Skip to content
DeFi Reference

Liquid staking tokens

1. TL;DR

Section titled “1. TL;DR”

A liquid staking token (LST) is a receipt a staking pool gives you in exchange for your deposit, so that you can keep trading, lending, or spending your stake’s value while the underlying coins stay locked up earning rewards. On Ethereum, one design (Lido’s stETH) dominates with roughly 70% of the liquid-staking market as of 2024 data cited by Helius; on Solana, no single liquid staking token dominates, and a “long tail” of over a thousand validator-specific tokens has emerged instead, aggregated through infrastructure like Sanctum.

2. Explain it simply

Section titled “2. Explain it simply”

Analogy

Section titled “Analogy”

Depositing coins into a staking pool is like checking a suitcase at the airport: you hand over the physical bag (your coins go to a validator and get locked up), but you get a claim ticket in return. That ticket is worth exactly your bag no matter who is holding the bag, and — unlike a real claim ticket — you can trade it, use it as collateral for a loan, or sell it to someone else entirely before your original bag ever comes back.

ELI5

Section titled “ELI5”

Locking up your coins to help secure a network usually means you can’t touch them for a long time. A liquid staking pool solves this by giving you a second token the moment you deposit — a stand-in that represents your locked coins plus the rewards they’re earning. You can use that stand-in token anywhere a normal token is accepted: trade it, lend it out, or hold it as savings. Meanwhile, the pool operator does the actual work of running the computers that earn the rewards. The tricky part is that you’re now trusting the pool and its operators to run things honestly and not lose your money.

Step-by-step walkthrough

Section titled “Step-by-step walkthrough”

Scenario: Mia deposits 10 ETH into a decentralized staking pool (Lido) when the pool’s stETH balance mechanism is 1:1 with ETH.

  1. Deposit (before: 10 ETH in wallet → after: 10 stETH in wallet). Mia sends 10 ETH to the pool’s smart contract and immediately receives 10 stETH. Her ETH is now queued to be assigned to one of the pool’s node operators.
  2. Pool delegates the ETH (before: ETH sits in the contract → after: ETH assigned to a validator). Once 32 ETH accumulates from various depositors, the pool activates a new validator using a professional node operator selected by its governance.
  3. Rewards accrue over a year (before: 10 stETH balance → after: ~10.32 stETH balance). The pool’s design is “rebasing”: each day, Mia’s stETH balance itself grows to reflect her share of new rewards, net of the pool’s fee, while the stETH-to-ETH ratio stays fixed near 1:1.
  4. Mia uses the token (before: 10.32 stETH sitting idle → after: 10.32 stETH posted as loan collateral). Because stETH is a normal, tradeable token, Mia deposits it into a lending market to borrow against it — something she could not do with locked-up, undelegated stake.

Common misconceptions

Section titled “Common misconceptions”
  • Myth: A liquid staking token is always redeemable 1-for-1, instantly, for the underlying coin. Reality: Redemption is rate-limited by an unbonding period and by how much instant-exit liquidity the pool keeps on hand; once that buffer is empty, exits queue behind it (Neuder & Chitra, ethresear.ch 18799, 2024-02).
  • Myth: All liquid staking tokens work the same way. Reality: Some grow your token balance over time (Lido’s stETH); others keep your balance fixed and grow the token’s exchange rate against the underlying coin instead (Rocket Pool’s rETH, and every major Solana liquid staking token) — Helius notes “Solana currently has no rebasing LSTs” (Helius, “Liquid Staking and LSTs on Solana,” n.d.).
  • Myth: More liquid staking tokens automatically means more decentralization. Reality: A dominant liquid staking token can create a single point of failure and a “stratum for cartelization” if it grows large enough to function as the de facto money of the network (Hasu & Konstantopoulos, Paradigm, 2021-04).
  • Myth: Solana lags Ethereum on liquid staking because the technology is worse. Reality: Helius attributes Solana’s under-5% liquid-staking share (versus Ethereum’s roughly one-third-plus of stake) mainly to path dependence, thinner on-chain collateral utility, and large holders preferring to run their own validators for legal reasons (Helius, n.d.).

If you only remember one thing

Section titled “If you only remember one thing”

A liquid staking token turns “locked and earning” into “liquid and earning,” but it can only do that by adding a layer of trust in whoever operates the pool and whoever else is willing to buy your ticket back.

3. How it works

Section titled “3. How it works”

Two accounting models

Section titled “Two accounting models”

Every liquid staking token needs to represent a growing pool of underlying value (deposits plus accumulated rewards, minus any losses) with a token that can be freely transferred. There are two common designs:

  • Rebasing. The token supply itself expands to match rewards. If Mia holds a fraction ff of total shares, her balance at time tt is balancet=f⋅Et\text{balance}_t = f \cdot E_t, where EtE_t is the pool’s total pooled ETH (principal plus rewards). The exchange rate against the underlying coin stays fixed near 1, and Mia’s balance visibly grows every day. Lido’s stETH uses this model.
  • Exchange-rate (reward-bearing). The number of tokens a holder owns never changes after their deposit; instead, the token’s price against the underlying coin rises: rt+1=rt⋅(1+yt)r_{t+1} = r_t \cdot (1+y_t), where yty_t is the net yield earned that period after the pool’s fee is deducted. Rocket Pool’s rETH and every major Solana LST (JitoSOL, mSOL, and the long tail of validator tokens) use this model — Solana’s SPL stake pool program tracks a pool-wide exchange rate and mints tokens against it rather than rebasing balances.

Both designs deliver the same underlying economics; they differ only in which number moves. Rebasing tokens can complicate tax reporting in jurisdictions that treat each rebase as a taxable event, which is one reason some designs deliberately avoid it (Helius, n.d.).

The stake-pool structure

Section titled “The stake-pool structure”

On both chains, a pool works by (1) accepting deposits and minting the LST, (2) delegating the pooled principal across a set of node operators according to some selection policy, and (3) collecting rewards net of a fee, split between node operators and (where applicable) a DAO treasury. Ethereum’s constructions differ mainly in how permissioned that operator set is:

  • Permissioned node operators (stETH). Lido’s DAO vets and approves roughly 30 professional operators, who post no collateral of their own — a design the Paradigm authors call efficient but exposed to a principal-agent problem, since operators have “nothing at stake” of their own capital (Hasu & Konstantopoulos, 2021-04).
  • Permissionless node operators (rETH). Rocket Pool lets anyone become an operator if they post collateral — 8 ETH of their own plus 2.4 ETH worth of RPL to match 24 ETH of pooled capital (Helius, n.d.), aligning operator and depositor incentives without a curation committee.
  • Centralized custodians (cbETH). An exchange runs the validators itself and issues a custodial receipt token.

Solana instead has a native delegation primitive built into the base protocol, so a stake pool LST (JitoSOL, mSOL) simply automates delegation across many validators the pool selects, while a validator LST — a design popularized by Sanctum — lets a single validator issue its own branded token with no pooling logic at all, trading diversification for simplicity and “tribalism” around one operator (Helius, “Liquid Staking and LSTs on Solana,” n.d.).

Why network effects concentrate the market

Section titled “Why network effects concentrate the market”

Hasu & Konstantopoulos argue LSTs are subject to the same dynamic as any money-like asset: the more places a token is accepted as collateral, the more valuable holding it becomes, which drives further adoption — a flywheel that tends toward one dominant winner per chain (Paradigm, 2021-04). This argument underlies the “rainbow staking” literature’s concern about LST dominance and motivates proposals for protocol-level primitives (Liquid Staking Modules, enshrined partial pools) that would let many liquid staking tokens compete on a level footing rather than converging to one (barnabe, ethresear.ch 18683, 2024-02). See /staking/restaking/ for how this framework extends to restaked positions.

4. Worked numeric example

Section titled “4. Worked numeric example”

Ethereum (matching §2). Mia deposits 10 ETH into Lido’s stETH pool, minting 10 stETH at a 1:1 balance ratio. Assume a 3.6% gross annual staking yield and Lido’s cited fee split (roughly a 10% cut of rewards shared between node operators and the DAO, per the fee comparison table in casparschwa & adietrichs, ethresear.ch 18751, 2024-02, footnote 5’s linked sources). Net yield to Mia: 3.6%×(1−0.10)=3.24%3.6\%\times(1-0.10)=3.24\%. After one year, her rebasing balance grows to 10×1.0324≈10.3210\times1.0324\approx10.32 stETH, each still worth approximately 1 ETH, so her position is worth roughly 10.32 ETH — a gain of 0.32 ETH for having done nothing beyond holding the token, with the pool’s chosen node operators doing the validating.

Solana analogue. Compare an exchange-rate design: depositing 10 SOL into a stake-pool LST at an initial exchange rate of r0=1.00r_0=1.00 SOL per token mints 10 tokens. If the pool nets an 6.5% annual yield after fees (illustrative, in the range cited for Marinade’s mSOL by Milkroad’s 2026 review, secondary, accessed 2026-08), the exchange rate rises to r1=1.00×1.065=1.065r_1 = 1.00\times1.065=1.065. The depositor’s 10 (unchanged) tokens are now worth 10×1.065=10.6510\times1.065=10.65 SOL — the same underlying gain as the rebasing case, but visible in the price of the token rather than in a growing balance.

5. Where it’s used

Section titled “5. Where it’s used”

Ethereum

Section titled “Ethereum”
  • Lido (stETH) — permissioned operator set, rebasing token, roughly 70% of the total LST market by size as cited by Helius (n.d., citing 2024 data), and about 31% of all staked ETH per Dune data referenced in the same piece. lido.fi
  • Rocket Pool (rETH) — permissionless, collateral-backed node operators; exchange-rate token. rocketpool.net
  • Coinbase (cbETH) — centralized custodial staking with an on-chain, tradeable receipt token.
  • EigenLayer / restaking protocols — accept ETH LSTs as the deposit asset for restaking, layering a second token (an LRT) on top; see /staking/restaking/ and /staking/lrt-risks/.

Solana

Section titled “Solana”
  • JitoSOL (Jito stake pool) — the largest Solana LST; delegates only to validators running Jito’s client and rebalances across the top-performing set, distributing MEV tips on top of staking rewards. jito.network/docs/jitosol
  • mSOL (Marinade) — Solana’s original liquid staking token; diversifies delegation across roughly 100 validators.
  • Sanctum and the validator-LST long tail — Sanctum’s shared liquidity pool (“Infinity”) reduces the marginal cost of launching a new LST to near zero, enabling over 1,361 distinct LSTs as of December 2025, including branded tokens like jupSOL (Jupiter), bbSOL (Bybit), and dfdvSOL. learn.sanctum.so/docs
  • hSOL (Helius) — a validator LST from a zero-commission validator, illustrating the “tribalism” model where a single operator’s brand anchors the token.

6. Risks, attacks, and incidents

Section titled “6. Risks, attacks, and incidents”
  • Real incident: the stETH depeg, June 2022 (as of 2022-06). After the Terra/UST collapse drained market liquidity, large holders (notably Celsius and, indirectly, Three Arrows Capital) needed to exit leveraged stETH positions through Curve — the only deep secondary market before Ethereum’s withdrawals were enabled — and pool imbalance pushed stETH to trade as low as roughly $0.93–0.95 per ETH around June 13, 2022 (CoinDesk, 2022-06-29, reporting Nansen research; secondary). Root cause: stETH could not yet be redeemed 1:1 on-chain, so its price depended entirely on secondary-market liquidity, which vanished under stress.
  • Real incident: Lido/RockLogic Prysm client slashing, April 2023 (as of 2023-04). A Lido node operator, RockLogic, restarted an updated Prysm validator client that contained a bug (Prysmatic Labs issue #12281) causing it to re-import validator keys that had supposedly been deleted, which then signed slashable messages; 11 validators were slashed, with an estimated 13.77 ETH in penalties and missed rewards — about 2.4% of that day’s rewards and 0.0023% of protocol TVL. Lido’s DAO voted on 30 June 2023 to use its cover fund to compensate affected stETH holders (Lido, “Post Mortem: Lido on Ethereum RockLogic GmbH Slashing Incident,” 2023). Root cause: a client software bug, not malicious operator behavior — illustrating the correlated-slashing risk of many validators running the same client.
  • Principal-agent risk from uncollateralized operators. Permissioned models like Lido’s let node operators run validators for delegators’ capital without posting their own collateral, which the Paradigm authors flag as the same “nothing at stake” problem that has historically worried Proof-of-Stake designers (Hasu & Konstantopoulos, 2021-04).
  • Too-big-to-fail dynamics. casparschwa & adietrichs argue that a sufficiently dominant liquid staking token could create an implicit expectation of a protocol bailout in the event of failure, distorting the market’s true perception of operator risk (ethresear.ch 18751, 2024-02).
  • Redemption freezes under stress. Lido’s own “bunker mode” mechanism can pause certain withdrawals during elevated slashing events, and Neuder & Chitra note that a “rush for the exit” scenario can force any liquid staking token to trade at a discount once its instant-liquidity buffer is exhausted (ethresear.ch 18799, 2024-02).

7. Open problems

Section titled “7. Open problems”
  • Can Ethereum avoid a single dominant LST? The “rainbow staking” line of research (barnabe, ethresear.ch 18683, 2024-02) proposes enshrining Liquid Staking Module-style primitives so that many liquid staking tokens can compete on equal footing rather than converging to a winner-take-most outcome.
  • Why hasn’t Solana converged on one LST? Helius offers path dependence and thin on-chain collateral utility as partial explanations but treats the question as open, especially as validator LSTs proliferate rather than consolidate (Helius, n.d.).
  • Rebasing vs exchange-rate design trade-offs. Jito’s own writing (cited by Helius) frames this as an unresolved design choice balancing tax treatment, DeFi composability, and user comprehension, with no consensus winner (Helius, n.d.).
  • How should protocol-level slashing interact with LST holders once Solana adds programmatic slashing? Helius notes Solana currently has no programmatic slashing, so today’s LST operators cannot unilaterally lose delegated stake — a property that will need re-examination once slashing ships (Helius, n.d.).

8. Ethereum vs Solana

Section titled “8. Ethereum vs Solana”
AspectEthereumSolana
Dominant designRebasing balance (stETH) alongside exchange-rate (rETH)Exchange-rate only; no rebasing LSTs
Market concentration~70% of LST market in one token (stETH), ~31% of all staked ETHFragmented: Jito ~35%, Marinade ~42% of the (small) LST segment, plus 1,000+ validator LSTs
LST share of total stakeLSTs are the dominant channel for new stakeUnder 5% of staked SOL is in LST form
Redemption pathOn-chain withdrawal queue since Shapella, plus instant secondary-market exitNative unstake (few-day cooldown) or instant redemption via aggregators like Sanctum’s Reserve
Operator modelPermissioned (Lido), permissionless-with-bond (Rocket Pool), or custodial (cbETH)Native delegated Proof of Stake; stake-pool or single-validator LSTs

Ethereum’s LST market is a story of concentration around one liquid, deeply DeFi-integrated token, which is precisely what motivates targeting and rainbow-staking research (see /staking/issuance-and-staking-economics/). Solana’s is the opposite: native delegation already gives every SOL holder liquidity-adjacent flexibility, so LSTs compete on marginal features (MEV capture, validator branding, DeFi points) rather than on solving illiquidity itself, and the market has stayed structurally fragmented.

9. Reference doc

Section titled “9. Reference doc”

The reference

Section titled “The reference”

On Staking Pools and Staking Derivatives — Hasu and Georgios Konstantopoulos (Paradigm), 23 April 2021. paradigm.xyz/writing/on-staking-pools-and-staking-derivatives

Summary of the reference

Section titled “Summary of the reference”

Writing before Ethereum’s Merge, the authors start from the frictions of solo staking: a 32 ETH minimum, the need to run and maintain validator infrastructure, and (at the time) a total inability to withdraw. They argue staking pools solve the first two problems by pooling small deposits past the minimum and outsourcing operations to professional node operators, while a fourth benefit — issuing a tradeable “staking derivative” token — solves illiquidity itself. The post then works through a decentralized pool’s mechanics using Lido as the worked example: users deposit ETH and receive stETH 1:1, and once 32 ETH accumulates, the DAO assigns it to a governance-approved operator using DAO-controlled withdrawal credentials (at the time, a 6-of-11 multisig, since replaced by a smart contract).

The core analytical claim is about systemic security. Standard intuition says liquid staking derivatives weaken Proof of Stake by separating capital from validation (“nothing at stake”), but the authors argue the opposite dominates in practice: because staking derivatives remove the opportunity cost of staking, they can push the staking ratio from an estimated 15–30% (without derivatives) to 80–100% (with them). They formalize this with a simple attack-cost argument across staking ratios — at 20% staked, an attacker needs to buy 40% of circulating ETH to control 66% of stake; the required open-market purchase share only rises as the staking ratio (and derivative liquidity) grows, so the paper concludes that above roughly 60% staked, derivatives strictly increase rather than decrease economic security.

The post closes by arguing the eventual winner of the pooled-staking market will most likely be a decentralized, non-custodial protocol rather than a centralized exchange, for three reasons: decentralized pools are more socially scalable (no cap on how much stake one entity can hold without raising centralization concerns), their derivative token is trustless rather than custodial (a stronger basis for network effects), and they face fewer reputational constraints on MEV extraction than regulated exchanges — potentially flipping the usual “decentralization premium” into a “decentralization discount.” The authors explicitly disclose that Paradigm held LDO tokens at the time of writing.

Key quotes

Section titled “Key quotes”

“Solo staking is trustless, but illiquid and inconvenient; liquid staking requires varying degrees of trust but is very convenient and importantly liquid.” (§ Unpacking the stETH token)

“We can conclude that if staking derivatives can increase the number of ETH staked above 60%, they would strictly increase Ethereum’s economic security instead of decreasing it.” (§ Unpacking the stETH token)

“The existence of stETH grants its pool an important network effect… it is possible that stETH will replace ETH in many use cases, and potentially even replace ETH altogether.” (§ Unpacking the stETH token)

“As long as there is a private benefit to creating and using them, they will exist and flourish.” (§ Conclusion)

How to read the original

Section titled “How to read the original”

Background needed: pre-Merge Ethereum staking mechanics (32 ETH deposits, no withdrawals yet) and the basic idea of an LP token as loose analogy. Skip the footnoted academic citations on staking-versus-lending equilibria unless you want the formal finance angle. The hardest part to reconcile with 2026 hindsight is the attack-cost argument: it assumes an attacker must acquire stETH on the open market and then unwind it to attack the underlying chain, which is a more complex, multi-step attack than the simple “buy 34% of ETH” baseline — read it as a lower bound on attacker cost, not a precise threat model.

What changed since

Section titled “What changed since”
  • Ethereum’s Merge (September 2022) and Shapella upgrade (April 2023) enabled staking withdrawals, closing the “cannot redeem” gap this post treated as a temporary limitation of the derivative model.
  • Lido’s withdrawal credentials moved from the multisig described in the post to a smart contract, as the authors anticipated.
  • The staking ratio reached roughly 25% by early 2024 — well short of the 80–100% the authors projected liquid staking derivatives would eventually enable — and subsequent research (casparschwa & adietrichs, ethresear.ch 18751, 2024-02; see /staking/issuance-and-staking-economics/) argues that outcome should be actively prevented, inverting this post’s implicit optimism about high staking ratios.
  • The “rainbow staking” framework (barnabe, ethresear.ch 18683, 2024-02) directly engages this post’s LST-dominance concern with concrete protocol-level countermeasures.

Secondary references

Section titled “Secondary references”
  1. Helius, “Liquid Staking and LSTs on Solana” (n.d.) — read for the Solana-side construction details and the Ethereum-vs-Solana LST comparison used throughout this page.
  2. barnabe, “Unbundling staking: Towards rainbow staking” (ethresear.ch 18683, 2024-02) — read for the protocol-level response to this post’s LST-concentration concern.
  3. Neuder & Chitra, “The risks of LRTs” (ethresear.ch 18799, 2024-02) — read next for how these same dynamics compound once LSTs are restaked; see /staking/lrt-risks/.
  4. Lido docs, “Introduction” (n.d.) — read for the current (post-Merge) mechanics of stETH.

10. Sources

Section titled “10. Sources”