Ethereum (ETH) is trading near $2,327 as of May 9, 2026, up roughly 0.47% on the day, but the price chart tells only part of the story.
Beneath the surface, Ethereum's economic model is under the most serious structural stress it has faced since the network launched proof-of-stake, and the pressure is coming from the very scaling infrastructure that Ethereum itself made possible.
The numbers are stark. Since the activation of EIP-4844 in March 2024, Layer 2 networks have absorbed the vast majority of user-facing transaction volume while paying Ethereum only a fraction of what full calldata fees once generated.
Base, Arbitrum, and Optimism collectively process multiples of Ethereum mainnet's daily transaction count, yet the blob fee revenue flowing back to the base layer has at times been measured in single-digit ETH per day. For a network whose bull case rests on being the settlement layer for a multi-trillion-dollar ecosystem, the question of whether that role generates sufficient economic return for ETH holders has never been more pressing.
TL;DR
- EIP-4844 slashed Ethereum's fee revenue by making L2 data posting dramatically cheaper, compressing the ETH burn rate that underpinned the "ultrasound money" narrative.
- Layer 2 networks now process more transactions than Ethereum mainnet by a wide margin, but most economic value, fees, MEV, sequencer revenue, is captured at the L2 layer, not by ETH holders.
- Ethereum's long-term value accrual thesis depends on blob fee markets maturing, restaking adoption via EigenLayer, and the eventual success of based rollups or enshrined PBS, none of which is guaranteed on a near-term timeline.
The EIP-4844 Bargain And What It Actually Cost ETH Holders
EIP-4844, colloquially known as "proto-danksharding," was activated on the Ethereum mainnet on March 13, 2024. Its stated goal was to reduce the cost of posting transaction data from Layer 2 rollups to Ethereum by introducing a new transaction type that carries data "blobs", large chunks of data that are stored temporarily and are not accessible to the EVM. The immediate effect on L2 users was dramatic and positive. Transaction fees on Base, Arbitrum One, and OP Mainnet dropped by as much as 95% within days of activation, making Ethereum-secured rollups genuinely competitive with Solana (SOL) and other monolithic chains on a cost basis.
But that bargain had a price. Before EIP-4844, Layer 2 networks paid Ethereum in full calldata fees to post their transaction batches. Those fees were denominated in ETH, were burned under EIP-1559, and contributed meaningfully to the deflationary pressure that the "ultrasound money" community tracked obsessively at ultrasound.money.
After EIP-4844, the same data posting shifted to blob transactions, which use an entirely separate fee market with its own base fee that can fall to near zero when demand is low.
In the months following EIP-4844's activation, daily ETH burn from blob fees frequently measured in the low single digits, compared to hundreds of ETH per day when L2s relied on calldata.
Ethereum Foundation researcher Justin Drake had framed proto-danksharding as the first step toward full danksharding, arguing the long-run vision would see thousands of blob slots per block, enormous data throughput, and a fee market large enough to compensate for the per-blob price reduction through sheer volume. That volume has not yet materialized at the scale needed to restore pre-EIP-4844 burn rates.
The result is a network that processes more data than ever before while collecting less ETH revenue from that data, a trade-off that is rational for ecosystem growth but genuinely complicated for ETH as a monetary asset.
Also Read: WOJAK Posts 14% Gain, $8M Volume In Crypto’s Meme-Nostalgia Trade
Layer 2 Transaction Volume Now Dwarfs Ethereum Mainnet
The scale of L2 activity relative to Ethereum mainnet is no longer a marginal observation. According to data from L2Beat, the combined daily transaction count across all tracked Layer 2 networks has consistently exceeded Ethereum mainnet transactions by a ratio of between 5:1 and 10:1 throughout the first quarter of 2026. Base alone, the L2 network launched by Coinbase in August 2023, has on multiple days processed more transactions in 24 hours than Ethereum mainnet.
Arbitrum One and OP Mainnet add further volume. The three leading Optimistic Rollups together represent the majority of tracked L2 activity. ZK-rollup ecosystems including zkSync Era and Polygon zkEVM contribute additional throughput. The aggregate daily transaction count across all L2Beat-tracked chains has surpassed 10 million transactions on peak days, a figure that Ethereum mainnet alone has never approached given its ~1 million daily transaction ceiling at current block gas limits.
L2Beat data shows total value locked across all Ethereum Layer 2 networks has exceeded $40 billion in 2026, with Arbitrum and Base commanding the largest individual shares.
What this volume distribution means for Ethereum's economic model is nuanced. L2 activity does eventually settle on Ethereum, state roots are posted, fraud proof windows expire, and withdrawal bridges are secured by Ethereum's validator set. But the users of those L2 networks pay their transaction fees to L2 sequencers, not to Ethereum validators directly. The MEV extracted on L2 blocks accrues to L2 block builders or sequencer operators. The gas fees paid by DeFi users on Base are denominated in ETH but collected by Coinbase's sequencer infrastructure first. Ethereum mainnet benefits from the periodic batch posting costs, and nothing more.
Also Read: Toncoin’s $6.7B Telegram Bet Holds Firm Despite Market Dip
The Fee Revenue Collapse In Hard Numbers
To understand the magnitude of the revenue shift, it helps to anchor the analysis in specific data points. Before EIP-4844's activation, Ethereum's annualized fee revenue regularly exceeded $1 billion USD equivalent during periods of elevated network activity. The Electric Capital Developer Report and on-chain fee trackers such as Token Terminal documented peak months in 2023 where Ethereum collected north of $300 million in fees in a single calendar month.
Post-EIP-4844, the fee environment has been structurally altered. Blob base fees are set by a separate EIP-1559-style mechanism targeting a fixed number of blobs per block. When the number of blobs submitted is below the target, the fee falls. In the weeks following activation, blob demand was frequently below target, driving the blob base fee to its floor value of 1 wei. Mainnet base fees also compressed as L2 batch-posting calldata, which previously consumed large amounts of gas, was replaced by the far more gas-efficient blob transactions.
Token Terminal data shows Ethereum's monthly protocol revenue fell by roughly 60-80% in the quarters immediately following EIP-4844 activation compared to equivalent activity periods in 2023.
The ETH issuance and burn dynamic shifted accordingly. Under proof-of-stake, Ethereum issues new ETH to validators as block rewards. EIP-1559 burns the base fee component of every transaction. When fee revenue is high, burns exceed issuance and supply is deflationary. When fee revenue is low, issuance exceeds burn and supply expands. Throughout much of 2025 and into 2026, Ethereum's supply has been in mild inflation rather than deflation, a significant departure from the narrative that propelled ETH's price during the 2021 and late 2023 cycles.
Also Read: Ondo Finance Posts 24% Rally, Turning RWA Tokens Into The Day’s Hot Trade
Sequencer Revenue And Who Actually Captures L2 Value
The concentration of sequencer revenue is one of the most under-discussed structural issues in the Ethereum ecosystem. Every major Optimistic Rollup currently operating in production, Arbitrum, Base, OP Mainnet, Blast, and others, operates with a single centralized sequencer. That sequencer collects user fees, orders transactions, and extracts MEV before submitting batches to Ethereum. The revenue generated in that process flows to the sequencer operator, not to ETH stakers or Ethereum validators.
Offchain Labs, the company behind Arbitrum, has disclosed that Arbitrum One sequencer revenue has been substantial, though exact figures are not broken out in public financial statements. OP Labs and Coinbase's Base have similarly benefited from sequencer fees during periods of elevated DeFi activity. Research from Flashbots on MEV extraction patterns suggests that sequencer-controlled ordering on L2 networks generates significant value that is structurally inaccessible to Ethereum's base layer under current designs.
Flashbots research has estimated that MEV on L2 networks could eventually rival or exceed mainnet MEV in absolute dollar terms as L2 activity scales, but under current architectures, Ethereum mainnet captures none of it.
The planned transition to decentralized sequencers and "based rollups", where Ethereum validators themselves sequence L2 transactions, is the proposed solution to this value leakage. Based rollups, as described in a research post by Justin Drake on ethresear.ch, would route MEV and sequencer fees back through Ethereum's proposer pipeline. But the timeline for based rollup adoption remains speculative. No major production rollup has committed to a firm based sequencing launch date, and the technical challenges, including cross-domain MEV, latency, and censorship resistance, are significant.
Also Read: Sui Gains 4.3%, Outpaces Aptos In The 2026 Move Ecosystem Race
EigenLayer Restaking As A Compensating Revenue Stream
The most prominent attempt to create a new ETH value accrual mechanism beyond transaction fees is EigenLayer, the restaking protocol that launched on Ethereum mainnet in 2024. EigenLayer allows ETH stakers to "restake" their staked ETH, or liquid staking tokens representing staked ETH, to provide cryptoeconomic security to third-party services called Actively Validated Services, or AVSs. In exchange, restakers earn additional yield denominated in the tokens of those AVS protocols.
The growth of EigenLayer has been substantial. Total restaked value reached billions of dollars in ETH-equivalent terms within months of launch, making it one of the fastest-adopted new primitives in Ethereum's history. EigenDA, the data availability layer built on EigenLayer, has attracted significant adoption from rollup teams seeking a cheaper alternative to posting blobs directly to Ethereum, which adds a layer of irony to the fee revenue dynamic, as EigenDA competes with Ethereum's native blob market for L2 data availability business.
EigenLayer's restaking model has attracted over $15 billion in restaked value at peak, though the actual yield generated for restakers depends entirely on the revenue models of individual AVS protocols, many of which remain early-stage.
The critical question for ETH holders is whether EigenLayer restaking meaningfully translates into sustained demand for ETH as an asset.
The mechanism does create an additional utility layer, ETH is required to participate in restaking and to provide slashable stake to AVSs. But the yield currently offered by most AVSs is modest relative to native staking yields, and the risk of slashing from AVS participation adds a complexity that many institutional stakers have been cautious about. Academic work on restaking systemic risk, including a paper published on arXiv by researchers examining correlated slashing scenarios, has flagged potential cascading failures if multiple AVSs face simultaneous slashing events.
Also Read: WOJAK Down 9.6% While PENGU Holds: What The Meme Token Divergence Says About The Market
The Competing L1 Threat From Solana And Sui
Ethereum's positioning challenge is not purely internal. The external competitive landscape has hardened considerably since 2024. Solana has established itself as the dominant chain for retail-facing consumer applications, memecoins, and high-frequency trading activity. Solana's daily active address counts and transaction volumes have, on multiple occasions in 2025 and 2026, exceeded all Ethereum L2s combined, a data point that would have seemed implausible two years ago.
Sui, currently ranked 28th by market cap with a price near $1.06 and a $4.26 billion market capitalization per CoinGecko data as of May 9, 2026, represents the newer wave of Move-language L1 challengers. Sui (SUI)'s object-centric execution model allows for parallel transaction processing that Ethereum's sequential EVM cannot match at the base layer. Daily volume on Sui has approached $724 million on active days, and its DeFi TVL has grown rapidly as developers port applications from Ethereum and Solana.
The combined DeFi TVL of Solana, Sui, and Avalanche (AVAX) has grown from roughly 5% of Ethereum's TVL in 2022 to over 30% by early 2026, according to DefiLlama aggregate data, a structural shift in where economic activity settles.
For Ethereum, the competitive threat from alternative L1s is fundamentally different from the L2 challenge. L2s are Ethereum-adjacent, they borrow security from the base layer even if they capture most of the fee revenue. Competing L1s represent a direct challenge to Ethereum's position as the settlement layer of choice.
If developers and users route activity to Solana or Sui rather than Ethereum L2s, even the attenuated fee revenue from blob posting is lost entirely. The network effects that once made Ethereum's lead appear insurmountable have weakened, even if Ethereum still commands a $280 billion market capitalization that no rival approaches.
Also Read: Monad Joins Trending Top 15, Putting Parallel EVM Claim Back In Focus
DeFi TVL Concentration And What It Reveals About Ethereum's Moat
Total Value Locked remains one of the most-cited metrics for assessing blockchain ecosystem health. By this measure, Ethereum still dominates decisively. DefiLlama data shows Ethereum mainnet holding roughly $50-60 billion in DeFi TVL in May 2026, with all L2s adding another $40+ billion in ETH-secured TVL. The combined Ethereum ecosystem, mainnet plus L2s, represents well over 60% of all DeFi TVL across all chains.
But TVL has important limitations as a value-accrual metric for ETH. The largest TVL categories on Ethereum mainnet are liquid staking protocols, primarily Lido Finance, which holds over $30 billion in stETH deposits, and lending protocols like Aave and Compound.
Much of this TVL generates relatively modest fee revenue compared to the dollar values it represents. Liquid staking protocols collect a thin spread on staking rewards. Lending protocols generate interest income, but the marginal ETH burned per dollar of TVL in these protocols is low compared to high-frequency DEX trading or NFT minting periods.
DefiLlama shows Ethereum's share of total DeFi TVL across all chains has declined from over 90% in 2021 to approximately 60-65% in early 2026, still dominant, but the directional trend is toward fragmentation.
The protocols driving the highest fee revenue per unit of TVL, perpetual DEXs, high-frequency spot markets, and memecoin launchpads, have migrated disproportionately to Solana and to Ethereum L2s. Hyperliquid (HYPE), with a market cap above $10 billion and daily DEX volumes over $200 million, operates its own L1 entirely outside the Ethereum security model. The premium DeFi activity that generated Ethereum's highest fee periods is the activity most likely to route to specialized, low-latency, low-cost environments. Ethereum mainnet increasingly serves as a settlement and collateral layer for large, infrequent transactions, a valuable role, but one with a different fee profile than a high-throughput trading environment.
Also Read: Real-World Asset Token Ondo Climbs 12%: What Is Driving The Move
The Institutional ETF Effect And Its Limits For Value Accrual
One genuine positive development for ETH in 2025 and 2026 has been the launch and sustained growth of spot Ethereum ETFs in the United States. The SEC approved spot Ethereum ETF applications from BlackRock, Fidelity, and several other asset managers in mid-2024, following the approval of spot Bitcoin ETFs earlier that year. Institutional inflows into Ethereum ETF products have added a layer of structural demand for ETH that did not exist in prior cycles.
However, spot ETF ownership does not translate into on-chain fee generation in the same way that active DeFi participation does. ETF holders are exposed to ETH price appreciation and depreciation, but their ETH is custodied off-chain by ETF trustees and does not participate in staking, restaking, DeFi protocols, or L2 bridge activity. The ETF structure also does not stake the underlying ETH, a deliberate structural choice by the SEC and by ETF issuers to avoid securities classification complexity, but one that removes a portion of the float from the productive staking economy.
BlackRock's iShares Ethereum Trust has attracted multi-billion-dollar inflows since launch, making institutional ETF demand a significant price support factor, but ETF holders generate zero on-chain fee revenue for Ethereum validators.
For the ETH value accrual thesis, ETF demand is a mixed signal. It supports price through demand-side pressure on supply, and it expands ETH's addressable market by making it accessible through brokerage accounts. But it does not address the fee revenue compression described in earlier sections of this piece. An ETH price supported primarily by ETF inflows and passive institutional demand, rather than by strong on-chain fee generation and deflationary burn dynamics, represents a structurally different asset than the "ultrasound money" narrative envisioned. The long-run sustainability of that price support depends on whether ETF investors' thesis is price appreciation, yield, or both.
Also Read: AURA Triples In 24 Hours As Solana's Low-Cap Memecoin Trade Heats Up Again
Roadmap Milestones That Could Restore ETH's Fee Economics
The Ethereum development roadmap contains several milestones that proponents argue will resolve the current fee revenue compression. The most significant near-term upgrade is full danksharding, which would increase the number of blob slots per block from the current 3-6 to potentially 64 or 128.
Under the assumption that L2 demand for blob space grows proportionally with transaction volume, a dramatically larger blob market could generate fee revenue that more than compensates for the per-blob price reduction introduced by EIP-4844.
The Ethereum Foundation's Pectra upgrade, which activated in May 2025, introduced EIP-7251 to raise the maximum effective validator balance, EIP-7549 for attestation improvements, and a suite of account abstraction improvements under EIP-7702. These changes improve staking efficiency and user experience but do not directly address the blob fee market. The subsequent Fusaka upgrade planned for late 2025 and into 2026 includes PeerDAS (Peer Data Availability Sampling), a step toward full danksharding that would increase blob throughput without requiring every node to download all blob data.
Ethereum's roadmap assumes that blob demand will scale with L2 adoption and that fee market competition for blob space will increase as L2 throughput grows, but that scaling of demand has not yet matched the pace of blob supply increases.
Beyond data availability, enshrined Proposer-Builder Separation (ePBS) and the broader "Beam Chain" redesign proposed by Justin Drake represent longer-horizon changes that could alter Ethereum's consensus layer economics. The Beam Chain proposal, presented at Devcon Bangkok in late 2024, envisions a ground-up redesign of Ethereum's consensus layer using SNARK proofs, faster finality, and reduced validator entry barriers. If implemented, it would significantly change the economics of Ethereum validation, but a realistic timeline for Beam Chain deployment extends several years into the future, well beyond any near-term investment horizon.
Also Read: BILL Surges 26% As Billions Network Sees Volume Outrun Its Market Cap
How Markets Are Pricing Ethereum's Value Accrual Uncertainty
The tension between Ethereum's dominant network effects and its compressed near-term fee economics is legible in market data. Ethereum's price-to-fees ratio, a rough analog of the price-earnings multiple applied to on-chain revenue, has expanded dramatically since EIP-4844. When Ethereum was generating hundreds of millions of dollars per month in fee revenue at a $200-300 billion market cap, the implied multiple was comparable to mature tech infrastructure companies. At current fee revenue levels and a similar market cap, the multiple implies either that markets expect fee revenue to recover substantially or that they are pricing ETH primarily on non-revenue factors such as store of value, ETF demand, and optionality on future upgrades.
Token Terminal data shows Ethereum's price-to-fees ratio reaching historically elevated levels in 2025 relative to its own history. This does not necessarily indicate overvaluation, infrastructure assets with strong network effects often trade at high multiples to current revenue, but it does indicate that the current price depends heavily on forward-looking assumptions about fee recovery. Markets are, in effect, pricing in the success of full danksharding, based rollup adoption, and EigenLayer AVS revenue growth simultaneously.
Ethereum's market capitalization of approximately $280 billion implies a dramatically higher revenue multiple than in prior cycles when on-chain fee generation was at peak levels, markets are explicitly pricing in a recovery thesis that has not yet materialized in the data.
The ETH/BTC ratio provides another lens on relative positioning. Bitcoin (BTC) has maintained and extended its dominance during the current cycle.
The ETH/BTC ratio has trended lower since the 2021 peak, reflecting both the relative maturity of BTC's investment thesis and the ongoing uncertainty around ETH's fee economics. Institutional capital allocating to crypto has skewed heavily toward Bitcoin in the current cycle, with ETF inflows into BTC products substantially exceeding those into ETH products by most accounts. For ETH to reclaim a higher ETH/BTC ratio, it likely needs either a demonstrable recovery in on-chain fee revenue or a compelling new value accrual narrative that resonates with institutional allocators at the scale now active in the market.
Read Next: Privacy Trade Roars Back: Zcash Climbs 110% In One Month
Conclusion
Ethereum at $2,327 is not a network in crisis. Its developer ecosystem remains the largest in crypto by almost any measure. Its security model, backed by hundreds of billions of dollars in staked ETH, is unmatched. Its DeFi TVL lead over competing blockchains, while narrower than in prior cycles, remains substantial. The Ethereum Foundation and its affiliated research community continue to produce the most technically ambitious roadmap in the industry.
But the current moment represents a genuine inflection point for the ETH value accrual thesis. EIP-4844 was the right call for ecosystem growth, cheaper L2 fees have driven real user adoption and made Ethereum-secured rollups competitive with monolithic chains.
The cost, however, was a structural reduction in the fee revenue that drives ETH burn and supports the deflationary monetary premium. The mechanisms designed to restore that revenue, full danksharding blob markets, based rollup sequencer fee routing, and EigenLayer AVS yield, are real but unproven at scale and are measured in years, not months.
Investors and analysts assessing ETH in 2026 are therefore facing a specific kind of uncertainty that differs from the risks of earlier cycles. This is not primarily a regulatory risk, a security risk, or a technology risk. It is a value accrual timing risk, a question of whether Ethereum's roadmap delivers fee market recovery before competitive pressure from Solana, Sui, and application-specific chains erodes the user base that would generate that revenue. The answer will likely define the trajectory of ETH's price and its share of crypto market capitalization for the remainder of this decade.
Read Next: Zcash Enters Top 15 Crypto Assets As ZEC Volume Surges Past $1B