ARKB vs BITB: the fee difference is real, the spread numbers aren’t

Compare

August 20, 2026

6 min read

Compare

On expense ratio, BITB is one basis point cheaper than ARKB – 0.20% against 0.21%, according to btcetfcalc.com, crypto.news, Synapse Trading and CoinCodex. On spread and on fund-specific tracking error, none of those four sources gives a number for either fund, despite several of them naming spread as a reason to prefer one over the other.

The one number every source agrees on

Four independent outlets – btcetfcalc.com, crypto.news, Synapse Trading and CoinCodex – all list the same pair of figures: ARK 21Shares Bitcoin ETF (ARKB) charges an annual expense ratio of 0.21%, and the Bitwise Bitcoin ETF (BITB) charges 0.20%. Synapse Trading’s page, last updated 3 July 2026, places both funds at the cheap end of the US spot bitcoin ETF field, well below Grayscale’s GBTC at 1.50%, a figure also confirmed by crypto.news. That single basis point is the only figure in this comparison that all four sources independently corroborate.

What the expense ratio actually costs, worked through

An expense ratio is deducted from the fund’s assets over the course of a year, not billed to the holder directly – which is why each fund’s bitcoin-per-share ratio only ever declines between filings, never rises, as btcetfcalc.com notes in its methodology. According to btcetfcalc.com, ARKB held 0.0003316 BTC per share as of 19 August 2026, per a snapshot of ARK 21Shares’ own daily holdings file. The same site’s snapshot of Bitwise’s own trust holdings page shows BITB holding 0.0005427 BTC per share as of 17 August 2026 – a different ratio because the funds have different share prices, not because one is a better tracker.

crypto.news ran the compounding arithmetic on a flat $100,000 position held for five and ten years, isolating the fee effect from any price movement in bitcoin. Over five years, in an article published 19 August 2026, it calculates BITB’s 0.20% fee eating $996, leaving $99,004, against ARKB’s 0.21% fee eating $1,046, leaving $98,954 – a $50 difference. Over ten years the gap widens only slightly: BITB’s fees total $1,982 against ARKB’s $2,080, a $98 difference. For comparison, the same crypto.news piece gives two different ten-year GBTC fee totals for a $100,000 position: its summary states approximately $14,017, while the body of the article, working through a step-by-step table, puts the figure at $14,093. crypto.news does not reconcile the two numbers within its own article; both are crypto.news’s figures, not this page’s calculation.

Separately, btcetfcalc.com’s own fee calculator, run over a five-year horizon on the investment amount and holding period a user selects, shows ARKB accumulating $105 in fees against BITB’s $100 – a $5 gap. The site’s FAQ frames the same comparison on a $10,000 investment as a difference of roughly $1 per year. Two different tools, two different framings, and both land on the same conclusion: at these fee levels, the dollar difference between ARKB and BITB is close to noise for most account sizes and holding periods.

Custody: no diversification between the two

btcetfcalc.com’s side-by-side table lists Coinbase Custody as the custodian for both ARKB and BITB, and its comparison text states plainly that choosing between the two funds gives an investor no custodial diversification, since both rely on the same custodian. That is a distinction some readers look for when comparing spot bitcoin ETFs – FBTC, for instance, is described by CoinCodex as using Fidelity Digital Assets, and HODL is described by CoinCodex as using Gemini – but it does not separate ARKB from BITB.

The AUM figures do not agree

btcetfcalc.com’s snapshot puts ARKB’s assets under management at roughly $2.5 billion (dated to its 19 August 2026 refresh) and BITB’s at roughly $2.7 billion (dated to its 17 August 2026 refresh) – figures reported by btcetfcalc.com itself, with no independent audit cited. crypto.news’s article, published the same day as the first snapshot date, separately states that BITB’s low-cost positioning has drawn in “over $3.8 billion” in assets, without specifying the date that figure was measured against. The two BITB figures – roughly $2.7 billion and over $3.8 billion – cannot be reconciled from the evidence reviewed for this page; they may reflect different measurement dates, different scopes, or both.

Where spread and tracking error should be, and are not

Three of the four sources reviewed here raise spread as something a reader should check before trading ARKB or BITB. CoinCodex tells readers to compare the spread on the two funds right before placing an order and to watch daily volume, and btcetfcalc.com states that trading volumes and bid-ask spreads are comparable between the two funds. Neither source, nor Synapse Trading, publishes an actual spread figure, on any date, for either fund. The advice to “check the spread” is sound trading practice, but it is not the same as having measured it, and no measurement appears in this set of sources.

Tracking error – the gap between a fund’s price performance and bitcoin’s own price performance, after fees – fares no better. crypto.news reports that tracking error across all 11 US spot bitcoin ETFs ranges from 0.03% to 0.42% annualized, in an article published 19 August 2026, and states that this hidden cost can exceed the fee difference between cheap and mid-priced funds. But that figure is an aggregate range across the entire category. Nothing in the four sources isolates a tracking-error number for ARKB specifically, or for BITB specifically. A reader cannot use this evidence to say which of the two funds tracks bitcoin more tightly, only that the category-wide range exists and that fees alone do not capture the full cost picture.

Tax-loss harvesting: a live legal question, not a settled one

btcetfcalc.com raises the possibility of using ARKB and BITB as a pair for tax-loss harvesting – selling one at a loss and buying the other to maintain bitcoin exposure – because the two hold the same asset with near-identical fee structures. Its own FAQ states that the IRS has not issued explicit guidance on whether two different spot bitcoin ETFs count as “substantially identical” for wash-sale purposes, and that some tax professionals argue they are not, given the different issuers and trust structures. That is a description of an open legal question, not a recommendation, and a reader considering the strategy should consult a tax professional, as btcetfcalc.com itself advises.

What this page does not tell you

This page cannot tell a reader what bid-ask spread they will actually pay buying or selling ARKB or BITB on a given day. None of the four sources reviewed measures it; three of them mention spread only as advice to check it yourself before trading.

This page cannot state a tracking error specific to either fund. The only tracking-error figure available – crypto.news’s 0.03% to 0.42% range, published 19 August 2026 – covers all 11 US spot bitcoin ETFs together and is not broken out by fund.

The bitcoin-per-share ratios cited from btcetfcalc.com are drawn from each issuer’s own holdings disclosure – ARK 21Shares’ daily CSV and Bitwise’s trust holdings page – refreshed on different dates (19 and 17 August 2026 respectively). The AUM figures cited from btcetfcalc.com are reported by the site itself; the evidence reviewed does not show them sourced to those same issuer disclosures, and neither figure is independently audited.

The AUM figures conflict depending on source and date: roughly $2.5 billion and $2.7 billion for ARKB and BITB respectively as of mid-to-late August 2026 per btcetfcalc.com, against “over $3.8 billion” for BITB alone per crypto.news, with no shared measurement date to reconcile the two.

Nothing in this evidence set covers spot ETFs outside the United States, options market depth beyond a yes/no flag, or any performance period beyond the flat-price, fee-only arithmetic that crypto.news ran.