Liquid Staking in 2026: Yield, Liquidity and the One-Third Problem

Ethereum’s staking market crossed a strange threshold this summer. In late July 2026, Lido, the largest liquid staking protocol, began moving more than 8 million ETH, worth roughly $16.5 billion, onto a new generation of higher-capacity validators, an overhaul expected to cut Ethereum’s total validator count by about a third and trim attestation traffic by close to 29% per epoch, according to CoinDesk. A single protocol was not merely staking a large slice of the network; it was rebuilding part of the network’s plumbing.

With ETH trading near $2,491 and about 33.98% of all supply now staked, a record, per CoinGecko and Coinpedia, liquid staking has become the most important primitive in Ethereum’s on-chain economy and, at the same time, its single largest governance question. This guide explains what liquid staking is, how a token like stETH is created and redeemed, where the yield really comes from, and the risks that matter now. It also takes seriously the problem most explainers skip past: liquid staking solved Ethereum’s liquidity problem so well that it created a concentration-of-power problem, and in 2026 the guardrails are being tested in the open.

What liquid staking actually solves

Staking is how Ethereum secures itself. Since the network switched to proof of stake at the Merge in September 2022, validators lock up ETH as collateral, propose and attest to blocks, and earn rewards for honest work or forfeit part of their stake for misbehavior. The design is elegant, but it carries a cost that is easy to miss: staked ETH is capital pulled out of circulation. While it secures the chain, you cannot sell it, spend it, or post it as collateral.

Liquid staking removes that trade-off. When you stake through a liquid staking protocol, you receive a token that represents your staked ETH plus the rewards it earns. That token, stETH from Lido or rETH from Rocket Pool, is freely transferable. You can hold it, trade it, lend it, or use it as collateral in other protocols while the underlying ETH keeps validating and earning. In one sentence: you get the yield and keep the liquidity. That combination is why liquid staking, rather than solo or plain custodial staking, has become the largest single category of staked ETH.

It helps to be precise about what the token is and is not. A liquid staking token is a claim on staked ETH plus its accrued rewards, not a promise from a company to pay you back. Its value comes from the ETH securing the chain and from the protocol’s rules, enforced by smart contracts, for how rewards flow and how you exit. That is a different risk profile from a savings account or a lending desk: there is no balance sheet behind it, only code, validators and a market. Understanding that distinction is the difference between using liquid staking well and being surprised by it.

How staking worked before the liquid token

To see why liquid staking caught on, start with the friction it removed. Running your own validator requires exactly 32 ETH, about $80,000 at current prices, plus a machine that stays online continuously and the technical skill to avoid penalties. Make a mistake that harms the network, such as double-signing, and a portion of your stake can be slashed. Fall offline and you leak small inactivity penalties. For most holders, that is too much capital and too much operational risk to take on directly.

Pooled staking answers the capital problem first. Instead of needing a full 32 ETH, you can stake any amount, even a fraction of one ETH, because the protocol combines many small deposits into whole validators run by professional operators. That alone widened staking from a few well-capitalized technicians to anyone with a wallet. Liquid staking then adds the second half of the answer, a transferable token, so the pooled position is not just accessible but also usable elsewhere while it earns.

There are also queues. Ethereum limits how quickly validators can join or leave, so both entry and exit are rate-limited by a per-epoch churn cap. In mid-August 2026, roughly 2.4 million ETH sat in the entry queue, a wait of about 39 days, driven by exchange-traded fund and corporate-treasury demand for yield, per ValidatorQueue. The exit side swings the other way and has at times cleared to near zero. Liquid staking sidesteps both queues entirely: buying or selling stETH on the open market never touches the validator queue, because you are trading a token, not activating or exiting a validator. That instant liquidity is the product.

How a liquid staking token is minted and redeemed

The lifecycle is simpler than the jargon suggests. You deposit ETH into the protocol’s smart contract. The protocol pools deposits and allocates them, in 32-ETH units or larger post-Pectra batches, to professional node operators who run the validators. In return, you receive the liquid staking token, roughly one for each ETH you deposit. As the validators earn, the value represented by your token grows.

Getting out works two ways. You can sell the token on the open market at whatever price it trades for, which is instant. Or you can redeem it through the protocol: you burn the token, the protocol exits enough validator balance to cover you, and you receive ETH once the exit queue clears. Native redemption only became possible after Ethereum’s Shapella upgrade enabled staking withdrawals in April 2023. Before that, liquid staking tokens could be sold but not redeemed, a distinction that turned out to matter enormously in 2022. The existence of a 1:1 redemption path is also what keeps the token’s market price anchored near the price of ETH: if stETH trades below its redeemable value, arbitrageurs buy it cheap and redeem it for full value, closing the gap.

Two ways out means two prices, and the difference is worth understanding. The redemption value is what the protocol will give you in ETH if you burn the token and wait for the exit queue; it only moves upward as rewards accrue. The market price is whatever a buyer will pay right now, and in calm conditions the two sit almost on top of each other. They separate only when everyone wants out at once and the exit queue is too slow to help, which is precisely the scenario that played out before native withdrawals existed. Who runs the validators matters too: reputable protocols spread deposits across a vetted set of operators so that no single failure drains the pool.

Rebasing versus reward-bearing: stETH, wstETH and rETH

Liquid staking tokens accrue rewards in one of two ways, and the difference matters for how you use them.

The first is rebasing. Lido’s stETH uses this model: your token balance increases a little every day as rewards land, and one stETH always aims to equal one ETH. It is intuitive to read in a wallet, since the number simply grows. The drawback is that a balance which changes daily is awkward for many DeFi contracts and complicates tax accounting in some jurisdictions.

The second is value-accruing, sometimes called reward-bearing. Here the token count stays fixed and each token becomes worth more ETH over time. Lido offers wstETH, a wrapped version of stETH that holds this shape: one wstETH is worth roughly 1.24 ETH today and that ratio only climbs. Rocket Pool’s rETH works the same way, trading at about 1.17 ETH per token, per CoinGecko, as do the custodial tokens from large exchanges. Value-accruing tokens are the ones DeFi protocols prefer, because a fixed balance is far easier to price and lend against.

Where the yield comes from, and why it fell to 2.7%

Staking yield is not a single number handed down by the protocol; it is a stack. The base layer is consensus issuance, the new ETH the network mints to reward validators for proposing and attesting to blocks. On top of that sit execution-layer priority fees, the tips users pay to have their transactions included, and MEV, the extra value a validator can capture from how transactions are ordered within a block. Validators running MEV-Boost typically add somewhere between half a point and a full point of yield from that last source.

The headline number has fallen. Ethereum’s base staking APR is around 2.7% in 2026, down from a peak above 5% in mid-2023, per Coinpedia. The reason is mechanical: issuance rewards are shared across all validators, so the more ETH is staked, the thinner each validator’s slice. With staking at a record share of supply, per-validator rewards have compressed to a three-year low even as the network has never been more secure. Liquid staking providers then take a cut of the rewards before passing them on. Lido, for example, charges a 10% fee split evenly between node operators and its treasury, per Lido, so the net yield a stETH holder sees sits a little below the gross validator rate.

The composition matters as much as the total. Issuance is newly printed ETH, which rewards you in nominal terms but dilutes every holder who is not staking; priority fees and MEV, by contrast, are real income paid by users of the network. As issuance thins, that second bucket becomes a larger share of what a validator actually earns, which is why providers compete on how efficiently they capture MEV rather than on the base rate alone. For a holder, the practical read is simple: compare net yields after fees, and treat any advertised rate well above the network base as a sign that leverage or restaking, not magic, is doing the work.

The market map: who runs liquid staking in 2026

One protocol dominates. Lido’s stETH carries a market capitalization near $24 billion across about 9.6 million tokens, per CoinGecko, which works out to roughly 23% of all staked ETH and around 62% of the narrower liquid staking segment, with Binance’s wBETH second at about 25% of that segment, according to figures reported by Bitcoin.com. Everyone else competes for what is left.

Rocket Pool’s rETH, at roughly $930 million, is the largest permissionless alternative: anyone can run a Rocket Pool node, and its Saturn I upgrade in February 2026 cut the minimum operator bond from 8 ETH to 4 ETH and introduced megapools that consolidate validators, as Crypto Briefing reported. Frax offers a two-token design, frxETH for liquidity and sfrxETH for concentrated yield. Coinbase’s cbETH and Binance’s wBETH are custodial: an exchange holds the keys and runs the validators, which is simpler for users but reintroduces counterparty risk. Beyond Ethereum, Solana has its own fast-growing liquid staking market led by JitoSOL, though it solves a different problem, since Solana never had a 32-ETH barrier to begin with.

The competitive frontier in 2026 is less about retail and more about institutions. Lido shipped a modular V3 design built around stVaults, segregated vaults with dedicated validators that let a large holder set its own operator, fee and reporting terms rather than pooling into the shared stETH. The pull is real money: SharpLink, a Nasdaq-listed treasury company, committed to stake roughly $200 million of ETH through Lido and hold the resulting wstETH, per crypto.news. That institutional demand is a big part of why Lido’s absolute size keeps growing even as its percentage share of all staked ETH slowly declines.

Liquid staking is not restaking

The two get confused constantly, so it is worth drawing the line clearly. Liquid staking does one job: it secures Ethereum and gives you a receipt token. Restaking, pioneered by EigenLayer, takes staked ETH or a liquid staking token and re-pledges it to secure additional services, from oracles to bridges to data-availability layers, in exchange for extra yield and extra slashing risk. Liquid restaking tokens such as weETH and rsETH are the receipts for that second layer, stacked on top of the first.

The distinction became sharper in 2026, when the restaking boom cooled and several issuers walked back their exposure; ether.fi, for one, moved to unwind restaking from its flagship token. Think of liquid staking as the foundation and restaking as an optional, riskier floor built above it. For most holders, a plain liquid staking token is the conservative choice, and the exotic yields, along with most of the danger, live one level up.

What a liquid staking token is really for: collateral

Yield is only half the story. The reason stETH and wstETH matter to the rest of DeFi is that they became prime collateral. Because a liquid staking token is nearly as safe and liquid as ETH itself but also pays a yield, lending markets treat it as a reserve asset. On Aave, wstETH is one of the largest collateral holdings, and its correlation mode allows very high loan-to-value ratios against ETH.

That opens a strategy called looping: deposit wstETH, borrow ETH against it, buy more wstETH, and redeposit, repeating until the base yield is amplified by leverage. Done carefully it turns a 2.7% base into high-single-digit returns; done carelessly it is a liquidation waiting to happen. The real danger is not the token’s price but the oracle that reports it. In March 2026, a safeguard on Aave meant to smooth wstETH’s value misfired on a stale timestamp, briefly marking the asset about 2.85% below its true worth and triggering roughly $27 million in avoidable liquidations across 34 accounts before it was corrected; there was no actual depeg and no bad debt, and the risk manager pledged full reimbursement, as The Block documented. The episode is a reminder that when a staking token becomes systemic collateral, its plumbing becomes systemic risk, a theme we traced in the 2026 reckoning over DeFi lending.

Collateral was only the start. Protocols such as Pendle let holders split a liquid staking token into its principal and its future yield and trade each separately, so one user can lock in a fixed rate while another speculates on where staking yields go next. Others wrap stETH into structured products, stablecoin backing and delta-neutral strategies. Each layer makes the token more useful and more deeply embedded, which is the point, but it also means a problem in one liquid staking token no longer stays in one place. The more of DeFi that rests on stETH, the more a wobble there radiates outward.

The one-third problem

Here is the tension liquid staking created. Ethereum’s consensus has hard safety thresholds tied to how much stake any single actor controls. An entity commanding one-third of all staked ETH can prevent the chain from finalizing blocks; one-half opens the door to certain forms of censorship; two-thirds could, in the worst case, finalize an invalid chain. Ethereum researcher Danny Ryan laid out these thresholds in a widely cited note years ago, warning that a single staking pool crossing them would put the network’s core guarantees at risk.

Lido’s roughly 23% share of all staked ETH sits below the one-third line, but not comfortably below it, and it has drifted nearer a third before. That is why the concern keeps resurfacing. Vitalik Buterin has repeatedly named staking centralization as “one of the biggest risks” facing Ethereum, listing it among the open problems in the research agenda he calls the Scourge, as The Block reported. The July consolidation drove the point home in a new way: when one protocol can reshape the validator set by moving $16.5 billion at once, its governance is no longer just its own business. It is the network’s.

Lido has faced this question before, and its answer at the time made the current guardrails necessary. In mid-2022, with the protocol already near a third of all staked ETH, the DAO put a proposal to self-limit its share to a vote. More than 99% of voting power chose to stay uncapped, per The Block. The reasoning was that an artificial cap would simply push stakers toward less accountable providers without fixing the underlying incentive. Tellingly, Lido flagged even then that if it would not limit its size, it would instead build structural checks, and dual governance is the promise from that vote finally delivered.

Lido’s dual governance: the check stakers voted for

For most of Lido’s history there was a mismatch at the heart of its governance. Decisions were made by holders of LDO, the protocol’s governance token, while the people whose money was actually at stake, the stETH holders, had no direct vote. If LDO holders ever approved something that harmed stakers or the network, stakers could complain but not block it.

Dual governance, which Lido’s DAO approved in June 2025 by 53.6 million LDO in favor against barely more than a single token opposed, per The Block, closes that gap by inserting a timelock between a decision and its execution and handing stETH holders a brake. It runs in two stages. If dissenting stETH holders escrow at least 1% of staked ETH, execution of the contested proposal is delayed by five days, and the delay grows as more stake piles in. If the escrow reaches 10% of staked ETH, the protocol enters a rage-quit state: the proposal freezes and objectors can withdraw their ETH before the change ever takes effect, as CoinDesk and Lido’s own explainer describe. In other words, stakers cannot outvote LDO holders, but they can escape before being harmed, and the credible threat of a mass exit is itself the discipline.

Lido paired the change with tighter operator accountability. Under its Curated Module v2, all 34 of its existing node operators are posting locked ETH bonds for the first time, replacing a reputation-only system with real financial skin in the game, per crypto.news. Dual governance is one of the clearest answers yet to the worry that a dominant liquid staking protocol is a single point of failure, and it echoes the broader debate over governance attacks, where the vote itself becomes the exploit.

DVT and the decentralization playbook

Governance is one lever; operations are the other. Even if Lido stays large, the actual work of validating does not have to sit on a handful of machines. Distributed validator technology, built by teams such as Obol and SSV, splits a single validator’s signing duties across several independent operators, so no one operator can act alone, go offline, or get the whole validator slashed. It turns each validator into a small committee.

Lido has been routing a growing share of new validators through a Simple DVT module that uses exactly this design, and the approach is spreading across the sector as the standard answer to operator concentration. If the one-third problem is about who controls the stake, DVT is about making sure that control is never truly held by one party in practice. Our deep dive on SSV, distributed validators and the restaking unwind covers how the technology is being deployed at scale.

The two fixes work in tandem. Dual governance limits what a dominant protocol can do to its own stakers; distributed validators limit what any operator inside it can do to the network. Neither shrinks Lido, and that is deliberate, because the sector’s researchers have largely concluded that the durable answer to concentration is accountability and distribution rather than a hard ceiling nobody can enforce. Whether that bet holds is the experiment now running in public.

The 2022 depeg and what “liquid” really means

Liquid does not mean risk-free, and the clearest lesson came in 2022. When the Terra stablecoin collapsed in May 2022 and dragged much of crypto down with it, stETH came under pressure. Three Arrows Capital pulled roughly $400 million, about 128,000 stETH and 73,000 ETH, out of the main stETH/ETH liquidity pool on Curve in a single transaction on 12 May 2022, thinning the exit route just as everyone wanted out. Celsius, meanwhile, had pledged around $400 million of stETH on Aave. stETH slipped from near parity to about 93 to 95 cents on the dollar, as CoinDesk reported at the time, citing Nansen.

Crucially, there was no hack and no failure of the staking itself. It was a pure liquidity and confidence event, made worse by one fact: in 2022, stETH was not yet redeemable, because staking withdrawals did not exist until the Shapella upgrade in April 2023. The only way out was to sell into a draining pool, so the price could wander from the value of the underlying ETH. Today the picture is different. Native redemption gives the token a hard backstop the 2022 market never had. A liquid staking token can still trade below par in a panic, but the redemption path means the gap now has a floor and a reason to close.

The human cost of that episode was not the discount itself, which arbitrage eventually closed, but the leverage stacked on top of it. Both Three Arrows Capital and Celsius collapsed into bankruptcy weeks later, and their forced stETH selling turned a manageable discount into a self-reinforcing spiral. The lesson that stuck was not to fear stETH, but to respect what leverage does to any asset when everyone reaches for the exit at once.

How the SEC learned to stop worrying about stETH

For years the biggest question hanging over liquid staking in the United States was legal, not technical. In February 2023, the SEC settled with Kraken for $30 million and forced it to shut its staking-as-a-service program for US customers, treating the offering as an unregistered securities product, per the SEC. For a while, every staking-linked token in the country carried that shadow.

The posture reversed in 2025. In May, the SEC’s Division of Corporation Finance stated that protocol staking, whether solo, delegated or custodial, is not itself a securities transaction, and Commissioner Hester Peirce underlined the point in a companion statement she titled “Providing Security is not a ‘Security’”. Then, on 5 August 2025, the staff extended that safe harbor explicitly to liquid staking and to receipt tokens like stETH, per the SEC. The relief is not unconditional: a provider that guarantees a fixed return, or that exercises open-ended discretion over how and when to stake, falls outside it.

The guardrails are narrow on purpose. The relief the staff described covers the act of staking and the receipt token that represents it, not every product that might wrap around them. A platform that pools user funds, promises a set yield and manages the strategy at its own discretion looks more like the arrangement the SEC penalized in 2023 than the protocol staking it blessed in 2025. For holders, the takeaway is that the token itself is on firmer legal ground than it has ever been, while the packaging sold on top of it still varies case by case.

That clarity is what let staked-ETH exchange-traded funds reach the market, a story we cover in our guide to how the SEC approved crypto yield. It also brought institutions in directly. Nathan McCauley, co-founder and chief executive of the federally chartered bank Anchorage Digital, which integrated Lido in 2026, put the shift plainly, calling liquid staking “one of the most important building blocks for institutional participation in Ethereum” in comments reported by Bitcoin.com. Treasury companies followed the money: public firms have committed hundreds of millions of dollars of ETH to staking through Lido, taking wstETH and custodying it at regulated venues.

How to choose a liquid staking token

If you are weighing a liquid staking token, a short checklist separates the serious choices from the risky ones:

• Redemption path: can you withdraw the underlying ETH natively through the protocol, or is selling on the market your only exit? Native redemption is a meaningful safety backstop.
• Custody model: a contract-based token like stETH or rETH is governed by code and a distributed operator set; a custodial token like cbETH or wBETH depends on an exchange holding the keys.
• Concentration and governance: a token from a provider approaching a third of all staked ETH carries network-level risk a smaller provider does not, even if it offers deeper liquidity.
• Fees and net yield: compare the provider’s cut, since a lower headline fee can matter more than a slightly higher gross rate.
• Oracle and DeFi risk: if you plan to lend or loop the token, the reliability of the price oracle feeding those markets is as important as the token itself.

For most holders the sensible default is a widely integrated, natively redeemable token used without leverage. The exotic returns live in looping and restaking, and so does most of the risk. Liquid staking earned its place as Ethereum’s base money by making staked ETH useful again; the open question for 2026 is whether the guardrails around its largest provider mature as fast as the capital pouring into it.

Frequently Asked Questions

What is liquid staking in simple terms?

Liquid staking lets you earn Ethereum staking rewards without locking your ETH away. You deposit ETH with a protocol like Lido or Rocket Pool and receive a token, such as stETH or rETH, that represents your staked ETH plus its rewards. You can trade, lend or spend that token freely while the underlying ETH keeps validating and earning, so you get the yield and keep the liquidity.

Is stETH the same as ETH?

No. stETH is a token that represents ETH staked through Lido, and it aims to stay close to a 1:1 value with ETH because it can be redeemed for ETH through the protocol. But it is a separate asset with its own risks, including smart-contract risk and the chance that its market price drifts below ETH during a liquidity crunch, as happened in 2022. Since Ethereum enabled staking withdrawals in 2023, native redemption gives stETH a stronger anchor than it had then.

How much can you earn from liquid staking in 2026?

Ethereum’s base staking yield is around 2.7% a year in 2026, down from more than 5% in 2023 because rewards are spread across a record amount of staked ETH. Liquid staking providers take a fee, often around 10% of rewards, so the net rate is a little lower. Users who lend or loop their tokens in DeFi can earn more, but that added return comes with liquidation and oracle risk.

What are the risks of liquid staking?

The main risks are smart-contract bugs, a market depeg where the token trades below the value of the underlying ETH, validator slashing passed on by the provider, and, for custodial tokens, exchange counterparty risk. There is also a network-level concern: because one provider controls close to a third of all staked ETH, its governance and reliability affect Ethereum as a whole, which is why mechanisms like Lido’s dual governance exist.

Is liquid staking legal in the United States?

As of 2026, US regulators have grown far more accommodating. The SEC’s staff said in 2025 that protocol staking is not itself a securities transaction and extended that safe harbor to liquid staking and receipt tokens like stETH, provided the offering does not guarantee returns or exercise open-ended discretion over staking. That shift helped clear the way for staked-ETH exchange-traded funds. It is staff guidance rather than settled law, so specifics can still change.

By Yuki Tanaka, senior DeFi correspondent at HOGE Wire.