Fees, in euros instead of percentages

Published 17 September 2026

Your fund charges 1.3% a year. It does not sound like much, and nothing ever leaves your account. This is what that figure is worth in euros over ten years, where the rest of it hides, and why the statement that already tells you does not land.

The number you know, and the two you don't

Most people who hold a fund know one cost figure. It sits in the fund's own paperwork, it is written as a percentage, and it covers running the fund: paying the managers, the administration, the audit. Across European equity funds in 2024, the EU markets authority puts that figure at 1.32% a year on average.¹ On a holding of €10,000, that is €132 in a year. That average leaves out exchange-traded index funds, which are a great deal cheaper; if you hold one of those, your own figure is a fraction of it, and the next few paragraphs show roughly where.¹

Two other layers sit around it. Your broker or bank charges for the service of holding and trading — a platform fee, a transaction charge, sometimes a fee for advice. And inside the fund, buying and selling the shares it owns costs money too. The Dutch financial regulator splits the world exactly this way: what you pay for the service, and what you pay for the product.⁴ The percentage you know is one part of the second half.

Within that one layer, the spread is wide. The EU markets authority finds average charges across European equity strategies running from roughly 0.5% to 2% a year — €50 to €200 on the same €10,000.¹ Funds that simply follow an index sit below that range altogether: on a one-year measure in 2024 they averaged 0.22%, against 1.28% for funds where someone picks the holdings.¹ In euros on €10,000: about €22 against about €128 (Figure 2).

We are not going to tell you which of those to hold. Whether the more expensive kind earns its keep is a separate question, and a harder one. What this piece does is put the percentages into euros, and then follow them for ten years.

One caution before we go further. The 1.32% and the 1.28% above come from two different tables with two different populations — the first covers private holdings, the second covers private and professional together. They are close, and they are not interchangeable. We use each where the regulator uses it, and we do not subtract one from the other.

Why it never feels like paying

No invoice arrives. No line appears on your bank statement. The charge is taken out of the fund's assets before the price is struck, so the price you see is already the price after costs. You are not asked to approve it, because you already paid it.

That design is not a trick — it is simply how a pooled fund works, and it saves everyone a monthly bill. But it removes the one moment when a cost normally registers. A €40 bank charge is annoying because you watch it leave. A €132 fund charge is almost invisible.

It also means the amount grows with you without any moment of decision. The percentage stays the same; the base does not. At €10,000 the 1.32% is €132 a year. At €60,000 it is about €790 a year, for the same fund and the same work. Your provider does have to account for it — we come back to that — but the change itself passes without anyone having to agree to it.

There is a second consequence of never seeing the bill, and it runs the other way. Because the charge is not an event, it is also not a decision. Costs you pay actively — a transaction fee, a subscription — get reviewed, because paying them takes a moment of attention. A charge deducted from a price asks nothing of you for as long as you hold the fund. Ten years of not deciding is still a decision.

What a charge actually costs you

Here is the part that the percentage hides, and the American markets regulator writes it in a single sentence: you lose the charge, and you lose the return that the charge would have earned.² The money taken out in year one is not just gone. It is also not invested for the nineteen years that follow.

Their worked example is a round one: $100,000, a charge of 1% a year, and an assumed 4% return, held for twenty years. Over that period the charges add up to almost $28,000. The return those charges would have made, had they stayed invested, is another $12,000 on top (Figure 3).²

So the charge itself is the larger half, and roughly another forty per cent arrives behind it. That second part is the one you are least likely to have seen put in money. It is not in the percentage, because it is not a charge at all — it is an absence. European rules do require your provider to illustrate this cumulative effect, and a later section is about why that requirement does not do the work you would expect.

Ten years, ten thousand euros

For a European number we can use one that was not built by us. The regulator publishes a worked case each year: €10,000 put into a representative mix of European funds — 40% equities, 30% bonds, 30% mixed — and held from 2015 to the end of 2024.¹

After ten years that €10,000 had become €15,530, after the funds' running charges were taken out. Over those same ten years, €1,687 went to those charges.¹ That is the percentage, in euros, for a decade: about a sixth of what you started with (Figure 4).

One more line belongs next to it, and it is not a cost. Prices rose over that decade too. In 2015 prices, the €15,530 is worth about €11,927.¹ The regulator keeps the two apart on purpose, and so do we: charges are something a firm takes, inflation is something that happens to everyone. Both change what you end up with; only one of them has a price list.

The regulator runs a shorter version of the same case, and it is worth a glance. Take €10,000 over the five years to the end of 2024 and you end with about €12,207 after charges, of which €704 went to the charges themselves. In 2020 prices, that €12,207 is worth about €9,956 — slightly less than the €10,000 you started with.¹ Five years of a rising market, and the purchasing power went sideways. Charges were the smaller reason; they were a reason.

Tax sits outside all of this. It differs by country and by the kind of account, and it lands on top of every figure in this article rather than inside them.

These are the running charges only. Entry and exit fees, the trading the fund does inside itself, and what your broker charges you are all outside this figure. ¹ The real total is higher than €1,687; the regulator does not say by how much.

What that number means

€1,687 over ten years is not a catastrophe. On a portfolio that grew by more than €5,000 it is real but survivable, and it bought something: somebody ran the fund. The useful question is not whether the number is big. It is what you can and cannot do about it.

Start with what you cannot. Over those ten years the markets did what they did, and that dominated everything. A single year can move the value of a fund by far more, in either direction, than a year of charges ever will, and against swings like that a difference of half a percent is invisible. If you are looking for the thing that decided your outcome, it was not the fee.

Now the part you can. The regulator ran the same ten years, the same €10,000 and the same strategy for a professional investor, who pays institutional rates. That version ended at about €16,400, against about €15,500 for an ordinary one — the same ten years as the €15,530 above, quoted to less precision because the regulator states the pair that way — roughly €900 apart, on the same strategy, from nothing but the price of access.¹ A private investor paid close to twice what a professional paid for it (Figure 5).

That is the shape of it. Charges do not decide whether a decade goes well — the market does. They decide how much of the decade's result you keep, and that part is quietly yours to influence, in a way the market never is. Across European equity strategies the average charge runs from about €50 to about €200 a year on €10,000 (Figure 6).¹

It is worth being honest about what that €900 is and is not. It is not evidence that the professional investor did better at investing; the strategy was identical by construction. It is the price of access, and that gap is not yours to close. The spread inside the ordinary market is — and it is not a small one.

One structural note belongs here. A large share of what you pay does not go to running the fund at all, but to the party that sold it to you. The regulator says so, quoting a separate study of its own — so we put no figure on it. The charges you are quoted depend on the chain you bought through.¹

You already get this figure

Here is the thing that surprised us while writing this. In the European Union, none of the above is hidden by law. The opposite: firms must give you the total cost as a cash amount as well as a percentage. Where they recommended or marketed the fund to you, or handed you the short summary document that comes with it, and have had an ongoing relationship with you during the year, they must send you a personal statement based on what you actually paid. And they must show you an illustration of the cumulative effect of those costs on your return — before you invest and afterwards.⁵

So a European investor is not short of euro figures. By the letter of the rules, much of what this article works out is already on paper somewhere in your account.

And yet the figure rarely registers. The reason is not concealment; it is framing. The rule says the cumulative effect must be shown — but it does not say over what period or on what amount. Your horizon might be thirty years. The statement covers the year just gone, on the money you had then. Nothing in it is aimed at the question you actually have, which is what this will come to by the time you need it.

There is a reason the rules leave those choices open, and it is not a bad one. Any period a regulator picks will be wrong for most people. The cost of that design is that the figures answer a question about the product rather than a question about you. You hold what you hold, for as long as you hold it, and the arithmetic that bridges those two things is left to you.

Why we underestimate it anyway

Suppose the framing were fixed. Suppose someone handed you the charges, plainly, on a single sheet, at the moment you had to choose. Would it change what you did?

Three economists tested more or less that. They asked 730 people to allocate $10,000 across four funds that all tracked the same American index. The funds held the same shares and would earn the same return before costs; the only thing separating them was the charge. The right answer was to put everything in the cheapest one. Some participants were also handed a one-page sheet listing the fees.³

Among university staff, 3% picked the cheapest fund without the sheet. With the sheet, 9% did. Among MBA students — people who study this — it went from 6% to 19%. Among undergraduates, from 0% to 10%.³ The sheet helped. It helped a lot in relative terms. And around nine in ten people still did not do the thing that was plainly better for them (Figure 7).

What they chose instead is the interesting part. Participants leaned on past returns, which for four funds tracking one index is noise. Among the staff and undergraduates who were given no fee sheet, around half spread their money across all four, as if diversifying between things that were already identical.³ The sheet did change how much weight people said they put on costs — it rose sharply in their own ranking. It changed what they did much less.

There is a reading of that result which is unkind to the participants, and we do not think it is the right one. Consider what the sheet actually asked of them. It gave a number in dollars, at a moment when four options looked broadly alike, and offered no account of what the number would come to. Past returns, by contrast, arrive as a story: this one did well. A figure has to be converted into a consequence before it can compete with a story, and nothing in the sheet did that conversion.

That is the same gap as the one in the previous section, seen from the other side. The statement you are sent is accurate, complete and on time. It reports a year. The question in your head is about a working life. Between those two sits an arithmetic step that most readers — reasonably enough — never take.

There is a real objection to leaning on this. The experiment is American, it is twenty years old, the sums were small, and a sheet in a lab is not a law. When American regulators imposed fee disclosure on workplace retirement plans in 2012, a study of a large sample of those plans found that savers did become measurably more attentive to charges afterwards. Money moved away from the dearer options. A fund charging 0.36 percentage points more lost about 0.17 percentage points of its share of the plan each year — around 6% of the 2.9% a typical fund holds.⁶ Real, measurable, and modest. Disclosure is not futile; it moved people when it arrived as a rule rather than as a leaflet, and it moved them a little.

What survives both findings is narrower, and it is the point of this article. The information was on the table, in money, at the moment of choosing — and most people still did not use it. Handing someone a number is not the same as handing them a reason to act on it, and the number has to be converted into a consequence before it can compete with a story about past performance. That conversion is the step the paperwork leaves to you.

Sources

1. European Securities and Markets Authority, Costs and Performance of EU Retail Investment Products 2025, ESMA50-1949966494-4065 (cover date 3 March 2026; reporting period 1 January 2015 – 31 December 2024). Our source for: average yearly charges of European equity funds (p. 12); charges by how a fund is run (p. 18); the €10,000 worked case over ten and five years (p. 6 and p. 15); the professional comparison (p. 15); the 0.5%–2% range (p. 15); the share of charges paid to distributors (p. 22) and distribution as a share of total cost (p. 4, note 1); and the note that trading and distribution costs could not be accounted for (p. 15, note 22) — entry and exit fees are reported separately (p. 14).
2. U.S. Securities and Exchange Commission, Office of Investor Education and Advocacy, Investor Bulletin: How Fees and Expenses Affect Your Investment Portfolio, Pub. No. 164 (2014). Our source for: the mechanism — you also lose the return the charge would have earned (p. 3) — and the $100,000 / 4% / twenty-year worked example with almost $28,000 in charges and a further $12,000 in forgone return (p. 3).
3. James J. Choi, David Laibson and Brigitte C. Madrian, Why Does the Law of One Price Fail? An Experiment on Index Mutual Funds, NBER Working Paper 12261 (May 2006, revised April 2008). Our source for: 730 participants, the four identical index funds, and the share who put everything in the cheapest fund with and without a fee sheet (p. 17), the ranking of fees among eleven factors and the spreading across all four funds (p. 19). A later version appeared in the Review of Financial Studies in 2010; we read the working paper.
4. Autoriteit Financiële Markten (Dutch financial markets authority), Wat kost beleggen?, afm.nl, read 17 September 2026. Our source for: the split between what you pay for the service and what you pay for the product, and the four categories of charge used in European product documents.
5. Commission Delegated Regulation (EU) 2017/565, Article 50, in the Official Journal of the European Union L 87, 31 March 2017. Our source for: the total must be given as a cash amount as well as a percentage (paragraph 2); an annual, personalised statement based on costs actually incurred (paragraph 9); and an illustration of the cumulative effect of costs on return, both before investing and afterwards (paragraph 10).
6. Mathias Kronlund, Veronika K. Pool, Clemens Sialm and Irina Stefanescu, Out of Sight No More? The Effect of Fee Disclosures on 401(k) Investment Allocations, NBER Working Paper 27573 (July 2020). Our source for: after the 2012 US disclosure requirement, participants "became significantly more attentive to expense ratios" (abstract); and that funds with an expense ratio 0.36 percentage points higher saw their plan share fall by 0.17 percentage points a year, "around 6% of the median fund plan share of 2.9%" (introduction and section 4). Sample: the 1,000 largest US plans plus an earlier research set, 2010–2013. Interests declared on the title page: Sialm is an independent contractor with AQR Capital Management, and Stefanescu is at the Federal Reserve Board.

How we checked. Verified 17 September 2026. Every figure from sources 1, 2 and 3 was pulled from its source document twice, independently, and the two readings were compared cell by cell; they agreed on every one. Sources 4 and 5 were read once, and neither carries a figure. The euro amounts marked as ours are single arithmetic steps on published figures — no projections, no models. Five things a careful reader should know. First, source 1 gives the ten-year charge total three times and not identically: €1,687 in its table, "approximately €1,700" in the text, and "around €1,500" in its summary; we use the table figure. Second, the $28,000 and $12,000 in source 2 are measured against holding the same portfolio with no charge at all, while the same bulletin elsewhere compares one charge against another — the two are different yardsticks and we have not mixed them. Third, source 6 was added late: our own completeness check found that our closing argument leaned on a twenty-year-old laboratory experiment while a field study of a real disclosure rule pointed the other way. It has had one reading, not the double one the first three sources had. Fourth, source 1's own charge figures exclude the fund's internal trading and the cost of distribution, and they sit alongside entry and exit fees and whatever your broker charges you, so the true total cost of those ten years is higher than the figure shown. Fifth, we could not retrieve a key information document for an ordinary index fund; we have therefore described only what the rules in source 5 require, and said nothing about the standard amounts or holding periods that such documents use. Think we got something wrong? Tell us at info@proofofreturns.com — we correct in public.