Liquid Staking in 2026: Custodians, DAOs, and the Receipt Token

Stake ETH the old-fashioned way and you lock it up. The coins sit behind a validator, earning a yield, but you cannot sell them, lend them, or post them as collateral until you exit a queue that can run for weeks. Liquid staking removes that trade-off. You hand your ETH to a provider, the provider stakes it, and in return you receive a token that stands for your deposit plus the rewards it is earning. That token is liquid: you can trade it, lend it, or drop it into a dozen other protocols while the original ETH keeps working behind the scenes.

By late 2026 this is no longer a niche. Roughly 39.7 million ETH, about a third of all circulating supply, is staked across more than 1.24 million validators, according to Datawallet, and a large slice of that sits behind a liquid staking token rather than a validator the owner runs personally. The single largest of those tokens, Lido’s stETH, is a top-ten crypto asset in its own right, with a market value near 24.8 billion dollars as of 8 October 2026, per CoinGecko.

The part that gets glossed over is that these tokens are not all the same thing. A liquid staking token is an IOU, a claim on staked crypto that someone else is holding. Who that someone is varies enormously: a Swiss-registered DAO, a permissionless smart contract with thousands of independent node operators, or a US-listed exchange that runs the validators on its own servers. The yield looks similar across all of them. The thing you are actually trusting does not. This piece is about that difference, and about the receipt-token economy it has built.

How the receipt-token machine works

The mechanics are simpler than the jargon suggests. You send ETH to a staking protocol or an exchange. Behind the scenes, that ETH is pooled with everyone else’s and used to fund validators, the software nodes that propose and attest to blocks on Ethereum’s proof-of-stake network. Each validator requires 32 ETH, but as a depositor you never see that boundary; you can stake 0.1 ETH or 10,000 ETH and the protocol handles the bundling into whole validators.

In exchange for your deposit you are minted a liquid staking token, usually shortened to LST. One stETH, one rETH, one cbETH or one wBETH each represent a position in the underlying staked pool. As the validators earn, the value backing each token grows. The reward itself has two parts: the consensus-layer issuance Ethereum pays for honest validation, and the execution-layer income from transaction priority fees and MEV, the value captured from ordering transactions inside a block. A good provider harvests both and passes most of it on, minus a fee.

Getting out used to be the weak point. Before Ethereum’s Shapella upgrade in April 2023, staked ETH could not be withdrawn at all, which is exactly why the secondary market for stETH mattered so much, as the next sections will show. Today every serious provider supports withdrawals, but they are not instant: unstaking routes through Ethereum’s exit queue, and the protocol keeps a liquidity buffer so that most redemptions settle quickly while larger ones wait their turn. The liquid token exists precisely so you rarely need that exit at all; you can simply sell the token to someone else who wants the staked position.

Three ways to package the same yield

Not every liquid staking token behaves the same way in your wallet, and the difference matters for taxes, for DeFi integrations, and for how the peg is read. Three designs dominate.

The first is rebasing. Lido’s stETH is the headline example: your balance grows a little each day, so if you hold 10 stETH this morning you might hold 10.0005 tomorrow, and one stETH is meant to stay roughly equal to one ETH. It is intuitive, but the constant drip of new units is awkward for smart contracts and, in many jurisdictions, generates a running series of taxable events.

The second is the value-accruing, or reward-bearing, design. Here your token count never changes; instead each token is worth progressively more ETH over time. Rocket Pool’s rETH works this way, and as of 8 October 2026 one rETH was worth about 1.1713 ETH, per CoinGecko, with the accumulated reward baked into that exchange rate. Lido offers a wrapped version of stETH called wstETH on the same principle, and crucially the two big custodial exchange tokens, Coinbase’s cbETH and Binance’s wBETH, use it too. That is why cbETH traded near 1.1436 ETH, well above the spot ETH price, in early October, per CoinGecko: the premium is roughly three years of compounded staking since the token launched in 2022, not a mispricing.

The third is Frax’s two-token split, which separates the liquid-but-yieldless frxETH from the yield-concentrating sfrxETH vault, letting the protocol route rewards to whoever opts in. The design is elegant, but it has stayed small next to the giants. The key point for a holder is that these are three ways of expressing the same underlying thing, and the one you pick changes how the reward shows up, and how your accountant treats it, long before it changes how much you earn.

The custody question that actually matters

Strip away the branding and every liquid staking token answers one question: who is holding your ETH and deciding how it is staked? The answers fall on a spectrum, and where a token sits on it tells you more about your real risk than the headline APR does.

At one end sit the custodial exchange tokens. When you stake ETH on Binance or Coinbase and receive wBETH or cbETH, the exchange takes custody of the underlying coins, runs the validators on its own infrastructure, and controls the keys. You are trusting a company’s balance sheet and operational competence, the same way you trust it to hold your spot balance. The upside is a clean user experience and, for most retail users, a familiar brand with a support desk. The downside is counterparty risk: the token is only as solvent as its issuer, and if the exchange freezes withdrawals, your liquid position is suddenly not liquid.

In the middle sits Lido, the DAO-curated model. Lido itself does not run validators; it delegates pooled ETH to a vetted set of professional node operators, with the LDO token governing who joins that set and how fees are split. It is decentralized in governance but still permissioned in practice, because you cannot simply show up and become a Lido operator. The trust you place is in the DAO’s process and in the honesty of its chosen operators.

At the far end sits Rocket Pool, which is permissionless. Anyone can run a Rocket Pool node by posting a bond in ETH and RPL alongside the pooled stake, which means the validator set is open and nobody needs the protocol’s approval to join. That openness is the strongest decentralization story in the sector, and it is also why Rocket Pool has stayed far smaller than Lido: a permissionless network of thousands of individual operators is harder to scale than a curated roster of institutions.

None of these is simply better than the others. They are different trust models, and the right one depends on whether your main worry is a company failing, a DAO being captured, or a smart contract carrying a bug. Picking a liquid staking token without knowing which of those risks you are signing up for is the single most common mistake newcomers make.

The exchange tokens nobody talks about

The custodial corner of the market gets the least coverage and is, by one measure, the second-largest segment of all. Binance’s Wrapped Beacon ETH (wBETH) carried a market value of roughly 9.5 billion dollars on 8 October 2026, spread across about 3.37 million tokens, according to CoinGecko. That makes wBETH one of the largest liquid staking tokens in existence, behind only Lido’s stETH and its wrapped form wstETH, and many times the size of Rocket Pool’s rETH. The biggest challenger to Lido’s dominance, in pure size terms, is not a decentralized protocol at all. It is an exchange, and that fact rarely makes it into the usual Lido-versus-Rocket-Pool framing.

Coinbase’s cbETH tells the opposite story. Despite Coinbase being the largest US-listed exchange, cbETH had shrunk to a market value near 486 million dollars across roughly 166,000 tokens by October 2026, per CoinGecko. Coinbase still mints the token, and it still earns a live yield, listed near 2.37 percent by StakingRewards, but the company has leaned toward plain custodial staking, where a user simply holds a staked balance rather than a wrapped ERC-20. The wrapper lost much of its reason to exist for retail customers once the regulatory cloud over US staking lifted, which it did in 2025.

The gap between the two is a clean illustration of the custodial bargain. wBETH’s scale shows how much staking demand flows through a single large exchange when the product is frictionless and the brand is trusted. cbETH’s decline shows how quickly a custodial token can fade when its issuer decides a wrapper is more trouble than it is worth. In both cases the holder’s exposure runs straight back to one company, which is the entire difference between a custodial LST and an on-chain one.

Lido, Rocket Pool and Frax: the on-chain field

On the decentralized side, three names still define the category. Lido is the giant: its stETH accounts for the bulk of the liquid staking market, with a market value near 24.8 billion dollars across about 9.69 million tokens as of 8 October 2026, per CoinGecko, which ranks stETH as a top-ten crypto asset. Lido’s share of all staked ETH has slipped to roughly 23 percent from a peak near 32 percent in late 2023, according to Datawallet, a decline the protocol frames as healthy for the network even as it defends its lead with new products and governance changes.

Rocket Pool is the decentralization purist. Its rETH carried a market value around 946 million dollars on 8 October 2026, per CoinGecko, a fraction of Lido’s, but its permissionless node network and its ETH-plus-RPL bond model make it the reference point whenever anyone argues about whether staking has grown too concentrated. If you care more about censorship resistance than about having the deepest liquidity, rETH is the usual answer.

Frax rounds out the trio with its frxETH and sfrxETH pair. It is the smallest of the three and has spent 2026 folding its staking products into a broader stablecoin-and-chain strategy, but it remains a useful example of how differently the same yield can be packaged and how a protocol can use staked ETH as one input into a larger machine rather than as the product itself.

These three have been compared endlessly, and the honest summary is that they optimize for different things: Lido for scale and liquidity, Rocket Pool for permissionlessness, Frax for composability within its own ecosystem. Add the two exchange tokens and you have the whole shape of the Ethereum market in five names.

The Ethereum liquid staking map at a glance

The table below collects the major Ethereum liquid staking tokens, their issuers, and how each one is backed. Market values are from CoinGecko as of 8 October 2026 and move with the ETH price, so treat them as a snapshot rather than a fixed figure.

Two things jump out. First, the custodial segment (wBETH plus cbETH) is large, and it is overwhelmingly a Binance story. Second, the value-accruing design has quietly won almost everywhere except Lido’s flagship rebasing token, which is why so many LSTs trade at a premium to ETH that has nothing to do with demand and everything to do with accumulated yield.

What you actually earn

The reason to stake at all is yield, and in 2026 that yield is thinner than newcomers expect. Ethereum’s base staking reward has drifted down toward the 2.5 to 2.7 percent range as the staked total has climbed, because the issuance curve pays less per validator the more ETH is locked up. Add execution-layer tips and MEV and an efficient provider lands somewhere near 3 percent all-in. Our breakdown of validator economics walks through why the number keeps compressing and what actually moves it from one quarter to the next.

That matters because 3 percent in ETH terms is not the same as 3 percent in dollars. When US Treasury bills pay more than a staked-ETH position, the staking reward is best read as a thin coupon layered on top of ETH price exposure, not as a reason to own ETH by itself. The ETH you stake can still fall 48 percent below its all-time high, as it had by early October 2026 per CoinGecko, and a 3 percent yield does little to cushion a move like that.

Providers compete over how much of that thin spread they let you keep. Lido and most decentralized peers take around 10 percent of staking rewards as a protocol fee, written transparently into code; Rocket Pool’s effective take varies with its node-operator economics; and the custodial exchanges have historically charged more, setting their cut as a business decision they can revise. Over years of compounding, a few percentage points of fee on an already-thin yield is not a rounding error, it is a meaningful share of what you came for.

Solana runs the same play differently

Ethereum is not the only chain with a receipt-token economy. Solana has one too, and it is structured differently. Because a very high share of SOL is already staked natively, the liquid slice is proportionally smaller, around 14 percent of staked SOL by recent counts, but the yields are higher, typically in the mid-single digits rather than near 3 percent.

The Solana leaderboard also looks different. Jito’s JitoSOL leads on MEV-boosted rewards, with roughly 1.2 billion dollars tracked by DefiLlama in late September 2026 and a supply APY near 4.85 percent. Sanctum, an umbrella that lets validators spin up their own branded staking tokens, collectively tracked above 2 billion dollars; Marinade, the early pioneer that spreads stake across more than a hundred validators by a public scoring algorithm, sat lower at a few hundred million. The structural point for a cross-chain reader is that the custodial pattern repeats here too, because Binance issues bnSOL on Solana just as it issues wBETH on Ethereum. The same exchange is a major liquid staking issuer on two different chains at once.

The lesson is that the receipt-token model generalizes cleanly, but the competitive shape, the yield level, and the degree of concentration are all chain-specific. A JitoSOL holder and a stETH holder own the same kind of instrument; they do not own the same kind of risk.

The receipt token as a DeFi building block

A liquid staking token would be useful even if all it did was stay tradeable. What turned it into the backbone of on-chain finance is that other protocols accept it. stETH and wstETH are among the most widely used forms of collateral in lending markets; you can deposit them, borrow against them, and in many cases loop the position to amplify your staking exposure. The same tokens anchor liquidity pools, back stablecoins, and serve as the base asset for structured products. A staked-ETH token is, in practice, DeFi’s reserve collateral.

That composability is also where risk compounds. The same token can be staked, lent, and re-pledged several times across different protocols, so a problem at the base, a depeg or a validator incident, can propagate upward through everything built on top of it. The clearest example sits one layer up, in restaking, where liquid staking tokens are re-pledged to secure additional services in exchange for extra yield. That market has had a rough 2026; our look at the security glut nobody is renting covers why the promised extra yield has largely not materialized, and why rehypothecating staked ETH stacks fresh slashing conditions on top of the ones you already carry.

The takeaway is not that composability is bad; it is the whole reason liquid staking beats locking ETH behind your own validator. It is that a liquid staking token is a foundation, and anything you build on a foundation inherits the foundation’s cracks. Each layer you add trades a little more yield for a little less clarity about what, exactly, could break.

When one stETH is not one ETH

The peg is the quiet assumption under the whole model: that one stETH can always be swapped for roughly one ETH, or that one rETH is really worth its stated 1.1713 ETH. Most of the time that holds, enforced by arbitrage and by the ability to redeem directly with the protocol. But it is a soft peg, not a hard one, and it has broken before.

The textbook case came in June 2022, when stETH traded as low as roughly 0.94 ETH on the secondary market, according to CoinDesk. There was no shortfall in the underlying staked ETH. The problem was that withdrawals were not yet enabled, since Shapella was still ten months away, so holders who needed cash in a hurry, Celsius and Three Arrows Capital among them, could only sell on the open market, and the forced selling pushed the price below parity. The token was fully backed the entire time. It simply could not be redeemed, and in a panic that distinction stops mattering to the person trying to get out.

Today withdrawals exist, which makes a 2022-style gap far less likely, because anyone can redeem near parity rather than dumping into a thin pool. But the episode is the permanent reminder that a liquid staking token is a claim, and a claim trades on confidence in redemption, not only on the assets sitting behind it. The deeper and more liquid the token’s markets, the smaller the gap tends to be when confidence wobbles, which is one concrete advantage the largest tokens hold over the smallest.

The risk stack, itemized

Because a liquid staking token sits at the base of so much else, its risks are worth itemizing rather than lumping into one word. They do not all apply to every token, which is the whole point of the custody spectrum: different models carry different failure modes.

Custodial counterparty risk is the one that most cleanly separates the exchange tokens from the rest, and it is worth being precise about what you are trusting when an exchange holds the keys. Key management is its own discipline, not an afterthought; our conversation with Trail of Bits on MPC and TEEs digs into how large custodians actually secure the private keys behind billions of dollars in staked assets, and why “the exchange holds it” is a sentence that hides a great deal of engineering, and a great deal that can go wrong.

The one-third problem

The concern that gets Ethereum researchers out of bed is not any single token failing. It is concentration. Ethereum’s consensus has meaningful thresholds: a single entity controlling more than one third of staked ETH can disrupt the chain’s ability to finalize blocks, and one controlling more than two thirds could finalize blocks on its own terms. Lido’s roughly 23 percent share of all staked ETH, per Datawallet, sits uncomfortably close to the one-third line for many observers, even though Lido spreads that stake across a set of independent operators rather than running the validators itself.

Vitalik Buterin has returned to this worry repeatedly. In its coverage of his writing on the subject, The Block summarized his concern as the potential for a single liquid staking token to take over the “money” network effects from Ethereum itself, a scenario in which the LST, rather than ETH, becomes the default asset people hold and its governance becomes a lever over the chain. Lido’s defenders counter that a diverse operator set and on-chain guardrails blunt the threat, and the protocol has added mechanisms, including a dual-governance system that lets stETH holders slow or veto contentious DAO decisions, precisely to answer the capture objection.

The custody spectrum reappears here in a sharper form. A custodial exchange that stakes several million ETH is also a concentration risk, arguably a more acute one, because it is a single company with a single set of keys rather than a spread of independent operators. When wBETH alone represents billions of dollars of staked ETH run on one firm’s infrastructure, concentration stops being only a Lido question and becomes a question about the whole top of the market.

What the SEC decided

For years the unanswered question hanging over US staking was whether a liquid staking token is a security. In 2023 the SEC sued both Coinbase and Binance, with their staking services among the activities it challenged. That posture reversed in 2025. The agency dismissed its case against Coinbase with prejudice in February 2025, part of a broader retreat that also dropped actions against Kraken, Ripple and Robinhood, as law firm Manatt documented, and its long-running Binance litigation was wound down as well, per Banking Dive.

The clearest signal came in August 2025, when the SEC’s Division of Corporation Finance stated that, in its view, “Liquid Staking Activities…do not involve the offer and sale of securities,” and that the receipt tokens themselves do not either, in the Division’s published statement. The safe harbor comes with a sharp condition: the provider has to stay administrative or ministerial. It must not “decide whether, when, or how much” to stake, and must not “guarantee or otherwise set the amount” of rewards. A provider that exercises that kind of discretion, the statement warns, falls outside the view entirely.

That condition maps almost perfectly onto the custody spectrum. A permissionless protocol is administrative by construction; it cannot choose anything because it is code. A custodial exchange has to be careful not to behave like a discretionary asset manager if it wants to stay inside the safe harbor. SEC Commissioner Hester Peirce framed the receipt token in exactly these terms, describing it as “a variant on the longstanding practice of depositing goods with an agent who performs a ministerial function in exchange for a receipt that evidences ownership of the goods,” in commentary collected in published writings on the guidance. The receipt, in that reading, is a warehouse ticket, not a share of stock.

The clarity opened a door that had been shut for years: staked-ETH exchange-traded funds. In March 2026 BlackRock listed its iShares Staked Ethereum Trust (ETHB) on Nasdaq, a spot-ETH fund that stakes its holdings and passes most of the reward through to investors, splitting rewards 82 percent to holders and 18 percent to the sponsor, as Cointelegraph reported. “By bringing together spot ether exposure and staking rewards in an ETP, ETHB provides investors with an important new avenue to participate in the ecosystem’s evolution,” said Robert Mitchnick, BlackRock’s global head of digital assets, in comments to CNBC; he added that the yield could draw in investors who had been “given pause” by its earlier absence. The arrival of such products was the theme of CoinDesk’s 2026 staking outlook, and our guide to who approves crypto ETFs now traces how the pipeline opened.

Europe took a different road. Under MiCA, the EU’s crypto rulebook, the staking protocol itself is generally out of scope while the service providers around it are regulated, and a custodial staking token can brush up against the rules on asset-referenced and e-money tokens in ways a US regulator would not frame the same way. We mapped the European regime in our MiCA explainer; the short version is that the very same receipt token can be treated as a non-security in New York and as a regulated instrument in Frankfurt, which matters a great deal if you are an exchange deciding where to offer it.

Where this goes next

Three trends are shaping the next phase. The first is consolidation at the validator layer: with Ethereum’s Pectra upgrade allowing a single validator to hold far more than 32 ETH, large providers are merging validators to cut overhead, which lowers the raw validator count even as the staked total rises. The second is institutionalization: staked-ETH ETFs, regulated custodians offering wrapped staking tokens, and professional operators now sit alongside the original DeFi-native crowd, and they bring both deeper liquidity and sharper concentration at the same time. The third is a slow repricing of trust, as holders learn to ask not just what a token yields but who, exactly, stands behind it.

What will not change is the core bargain. A liquid staking token hands you liquidity and yield in exchange for trusting someone, or some code, to hold and stake your ETH honestly. The entire 2026 debate, custodial versus decentralized, concentrated versus diffuse, security versus not, is really one argument about how to price that trust. The receipt is only ever as good as whoever stands behind it, and the most important thing you can know about your liquid staking token is whose promise it actually is.

Frequently Asked Questions

Is a liquid staking token the same as actually owning ETH?

Not quite. A liquid staking token is a claim on staked ETH that a provider holds for you, plus the rewards it is earning. It usually tracks the value of ETH closely, but it carries extra risks ETH does not, including smart-contract bugs, validator slashing, and, for exchange-issued tokens, the solvency of the issuer.

Why does one rETH or one cbETH cost more than one ETH?

Because those tokens are value-accruing. Instead of increasing your token count, they let each token represent a growing amount of ETH over time, so the accumulated staking reward shows up as a higher exchange rate. In early October 2026 one rETH was worth about 1.1713 ETH and one cbETH about 1.1436 ETH, which reflects years of compounded rewards rather than a premium you are overpaying.

Which is safer, a custodial token like wBETH or a decentralized one like stETH?

They trade different risks. Custodial tokens from Binance or Coinbase depend on the exchange staying solvent and honest but spare you smart-contract exposure. Decentralized tokens remove the single-company risk but rely on audited code and, in Lido’s case, on DAO governance. Neither is universally safer; the right choice depends on which failure worries you more.

Can a liquid staking token lose its peg to ETH?

Yes, though it is less likely than it used to be. In June 2022 stETH briefly traded near 0.94 ETH because withdrawals were not yet enabled and distressed holders had to sell into a thin market. Now that direct redemptions exist, a token can be swapped close to its backing, which makes a sustained depeg much harder, though a soft peg is never guaranteed in a genuine panic.

Does staking through an ETF count as liquid staking?

Not exactly, but it is a close cousin. A staked-ETH ETF such as BlackRock’s ETHB holds ETH, stakes it, and passes most of the reward to shareholders, giving you staking yield through a regulated wrapper rather than an on-chain token. You get exposure and yield without holding keys or a liquid staking token yourself, but you depend on the fund sponsor and its custodian instead.

By Yuki Tanaka, staking and DeFi correspondent, HOGE Wire.