Base and Arbitrum take over 80% of Ethereum L2 DeFi TVL as smaller rollups bleed
Linea bridge deposits fell from $976M to $367M in six months as the general-purpose L2 shakeout turns measurable.
Ethereum’s layer-2 DeFi liquidity is concentrating into a small set of winners, with Base and Arbitrum now controlling more than 80% of L2 DeFi TVL. Bridge-deposit drawdowns and at least one shutdown are sharpening the market’s read that weaker general-purpose rollups are in a consolidation phase, not that L2s are broadly failing.
Key Takeaways
- Base and Arbitrum now represent more than 80% of Ethereum layer-2 DeFi TVL, per DefiLlama data.
- Linea’s bridge deposits dropped from $976 million in November 2025 to $367 million in May 2026, a decline of more than 60%, per Token Terminal data.
- Zero Network announced it was shutting down last month, feeding the narrative that some general-purpose L2s are exiting.
- Post-Dencun, rollup data availability costs fell via blobs, and Messari research indicates DA is now a small slice of operator expenses for many OP Stack chains.
Base and Arbitrum Pull Away as L2 DeFi TVL Concentrates
The cleanest signal in Ethereum’s layer-2 market is where DeFi liquidity is actually sitting. Base and Arbitrum together account for more than 80% of Ethereum layer-2 DeFi total value locked (TVL), per DefiLlama data. For traders, that is the scoreboard. TVL is not a perfect measure of usage, but it is a direct proxy for where capital is willing to take smart contract risk and where liquidity is deep enough to support size.
This is less about “L2s dying” and more about winner-take-most dynamics asserting themselves. General-purpose rollups are increasingly substitutable products. If two venues offer similar execution environments, liquidity tends to cluster where distribution, integrations, and existing flows are strongest. The result is a market structure where the marginal L2 has to pay up to attract capital, then keep paying to prevent it from leaving.
Liquidity Leakage Shows Up in Bridge Deposits, With Linea as the Clearest Example
Bridge deposits are a practical stress gauge because they track whether assets are being parked on a chain or pulled back out. Over the past six months, several networks have seen declining bridge deposits, including Linea, World Chain, Starknet and Mantle.
Linea provides the clearest quantified example. Its bridge deposits fell from $976 million in November 2025 to $367 million in May 2026, a decline of more than 60%, per Token Terminal data cited. That kind of drawdown is hard to spin as noise. It reads as liquidity failing to stay anchored on a non-dominant rollup, which typically shows up next as thinner markets, weaker incentives, and a harder time sustaining DeFi activity.
Shutdowns Add to the Shakeout Narrative: Zero Network Exits
Zero Network announced it was shutting down last month, reinforcing the perception that another Ethereum L2 “bit the dust.” The exact date was not specified beyond “last month” relative to June 4, 2026, but the timing matters less than the direction of travel. Closures tend to cluster when the economics of staying live no longer clear the hurdle rate.
Ben Fisch, co-founder and CEO of Espresso Systems, framed the moment as “a consolidation phase for general-purpose layer twos, not layer twos broadly.” That distinction is important for positioning. The tech category can be healthy while the long tail of undifferentiated chains gets compressed.
Signals Traders Can Track as General-Purpose L2s Consolidate
The first threshold is whether Base plus Arbitrum keep their share of Ethereum L2 DeFi TVL above 80% on DefiLlama, or whether that concentration starts to unwind.
Second, Linea’s bridge deposits around the $367 million May 2026 level are a near-term tell. Stabilization would suggest the outflow wave is slowing. Continued declines would confirm that liquidity is still migrating.
Third, watch for additional shutdown announcements following Zero Network’s exit, especially among general-purpose rollups without a clear distribution edge.
Finally, post-Dencun operator economics are shifting. Ethereum’s 2024 Dencun upgrade reduced rollup data-posting costs via blobs, and Messari research indicates data availability costs are now only a small fraction of operator expenses for many OP Stack chains. Any updated breakdowns that quantify where costs moved next will matter, because they clarify what chains must monetize to survive.
Consolidation Risk Is a Liquidity Problem, Not a Tech Problem
I don’t read this as an indictment of rollups. The evidence points to a capital allocation problem. When Base and Arbitrum hold more than 80% of L2 DeFi TVL, smaller general-purpose chains are competing for the same liquidity pool with fewer structural advantages.
The threshold that matters is whether bridge deposits and TVL can stop bleeding once incentives fade and DA costs are no longer the bottleneck post-Dencun. If concentration stays extreme and bridge balances keep trending down on the long tail, the setup starts to look structural rather than narrative-driven, because the gating factor becomes distribution and sustained demand, not the viability of the technology.
