Balancer's CEO Just Asked Token Holders to Vote the Protocol Dead. Its $9M Treasury Is Worth More Than Its Own Token.

Balancer’s treasury is worth at least $9 million. The entire BAL token, as of last week, was worth about $7.7 million.

So the guy running the protocol did the math and put it on the forum: shut it down, and give the treasury to the holders.

On Sunday night, Balancer Labs CEO Marcus Hardt posted a governance proposal titled “Orderly Winddown of Balancer and Distribution of the Treasury.” It does exactly what it says. Every pausable pool goes withdrawals-only on October 30. Contributors’ notice period, which quietly started on August 27, ends the next day. From November 1 a small hourly transition team keeps the lights on for withdrawals, and starting in May 2027 BAL holders can burn their tokens in exchange for a pro-rata slice of whatever is left.

The Snapshot vote runs September 25 to 29. Quorum is 5 million BAL.

How a 2020 blue chip ends up here

Balancer launched in March 2020. It was one of the first AMMs on Ethereum, it pioneered weighted multi-asset pools, and its June 2020 liquidity mining program helped kick off DeFi Summer. At its late-2021 peak the protocol held more than $3 billion in TVL, per DefiLlama. Today it holds roughly $59 million.

The thing that broke it was nine wei.

On November 3, 2025, an attacker exploited a rounding error in the upscale function of Balancer v2’s Composable Stable Pools. When token balances were pushed into the 8 to 9 wei range, Solidity’s integer division lost precision, and batched swaps compounded that tiny loss into full invariant manipulation. Around $128 million left the pools across Ethereum, Base, Arbitrum, Optimism, Gnosis, Polygon and other chains in under 30 minutes.

Revenue collapsed with it. DefiLlama data cited by Cointelegraph shows monthly protocol revenue went from $1.13 million in October 2025 to $371,000 in November. By August 2026 it was $56,781. Hardt’s own proposal uses an even lower all-in figure of about $30,000 for August.

In March, Balancer tried the restructuring playbook. Balancer Labs, the Estonian entity, was wound down and operations moved under a DAO-controlled OpCo. Headcount went from roughly 25 to 12.5 full-time equivalents. BAL emissions stopped. veBAL was sunset. The annual budget was cut by a third to $1.9 million. In April, BAL holders approved BIP-919, a buyback program. Hardt at the time said “the technology works… what stopped working was the economic model.”

Six months later, the new economic model didn’t work either. From the proposal:

“With a working product suite, a lean structure and full effort from everyone who stayed, the traction the plan required did not come. I do not see a funded path that changes this picture.”

And on why now rather than later:

“Continuing on the current path spends the treasury to arrive at the same place later. That treasury belongs to BAL holders. The question is whether what remains reaches holders while it is still substantial, or is spent first on a path that has already been tried.”

The core problem is that v3, the rebuilt architecture that was supposed to be the future, never replaced v2’s fee income. Hardt writes that most revenue still comes from v2, and “v3 revenue has not grown to replace it.” The November exploit hit legacy v2 pools, not v3, but as he puts it, the event “followed the name into every conversation since.”

The numbers that make the decision for you

Here is the monthly picture from the proposal:

Line item	Monthly
All-in burn	~$150,000
Protocol revenue (August)	~$30,000
Treasury yield (managed by kpk)	~$25,000
Net	roughly -$95,000

At that rate, $9 million is about eight years of runway. The proposal’s argument is that eight years of slow bleed on a plan that has already failed is a worse outcome for holders than an orderly liquidation now.

The wind-down itself is capped at $400,000: $150,000 for the transition team through May 2027, $30,000 for running the distribution, and a $220,000 reserve that only gets touched if needed. Anything unspent flows back into the pot for holders. September and October contributor pay, around $300,000, comes out of the existing budget.

How holders actually get paid

This is not a simple airdrop. It’s a three-round burn-and-redeem:

• Round one opens end of May 2027. The date is timed to the last veBAL lock expirations. Holders burn BAL and receive their pro-rata share of the treasury in-kind, meaning the actual tokens the treasury holds, not a USD equivalent. The claim window is six months.
• Round two, roughly January 2028. Unspent wind-down budget, post-pause protocol revenue, and any unclaimed round-one shares get airdropped to addresses that redeemed in round one, proportional to what they burned.
• Final sweep, roughly July 2028. Whatever trickles in afterward goes out to the same addresses. Then the treasury and distribution controls are retired.

Circulating supply for the denominator excludes treasury-held BAL and the Balancer Labs vesting and fundraise safes. Wrapped positions like auraBAL and sdBAL have to be unwound to raw BAL before round one closes or they don’t redeem. Funds recovered from the exploit are explicitly excluded from the distribution. Those belong to the affected LPs.

BIP-919’s buyback is cancelled. The ring-fenced $1 million bug bounty earmark gets released after coverage ends on October 30.

What doesn’t get decided by this vote

The contracts are non-custodial, and Hardt stresses that “withdrawing does not depend on Balancer or on anyone continuing to operate.” Pools that can’t be paused keep running with protocol fees set to zero.

Code, licenses and deployments are not being handed to anyone. Each transfer would need its own Snapshot vote. Hardt says he supports contributors who want to fork the infrastructure and keep it alive, but they have to bring their own proposal.

There’s already a counter-proposal on the forum from a user calling themselves Wise_Enthusiast, pitching a partnership with WiseSoft LLC to put the treasury’s stablecoins to work instead of shutting down. The author discloses they founded WiseSoft. Take that for what it’s worth.

Why This Matters for Crypto Jobs

The layoffs already happened, and nobody announced them. The proposal reveals that contributor notice periods started on August 27, more than two weeks before the public post. After March’s cuts, the team was about 12.5 full-time equivalents. By October 31 it’s an hourly transition crew. Hardt’s send-off line, “Everyone leaves with references and introductions from me, and public credit for what they built,” is the kind of thing you write when you know the people reading it are updating their résumés tonight.

A $128M exploit is now a documented 10-month kill chain. Hack in November, restructure in March, wind-down in September. That is the template every DeFi risk lead, protocol treasurer and DAO contributor should be studying. The exploit wasn’t in the new code. It didn’t matter. Reputational damage attached to the brand, LPs left, and revenue never came back. If you’re at a protocol that’s been hit this year, and there were plenty, the question isn’t “did we patch it,” it’s “did TVL and fees recover within two quarters.” If not, this is the path.

“DAO wind-down specialist” is a real job now. Balancer’s plan involves a Treasury Council, Foundation directors, a treasury manager (kpk), an OpCo, and an emergency subDAO whose permissions have to be inventoried and retired. Multi-chain asset inventories, audited snapshots, a distribution spec due February 2027, three redemption rounds through mid-2028. Someone has to run all of that. As more 2020-era protocols hit the same wall, the people who can execute an orderly shutdown without getting sued or drained will be in demand.

The v2-to-v3 migration story is a warning for every “rebuild” roadmap. Balancer shipped v3. It shipped new products. The tech worked. Users didn’t move. If your protocol’s hiring plan is built around a next-gen version replacing legacy revenue, you should be asking hard questions about what the migration incentives actually are. “Better architecture” alone didn’t do it here.

AMM talent is about to be on the market. The people who built weighted pools and boosted pools and the v3 vault architecture are some of the deepest AMM engineers in the space. Curve, Uniswap, Aerodrome, Fluid and every L2 trying to bootstrap native liquidity should be in their DMs already. If you’re one of those engineers, you have leverage. Use it.

---

Whether you’re leaving a protocol that’s winding down or hiring the people who are, Cryptogrind lists the DeFi, security, and protocol engineering roles that are actually open right now. The talent from a 2020 blue chip is about to hit the market. Be the team that catches it.