Balancer's own council member proposes shutting the protocol and handing the treasury back

What happened

A proposal to shut Balancer down and distribute its treasury to BAL holders was posted on the protocol's governance forum at 19:28 UTC on 14 September 2026, under the account name Marcus. The author discloses in the post that he sits on the Treasury Council and on the multisigs that execute what governance approves, that he was added to the Council under BIP-918, that he told the Council and the largest holders what he intended to propose before publishing it 'so they could push back before it was public' and asked them not to trade on it, and that he will redeem on the same terms as any other holder. The Block reports him as a former chief executive of Balancer Labs. The case is made on three numbers. 'Monthly burn today, all in: about $150k a month. Monthly protocol revenue: about $30k in August, down from $97k in June, most of it from v2.' The treasury is 'at least $9M at current token prices', which the post attributes to kpk, the DAO's treasury manager, as the managed treasury, with further assets across other addresses being inventoried and to be published before the distribution opens. The post says the April 2026 restructuring was executed as approved, that v3 is live and operating as designed, and that 'Some generated interest. None converted into sustained revenue growth.' It says BIP-918 set a review for this case and that this proposal is that review brought early, because 'Waiting for the calendar would change the numbers, not the conclusion, and every month of waiting is spent from the treasury.' The timetable is specific. Contributor notice started on 27 August and ends on 31 October. On 30 October 2026 pools that can be paused are paused and move to withdrawals only, with recovery mode enabled where the contracts require it and the protocol fee set to zero on pools that cannot be paused; bug bounty coverage ends the same day. From 1 November a transition team runs on a wind-down budget of $150,000 to May 2027, $30,000 thereafter, and a $220,000 reserve drawn only if needed, $400,000 in total, replacing the remainder of the BIP-918 allocation. An implementation specification for the claim contract is published for comment by the end of February 2027 and the contract is audited before use. Distribution runs in two rounds and a sweep. Round one opens at the end of May 2027, by which time every veBAL lock in existence at the time of the post has expired, and runs six months: a holder burns BAL and receives a pro rata share of the treasury in kind, in the tokens actually held, measured at an opening snapshot announced at least two weeks ahead and audited. The denominator is circulating supply as BIP-919 defined it, total supply less BAL the treasury holds and less the two Balancer Labs safes. Round two, by the end of January 2028, airdrops the unspent budget, later arrivals and the unredeemed share to the addresses that redeemed in round one in proportion to what they redeemed, and 'an address that did not redeem in round one has no share in round two'. A final sweep follows at the end of July 2028. The BIP-919 buyback, which committed the DAO to buy BAL at net asset value up to 35 per cent of the treasury, is cancelled, and the BIP-687 bug bounty earmark of $1m USDC is superseded. tetuBAL, an immutable permalock that never becomes BAL, is handled by paying its holders of record at the block of the post BAL equivalent to half the measured amount. The Snapshot vote runs from 25 to 29 September against a quorum of 5m BAL. A second post appeared on the forum at 02:17 UTC on 15 September, titled KEEP BALANCER ALIVE!, which The Block presents as a competing proposal. It is from a trust-level-zero account whose first two lines read 'Author: Wise Enthusiast (@Wise_Enthusiast on X), founder of Wise / WiseSoft LLC' and 'Conflict of interest: I founded the company on the other side of this deal', and it asks the DAO to put up to $7m of USDC into that company's product at a stated 20 per cent yield plus a share of its net revenue. The wind-down post separately says contributors are preparing a continuation 'under a new name', that the author supports it 'as a fork', and that it is posted and decided on its own; that document is not the one The Block is describing.

Why it matters

This is the clearest statement of what a protocol treasury is for that the governance record has produced. The post's argument is not that Balancer is insolvent, and it says so: 'The argument is not that Balancer is running out, but that what remains belongs to holders and should reach them.' It is that continuing spends other people's money to arrive at the same place later, and that the choice in front of holders is whether the treasury reaches them while it is substantial or is spent first on a path already tried. A DAO that can put that question to a vote, and price both branches, is exercising something most token structures assert and few demonstrate. The revenue numbers are the part worth keeping. Protocol revenue of about $30,000 a month against a burn of about $150,000 is a ratio, and the detail that most of the revenue still comes from v2 rather than v3 is the finding: the newer architecture shipped, worked and did not earn. That is a result about automated market making as a business rather than about one team, and it arrives with a specific counterfactual, since the April restructuring had already cut costs, ended emissions and routed revenue to the DAO. The plan was executed and the revenue did not follow. The mechanics are where the economics get interesting, because a burn-for-treasury redemption converts a governance token into a claim on a wind-up estate, and the post is careful that what a holder receives 'is not BAL and does not resolve into BAL'. That gives BAL an arithmetic floor and a deadline at the same time: from the end of May 2027 a holder can exchange it for an audited pro rata share of measured assets in kind, and from the end of November 2027 cannot. The round two rule then makes non-participation total rather than partial, because an address that skips round one forfeits its share of everything that arrives afterwards. Anyone holding through a wrapper has to unwind to BAL first or lose the claim, and the post says so for auraBAL and sdBAL by name. What the proposal also does is retire the protocol's own privileged roles: admin permissions, multisig roles and the Emergency subDAO are to be inventoried and then revoked or transferred, with 'The end state is a protocol that needs no one from Balancer'. Contracts that cannot be paused keep running with the fee set to zero. So the outcome contemplated is not a deletion but an unowned residue, still functional and earning nothing for anybody, which is a distinct thing from a company closing and worth naming as such.

What is not settled

The vote has not happened, and the post is explicit that a no vote leaves the BIP-918 mandate and budget, the BIP-919 buyback and the BIP-687 bounty in place, and that contributors who want to carry on under the existing mandate can, because it is funded. A 5m BAL quorum over four days is the immediate test, and the author says he will not front a continuation plan and that nothing in the proposal funds one. The treasury figure is a floor, not a number. 'At least $9M at current token prices' is the managed treasury kpk reports; the DAO holds further assets across other addresses, wallets and positions on several chains, those are being inventoried, and the base that governs the distribution is whatever is measured and audited at the opening snapshot in May 2027, net of costs until then, third-party funds and recoveries. A holder voting this month is voting on a rule rather than on an amount. Several pieces are promised rather than drafted. The claim contract, the snapshots and the distribution mechanism are to be published as an implementation specification by the end of February 2027, for comment and not for a vote, with the contract audited before round one. The post says anything that would change an allocation comes back to a vote, which leaves the line between implementing an allocation and changing one to be drawn later. The legal path is acknowledged and unresolved. The DAO vote governs the protocol and the treasury; the legal entities execute their own closures under the law that governs them, and the post says that where that law diverges from the plan, 'the law wins and the divergence is reported to holders'. What belongs to the DAO and what sits with the entities is to be established before any transfer vote. And funds recovered from the November 2025 exploit are excluded from this distribution and belong to the affected liquidity providers, with how the unclaimed remainder reaches them left to a separate proposal. The post gives no figure for that exploit; reports put it at $128m, and that figure is theirs.