The verdict is in. Fidelity is turning its Ethereum ETF into a yield-generating machine. But the fine print reveals a centralized staking architecture that retail investors should scrutinize.
Context: Why Now?
The catalyst is the IRS safe harbor rule from November 2025. It permits qualifying crypto trusts to stake without losing grantor trust status—provided they distribute net rewards at least quarterly. Fidelity’s amendment to its FETH fund is a direct response. Grayscale already activated staking in October 2025 and paid its first distribution in January 2026. 21Shares followed. BlackRock chose a different path: a standalone staking ETF in March 2026.
Fidelity’s move is not a technical breakthrough. It is financial engineering. The underlying staking mechanism—ETH proof-of-stake—has been running for years. The innovation is packaging it into a SEC-registered, 1940 Act ETF that can be held in retirement accounts. The ledger remembers what the market forgets: this is a product of regulation, not code.
Core: The Staking Architecture
Fidelity proposes to stake up to 100% of its ETH holdings, currently $903 million. There is no minimum staking requirement. The fund will reserve assets for redemptions, fees, and liquidity.
The staking is executed through a two-layer structure. Three custodians—Anchorage Digital Bank, BitGo Bank & Trust, and Fidelity Digital Assets—hold the ETH and arrange staking. Three node operators—Blockdaemon, Figment, and Galaxy—run the validator infrastructure. This is a deliberate separation of duties: custody and validation.
The fee structure is fixed. 15% of all staking rewards is allocated to the sponsor, custodians, and node operators. The remaining 85% stays in the trust. After paying fund expenses, the net is converted to USD and distributed quarterly to shareholders. Distributions are not guaranteed. If liabilities exceed rewards, the fund can suspend payments.
This is where the technical analysis gets interesting. The reward rate for ETH staking is currently around 3-5% annually. Assuming a 3.5% yield on $903 million, gross annual rewards are approximately $31.6 million. After the 15% fee, $26.9 million remains. After the 0.25% management fee ($2.26 million), about $24.6 million is available for distribution. That translates to roughly a 2.7% net yield on fund assets. Compare this to self-staking: a solo validator earns the full 3.5% minus hardware and operational costs. The ETF offers convenience at a cost.
But the real risk is not yield compression. It is centralization. The three custodians represent a single point of trust failure. True, Fidelity uses three different entities to avoid a single point of collapse. But these are still institutional custodians, not decentralized smart contracts. Power lies in the code, not the community. Here, the code is replaced by a legal agreement. Slashing risk is disclosed but not quantified. The custodian’s liability for node operator misconduct is limited. This is a gap in the risk model.
Based on my experience auditing the 2020 Aave governance shift, I saw how centralized control over key parameters can undermine user expectations. In Aave, the DAO’s voting power concentrated in a few whales. Here, Fidelity’s sponsor controls the staking percentage, fee structure, and distribution schedule. Shareholders have no direct vote. Governance is theater. Execution is reality.
Contrarian: The Unreported Blind Spot
The market is bullish on staking ETFs. But the contrarian angle is that this product may actually increase systemic risk for Ethereum. Why? Because the ETF structure locks ETH into a centralized staking pool controlled by a few node operators. Blockdaemon, Figment, and Galaxy are also major node operators for Lido, Rocket Pool, and other protocols. If Fidelity’s ETF adds significant staking volume to these operators, it further concentrates Ethereum’s validator set.
Consider the numbers. FETH’s $903 million at current ETH prices (~$3,000) represents about 301,000 ETH. If fully staked, that is roughly 9,400 validators. That is a small fraction of the total validator set (over 1 million). But the trend is concerning. As more ETFs adopt staking, the same custodians and node operators accumulate control. This is the opposite of the decentralization ethos.
Another hidden risk: the “no minimum staking” clause. In practice, Fidelity can choose to stake less than 100%. During market stress, when redemptions spike, the sponsor may reduce staking to maintain liquidity. This would lower the yield for remaining shareholders. The fund’s prospectus allows this. The market is focused on the upside of yield, but the downside of variable staking is not priced in.
Finally, the competitive landscape. Fidelity’s 0.25% management fee is low, but Grayscale’s ETHE charges 2.5%. That means Grayscale’s staking yield is heavily diluted. Fidelity’s product will likely attract flows from Grayscale, putting pressure on Grayscale to cut fees. But BlackRock’s independent staking ETF offers a choice: investors can buy the non-staking ETFA or the staking version. Fidelity’s approach forces all ETF holders to accept staking. This could alienate investors who prefer pure ETH exposure without the complexity.
Takeaway: The Next Watch
The real test will be the first distribution. If Fidelity pays out less than expected due to slashing or operational costs, the narrative will shift from “yield enhancement” to “yield disappointment.” The market is pricing in a smooth launch. But the ledger remembers the 2022 Terra collapse, where centralized staking and yield promises unraveled. This time, the structure is more robust, but the risks are still human.
Watch for two things: the SEC’s approval timeline and the actual staking percentage Fidelity adopts. If it stays below 50% initially, that signals caution. If it goes to 100%, that is a bet on low volatility and low redemptions. Either way, the ETF is no longer a simple ETH proxy. It is a leveraged play on the staking infrastructure’s reliability.
Power lies in the code, not the community. But here, the code is the prospectus. And the community is the shareholders who cannot vote. The ledger remembers what the market forgets: centralized yield is a promise that can be broken.