Why Bundler Extractable Value is Inherently Toxic

Bundlers are extractors by design. The ERC-4337 standard grants bundlers the power to order transactions within a UserOperation, creating a direct analog to block builders in MEV supply chains like Flashbots. This ordering power is a financial asset, not a neutral utility.

User intents are monetizable liabilities. When a user submits a gasless intent via a Paymaster, the bundler sees a raw profit opportunity. This is identical to the intent-based MEV extracted by solvers on CowSwap or UniswapX, but now embedded in core wallet infrastructure.

The 'abstraction' hides the cost. Users perceive seamless UX, but the economic reality is a forced arbitrage. A bundler will always reorder and insert its own profitable transactions, capturing value that would otherwise go to the user or the destination DApp.

Evidence: The rapid emergence of bundler-as-a-service providers like Stackup and Alchemy, which operate private mempools, proves the value is already being captured. This is pre-MEV, not post-MEV.

BEV inverts the user-validator relationship. In traditional MEV, searchers compete to serve users for a fee. BEV allows the bundler (validator) to become the user, creating transactions solely to extract value from the system they are supposed to secure.

This creates a tax on honest users. Bundlers like EigenLayer operators or AltLayer sequencers must be compensated for forgoing BEV opportunities. This cost is passed to users as higher fees, subsidizing the very activity that degrades the network.

It directly attacks consensus security. The proposer-builder separation (PBS) model in Ethereum L1 exists to isolate consensus from profit motives. BEV re-entangles them at the L2/rollup layer, creating a centralizing force where the most extractive validator wins.

Evidence: The existence of MEV-Boost on Ethereum proves the danger of unmanaged extractive value. Rollups without PBS, like many OP Stack chains, bake this risk directly into their consensus mechanism from day one.

Bundler Extractable Value (BEV) is the profit a bundler earns by reordering, inserting, or censoring user operations before submitting them to an EntryPoint. This is a direct analog to Miner/Maximal Extractable Value (MEV) in block production.

BEV is structurally toxic because it incentivizes bundlers to act against user interests. A bundler will always prioritize a backroom deal with a DEX aggregator like 1inch over a user's original transaction intent, extracting value that belongs to the user.

This leads to centralization pressure. Just as MEV led to the rise of specialized builders like Flashbots and block builders, BEV will create a market for proposer-builder separation (PBS) for bundlers. Only large, sophisticated bundlers with private orderflow deals will survive.

Evidence: The existing MEV supply chain proves this outcome. On Ethereum, over 90% of blocks are built by a handful of entities. The bundler layer, without PBS, will replicate this centralization, making the user-facing promise of permissionless participation a fiction.

BEV is a negative-sum game. The value extracted by a bundler is directly siphoned from user transactions, creating a tax on the ecosystem. Unlike MEV, which can be neutral or positive-sum via arbitrage, BEV's extraction is pure rent-seeking.

Bundlers control transaction ordering. This power, combined with paymasters and account abstraction, lets them front-run, censor, or delay user ops for profit. The user's intent, as defined in ERC-4337, is subverted by the entity meant to serve it.

The system's design guarantees conflict. The paymaster subsidy model creates a direct incentive for bundlers to exploit the very users they are paid to help. This is a fundamental flaw, not an edge case.

Evidence: In early ERC-4337 implementations, bundlers like Pimlico and Stackup must run complex, proprietary algorithms to capture BEV just to remain economically viable, proving the incentive is inescapable.

Reputation is not a deterrent for extractive behavior. The economic upside from a single successful MEV capture dwarfs any long-term reputational penalty. This is a prisoner's dilemma where the dominant strategy is to extract.

Competition does not lower fees. It merely shifts value from users to a sealed-bid auction between bundlers. This is identical to the miner extractable value (MEV) problem on L1, where more miners increased competition but not user welfare.

The system is structurally opaque. Unlike transparent mempools, the bundler's private transaction pool and order flow agreements with entities like Flashbots Protect or BloxRoute create information asymmetry users cannot audit.

Evidence: The existence of PBS (Proposer-Builder Separation) on Ethereum is a direct admission that reputation and competition failed. The protocol had to be redesigned to mitigate extractive behavior at the builder level.

BEV is toxic MEV: Bundler Extractable Value is a direct analog to Miner Extractable Value, but with higher centralization risk. A bundler's role in ordering user operations creates an identical profit motive to reorder, censor, or front-run transactions.

Protocols will centralize: The economic incentive to capture BEV will drive bundler infrastructure towards a few dominant players like Ethereum's PBS builders or specialized services from EigenLayer AVSs. This recreates the validator centralization problem Account Abstraction aims to solve.

In-protocol PBS is necessary: Sustainable economics require a Proposer-Builder Separation (PBS) model enforced at the protocol level, similar to Ethereum's roadmap. Without it, user experience and security are auctioned to the highest bidder.

Evidence: Flashbots' dominance in MEV capture demonstrates how economic incentives inevitably consolidate infrastructure. A bundler market without PBS will follow the same path, making decentralized bundlers a marketing term.