The Ethena fee switch is close to approval, and the Ethena Foundation will still not buy back a single ENA for now. The reason has been missing from the coverage of the past few days: the buy
The Ethena fee switch is close to approval, and the Ethena Foundation will still not buy back a single ENA for now. The reason has been missing from the coverage of the past few days: the buybacks only begin once the circulating supply of the synthetic dollar USDe reaches $7.5 billion. On August 30, 2026 it stands at roughly $4.07 billion. That leaves a gap of just over $3.4 billion to the first tier, about 84 percent of growth.
Anyone searching for “Ethena fee switch buyback” wants that one number. This article gives it, explains the four-tier schedule behind it, shows where the money for the ENA buybacks comes from, and sets out what two members of the Risk Committee have calculated in their own analyses.
A fee switch is a governance decision that redirects part of a protocol’s revenue away from its current recipients and into purchases of the protocol’s own token. For Ethena that means a fixed share of gross revenue goes to the Ethena Foundation, and 95 percent of the foundation’s net income flows into open-market ENA buybacks. This is revenue sharing in its most direct form: protocol revenue turned into buying pressure for the governance token. The proposal names the basis explicitly as gross protocol revenue, the top line before any distributions are deducted.
Three terms make the arithmetic legible. USDe is Ethena’s synthetic dollar, which holds its value through hedged positions on derivatives exchanges and draws most of its yield from the funding rates on those positions, the balancing payments between the long and the short side. sUSDe is the staked version that passes this yield on to holders. sENA is the staked form of the governance token ENA. How the USDe mechanism works in detail, and what BaFin’s wind-down of the German offering means for it, we took apart in our explainer on the Ethena USDe yield.
The proposal comes from Ethena Labs Research and was published in the foundation’s governance forum on August 27, 2026 at 13:59 UTC. It explicitly replaces every parameter the Risk Committee had put forward before. The old framework from late 2024 set three success criteria that were meant to be met before a fee switch was discussed at all: USDe circulating supply above $6 billion, cumulative protocol revenue above $250 million, and USDe distribution across the five largest centralised derivatives exchanges. That hurdle now falls away and gives way to a pure supply schedule.
The short answer: on the day USDe circulating supply clears the $7.5 billion mark. Nothing happens before that, whatever the vote decides. Blockworks Advisory puts it plainly in its own forum statement: at today’s supply of $4.07 billion the schedule takes nothing, so no part of the document has any immediate effect.
Circulating supply here is neither a price target nor market capitalisation. It is simply the sum of all USDe issued. The figure grows when users mint new units and shrinks when they redeem them. That is exactly what makes the metric checkable for you: Ethena publishes the running numbers on its own transparency dashboard, and independent data providers such as DefiLlama measure the same quantity daily.
For context, $7.5 billion is no invented mark. Circulating supply has passed that level before, first on July 27, 2025 according to Blockworks Advisory. The proposal openly targets a return to the region of $15 billion and names more than $100 billion within five years as its longer horizon. Whether that growth arrives is an open question, and it is the question that decides whether the fee switch ever moves any money at all.
The levy climbs in four tiers alongside USDe circulating supply – the lowest tier has not been reached. The proposal ties the levy rate to USDe circulating supply. The larger the stablecoin, the larger the share of gross revenue that goes to the foundation and from there into ENA buybacks. The schedule reads as follows in the statements from Blockworks Advisory and OAK Research, both members of the Risk Committee:
USDe circulating supplyLevy rate on gross revenuefrom $7.5 billion5 percentfrom $10 billion10 percentfrom $15 billion15 percentfrom $20 billion20 percentOAK Research converts the tiers into dollars using an assumed protocol yield of 6 percent. At 7.5 billion USDe that equates to annual gross revenue of roughly $450 million, of which $22.5 million would go into the buyback pot. At 10 billion it is $600 million of revenue and $60 million of levy, at 15 billion $900 million and $135 million, at 20 billion $1.2 billion and $240 million. Treat these as model values: every line depends entirely on the assumed yield of 6 percent actually materialising.
The schedule follows a stated logic that the proposal calls growth mode. The low entry rate is meant to avoid slowing protocol growth, because a high levy at small supply would make USDe more expensive against competing stablecoins. Only with scale does the share that reaches ENA holders rise.
What matters for understanding ENA tokenomics is that the levy applies to gross revenue, and not to a surplus. The money is not ownerless: it already has recipients today. Who those recipients are decides how expensive the fee switch makes the USDe product.
The vote is being held in the Snapshot space ethenagovernance.eth under the title “ENA Fee Switch”. Snapshot is a voting tool in which voting power is derived from token holdings at a fixed block of the blockchain; casting a vote itself costs no transaction fees. Voting opened on August 27, 2026 at 13:59 UTC and closes on September 2, 2026 at 13:59 UTC.
The interim result shows how little tension the matter itself carries. Queried through the Snapshot programming interface on August 30, 2026, 17,014,325 ENA stood in favour, zero against and zero abstentions, spread across 76 votes cast. The quorum, the minimum voting power without which a proposal fails whatever the result, sits at 5,000,000 ENA and is therefore more than three times covered. The three Risk Committee members who have spoken up in the forum back the proposal as well: Blockworks Advisory, OAK Research and Kairos Research.
For you that means two things. If you hold ENA and want to object, the window runs only until September 2 at 13:59 UTC. And if you are minded to react to a price move that recent coverage has pinned on the word buyback, it is worth knowing that this vote triggers no buyback at all and merely fixes the conditions for one. Between the decision and the first token bought lie $3.4 billion of growth.
Blockworks Advisory has disclosed which pots the levy is drawn from. Over the past 30 days the protocol distributed an annualised $175 million across three running lines: 34.3 percent to sUSDe staking, 34.3 percent to partner payouts and 31.4 percent to the Aave liquid leverage line. The levy is deducted proportionally from all three. On this calculation a rate of 10 percent takes $6.0 million a year out of the sUSDe distribution and $11.5 million out of the two reward lines combined.
Blockworks puts the total withdrawal per tier at $8.8 million a year at 5 percent, $17.5 million at 10 percent, $26.3 million at 15 percent and $35.1 million at 20 percent, each based on the distribution of the past 30 days. Minting fees of $8.4 million a year are not included; they make up 4.7 percent of gross revenue and are carried as a line of their own.
Savers should hold on to a distinction that coverage regularly gets wrong. The levy works proportionally: on the Blockworks reading, a rate of 10 percent lowers the sUSDe APY, the annualised yield on staked USDe, by one tenth of whatever its value happens to be, regardless of how the distribution looks in detail. At an assumed yield of 8 percent, roughly 7.2 percent would remain. There is no deduction of 10 percentage points.
That holds equally at every tier. If you spread your stablecoin yield across several providers anyway, a look at our comparison of the best staking and rewards platforms is worth the time before you reallocate; the gaps between the offers are wider than the deduction discussed here.
Blockworks names one side effect that the proposal itself omits. Across the 118 days on which the schedule would have been active in the backtest, the levy pushed the sUSDe yield below that of the competing sUSDS product on eleven days. The analysts therefore recommend publishing this figure on an ongoing basis, while judging a hard floor more expensive than the benefit it would bring. Anyone who switches provider on the strength of such comparisons should watch the supervisory side; our overview of regulated crypto exchanges shows which venues in Germany operate under regulation.
An unlock is the scheduled release of tokens that were previously locked for team, investors or programmes; the selling pressure that results from it is known as the overhang. That is precisely the measure of whether a buyback programme amounts to more than a headline.
Blockworks Advisory ran the draft over 705 days, from September 19, 2024 to August 24, 2026, day by day with the revenue that actually accrued and the prices that actually applied. The result: during the active phases the schedule captures $52.7 million a year, and averaged across the whole period $8.82 million. Set against an ENA market capitalisation of $1.57 billion, that is a buyback yield of 3.36 percent in the active state and 0.56 percent on average.
Against those amounts stand planned gross releases of $512 million a year through to April 2028. On the Blockworks calculation the buyback absorbs roughly one tenth of that in the active state, and less than 2 percent across the full cycle. That is the most honest number in the whole exercise, and it comes from a supporter of the proposal: the fee switch shares protocol revenue with token holders, and it was never built to hold up the ENA price.
Approved and still shut: only from 7.5 billion USDe does the line towards the ENA buyback open. A schedule that reads the daily value of circulating supply can be jumped with one large mint. OAK Research states the problem clearly: a single dollar then decides between an annual buyback budget of around $30 million and one of around $60 million, and the moment of crossing becomes visible to the market in advance.
Blockworks Advisory therefore proposes a rolling 14-day average as the measure, meaning the mean of circulating supply over the preceding two weeks in place of the daily close. In the backtest this window would have started the buyback seven days after the $7.5 billion mark was first crossed; a 90-day window only 34 days after. At the same time the shorter window captures more revenue, $8.82 million against $7.66 million per panel year, without triggering additional tier changes. OAK Research has since withdrawn its own proposal of a 30-day ramp and supports the 14 days.
For you as an observer that is the decisive reading aid: if USDe circulating supply briefly jumps above $7.5 billion at some point, the buyback is not yet triggered. Only a two-week average above the mark counts, assuming the vote is implemented in this form.
The vote under way decides on the supply thresholds alone. Everything that carries the mechanism through daily operation has so far only been announced, by OAK Research, as a series of separate proposals over the coming weeks. Five points are outstanding:
Ethena points to its own transparency dashboard for tracking. Until the points above are settled, the fee switch remains an approved framework whose operating manual has yet to be filled in.
The mechanism ties the value of ENA to the protocol’s business volume in a binding way for the first time. That is a genuine change to the ENA tokenomics, and it is publicly verifiable, because the trigger quantity, USDe circulating supply, can be measured at any moment. Until then the plain position holds: no revenue share, no purchases.
Keep three things apart. First, the vote on thresholds is no buy signal, because the lowest tier sits around 84 percent of growth away. Second, the levy costs sUSDe part of its yield, which works in opposite directions for savers and for holders. Third, the captured amounts come out small measured against the coming token releases, even in the active state.
None of these statements is a price forecast, and none of the analyses quoted names a price target. If you are reviewing your holdings anyway, watch the tax side: buyback programmes change nothing about your holding periods, but every reallocation does. Anyone who reallocates repeatedly should record each move with its date and its value, or the tax return turns into reconstruction work.
You can read the primary sources yourself: the running Snapshot vote “ENA Fee Switch” with its interim result and quorum, and the proposal together with the Risk Committee statements in the Ethena governance forum.
(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.)
