Skip to content
Menu
Finabase Learn

Strategies · Long-term investing

Small-cap investing and the size effect: a 26-year test

The size effect as rules: what a small-cap index holds, and IWM against SPY from 2000 to 2026, with a strong first half, a weak second and more volatility throughout.

On this page
  1. The idea
  2. The rules
  3. What the backtest shows
  4. Where it fails: the losing periods
  5. Compared with its neighbours
  6. How Finabase does it
  7. Frequently asked questions
  8. Summary
  9. Check your understanding

The size effect is the observation that small companies' stocks have, over long periods, returned more than large companies' stocks. Rolf Banz reported it in 1981, it became one of the three factors in the Fama and French model, and it is the reason many long-term portfolios hold a separate small-cap fund.

It is also the factor whose evidence has aged worst. This lesson states the rules a small-cap index follows and tests the most heavily traded small-cap fund, IWM, against SPY from June 2000 to October 2026. On price alone the two ended almost exactly level. The road there was a strong first half, a weak second half and about a third more volatility the whole way.

The idea

A small-cap stock is one whose company has a low market value, typically from a few hundred million dollars to a few billion. Why would such stocks earn more?

The risk explanation says they should. Small companies fail more often, borrow on worse terms, depend on fewer products and customers, and are harder to sell in a hurry. Investors demand a higher expected return for holding them. The neglect explanation says small companies are followed by few analysts and owned by few institutions, so their prices are more often wrong, and wrong in a direction that rewards buyers.

Banz found that the smallest New York Stock Exchange companies had out-earned larger ones from 1936 to 1975, even after adjusting for beta. The doubts began soon after. The effect was concentrated in the very smallest companies, the kind a large fund cannot buy in size. Much of it arrived in January. And it weakened after it was published. In 2018 Asness and co-authors argued that the premium is still there once unprofitable, low-quality small companies are set aside, which turns out to matter in the test below.

The rules

These are the rules of the Russell 2000 Index, which IWM tracks, in outline.

  1. Rank every US-listed company by market value.
  2. Set aside the largest 1,000.
  3. Hold the next 2,000, each weighted by its market value.
  4. Rebuild the list at each scheduled reconstitution: once a year in June until 2025, and in June and December from 2026. Companies that have grown too large move up and out; companies that have shrunk enough move in from above.
  5. Reinvest dividends.

A person holding the fund follows one rule: buy it and keep it.

Notice that rule 3 has no test of the business. If a company is the right size, it is in. A large minority of Russell 2000 companies report losses, and the index holds them all.

The S&P SmallCap 600, which the neighbour fund IJR tracks, adds a rule: to be added, a company must have positive reported earnings over its latest quarter and over the latest four quarters together. That one extra line makes it a different strategy.

Rule 4 also has a cost that does not appear in any fee. Everyone knows in advance which stocks will enter and leave on the reconstitution date, and traders buy and sell ahead of the index funds. A small-cap fund also loses its winners by design: a company that grows large enough is handed to the large-cap index, and the gains after that belong to someone else.

What the backtest shows

The test, stated in full so it can be checked:

  • Data. Month-end closes of IWM and SPY from June 2000, IWM's first full month, to October 2026: 317 months.
  • Trading. Each fund is bought once and held.
  • Costs. One purchase each. The funds' costs are inside their prices: IWM charges 0.19% a year, IJR 0.06% and SPY 0.0945% (the issuers' fund pages, November 2026). The cost of trading at each reconstitution is inside the fund's price as well. Small stocks have wider spreads than large ones, so buying them directly costs more than buying the fund.
  • Left out. Dividends on both sides, and taxes. SPY has usually yielded a little more than IWM, so the omission flatters small caps by a few tenths of a point a year.
IWM against SPY, monthly, 2000–2026: price only, no dividends $0.5 $1 $2 $5 Growth of $1 SPY (the market) IWM (Russell 2000) 0% −20% −40% −60% Fall from the peak 2005 2010 2015 2020 2025
IWM (Russell 2000) against SPY, month-end prices, 2000 to 2026. Price only: dividends are not included for either fund. Top: growth of $1 (log scale). Bottom: the fall from the highest value so far.
Measure, price only IWM SPY
Total return 440.5% 420.2%
Return per year (CAGR) 6.6% 6.5%
Deepest fall from a peak −53.5% −52.2%
Longest time below a peak 46 months 80 months
Volatility, annualized 19.9% 15.1%
Beta to SPY 1.15 1.00
Return per unit of volatility 0.33 0.43
Value of $1 at the end $5.40 $5.20

One dollar became $5.40 in small caps and $5.20 in large. After the dividend difference, that is a tie. What was not tied is the risk. Small caps were about a third more volatile. In the 122 months when SPY's price fell, it fell 3.6% on average and IWM 4.2%. In the months when SPY rose, it rose 3.3% and IWM 3.8%. Small caps behaved like the market with the volume turned up, and over 26 years the extra movement bought no extra return.

Two halves

The period is split at its midpoint, August 2013, so the split is not chosen to flatter either side.

Period IWM, per year SPY, per year Difference
June 2000 – August 2013 5.2% 0.9% +4.3
August 2013 – October 2026 8.1% 12.3% −4.2
IWM price divided by SPY price, month-end, 2000–2026 (start = 1) 1.00 1.25 1.50 1.75 IWM ÷ SPY 2005 2010 2015 2020 2025
IWM’s price divided by SPY’s price, month-end, 2000 to 2026 (start = 1). Small caps gained 79% on large caps by March 2011 and gave almost all of it back by 2025.

The first half looks like the textbook. The test opens just after the dot-com peak, when the largest companies were the most overpriced. From August 2000 to September 2002 SPY's price fell 46.3% and IWM's 33.4%, and small caps then led the recovery. By March 2011 a dollar in IWM was worth 1.785 times a dollar in SPY.

The second half reversed nearly all of it. The ratio fell 45.3% from that high to July 2025. The largest companies in the market grew faster than the small ones, which is the opposite of what the size effect assumes. Over every five-year window in the data, IWM's price was ahead of SPY's in 130 of 257, or 51%: a coin toss.

Other small-cap funds

Fund Index and rule From Fund, per year SPY, same months Deepest fall
IWM Russell 2000: size only June 2000 6.6% 6.5% −53.5%
IJR S&P SmallCap 600: size plus positive earnings June 2000 8.2% 6.5% −52.8%
VB CRSP US Small Cap: size only, somewhat larger companies February 2004 8.1% 8.7% −55.4%
VBR CRSP US Small Cap Value: small and cheap February 2004 7.2% 8.7% −58.3%

IJR is the row to look at. Over the same 317 months, the index that refuses unprofitable companies returned 8.2% a year against 6.6% for the one that takes them all. That 1.6-point gap is larger than any difference between small and large in the main table, and it agrees with the argument of Asness and co-authors that size works better with the junk removed. IJR still followed the same two-half pattern: 7.8% a year against 0.9% for SPY to August 2013, then 8.6% against 12.3%.

VB and VBR start in 2004 and so miss the years when small caps did best. Over their own histories both trailed SPY's price, and IWM returned 7.2% a year over those same months. Adding a value screen (VBR) did not help in this period.

Where it fails: the losing periods

Since 2013. The second half of the test is one long losing period against large companies: 8.1% a year against 12.3%. In the twelve full calendar years from 2014 to 2025 IWM's price finished ahead of SPY's twice, in 2016 and 2020.

Year IWM SPY
2014 3.7% 11.3%
2018 −12.2% −6.3%
2021 13.5% 27.0%
2023 15.1% 24.3%
2024 10.1% 23.3%
2025 11.4% 16.4%

Crashes. Small companies have less cash and less access to credit, and it shows when credit tightens. From December 2019 to March 2020 SPY's price fell 19.9% and IWM's 30.9%. In the financial crisis both fell by about half (52.4% for IWM from October 2007 to February 2009, 52.2% for SPY). Since 2013 IWM's deepest fall from a peak has been 33.8%, against 24.8% for SPY.

The rebounds are real, and they are the other half of the same thing. In the twelve months after the March 2020 low IWM rose 93.0% and SPY 53.8%; after February 2009, 60.4% and 49.8%. A holder who sold small caps during the fall kept the larger loss and missed the larger recovery.

Other ways it fails or misleads:

  • Higher interest rates. Small companies carry more floating-rate and short-term debt than large ones. When rates rose from 2022, their interest costs rose faster, which is one reason they lagged in 2023 and 2024.
  • The index is not the research. Banz's result came from the very smallest listed companies. The Russell 2000's holdings are weighted by size, so most of the money is in its largest members, companies worth several billion dollars.
  • Winners leave. A company that succeeds graduates to the large-cap index. The small-cap fund sells it and replaces it with something smaller.
  • The start date. Beginning in June 2000 hands small caps their best decade first. A test starting in 2011 would show a strategy that lost almost half its value against the market.

Compared with its neighbours

Strategy Selects on Price result against SPY in its test Risk against SPY
Small cap (IWM) Small size only +0.1 points a year, 2000 to 2026 Volatility a third higher
Small cap, profitable (IJR) Small size, positive earnings +1.7 points a year, 2000 to 2026 Similar deepest fall to IWM
Value (IWD) Low price to book −0.4 points a year, 2000 to 2026 Deeper fall in 2007 to 2009
Low volatility (USMV) Calm prices −3.1 points a year, 2011 to 2026 Volatility a fifth lower

How Finabase does it

Finabase does not classify stocks by size style and no Finabase tool says what to buy.

  • Screener. The screener filters on market capitalization and can be limited to an index, including the S&P SmallCap 600 (the link above) and the S&P MidCap 400. Within a scope it can keep companies with a positive P/E, which is a rough version of the earnings rule that separates IJR from IWM.
  • Stock pages. Each stock's page shows its market capitalization and beta.
  • Heatmap. The heatmap sizes each company by market value, which makes plain how little of the market small companies are.
  • Comparison. The comparison chart linked above draws IWM against SPY as percentage change, on prices, like this page.
  • Backtest and Strategy. These tools replay indicator signals and if/then rules on one symbol, buying a fixed $10,000 per signal with no trading cost; Strategy covers two years. They cannot hold an index of 2,000 stocks, and two years is one small piece of either half of this test.

Frequently asked questions

Do small-cap stocks outperform large caps?

Not reliably. In this test IWM's price returned 6.6% a year from 2000 to 2026 and SPY's 6.5%. Small caps led by 4.3 points a year until August 2013 and trailed by 4.2 points a year afterwards, with about a third more volatility throughout.

Are small caps riskier than large caps?

Yes, on every measure here. IWM's volatility was 19.9% a year against 15.1%, its beta was 1.15, and it fell 30.9% in the 2020 crash against 19.9% for SPY.

What is the difference between the Russell 2000 and the S&P 600?

The Russell 2000 holds the 2,000 companies ranked below the largest 1,000, whatever their profits. The S&P SmallCap 600 requires positive earnings before a company is added. Over the same 26 years the S&P 600 fund IJR returned 8.2% a year and the Russell 2000 fund 6.6%.

Why have small caps lagged since 2013?

The largest companies grew their profits faster than small ones, which is unusual. Small companies also carry more debt that reprices when interest rates rise, and many of the most successful young companies now stay private longer and reach the stock market already large.

Summary

Small-cap investing holds the companies ranked below the largest 1,000, weighted by size, with no test of the business. Tested through IWM from June 2000 to October 2026, on prices without dividends, it returned 6.6% a year against 6.5% for SPY: level, with a third more volatility and deeper falls in 2020 and 2022. The first half of the period showed a size premium of more than four points a year and the second half reversed it. The clearest finding is among the small-cap funds themselves: the index that requires profits beat the one that does not by 1.6 points a year. The test starts at a moment that favored small companies and covers funds of mid-sized "small" companies, so it neither confirms the original size effect nor rules it out.

Check your understanding

4 questions on this lesson. Each answer comes with the reason, and nothing is scored.

  1. IWM and SPY ended 26 years with almost the same price return. Why is that not a neutral result for small caps?

    1. Because small caps pay higher dividends
    2. Because small caps carried about a third more volatility and deeper falls to get the same return
    3. Because the test used the wrong index
    Show the answer

    B. Because small caps carried about a third more volatility and deeper falls to get the same return The size effect is supposed to be extra return for extra risk. Here the extra risk was present (volatility 19.9% against 15.1%, beta 1.15) and the extra return was not.

  2. The test is split at its midpoint, August 2013, and not at the month small caps peaked against large caps. Why?

    1. A midpoint is chosen before looking at the result, so it does not exaggerate the contrast
    2. Because August 2013 was a market low
    3. Because the index changed its rules that month
    Show the answer

    A. A midpoint is chosen before looking at the result, so it does not exaggerate the contrast Splitting at the relative high would produce the largest possible "before and after" by construction. A split fixed in advance still shows a +4.3 and −4.2 point difference, which is the more credible evidence.

  3. IJR returned 8.2% a year and IWM 6.6% over the same months. What rule separates their indexes?

    1. IJR holds larger companies only
    2. IJR’s index requires positive earnings before a company is added; IWM’s takes every company of the right size
    3. IJR rebalances monthly
    Show the answer

    B. IJR’s index requires positive earnings before a company is added; IWM’s takes every company of the right size The Russell 2000 has no test of the business, so it holds a large minority of loss-making companies. The S&P 600’s earnings requirement screens those out, and that one rule was worth more than the whole small-against-large difference.

  4. From December 2019 to March 2020 IWM fell 30.9% and SPY 19.9%; over the next twelve months IWM rose 93.0% and SPY 53.8%. What does that pattern show?

    1. Small caps are safer because they recover faster
    2. Small caps moved like the market with the swings enlarged, in both directions
    3. Small caps should be bought only after crashes
    Show the answer

    B. Small caps moved like the market with the swings enlarged, in both directions A beta above 1 means larger moves both ways. The bigger rebound is the same property as the bigger fall, and someone who sold during the fall kept the loss and missed the recovery.

Sources and further reading

  1. Rolf W. Banz, “The Relationship Between Return and Market Value of Common Stocks,” Journal of Financial Economics 9(1), 1981
  2. Eugene F. Fama and Kenneth R. French, “Common Risk Factors in the Returns on Stocks and Bonds,” Journal of Financial Economics 33(1), 1993
  3. Clifford Asness, Andrea Frazzini, Ronen Israel, Tobias Moskowitz and Lasse H. Pedersen, “Size Matters, If You Control Your Junk,” Journal of Financial Economics 129(3), 2018
  4. FTSE Russell — Russell U.S. Equity Indexes: construction and methodology
  5. FTSE Russell — Russell US Indexes move to semi-annual reconstitution (announcement, January 16, 2025)
  6. S&P Dow Jones Indices — S&P U.S. Indices methodology (eligibility, including the earnings requirement)
  7. iShares — Russell 2000 ETF (IWM) and Core S&P Small-Cap ETF (IJR) fund pages: expense ratios and yields
  8. State Street — SPDR S&P 500 ETF Trust (SPY), fund page: gross expense ratio 0.0945%
  9. Finabase chart data — IWM, IJR, VB, VBR and SPY monthly bars to October 2026; results computed by Finabase