LAYER2.REPORT
DISPATCH — 2026

RESEARCH BRIEF / SETTLEMENT SYSTEMS

Layer 2 markets have stopped being a scaling story. They are now a balance-sheet story.

A research dispatch on rollup economics, liquidity migration, sequencer concentration, and the quiet operational risks emerging beneath Ethereum's execution frontier.

The public narrative still treats Layer 2 networks as interchangeable pipes: cheaper blockspace, faster confirmation, more consumer throughput. That description is now out of date. The leading systems have become capital venues with distinct risk profiles, operating leverage, and monetary dependencies.

What matters is no longer whether a rollup can settle transactions at lower cost. The relevant question is whether its economic loop can survive a period of flat demand while paying for data availability, liquidity incentives, ecosystem grants, and security operations.

Execution capacity is abundant. Durable settlement demand is scarce.

Across the networks tracked in this brief, revenue quality varies more than throughput. The strongest chains show recurring application flow, resilient bridge balances, and declining incentive intensity. The weakest show high nominal activity with little retained value after subsidies are stripped away.1

EXHIBIT 01

Fee compression versus retained settlement margin

Lower user fees did not translate uniformly into lower operator margin. Networks with dense internal order flow retained more value per unit of posted data.

0255075100
Base
88
Arbitrum
81
Optimism
64
zkSync
49
Starknet
37
02
Liquidity is the new throughput

Liquidity remains the most reliable indicator of economic gravity. Transaction counts can be manufactured by campaigns, wallets, and automation. Stable balances are harder to fake because they must be useful somewhere else in the system.

The leading networks are increasingly differentiated by the quality of assets they attract. Chains with native exchange depth, payments flow, and lending collateral show slower but more durable expansion. Chains dependent on rotating farms produce impressive bursts that decay once incentives move.

Bridge flow has also become less directional. In earlier cycles, assets moved outward from Ethereum and stayed. In the current regime, liquidity behaves like treasury cash: it seeks yield, rotates quickly, and penalizes networks that cannot maintain credible exit liquidity.

EXHIBIT 02

Net bridge liquidity, indexed by quarter

Q1Q8Q16+186%
03
Sequencers remain a concentration point

Decentralization promises remain uneven. Most Layer 2 networks have reduced user cost faster than they have reduced operator discretion. The gap is defensible in early deployment, but it becomes harder to justify as networks secure larger balances and institutional flows.

Sequencer design is now a governance and market-structure issue. Whoever orders transactions controls latency rents, MEV capture, outage response, and in some cases the practical ability to censor. The market has begun to price this implicitly through liquidity preference rather than explicitly through risk disclosures.

Security assumptions are moving from whitepapers into treasury policy.

The next phase of competition will reward networks that can make operational guarantees legible: fault proofs that work, escape hatches that can be used, and sequencing roadmaps backed by shipped code rather than forum language.

EXHIBIT 03

Operating profile comparison

Network
Proof status
Liquidity trend
Risk flag
Base
Active
+42%
Low
Arbitrum
Active
+31%
Moderate
Optimism
Stage 1
+18%
Moderate
Blast
Limited
-12%
High
04
Implications for allocators

The practical conclusion is narrow but important: Layer 2 analysis should move from technology taxonomy to financial underwriting. Rollup design still matters, but it matters through cost curves, user retention, liquidity depth, and credible control reduction.

For allocators, builders, and protocol treasuries, the winning question is not which network is fastest. It is which network can maintain useful blockspace when incentives normalize, data costs rise, and users become less willing to bridge for marginal rewards.

The sector is not consolidating because scaling failed. It is consolidating because scaling succeeded enough to make execution capacity common. Scarcity has migrated upward: from blocks, to liquidity, to trust.