The New Ethereum Staking ETFs Pay You Less Than Lido Does

You own ether and you want it to earn something. As of this month you have two live routes that did not both exist a year ago. You can buy shares in an exchange-traded fund that stakes the ETH on your behalf and sends you cash, or you can deposit ETH into a liquid staking protocol like Lido and hold a token that grows in value while you keep control of it.

The short answer: the funds pay less, but not by as much as the gap in the headline suggests, and the difference is smaller than the difference in what you are exposed to. Everstake's April 2026 survey of the category put net distributions to ETF shareholders at 1.9% to 2.6% a year against gross network rewards of 3.1% to 3.3%. Lido's stETH netted roughly 2.16% after its 10% protocol fee in CoinLaw's May 2026 data. A mid-table ETF and Lido land within a rounding error of each other. Fidelity's new fund, projected at 1.5% to 2%, does not.

Two definitions before the numbers. Staking means locking ETH with a validator, a computer that helps confirm Ethereum's transactions, and collecting a share of the network's rewards for it. Liquid staking means doing that through a pooled protocol that hands you a tradable token representing your stake, so your ETH is working but not frozen.

One number to keep in mind while reading the rest: every product in that table stakes into the same Ethereum consensus layer and earns the same base rate. Nobody has a yield edge. All that varies is how many hands the reward passes through on its way to you.

Grayscale went first. It switched on staking in October 2025 and on 6 January 2026 paid out $0.083178 per share, about $9.4 million in total, covering rewards earned from 6 October to 31 December. That was the first time a US spot crypto fund had distributed staking rewards to shareholders. BlackRock followed on 12 March 2026 with iShares Staked Ethereum Trust (ETHB), a separate product rather than a change to its existing ETHA fund, and pulled in $254 million in its first week.

Fidelity is the one moving now. On 11 August 2026 it filed an amended registration statement to add staking to its Ethereum Fund, then quantified $898 million in net assets. A prospectus dated 21 August grants authority to stake up to 100% of holdings with no minimum. The staking cannot begin until the SEC declares the amendment effective. Franklin Templeton, Invesco, 21Shares and VanEck have amendments of their own in the queue.

The fee mechanics are where the yield goes. Fidelity's sponsor agreement, filed with the SEC on 10 August, says the trust keeps 85% of gross staking rewards and pays 15% to the sponsor, to be shared with custodians and node operators. That 15% comes off before the fund's 0.25% expense ratio does, and the expense ratio is paid out of staking income before any dividend reaches you. ETHB passes through 82%, with Coinbase taking 10% of rewards as primary custodian and staking provider, falling to 6% if the fund reaches $20 billion. Grayscale's Mini ETF is the most generous in the category at 94% pass-through on a 0.15% fee.

Lido is not a close race. As of CoinLaw's May 2026 data it held $18.7 billion in ETH, 8,863,785 coins across 621,094 staker addresses and more than 900 node operators, amounting to 22.7% of all staked ETH and 62.3% of the liquid staking market specifically. It has paid $2.446 billion in cumulative rewards since 2020. The mechanics are plain: deposit ETH, receive stETH, watch the balance grow, swap back whenever you want. The protocol takes 10% of rewards, split between node operators and the Lido DAO.

Rocket Pool is the decentralised alternative, and it is where the data gets messy. CoinLaw lists rETH at a 2.00% APR with a 14% fee. Passive Yield Lab's 2026 comparison puts it at roughly 3.46% net. Those cannot both be right, and the difference is bigger than the entire gap between ETFs and liquid staking. The likeliest explanation is timing and methodology, since rETH accrues value through a rising exchange rate rather than a growing balance, and a short measurement window can flatter or flatten it. Until someone publishes a like-for-like trailing-12-month figure, treat 3.46% as unverified. Rocket Pool is also shrinking: its $922.87 million TVL was down 22.75% over the trailing 30 days at the time of that data.

What is not in dispute is that net yields across liquid staking tokens cluster tightly between about 2.0% and 2.2%, whatever the headline APR says. Coinbase's cbETH shows 2.83% gross but a 25% fee, netting about 2.12%. Frax, Binance and StakeWise all land in the same band. Same network, same rewards, different cuts.

The fee stack, not the fee. An ETF charges you twice on the same reward stream. Fidelity's 15% comes off the gross, then the 0.25% expense ratio comes off what is left, and only the surplus becomes a dividend. Lido charges 10% once. BlockEden's January 2026 analysis estimated ETF investors capture 60% to 70% of raw staking yield after everything. That is the whole story of the spread.

Getting out is not symmetrical. Selling ETF shares on an exchange is instant during market hours, but that is the secondary market, not your ETH. If a fund faces heavy redemptions while heavily staked, its validators must exit Ethereum's active set and clear a network-imposed waiting period first. CryptoSlate's read of Fidelity's prospectus found no guaranteed exit timeline for FETH and a four-step ladder of contingencies, including extending settlement and delivering "cash in place of some or all of the crypto owed." Fidelity's Solana fund was 99.64% staked as of 30 June 2026, which suggests how close to fully deployed these funds intend to run. With stETH you swap on-chain in seconds, assuming the pool has depth.

You are swapping one counterparty for another, not removing one. The ETF route replaces smart contract risk with custodian and issuer risk, plus the absence of Investment Company Act protections that Grayscale's own filings disclose. Neither route is the safe one. They fail differently.

If you already self-custody your ETH and are comfortable signing transactions, Lido at roughly 2.16% net is simply the cheaper product. You pay one fee instead of two, your rewards compound in ETH rather than arriving as cash you have to redeploy, and you can exit at 3am on a Sunday. The trade is that you are underwriting a smart contract with $18.7 billion in it.

If your ETH exposure lives in a brokerage account and you have no intention of running a wallet, the honest framing is that you are paying 30 to 60 basis points a year for someone else to handle custody, validators and paperwork. Grayscale's Mini ETF at 0.15% with 94% pass-through is the least expensive way to buy that convenience. Fidelity's projected 1.5% to 2% is the most expensive of the named products, and it is the one you should look hardest at once actual distributions start.

If you are shopping for the highest number, wait for better data. Rocket Pool's reported 3.46% would beat every ETF in the category comfortably, and it may well be correct, but a single blog figure contradicted by a statistics aggregator is not a basis for moving money. What would settle it is a trailing-12-month rETH/ETH exchange rate comparison, which anyone can compute from on-chain data.

And if the answer is that you want yield without the volatility, neither of these is that. Both pay you in an asset whose price moves far more in a week than either yields in a year.