On 15 September 2026, Balancer said a proposal to wind down the protocol and distribute the treasury to BAL holders is live on the forum. Authored by former Balancer Labs CEO Marcus Hardt, the proposal is currently open for discussion and a Snapshot vote is expected from 25 to 29 September.
The X post states that nothing changes today: pools and withdrawals work as they do now, and any wind-down action waits for the vote.
The full text is titled [BIP-XXX] Orderly Winddown of Balancer and Distribution of the Treasury. What follows stays with that proposal, Balancer’s official account, and earlier Balancer forum posts the new vote would change.
What holders are being asked to decide
The proposal is a direction vote, not an instant close. If it passes, Balancer would stop new business development, run a phased sunset with a defined exit window, and close the DAO to the extent legally and practically possible.
Passing would also set the timeline, the treatment of DAO assets beyond the managed treasury, the allocation rules for BAL holders, and a wind-down budget. It would cancel the BIP-919 buyback and supersede BIP-687. Spending through 31 October 2026 would still follow BIP-918. That remainder would not roll forward. After 1 November it would be spent against the wind-down budget, with unused funds returning to holders.
Marcus discloses that he sits on the Treasury Council and on the multisigs that execute what governance approves. He was added to the Council under BIP-918, is paid through the notice period on the same terms as other contributors, and would redeem on the same terms as other holders. No new vesting is proposed.
If the vote fails, the current framework remains. The protocol would not move to the withdrawals-only date described below. The buyback approved in BIP-919 would not be cancelled by this text. Requests to carry a deployment or product forward, the proposal says, would need their own Snapshot vote.
The implementation specification for the claim contracts is not part of this vote. It would be published later for comment. Transfers of DAO assets beyond the treasury would also need separate votes. Changes to the allocation rules would return to a vote.
From the November 2025 exploit to the wind-down vote
On 3 November 2025, Balancer was hit by a major exploit and attackers drained roughly $128 million in funds from legacy Balancer v2 pools across multiple chains. The event, later cited in Balancer’s own Labs and governance posts, sat in the $110–128 million range and hit the v2 architecture that still produced most protocol revenue. v3 was a different design, but the exploit followed the name.
Recovery talks, bounty caps, and chain-level fund freezes played out in public while LPs on affected pools waited on separate recovery allocations. Those recovered funds were later ring-fenced from any DAO treasury split.
After the hack, Balancer tried a lean reset rather than an immediate close. In March 2026 Fernando Martinelli said Balancer Labs was shutting down, pointing to legal exposure and a shift to the DAO and OpCo model.
The same week, BIP-918 cut the team and budget and BIP-919 halted BAL emissions, ended veBAL economic rights, sent all protocol fees to the treasury, and offered a NAV buyback. Revenue did not recover: official later framing is that v3 never replaced v2 income and key contributors stepped back.
On 14 September 2026 Marcus Hardt posted the orderly wind-down proposal, with a Snapshot vote set for 25–29 September and no pool changes until that vote.
Why the author says the review is coming now
In March and April 2026, holders approved a lean reset. BIP-918 consolidated operations under Balancer OpCo Limited after Balancer Labs ceased operations, right-sized the team to 12.5 FTE, and set a $1.9 million annual budget, a cut from about $2.87 million under BIP-873. Combined with fee routing in the companion tokenomics vote, that post projected a smaller deficit and a longer runway in a neutral case.
BIP-919, posted 23 March 2026, halted BAL emissions, discontinued veBAL economic rights, routed 100% of protocol fees to the DAO treasury, and authorized a buyback-and-burn capped at 35% of treasury holdings at snapshot, then about $3.6 million, at NAV. It also included a $500,000 stablecoin campaign for veBAL holders.
On the same day, Fernando Martinelli wrote that Balancer Labs was shutting down, citing legal exposure after the 3 November 2025 v2 exploit and the shift to a DAO, Foundation, and service-provider structure.
The new proposal says holders backed that plan and the team executed it. v3 is live. Boosted Pools remain a category Balancer defined. AutoRange Pools shipped. Integrations and partnerships were pursued. Some drew interest. None, the author writes, converted into sustained revenue growth. Key people left or stepped back. Most revenue still comes from v2. v3 revenue has not grown to replace it.
BIP-918 set a review for this case. This text is that review, brought early. Waiting, the author writes, would change the numbers, not the conclusion, and every month of waiting is spent from the treasury.
The November 2025 exploit on legacy v2 pools is described as part of why traction was harder after April, and not as the basis of the proposal. “I do not see a funded path that changes this picture,” the text says. Continuing on the current path “spends the treasury to arrive at the same place later.”
The treasury, it says, belongs to BAL holders. The question posed is whether what remains reaches them while it is still substantial, or is spent first on a path that has already been tried.
How the treasury would be split, and who must unwrap first
The proposal puts the managed treasury, as reported by kpk on Octav, at least at $9 million at current token prices. Other DAO addresses and positions are to be inventoried and published before round one. The figure that counts is the amount measured and audited at the block round one opens.
Positions would be moved into yield-bearing forms where possible until then. Funds in operational safes and fee wallets would be consolidated. Open receivables in the DAO’s name would be recovered. The base is everything the DAO holds and is owed, net of what belongs to others.
BAL, and anything that resolves into BAL, is excluded from that pot, with one stated exception for tetuBAL. The BIP-919 buyback would be cancelled, releasing earmarked funds into the distribution.
Nothing is paid before round one. After the wind-down budget is set aside, holders would burn BAL and receive an in-kind, pro-rata share of what remains. Round one would open at the end of May 2027 and run six months, to the end of November 2027. The claim contract would record each redeeming address and amount. That record is the basis for later rounds.
Round two would be an airdrop, within two months of that close, to addresses that redeemed in round one, in proportion to what they burned. It would include unspent budget, value that arrived after the first snapshot, and the share not redeemed in round one. An address that missed round one would have no round-two share. A final sweep, six months later, would send later arrivals to the same addresses in the same proportion. The process would end there.
veBAL unlocks into the 80/20 BAL/WETH BPT. That pool stays exitable. Holders would exit to BAL and redeem. Locks that exist when the proposal was posted are described as expired by round one. A lock extended after the post redeems when it unlocks, inside the window or not at all.
tetuBAL is treated as an immutable permalock. The holder set and amount would be fixed at the block of the forum post. When round one opens, tetuBAL holders would receive BAL equal to half of that measured amount from the treasury and could redeem it like other circulating BAL. The proposal calls this the one case where treasury BAL is distributed rather than excluded.
auraBAL, sdBAL, and other wrappers would have to unwind to BAL on those protocols’ own calendars before the close of round one. A position not unwound by then would not redeem.
Funds recovered from attacks on the protocol are stated to belong to LPs of the affected pools. They sit outside the treasury and outside this distribution. This text does not change prior allocations for those recoveries.
The wind-down budget in the vote is $150,000 from 1 November 2026 to May 2027, $30,000 from then to the final sweep, and a $220,000 reserve drawn only if needed.
What would change for pools, partners, and the calendar
Balancer says pools and withdrawals work as they do now until a vote passes. If it does, the contributor notice, already running from 27 August 2026 in the timeline, ends on 31 October 2026.
On 30 October 2026, pausable pools would be paused and moved to withdrawals only. Recovery mode would be enabled where contracts require it. Protocol fees would be set to zero on the rest where allowed. Bug-bounty coverage would end that day.
LP funds stay withdrawable. Withdrawing does not depend on Balancer operating. Withdrawal paths are to be documented before 30 October. Partners would be told directly and asked to migrate liquidity in that window. Obligations would be honored or wound down explicitly.
From 1 November 2026 a small transition team would run a minimal stack through the distributions. An implementation specification would be published at the end of February 2027 for comment. Round one would open at the end of May 2027. Round two would follow at the end of January 2028. The final sweep would be at the end of July 2028. Entity closures would come after that sweep.
Governance docs still describe the Q2 2026 stack: BIP-918, BIP-919, the veBAL compensation airdrop, and later vote-weight changes. This new proposal is the review those posts left room for. It is not in force. Holders decide on Snapshot from 25 to 29 September.
Also read: CoinEx Announces Shutdown After 9 Years, Withdrawals Open Until December 2026
