Why Equal-Weight Indices Beat Cap-Weight Over the Long Run — and When They Do Not

Equal-weight sounds boring. It is not. In the S&P 500 Equal Weight Index, every constituent starts at roughly 0.2% and is reset back to that weight each quarter, while the cap-weighted S&P 500 lets the biggest winners keep growing until they dominate the index ^[5]. That difference has mattered a lot. From 1990 through 2024, equal-weight U.S. large-cap portfolios often outpaced cap-weighted benchmarks, but the source of that outperformance was not magic. It came from three places: a persistent size tilt, a small rebalance premium, and a built-in anti-momentum drag that sometimes helped and sometimes hurt ^[1][2][3].

The uncomfortable part is that equal-weight is not a free lunch. It can lag for years when the market is rewarding the same giant stocks over and over, as it did in 1998–1999 and 2020–2021 ^[3][4]. If you want the extra return, you are implicitly accepting more turnover, more exposure to smaller names, and a willingness to sell strength before the crowd does. If you want the benchmark with the least drama, cap-weight still wins on simplicity and usually on trading efficiency. For a broader framing of that tradeoff, see active vs passive investing and what an index fund actually does.

The edge is real, but it is not one thing

Equal-weight outperformance is often described as if it were a single factor. That is sloppy. The better way to think about it is as a bundle of exposures and trading rules. Research Affiliates has long argued that equal-weighting captures a size tilt because it gives the same capital to a $20 billion company and a $500 billion company, which mechanically overweights the smaller names relative to cap-weighting ^[1]. DFA’s long-run factor work points in the same direction: smaller stocks have historically earned a return premium, though the premium is unstable and can disappear for long stretches ^[7].

Then there is the rebalance effect. Equal-weight indices must sell stocks that have run up and buy stocks that have fallen behind. That is a contrarian rule, and contrarian rules can harvest a premium when prices mean-revert. Plyakha, Uppal, and Vilkov (2014) found that equal-weighting’s excess return can be decomposed into a rebalancing component and a size component, with the rebalance piece tied to cross-sectional mean reversion rather than some mystical “rebalancing bonus” ^[2]. The catch is obvious once you say it plainly: if momentum is working, equal-weight is fighting the tape.

That is why the right comparison is not “equal-weight versus cap-weight” in the abstract. It is “which market regime is paying for size and mean reversion right now?” If you want a deeper primer on the factor side of that question, AIBROKER’s factor investing overview and momentum premium explainer are the right companions.

That table is the whole story in miniature. Equal-weight is not “better.” It is a different bet.

Why the size tilt did most of the heavy lifting from 1990 to 2024

Research Affiliates’ work on equal-weighting argues that much of the long-run excess return comes from the embedded size exposure, not from some magical portfolio construction trick ^[1]. That matters because size is a real factor, but it is not a stable one. DFA’s research on small-cap premiums shows that small stocks have outperformed over very long horizons, yet the premium is lumpy and can be negative for extended periods ^[7]. Equal-weight inherits that lumpiness. It does not diversify it away.

From 1990 through 2024, the S&P 500 Equal Weight Index and similar equal-weight large-cap portfolios generally posted higher long-run returns than the cap-weighted S&P 500, but they also carried a different risk profile: more turnover, more exposure to mid- and smaller-cap names, and a less concentrated return stream ^[5][8]. That concentration difference is not cosmetic. In cap-weighted indices, the top ten names can dominate a large share of performance in a given year. In equal-weight, no single stock can hijack the result. That is a feature if you dislike concentration. It is a bug if you are trying to ride the market’s biggest winners.

The uncomfortable implication is that equal-weight can look brilliant in hindsight precisely because the market’s leadership was broad enough to reward smaller names. When leadership narrows, the same structure becomes a headwind. Investors who buy RSP or EQAL because they think they are getting “the market, but smarter” are usually making a category error. They are buying a factor bundle with a size tilt. That is fine. Just call it what it is.

If you are comparing ETFs, the fund wrapper matters too. Expense ratios, spreads, and turnover all eat into the theoretical edge. A useful companion piece is understanding fees, expense ratios, commissions, and the costs you don’t see.

The rebalance premium is real, but it is smaller than the sales pitch

The rebalance premium is the part of equal-weighting that sounds too neat to be true, because it often is oversold. Plyakha, Uppal, and Vilkov (2014) showed that equal-weighting can earn a rebalancing return when prices mean-revert across constituents, and that this effect is distinct from the size effect ^[2]. But the premium is not a guaranteed gift from the market. It depends on dispersion, volatility, and the degree of reversal after relative price moves. When winners keep winning, the premium turns into a tax on trend-following.

Here is a simple worked example. Suppose a stock starts the quarter at 1.00% of an equal-weight portfolio. By quarter-end, it has risen to 1.40% because it outperformed the rest of the basket. At the rebalance date, the index trims it back to 1.00%. That means the portfolio sells 0.40 percentage points of a recent winner. If the stock then mean-reverts and falls back toward the pack, the rebalance helped. If it keeps rising, the rebalance hurt. That is the whole game. No mystery. No free money.

Now scale that up. In a 500-stock index, the aggregate amount sold from winners and bought into laggards can be meaningful, especially after a quarter with wide dispersion. But the premium is still modest relative to the market’s total return. It is best thought of as a small edge that compounds over time, not as a source of double-digit annual alpha. Investors who expect more are usually confusing a structural tilt with a forecasting machine.

Assumptions: illustrative one-stock example; quarterly rebalance; ignores transaction costs, bid-ask spread, and taxes. It is not audited performance.

For readers who want the mechanics of why prices move enough for this to matter, AIBROKER’s how stock prices are set and transaction costs and slippage explain the plumbing.

Equal-weight does not harvest a premium because it rebalances. It harvests a premium only when the market’s losers later stop losing and the winners stop winning.

1998–1999 and 2020–2021 were the wrong years to own the rebalance trade

Equal-weight’s worst stretches are not random. They cluster in momentum-led regimes. In 1998–1999, the late-1990s growth and technology boom pushed a narrow set of large-cap winners far ahead of the rest of the market. Equal-weight, which kept trimming those winners, lagged badly ^[3][4]. The same pattern repeated in 2020–2021, when the pandemic recovery and the rise of mega-cap technology created another winner-take-most market ^[3][5].

This is where the anti-momentum drag becomes visible. Equal-weight is structurally late to the party. It sells the names that are already working and buys the names that have not yet caught up. That can be a virtue in a choppy market. It is a liability when the market is rewarding persistent trends. If you want a deeper dive into that dynamic, the logic is the same one discussed in momentum versus value investing and mean reversion versus trend following.

Most investors miss the timing risk because they look only at long-run averages. That is a mistake. A strategy can have a positive long-run premium and still be miserable for five-year stretches. Equal-weight is a good example. If you bought it in 1998 or 2020 and expected immediate vindication, you got a lesson in humility instead. The strategy was not broken. The regime was hostile.

S&P Dow Jones’ quarterly reset is the whole point — and the whole cost

The S&P 500 Equal Weight Index is not equal-weight by accident. S&P Dow Jones Indices rebalances it quarterly and reconstitutes it to keep weights near equal, which is what creates the contrarian trade in the first place ^[5]. That design choice is the source of both the edge and the drag. Without the quarterly reset, the portfolio would drift toward cap-weighting over time and lose the equal-weight effect. With the reset, it keeps selling strength and buying weakness.

That trade has costs. Turnover is higher than in cap-weighted indices, and higher turnover means more trading costs, wider implementation slippage in less liquid names, and more taxable distributions in taxable accounts ^[5][9]. The costs are not always huge, but they are real. Investors who compare only headline expense ratios are missing half the bill. If you want a framework for that, AIBROKER’s rebalancing guide and tax-aware rebalancing article are worth reading before you trade.

There is also a subtle implementation issue. Equal-weight ETFs such as RSP and EQAL do not own the index in a vacuum; they trade in the real market, where spreads, market impact, and creation/redemption mechanics matter. For liquid mega-caps, that friction is small. For smaller index constituents, it is not zero. The more the strategy leans into rebalancing, the more it depends on efficient execution. That is why the same idea can look cleaner on paper than in a brokerage account.

For ETF mechanics, see ETFs vs mutual funds and bid-ask spread.

Equal-weight’s quarterly rebalance is not a side effect. It is the strategy. If you remove the rebalance, you remove most of the reason to own it.

A simple decision tree for RSP, EQAL, or plain S&P 500

Investors usually ask the wrong question. They ask which ETF is “better.” Better for what? Equal-weight is not a universal upgrade. It is a specific bet on breadth, mean reversion, and a less concentrated return stream. Cap-weight is a bet on letting winners run and minimizing trading friction. Both are rational. Neither is free.

Use this decision tree instead:

1. If you want the lowest-maintenance core holding and the market’s natural concentration, choose cap-weight.
2. If you believe mega-cap leadership is stretched, or you want less single-stock concentration, equal-weight deserves a look.
3. If you are investing in a taxable account and hate turnover, cap-weight usually has the cleaner after-tax profile.
4. If you are already tilted to small caps elsewhere, equal-weight may duplicate exposure you already own.

That last point matters more than most people think. Equal-weight is often sold as diversification, but diversification is not just “more names.” It is exposure to different return drivers. If your portfolio already has a small-cap sleeve, adding equal-weight may simply stack another size tilt on top of the one you already own. A better way to think about portfolio construction is in terms of the three numbers that matter: expected return, volatility, and correlation.

Here is a compact checklist you can use before buying RSP, EQAL, or a similar fund:

• Do I already own a small-cap fund?
• Am I in a taxable account?
• Can I tolerate multi-year underperformance versus the S&P 500?
• Do I understand that I am buying a rebalance rule, not just “more diversification”?

If the answer to the first three is yes, equal-weight may be redundant or costly. If the answer to the last one is no, you probably do not want the strategy yet.

What the 1990–2024 evidence says when you separate the pieces

The cleanest way to read the evidence is to stop treating equal-weight as a single return stream. Research Affiliates’ decomposition work, Plyakha-Uppal-Vilkov’s rebalancing analysis, and S&P’s methodology all point to the same conclusion: the long-run edge is a combination of size exposure and disciplined rebalancing, offset by a momentum headwind ^[1][2][5]. That is why equal-weight can beat over decades and still look foolish in specific regimes.

For investors, the practical implication is not “buy equal-weight.” It is “know what you are buying and when it is likely to disappoint.” If you are benchmarking a portfolio against the S&P 500, equal-weight is a legitimate alternative benchmark only if you are willing to accept a different factor mix. If you are using it as a tactical trade, you need a view on market breadth and momentum. Without that, you are just hoping the last decade repeats.

That is a weak plan. The market rarely rewards weak plans.

For readers who want to test whether a strategy is genuinely robust, AIBROKER’s backtest checklist and regime detection primer are the right next stops. Equal-weight is a good case study in why regime awareness matters: the same rule can be a winner in one decade and a laggard in the next.