Photo by Willfried Wende on Pexels. Euro banknotes and coins.
An average fund fee can fall while the charge on an existing holding barely moves. The distinction sits inside the European Securities and Markets Authority’s latest annual cost study: cheaper additions can improve the market’s headline number without delivering an equivalent saving to people already invested.
For a reader seeing lower fees advertised in 2026, that is a useful tension. The report was published on 3 March this year, but its observations run to the end of 2024. It describes a historical change in the market, rather than today’s bill for a particular account.
Fact: the market and the continuing funds differ
ESMA’s 2025 costs and performance report says ongoing costs for retail equity funds excluding exchange-traded funds fell by about 8% between 2020 and 2024. For incumbent share classes, the fall was about 3%. Bond funds showed reductions of almost 15% across the wider sample and 9% for incumbents.
These are relative reductions in costs, measured using a one-year investment horizon. They are not percentage-point reductions in a fee rate, annual discounts repeated every year, or increases in investment returns. ESMA links part of the wider decline to new funds with lower costs. Its incumbent comparison concerns share classes already offered over the preceding five years.
The difference matters because “fund fees are falling” can describe two separate developments: existing products becoming less expensive and the market acquiring a cheaper set of products. Both may occur together. Neither tells us the change in a named investor’s charge.
Interpretation: a changing menu is different from a cheaper bill
Consider a deliberately simplified market with three fictional share classes, charging 1.00%, 1.00% and 0.40% annually. Their unweighted average charge is 0.80%. Add a fourth charging 0.20%, and the average falls to 0.65%, although none of the original charges has changed.
That is an illustration of how a changing product mix can affect an average. It does not reproduce ESMA’s weighting method or estimate the contribution of new launches to its results. Real fund markets also have different amounts invested in each product and investors entering and leaving.
The illustration nevertheless identifies the missing step in a broad claim about savings. Evidence that the menu is cheaper must be connected to evidence about the particular share class held. A share class is a version of a fund with its own terms; the fund name alone may not identify the applicable charge.
For reporting purposes, a fee reduction needs a starting rate, an ending rate, effective dates and the same product identifier. Without those details, a market-wide improvement is easily mistaken for a saving received by everyone. Our interpretation is that the incumbent comparison is the more relevant starting point for examining that claim, while still being too broad to settle an individual case.
A small percentage needs its own denominator
Suppose a fictional annual charge falls from 1.00% to 0.92%. The charge has declined by 8% relative to its original level, and by 0.08 percentage points. On an assumed constant €10,000 balance, that difference represents €8 for one year: €100 compared with €92.
A 3% relative reduction from the same starting rate would produce a 0.97% charge, or €97 on the same constant balance. Neither calculation is an estimate of what any actual fund charged. They translate two different kinds of percentage into comparable units.
The constant balance is doing work here. In a real portfolio, the amount exposed to an annual percentage charge changes as asset values, contributions and withdrawals change. An investor’s cash saving therefore cannot be inferred from a fee-rate change without a balance and time period.
This is also why a percentage decline in costs should not be added directly to a fund’s return. Cost rates and investment returns measure different things. A lower recurring charge improves the arithmetic under otherwise identical assumptions; it does not determine the movement in the underlying holdings.
Fact: the product charge is only part of the structure
A separate ESMA distribution-cost assessment published on 6 November 2025 found distribution accounted for 48% of total UCITS costs in its analysis. UCITS are funds operating under the EU’s harmonised investment-fund framework. Distribution concerns the route through which the product reaches the investor.
That proportion is an aggregate finding from a separate exercise. It is not a charge of 48% on invested money, nor a rule that an investor should add 48% to a displayed ongoing charge. ESMA’s annual report flags missing explicit distribution costs paid directly to distributors as a limitation.
The interpretation is straightforward: comparisons need a boundary. A product-level figure and a figure including services around the product can both be correctly reported and still answer different questions. Adding them without checking their contents can also count the same cost twice.
A useful cost account identifies who receives each charge, whether it is already included elsewhere, and whether it recurs or arises on a transaction. That structure matters before any verdict about which total is larger.
Different wrappers complicate the comparison
The US Securities and Exchange Commission’s July 2025 investor bulletin on fund fees distinguishes annual operating expenses from shareholder fees. It also explains that a prospectus fee table may omit brokerage commissions and other intermediary charges, and that multiple share classes can hold the same securities while carrying different costs.
Those are US disclosure explanations, not a statement of EU or UK rules. Their relevance here is the comparison method: identify the charge being measured before treating a prominent number as total cost.
For an exchange-traded fund, the trading price supplies another condition. The SEC’s ETF bulletin explains the bid–ask spread: the difference between the quoted buying and selling prices.
In a fictional unchanged market, buying 100 shares at €100.10 and immediately selling them at €99.90 would produce a €20 difference before commissions. That illustration concerns one round trip. It is not an annual fee and cannot be turned into one without assumptions about trading frequency. Nor does the quoted spread necessarily describe the execution price of every trade.
A comparison that puts an annual charge beside a transaction cost should retain those different time bases. Otherwise, a precise-looking total may depend more on hidden assumptions than on the displayed numbers.
Who could benefit?
Fact. Amundi (Euronext Paris: AMUN) reports continued ETF-platform expansion in its second-quarter 2026 update. DWS (Xetra: DWS) says Passive, including Xtrackers, helped drive inflows in its second-quarter results. BlackRock (NYSE: BLK) describes a European iShares ETF business in its 2025 annual report. These are listed managers of funds, rather than the funds themselves.
Interpretation. All three could gain assets if investors direct new money towards their competitively priced products. A lower ongoing charge can help attract business, but a manager needs enough additional fee-paying assets to offset the lower revenue earned on each unit. Distribution access, product choice and servicing costs also matter. Over several quarters, net flows and management-fee revenue would test that mechanism; larger holdings alone would not establish its cause.
Fund management is central to these businesses, but the exposure is broader than EU retail fund charges. Active strategies, institutional mandates and other services can change the group outcome. None of the cited disclosures establishes that these companies caused ESMA’s historical decline, or that their existing clients received the average saving.
Who’s at risk?
Fact. BlackRock’s annual report identifies competitive fee reductions as a possible threat to assets, revenue and earnings. DWS attributes its latest quarter’s higher management-fee revenue mainly to higher average assets under management. Fee revenue can therefore rise while individual product charges face pressure: the asset base and the rate are different parts of the calculation.
Interpretation. Amundi, DWS and BlackRock face pressure if competition lowers charges on existing assets without sufficient new business or cost savings. Migration from higher-fee products into lower-fee products within the same group can also change revenue even when the group retains the assets. That is a conditional business mechanism, not a claim that such migration explains any reported result. Market movements, exchange rates and the mix of mandates complicate the comparison.
Speculation. Cheaper launches could prompt broader reductions on continuing share classes, or leave existing terms largely intact. No probability is assigned. Over coming reporting periods, dated product-fee changes, flows by strategy, fee revenue and operating costs would help distinguish those paths. This article estimates no earnings effect and forecasts no share price; a manager’s potential gain is separate from the documented saving on a particular holding.
Speculation: whether cheaper launches will change existing charges
New products could increase competitive pressure on established funds. Existing providers might reduce charges, introduce different share classes or retain their current terms. The historical comparison does not establish which response will dominate, when it would happen, or how widely any saving would reach investors.
Evidence of that transmission would include dated changes to continuing share classes, supported by fund documents, alongside comparable information about distribution and account costs. A subsequent report showing a narrowing gap between the wider market and continuing products would also change the assessment.
The present limitation is substantial: data ending in 2024 cannot settle current 2026 charges, and aggregate categories cannot replace a product-specific comparison. This article has not audited a named fund or an investor’s complete account costs.
The useful conclusion is narrower than a product ranking. Lower market averages are evidence of change, but the saving attached to an existing holding needs its own documentation. The next meaningful number is the comparable charge on the same share class, with the cost boundary and effective date made clear.
Sources checked on 12 October 2026. Informational analysis; no personalised investment, product or allocation recommendation.