Crypto ETF fees: the four costs, not one

Every comparison table you will find sorts crypto funds by expense ratio, as though that were the cost of ownership. It is not. It is one of four, it is the only one that is published, and for a frequent buyer on a percentage-fee platform it is often the smallest of the four.

The four costs

One: the sponsor fee

This is the number everyone quotes. It is deducted from fund assets rather than charged to you, which is why it never shows up on a statement and why people routinely forget it exists.

Two features of that distribution are worth pausing on. First, the cheap end is now populated mostly by London listings, which is a genuinely unusual outcome and a direct consequence of several issuers arriving into the UK market simultaneously in late 2025. Second, the expensive tail is almost entirely converted trusts — GBTC, ETHE, GSOL — where the sponsor is pricing to a captive base rather than competing for new money.

Two: the bid-offer spread

The spread is the gap between what you pay to buy and what you receive to sell at the same moment. You pay it every time you deal, in both directions, and nobody publishes it.

In the largest funds it is a couple of basis points and effectively invisible. In smaller products — a $200m altcoin fund, a newly listed note, anything outside the main lines — it can be an order of magnitude wider, and it widens further precisely when markets are stressed and you most want to deal.

The practical rule: your sensitivity to spread scales with how often you trade, and your sensitivity to fee scales with how long you hold. A once-a-decade buyer should ignore spread and chase the fee. A monthly contributor should do close to the opposite. Very few comparison pages make that distinction, and it inverts the ranking.

Three: your platform

This is where the UK market differs sharply from the American one, and where the biggest available saving usually sits.

UK investment platforms charge in two broad ways. Some levy a flat annual fee regardless of portfolio size. Others charge a percentage of assets held — commonly up to 0.45% on the first tranche, tapering above it. If you are on a percentage-fee platform, that charge can be larger than the fund's own fee by a factor of three or more.

Put concretely: a 0.05% note held on a 0.45% platform costs you 0.50% a year all in. The same note on a flat-fee platform might cost 0.05% plus a fixed amount. The fund is identical. The choice of platform is doing nine-tenths of the work. Dealing commissions add to this on a per-trade basis and range from nothing to around £12 depending on the broker — we set out who charges what on the UK brokers page.

Four: currency

A UK investor buying a line priced in US dollars converts sterling on the way in and back on the way out. Broker FX charges typically run from around 0.15% at the competitive end to 1.50% at the expensive end, applied each way.

At the top of that range, a round trip in and out costs you up to 3% in currency conversion alone — more than a decade of a 0.25% fund fee. Several London-listed crypto notes have a sterling-denominated line precisely to avoid this. Check the currency on the order ticket, not just the ticker. Note that buying a GBP-denominated line does not remove your underlying currency exposure, since bitcoin is priced globally in dollars; it removes the conversion charge, which is a different and more immediately controllable thing.

Tracking difference: the number that sums it up

If you only look at one figure, make it tracking difference rather than expense ratio. It measures what actually happened — how far the fund's result diverged from the asset over a period, after every cost and friction inside the product.

In a large spot fund with healthy creation and redemption, tracking difference lands close to the sponsor fee, because there is very little else going on. That is why fee is a workable shorthand for the main bitcoin funds. Where it stops being a workable shorthand is in products with thin arbitrage, cash-settled creation, or a derivative overlay — and in futures-based products, where roll cost has historically produced divergence far larger than any fee. We cover that mechanism on spot vs futures.

A worked example

Take two UK investors, both putting £10,000 into bitcoin exposure through a London-listed note and holding for five years without dealing again.

Investor A picks a 0.05% note on a flat-fee platform costing £120 a year, with a £5 dealing commission and a GBP-denominated line. Annual cost: roughly £5 in fund fee plus £120 platform. Over five years, in the region of £630 including the deal.

Investor B picks a 0.25% note on a platform charging 0.45% custody, deals in a dollar line with a 1.00% FX charge each way. Annual cost: roughly £25 fund fee plus £45 platform, plus about £100 in FX on entry and a similar amount on exit. Over five years, in the region of £550 in running costs plus £200 in currency, so around £750.

The fund fee difference — the only thing the comparison tables show — accounts for £100 of the gap. Everything else accounts for the rest. These are illustrative arithmetic on stated tariffs, not a projection of any investment outcome, and they ignore the price of bitcoin entirely, which will dwarf all of it in either direction.

Fee waivers and what happens when they end

A recurring pattern in this market is the introductory waiver, and it is worth reading the expiry terms rather than the headline.

BlackRock's iShares Bitcoin ETP in Europe ran at 0.15% until 1 January 2026 before reverting to 0.25%. Invesco's UK bitcoin product ran a 0.10% discount through the end of 2025. Bitwise's Solana staking fund waived the fee on the first $1bn for three months before a 0.20% net ratio applied. VanEck's solana fund carries a 1.50% headline with a waiver on the first $1bn. Bitwise's reduced UK Core Bitcoin rate was stated as continuing "until further notice", which is not the same as permanent.

None of this is improper — it is standard fund launch practice and it is disclosed. But a table built on waiver-period pricing will be wrong within a year, which is another reason to check the Key Information Document on the day you deal rather than trusting a figure on any website, including this one.