Fees, Holdings & Direct Staking Math

The Decision You Are Actually Making

The question of which crypto ETF to buy is the wrong question for most holders. The question that matters is whether you should use an ETF wrapper at all, or whether direct ownership plus self-custody plus staking earns you more after fees. The answer depends on three numbers: the expense ratio you pay annually, the yield you forfeit by holding an ETF instead of the underlying asset, and the tax wrapper you are using. For holders with more than six months of experience, direct ownership wins in almost every scenario outside of a tax-advantaged retirement account.

If you must use an ETF, the product landscape has clarified significantly since the spot Bitcoin ETFs launched in January 2024. Morgan Stanley Bitcoin Trust (MSBT) charges 0.14% annually, one basis point below Grayscale Bitcoin Mini Trust (BTC) at 0.15%. iShares Bitcoin Trust (IBIT) charges 0.25% but has superior tracking accuracy, meaning the all-in cost may be lower. Grayscale Bitcoin ETF (GBTC), the converted trust, still charges 1.50%. On the Ethereum side, Bitwise Ethereum ETF (ETHW) charges 0.20%, while Grayscale Ethereum Staking ETF (ETHE) charges 2.91% all-in when you include the sponsor fee and staking commission.

Those numbers matter because fees compound. The expense ratio is not a one-time cost. It is a permanent drag on your returns, accruing every year you hold the position, and it determines whether the convenience of an ETF wrapper justifies the loss of yield and control.

Expense Ratios and Tracking Error

MSBT at 0.14% is the lowest-fee Bitcoin ETF in the U.S. market. For a $100,000 position, that costs $140 in the first year, compounding to approximately $1,400 over ten years assuming flat Bitcoin prices. IBIT at 0.25% costs $250 in year one, approximately $2,528 over ten years. GBTC at 1.50% costs $1,500 in year one, approximately $14,017 over ten years.

The difference between IBIT and GBTC over a decade is $11,489 for identical Bitcoin exposure. That spread exists because GBTC converted from a closed-end trust with embedded illiquidity premium and has not reduced its fee to competitive levels. If you hold GBTC in a taxable account, you are paying $11,489 more than necessary over ten years for the same Bitcoin exposure.

Tracking error adds a second layer of cost that expense ratios alone do not capture. Across the 11 U.S. spot Bitcoin ETFs, tracking error ranges from 0.03% to 0.42% annualized. When tracking error is included, IBIT at 0.25% may have the lowest total cost of ownership because BlackRock’s execution and custody mechanics produce less drift between the ETF’s net asset value and Bitcoin’s spot price. That means the listed expense ratio understates the true cost for high-tracking-error products and overstates it for funds with tight execution.

On the Ethereum side, ETHW at 0.20% is the lowest-cost wrapper. ETHA charges 0.25%. ETHE charges 2.50% sponsor fee plus 0.41% staking fee, for a total annual drag of 2.91%. The fee difference between ETHW and ETHE is 2.71% per year. Over ten years, that is $27,100 in foregone returns on a $100,000 position, assuming flat Ethereum prices.

The Yield You Forfeit By Holding an ETF

Ethereum staking yield in 2026 sits at approximately 2.78% base APR. Validators running MEV-Boost capture an additional 0.5-1% from maximal extractable value, meaning realistic all-in yield for a well-operated node is 3.3-3.8%. Solo stakers with 32 ETH capture the full base APR plus MEV. Liquid staking providers like Lido and Rocket Pool take a commission, typically 10%, so a holder using those services receives approximately 90% of gross yield, or 3.0-3.4% net.

Exchange staking is worse. Coinbase takes a 25-35% commission on staking rewards, meaning if the base Ethereum rate is 3.5%, you receive approximately 2.5% after Coinbase’s cut. Binance is more competitive at 10-20%, but availability varies by region.

If you hold ETHW at 0.20%, you receive zero staking yield. The ETF holds raw Ethereum, and U.S. regulations as of early 2026 still prevented most Ethereum ETFs from staking the underlying collateral. BlackRock’s ETHB product is an exception, distributing approximately 82% of staking rewards to investors as monthly cash payments, producing a net yield of approximately 3.1% before the 0.25% expense ratio.

ETHE charges 2.91% all-in and distributes staking yield, but the net return to holders after fees is approximately 0.4-0.9% lower than what a self-custodied holder running a solo validator would capture. That gap widens when you compare it to liquid staking at 10% commission, which delivers 3.0-3.4% net. Over ten years, the yield difference between ETHE and direct staking is $4,000 to $9,000 on a $100,000 position, assuming flat Ethereum prices and stable staking yields.

Bitcoin does not support native proof-of-stake, so there is no staking yield to forfeit. However, direct Bitcoin holders can lend their BTC on platforms that pay interest, typically 2-5% depending on the platform, duration, and counterparty risk. ETF holders cannot lend the shares because the Bitcoin is held in custody by the fund. That means the opportunity cost of holding a Bitcoin ETF instead of direct BTC is whatever yield you could earn by lending or using the asset in a productive way. For holders who never lend and never use their Bitcoin, the cost is zero. For holders who do, the cost is 2-5% annually.

The Tax Argument For ETFs Falls Apart In Most Cases

The usual argument for ETFs is tax simplicity. Brokers handle cost basis reporting automatically. Starting with the 2026 tax year, the IRS 1099-DA regime requires custodial brokers to report both gross proceeds and cost basis, so Bitcoin held through an ETF or on a U.S. exchange is fully legible to the tax authority by design. That makes ETF reporting slightly easier than self-custody reporting, but the gap is smaller than it was in 2023.

The tax argument weakens further when you consider that most tax professionals treat spot Bitcoin ETF shares as securities subject to wash sale rules, while direct BTC, classified as property under IRS Notice 2014-21, is currently exempt. That means direct holders can sell at a loss and immediately repurchase to harvest tax losses, while ETF holders must wait 30 days. For active tax-loss harvesters, the wash sale exemption on direct crypto is worth more than the reporting convenience of an ETF.

If you hold crypto in an IRA or 401(k), the tax argument reverses. In those accounts, you cannot hold direct crypto unless your custodian supports it, which most do not. Fidelity and a handful of others allow direct crypto in IRAs, but the majority of retirement accounts only allow ETFs or other securities. In that narrow case, the ETF is the only way to gain Bitcoin or Ethereum exposure inside the tax wrapper, and the expense ratio is the price you pay for access.

Estate Planning and Operational Risk

If you die holding an ETF, your heirs receive the shares through your brokerage account with standard estate procedures. If you die holding direct crypto, your heirs need the private keys, the hardware wallet PIN, and the knowledge to access the funds. If you did not document that information, the crypto is unrecoverable.

That operational risk is real. It is also solvable. Multisignature wallets, key-splitting protocols, and estate planning services designed for crypto holders now exist. The cost of implementing those systems is approximately $500 to $2,000 upfront, depending on the service. For holders with more than $50,000 in crypto, that one-time cost is less than the expense ratio you pay in year two of holding an ETF.

The Futures ETF Trap

ProShares Bitcoin Strategy ETF (BITO) charges approximately 0.95% annually and holds Bitcoin futures contracts rolled monthly. When futures trade in contango, which is typical for Bitcoin, rolling costs bleed 6-8% per year of return. That cost does not appear in the expense ratio. It appears as tracking error between the ETF’s returns and Bitcoin’s spot price.

Buying a futures Bitcoin ETF when a spot equivalent exists is, in almost every retail scenario, paying a hidden 6-8% per year for no benefit. The only use case for a futures-based crypto ETF is if you need 60/40 tax treatment under Section 1256, which benefits active traders. Long-term holders should never hold a futures ETF when a spot ETF is available.

Custody Concentration and Regulatory Risk

Coinbase Custody holds Bitcoin for 9 of the 13 U.S. spot Bitcoin ETFs. Gemini serves as custodian for VanEck’s Bitcoin ETF. That means if Coinbase experiences a custody failure, operational issue, or regulatory action, the majority of U.S. Bitcoin ETF holders are exposed to the same risk simultaneously. Self-custody eliminates that concentration risk entirely.

ETF holders also face regulatory risk that direct holders do not. If the SEC reverses its approval of spot ETFs or imposes new restrictions, ETF holders are locked into whatever the fund’s board decides to do. Direct holders can move their Bitcoin across borders, across wallets, or into cold storage without asking permission.

Breakeven Holding Periods

For Bitcoin, the breakeven calculation is straightforward. A hardware wallet costs $79 to $219 one-time. IBIT charges 0.25% annually. On a $100,000 position, IBIT costs $250 in year one. After one year, the hardware wallet has already paid for itself. By year ten, the hardware wallet holder has saved $2,378 relative to IBIT, and $13,867 relative to GBTC.

For Ethereum, the math includes staking yield. If you can earn 3.3-3.8% all-in via solo staking or 3.0-3.4% via liquid staking at 10% commission, and you hold ETHW at 0.20% with zero staking yield, you forfeit 3.0-3.8% annually. On a $100,000 position, that is $3,000 to $3,800 per year. Over ten years, that is $30,000 to $38,000 in foregone yield.

If you hold ETHE at 2.91% all-in with staking yield distributed, the comparison is closer. ETHE delivers approximately 0.4-0.9% less than solo staking, which costs $400 to $900 per year on a $100,000 position, or $4,000 to $9,000 over ten years. A hardware wallet plus the operational cost of running a validator node costs approximately $150 one-time for the wallet plus $100-$200 per year in electricity and maintenance. That means the breakeven period for direct staking versus ETHE is approximately six months to one year, depending on yield assumptions.

Who Each Option Is Right For

ETFs make sense for:

• Holders using an IRA or 401(k) where direct crypto custody is unavailable
• Holders with less than $5,000 invested, where hardware wallet overhead is disproportionate
• Holders who cannot manage private keys, PIN codes, and seed phrase backups
• Holders who need exposure for less than six months and do not want to set up custody infrastructure

Direct ownership makes sense for:

• Holders with more than $5,000 invested
• Holders with a time horizon longer than 12 months
• Ethereum holders who want to capture staking yield
• Holders who value tax-loss harvesting without wash sale restrictions
• Holders who want to eliminate custodial concentration risk
• Holders who want the ability to lend, use, or move assets without fund board approval

The Recommendation

If you are using a tax-advantaged retirement account and your custodian does not support direct crypto, use MSBT at 0.14% for Bitcoin exposure and ETHW at 0.20% for Ethereum exposure. Those are the lowest-cost wrappers available.

If you are holding in a taxable account with more than $5,000 and a time horizon longer than 12 months, buy direct. Use a hardware wallet. Set up a validator or use a liquid staking provider at 10% commission. The yield you capture and the fees you avoid will exceed $10,000 over ten years on a $100,000 position. That is the real cost of convenience.

If you are holding Ethereum and you want staking yield inside an ETF wrapper, ETHB at 0.25% distributes approximately 3.1% net. That is competitive with exchange staking at 25-35% commission, but it is still 0.2-0.7% lower than liquid staking at 10% commission. The gap is small enough that the convenience of the ETF may be worth it if you are holding in an IRA, but not if you are holding in a taxable account where you could stake directly.

The Takeaway

Expense ratios are permanent drags. GBTC at 1.50% costs $14,017 over ten years on a $100,000 position. IBIT at 0.25% costs $2,528. Direct ownership costs $150 one-time for a hardware wallet. On Ethereum, the yield gap between direct staking and ETHW is $30,000 to $38,000 over ten years. ETFs solve a custody problem for retirement accounts and short-term holders. For everyone else, they are expensive wrappers around assets you could hold more efficiently yourself. The convenience is real. The cost is higher than most holders realize. The breakeven period is shorter than one year.

Frequently Asked Questions

What is the lowest-fee Bitcoin ETF available in 2026?

Morgan Stanley Bitcoin Trust (MSBT) charges 0.14% annually, one basis point below Grayscale Bitcoin Mini Trust at 0.15%. iShares Bitcoin Trust (IBIT) charges 0.25% but may have lower total cost of ownership when tracking error is included, because BlackRock’s execution produces less drift between the ETF’s net asset value and Bitcoin’s spot price. Avoid Grayscale Bitcoin ETF (GBTC) at 1.50%, which costs $11,489 more than IBIT over ten years on a $100,000 position.

Do Ethereum ETFs allow me to earn staking rewards?

Most U.S. Ethereum ETFs hold raw Ethereum and distribute zero staking yield. Bitwise Ethereum ETF (ETHW) at 0.20% and iShares Ethereum Trust (ETHA) at 0.25% do not stake. BlackRock’s ETHB distributes approximately 82% of staking rewards as monthly cash payments, producing net yield of approximately 3.1% before the 0.25% expense ratio. Grayscale Ethereum Staking ETF (ETHE) charges 2.91% all-in and distributes staking yield, but delivers 0.4-0.9% less than solo staking after fees.

When does direct crypto ownership beat holding an ETF?

Direct ownership beats ETF wrappers for holders with more than $5,000 invested, a time horizon longer than 12 months, and basic custody competence. A hardware wallet costs $79-$219 one-time. IBIT at 0.25% costs $2,528 over ten years on a $100,000 position. For Ethereum, the yield gap between direct staking at 3.3-3.8% and ETHW at 0% is $30,000-$38,000 over ten years. The breakeven period is less than one year unless you are holding in an IRA where direct custody is unavailable.

Are futures-based Bitcoin ETFs a good option?

No. ProShares Bitcoin Strategy ETF (BITO) charges 0.95% and holds Bitcoin futures rolled monthly. When futures trade in contango, rolling costs bleed 6-8% per year in hidden tracking error. That cost does not appear in the expense ratio but shows up as underperformance versus Bitcoin’s spot price. Buying a futures ETF when spot ETFs exist is paying 6-8% annually for no benefit. The only use case is if you need Section 1256 tax treatment, which benefits active traders, not long-term holders.

What is the main risk of holding crypto through an ETF?

Custodial concentration risk and regulatory exposure. Coinbase Custody holds Bitcoin for 9 of 13 U.S. spot Bitcoin ETFs. If Coinbase experiences a custody failure or regulatory action, the majority of ETF holders are exposed simultaneously. ETF holders also cannot move assets during off-market hours, cannot lend or stake their holdings unless the fund permits it, and are locked into whatever the fund’s board decides if SEC rules change. Self-custody eliminates those risks entirely.