GBTC vs the Grayscale Bitcoin Mini Trust

What the Fee Difference Costs, in Bitcoin

The Sponsor's Fee is paid in bitcoin, out of the trust's holdings. It is not an abstract expense line — it is a steady reduction in how much bitcoin each share represents. The Mini Trust's 10-K states it directly: because the shares reflect accrued but unpaid expenses, the amount of bitcoin represented by a share gradually decreases over time as the trust's bitcoin is used to pay them. (Mini Trust Form 10-K FY2024) GBTC's 10-K uses identical language. (GBTC Form 10-K FY2025)

At the stated rates, accruing daily:

Arithmetic on the filed fee rates, not a projection. It assumes only that both funds continue charging what they currently charge. Model your own position with the fee-drag calculator.

The Filings Already Show It Happening

This is the part no other comparison can offer, because it requires tracking bitcoin-per-share from filings rather than quoting a fee sheet.

At the Initial Distribution, GBTC retained roughly 90% of its bitcoin against an unchanged share count while the Mini Trust received roughly 10% against a 1:1 share issuance — so GBTC represented 9× the bitcoin per share. The Mini Trust's 1-for-5 reverse split in November 2024 moved that ratio to 1.8.

Working from the June 30, 2026 filings: GBTC held 0.0007755946 bitcoin per share and the Mini Trust held 0.0004422030 — a ratio of 1.753933. Over the 699 days since July 31, 2024, the decline from 1.800000 to 1.753933 implies an annualised divergence of 1.3547%, against a theoretical fee differential of 1.3500%.

Five thousandths of a percentage point apart, over two years, derived entirely from filed figures.

The fee does exactly what the arithmetic says it does. You do not have to take the projection on faith; the divergence is already in the record. That auditability is a property of the trust structure, not of bitcoin-per-share figures in general — our IBIT vs MSTR comparison shows a filed bitcoin-per-share that cannot be verified this way.

One caveat, stated plainly: the 1.800000 starting ratio is inferred from the distribution mechanics, not stated in any filing. It corroborates the arithmetic rather than independently measuring it.

How the Mini Trust Was Actually Created

This is described loosely almost everywhere, and the details matter for anyone working out their own tax position. The distribution's place in the wider fee war is covered in our history of the road to the spot Bitcoin ETF; the mechanics, from the filings:

• The Mini Trust was formed on March 12, 2024. Its Form S-1 (File No. 333-277837) became effective, and the SEC approved NYSE Arca's rule change to list the shares on July 26, 2024. (Mini Trust Form 10-K FY2024)
• Record date: 4:00 p.m. ET, July 30, 2024. Distribution date: July 31, 2024.
• GBTC contributed 26,935.83753443 bitcoin — approximately 10% of what it held at the record date — in exchange for newly created Mini Trust shares.
• Those shares were distributed to GBTC shareholders pro rata, 1:1: one Mini Trust share per GBTC share held at the record date. 303,690,100 shares in aggregate. (GBTC Form 8-K, July 31, 2024)
• Shareholders paid nothing, exchanged nothing, surrendered nothing, and took no action. Total bitcoin exposure was unchanged — the same bitcoin, held across two vehicles instead of one.
• On November 19, 2024 the Mini Trust completed a 1-for-5 reverse share split. Restated for the split, the Initial Distribution was 60,738,020 shares at $28.92 per share, for contributed bitcoin valued at $1,756,821,047. (Mini Trust Form 10-K FY2024)

Two practical consequences.

If you held GBTC on July 30, 2024, you already own Mini Trust shares — unless you have since sold them. Many holders received them without noticing. Check your account before treating a switch as a new allocation.

The 1:1 ratio you may have read about is pre-split. Post-split it is one Mini Trust share per five GBTC shares. Anyone reconstructing cost basis from July 2024 press coverage will get it wrong.

The Tax Problem

The Distribution Was Not the Taxable Event

Grayscale's announcement of the record and distribution dates stated it in one long, carefully hedged sentence: “Subject to the limitations and qualifications set forth in the Preliminary Information Statement filed by GBTC (including with respect to the qualification of both GBTC and the BTC Trust as grantor trusts for U.S. federal income tax purposes and the proper allocation of existing tax basis between GBTC Shares and BTC Shares), it is expected that neither GBTC nor any beneficial owner of GBTC Shares will recognize any gain or loss for U.S. federal income tax purposes as a result of the Initial Distribution.” (press release, Exhibit 99.1 to GBTC Form 8-K, July 19, 2024)

Note what that sentence does. It hedges on grantor-trust qualification, and it makes basis allocation the shareholder's problem. The Information Statement itself carries the allocation framework (Definitive Information Statement, Schedule 14C, July 30, 2024), and Grayscale's worked cost-basis example applies it as a fixed split: 90% of a holder's existing GBTC basis retained in the GBTC shares, 10% reallocated to the Mini Trust shares. (Grayscale cost-basis example, July 30, 2024)

Worth being precise about one thing that is widely misdescribed: the non-recognition analysis rests on grantor-trust treatment under Subpart E of Subchapter J, not on Section 355, the corporate reorganisation provision. Secondary coverage calling this a “Section 355 spin-off” is using the term loosely.

The material point for a switching decision: the distribution did not reset anyone's basis upward. Your original GBTC basis was allocated across the two holdings. The embedded gain did not go away — it was split.

Selling GBTC Is the Taxable Event

Both trusts are treated as grantor trusts, meaning a shareholder is treated as owning a proportionate share of the underlying bitcoin. Selling GBTC shares is a disposition; gain or loss is measured against your basis in the shares sold.

For a long-time holder this is where the arithmetic turns hostile. GBTC was created in 2013 and traded on OTCQX from 2015 until its NYSE Arca uplisting on January 11, 2024. Anyone who acquired shares in that era holds a basis set against a much lower bitcoin price.

There is also a wrinkle specific to GBTC. For years before conversion the shares traded at a substantial discount to net asset value — the D.C. Circuit recorded a discount of as much as 30% in Grayscale v. SEC. Buying at a discount means your basis sits below the NAV you were acquiring, which increases embedded gain relative to someone who bought an equivalent amount of bitcoin directly.

The Fee Itself Creates Taxable Events

An underappreciated point, stated identically in both 10-Ks: assuming grantor-trust treatment, each delivery or sale of bitcoin by the trust to pay the Sponsor's Fee or additional expenses is a taxable event for shareholders. (GBTC Form 10-K FY2025; Mini Trust Form 10-K FY2024)

Both trusts pay the fee in bitcoin, daily. Both therefore generate a continuous stream of small deemed dispositions requiring basis tracking. GBTC's fee is ten times larger, so the outflow — and the associated shareholder-level disposition activity — is roughly ten times larger too. This is an administrative cost of staying, not only of switching.

The Breakeven

The comparison has a clean structure, and it requires no view on the price of bitcoin.

Sell GBTC and you pay tax now, in dollars, on the embedded gain. Convert that tax to bitcoin at today's price and you have a fixed, one-time reduction in the bitcoin you control. From that point forward the remaining position erodes at 0.15% a year instead of 1.5%. Because both funds hold the same asset, the whole comparison can be expressed in bitcoin terms — which means the breakeven horizon does not depend on what bitcoin does next. That is what makes this question answerable without forecasting anything.

The condition:

(1 − tax cost as a fraction of position) = e^(−0.0135 × years)

Two inputs, and only one of them is ours to supply.

Your tax cost as a percentage of your position is your embedded gain as a fraction of current value, multiplied by your effective rate on that gain. Both are personal. Your gain fraction comes from your own basis records; your effective rate depends on your bracket, your holding period, your state, and whether short-term rates apply. We are deliberately not printing a rate table — a specific rate on a page like this reads as advice, and the composite figures in circulation do not fit every holder.

The recovery period falls out of the first number, and this table is rate-agnostic:

Work out the first column from your own numbers and read across. A holder whose tax bill comes to a tenth of the position recovers it in under eight years. A 2013-era holder with a basis near zero is at the bottom of the table.

Three Things This Arithmetic Does Not Capture

Holding in a tax-advantaged account. In an IRA or 401(k) there is no taxable disposition on sale. The embedded-gain problem does not exist, the tax cost is zero, and the breakeven is immediate. If your GBTC sits in a retirement account, everything above about tax is irrelevant to you and the decision is just 1.5% versus 0.15%.

Never selling. Under current US federal law, assets held until death generally receive a basis adjustment to fair market value under Internal Revenue Code section 1014. A holder who genuinely intends never to sell may never pay the gain at all — in which case switching converts a tax that would never have been paid into one paid today. Highly circumstance-specific, and dependent on the law at the relevant time.

Partial switching. The decision is not binary. Selling a portion — particularly the highest-basis lots, if your broker supports specific identification — realises proportionally less gain while moving part of the position to the lower fee. Lot-selection method materially changes the tax cost of any partial sale.

What Changed in 2026

The bitcoin price decline through the first half of 2026 has altered this calculation for a large group of holders, in their favour.

From the filings: bitcoin was $87,549.41 at December 31, 2025 on the CoinDesk index used by both trusts, and $58,745.18 at June 30, 2026 on the same index. (GBTC Form 10-K FY2025; GBTC Form 10-Q)

Anyone who acquired GBTC above roughly the current level has no embedded gain to realise. Their tax cost is zero or negative and the breakeven is immediate. For that cohort the tax argument for staying has simply evaporated, and most of them do not know it.

The Mini Trust's own balance sheet shows how widespread unrealised losses have become: at June 30, 2026 it carried bitcoin at a cost of $4,292,046 thousand against a fair value of $3,188,712 thousand. (Mini Trust Form 10-Q)

The flows are consistent with holders acting on it. Across the first half of 2026 GBTC's shares outstanding fell while the Mini Trust's rose, from 109,539,963 to 122,749,963, even as bitcoin declined.

Working Out Your Own Answer

Four steps, all of which need numbers this article cannot have:

1. Find your adjusted basis in GBTC, allocated after the July 2024 distribution and reduced by fee accruals since. Your broker's figure may not account for the basis allocation correctly — check it.
2. Compute your embedded gain as a fraction of current value. If it is zero or negative, stop here; switch.
3. Apply your own effective rate to get tax cost as a percentage of position.
4. Read the breakeven table and compare it against how long you actually intend to hold.

None of this is tax advice, and the mechanics above are general. Basis, holding period, account type, state of residence and lot selection all change the answer, and a tax professional with your actual records is the right place to settle it. For how the two funds compare on everything other than the switch decision, see the GBTC and Mini Trust fact sheets and our full fund comparison.

FAQ