What You Pay, You Keep

The fee is charged on the whole balance

An expense ratio is not a purchase, it is a deduction taken from the fund’s assets every year — which is why it behaves like a slightly lower return rather than a bill. A 1% fee on an 8% market turns 8% into 7%. That sounds survivable until it compounds: over thirty years, the difference between 1.0795³⁰ and 1.07³⁰ is roughly $231,000 on a $100,000 start. The size of the effect surprises people because the fee is quoted as a percentage of assets while the outcome is a multiple of the money. One point off a return is not one point off the ending balance; it is one point off every year, applied to every dollar, for the whole horizon. In the arithmetic, a fee is indistinguishable from a negative expected return — and the same doubling logic that turned 7% into three doublings over thirty years turns a 1% fee into a material change in the count. This is why cost is treated as the most controllable input in the plan. The market will do whatever it does; the return assumption is a guess. But the fee is printed in the fund’s prospectus, it does not vary with skill or luck, and a household can change it in an afternoon. It is the only input where improvement is certain. The same market, two wrappers, thirty years — Gross return: 8.00% for both · Fund A: 0.05% cost: net 7.95% → ≈ $992,000 · Fund B: 1.00% cost: net 7.00% → ≈ $761,000 · Difference: ≈ $231,000 — 23% of the outcome ← The comparison holds the market constant on purpose. Nothing here requires a view about returns; the only variable is a price you can read in a document.

The sticker is never the whole bill

Expense ratios are the visible cost, and they are not the only one. An advisory fee is usually charged separately on assets under management, so a 1% advice fee on top of a 0.6% fund is a 1.6% drag. Trading costs are inside the fund for a high-turnover manager and are paid by the household for its own trades, arriving as spreads and slippage (Markets M17). And in a taxable account there is tax drag: distributions and realised gains are taxed as they occur, so the money compounds on an after-tax base. The useful discipline is to add the pieces before comparing two options. A fund with a 0.05% expense ratio inside a tax-deferred account is close to costless. The same fund in a taxable account still pays tax on its dividends every year, which is why asset location — putting the tax-inefficient assets in the sheltered account — is worth planning rather than improvising (PF10). And an advisor charging 1% is not automatically expensive: if the advice prevents one panic sale in a decade, it has paid for itself several times over, because the behaviour it prevented was worth far more than the fee. That last point is the honest one, and it is where this lesson stops being an argument for the cheapest possible product. Costs are a certain drag; the benefit of advice is an uncertain offset. Where the benefit is large and measurable — a plan that keeps a household invested through a drawdown, a tax strategy that actually gets executed — a fee can be good value. Where it is a wrapper around an index fund the household would have bought anyway, it is pure cost. Where an all-in cost comes from — Fund expense ratio: 0.05–1.00% · Advisory fee on assets: 0–1.00%, charged separately · Trading, spreads and slippage: small per trade, invisible per year · Tax drag in a taxable account: paid annually on dividends and gains ← Compare costs only after adjusting for what the money buys. A 0.30% index fund and a 0.90% active fund are not the same product, and the honest question is whether the extra 0.60% has any plausible mechanism for earning itself back — which the long-run evidence suggests is rare.

What the evidence says about paying for skill

The reason to default to low-cost, broad ownership is not an opinion about markets; it is the persistence of the results. S&P Dow Jones Indices’ SPIVA U.S. Scorecard for year-end 2025 reports that 79% of active large-cap U.S. equity funds underperformed the S&P 500 in 2025 — worse than the 65% rate in 2024, and the fourth-worst year in the scorecard’s twenty-five-year history. Most of the exceptions do not persist, and the fund’s own costs are the part of the shortfall that is certain. The arithmetic of that is worth stating plainly: the average active manager must beat the market by the size of the AUM-weighted cost advantage before the household sees anything, and the household that owns the index pays the cost once instead of twice. This is why “just buy the index” is the default and not a compromise — it is the position that starts with the wind at its back. Two caveats keep it honest. First, fees at the cheapest end have fallen a long way, and the difference between 0.03% and 0.10% is not the difference between 0.05% and 1.00% — chasing the last basis point has no teaching value. Second, the fee is not the only thing that decides an outcome: behaviour is at least as large (PF15), and a household that trades a cheap fund constantly can end up worse off than one that leaves an expensive fund alone. Cost is the certain part; it is not the whole story. The scorecard, one line — Active large-cap U.S. equity funds that underperformed the S&P 500 in 2025: 79% · The same measure in 2024: 65% · Where 2025 ranks in 25 years of the scorecard: fourth-worst ← Read a scorecard as a statement about the average before costs, not a prediction about any one manager. The lesson is the base rate, and the base rate is what a default should be built on.

The advice layer, and what it is worth

Fees come in two sizes and only one of them is usually visible. The fund expense ratio is a number on a fact sheet. The advice fee — paid to a planner, an adviser or a platform — is often a percentage of the whole balance, sometimes a similar size to the fund costs, and it is the larger of the two for many households. Reading the cost of investing without pricing the advice layer leaves out the line that dominates. The structures differ in a way that changes the incentives. A **fee-only** arrangement charges the client directly, by the hour, by a flat retainer or as a percentage of assets, and the adviser earns the same whether the client buys a product or not. A **commission** arrangement pays the adviser from the product, which means the recommendation and the payment come from the same place — legal, disclosed, and structurally conflicted. The intermediate case is the most common: an assets-under-management fee that pays for advice and none of the products, which removes the product conflict while creating a different one, since the fee rises with the balance and therefore rewards accumulating assets over reducing debt or spending. Then the question that decides whether any of it is worth paying: what does advice actually deliver? The measurable part is not portfolio construction — a passive three-fund portfolio can be assembled in an afternoon. The documented value is in the parts that are behavioural and structural: preventing the sale at the bottom, keeping the savings rate up, choosing an asset location, choosing a withdrawal order, coordinating insurance and estate documents, and stopping the small number of decisions that are irreversible. Studies that attempt to price this put it in the range of a few tenths of a percent a year to several percent in the years when a mistake would have been made — which is a real range and an honest one, because the benefit is concentrated in a minority of years and is nearly invisible in the rest. That asymmetry creates the practical test, and it is a scaling one. A portfolio of a certain size pays for a good amount of advice and a small one does not, because the fee is proportional while the questions the advice answers are much the same at any size. Below a threshold, the honest answer is that a fee-only hourly consultation a few times a decade is better value than an ongoing percentage. Above it, the ongoing relationship can be worth its cost, provided the advice covers the things above and not only the portfolio. The last piece is the one that never appears in a fee table: the cost of the products recommended instead of the funds chosen, and the change in behaviour a relationship produces. Both are large and neither fits on a disclosure. The way to see them is to ask what the adviser does that a low-cost index fund and a written plan would not, and to insist on an answer that is specific rather than reassuring. • The advice fee is often the larger of the two layers, and the less visible. • Fee-only separates the payment from the product; commission does not. • Assets-under-management removes the product conflict and creates a balance incentive. • The measurable value of advice is behavioural and structural, so it scales poorly to small portfolios. A useful question before signing anything: for this portfolio size, what is the fee in dollars, and which decisions in the last three years would have gone differently without it? The second half of the question is where the value either appears or does not.

Sources

• Fund expense ratios and the arithmetic of fee dragBogle, “The Little Book of Common-Sense Investing”
• Active managers against their benchmarksS&P Dow Jones Indices — SPIVA U.S. Scorecard, year-end 2025 (79% of active large-cap funds underperformed in 2025)
• Cost matters: the evidence on fees and outcomesMorningstar — U.S. Fund Fee Study
• Tax drag and asset location in a taxable accountBogleheads wiki — tax-efficient fund placement