“ETF vs index fund” is the kind of question where the search bar assumes there are two products, and that assumption is the error. There is one product, and the index fund and the ETF are two wrappers around it. The index inside is the same. What the wrapper changes is how you buy in, what the entry costs, and what the wrapper charges every year. That last number, the fee, is the one that compounds against you, and it is the only number this article counts.
What the index fund is, and what the ETF is
Both are index products: they hold the stocks of an index in the same proportions, and their job is to match the index, not to beat it. The wrapper is the difference.
The index fund is a mutual fund. In the United States, the line runs back to December 31, 1975, when the First Index Investment Trust, now the Vanguard 500 Index Fund, started with $11 million to track the S&P 500. You buy and sell it at a single price at the end of the trading day, the net asset value, the fund’s per-share price, calculated after the close.
The ETF does the same thing and trades on an exchange all day, like a stock. The first one listed in the United States began trading on January 29, 1993, the SPDR S&P 500 ETF Trust, the fund everybody knows as SPY.
The holdings are close enough that the difference does not move your return. A popular S&P 500 index fund tracks the index within 0.01 percent, and the 0.01 is the fee, not a difference in skill. What the wrapper actually changes is three things: the entry minimum, the price you buy at, and the annual fee.
The scale is worth one line. As of May 2026, there was $22 trillion in index funds in the United States. Vanguard alone held $13.3 trillion as of 2025, the largest mutual fund provider and the second-largest ETF provider in the world. In February 2025, Vanguard’s S&P 500 ETF, VOO, overtook SPY as the world’s largest ETF by assets. The wrapper war is real, and it is fought in basis points, a tenth of a percent.
The 2026 fee table, fund by fund
Here is the dated table. Two matched pairs, the same index offered as an index fund and as an ETF, and one legacy wrapper for scale. Every ratio is a 2026 figure from the fund’s own documents and fund data, and every minimum is the initial investment the provider requires.
| Index | Index fund (mutual fund) | 2026 fee | Minimum | ETF, same index | 2026 fee | Minimum |
|---|---|---|---|---|---|---|
| US total market (all US small, mid and large-cap stocks) | Vanguard Total Stock Market Index Fund Admiral, VTSAX (fund since 1992) | 0.04% | $3,000 | Vanguard Total Stock Market ETF, VTI (since 2001) | 0.03% | $0 |
| S&P 500 (the 500 largest US companies) | Vanguard 500 Index Fund Admiral, VFIAX (since 2000) | 0.04% | $3,000 | Vanguard S&P 500 ETF, VOO (since 2010) | 0.03% | $0 |
| S&P 500, legacy wrapper | — | — | — | SPDR S&P 500 ETF, SPY (since 1993, the first US ETF) | 0.0945% | $0 |
The sources and dates, because the date is what makes the table re-datable. The VTSAX and VTI figures are from the fund data at the September 25, 2026 close: VTSAX held $2.34 trillion across 3,514 companies, VTI held $699.67 billion, and the fund data lists VTI as VTSAX’s equivalent ETF at 0.03%, against 0.04% for VTSAX. VFIAX at 0.04% with its $3,000 minimum comes from the fund’s 2026 fact sheet, about $600 billion in assets as of August 2026. Its ETF twin, VOO, at 0.03% with no minimum beyond one share, is from the September 25, 2026 fund data, $1.04 trillion. SPY’s 0.0945% is the figure in the trust’s own document, and its September 25, 2026 assets were $817.16 billion.
Read the two matched pairs first. In both of them, the ETF is one basis point cheaper than the index fund, 0.03% against 0.04%, and the entry minimum goes from $3,000 to $0. One basis point is a tenth of one percent. It looks like a rounding error. It is not, and the next section counts it.
Then read the legacy row. SPY tracks the same S&P 500 index that VFIAX and VOO hold, and it still costs 0.0945%, three times the Vanguard ETF’s fee for the identical 500 companies. It became the first ETF to reach $500 billion in assets in February 2024 and the first past $600 billion in October 2024. In 2026 it was renamed the State Street SPDR S&P 500 ETF Trust, on January 26, which tells you how old the wrapper is without telling you how much it still costs. The first ETF in the country is still one of the most expensive ways to own the S&P 500.
The 30-year cost, counted on a fixed $10,000
The fee is a percentage, and a percentage is not a number you can feel. So here is the count. $10,000, invested once, no contributions added, thirty years. Growth is the dated set this site runs on: 6 percent real, after inflation, the long-run record of the US equity market since 1926, as the center case, and 3 percent real as the planning floor, the lower end of the range. The fee is deducted from growth every year, because that is how a fee works. It does not leave a line on a statement. It stays in the fund, and it compounds against you.
| 2026 wrapper (fee) | $10,000 after 30 years at 6% real | 30-year fee cost at 6% real | $10,000 after 30 years at 3% real | 30-year fee cost at 3% real |
|---|---|---|---|---|
| VTI and VOO (0.03%) | $56,949 | $486 | $24,061 | $211 |
| VTSAX and VFIAX (0.04%) | $56,788 | $647 | $23,991 | $281 |
| SPY (0.0945%) | $55,918 | $1,516 | $23,613 | $659 |
For the baseline, the same $10,000 at a zero-fee wrapper reaches $57,435 at 6 percent real and $24,273 at 3 percent. The fee cost is the gap between those two numbers.
The counted deltas, the ones that answer the question the search bar actually typed:
- VTI versus VTSAX, the ETF’s one basis point on the total market, is $161 on $10,000 over thirty years at 6 percent real. At the 3 percent floor, $70.
- VOO versus SPY, the legacy wrapper on the S&P 500, is $1,031 at 6 percent real and $448 at the floor. The same index, the same 500 companies, and a thousand dollars of wrapper.
- Year by year, the gap between 0.03% and 0.04% is $10 on $10,000. The $161 is what thirty years of that $10 becomes.
A client from my branch-office years asked me the same question in 2004, with a fund at 1.5% and an index fund at 0.04%, and the difference she lost over twenty years was the fee, not the market. The count above is the same arithmetic at a smaller scale: 2026’s wrappers are cheaper, and the count says how much cheaper is worth.
The point of the count is also its limit. On a fixed $10,000, the wrapper fee is a small number at 2026’s prices, and that is true. The fee compounds the same way on every dollar you add, so the $10,000 count is the floor of the cost, not its shape. A $100,000 account at the same fee pays ten times the fee, and the thirty-year gap scales with it.
Where the ETF wrapper actually costs more than the index fund
Here is the question the table alone does not answer, and the honest answer from the 2026 numbers: in the annual fee, it does not. On both matched pairs, the ETF is one basis point cheaper or equal. Where the ETF does cost more is in the per-trade layer, and there are three of them.
The commission. If your broker charges per trade, every ETF purchase and sale carries it. The mutual fund order is executed once, at the end-of-day price, without a trade price. The commission is a per-trade number, so it never appears in the annual fee at all, which is why a fee table looks closer than the real cost for somebody who trades.
The market price. The ETF trades all day at the market’s price, which means you buy at whatever the market is paying at that minute, spread included. The index fund buys at the end-of-day NAV, the fund’s actual per-share value, calculated after the close. The buyer who clicks at the top of a volatile afternoon pays a price the index fund’s buyer never sees. The founder of the world’s largest mutual fund company was skeptical of the wrapper for exactly this reason. He thought anything that could be traded mid-day would be traded mid-day, and that the trading would eat the return.
The legacy wrapper. SPY is the standing example. 0.0945% for the same index that VOO holds at 0.03% is $64.50 a year per $10,000, about $1,031 over thirty years on a fixed stake. If the ETF you own costs more than the index fund for the same index, you have your answer: the index is fine, the wrapper is not.
And the reverse direction, so the count stays honest. The index fund’s cost is the gate. Both of the Vanguard admiral funds above require $3,000 to start, and Vanguard’s ETF fractional program, in place since February 2021 from $1, means that below the gate the ETF is the only way in. The $3,000 minimum is the price the mutual fund wrapper charges the beginner who has less than $3,000. For a beginner with $500, the index fund is not cheaper. It is unavailable.
“Why index funds are bad”
The question comes up, so here is the counted answer. “Index funds are bad” is usually one of two claims.
The first is that they cannot dodge a down market. True, they track the index, so a bad year is a bad year, the same in both wrappers. That is the price of the boring default, not a flaw. The protection from a crash is the same number, and it is why the index is the default in the first place. The wrapper changes nothing about it.
The second is that the fee is too high for what the product is. That one is checkable, and it is the one that matters. The 2026 fact sheets for the big US indexes read 0.03% to 0.04%, ETF and index fund alike, and 0.0945% for the legacy 1993 wrapper. If the “index fund” you own charges 0.5%, 1% or more, it is charging an active fund’s fee for an index product, and that is the real version of bad. The index does the job. The wrapper sets the price.
One aside, because the 2026 fact sheets carry it. The S&P 500’s ten largest holdings, per the fund’s May 2026 fact sheet, are the technology and megacap names, with information technology at 32.9 percent of the fund. That is the index’s composition, not a wrapper choice, and every wrapper that holds the index carries it, ETF and index fund the same.
Which one, with the dated numbers
The boring default, stated with the numbers:
- Starting with less than $3,000: the ETF. Fractional, from $1, at 0.03% a year for the S&P 500 or the total market. The wrapper’s real cost is one basis point, and there is no gate.
- Already holding VFIAX or VTSAX in a 401(k) or IRA: the gap to the ETF twin is one basis point, $10 a year per $10,000, $161 over thirty years on a fixed stake at 6 percent real. Do not churn a tax-advantaged account over a rounding error. The fee is already as low as the index gets.
- Opening a taxable account today: the 0.03% ETF wrapper. The structure can shed low-basis positions through in-kind redemptions, swapping shares out without selling them, which the mutual fund cannot do, and that makes it the more tax-efficient of the two for the same index.
- Owning a wrapper that costs more for the same index: SPY at 0.0945% is the standing example. It is a fine fund. It is just a 1993 price tag, and the 2026 count says the price tag is a third of that.
You do not need a clever answer here. You need the boring one, dated: the same index, the cheaper of the two wrappers, the $0 entry if you are starting small, and the fee, 0.03% in 2026, as the number that compounds against you from day one.
The sequence, which is the whole point of the order: the spend plan first, the emergency fund next, sized and counted here, and then the index fund, in that order. And when the one fund becomes a portfolio, the three-fund build, costed the same way fund by fund, is the next article in this site’s tree: the 3-fund portfolio, built and costed in 2026. Start with the fee, not the ticker.