Finvest · ETFs & Funds
Part II · Choosing well · Chapter 5 of 13

Chapter 5: The expense ratio

9 min read · Evidence current as of June 2026 · Updated June 17, 2026
The fee leak $10,000 goes in. Both funds earn 7% a year before fees for 30 years. The only difference is the fee. $10,000 goes in 0.05% a year $75,063 the 0.05% fund 1.00% a year the leak: $17,628 $57,435 the 1.00% fund
Figure 5.1. Two funds receive the same $10,000 and earn the same 7% a year before fees for 30 years. The 0.05% pipe delivers $75,063. The 1.00% pipe delivers $57,435, and the missing $17,628 sits in a pool that never appeared on any statement. The 7% is an assumption for illustration, never a forecast.

Two funds receive the same $10,000. Both portfolios earn the same 7% a year before fees, every year for 30 years. The only difference is the price tag: one fund charges 0.05% a year, the other 1.00%. The cheap fund grows to $75,063. The expensive fund grows to $57,435. The gap, $17,628, is larger than the original deposit, and at no point in those 30 years did it show up on a statement, an invoice, or a confirmation email. The money simply failed to arrive.

This chapter is about that gap: how the fee is collected without you ever seeing it leave, what funds actually charge, and why the expense ratio is the one number on a fund page you fully control. Future returns are guesses, manager skill is a hope, and the next decade of markets is anyone's draw. The fee is printed on the page today, and you choose it the moment you choose the fund.

A fee collected without a bill

The expense ratio is the percentage of the fund's assets that the fund company keeps each year to pay managers, lawyers, custodians, and itself. Funds do not send bills. The fee accrues daily, shaved off the fund's NAV in slivers, so every price you ever see is already net of it. Your statement shows the fund's return after the skim, and a return that is quietly 1% lower than it could have been looks exactly like a slightly disappointing market.

Chapter 2 traced the tax plumbing and showed how the ETF wrapper lets you sidestep other investors' capital-gains bills. No wrapper sidesteps this charge. Mutual fund or ETF, index or active, taxable account or 401(k), the expense ratio comes out the same silent way. That is what makes it the universal line 1 of Chapter 4's checklist.

Why does a small percentage cost so much money?

Because it is charged on everything you have, every year, and because every dollar removed stops compounding. A 1.00% fee is a percentage of your whole balance, not of your gains: in a year the market returns 7%, handing over 1 point means handing over about a seventh of that year's growth. Then the damage compounds, since each dollar skimmed in year 3 would have earned returns in years 4 through 30. That is why the pool in Figure 5.1 holds $17,628 even though each individual year's charge looked trivial: the expensive tank is missing both the fees and everything the fees would have grown into. Small percentage, big base, long time. Two of those three are your money's best friends, and the fee borrows both.

Dollars, not percentages

Percentages anesthetize. The same fees in dollars, charged at today's balance and again every year after:

Balance At 0.05% At 0.40% At 1.00%
$10,000 $5 $40 $100
$100,000 $50 $400 $1,000
$500,000 $250 $2,000 $5,000
Total: $610,000 $305 $2,440 $6,100

A household with $610,000 spread across those three accounts pays $305 a year at index pricing and $6,100 a year at 1.00%: a $5,795 annual difference for owning, in many cases, nearly the same stocks. The table also shows why the fee gets more dangerous as you succeed. At $10,000 the worst column costs $100 a year, an annoyance. At $500,000 the same percentage costs $5,000 a year, a vacation, a used car, a year of a child's college savings, quietly rerouted. Percentages stay still while the dollars they describe grow with your balance, which is precisely backwards from how most services price themselves: the fund does no extra work on a bigger account, yet charges proportionally more for it.

What funds actually charge

The industry's own data, from ICI's "Trends in the Expenses and Fees of Funds, 2025," asset-weighted so it reflects where real dollars sit:

INDEX MUTUAL FUNDS
0.05%

Asset-weighted average expense ratio of index equity mutual funds, 2025. The giant funds inside retirement plans anchor this average.

INDEX EQUITY ETFS
0.14%

Asset-weighted average for index equity ETFs, 2025. Broad total-market funds go as low as 0.03%; pricier specialty products pull the average up.

ACTIVE EQUITY FUNDS
~0.40%

Asset-weighted average for actively managed equity mutual funds, 2025: roughly eight times the index mutual fund average.

Two readings are worth a pause. First, the surprise: the average index mutual fund dollar (0.05%) is cheaper than the average index ETF dollar (0.14%), mostly because enormous, dirt-cheap index mutual funds dominate retirement plans while many newer, narrower, pricier products live in the ETF wrapper. The wrapper does not set the price; the product does. Second, the trend: equity mutual fund average fees fell 62% from 1996 to 2025 (ICI, same report), and the explanation is unglamorous. Investors moved their money toward cheap funds, and expensive funds either cut prices or shrank. Every transfer out of a 0.95% fund is a vote, and the industry counts votes in dollars.

Try the leak yourself

The calculator below runs the two pipes from Figure 5.1 with your own numbers. At its default settings, $10,000 at 7% a year before fees over 30 years, the 0.05% fund ends at $75,063, the 1.00% fund ends at $57,435, and the fee takes $17,628. Stretch the years and watch the leak grow faster than the tanks; the fee's favorite ingredient is time. The return assumption is stated inside the widget and it is an illustration, never a promise.

The fee is certain; the edge is not

A fair question at this point: if a skilled manager could earn an extra point or two a year, would a higher fee matter? The asymmetry is the answer. The fee is contractual, collected in good years, bad years, and sideways years, with perfect reliability. The thing the fee supposedly buys, a manager's edge over the index, is a hope, and Chapter 6 counts exactly how often the hope pays. You are trading a certainty for a maybe, which can be rational only when the maybe is large, durable, and cheap. Paying less is the one part of your future return you can lock in today, and it is why cost sat at line 1 of Chapter 4's checklist instead of line 5.

Dev checks his blended cost once a year, in one line of a spreadsheet. His $140,000 sits in three funds: $78,400 in a US total-market fund at 0.03%, $33,600 in an international fund at 0.07%, and $28,000 in a bond fund at 0.03%. The math: $23.52 plus $23.52 plus $8.40 is $55.44 a year, a blended 0.04%. His climbing gym costs more than that every month. When a coworker pitched him a five-star fund charging 0.85%, Dev ran the one-liner on his own balance instead of arguing: 0.85% of $140,000 is $1,190 a year against his $55.44. The new fund would need to beat his portfolio by 0.81 points a year, every year, just to break even. He kept the spreadsheet open and the conversation short.

Pay under 0.10% for broad index exposure; 0.03% funds exist and they are not lesser products. Any fund charging over 0.50% carries the burden of proof and must show you, in its holdings rather than its brochure, what it does that a 0.03% fund cannot. Once a year, multiply each fund's expense ratio by your balance in it, and let the dollar answer move money.

Where people go wrong

Judging cost by the share price. A $480 ETF share is not more expensive to own than a $24 one; Chapter 1 separated price from value, and the same lesson applies to costs. The only ownership cost that scales with your money is the expense ratio, and it is the line to read.

Assuming a higher fee buys higher quality. In most of the economy, price tracks quality at least loosely. In funds the relationship has tended to run the other way, because every basis point of fee is subtracted directly from the return you keep. Chapter 4's tracking test made the point concrete: the expensive fund in Mara's plan delivered the same basket, minus more.

Ignoring fees inside a 401(k). The plan paperwork looks official and the menu feels pre-approved, so people assume someone negotiated every fund's price. Mara's 0.95% fund sat on an employer menu for years, next to a 0.04% twin. The dollars come out of your balance either way. Run line 1 on every fund you hold, employer-chosen or not, and remember from Chapter 4 that the discovery reads in dollars: $95 per $10,000 each year against $4.

Waiting for a better moment to switch. The fee is charged every year you delay, and inside a retirement account the swap from a dear fund to a cheap one usually has no tax cost at all (Chapter 2). In a taxable account there can be a capital-gains bill, which deserves a real comparison: a one-time tax against a leak that runs every year for decades. Often the leak is the larger number; the point is to do the arithmetic instead of letting inertia do it for you.

Fund fees are one room of a bigger house. Advisor fees, account fees, and the rest of the household version of this arithmetic live in the Finvest Personal Finance Guide's fee chapter, which applies the same dollars-not-percentages discipline to everything you pay for financial help.

Key takeaways

  • $10,000 at 7% a year before fees for 30 years ends at $75,063 in a 0.05% fund and $57,435 in a 1.00% fund. The $17,628 difference never appears on any statement.
  • The expense ratio is skimmed from NAV daily, with no bill and no line item, in every wrapper and every account type. Chapter 2's tax plumbing has an ETF escape hatch; the fee does not.
  • In dollars, a $610,000 household pays $305 a year at 0.05% and $6,100 a year at 1.00%, charged again every year.
  • ICI's 2025 fee data, asset-weighted: index equity mutual funds 0.05%, index equity ETFs 0.14%, active equity funds about 0.40%. Equity fund fees fell 62% from 1996 to 2025 because investors moved toward cheap funds.
  • The fee is certain; the manager's edge is a hope. Pay under 0.10% for broad index exposure, and make any fund over 0.50% explain itself before it touches your money.

Sources: ICI fee trends 2025 (Research Perspective) · ICI 2026 Fact Book · Vanguard indexing history · Investor.gov on mutual funds · Finvest Personal Finance Guide