ETF vs. Mutual Fund: Same Basket, Different Plumbing

12 min read

Two funds can own the same 500 stocks, in the same proportions, and still be different products in every way that touches the holder. One is a mutual fund. The other is an exchange-traded fund, an ETF. The portfolio is identical. What differs is the plumbing: how a share is created, when it is priced, who you trade with, what it costs to run, and how the tax consequences of other people’s decisions reach your account. Most explanations start with fees and stop there, which gets the order wrong. Fees matter, and the arithmetic below shows how much, but the fee is a consequence of the plumbing rather than a substitute for understanding it. In the most common comparison investors actually face — an index fund in either wrapper — the cheaper product is not always the one the shorthand predicts.

The scale is easy to understate. By year-end 2024, index mutual funds and index ETFs together held $16.3 trillion and accounted for 51 percent of assets in long-term funds, up from 19 percent at year-end 2010, according to the Investment Company Institute. ETF net assets alone had grown from $151 billion in 2003 to $10.3 trillion in 2024. The mutual fund is the older and still larger structure, the template on which American retirement saving was built. The ETF is the faster-growing challenger, and the growth has come at the level of the wrapper rather than the strategy, because the dominant strategy inside both is now the same one: owning an index, a construction Market Briefer’s explainer on what a stock index is sets out in full.

Same basket, different plumbing

A mutual fund deals directly with its own investors. You send money to the fund, the fund issues you shares, and when you leave, the fund buys those shares back. Every transaction happens at one price, the net asset value — the value assigned to the fund’s assets, minus its liabilities, divided by the shares outstanding, typically calculated at the end of each business day. The Securities and Exchange Commission’s investor bulletin states the consequence plainly: an investor who places a mutual fund order during the day does not know the exact price until that day’s calculation is done. Orders are filled forward, at the next computed value, which is why a sharp afternoon move changes the price of an order placed calmly in the morning. There is no spread to cross and no intraday price to chase. There is also no way to insist on a price, place a limit order, or exit at 11 a.m. because the news has changed.

An ETF splits that transaction into two markets. Retail investors buy and sell ETF shares from each other on a stock exchange, throughout the session, at prices that move continuously and can sit slightly above or below the fund’s own net asset value — a premium in the first case, a discount in the second. The fund itself deals only with large institutions called authorised participants, which typically create new shares by delivering a basket of securities to the fund, and redeem shares by taking the basket back. That traffic keeps the exchange price near the value of the holdings, because an institution that can assemble the basket cheaply and sell the resulting shares at a premium, or do the reverse at a discount, will keep doing so until the gap closes. What the retail holder experiences is a fund that trades like a stock, with a stock’s conveniences and frictions: a bid-ask spread on every trade, a price that can drift from value when markets are disorderly, and the ability to act at any minute of the trading day.

History explains why the second structure exists. The first U.S.-listed ETF, State Street’s SPDR S&P 500 ETF Trust, trading as SPY, was launched on January 22, 1993, and State Street marks it as a product born out of the stock market crash of 1987, when investors holding diversified portfolios through conventional funds had no way to trade the whole portfolio, at a visible price, in the middle of a falling session. SPY held more than $355 billion at its thirtieth anniversary in January 2023 and was, by State Street’s account, the most heavily traded security in the world. The mutual fund answered an older problem, how to pool small savings into a managed portfolio at a fair end-of-day price. The ETF answered a newer one, how to make that portfolio tradable. Neither answer cancels the other, which is why retirement plans still run overwhelmingly on the older structure: they were built around end-of-day pricing, purchases of exact dollar amounts including fractions, and automatic contributions. Mutual funds have long supported exact-dollar purchases and automatic contributions natively. ETFs can now offer those conveniences too where the broker supports recurring purchases and fractional shares — Fidelity’s recurring investments, for example, begin at $1 for stocks and ETFs as well as mutual funds. Where the broker does not support them, whole-share ETF purchases can still leave some cash uninvested.

What the wrapper costs — and what it does not decide

Every fund in either wrapper charges an expense ratio, an annual percentage taken from fund assets before returns reach the holder, so it never arrives as a bill. The industry’s own averages, compiled by the Investment Company Institute on an asset-weighted basis for 2024, show both the genuine gap and the trap inside it. Investors in equity mutual funds overall paid an average of 0.40 percent. Investors in index equity ETFs paid an average of 0.14 percent. Investors in index equity mutual funds paid an average of 0.05 percent — less than the index ETF average, not more. The same pattern held in bonds, where index bond mutual funds averaged 0.05 percent against 0.10 percent for index bond ETFs. The wrapper alone does not determine the fee. Both mutual funds and ETFs can follow active or passive strategies, and what the ICI figures actually show is narrower than the usual slogan: investors in index mutual funds paid lower asset-weighted average expense ratios in 2024 than investors in index ETFs. A category average across thousands of products does not establish which wrapper offers the cheapest fund for a particular benchmark and account — that takes comparing the specific products available. Scale is part of the story: the largest index mutual funds are so large that their cost per dollar runs below that of the average index ETF, most of which are far smaller products.

A worked example shows what those differences are worth, and the figures can be checked on Market Briefer’s compound interest calculator. Take $100,000 invested for 30 years in portfolios that each earn 7 percent a year before expenses, an assumption chosen for arithmetic rather than as a forecast — an illustrative annual-return model that excludes taxes, trading costs, contributions and withdrawals. At the all-equity-mutual-fund average of 0.40 percent, the net return is 6.60 percent, and $100,000 grows to about $680,300. At the index equity ETF average of 0.14 percent, the net return is 6.86 percent, and the same sum grows to about $731,900. At the index equity mutual fund average of 0.05 percent, the net return is 6.95 percent, and the sum grows to about $750,600. The annualized expense on a constant $100,000 balance is $400, $140 and $50 respectively — figures small enough to ignore, which is exactly the error. Fund expenses are not billed; they are deducted from assets as their value changes through the year, quietly reducing what the holder keeps. Compounded over 30 years, the gap between the 0.40 percent fund and the 0.05 percent fund is about $70,300, and the gap between the average index ETF and the average index mutual fund runs about $18,700 in the mutual fund’s favour. An investor who chose the ETF on the blanket assumption that ETFs are the cheap option, and paid 0.14 percent where a 0.05 percent mutual fund tracking a comparable index was available in the same account, would have bought the more expensive product while believing the opposite.

The expense ratio is not the whole cost of either wrapper. The ETF holder pays the bid-ask spread on entry and exit, trivial in a large index fund and material in a small one, and pays any premium over net asset value in the purchase price. The mutual fund holder pays no spread and always transacts at net asset value, but may face purchase minimums, short-term redemption fees, and sales charges outside the expense ratio entirely. The honest comparison is total cost for the specific two funds, in the specific account, over the intended holding period. No wrapper decision changes what the contents do. A bond allocation held through either structure reprices as yields move, for the reasons set out in the guide to how bond yields work, and its current level can be followed on the U.S. Treasury yields page. The wrapper governs access, price formation and tax mechanics. It does not govern risk, a point the site’s examination of when diversification fails makes from the portfolio side.

Where the tidy comparison breaks down

Tax is the difference most often cited in the ETF’s favour, and it is real, conditional and routinely overstated. When a mutual fund manager sells holdings at a profit, or departing investors force the fund to sell holdings to meet redemptions, the resulting capital gains are distributed to everyone still holding the fund, who owe tax on gains they did not choose and may receive in a year when the fund’s value fell. The ETF’s in-kind machinery largely avoids this, because authorised participants redeeming shares take securities rather than cash, so the fund can pass out holdings without a taxable sale inside the fund. ETFs can still distribute taxable capital gains; their structural advantage is typically fewer such distributions, not their elimination. The SEC’s bulletin states the result carefully: because many ETFs exchange portfolio securities in kind, they typically have fewer capital gains distributions, and therefore lower taxes, than mutual funds in a taxable account. The same bulletin states the boundary in the next breath. Held inside a tax-advantaged account such as a 401(k) or an IRA, there is no tax difference between an ETF and a mutual fund at all. Since a large share of American fund ownership sits inside exactly those accounts, the ETF’s signature advantage applies to only part of most households’ money — the taxable brokerage account — and is worth nothing on the rest.

March 2020 supplied the stress test for the other signature claim, intraday liquidity. As the pandemic sell-off reached the corporate bond market, the underlying bonds largely stopped trading at reliable prices, while the bond ETFs holding them kept trading all day. The two prices separated. The iShares investment-grade corporate bond ETF known by its ticker LQD, a fund of roughly $30 billion at the time, closed on March 12, 2020 at a 5.0 percent discount to its stated net asset value, having never closed beyond a 1 percent discount or premium in 2018 or 2019, and the discount narrowed to 0.19 percent the following day. The Bank of England’s interim financial stability report recorded that in mid-March some of the largest investment-grade and high-yield corporate bond ETFs recorded discounts in excess of 5 percent, having been no larger than 0.1 percent in January. The Bank for International Settlements, reviewing the episode in April 2020, drew the conclusion that has since become the standard reading: the discounts did not show the ETF price failing to find the value of the bonds, they showed it finding that value faster than a net asset value built from bonds that were barely trading. The exchange price was the live signal. The once-a-day value was the stale one.

That episode cuts in both directions. For the investor who needed to sell a bond allocation in the worst week of March 2020, the ETF offered an exit at a visible, if painful, price while the bond market itself offered almost none. But the size of that discount to the reported net asset value was not a simple measure of how far below fair value a seller sold, because the NAV itself could lag changing conditions in a market where the underlying bonds were barely trading. The episode showed both the usefulness of exchange liquidity and the difficulty of measuring its cost during stress. Liquidity is an option, and its value depends on how it is used. Trading a portfolio at 11 a.m. is worth a great deal to an institution rebalancing under pressure. To a household accumulating for a retirement decades away, it is worth close to nothing, and it tempts the holder to treat a long-term holding as a trading instrument. The mutual fund’s single daily price functions for that household as a mild and useful friction.

The final breakdown is the framing itself. ETF versus mutual fund is a question about containers, and containers are the second decision, not the first. The first decision is what to own: which index or strategy, and how much risk the purpose of the money can carry. An expensive active strategy costs roughly the same in either wrapper, an index strategy is cheap in both, and the ICI averages show investors in index mutual funds paying lower asset-weighted average expense ratios than investors in index ETFs — without establishing which wrapper offers the cheapest fund for a particular benchmark and account. Settled in the right order, the choice becomes mostly administrative. In a taxable account, an ETF’s in-kind structure can reduce capital-gains distributions relative to a comparable mutual fund. In a tax-advantaged account, that distinction falls away, and the choice turns on the particular funds’ expenses, available share classes, transaction costs and the account’s purchasing features — not the wrapper’s reputation. In both cases the prospectus figures for the two specific funds, not the reputation of either wrapper, decide what the plumbing costs.

Market commentary for informational purposes only, not investment advice.