Balancer Labs, the Estonian corporate entity that built and maintained the Balancer decentralized exchange, announced on March 24 that it will cease operations. The decision follows a $128 million exploit on November 3, 2025, that drained assets across six blockchain networks in under 30 minutes....
"Balancer Labs, as a corporate entity, has become a liability rather than an asset to the protocol's future." — Fernando Martinelli, Co-founder, Balancer Labs
Balancer Labs, the Estonian corporate entity that built and maintained the Balancer decentralized exchange, announced on March 24 that it will cease operations. The decision follows a $128 million exploit on November 3, 2025, that drained assets across six blockchain networks in under 30 minutes. Co-founder Fernando Martinelli disclosed the wind-down in a governance forum post, citing mounting legal exposure and an unsustainable revenue model.
The protocol itself will not shut down. A new entity, Balancer OpCo Limited, incorporated in the British Virgin Islands, will assume operations with a skeleton crew of 12.5 full-time equivalents — roughly half the prior headcount — on an annual budget of $1.9 million. Two linked governance proposals would eliminate all BAL token emissions, redirect 100% of protocol fees to the DAO treasury, and commit $3.6 million to a BAL buyback-and-burn program. The Balancer case is now the largest DeFi corporate dissolution triggered by a single exploit, and it establishes a template that other protocols facing similar legal exposure may follow.
On November 3, 2025, an attacker exploited a rounding-direction mismatch in the scaling math of Balancer V2 Composable Stable Pools. According to a post-mortem by security firm Halborn, the _upscaleArray function used mulDown for upscaling while downscaling used divUp/divDown, creating a precision asymmetry that could be weaponized through carefully crafted batchSwap operations.
The exploit mechanics were precise. The attacker deployed a smart contract whose constructor executed more than 65 micro-swaps in sequence, compounding precision loss at each step. When pool balances were driven to the 8–9 wei range, Solidity's integer division caused rounding discrepancies reaching approximately 10% per operation, according to Check Point Research's analysis. The attacker artificially deflated pool invariants and extracted value through discounted Balancer Pool Token (BPT) redemptions.
Total losses reached $128.64 million across six chains:
| Chain | Loss | |-------|------| | Ethereum | $99M | | Arbitrum | Partial | | Polygon | Partial | | Base | Partial | | Optimism | Partial | | Sonic (forks) | Partial |
Ethereum bore the majority of the damage, with $99 million drained from pools holding osETH, WETH, and wstETH. The entire attack completed in under 30 minutes. No V3 pools were affected; security firm Certora confirmed the rounding logic was correct in the V3 codebase.
The exploit accelerated a decline already underway. Balancer's total value locked fell from $775 million immediately before the hack to $422 million — a 46% drop. As of March 23, 2026, TVL stands at $157 million, according to DefiLlama. For context, Balancer's TVL peaked at $2.96 billion in October 2021. The current figure represents a 95% contraction from that high.
The BAL token reflects the deterioration. It trades at approximately $0.16, down from an all-time high of $74.45. Market capitalization sits at roughly $10 million with a fully diluted valuation of $11 million. Circulating supply is approximately 65 million tokens.
Revenue generation, while still functional, is marginal. Martinelli stated in his forum post that the protocol generated "over $1 million in fees annualized" in the three months preceding the announcement. Under the current fee structure, where the DAO captures 17.5% of protocol revenue, this translates to approximately $175,000 annually flowing to the treasury — insufficient to sustain a corporate operation.
The restructuring involves dissolving Balancer Labs OÜ, the Estonian entity, and transferring operational control to Balancer OpCo Limited, a BVI-incorporated entity that operates as a direct agent of the DAO. The governance proposal was posted March 23; a snapshot vote is pending.
Key structural changes:
Martinelli's forum post framed the decision starkly: "Maintaining a corporate entity that carries the liability of past security incidents, while the protocol itself needs to move forward unburdened, is not responsible stewardship." He disclosed that he "seriously considered" shutting down the entire protocol before opting for restructuring.
The proposed changes to BAL tokenomics are substantial and, if approved, will fundamentally alter the protocol's incentive structure.
Emissions elimination. Current annual BAL emissions run at 3.78 million tokens per year. The proposal eliminates them entirely. Martinelli characterized the existing system as a "circular bribe economy" where emissions incentivized liquidity that generated fees that funded more emissions — without producing durable value for token holders.
Fee redistribution. The current model allocates 50% of swap fees to liquidity providers, with the DAO capturing 17.5%. The proposed structure increases the LP share to 75% and routes all remaining fees (25%) to the DAO Treasury. This inverts the prior dynamic: LPs receive a larger share, and the DAO captures all non-LP revenue rather than splitting it.
veBAL wind-down. The veBAL governance and yield-bearing token system will be sunset. A $500,000 compensation campaign, paid in stablecoins over six months, is earmarked for locked veBAL holders who will lose economic rights under the new model.
Buyback and burn. The DAO will commit $3.6 million — approximately 35% of the treasury — to a BAL buyback at net asset value, currently about $0.16 per token. At that price, this would retire approximately 22.7 million BAL tokens, or roughly 35% of circulating supply. The buyback window opens approximately 12 months after the governance snapshot passes.
Martinelli positioned the buyback as an exit mechanism: "If you believe in the restructured Balancer, you stay. If you don't, you get a fair exit. That's honest dealing, and it clears the overhang."
Recovery from the November exploit has been partial. According to The Block, Balancer proposed distributing approximately $8 million in rescued assets. Six white hat actors recovered roughly $3.86 million during the attack, with the largest single recovery — $2.68 million on Polygon — attributed to an anonymous actor labeled "Anon #1."
Separately, liquid staking protocol StakeWise recovered 5,041 osETH tokens worth approximately $19.3 million via a contract call, according to on-chain analyst EmberCN. The Berachain Foundation recovered $12.8 million after the exploit impacted its native DEX (Bex), which ran on Balancer V2 infrastructure. Berachain executed an emergency hard fork to freeze attacker funds on its chain.
Total confirmed recoveries stand at approximately $26.4 million. The claiming process for affected users remains active. That leaves an estimated $102 million in unrecovered funds.
Balancer is not an isolated case. Step Finance, a Solana-based DeFi portfolio aggregator, shut down in early 2025 after a $27 million exploit drained 261,854 SOL. Tally, a DAO governance platform, also ceased operations during the same period. The pattern is consistent: exploit, legal exposure, revenue collapse, corporate dissolution.
The Balancer restructuring tests a specific hypothesis — that dissolving the corporate entity while maintaining the protocol through DAO governance can insulate a project from the legal fallout of a security breach. Whether this structure holds under the SEC-CFTC joint interpretation issued March 17, 2026, which established crypto asset classification guidelines, remains untested. The EU Product Liability Directive (2024/2853), which explicitly includes software in its definition of "product" and imposes strict liability for defects, adds further uncertainty for any entity — corporate or DAO — operating DeFi infrastructure in European jurisdictions.
The economic question is equally pressing. Balancer's annualized revenue of approximately $1 million places it well below the operational sustainability threshold for a protocol that once managed nearly $3 billion in TVL. The proposed $1.9 million annual budget for OpCo exceeds current annualized fee revenue, implying the protocol must either grow fees or draw down treasury reserves. At current burn rates, the treasury (estimated at approximately $10 million) provides roughly five years of runway — assuming no further security incidents.
The Q1 2026 DeFi exploit total reached $138 million, according to a separate webthreepedia analysis. The Balancer episode accounts for the majority of 2025's single-incident losses and contributed to what security researchers describe as a structural shift in DeFi risk modeling, where rounding errors and arithmetic precision attacks now rank alongside flash loan exploits and oracle manipulation in the threat taxonomy.
Balancer's corporate dissolution is the clearest case yet of a DeFi protocol severing its corporate shell to escape the legal gravity of a major exploit. The strategy is rational: the corporate entity generated no independent revenue, carried substantial legal liability, and consumed resources that could otherwise flow to protocol development. By shifting operations to a DAO-controlled OpCo, Balancer attempts to preserve the functional protocol while discarding the legal entity that built it.
The next 12 months, as Martinelli noted, will determine whether this works. The protocol must demonstrate that a team half its former size, operating on a budget that exceeds its current revenue, can ship product improvements sufficient to reverse a 95% TVL decline. The $3.6 million buyback provides token holders with a floor but also drains more than a third of the treasury. If TVL and fee revenue do not recover, the math constrains OpCo's runway to approximately five years before the treasury is depleted.
For the broader DeFi sector, Balancer's restructuring raises a question that has no settled answer: does dissolving a corporate entity actually insulate a protocol from liability, or does it simply redistribute that liability to DAO participants? The legal frameworks — both the U.S. SEC-CFTC joint interpretation and the EU Product Liability Directive — remain untested on this specific structure. Until a court rules, the Balancer model is a bet, not a precedent.