A rebasing token changes every holder's balance on a schedule instead of changing its price. In liquid staking that means stETH balances grow each day as rewards arrive, keeping one unit close to one ETH. The balance moves without any transfer being sent, which is exactly what makes the design intuitive in a wallet and hostile to every system that expects a balance to change only when a transaction moves it.
A rebase is a state change with no transaction attached. Anything that cached a balance, priced a position, or recorded a cost basis at the last block is now wrong, and it will not know it.
Scroll to see the full diagram
The balance moves and no transaction happened
Lido's own description of the mechanism is the sharpest one available. Because the rewards are embodied through a balance rebase, users holding stETH will not see a transaction sent to their wallet; the balance simply changes without an accompanying transaction.2 Lido states the cadence as a daily oracle report at 12:00 UTC, and that the balance increases with the current APR.3
That single property is the source of every downstream problem and every downstream benefit. The benefit is legibility: a holder opens a wallet, sees more tokens than yesterday, and needs no explanation of what a redemption rate is. The problem is that transfer events are how the rest of the ecosystem learns that anything changed.
Lido flagged the consequence in the same 2021 post, naming Uniswap, 1inch and SushiSwap as platforms not designed for rebasable tokens where a liquidity provider risks losing part of the daily rewards.2 Five years on, that has hardened into a rule rather than a caveat: the academic survey of the category records that rebase tokens are largely unsupported for trading on automated market makers and are not used as lending collateral at all.4
Shares are the ledger, the balance is a view
Under the surface nothing rebases. Lido's core contract documentation defines the balance as the holder's shares multiplied by total pooled ether, divided by total shares.1 A deposit buys shares. Shares stay fixed. The pooled ether grows, and every holder's displayed balance moves with it because the same numerator is shared out over the same denominator.
Run one round number. Deposit 100 ETH into a pool holding 1,000 ETH against 1,000 shares and you receive 100 shares reading as 100 stETH. The pool earns 30 ETH, so pooled ether becomes 1,030 while shares are unchanged, and your balance reads 100 times 1,030 divided by 1,000, which is 103.1 The price never moved.
This matters for integration work more than it looks. Any contract that needs a stable quantity should hold shares, not balances, and any contract that quotes a balance is quoting a derived number that will be different next block for reasons unrelated to anything either party did. That is the whole integration problem in one sentence.
Rebases go down, and the documentation says so
One line in Lido's contract documentation deserves far more attention than it gets. When an oracle report is applied, the supply of the token is increased or decreased algorithmically, based on staking rewards or slashing penalties on the beacon chain, execution-layer rewards, and fulfilled withdrawal requests.1 Increased or decreased.
Take the same pool and apply a loss. If penalties reduce pooled ether from 1,000 to 990 while shares are unchanged, the 100-share holder now reads 99 stETH. No transfer. No notification. A number that was 100 yesterday is 99 today, and the accounting system on the other side of the integration has to work out on its own that this was a loss rather than a bug.
In our view this is why the rebasing model is a poor fit for anything that has to reconcile. Every explainer describes the mechanism in the direction that goes up. The contract makes no such distinction, and a design that only reads correctly during good periods is a design that has not been tested against the periods it exists for.
Every rebasing token ends up shipping a wrapper
Lido describes wstETH as a value-accruing wrapper that exists mainly as a compatibility layer for integrating stETH into protocols that do not support rebasable tokens, and particularly for bridges to layer 2 networks.6 The survey explains the economics behind that: with a rebasing token the staking reward would have to be divided between borrower and lender in a lending market, whereas with an exchange-rate token all of the accrual lands on the collateral provider, which is why reward-bearing tokens became a dominant collateral asset.4
So the practical shape of the category is settled. The rebasing token is the user-facing surface and the wrapped exchange-rate token is where the liquidity, the collateral integrations and the bridge deployments live. Founders discover this after shipping, usually when the first serious integration partner asks for a version without rebasing.
Decide it before instead. If your venue list contains lending collateral, a layer 2 deployment, or a generic AMM pool, you are shipping two tokens whether you planned to or not, and the exchange-rate token entry covers the one that will end up carrying the volume.
Four things to decide before the first rebase
Publish the share formula and tell integrators to hold shares, not balances. Say plainly, in the same paragraph as the reward mechanic, that the balance can fall and name the three inputs that make it fall. Ship the wrapper at launch rather than as a follow-up, because the integrations you want will ask for it in week one. And settle the reward-recognition question with counsel first: Revenue Ruling 2023-14 holds that validation rewards are includible in gross income in the year the taxpayer gains dominion and control over them, and a design crediting balances daily presents that standard very differently from one where value accrues into a redemption rate.5 Nothing here is tax advice.
There is also a distribution effect worth naming. Lido's early documentation notes that rewards are socialised across all holders, so rebases reach everyone including depositors whose ETH has not yet cleared the activation queue.2 That smooths the experience and it dilutes the rate for holders already active. Socialised reward accounting is a real design choice with a real cost, and it should be stated rather than discovered.
This page is reference material for design work. It is not investment advice, not tax advice, and not a recommendation about any token.
Common questions
How does a rebasing token work?
It changes balances instead of price. Under the surface a holder owns shares, and the displayed balance is shares multiplied by the pool's total ether divided by total shares.1 When rewards land, the pool grows and every balance grows with it. Lido applies this daily at 12:00 UTC through an oracle report, and because no transfer is sent, the balance moves without any transaction appearing in the holder's history.2
Can a rebasing token balance go down?
Yes. Lido's contract documentation states that supply is increased or decreased algorithmically on each oracle report, based on staking rewards or slashing penalties on the beacon chain, execution-layer rewards, and fulfilled withdrawal requests.1 A holder with a fixed number of shares sees a smaller balance when the pool shrinks, with no transaction and no notification. Most explanations of rebasing describe only the direction that goes up.
Why do rebasing tokens break DeFi integrations?
Because most contracts assume a balance changes only when a transaction moves it. A rebase changes it with no transfer event, so cached balances, pool accounting and collateral valuations drift. Rebase tokens are largely unsupported for trading on automated market makers and are not used as lending collateral, since the reward would have to be split between borrower and lender.4 Lido flagged the AMM liquidity risk in its own early documentation.2
Should a new staking protocol use rebasing or an exchange rate?
Decide it against a written list of the venues you need to reach, not against which reads better in a wallet. Rebasing is more legible to end users and blocks most lending, AMM and bridge integrations without a wrapper. An exchange-rate token reaches those venues directly. In practice a rebasing protocol ships both, so the real question is which one you intend to be canonical.
See LST and LRT Tokenomics Design for how this applies in practice.
Sources
- Lido core contract documentation
Lido Docs, 2026
stETH as an ERC-20 rebasing token, the shares formula defining balanceOf, and the statement that supply is increased or decreased algorithmically on rewards, slashing penalties, execution-layer rewards and fulfilled withdrawals. - Lido's stETH: The mechanics of staked ETH
Lido Blog, 2021
Published 1 February 2021, read 3 August 2026. States that a rebase sends no transaction to the wallet, names AMMs not designed for rebasable tokens as a liquidity-provision risk, and describes rewards as socialised across all holders including those still in the activation queue. - What is stETH?
Lido Help Center, 2024
The daily rebase at 12:00 UTC and the statement that the balance increases with the current APR. - SoK: Liquid Staking Tokens (LSTs) and Emerging Trends in Restaking
arXiv preprint 2404.00644v3, 2024
Rebase versus reward LST taxonomy, the finding that rebase tokens are largely unsupported on automated market makers and unused as lending collateral, and the reason accrual splitting makes them unsuitable as collateral. - Revenue Ruling 2023-14
U.S. Internal Revenue Service, 2023
Validation rewards are includible in gross income in the taxable year the taxpayer gains dominion and control over them. - wstETH contract documentation
Lido Docs, 2026
wstETH as a value-accruing wrapper whose balance is unchanged by oracle reports, and its stated purpose as a compatibility layer for protocols and bridges that do not support rebasing.
Last reviewed 2026-08
Know the terms but not sure how they apply to your project? That is what an engagement is for. We design, document, and stress-test the whole token economy inside the Tokenomics Data Room.
100+ projects advised. Complete tokenomics in 4 to 6 weeks.