Until 3 September 2026, 01:00 UTC, the Sushi DAO is voting on how much of the protocol's fee income still reaches xSUSHI holders. The proposal caps buybacks at 1 percent of protocol fees plus
Until 3 September 2026, 01:00 UTC, the Sushi DAO is voting on how much of the protocol's fee income still reaches xSUSHI holders. The proposal caps buybacks at 1 percent of protocol fees plus 1 percent of perpetuals revenue. Everything above that moves into a newly created protocol reserve and to the operations team. Anyone staking SUSHI has until that date to act on it.
The process has drawn little coverage so far. This text therefore relies on the voting record itself, on independent fee data, and on an own measurement of the voting-power contract taken on 30 August 2026 at 06:37 UTC. Every figure below was recorded at that moment.
The proposal on the table is SUSHI Tokenomics Restructure: SUSHI Reserve, Revised xSUSHI Buybacks, and Protocol-Deployed Liquidity Deployment to Robinhood Chain, filed in the governance space sushigov.eth. It opened on 27 August 2026 at 01:00 UTC and closes on 3 September 2026 at 01:00 UTC. Three options are on the ballot: for, against, abstain. In DAO parlance, filings of this kind are proposals: drafted texts that a protocol's community votes on under fixed rules.
xSUSHI is the token you receive when you deposit SUSHI in the so-called Sushi Bar; it represents your share of the pot from which the protocol funds buybacks. Protocol fees are the trading fees the protocol charges the users of an exchange and then distributes under fixed rules. Those rules are exactly what the proposal changes.
In the proposal's own wording, revenue has so far flowed primarily toward xSUSHI buybacks. A fixed split is to apply in future, and in it stakers are the smallest of the three items.
Tokenomics is the full set of rules by which a crypto project issues, distributes and reclaims its tokens: supply, distribution, buybacks, lock-up periods. For you as a holder it is the difference between a token that is continuously assigned revenue and one that carries nothing but voting rights. What a protocol pays out in staking rewards is therefore a question of allocation, of how income is divided, and not a question of technology.
A DEX such as Sushi is an exchange without a central custodian: prices form in liquidity pools, contract balances that other users contribute as liquidity providers. Those users earn from the trading fees. What is passed on to stakers on top of that is a pure governance decision, not a technical property of the protocol. That is why a vote can change it.
This is the point at which a DAO formality becomes a question about your portfolio. Anyone budgeting for staking income should know that at many protocols the payout rests on a resolution rather than on a contract that guarantees it. How widely the models differ is clear from the comparison of staking platforms.
The proposal fixes three streams. First, 1 percent of protocol fees plus 1 percent of all revenue from the perpetuals business goes into monthly xSUSHI buybacks. Second, 4 percent of protocol fees goes into weekly SUSHI purchases for a newly created SUSHI Reserve, whose holdings remain protocol-owned and are to be disclosed through a public dashboard. Third, all remaining protocol fees, migration fees and other income goes to Sushi Ops, meaning development, incentives, partnerships, liquidity programmes, security and operations.
Perpetuals are futures contracts with no expiry date, in which a recurring settlement payment between the long and short side keeps the price anchored to the spot market. For Sushi they are a separate revenue stream, and the proposal treats them separately from trading fees. What matters most for holders is that this revenue stream, too, feeds the buybacks at only 1 percent.
One point of context: the reserve is described in the proposal text explicitly as a treasury function, not as a means of supporting the SUSHI price. No commitment to holders comes with it.
The proposal is not limited to future income. It names explicitly all protocol fees already accrued and held by or for Sushi that have not yet been distributed. According to the text, the same applies to accrued perpetuals revenue as well as to migration fees and other protocol income.
In practice this means that a balance still awaiting distribution under the old rule today would, if the proposal passes, be distributed under the new split. Anyone who assumed accrued fees would still reach stakers under the previous formula should be aware of this paragraph. Historical claims from the MasterChef LP migration remain expressly unchanged under the proposal text; they are unaffected by the switch.
One large lever, six small ones: seven wallets voted in the 25 August signal vote, and one of them supplied 99.76 percent of the result. A signal vote, a non-binding sounding of opinion, ran ahead of the binding ballot. That preliminary stage closed on 25 August 2026 at 20:00 UTC with 5,213,018.14 votes in favour, none against and no abstentions, cast by seven wallets.
Of those, 5,200,299.37 votes came from a single address, 0xFf467361cC46dB493588cCd60733e391e856E492. That is 99.76 percent of the result. The remaining six wallets contributed 12,718.77 votes between them.
That order of magnitude can be checked independently of Snapshot. Voting power in the space is measured through the contract 0x62d11bc0652e9D9B66ac0a4c419950eEb9cFadA6 on Ethereum. An own query of that contract on 30 August 2026 returned a balance of 5,200,299.37 units for the address in question against a total of 19,399,067.14 units, the total supply of the voting right. That equals 26.81 percent of all voting power.
Concentration of this kind is not unusual in DAOs with low participation and is no accusation in itself. The finding says nothing about the intentions behind the address and nothing about whether the proposal is good or bad on the merits. It describes the mechanics alone. How consequential those mechanics can become when voting power and participation diverge was shown by the governance attack on the BONK DAO, in which a formally correct vote moved 20 million dollars out of the treasury.
A quorum is the minimum number of votes that must be gathered for a ballot to be valid at all. In the Sushi space it stands at 5,000,000 votes.
From the two measurements above follows the statement that carries the whole process: the address holding 5,200,299.37 votes meets the quorum on its own, at 104 percent. Without it, a vote in this space comes together arithmetically only if a great many small holders join forces.
A look at the space's voting history supports this. The four closed proposals that preceded the current one in the same space ended at 5,200,359, 5,215,215, 5,342,865 and 5,216,119 total votes. All four sit just above the quorum threshold and in the order of magnitude of this one address. That is an observation drawn from public voting data, not a claim about coordination.
The binding implementation vote stood on 30 August 2026 at 06:37 UTC at 12,289.49 votes in favour, none against and no abstentions, cast by two wallets. That is 0.25 percent of the quorum.
The large address from the signal vote had not voted at that point. Two things follow, and both are relevant to you as a holder. First, four days before the close the proposal is far from valid. Second, a single vote would clear the quorum in one move.
The difference between a signal vote and an implementation vote is no formal aside here. The first measures sentiment, the second puts the rule into force. The proposal has already cleared the first stage.
The proportions the percentages conceal: the fee pot of the past 30 days and the share of it that reached xSUSHI holders. Sobriety pays here, or the news turns into scaremongering. Independent fee data put Sushi's protocol fees over the past 30 days at 1,331,834.81 US dollars, and at 6,690.43 US dollars over the past 24 hours. The portion of that reaching holders as revenue came to 9,098 US dollars in the same 30-day window, 2,492 US dollars over the past seven days and 199 US dollars over the past 24 hours.
What counts, then, is less the size of the fees than the part of them that lands with xSUSHI holders as a distribution. For SUSHI holders who keep their SUSHI tokens unstaked in a wallet, the proposal changes nothing directly.
The ratio is the actual finding: around 0.68 percent of protocol fees have lately been reaching holders. The proposal fixes 1 percent. The honest summary is therefore less dramatic than the headline of the process suggests: the vote largely cements a state of affairs that already applies in practice, and shifts the rest permanently and explicitly to the reserve and to operations.
For a sense of scale: total capital held in the protocol stood at around 95.9 million US dollars, with the SUSHI price at 0.1952 US dollars. What the resolution does to the price is expressly not stated here; this text makes no price forecast.
The case works as an object lesson because it shows three risks that sit in almost every staking model and have nothing to do with the token's price.
Rule-change risk. A payout introduced by resolution can be cut by resolution. Anyone calculating a yield over years is calculating a rule that can go to a vote at any time.
Participation risk. When two to seven wallets decide over a protocol holding capital in the tens of millions, your voting right as a small holder is effectively without force. The right remains formally intact and loses its economic effect.
Expectation risk. Between the advertised yield of a staking offer and what actually flows there can be a gap you only see in the raw data. At Sushi it was lately 199 US dollars in a day, spread across all stakers combined.
On the tax and regulatory side, who offers the staking and under which licence it happens matters on top of all this.
There is no blanket answer, but the question can be sharpened. Staking is worth it where the payout is tied to a durable revenue source and the rule cannot be toppled by a simple majority of a handful of wallets. Both are checkable before you deposit.
Check first where the returns come from: from genuine protocol fees or from the issuance of new tokens. Check second how many wallets carried the quorum in recent votes. Check third whether the payout rule sits in the contract or in a governance resolution. And check fourth how long your capital is locked up should you want to exit while a process is running.
Anyone working through those four points on an offer has learned more about their risk than any advertised annual yield could teach them.
The same proposal contains a second resolution point that is easily missed. It authorises the protocol to shift part of its protocol-owned liquidity from existing EVM networks to Robinhood Chain, starting with an ETH-USDG position in the V3 model. The text names a target of roughly 10 to 20 million US dollars of protocol-owned capital on that chain, executed in tranches and limited to the protocol's own pools.
For you as a holder this is notable for two reasons. For one, it moves capital into an ecosystem whose use you would have to assess separately. For another, this point sits in the same proposal as the fee cut: whoever votes on the yield votes on the move at the same time. The procedure makes no provision for a separate decision.
Concretely and without haste: open the ballot page first and see whether the quorum has been reached in the meantime; the state of play may change within minutes once the large address votes. Then check in your wallet whether you hold xSUSHI or have SUSHI sitting unstaked in circulation, because only the staked position is affected by the buyback rule.
Next, record what your position actually returned over recent weeks. That figure, not the advertised yield, is the yardstick for the decision. And document deposits and withdrawals cleanly, because staking income is recorded for tax purposes and reconstructing it later is laborious.
The primary sources to read up on: the voting record in Sushi's governance space and the accompanying forum post on the proposal.
(As of August 30, 2026. This article is not investment advice. Prices and fee structures change; check the terms with the provider before you buy.)
