“Keep fees low” is the least controversial advice in personal finance, which is exactly why it’s usually stated without numbers. Here are the numbers.
The receipt
Take $10,000, invest it for thirty years, and assume a 7.00% nominal annual return before fees. Now charge two different prices for the same exposure to the same market:
| Fund type | Expense ratio | Value after 30 years |
|---|---|---|
| Index ETF (e.g., VOO) | 0.03% | $75,470 |
| Average active equity fund | 0.65% | $63,410 |
| Typical advisor-sold A-share fund | 1.00% | $57,434 |
Same market, same thirty years, same starting balance. The 0.97-point gap between the index fund and the 1% fund is $18,036 — on a $10,000 investment. That’s not a rounding error; it’s a used car.
The math is one formula: $10,000 × (1 + 0.07 − fee)^30. You can run it in any spreadsheet in four seconds. I’m showing it because “fees matter” without the exponent is just a slogan.
Why the cheap fund also usually wins outright
The fee gap only tells you what you’d lose if both funds picked stocks equally well. They don’t. The SPIVA U.S. Year-End 2023 Scorecard found that 86.92% of large-cap mutual funds underperformed the S&P 500 over the trailing ten years. Over fifteen years it’s worse: 88.06%.
So the realistic comparison isn’t “0.03% versus 1.00% for the same return.” It’s 0.03% for the index return versus 1.00% for a return that, with roughly 87% probability, is lower than the index. The fee math above is the optimistic case for the expensive fund.
The expense ratio is the only number in a fund prospectus that’s guaranteed. Everything else is a forecast.
The two questions that replace fund picking
- What market am I buying? (Total U.S., S&P 500, total international, bonds.)
- What does it cost? (One line in the prospectus. If it’s above 0.20% for plain index exposure, close the tab.)
That’s it. There is no third question, and anyone asking you a third question is usually being paid by the answer.
The one honest caveat
7% nominal for thirty straight years is a simplification — real returns arrive in ugly sequences with drawdowns of 30% and more, and a decade can finish below 7% (2000–2009 for the S&P 500 total return was slightly negative). The ordering survives every sequence I’ve run, though: at every horizon, the lower-fee fund keeps more of whatever the market actually delivers. That’s the whole claim, and it doesn’t require a forecast to be true.
Next: how I applied this to my own rollover IRA — three funds, one spreadsheet, six years of receipts.