Ethereum staking yield has halved since 2021. Two things did it.
Lido has posted a rewards report almost every day since April 2021. We read all 1,978 of them, then went to Ethereum's own data to find out why they now pay less than half of what they did.
- 2.38%
- 10.8x
- 6.4%
- 74%
- 1,978
- 136,989
01 · One reportThe same report, three years apart
On the last day of September, a little after noon UTC, Lido posted its daily report. It told the stETH contract how much the pool's validators, the nodes that do the staking, had earned since the day before. Every stETH balance grew to match. The pool held just under ten million ETH, and that day it added about 670.
Three years earlier, to the day, the same report brought in about 945 ETH on a smaller pool.
Lido is the largest liquid staking pool on Ethereum. You send it ETH to stake, and you get stETH back, a liquid staking token (LST) you can trade or use in DeFi while the stake keeps earning. The earning shows up as a rebase: once per day your balance ticks up.
That tick is far smaller than it used to be. For a holder, the one on the last day of September was less than a third of what the first report in our data paid, back in April 2021. Over a full year, Lido's staker APR was 5.26% when our data starts and 2.38% this year, so the Ethereum staking yield has more than halved. We read every report Lido posted in that time, then looked at Ethereum itself for the cause. There are two, and they are not Lido's doing.
02 · The basicsWhere staking yield comes from
ETH staking rewards come from two places. The first is the protocol reward, which is new ETH that Ethereum pays every validator for doing its job. The second is tips, the extra that users pay to get a transaction into a block. A validator collects tips only when it is picked to propose a block, and our piece on the MetaMask Staking incident shows what happens when those tips are diverted.
Lido keeps 10% of the two together and passes the rest to stETH holders. Lido's docs state that fee, and the reports show exactly 10%.
Tips have not always gone to stakers. Until the Merge, the switch to proof of stake in September 2022, blocks were made by miners and tips went to them, so stakers earned the protocol reward and nothing else. From that day tips went to stakers too.
03 · The fallFrom over 6% to just over 2%
When our data starts, stETH was paying over 6% per year. A year later it was under 4%, with no drama on the way. The reports just kept getting a little smaller.
The Merge turned that around for a while, because tips arrived, and two months later the yield was back over 6%. Then both parts shrank together. In the last month of our data the yield was 2.25%.
| Year | Yield | Protocol | Tips | Tips share |
|---|---|---|---|---|
| 2021 | 5.26% | 5.26% | 0.00% | 0.0% |
| 2022 | 4.48% | 3.85% | 0.63% | 14.1% |
| 2023 | 4.35% | 3.04% | 1.31% | 30.2% |
| 2024 | 3.17% | 2.56% | 0.61% | 19.2% |
| 2025 | 2.81% | 2.48% | 0.33% | 11.7% |
| 2026 | 2.38% | 2.24% | 0.14% | 5.9% |
"Halved" depends on where you start. Against the first year in the table the yield is down by more than half. Against the second it is down by a little less than half, and against the best month since tips arrived by almost two thirds. In all that time the rate never went backwards. Each report is an event on-chain, and not one of them lowered it.
04 · Cause oneMore ETH staked, less paid per ETH
The first reason is built into Ethereum. It sets the protocol reward by a rule: the more ETH is staked in total, the less each staked ETH earns. Quadruple the total and the reward per ETH halves.
The total did far more than quadruple. When Lido's reports begin, about 4 million ETH was staked, and at the start of this October it was 43.7 million, almost eleven times as much. By the rule alone, that cuts the reward per ETH to less than a third of what it was.
The steep climb in 2023 came after withdrawals opened in April of that year. Staking became a two-way door, and the staked total almost doubled that year.
Lido's stakers got what the rule says, give or take a little. In every month we could check, the protocol reward they received sat just under the rule's figure, usually by a few percent. A gap like that is no surprise, since a pool always has some ETH waiting to start or to leave and validators miss a duty now and then. We did not measure which of the two matters more.
05 · Cause twoThe tips dried up
The second reason moved faster. Tips are what users pay on top of the base fee, which is burned. We added them up for every block since the burn began. In the first year after the Merge, users paid more than 200,000 ETH in tips. In the latest year they paid about a quarter of that.
Ethereum did not go quiet. Blocks carried almost twice as much gas in the latest year as they did three years before. Blockspace just got cheap. The average base fee went from about 25 gwei to about 0.3 gwei, and the average priority fee from about 5 gwei to under 1.
Two changes on the record line up with that. Blobs, added in early 2024, gave rollups a cheaper place to post their data. And blocks got bigger. A block now carries about 30 million gas, twice what it did until early 2025. We did not measure how much of the fall each one explains.
Stakers get that pot second hand. Since the Merge a validator can hand the job of building its block to a specialist block builder, who takes the tips and whatever else it can make from ordering transactions (MEV), then pays the validator a bid for the slot. So the pot is the size of the fee market, and Lido's reports show what reached stakers.
What reached them fell even faster, from about 100,000 ETH of tips at Lido in the first year after the Merge to under a sixth of that in the last twelve months. As a share of all rewards, tips went from a third to about 6%.
06 · The splitWhich part of the yield fell more
Take the first year after the Merge and the latest year. The yield fell by almost half between them. About three fifths of that drop came out of the tips part and the rest out of the protocol reward. The two causes overlap here, because the tips that are left are shared across about three times as much staked ETH as at the Merge.
Go back to the start of our data and the answer changes, because there were no tips for stakers then. The whole yield was protocol reward, and that part alone has since more than halved. Over the full five years, then, the growing staked total did the damage. Tips arrived with the Merge and lifted the yield for about a year. They have mostly gone again.
07 · Lido's cutThe fee did not move
Could Lido simply be taking more? No. Its cut did not change in five years, so it explains none of the fall. In ETH the cut has come to about 137,000 since our data starts, while stETH holders received about 1.24 million. Both peaked three years ago and have been shrinking since, though the pool holds more ETH now than it did then.
08 · Everyone elseThe same fall in every large staking token
The other big pools show the same fall, since they all run validators under the same rule and in the same fee market. We read the rate of the ten largest other ETH staking tokens, and this year they paid between about 2% and 3% per year. All of them paid less than the year before.
| Token | 2024 | 2025 | 2026 |
|---|---|---|---|
| wstETH | 3.23% | 2.85% | 2.40% |
| wBETH | 3.08% | 2.72% | 2.49% |
| weETH | 2.80% | 2.69% | 2.45% |
| cbETH | 2.81% | 2.73% | 2.54% |
| rsETH | 3.15% | 2.75% | 2.46% |
| rETH | 2.81% | 2.59% | 2.19% |
| LsETH | 3.04% | 2.89% | 2.41% |
| mETH | 4.17% | 2.92% | 1.98% |
| osETH | 3.15% | 2.63% | 2.22% |
| ETHx | 3.08% | 2.79% | 2.50% |
| sfrxETH | 3.67% | 3.28% | 2.90% |
That list includes two liquid restaking tokens (LRTs), weETH and rsETH. The issuer of weETH pulled its stake off EigenLayer this year. Whatever the wrapper, the collateral people borrow against now earns about the same rate.
09 · The answerWhy the yield halved, and what moves it now
So why has the yield on staked ETH halved? Because almost eleven times as much ETH now shares the protocol's reward, and because the tips that propped it up for a while have mostly gone.
If you hold stETH, about 96% of what you earn today is protocol reward, and it follows the staked total. More stake pushes your yield down, and stake leaving pushes it up. Switching to another LST changes little, since the big ones all pay within a point of each other. The other 4% is tips, and it follows demand for blockspace. A busy month for Ethereum is a better month for stakers, which was true in May 2023 and is still true at a smaller scale.
Both are set by Ethereum and the people using it. No pool controls either one.
10 · How we measuredWhat we counted, and what we left out
We read Lido's own reports on-chain. Each one logs how much the pool earned and, since the Merge, how much of it was tips. We checked the result against the stETH rate itself, and the two agree in every year.
For the fee market we added up what users paid and what was burned in every block. For the staked total we counted validators.
All figures run to September 30, 2026. The 2021 figures start on April 30, 2021, the first Lido report that logs the pool's totals; Lido itself began in December 2020. The 2026 figures cover nine months. Days are UTC.
Yield here is APR: each report's growth added up and scaled to a year, after Lido's fee, with no compounding. A compounded figure runs slightly higher, which is why the wstETH rate in the token table shows 2.40% for 2026 against 2.38% here.
Three figures here describe the recent yield, each for its own period: 2.38% for January to September 2026, 2.45% for the twelve months to September 2026, and 2.25% for September 2026 alone.
Lido's reports log its fee from May 2023. Before that the fee is the 10% the contract applied. Tips are logged from the day of the Merge. On five busy days in November 2022 and March 2023, Lido paid part of the tips out over the following days.
For the staked total we counted validators at each month end. Until May 2025 each held 32 ETH. Since then a validator can hold up to 2,048 ETH, so a count no longer gives the total and our series stops in April 2025. The last point is one reading taken on October 1, 2026.
The rule assumes a validator that never misses. In the months we could check, the protocol reward paid to stETH holders ran between 1.5% and 14% under the rule after Lido's fee. We did not measure why.
Tips on Ethereum are the fees users paid minus the fees burned, added up over every block. Payments made to block builders outside the fee are not in that figure. We found and corrected two small faults in our own fee data; together they came to under 2% of the latest year's tips.
Token yields come from the growth of each token's rate between year-end blocks and are compounded. They are after each issuer's own fee.
We did not speak to Lido or to any other party named here. Nothing here is a forecast.
11 · The recordThe contracts and reports behind the numbers
| What | Address or transaction |
|---|---|
| stETH (Lido) | 0xae7a…fe84 |
| wstETH | 0x7f39…2ca0 |
| Lido's report contract | 0x852d…3cee |
| Its report contract to 2023 | 0x442a…43fb |
| Lido's tips vault | 0x388c…9297 |
| Staking deposit contract | 0x0000…05fa |
| First report, Apr 30, 2021 | 0xf88e…cc22 |
| Merge day, Sep 15, 2022 | 0x6415…c45e |
| First report, new format | 0x7cdd…a263 |
| Report, Sep 30, 2023 | 0x5f2d…6d90 |
| Report, Sep 30, 2026 | 0xa715…f31b |
FAQ
What is the Ethereum staking yield in 2026?
Lido's stETH paid 2.38% per year in the first nine months of 2026, after Lido's fee. The ten largest other ETH staking tokens paid between about 2% and 3% over the same months.
Why is the Ethereum staking yield falling?
For two reasons. The ETH staked on Ethereum grew almost elevenfold in five years, and the protocol pays less per ETH as that total rises. Tips also shrank, from about a third of Lido's rewards in the first year after the Merge to about 6% in the twelve months to September 2026.
How much has ETH staking yield dropped since 2021?
At Lido it went from 5.26% per year in 2021 to 2.38% in the first nine months of 2026, a fall of more than half. Measured from 2022, the fall is a little under half.
What are tips in Ethereum staking?
Tips are what users pay on top of the burned base fee to get a transaction into a block. Since the Merge in September 2022 they go to block builders and validators. Before it they went to miners.
How much does Lido charge?
Lido keeps a tenth of staking rewards. Every report that logs the fee shows exactly that share, and since April 2021 it has come to about 137,000 ETH.
How much ETH is staked on Ethereum?
About 43.7 million ETH at the start of October 2026. That is almost eleven times what was staked five years earlier, when Lido's reports begin.
Does every liquid staking token pay the same yield?
Close to it. In the first nine months of 2026 the eleven largest ETH staking tokens paid between about 2% and 3% per year, because they all run validators under the same rules and share the same fee market.
Ask about any staking token in plain English
Every figure above came from data anyone can query. The Bitquery MCP server puts it behind an AI assistant, so you can ask what a staking pool reported today, how much a block paid in tips, or who holds a token, without writing the query yourself.
This article is provided for informational and educational purposes only and reflects analysis of publicly available on-chain data as of the dates indicated. It does not constitute legal, financial, compliance, or investment advice, and nothing in it is a recommendation to buy, sell, hold or stake any token or asset.
The findings describe contracts, reports and transactions observed in Ethereum data between April 30, 2021 and September 30, 2026. Yields are historical and say nothing about future returns.
References to Lido, Binance, Coinbase, ether.fi, Kelp, Rocket Pool, Liquid Collective, Mantle, StakeWise, Stader, Frax or any other named party describe what the record shows about contracts and tokens linked to them. They are not statements about any company's solvency, security practices, disclosures or conduct, and nothing here asserts that any party acted unlawfully or improperly.
Nothing herein should be relied upon as a definitive determination of fact. Readers should conduct their own independent verification before taking any action. The authors and publisher accept no liability for any loss or damage arising from reliance on this material. All trademarks and company names are the property of their respective owners.
Reported by Gaurav Agarwal for Bitquery Research, with AI tools; every figure was checked against the raw data.