As protocols increasingly move back to the mainnet, how is Ethereum’s value logic changing?

Author: imToken

Are L2s cannibalizing L1's value? Is Ethereum losing its global composability?

During the two years when the Rollup narrative was at its peak, such anxiety almost permeated the entire Ethereum community. The market logic at the time was straightforward: since mainnet transaction costs were rising, more applications, assets, and users should be migrated to independent L2s.

However, as we entered 2026, some interesting changes began to emerge.

From Synthetix deploying its new perpetual exchange back to the Ethereum mainnet, to ENS abandoning the long-in-the-works Namechain, to projects like Trueo and Superseed returning to L1, project teams seem to feel that maintaining an independent L2 is no longer the optimal solution.

Some L2s have even begun to contract operations or shut down. For example, in early October, Blast, which once held over $2 billion in TVL, announced it would gradually wind down operations, followed shortly by Abstract announcing plans to shut down its network in December.

On the other hand, by integrating three core strategic pillars—"Scale," "Improve UX," and "Harden the L1"—the Ethereum Foundation has pushed mainnet performance into a new phase of engineering delivery. So, from "everything can be a Rollup" to "returning to the Ethereum mainnet," what exactly has happened to Ethereum?

From the "Migration Wave" to the "Return to Mainnet Wave"

In recent years, constrained by high mainnet gas fees, building custom L2s or deploying core business to other Rollups has become a standard move for many leading protocols.

But recently, the on-chain ecosystem has been undergoing an interesting reverse adjustment, with different types of projects re-evaluating their choice between L1 and L2 from perspectives such as infrastructure costs, asset security, and ecosystem liquidity.

The most representative case is ENS.

In 2024, ENS Labs planned a dedicated ZK Rollup (Namechain) for ENSv2, hoping to reduce domain registration and management costs through an independent execution environment.

But by February 2026, ENS co-founder Nick Johnson announced that the project would halt development on Namechain and deploy ENSv2 directly on the Ethereum mainnet. His reason was straightforward: Ethereum L1's scaling progress has far exceeded expectations from two years ago.

According to data disclosed by ENS Labs, with the increase in Ethereum's Gas Limit, the average gas cost for ENS domain registration has dropped by approximately 99% over the past year, falling from nearly $5 to under $0.05.

Meanwhile, maintaining an independent L2 not only requires bearing node and infrastructure costs, but also demands additional work in cross-chain resolution, bridging, security audits, and ongoing maintenance.

If a service can already be executed on the mainnet at extremely low cost, continuing to maintain an independent Rollup may actually make the product architecture more complex.

Similar shifts are also occurring in the DeFi space.

In June 2025, Synthetix, which had previously championed the Ethereum L2 ecosystem, published "L2s are Dead, Long Live Ethereum," pointing out that the past L2 scaling approach brought issues like liquidity fragmentation and unstable infrastructure to applications.

In December of the same year, Synthetix's next-generation perpetual exchange officially launched on the Ethereum mainnet.

In the second half of 2026, several more specific migration actions took place:

• Trueo: In September, announced it would migrate its primary deployment back from Base to Ethereum L1, aiming to leverage the mainnet's stronger integration capabilities, network effects, and neutrality to build a more trustworthy prediction market and result oracle;
• ShapeShift: On October 1, migrated its rFOX staking program from Arbitrum to the Ethereum mainnet. The rationale was practical: the vast majority of FOX tokens are already held on the mainnet, making additional cross-chain transfers for staking largely unnecessary;
• Superseed: In August, halted its original L2 network and plans to shift its ecosystem's core products to the Ethereum mainnet, no longer relying on independent chain gas revenue to support lending repayments, but instead achieving its original product goals through application-level revenue strategies;
• Nillion: Launched Blacklight L1 in October, migrating node operations originally run on Nillion L2 to the Ethereum mainnet, with plans to gradually decommission the original L2 network.

Of course, these cases shouldn't be simplified as all protocols fleeing L2, but they do indicate that an increasing number of project teams are beginning to reassess infrastructure choices based on the actual security, liquidity, and execution efficiency their businesses require.

More notably, operational pressures for some independent L2s are also starting to surface.

On October 2, Blast announced it would cease operations, citing that network maintenance costs had surpassed revenue generated by the L2, with no reliable path to economic sustainability in sight.

Subsequently, Abstract also announced plans to close its network on December 15. Despite having integrated over 144 applications and attracting more than 400,000 users, insufficient liquidity, a limited DeFi ecosystem, and a lack of institutional participation ultimately made independently operating a consumer-grade public blockchain unsustainable.

This also demonstrates that for many L2s, owning a chain does not equate to possessing a sustainable business model.

Especially against the backdrop of Ethereum continuously lowering L2 data publication costs and steadily improving mainnet performance, relying solely on low gas fees and token incentives to attract users is becoming increasingly difficult to sustain as a long-term competitive advantage.

Reconfiguring the L1-L2 Relationship: "State Remains on L1, Computation is Offloaded"

Why are project teams no longer fixated on maintaining a comprehensive, all-in-one L2?

Besides the economic equation, there is a more fundamental issue: as more applications and liquidity are dispersed across different networks, what impact is being placed on Ethereum's once-critical global composability?

In the initial Rollup-centric roadmap, the division of labor was actually very clear: L1 handles security, data availability, and final settlement, while L2 handles extreme scalability and low gas fees.

In that era when mainnet gas routinely cost dozens of dollars, this was virtually the only viable answer.

But reality proved far more complex. As of November 2026, data from ETHBEAT (formerly L2BEAT) lists dozens of different types of L2s, but only a single-digit number has reached Stage 2, with asset volumes remaining negligible.

In contrast, major Rollups like Base, Arbitrum One, and OP Mainnet remain at Stage 1.

Ultimately, L2s have progressed toward Stage 2 much slower than expected; meanwhile, L1's own scaling progress has accelerated, as Vitalik Buterin summarized: with the mainnet's internal scaling picking up speed, "the original vision for L2s and their role in Ethereum no longer hold."

More importantly, some L2s may proactively choose to remain at Stage 1 long-term due to regulatory, business model, or customer control considerations.

This doesn't necessarily mean they are wrong. For institutional applications requiring compliance with specific regulations, retaining certain management and upgrade privileges might actually be part of the product design. However, if a system relies long-term on a few entities holding key upgrade and state adjudication rights, it shouldn't be simply equated with a fully decentralized shard of the Ethereum mainnet.

Therefore, in concrete application architectures, a consistent pragmatic approach has emerged: keep assets and critical state on L1, while offloading high-frequency, complex computation to off-chain or alternative execution environments.

Synthetix's next-generation perpetual exchange is a prime example. It adopts a hybrid architecture where user assets remain custodied by Ethereum L1 contracts, order matching is handled by an off-chain high-performance CLOB (Central Limit Order Book) engine, and trade results are written back to the mainnet via batch settlement.

This way, users don't need to bridge large amounts of assets to another chain just to achieve trading speed, while still enjoying an order matching experience comparable to centralized exchanges.

Of course, this does not completely eliminate trust assumptions. Synthetix currently still relies partially on off-chain order books for margin management, and the off-chain matching system itself warrants scrutiny. However, asset custody and final withdrawals are still guaranteed by mainnet contracts, meaning users do not have to entrust all their fund security to a single L2's sequencer and bridge system.

This effectively offers an alternative perspective: it is not strictly necessary to move all computation on-chain to achieve true on-chain finance; the key lies in identifying which components require irreplaceable trust guarantees provided by on-chain rules.

Looking further ahead, the Native Rollup research direction championed by Vitalik could represent a more ideal solution.

Simply put, it uses Ethereum's native ZK-EVM verification mechanism to allow the mainnet to directly verify Rollup EVM execution results, reducing reliance on external safety committees, and creating conditions for more trustworthy interoperability and synchronized composability among different Rollups.

Meanwhile, solutions like the Ethereum Economic Zone (EEZ) are attempting to enable tighter synchronous interaction between different Rollups and the mainnet.

These technologies are currently at various stages of development and validation, but they collectively point in one direction: ensuring users are not forced to accept fragmented assets, liquidity, and trust systems simply because they use different execution environments.

Of course, L2s will not disappear. For high-frequency trading, gaming, privacy computing, and applications requiring autonomous control over execution environments, L2s remain highly valuable.

What truly needs to change is the past mindset that treated "scalability" as L2's sole core mission.

When Both L1 and L2 Become Cheap, How Does ETH Capture Value?

The preceding discussion focused on infrastructure and application architecture shifts, but it ultimately circles back to a more practical question: if the Ethereum mainnet grows stronger and more applications return to L1, will ETH's value thesis also evolve accordingly?

In recent years, the crypto market's valuation of ETH has relied heavily on the fee-burning mechanism introduced by EIP-1559.

Under this mechanism, the base fee portion of the transaction fee paid by users is burned, while the priority fee goes to validators. When network transaction demand is high and block space is tight, the base fee often rises, and ETH burn volume increases accordingly.

This formed the foundation of the past "Ultrasound Money" narrative. If the network continues to grow, fee burns have the potential to offset or even exceed ETH's new issuance, pushing ETH into a net deflationary state.

But the problem lies precisely here.

As upgrades like Dencun and Fusaka continuously lower the cost for L2s to post data to the mainnet, and the mainnet itself continuously raises the Gas Limit, the cost per transaction and fee revenue may consequently decline.

Even if the overall ecosystem's transaction volume continues to expand, it does not guarantee a proportional increase in ETH burns. Relying solely on higher gas fees to support ETH's valuation could instead force ordinary users and real-world applications to bear higher costs, ultimately undermining the network's long-term competitiveness.

Vitalik Buterin has specifically addressed ETH's value positioning, and his view is quite clear: rather than primarily understanding ETH as an asset whose value is derived solely from fee burning, he believes its positioning as a Store of Value deserves Ethereum's long-term focus.

Vitalik does not deny EIP-1559, nor does he advocate abolishing the fee-burning mechanism.

On the contrary, he considers this mechanism still reasonable and worth retaining. However, even if fee burns in the future consistently cover less than 10% of break-even issuance, holders should not be disappointed, because ETH's value derives from more than just burning.

Why?

First, the periods of highest ETH burn in the past largely stemmed from speculative activity within the crypto industry itself.

For instance, Yuga Labs' Otherdeed NFT minting in 2022 once generated around $180 million in fee burns in a single day. However, this revenue does not represent genuine external demand. If the broader crypto market shrinks and user speculative capacity declines, such transactions and associated fee revenues could vanish simultaneously.

Therefore, deriving ETH's long-term valuation solely from historical burn data cannot establish a reliable value floor.

Secondly, if the network is required to continuously generate sufficiently high fee revenues to justify ETH's multi-hundred-billion-dollar market cap, Ethereum might be forced to act like a large commercial corporation, continually increasing R&D investment, pivoting product directions, and engaging in more active governance intervention.

This is not entirely aligned with Ethereum's long-standing goals of decentralization and resistance to capture.

More importantly, deliberately maintaining high gas fees to boost burn volume inherently risks excluding the everyday users who most need privacy, security, and open financial services.

So, setting aside fee burns, what should form the foundation of ETH's long-term value?

In Vitalik's view, the answer lies in Ethereum's truly unique competitive advantages: decentralization, and what he summarizes as the CROPS characteristics—Censorship and Capture Resistance, Open Source, Privacy, and Security.

These attributes ultimately point to a more pragmatic value: stability, and the network's ability to continue operating reliably despite constantly shifting external environments.

Especially in an era of rapidly advancing AI technology and continuously evolving traditional finance and digital infrastructure, a financial infrastructure capable of maintaining long-term neutrality, reliability, and resistance to external control may hold far greater value than temporary spikes in fee revenues.

From this perspective, the strengthening of L1 is practically consolidating ETH's long-term value foundation across three dimensions.

• Core State and Settlement Capabilities Build Ethereum's Most Critical Network Effects: Computing power can constantly be commoditized, and execution environments can be replicated, but a shared settlement layer hosting vast amounts of assets, mature financial protocols, and deep liquidity is exceedingly difficult to rebuild in a short timeframe. This means ETH's position as a foundational gas asset, native collateral, and settlement asset stands to be reinforced as core financial activities converge back;
• Mainnet Scaling Allows ETH Value to Decouple From High Single-Transaction Fees: Over the past two years, Ethereum has been working to resolve a long-standing contradiction: how to increase throughput without letting node hardware barriers spiral out of control. If a series of technologies continue to mature, Ethereum has the opportunity to evolve from a historically "low-frequency, high-cost" settlement network into a global financial infrastructure featuring higher throughput and lower usage costs;
• Staking Mechanisms and Economic Security Establish More Robust Underlying Demand for ETH: Viewed across the entire network, ETH's value is gaining more diversified backing. It serves as both the native resource required for network operation and the economic collateral for PoS consensus, while also acting as crucial collateral and liquidity within DeFi and on-chain financial systems. This paints a far more complete picture than relying exclusively on fee burns to explain ETH's long-term value.

Final Thoughts

If the development trajectory of Ethereum over the past few years is viewed as a scaling experiment, then the changes in 2026 resemble Ethereum beginning to recalibrate its ecological balance points.

L2s will not disappear, nor should they. They remain integral components for Ethereum to expand application scenarios, host differentiated execution environments, and serve more users.

At the same time, however, L1 cannot remain indefinitely confined to a foundational role limited solely to providing data availability and final settlement.

It needs to reinforce its execution capabilities, liquidity hub status, censorship resistance, and economic security, allowing the most critical assets and financial states to settle within a sufficiently trustworthy environment.

Just as the philosophy of Lean Ethereum advocates: subtracting redundancy and friction, and returning the core elements of security, censorship resistance, and state consensus to the protocol center.

And as the most valuable states and liquidity converge back onto the mainnet, a more powerful, uniformly experienced, and fundamentally solid Ethereum is opening a new chapter for its next decade.