ETF expense ratio vs. total cost: when the cheaper fund costs more

Compare funds with equivalent exposure, benchmark, currency treatment, and income policy before comparing costs. A cheaper price tag on a different investment is not a like-for-like saving.

The expense ratio describes annual fund operating expenses relative to average net assets. It is not a charge on each purchase and is not the same as your broker's commission.

[Vanguard's expense-ratio definition](https://investor.vanguard.com/investment-products/etfs/etf-fees) explains that denominator. Confirm whether an advertised management fee includes all the operating costs you intend to compare.

Build three separate buckets: recurring fund expenses, your entry and exit costs, and any account, advice, or currency-conversion charges. Identify costs outside the headline ratio rather than assuming it covers everything.

This guide uses fully paid, unleveraged holdings. The arithmetic is global; local fee labels, tax treatment, and the contents of disclosure tables are not interchangeable. All numerical examples below are invented, not live ETF quotes.

Use average holding value × annual expense rate × fraction of a year as a first-order estimate. An annual rate of 0.20% is 0.002 in decimal form, not 0.20.

For an approximately constant 10,000 holding over 90 days, using a stated 365-day year gives 10,000 × 0.002 × 90 ÷ 365 ≈ 4.93. The full-year estimate on that same constant base is 20.

The 4.93 is an estimate of the economic expense, not a separate bill to subtract from the broker's cash balance. Actual value changes, accrual conventions, and fee changes affect the result.

[Investor.gov's ETF overview](https://www.investor.gov/introduction-investing/investing-basics/investment-products/mutual-funds-and-exchange-traded-2) explains that fund expenses reduce NAV indirectly.

A partial-year holding does not incur an entire year's operating expense merely because the disclosed ratio is annual. Conversely, buying and selling several times can repeat transaction costs without extending the year.

Consider a separate, unchanged market with a bid of 99.90 and an ask of 100.10. Assume those prices can fill all 100 shares and exclude every other cost.

Relative to the midpoint of 100, entry costs 10 and exit costs another 10. The round trip crosses one full unchanged spread, not a full spread twice.

This is an isolated example, not a promise of future liquidity. If prices move between trades, the observed P/L contains a market move as well as execution effects.

[The SEC's ETF bulletin](https://www.investor.gov/introduction-investing/general-resources/news-alerts/alerts-bulletins/investor-bulletins-24) explains the bid-ask spread as an investor cost.

Actual fills already contain their execution-price effect. Do not subtract another estimated spread from a P/L calculated using those fills. A separate broker commission still belongs in the cash ledger.

• Buying 100 shares at the ask costs 10,010.
• Immediately selling 100 at the unchanged bid returns 9,990.
• The round-trip loss is 20, or approximately 0.1998% of the 10,010 purchase amount.

Assume both funds offer equivalent exposure and use the same unchanged 10,000 reference amount. Ignore market returns, taxes, financing, and all costs except those explicitly listed. This is a linear cost estimate, not a return forecast.

The trading allowances include both sides once and remain fixed in this example. One basis point, or bp, is 0.01 percentage point. A therefore has 40 bp upfront in the comparison and 10 bp per year; B has 10 and 30.

Over one quarter, A costs 40 + 10 × 0.25 = 42.5 bp, or 42.50. B costs 10 + 30 × 0.25 = 17.5 bp, or 17.50. The higher annual-fee fund is cheaper in this short-horizon scenario.

Over three years, A costs 40 + 10 × 3 = 70 bp, or 70. B costs 10 + 30 × 3 = 100 bp, or 100. The lower annual fee now outweighs the assumed trading-cost disadvantage.

These are modeled total costs, not extra amounts to deduct from already net reported returns. The exit allowance is prospective even though the comparison includes it from the start.

• Fund A: annual operating expense 0.10%; combined entry-and-exit allowance 0.40% of the reference amount.
• Fund B: annual operating expense 0.30%; combined entry-and-exit allowance 0.10% of the reference amount.

Let q denote the combined trading allowance and e the annual expense rate, both as decimals on the same fixed reference amount. Estimated cost fraction over T years is q + e × T.

Set the two funds' estimates equal. If e_B exceeds e_A, the crossover is T = (q_A − q_B) ÷ (e_B − e_A).

Here, T = (0.004 − 0.001) ÷ (0.003 − 0.001) = 1.5 years. Both estimates equal 55 bp, or 55 on 10,000. This is a cost-equality point, not the investment's profit break-even.

If A's combined trading allowance rises to 0.70% while B's stays at 0.10%, the crossover moves to (0.007 − 0.001) ÷ 0.002 = 3 years. A worse exit assumption can erase an apparent saving.

[Vanguard's holding-period analysis](https://corporate.vanguard.com/content/corporatesite/us/en/corp/articles/how-low-etf-fees-benefit-investors-over-time.html) illustrates why time matters to a cost comparison.

The linear estimate ignores changing asset values, reinvestment, compounding, and future fee changes. For a detailed cash projection, model dated trades and expense accruals instead of extending the straight line indefinitely.

If annual rates are equal, this equation has no finite crossover unless the trading allowances also match. A negative solution does not describe a future crossover under the chosen assumptions.

When a fund's published NAV total return is after operating expenses, subtracting its expense ratio again double counts them. Likewise, tracking difference is an observed return gap, not an additional invoice.

Use either a transparent forward cost model or a reconciled historical net return. Mixing assumed gross returns, net fund returns, and the same fee twice gives neither a valid estimate nor a valid historical result.

Tracking metrics explain benchmark comparisons. Total return addresses income and reinvestment.

The headline ratio need not include all portfolio transaction costs or your intermediary's charges. A zero stated expense ratio therefore does not establish zero ownership cost.

[The SEC's fee guide](https://www.investor.gov/introduction-investing/general-resources/news-alerts/alerts-bulletins/investor-bulletins-20) identifies exclusions. Check waiver end dates; a temporary rate may change.

A hypothetical fixed commission of 2 costs 1% of a 200 purchase, but 0.10% of a 2,000 purchase. Twelve purchases charged 2 each cost 24 before any future sale charges. These are not a broker's current prices.

That arithmetic does not prove waiting to invest is better. Different purchase dates change market exposure and cash returns. Separate the fee comparison from the investment-timing decision.

Currency conversion and platform charges can also depend on the broker and account. Keep a percentage conversion charge distinct from an exchange-rate movement; they are different effects.

See ETF trading currency and exposure. Choosing another trading currency can change payment costs without removing the underlying currency risk.

For a switch, compare future costs of staying with future costs of switching. Include selling the old holding and buying the new one; do not count the original purchase's sunk costs as a new switching expense.

Any taxes or time out of the market need separate analysis. A small annual saving alone does not show that selling an existing position is worthwhile.

Record the assumptions at a named comparison date. Your estimate should be reproducible from fund documents and broker terms, not just a ranking of headline expense ratios.

Execution takes time, and a quote need not be your fill. [Investor.gov's execution guide](https://www.investor.gov/introduction-investing/investing-basics/how-stock-markets-work/executing-order) explains the distinction. [!TRYMARK] TryMark ownership-cost review Before the next purchase, compare costs through a stated exit horizon. Recalculate when trade size, annual fees, bid-ask quotes, conversion charges, or the expected sale date changes. [!WARNING] Cost equality is not a reason to take more risk Lower modeled costs do not make an ETF safe or guarantee better performance. Check investment exposure and liquidity separately. This guide explains comparison mechanics, not a product recommendation.

• Match exposure, benchmark, share class, currency policy, and income treatment.
• Record annual expenses, included items, temporary waivers, and the source date.
• Estimate buy and sell costs separately, with trade size and quote timestamps.
• State the holding period, reference amount, account charges, and omitted taxes.
• Check which costs are already included in NAV returns or actual execution prices.