What is a liquid staking token?

A liquid staking token is an ERC-20 that represents a claim on staked ETH. The staker gets staking yield; the token is freely transferable and can be used throughout DeFi as if it were plain ETH. LSTs solve one of the least glamorous but most important problems in proof-of-stake: staking normally locks your capital, and locked capital is expensive capital.

The problem LSTs solve

Native ETH staking requires depositing 32 ETH to the staking contract. The ETH is then committed to validating the network, and until Ethereum's Shanghai upgrade in April 2023, it could not be withdrawn at all. Even now, withdrawing from native staking goes through a queue that can take days to weeks during periods of heavy exits.

For a long-term holder willing to earn 3 to 4 percent staking yield, that lockup is fine. For anyone who wants to use their ETH as collateral in DeFi, hedge it, or move it quickly, the tradeoff is awful. The choice was either: stake and give up liquidity, or do not stake and give up yield.

Liquid staking resolves the dilemma by separating custody of the underlying asset from the right to transfer a claim on it. The user deposits ETH with a protocol like Lido. The protocol runs validators with the ETH. The user receives an equal amount of stETH in return. The stETH is liquid — tradable on Uniswap and Curve, usable as collateral on Aave, stackable in yield vaults.

stETH and its cousins

Lido's stETH, launched in December 2020, is the category-defining LST and remains the largest by a wide margin. It is rebasing: balances grow daily as staking rewards accrue. wstETH is a wrapped version with a static balance and an ever-increasing exchange rate — more convenient for DeFi integrations that dislike rebasing tokens.

Rocket Pool's rETH is the decentralized alternative. Rocket Pool node operators put up half the collateral for each validator, and users supply the other half; rETH's design makes it more distributed across operators and less vulnerable to a single protocol failure, at the cost of lower scale.

Coinbase's cbETH and Binance's bETH are custodial LSTs. They work, they have deep liquidity, and they reintroduce exactly the centralized-custody risk that crypto was supposed to avoid. Frax's frxETH splits the role: sfrxETH earns the staking yield, while frxETH is the liquid wrapper — a cleaner separation than stETH's rebasing design.

The depeg risk

An LST is supposed to trade at the value of the ETH it represents. It usually does, but not always. stETH traded at a 4-to-7 percent discount to ETH in mid-2022 during the collapse of Three Arrows Capital and Celsius, because those firms held enormous stETH positions financed with leverage, and liquidations forced them to sell stETH they could not redeem (Ethereum's Shanghai withdrawals were not yet live).

Since Shanghai, stETH can be redeemed at 1:1 through the Lido protocol itself (with a queue), which pins the price. Depeg risk is much lower than it was in 2022, but it is not zero. Smaller LSTs with thinner liquidity and less robust redemption paths can still depeg meaningfully in stress scenarios.

The composability multiplier

LSTs are where most of DeFi's yield stacking comes from. A user can: deposit ETH into Lido, receive stETH, deposit stETH into Aave as collateral, borrow USDC against it, swap the USDC for more ETH, and deposit that into Lido again. This loop — looped staking — produces leveraged exposure to staking yield. It also produces leveraged exposure to an stETH depeg, which is how several leveraged stakers got liquidated in 2022.

LSTs in Curve pools (stETH/ETH, wstETH/ETH) are among the deepest pools in all of DeFi. They provide cheap conversion between LST and ETH and earn LPs trading fees. The stETH/ETH pool has traded multiple billions per day in active periods.

Centralization concerns

Lido's dominance — at various points controlling over 30 percent of all staked ETH — is one of the most-discussed governance issues in Ethereum. If a single staking provider exceeds 33 percent of the validator set, it could unilaterally disrupt finality. At 50 percent, it could censor the chain. At 67 percent, it could finalize whatever it wanted.

Lido itself is a DAO with its own governance and a curated operator set, which is better than one company running it, but still concentrated relative to Ethereum's core value of decentralization. Ongoing debates concern whether Lido should self-limit, whether the protocol should include cap-like mechanisms, and whether the broader LST market should fragment.

Restaking and LRTs

Liquid restaking tokens (LRTs) are the LST model extended to restaking. EtherFi's eETH, Renzo's ezETH, Kelp's rsETH are LRTs that represent restaked ETH inside EigenLayer (see the restaking explainer). They stack another layer of yield and another layer of risk. The composability is powerful; the systemic exposure is real.

Why it matters

LSTs are one of the few DeFi primitives that genuinely delivered what they promised: liquid, composable exposure to staking yield, usable across the ecosystem as a base asset. They are now foundational infrastructure for Ethereum DeFi — measured in tens of billions of dollars of deposits, with pools among the deepest on-chain.

They are also a concentration risk and a systemic dependency. Any user holding stETH as collateral, any protocol using stETH pricing for liquidations, any position looped through LST yield is exposed not just to ETH but to the operational health of a specific staking provider and its liquidity. Understanding what is actually under the hood of an LST is the difference between using a powerful tool and betting on a brand.