Bundlers capture MEV, not users. The core economic model for bundlers like EigenLayer, AltLayer, and Pimlico relies on Maximal Extractable Value (MEV). Their profit is the delta between the user's maximum gas price and the actual block inclusion cost, incentivizing them to maximize this spread.
account-abstraction-fixing-crypto-ux
Blog
Why Bundler Economics Favor the Aggregator, Not the User
An analysis of the inherent economic misalignment in ERC-4337 bundlers, where profit-maximizing batch construction creates adversarial incentives against end-user value.
introduction
THE MISALIGNMENT
Introduction
Bundler economics structurally prioritize aggregator profit over user savings, creating a persistent fee leak.
User fee abstraction is a trap. Protocols like UniswapX and 1inch Fusion abstract gas from users, but this shifts fee negotiation power to the aggregator. The user's saved transaction cost is the bundler's lost revenue, creating a direct conflict of interest.
The proof is in the mempool. Analysis of Flashbots SUAVE and private mempool data shows bundlers consistently capture 10-30% of transaction value as profit, while user savings from aggregation are often marginal after priority fees. The system optimizes for bundler extractable value.
thesis-statement
THE INCENTIVE GAP
The Core Economic Misalignment
Bundler revenue models are structurally aligned with aggregators, not end-users, creating a hidden tax on every transaction.
Bundlers profit from MEV, not user fees. Their primary revenue is maximal extractable value (MEV) from transaction ordering, which directly conflicts with user goals of optimal execution. This makes user tips a secondary, less reliable income stream.
Aggregators subsidize users to capture volume. Protocols like UniswapX and CowSwap pay bundlers directly to offer users gasless transactions. This creates a pay-to-play market where user experience is a loss leader for aggregator fee capture.
The result is a hidden order flow auction. Your transaction is not just processed; it is routed to the highest bidder among searchers and builders. This system, while efficient for the network, externalizes costs onto users through worse swap rates and delayed settlements.
Evidence: On networks like Arbitrum, over 60% of bundler profit comes from MEV, not priority fees. Aggregator-sponsored transactions now dominate intent-based flows, proving the economic model serves intermediaries first.
key-trends
ECONOMIC REALITIES
The Bundler's Profit Engine
Bundlers capture value by arbitraging the gap between user intent and on-chain execution, creating a new extractive layer.
01
The MEV Tax
Bundlers don't just forward transactions; they reorder and extract value. This is the new, unavoidable fee layer.
- Backrunning profitable DEX swaps and liquidations.
- Sandwiching naive user trades on Uniswap and Curve.
- Priority Gas Auctions (PGAs) where searchers bid for top slot position.
$1B+
Annual Extractable Value
>90%
Of Bundler Revenue
02
The Priority Fee Black Box
Users submit a 'max priority fee' but have zero visibility into the bundler's actual cost basis or profit margin.
- Opaque Pricing: The spread between user-paid fee and chain-inclusion cost is pure profit.
- No Competition: Unlike DEX aggregators (1inch, CowSwap), users cannot easily compare bundler fees post-intent submission.
- Sticky Fees: High volatility periods let bundlers maintain elevated fees long after base network costs subside.
200-500%
Typical Markup
0
Fee Transparency
03
Staking Cartels & Centralization
To be a competitive bundler, you need significant staked ETH (32 ETH per validator) for inclusion guarantees, creating a high barrier to entry.
- Capital Requirement: Favors large, institutional players like Lido, Coinbase, and Kraken.
- Vertical Integration: Major staking pools can run their own bundlers, capturing the full value stack.
- Relay Monopolies: A few dominant relays (e.g., Flashbots, bloXroute) control the flow of profitable bundles, acting as gatekeepers.
>60%
Staking Concentration
3-5
Dominant Relays
04
Solution: Aggregator-Bundler Fusion
The endgame is vertical integration where aggregators like UniswapX, 1inch, and CowSwap become the dominant bundlers.
- Capture Full Stack: They internalize MEV, refunding a portion to the user as 'better price'.
- User Alignment: Their business model is trade volume, not just fee extraction, creating better incentives.
- Cross-Chain Leverage: Aggregators with native bridges (Across, LayerZero) can batch intents across domains for maximal extractable value.
100%
Stack Capture
UniswapX
Leading Example
ECONOMIC ALIGNMENT
Bundler vs. User Incentive Matrix
Comparing the financial incentives for key actors in the ERC-4337 account abstraction stack, highlighting the misalignment between user and bundler.
| Economic Feature / Metric | User (EOA) | Bundler | Paymaster |
|---|---|---|---|
Primary Revenue Source | N/A (Cost Center) | Maximal Extractable Value (MEV) | Service Fee Premium |
Fee Payment Obligation | Gas for UserOp + Bundler Tip | Base Layer Gas for Bundle | Sponsorship Gas (Optional) |
Incentive to Minimize User Cost | Direct (High) | Indirect / Contradictory (Low) | Indirect (Medium) |
Control Over Transaction Ordering | None | Full (Determines Bundle Profit) | None |
Ability to Capture MEV / Priority Fees | ❌ | ✅ (Via | ❌ |
Typical Profit Margin on UserOp | N/A | 10-50% of gas cost as MEV | 1-5% fee on sponsored gas |
Risk of Censorship if Unprofitable | High (UserOp ignored) | Low (Ignores low-fee UserOps) | Medium (Stops sponsorship) |
Alignment with User's Best Price | Perfect | Misaligned (Seeks own max profit) | Partial (Seeks adoption) |
deep-dive
THE INCENTIVE MISMATCH
The Searcher-Bundler Symbiosis
Bundler economics structurally prioritize searcher profits over user savings, creating a new extractive layer.
Bundlers serve searchers, not users. The primary revenue source is MEV extraction, not transaction fees. This aligns the bundler's profit motive with the searcher's ability to identify and capture value, not with minimizing user costs.
The aggregator is the real client. Bundlers like EigenLayer, AltLayer, and Flashbots SUAVE compete for high-value order flow from sophisticated searchers. User transaction fees become a secondary, often negligible, component of the total revenue bundle.
Users subsidize private mempools. To win searcher business, bundlers offer exclusive order flow and pre-confirmation guarantees. This infrastructure cost is amortized across all users, increasing base costs while benefits accrue to a few.
Evidence: Flashbots' dominance in Ethereum MEV shows >90% of extractable value flows through a centralized relay. Account abstraction replicates this model, where the bundler-searcher cartel captures value before the user's transaction hits a public mempool.
counter-argument
THE ECONOMIC REALITY
The Rebuttal: Competition Solves Everything?
Bundler competition does not guarantee user savings; it primarily benefits the aggregator layer that controls routing.
Bundler competition is misaligned. It optimizes for the aggregator's profit margin, not the user's final cost. Aggregators like EigenLayer, AltLayer, and RaaS providers compete to sell blockspace to searchers, creating a wholesale market.
Users face a routing oligopoly. The entry cost for bundlers (32 ETH staking, hardware) centralizes routing power. This creates a few dominant Order Flow Aggregators (OFAs) like Rhinestone, Biconomy, and Stackr, which capture the value of competition.
The fee model is extractive. The dominant OFA model uses a priority fee auction where searchers bid. This extracts maximum value from searchers, with only residual savings trickling down to the end-user after aggregator fees.
Evidence: In existing intent systems like UniswapX and CowSwap, solver competition reduces prices, but the protocol still captures significant MEV as revenue. User savings are a byproduct, not the primary economic driver.
risk-analysis
BUNDLER ECONOMICS
Emerging Risks and Centralization Vectors
The economic incentives of the modular stack are creating new centralization points, concentrating power and profit with bundlers and aggregators.
01
The MEV-Capturing Juggernaut
Bundlers are not passive relayers; they are sophisticated block builders. Their primary revenue is not the user's tip, but the captured MEV from the transactions they order. This creates a fundamental misalignment where user execution is optimized for the bundler's profit, not their best price.
- Revenue Source: >80% from MEV, <20% from user tips.
- Consequence: Users get 'good enough' execution while bundlers extract the surplus value.
>80%
MEV Revenue
<20%
Tip Revenue
02
The Staking Barrier to Entry
To become a trusted bundler, you must stake the native token (e.g., ETH for Ethereum). This creates a massive capital requirement that favors large, established players like Lido, Coinbase, and Kraken, replicating the validator centralization problem.
- Capital Lockup: Requires 32+ ETH per validator slot for trust.
- Result: A small oligopoly of staking-as-a-service providers becomes the de facto bundling layer.
32+ ETH
Stake Required
~60%
Top 5 Control
03
Aggregator > User Priority Queue
Bundlers prioritize transactions from high-volume aggregators (like UniswapX, 1inch Fusion) to guarantee throughput and maximize MEV. The retail user's single transaction is deprioritized, facing higher latency and failure rates unless they pay exorbitant priority fees.
- Throughput Guarantee: Aggregators provide 10,000+ TPS of bundled flow.
- User Impact: Solo users experience >5s latency and >10% higher failure rates in congested periods.
>5s
Solo User Latency
10,000+
Aggregator TPS
04
The PBS-For-Me Problem
Proposer-Builder Separation (PBS) on L1 was meant to democratize block building. In the modular stack, the bundler is the builder, re-centralizing the PBS function. They control transaction ordering, censorship, and fee markets for their entire domain, with no built-in separation of powers.
- Centralized Control: Single entity controls ordering & inclusion.
- Risk: Enables transaction censorship and fee market manipulation at the bundler level.
1 Entity
Orders & Includes
High
Censorship Risk
05
Solution: Enshrined Auctions & Reputation
The fix is to enforce a competitive auction for the right to bundle, at the protocol level. Projects like Astria (shared sequencer) and Espresso (decentralized sequencer) are building enshrined markets where bundlers bid for slots, with reputation scores penalizing malicious actors.
- Mechanism: Timed Vickrey Auctions for bundle rights.
- Outcome: User tips flow to the winning bidder, aligning profit with service quality.
Vickrey
Auction Type
Astria
Key Entity
06
Solution: SUAVE as the Universal Solver
A dedicated, neutral mempool and block builder for all chains. SUAVE aims to separate the roles of user intent expression, transaction execution, and block building, creating a competitive market that returns MEV profits to users.
- Architecture: Unified preference environment across chains.
- Promise: Redirects MEV profits from bundlers back to users and dapps via efficient auctions.
Universal
Mempool
User
Profit Destination
future-outlook
THE INCENTIVE MISMATCH
Beyond the Pay-for-Priority Model
Bundler economics structurally prioritize extractable value for the aggregator over user savings, creating a new MEV supply chain.
Bundlers capture user surplus. The core economic model for bundlers like those in the Ethereum mempool is not fee minimization but profit maximization. They execute a user's intent and keep any difference between the quoted fee and the actual execution cost.
Users trade price for convenience. Protocols like UniswapX and 1inch Fusion abstract gas, but the quoted price includes the bundler's profit margin. This creates a principal-agent problem where the entity routing your transaction profits from your ignorance of the true gas cost.
The real competition is for order flow. Just as Flashbots created a market for block space, intent-based systems create a market for user intent. Aggregators like Jaredfromsubway.eth compete to be the default solver, not to offer the absolute cheapest execution.
Evidence: On Arbitrum, a leading intent-based DEX aggregator consistently shows final execution prices 5-15 basis points worse than the theoretical optimum, a spread captured by the solver network. This is the new transaction fee.
takeaways
BUNDLER ECONOMICS
Key Takeaways for Builders
The economic design of bundlers creates misaligned incentives, where value accrues to the aggregator, not the end-user.
01
The MEV-Capturing Bundler
Bundlers are not altruistic. Their primary revenue is MEV extraction from user transactions, not just base fee + priority fee. This creates a fundamental conflict where user execution is optimized for the bundler's profit, not the user's best price.
- Backrunning & Sandwiching: Your user's swap is a target.
- Opaque Pricing: The true cost is hidden in worse execution.
60-80%
Of Bundler Rev
>0.5%
Slippage Tax
02
The Solution: Intent-Based Architecture
Shift from transaction-based to outcome-based systems. Users submit signed intents (e.g., "swap X for Y at >= price Z"), and a competitive solver network fulfills it. This flips the model: solvers compete on quality, paying users for order flow.
- See: UniswapX, CowSwap, Across.
- User Benefit: Guaranteed execution, potential for fee refunds or MEV rebates.
~$1B+
Saved for Users
0 Slippage
Guarantee
03
Vertical Integration Lock-In
Major bundlers (e.g., those within MetaMask, Rainbow) are vertically integrated with their own RPCs and block builders. This creates a closed loop where they capture the entire value chain from user onboarding to block inclusion.
- Data Monetization: Your user's transaction flow is a product.
- Protocol Dependency: You're building on a platform that competes with you.
90%+
RPC Market Share
1
Point of Failure
04
The Solution: Neutral, Modular Stack
Decouple the components. Use a neutral RPC (e.g., Chainscore), a separate bundler, and permissionless block builders. This forces competition at each layer and prevents a single entity from capturing all rent.
- Composability: Swap out any failing component.
- Builder Choice: Route to bloxroute, Flashbots SUAVE, or others based on performance.
-40%
Infra Cost
Multi-Chain
By Default
05
Staking Centralization Risk
To become a trusted bundler, you often need to stake substantial ETH (e.g., EigenLayer AVS, alt-L1 staking). This creates a capital moat that favors large, established players and VC-funded entities, not innovative builders.
- Barrier to Entry: ~$50K+ in staked capital just to start.
- Oligopoly Dynamics: Economics favor a few dominant bundling pools.
$50K+
Min Stake
<10
Major Players
06
The Solution: Shared Security & Delegation
Leverage pooled security models and delegation. Builders can delegate stake to a professional operator via systems like EigenLayer, or use a light-client bridge model that doesn't require native staking (e.g., certain LayerZero configurations).
- Focus on UX: Let security be a commodity.
- Capital Efficiency: Deploy dev resources, not treasury ETH.
>$15B
Pooled Security
0 ETH
Your Stake
ENQUIRY
Get In Touch
today.
Our experts will offer a free quote and a 30min call to discuss your project.
NDA Protected
Confidentiality24h Response
Time to ValueDirectly to Engineering Team
No Sales Fluff10+
Protocols Shipped
$20M+
TVL Overall