Explainers
Bitcoin ETF fees: what you actually pay
Sponsor fees run from 0.14% to 1.50%, but that is one of four cost layers. We break down how the fee is charged, the spread nobody counts, and the ten-year arithmetic.
Everyone compares expense ratios, because it is the one number issuers publish prominently and it fits in a table. It is also, for a meaningful share of buyers, not the largest cost they pay.
This page separates the four layers of cost in a spot Bitcoin ETF, shows how the fee is physically collected — which is stranger than most people expect — and runs the ten-year arithmetic so you can see when eleven basis points matter and when they are noise.
Four layers of cost, only one of them advertised
| Cost layer | What it is | Typical size | When you pay it |
|---|---|---|---|
| Sponsor fee | Annual charge for running the trust, custody and administration | 0.14% – 1.50% a year | Continuously, deducted in kind |
| Bid-ask spread | The gap between the best bid and the best offer | 0.02% – 0.20%+ per trade | On every entry and every exit |
| Tracking difference | Gap between fund return and bitcoin return | Roughly the fee, plus operational friction | Continuously, as performance |
| Commission | Broker charge per trade | $0 at every major US broker | Per trade, where applicable |
Only the sponsor fee appears in marketing material. The spread is the one most often underestimated by active buyers.
The sponsor fee
This is the annual percentage the issuer takes to operate the trust: custody, administration, transfer agency, audit, legal, marketing, and profit. In an equity ETF the same concept is called the expense ratio and covers a portfolio manager. Here there is no manager — the fund holds one asset and does nothing to it — which is why competitive pressure has pushed fees down so hard.
Morgan Stanley entered in April 2026 at 0.14%, undercutting BlackRock's 0.25% deliberately, and became the first spot Bitcoin ETF issued under a major US bank's own name. Grayscale's Mini Trust sits at 0.15%. Most of the category clusters between 0.19% and 0.25%.
And then GBTC charges 1.50% — ten times the cheapest, for bitcoin held at the same custodian Grayscale uses for its own 0.15% fund. That is not an error in the table. The explanation is covered in the GBTC profile.
How the fee is physically paid, and why it matters at tax time
You never receive a bill. The trust sells a small amount of its own bitcoin periodically and uses the proceeds to pay the sponsor. Your share count is untouched; the amount of bitcoin standing behind each share falls slightly over time.
That mechanism has a consequence people discover in April. Because most spot Bitcoin ETPs are grantor trusts, shareholders are treated as owning a pro-rata interest in the underlying bitcoin. When the trust sells bitcoin to pay expenses, that is technically a disposal of your pro-rata share — a small taxable event, repeated through the year, on a position you never traded. Brokers generally handle the reporting, but it belongs on your radar. Full treatment in Bitcoin ETF taxes.
The spread: the cost nobody counts
When you buy, you pay the ask. When you sell, you receive the bid. The difference is a real, immediate cost that never appears as a fee.
IBIT's 30-day median bid-ask spread has run around 0.02% — roughly $2 on a $10,000 trade. ARKB has been quoted nearer 0.05%. Smaller funds can be several times wider still, and every fund's spread blows out in the first and last fifteen minutes of the session and during volatile moves.
Here is where it flips the fee comparison. Buy once and hold ten years: you pay the spread twice and the fee 3,650 times, so fee dominates and the cheapest fund wins. Trade monthly: you pay the spread 24 times a year. A 0.03 percentage-point spread disadvantage paid 24 times is 0.72% a year — which comfortably swamps the 0.11 percentage-point fee saving you chased. Order-placement tactics are in limit orders on crypto ETFs.
Tracking difference
A spot fund should return bitcoin's return minus its fee. In practice small gaps open from cash drag between trade and settlement, execution costs when the trust buys or sells, and the timing mismatch between a once-daily NAV and a 24/7 underlying market.
The July 2025 approval of in-kind creations and redemptions narrowed this measurably. Letting authorised participants deliver actual BTC instead of cash removed a conversion step, improved tax efficiency at fund level and tightened spreads. The SEC's own order cited enhanced tax efficiency and lower transaction costs as reasons. Mechanics in creation and redemption.
As a rule of thumb: a spot fund lagging bitcoin by roughly its fee over a year is working correctly. Lagging by substantially more is worth investigating.
Sponsor fees, all twelve funds
| Ticker | Issuer | Sponsor fee | Net assets | Notes |
|---|---|---|---|---|
| MSBT | Morgan Stanley | 0.14% | New — Apr 2026 | Lowest headline fee; first fund from a major US bank. |
| BTC | Grayscale | 0.15% | ~$3.4B | Spun out of GBTC; cheapest Grayscale option. |
| EZBC | Franklin Templeton | 0.19% | ~$545M | Low fee, thin volume. |
| BITB | Bitwise | 0.20% | ~$3.7B | Publishes its on-chain wallet addresses. |
| HODL | VanEck | 0.20% | ~$1.4B | Has run promotional fee waivers. |
| ARKB | ARK Invest / 21Shares | 0.21% | ~$3.6B | Spreads slightly wider than IBIT. |
| IBIT | BlackRock | 0.25% | $60.0B | Deepest liquidity; listed options. |
| FBTC | Fidelity | 0.25% | ~$13.2B | Only major fund that self-custodies. |
| BTCO | Invesco / Galaxy | 0.25% | ~$560M | Sub-scale versus the leaders. |
| BRRR | CoinShares | 0.25% | ~$544M | Dual custodians. |
| BTCW | WisdomTree | 0.25% | <$500M | Smallest of the January 2024 cohort. |
| GBTC | Grayscale | 1.50% | ~$14.9B | Legacy 2013 trust; highest fee in the group. |
Sorted by fee, cheapest first. Sponsor fee as disclosed by each issuer. Full detail including custodians on the spot Bitcoin ETF list.
What this costs over ten years
On a flat $10,000 position — held constant to keep the comparison clean — ten years of sponsor fees works out roughly as follows.
| Fee | Per year | Ten years | Example fund |
|---|---|---|---|
| 0.14% | ~$14 | ~$149 | MSBT |
| 0.15% | ~$15 | ~$160 | BTC (Grayscale Mini) |
| 0.20% | ~$20 | ~$213 | BITB, HODL |
| 0.25% | ~$25 | ~$265 | IBIT, FBTC |
| 0.95% | ~$95 | ~$1,020 | BITO (plus roll cost) |
| 1.50% | ~$150 | ~$1,540 | GBTC |
Illustrative arithmetic on a static balance for comparison, not a return projection. On a growing balance the absolute gaps widen substantially.
The IBIT-versus-MSBT gap is about $116 over a decade. Real, but small enough that liquidity, options availability or simple platform convenience can reasonably outweigh it. The GBTC-versus-Mini-Trust gap is about $1,380 for the same asset from the same issuer, and no convenience argument survives that.
Self-custody has no annual fee at all
Every layer of cost on this page exists because someone else is holding the asset for you. Buy bitcoin on a licensed exchange and hold it yourself and the ongoing fee is zero — CEX.IO is FinCEN-registered, licensed for money transmission across US states, and authorised in Gibraltar as a DLT provider.
Waivers: read the expiry date
Issuers use temporary fee waivers to buy market share at launch, and the headline number you see may not be the number you pay next year. BlackRock ran 0.12% on IBIT's first $5 billion for its first twelve months; that waiver has expired and the standing fee is 0.25%. On the ether side, 21Shares' TETH has been waived until 8 October 2026, and BlackRock's ETHB until March 2027 or $2.5 billion in assets, whichever comes first.
Before you switch funds for a fee advantage, check the prospectus for whether it is the standing fee or a waiver, and when it lapses. Switching in a taxable account triggers a capital gain — paying tax to chase a discount that expires in four months is a bad trade.
Our take from the desk
We think the fee war is close to over and largely won by investors. At 0.14% the wrapper is cheap enough that arguing over basis points is a poor use of attention. The two decisions still worth real effort are getting out of GBTC if you are in it without a large embedded gain, and never using a market order on a fund outside the top handful. Those two are worth more than every fee comparison on this page combined.