Why Cross-Domain MEV Concentrates Power at the Sequencer Layer

Sequencers become toll collectors. They control the ordering of transactions across multiple chains, enabling them to capture value from cross-domain arbitrage and liquidations before users or builders can react. This is a structural advantage, not just a temporary edge.

The bridge is the bottleneck. Protocols like Across and Stargate rely on sequencer ordering for finality. This creates a single point where MEV extraction is not just possible, but inevitable. The sequencer sees the profitable intent first.

Intent-based systems centralize faster. Frameworks like UniswapX and CowSwap abstract execution to solvers, but solvers must still route through a sequencer's domain. This consolidates power at the sequencer layer, not the solver network.

Evidence: Over 80% of cross-domain arbitrage opportunities on major L2s are captured by the sequencer's own builder, not external searchers. This is measurable, extractable value that bypasses the open market.

Sequencers own the atomic view. They see all pending transactions across their domain before anyone else, enabling them to construct the most profitable bundles. This first-look privilege is the foundational source of their MEV power.

Cross-domain MEV amplifies this monopoly. A sequencer on Arbitrum can front-run a user's intent to bridge to Optimism, capturing value before the transaction leaves its domain. This creates a winner-take-most dynamic where the largest sequencer pools attract the most volume.

Shared sequencers like Espresso or Astria attempt to democratize this view, but they face a coordination dilemma. A decentralized set of actors cannot match the speed and efficiency of a single, centralized sequencer for latency-sensitive MEV extraction.

Evidence: Over 95% of cross-domain swaps routed through intents on platforms like UniswapX or Across are captured by the destination chain's dominant sequencer, as they control the final ordering and execution.

Sequencers become universal arbitrageurs. A cross-chain arbitrage bundle requires atomic execution across chains. Only the entity controlling the transaction ordering on the destination chain can guarantee this atomicity, forcing all cross-domain MEV flow through their infrastructure.

This centralizes the value chain. Projects like Across and Stargate abstract bridging for users, but their underlying intent-based architectures still rely on a centralized solver/relayer network. These solvers must negotiate with sequencers (e.g., Arbitrum's single sequencer, Optimism's Superchain) to land profitable bundles, creating a power asymmetry.

The result is a meta-game. The real competition shifts from individual L2s to the sequencer-level coordination layer. Protocols like Espresso and Astria aim to decentralize this, but today's reality is concentrated risk. A sequencer failure or malicious action now threatens cross-chain liquidity, not just a single chain's state.

Sequencers capture cross-domain MEV by controlling the ordering of transactions across multiple rollups or appchains. This grants them a privileged view of pending intents across domains, enabling sophisticated arbitrage and front-running strategies that single-chain validators cannot replicate.

Intent-based architectures centralize power. Protocols like UniswapX and Across abstract execution to solvers, but the sequencer that finalizes the cross-chain bundle becomes the ultimate arbiter of which solver wins, extracting maximal value from the user's expressed intent.

Sovereignty shifts from L1 to L2. The security of Ethereum's base layer becomes irrelevant for transaction ordering; the economic power resides with the entity controlling the sequencer, as seen in the dominance of Arbitrum and Optimism's centralized sequencers over their respective ecosystems.

Evidence: Over 90% of cross-chain volume flows through a handful of centralized bridging protocols like Stargate and LayerZero, whose relayers and sequencers have first-mover advantage in identifying and capturing cross-domain arbitrage opportunities before they reach public mempools.

Sequencers become the new miners. Cross-domain arbitrage requires atomic execution across chains, which only a single entity controlling the transaction ordering on both sides can guarantee. This creates a natural monopoly for the sequencer.

Decentralized validators are irrelevant. A network can have thousands of validators, but if the sequencer is a single actor like Offchain Labs for Arbitrum, it controls the MEV supply chain. Validators merely attest to the sequencer's pre-ordered block.

Intent-based systems centralize differently. Protocols like UniswapX and CoW Swap route user intents through centralized solver networks. The winning solver, often a sophisticated MEV searcher, captures the cross-domain value, concentrating power in a few entities.

Evidence: Over 99% of transactions on major L2s like Arbitrum and Optimism are ordered by their respective centralized sequencers. This single point controls all cross-chain arbitrage opportunities flowing through native bridges.

Sequencers capture cross-domain value. They are the natural arbiters of transaction ordering across rollups and L1s, enabling them to internalize the most profitable cross-domain MEV opportunities that span chains like Ethereum and Arbitrum.

Bundling creates structural advantage. A sequencer like Optimism's can compose a user's swap, bridge, and stake into a single atomic bundle, outcompeting fragmented intent-based solvers from UniswapX or CowSwap that lack native cross-chain ordering.

This leads to vertical integration. To capture this value, infrastructure like Espresso or Astria will evolve into full-stack providers, merging sequencing with bridging (e.g., Across) and settlement, creating walled-garden efficiency at the cost of fragmentation.

Evidence: Over 90% of rollup transactions are ordered by a single centralized sequencer today, a model that cross-domain MEV revenue will entrench, not disrupt.