Three Ethereum Layer 2 infrastructure projects — Zero Network, Everclear, and Syndicate Labs — announced shutdowns on the same day, May 21, 2026. The simultaneous wind-downs mark the most visible casualties yet in a consolidation cycle that has eliminated over 50 rollups since late 2024 and conce...
"The projects that have thrived are highly customised, with their execution environments built entirely from scratch. Our framework does not fall into either category." — Will Papper, Co-founder, Syndicate Labs
Three Ethereum Layer 2 infrastructure projects — Zero Network, Everclear, and Syndicate Labs — announced shutdowns on the same day, May 21, 2026. The simultaneous wind-downs mark the most visible casualties yet in a consolidation cycle that has eliminated over 50 rollups since late 2024 and concentrated roughly 77% of all L2 DeFi liquidity in two networks: Arbitrum One and Base.
Total value secured across the rollup ecosystem has fallen 36% from its October 2025 peak above $50 billion. Arbitrum One holds approximately $16.9 billion in TVL (40–44% market share), Base commands roughly $12.8 billion (46.6% of L2 DeFi TVL), and OP Mainnet takes a distant third at approximately 6%. The remaining 100-plus rollups share the scraps. 21Shares research describes many of these as "zombie chains" — technically operational but economically dead.
The shakeout has structural causes. After the Dencun upgrade reduced L2 data-posting costs by 80–90%, transaction fees ceased to be a differentiator. Competition shifted to liquidity depth, exchange integrations, wallet distribution, and developer ecosystem density — advantages that compound over time and resist new entrants.
Zero Network (Zerion): The gasless EVM rollup, operational since November 2024, will cease block production on July 31, 2026. Inbound bridging has been disabled immediately; outbound bridging remains open until the cutoff. Zerion's paymaster model subsidized every user transaction directly — a cost structure that proved unsustainable without sufficient wallet and API revenue to offset chain operations. The team will refocus on its core wallet and API products.
Everclear (formerly Connext): The cross-chain settlement protocol, founded in 2017 with early Ethereum Foundation support, is shutting down its Foundation, Labs unit, and all product development. Despite processing over $1.5 billion across 23 networks and reaching $500 million in monthly volume, the team stated that "the cross-chain solvers segment never developed the commercial depth we needed — users proved highly price-sensitive, and we were unable to convert that volume into meaningful revenue." The CLEAR token fell 48% to $0.0002332 in a single session. The protocol has been sunsetted; no user funds are reported stuck.
Syndicate Labs: Backed by a $20 million Series A led by Andreessen Horowitz in 2021, Syndicate spent five years building customizable rollup infrastructure for application-specific chains. The team stated the "rollup market has fundamentally shifted" and that for every new rollup launching, several more are quietly shutting down. Co-founder Will Papper noted the company explored pivoting to consulting-style rollup-as-a-service but concluded its core product could not adapt. The SYND token hit an all-time low on the announcement.
The data paints a clear picture of market stratification:
| Metric | Arbitrum One | Base | OP Mainnet | Rest (~100+ chains) | |--------|-------------|------|------------|---------------------| | TVL Share | 40–44% | ~33% | ~6% | ~17% | | TVL (est.) | $16.9B | $12.8–15B | ~$2.9B | Fragmented | | Daily Txns | High | ~15M | Moderate | Minimal | | Stage (L2Beat) | Stage 1 | Stage 0 | Stage 1 | Mostly Stage 0 |
Three optimistic rollups — Arbitrum, Base, and OP Mainnet — handle roughly 90% of all L2 transaction volume. The combined L2 ecosystem regularly processes 5–8 times Ethereum mainnet's daily transaction count, but this activity is heavily concentrated.
Total value secured across all rollups peaked above $50 billion in October 2025 and has since fallen to approximately $32 billion — a 36% decline. The top three networks have maintained or grown their share during this contraction, meaning the losses are concentrated almost entirely in the long tail.
Post-Dencun fee commoditization. The March 2024 Dencun upgrade (EIP-4844) introduced blob transactions that reduced L2 data-posting costs to Ethereum by 80–90%. Average transaction fees dropped from $0.10–$0.50 to $0.001–$0.01. When cheap fees become the default, they cease to be a competitive advantage. Every rollup using the same data availability layer costs roughly the same to operate.
Distribution trumps technology. Base benefits from Coinbase's 110+ million verified users and direct fiat-to-L2 onramp. Arbitrum has the deepest on-chain liquidity pools and was the first L2 to reach Stage 1 security maturity on L2Beat. These are network effects that compound — deeper liquidity attracts more protocols, more protocols attract more users, more users generate more revenue, and more revenue funds infrastructure investments that widen the moat.
Incentive exhaustion. Dozens of rollups launched in 2024–2025 with nearly identical tech stacks (OP Stack or Arbitrum Orbit forks) and competed on points programs and airdrop promises rather than product differentiation. When token generation events concluded and incentives dried up, users left. TVL collapsed 70–90% within weeks on many of these chains.
Capital efficiency pressure. DeFi liquidity is finite. Capital parked on a chain with $10 million TVL earns lower yields and faces higher slippage than the same capital on a chain with $10 billion TVL. Rational allocators consolidate toward the deepest pools, accelerating concentration.
The L2 landscape increasingly resembles a settled oligopoly rather than an open, competitive market. Technical barriers to entry are lower than ever — spinning up an OP Stack or Orbit fork takes days — but commercial barriers are higher than ever.
Arbitrum One holds $16.9 billion TVL, captures 40–44% market share, and is the only L2 among the top eight to have achieved Stage 1 on L2Beat's security maturity framework. Fees collected total approximately $308.5 million, though protocol revenue retained is significantly lower at roughly $4.8 million — a figure that underscores the thin margins in L2 operations even at scale.
Base processes approximately 15 million daily transactions, serves roughly 400,000 daily active addresses, and benefits from seamless Coinbase integration. In 2025, Base generated $75.4 million in revenue, accounting for 62% of total L2 revenue of $120.7 million. Its gasless transaction infrastructure and consumer-facing distribution have made it the primary retail L2.
OP Mainnet maintains a smaller but defensible position. OP token holders approved a proposal to allocate 50% of sequencer revenue to buybacks, signaling confidence in long-term viability. However, Base's announced departure from the OP Stack shared sequencer model to build its own unified tech stack has raised questions about Optimism's Superchain thesis.
In March 2026, OP Labs cut 20 employees — approximately 20% of its staff. CEO Jing Wang stated this was "not about finances" and that "OP Labs is well capitalized with years of runway," framing it as a decision to "do fewer things well."
The layoffs coincided with Base's announcement that it is building its own unified tech stack, effectively departing from the shared Optimism Superchain architecture. Base is the largest chain built on the OP Stack with billions in TVL, and its departure caused a sharp drop in the OP token price. The move undercuts Optimism's core value proposition: that multiple chains built on the OP Stack would share sequencing, security, and interoperability benefits.
This fracture illustrates a broader pattern: even among the winners, competitive dynamics are intensifying. The L2 oligopoly is not stable — it is an oligopoly in formation, where the remaining players are still jockeying for position.
21Shares research published in December 2025 introduced the term "zombie chains" to describe L2s that are technically operational but economically dead — minimal user activity, evaporating liquidity, and negligible developer engagement.
Blast is the canonical example. After launching with heavy incentive programs in 2024, its TVL peaked and then collapsed 97% to roughly $55 million by December 2025. The chain still produces blocks, but its DeFi ecosystem has hollowed out.
Kinto has shut down entirely. Loopring closed its wallet. The 21Shares report predicts the L2 ecosystem will consolidate around "a leaner, more resilient set of networks" by the end of 2026.
The pattern is consistent: launch with incentives, attract mercenary capital, fail to develop organic usage, lose liquidity when incentives end, enter a death spiral of declining TVL and developer departure.
From an economic value distribution perspective, the L2 consolidation reveals a fundamental tension in blockchain scaling strategy.
Value accrual concentrates at the distribution layer, not the infrastructure layer. Syndicate, Everclear, and Zero Network all built functional infrastructure. Their technology worked. But in a post-Dencun world where data availability is cheap and execution environments are commoditized, the economic value flows to whoever controls user access — wallets, exchanges, and fiat onramps.
Revenue remains thin even for winners. Arbitrum's roughly $4.8 million in protocol revenue against $308.5 million in fees demonstrates that L2 sequencing is not a high-margin business. Base's revenue position is stronger due to Coinbase's vertically integrated model, but standalone L2s face persistent pressure on unit economics.
The blob fee floor matters. EIP-7918, shipped in the Fusaka upgrade, ties the minimum blob base fee to the L1 execution base fee, ensuring that data availability is never priced at zero. This provides a floor for Ethereum's revenue capture from L2 activity, but it also means L2 operators face a structural cost floor they cannot negotiate below.
The Ethereum L2 market is undergoing the consolidation phase that many predicted but few prepared for. The May 21 triple shutdown is a symptom, not the cause. The root dynamics — fee commoditization, distribution-driven moats, and capital efficiency pressure — have been building since Dencun.
What remains is less a competitive open market than a forming oligopoly. Arbitrum has liquidity depth and security maturity. Base has Coinbase distribution and retail flow. OP Mainnet has the Superchain thesis, though its coherence is tested by Base's departure. The remaining 100-plus chains face a narrowing set of options: find a defensible niche, merge into a larger ecosystem, or wind down.
The economic lesson is consistent with every prior infrastructure cycle: when a layer commoditizes, value migrates up the stack to whoever controls demand. In L2s, that means wallets, exchanges, and applications — not execution environments.