe any time.
Try it in Summitward
See portfolio factor analysis in action with your own financial data. Free to start, no credit card required.