Every blockchain, including Ethereum, has inherent limitations on how much data and how many transactions it can process. While Ethereum has been a pioneering platform for decentralized applications, its limitations, particularly around transaction speeds and costs, have led to the rise of L2 solutions and rollups.
However, the perception of L2s as "parasitic" entities has emerged due to the belief that they siphon value away from Ethereum by handling the bulk of user activity on their networks, leaving Ethereum with fewer transactions and lower fee revenue. The fear is that Ethereum’s economic value could be diminished as L2s, which operate as independent blockchains, capture most of the user base and transaction volume.
The economic relationship between Ethereum and its L2s is complex. L2s often rely on Ethereum for security, periodically settling their transactions on the Ethereum mainnet. However, Ethereum collects only a small fraction of the fees generated by L2 activity, leading to reduced revenue for the main chain. The fear is that as more users and developers migrate to L2s, Ethereum will struggle to justify its current valuation.
The migration to L2 has led to a significant decrease in transaction fee revenue generated on L1. This reduction is a double-edged sword. On one hand, it demonstrates the effectiveness of L2s in reducing costs for users, which is positive for user adoption and network growth. On the other hand, it raises concerns about the long-term economic sustainability of the L1 network, which has traditionally relied on transaction fees as a significant revenue stream.
This report dives into the numerous debates around the central question, “Are rollups parasitic to Ethereum?”
Things We Can All (Hopefully) Agree On
1. The “Are L2s parasitic to Ethereum?” narrative only emerged because of sluggish ETH price action.
If ETH was ~$4000+, this narrative/debate would almost certainly not exist. During downtimes and lulls in price action, people become frustrated and/or apathetic and need something to latch onto (example below). Add in the fact that the SOL price surged over the last 1.5 years, and a natural SOL vs ETH debate was bound to arise. Only this time, the proliferation of rollups are the scapegoat.
Note the date, which was two years ago, during the depths of the FTX bear market. While the Polygon PoS chain is completely architecturally different from Ethereum rollups, the point is that these debates are not new, and neither is the timing.
2. It is inarguable that ETH's price has lagged behind BTC and SOL for much of the last ~ two years.
However, the performance is HIGHLY dependent on where you start the chart. Starting the comparison at the market highs in November 2021 vs. the market lows in 2022 produces a very different picture/narrative. Regardless, we are left to figure out if this recent price underperformance is BECAUSE of the rise of L2s or merely coincidental, i.e., a convenient narrative.
Comparing BTC, ETH, and SOL price performance from Nov 2021 tops paints less of an ETH emergency narrative.
3 An endless number of things drive price action in crypto, some innate to the protocol and many, many more exogenous to the protocol.
Often the least of which is anything resembling “fundamentals” like users, fees, revenue, developers, dApp usage, etc. One thing is for sure: protocol fees are NOT one of the more significant aspects driving price appreciation. Every L1 in existence (outside of maybe Ethereum) had negligible fees prior to 2023. However, that has not stopped the market cap of many protocols from ballooning to comical valuation. Celestia was the poster child for this earlier in 2024 (image below). It was “earning” ~$200 a day in revenue, yet the protocol was valued at ~$18 billion. Therefore, attributing price movements to fees/revenue is, at the absolute best, a fraction of the overall story.
While pessimism surrounding Ethereum may arise from concerns about questionable scaling solutions or competition from other blockchain ecosystems, many of these challenges are not exclusive to Ethereum. As the space evolves, competition is/was inevitable. Let’s look to Bitcoin as an example. BTC was the first cryptocurrency; because of this, it was ~100% of the crypto market for years. However, over time, new, interesting protocols were built, and with each subsequent crypto bull/bear cycle, its market cap dominance has been diminished.
Ethereum is in a similar position within the smart contract blockchain market. It was the first and, therefore, enjoyed the lion’s share of the market for years. However, as the concept of smart contracts was proven out and accepted as a legitimate use case, new chains with varying tradeoffs and advancements have been created. It is irrational to believe that one chain, be it Bitcoin or Ethereum, was never going to cede market share to competition.
Right or wrongly, ETH’s underperformance has been largely attributed to a decline in Ethereum’s total transaction fees and activity, especially following the Dencun upgrade in March 2024 (despite the fact that SOL had been outperforming ETH for months at this point). The upgrade introduced binary large object (blob) storage, which shifted much of the transaction activity to Ethereum's rollups.
However, some believe that the migration to L2s is not the sole cause of ETH's underperformance. While only base and blob fees are burned according to EIP-1559 (and L2-generated fees have been minimal), even burning all L2 fees wouldn’t prevent ETH from becoming inflationary. This suggests that factors beyond L2 growth are at play. Additionally, while on-chain activity on Ethereum's layer-1 hit a four-year low by August 2024, it has since rebounded over the course of September 2024.
One example of out-of-protocol factors that may have contributed to ETH's recent price weakness is the unwinding of the Japanese yen (JPY) carry trade in early August. While this event triggered deleveraging across multiple asset classes, ETH may have been particularly vulnerable due to another exogenous factor, disappointing ETF flows. The launch of spot ETH ETFs failed to generate strong inflows, with a net outflow of $57M in the first two weeks, in contrast to a small net inflow of $69M for U.S.-based spot BTC ETFs during the same period.
4. If Ethereum did not have the rollup roadmap and was still stuck trying to scale the L1 with $50+ fees, sentiment would be a whole lot worse.
Ethereum’s shift from a single-layer (monolithic) blockchain attempting to solve scalability issues to a more modular structure, where Layer 1 functions as a settlement layer and Layer 2s take on additional responsibilities (execution), has been technically smooth but less successful in the court of public opinion. The issue is that this scaling strategy was always going to create profound shifts in its user activity and economic dynamics for Ethereum. That was by design because, literally, the only thing crypto Twitter could all agree on for years was that Ethereum fees were too high.
These rollup developments are a deliberate part of Ethereum's strategy to address scalability issues that have long plagued its network. By offloading a substantial portion of transaction processing to L2 solutions, Ethereum aims to birth countless new L2 environments while remaining the core, central piece that enables it all to work. This transition, however, has sparked intense discussions regarding its economic impacts and the future role of ETH in this design.
5. Overall activity (L1 + L2 transactions) is way up, and fees are way down. This was the goal.
Fees/revenue were always supposed to go down. Like way down. Whether or not you like how rollups are currently constructed, they have inarguably accomplished what they set out to do: reduce the burden on the L1, offload execution, and provide a means of scaling for Ethereum.
Since the Denun upgrade in March 2024, which significantly reduced the cost of rollup transactions, the combined Ethereum+rollup daily transaction count has roughly tripled while L2 TVL has been near all-time highs.
This illustrates that demand was there, but users were cost-sensitive. If rollups did not exist or remained cost-prohibitive, where would these users have gone? Would they have moved to an alt-L1 like Solana or BSC? It’s obviously impossible to know, but it's almost inarguable that having those users and that activity on a rollup is better than moving to an alt-L1 where there is no connection to Ethereum.
6. Let’s acknowledge the paradox we have constructed for Ethereum. High fees are bad, but so are low fees.
This contradictory thinking is, IMO, just another offshoot of topic #1. People are mad that the ETH price is lagging and thus need something to blame it on. It was high fees during the previous dull periods, and now it is too low fees. Again, fees are not the primary driver of token price—they never have been. But when a token isn’t mooning, people will find anything to blame.
Obviously, there are extremes on both ends that are not productive for a blockchain. If the fees are too high, you exclude certain people and use cases. If the fees are too low, the chain becomes insecure. But Ethereum is nowhere close to succumbing to either of these, especially compared to its competitors.
7. Ethereum L1 is not making as much revenue as before Dencun, but it is still competitive with the top L1s in terms of revenue while also experiencing far less inflation.
While Ethereum’s lead in fees is not as large as it once was, even with the addition of 100+ rollups over the last ~ three years, Ethereum mainnet is still generating 2x the fees of the second-place chain, Tron.
Meanwhile, Ethereum’s inflation rate is ~0.5%, better than nearly all alt-L1 competitors. This means that despite the recent histrionics around Ethereum, it still makes more revenue and issues less supply than its competitors. This leads to the question: Why all the worry about Ethereum and not for all the other L1 darlings like Solana, Aptos, and others? It is valid to critique the Ethereum-rollup model, but it is not like there is some proven, bulletproof alternative model that we know works today.
For reference, there are still questions about Bitcoin’s revenue/fee model. Additionally, Solana issues ~$6.35 M worth of SOL daily while generating less than $1.5 M in fees. This would make it far less economically sustainable than Ethereum, yet there is little concern for Solana (if you want to know why, just check the SOL price chart since 2023).
8. NONE of the chains today are valued on anything resembling traditional cash flows, revenue, DCF models, or business-like valuations. So, why are we sweating Ethereum’s revenue so much?
The “intrinsic” value of far more mature and socially accepted assets like BTC and even gold cannot be explained by any traditional financial model. Their market capitalizations seem disproportionate when viewed through traditional financial lenses such as earnings, productivity, or utility. Instead, these assets derive value primarily from their status as stores of value (SoV) and their perceived monetary premium. This concept prompts a broader question: what if the L1 blockchain that ultimately prevails is not the one that maximizes extraction of miner extractable value (MEV) but rather one that balances utility with its role as a store of value?
Unlike conventional businesses, which generate value through earnings, L1 blockchains will increasingly see their fees and MEV trend toward zero as network efficiency improves. Relying on cash flow measures to value these networks would, therefore, lead to a valuation of zero—an outcome that disregards their broader role in the ecosystem. Therefore, there must be other factors at play. Obvious qualities (at least to me) that command some sort of premium but are hard to model or quantify are SoV, “moneyness,” decentralization and censorship-resistance, brand and reputation, acceptance from TradFi and governments, and others.
Things We’ll Probably Argue About
So, we’ve established that Ethereum L1 fees are down dramatically, but this has always been the plan. $100 L1 transaction fees without an alternative solution for users was never going to enable Ethereum’s vision of a global, neutral, permissionless economic settlement layer for all. We’ve also established that a user moving from the Ethereum mainnet to a rollup is better for Ethereum than losing that user and activity to an alt-L1. Rollups are a significant factor driving demand for ETH. Over 34 million ETH has already been transferred into various L2 ecosystems, effectively creating another liquidity drain for ETH. While the ETH moved to these L2s may not be subject to direct burning, the need for users to hold native tokens to cover transaction fees contributes to an indirect lock-up of ETH. This growing reserve of ETH, held by new wallets to manage fees, represents a soft lock on a substantial and increasing portion of the overall token supply.
Therefore, the logic looks something like this:
1. Are L2s Symbiotic with the Ethereum L1 or Parasitic?
First, a definition!
Parasite: an organism that lives in or on an organism of another species (its host) and benefits by deriving nutrients at the other's expense.
Ok, but we’re not dealing with animals…
A parasitic business relationship is one where one party gains at the expense of another, and is often characterized by an imbalance of power. An example of a parasitic business relationship would be a company that only extracts profit, and doesn't create/preserve more value than it takes.
While blockchains are not 1-to-1 analogous to companies with traditional revenue metrics (as discussed in previous sections), let’s use this framework to dissect the mainnet-rollup relationship and determine if they are, in fact, parasitic (e.g., they extract more value than they create/preserve.)
Before attempting to determine whether or not rollups are parasitic to the L1, we must first address another common debate in this narrative and that is “Are rollups even an extension of Ethereum?”
In essence, rollups—whether optimistic or zk—quite literally are extensions of Ethereum. They are smart contracts on the Ethereum mainnet. They also hold significant amounts of ETH, submit data and proofs back to the Ethereum network, and use ETH for gas fees. This is precisely the difference between a rollup and an alt-L1. There is no connection between Ethereum and an alt-L1 in terms of native token, security, consensus, fees, etc. There are for Ethereum and rollups.
An analogy often used to explain the L1-L2 relationship is that of states within a federal system. Just as Texas is part of the United States but retains its own governance and tax structures, L2s operate under Ethereum’s security and infrastructure while maintaining some level of independence. However, this relationship is not unconditional. For example, if an L2 decides to retain the EVM but uses its own gas token and settles transactions on another blockchain, it would no longer be considered a part of Ethereum.
The L1-L2 Economic Relationship
The relationship between Ethereum and rollups revolves around the economics of gas fees. As a quick refresher, Ethereum's native token, ETH, is critical to the network. As the principal medium for paying transaction fees, ETH is essential for any interaction on the network, including transfers of stablecoins, decentralized exchange (DEX) transactions, non-fungible token (NFT) minting, and on-chain gaming. This foundational utility drives persistent demand for ETH, much like the demand for oil in global markets.
In contrast to ETH, Layer 2 (L2) tokens primarily serve governance functions. While some L2 networks permit the use of their tokens for gas fees, the majority of these fees are still paid in ETH. L2s also need to pay for block space on Ethereum using ETH, indicating that L2 tokens lack an intrinsic utility that can drive demand. Rollups, which offer scalability solutions to Ethereum, generate their revenue primarily through user transaction fees but must also contend with certain costs.
The cost of producing block space is a key consideration in blockchain economics, with distinct differences between Layer-1 (L1) and Layer-2 (L2) systems. L1 Blockchains, like Ethereum, rely on token incentives to maintain Sybil resistance and achieve consensus through Proof-of-Stake (PoS). Ethereum’s block production involves transaction fees (Base Fee and Priority Fee) and Block Rewards. Ethereum's scalability challenges highlight a paradox: its supply-constrained block space increases transaction fees but contradicts the goal of providing low-latency, affordable compute.
Rollups purchase block space from Ethereum, offering higher scalability by offloading transaction execution off-chain and batching data onto Ethereum. This enables rollups like Arbitrum to scale more efficiently, with transaction throughput uncapped by Ethereum’s constraints. L1 blockchains face fixed costs (e.g., block rewards) and floating income (transaction fees), with constrained scalability and inconsistent margins. L2 rollups, by contrast, experience floating costs and income, offering scalable demand-driven throughput and consistent, higher profit margins.
One of the largest costs for L2s is the expense of storing transaction batches on Ethereum’s L1. Despite these expenses, many L2 projects have begun to demonstrate profitability, thanks in large part of the Dencun upgrade and EIP-4844, which introduced “blobs.” Blobs drastically reduce the cost of posting data to Ethereum, an aspect that was previously costing rollup teams millions per day.
Despite the dramatic cost savings offered by rollups, Ethereum’s mainnet remains integral to the overall structure, even as L2 rollups gain traction. Several activities will likely remain anchored on Ethereum's L1 for the foreseeable future. For example, protocols such as EigenLayer require restaking activities, and governance decisions for key decentralized finance (DeFi) protocols like Aave, Maker, and Uniswap remain firmly tied to the mainnet. Additionally, users with significant security concerns—often those managing large amounts of capital—are more inclined to maintain funds on L1 due to the enhanced security it offers. Despite the advances in scalability and efficiency brought about by L2s, ETH will continue to play a foundational role in these protocols, acting as both the primary unit of account and as necessary collateral for paying "rent" on Ethereum’s L1 infrastructure.
When L2 operators begin to generate substantial profits from user fees (image below), it suggests that Ethereum may be undervaluing itself. In August, L2 operator revenue increased by ~10%, which marks the first instance of rising revenue despite declining operational costs. This development points to the possibility that L2 projects can remain profitable in a sustainable manner, particularly as they benefit from new cost-saving mechanisms like data blobs.
2. What Do The Rollups Get Out of the Ethereum-Rollup Relationship?
This has been explained again and again over the years (including here and here) but the TL;DR is:
- Rollups inherit Ethereum’s security and no longer need to expend money, time, and resources on trying to bootstrap a decentralized community, validator set, and global consensus mechanism. Instead, they can deploy their chain and focus on the other aspects of building a successful blockchain.
- Once the rollup is live, it can easily tap into Ethereum’s industry-leading TVL, liquidity, dApps, users, developer tools, and network effects.
3. What Does Ethereum Get Out of This Relationship?
Retaining Fee-sensitive Users
If Ethereum had not scaled through the introduction of rollups and blobs, some portion of its activity would have shifted to other alt-L1 blockchains rather than to L2 solutions, which would have been a far worse scenario for Ethereum than just a loss in fee revenue. While there's currently a disparity between the revenue generated by rollup sequencers and what Ethereum’s layer 1 collects, this is likely only a short-term issue. Ethereum is in a phase of hyper-scaling, where it has more blockspace available than current demand can fill. However, now that numerous rollups offer sub-$.10 transactions, entirely new use cases, and users can stay within the Ethereum-rollup ecosystem.
High-frequency dApps and Use Cases
The advent of L2 technologies not only facilitates scalability and privacy enhancements but also introduces new application possibilities. By driving down fees and boosting transaction capacity, L2 solutions generate benefits across the board—for app developers, users, and ETH validators and holders. Looking forward, the ecosystem is likely to witness the emergence of numerous L2 blockchains and application-specific platforms. This diversified landscape is expected to feature a mix of large, general-purpose L2s, such as Arbitrum and Optimism, alongside others catering to specialized niches that require micro-transactions, streaming transactions, or simply numerous onchain interactions that would otherwise be too expensive on the mainnet.
Experimentation
Rollups and other Ethereum L2s are less constrained than the L1, in a technical sense but also in a social, economic, and development sense. Because rollups are not the central pillar responsible for security and consensus (that’s Ethereum mainnet) and because they are isolated environments, the stakes (while still high) are not existential to the Ethereum economy. This means they can conduct experiments from which Ethereum can learn, serving as a pseudo “testing ground.” For instance, Gnosis implemented the Dencun hard fork before Ethereum did, allowing the Ethereum Foundation to observe and potentially avoid mistakes when forking the mainnet. Others are experimenting with new virtual machines, parallelizing the EVM, using a UTXO structure rather than accounts, and more. Much of this experimentation could eventually migrate to Ethereum's L1. zkEVMs, in particular, are gaining attention as a potential integration into Ethereum’s main network.
Rollups can also present a new range of perspectives on scaling Ethereum due to their developers, priorities, and users. With the flexibility offered by L2 networks, each group can effectively design its own "state," unlike in the physical world, where land is limited. For instance, platforms like ZORA cater to artists, while Arbitrum focuses on DeFi and XAI on gaming.
4. Solana is the obvious solution
Over the past two years, Solana has emerged as the clear Ethereum alternative, experiencing tremendous growth in onchain metrics as well as offchain (project launches, developers, new tech advances). It differentiates itself with a unique focus on high-speed performance, cost efficiency, and an integrated/monolithic design. The platform's near-instantaneous transactions, ultra-low fees, and Web 2-style speed make the network intuitive and easy to navigate, enhancing its usability. This user-centric design eliminates the need for switching between networks or bridging assets, offering a more streamlined experience that mirrors traditional digital platforms.
Solana’s original vision aimed to create a decentralized marketplace with the efficiency of traditional financial systems, akin to an on-chain Nasdaq. A critical element that sets Solana apart is its integrated architecture, where all applications operate within a shared environment. This setup promotes composability, a feature that is highly valued in the DeFi space. Composability allows developers to interconnect various applications, creating a synergistic effect where protocols can be built upon one another, often referred to as "money legos." Solana’s unified state architecture facilitates this process by enabling all apps to work within the same state, eliminating the need for cross-network interoperability solutions.
One of Solana’s most significant advantages lies in its high throughput, allowing the network to process thousands of transactions per second. This scalability makes it ideal for applications requiring quick transaction times and low latency.
However, the network has not been without its challenges. Despite the inarguable performance enhancements and scalability advantages it offers, it is by no means a panacea. The chain currently handles < 4,000 TPS and has even struggled under this demand load. In early 2024, transactions regularly failed during peak congestion, raising concerns about reliability. Additionally, the chain has a long history of (short-term) downtime and outages. While a new client, Firedancer, has made real progress and looks like it has a legitimate shot at ten or 100x’ing the chains performance, it is still a work in progress and has yet be be released on mainnet. Even with Firedancer, it is a tremendous leap of faith to assume that the chain we’ve all experienced for the last ~five years has any shot of running the entire world’s financial activities atop it.
And that is, largely, the crux of many of the crypto debates: how much credit/goodwill are you willing to extend to one side and their ability to execute on their vision and not the other?
SOL bulls remain confident in the network's potential to eventually scale and handle the entirety of the world's financial activity despite numerous data points (outages) suggesting otherwise while operating at 1/1,000,000th of the needed throughput. At the same time, they often downplay the possibility that Ethereum could find an effective balance between its Layer 1 and rollup solutions. Such wildly imbalance positions reflect bias rather than a grounded assessment based on data or logic.
At present, many impartial observers would agree that Solana has matured into a more robust and decentralized platform than most Ethereum rollups. However, this was not always the case. In its early stages, Solana faced a series of challenges, including a lack of a functional fee market, frequent downtime, issues with initial token distribution, excessive reliance on a single entity (Sam Bankman-Fried and FTX), dependence on a single client, and limited scalability, capped at approximately 10,000 transactions per second. Remarkably, Solana has addressed nearly all of these issues within a three-year period. This raises the question of whether rollups should be afforded the same developmental leeway or be judged based on their current limitations.
It is true that Ethereum rollups currently face issues such as centralized sequencers, fragmented liquidity, and transaction throughput limitations. However, many leading projects have clear plans to mitigate or resolve these challenges. Ultimately, the debate centers on how much trust one is willing to place in these projects' ability to successfully execute their roadmaps and overcome their present obstacles.
Ethereum’s Solana’s Validator Revenue Problem
In the Solana ecosystem, the allocation of inflationary rewards is largely determined by the stake-weight of validators. Validators that accumulate more SOL tokens staked to their nodes experience a greater chance of being selected to produce blocks and, as a result, stand to earn more inflationary rewards than those with lower stake-weight. This relationship between a validator’s stake-weight and its probability of selection operates on a proportional basis. Specifically, a validator’s chance of being selected as the leader for a particular slot is calculated using a straightforward formula:
This linear correlation incentivizes validators to accumulate as much stake as possible since there is no diminishing return or penalty for amassing more SOL. Validators that command higher stakes directly increase their likelihood of being selected, thus allowing them to secure more rewards.
Jito has introduced a new level of sophistication to the Solana validator ecosystem by integrating a Maximum Extractable Value (MEV) marketplace directly into the validator client’s codebase. This customized validator client, a fork of Solana Labs’ official client, allows validators and their delegators to participate in MEV rewards that would otherwise be inaccessible. Jito enables validators to capture these rewards and pass them on to stakers, offering an additional 10-15% in SOL rewards.
The mechanism behind Jito's MEV marketplace operates through competition among MEV searchers. These participants identify profitable opportunities and submit a bundle of transactions to the Jito MEV marketplace, bidding for the chance to have their transactions included in the next block. The slot leader (validator) selects the highest bidder, and the associated MEV tip is distributed among the validator and its delegators. This process adds a layer of complexity to staking but offers significant benefits by increasing overall returns for stakers who delegate to validators running the Jito client.
Solana validators earn rewards not only through inflationary mechanisms but also from block rewards. These block rewards consist of two main components: base fees and priority fees. When a validator successfully proposes a block, they collect a base fee of 0.000005 SOL per signature. Priority fees, on the other hand, vary based on transaction demand and are determined by the sender of the transaction. Currently, Solana burns 50% of priority fees, base fees, and vote fees.
However, the revenue model for validators in 2024 has become increasingly constrained. A Dune dashboard reveals that ~90% of Solana validator revenue comes from SOL issuance, which is designed to only decrease over time. Fees and Maximal Extractable Value (MEV)—the additional revenue validators can earn from reordering transactions in a block—constitute the remaining ~10% of their income.
Further exacerbating the problem for smaller validators are the recent changes by the Solana Foundation’s delegation program. The Solana Foundation delegation program is designed to bolster small validators by covering some operational costs and staking SOL tokens with them to ensure the network remains decentralized. According to a report from Helius, approximately 72% of Solana validators benefit from this initiative. However, this financial support does not fully alleviate the burden on these validators. Their costs, many of which are fixed, increase the pressure as revenue depends largely on the stake they control and the value of SOL, both of which are in decline.
Profitability got even harder in 2024 due to the fact that the Solana Foundation recently reduced the maximum commission that validators participating in its delegation program can charge. The previous limit of 7% has been lowered to 5%, and a new cap of 10% has been imposed on Jito MEV, a form of validator income. Under prior conditions, validators could claim 100% of Jito MEV, but that potential has now been curbed.
For smaller validators reliant on the Foundation’s delegation program to stay operational, this policy change directly impacts their ability to generate sufficient revenue, as they can no longer charge as much for their services.
The financial implications of these changes are profound. One small validator estimated that, without Foundation support, a validator would need to control around 65,000 SOL to break even—up from 40,000 SOL previously. The relative profitability of recent months for small validators appears to be unsustainable, and the current trajectory suggests that continued reliance on the Solana Foundation's delegation program may not be viable in the long term. While the program has played a key role in decentralizing Solana’s validator set, it is unlikely to provide indefinite support. The recent reduction in allowed commissions is a signal that small validators must find ways to stand on their own. This shift presents a critical juncture for Solana’s network, as small validators must adapt or risk being left behind. Decentralization is a key strength of the network, but it remains to be seen whether Solana can sustain its decentralized nature in the face of such financial pressures.
However, with the upcoming protocol upgrade Agave 2.0 100% of the priority fees will be allocated to the block proposer. This shift in protocol design aims to further incentivize validators by ensuring they capture all priority fees, rather than burning a portion. The change is expected to be implemented in late 2024. However, it still does not address the centralizing, economies-of-scale issue at the heart of the Solana validator ecosystem. Larger validators will still be more profitable than smaller validators and over time, will consolidate down to just a select few.
Solana Extensions
Despite Solana’s current industry-leading (actual) TPS and promises of 1 million TPS in the near future, thanks to the Firedancer client, the Solana ecosystem is experiencing the beginnings of its own “L2” revolution. These new projects are known as Solana “Extensions” (L2s), and 5-10 are being developed/released in 2024.
*Side note*: Notice the careful language used: Extensions. Kudos to Solana for getting in front of the exact same future debate. No one is going to pose the question, “Are Solana Extensions an extension of Solana?” Just look at the name! Debate over.
While there are very real and distinct technical differences between Solana Extensions and a traditional Ethereum rollup, the fact that they are emerging may suggest that no matter your base layer scalability, there are valid reasons for “breaking off” from the primary L1. A promising alternative that maintains Solana’s composability while improving scalability is Magic Block’s concept of ephemeral rollups. These rollups offer a solution that allows applications to scale specific operations without disrupting the composability of the base layer. By leveraging Solana’s SVM account structure and parallelization capabilities, developers can temporarily lock one or more accounts and shift their state to an ephemeral rollup. During this period, the sequencer can modify these accounts within the rollup, while other non-delegated accounts continue to interact with programs on the base layer.
The benefits of ephemeral rollups are multifaceted. First, they preserve the composability of the base layer, allowing programs to interact with existing protocols and assets without interruption. This ensures that the foundational interconnectivity that distinguishes Solana from other blockchains remains intact. Developers also gain the advantage of using existing Solana infrastructure, such as the Anchor Framework, RPCs, and libraries, which reduces the friction of adopting new scalability solutions.
Should these new Extensions garner meaningful adoption, questions will naturally arise. Like is this an invalidation of the integrated/monolithic narrative? If these Extensions grow to be large enough, will they eventually spin off into their own chain? These questions are part of the natural progression of a successful L1. Much like the Ethereum debates around fees, MEV, governance, etc. that have already taken place, any chain that becomes significant enough with meaningful adoption and usage will have tough questions to answer across various entities with different priorities.
Conclusion
In conclusion, the rise of Layer 2 solutions, while presenting significant benefits in scalability and cost reduction, has sparked heated debates about their impact on Ethereum’s economic model. Despite concerns that L2s might be parasitic, extracting value from Ethereum without proportionally benefiting the network, the reality is more nuanced. Rollups, as extensions of Ethereum, rely on the mainnet for security and settlement, ensuring that Ethereum remains integral to the broader ecosystem. While fee revenue on Layer 1 has declined, Ethereum’s scalability strategy through rollups has helped retain users who might otherwise migrate to alternative L1 blockchains.
The reduction in fees and the shift in activity to L2s were intentional and reflected Ethereum’s vision of a modular, scalable blockchain. As Ethereum continues to evolve, balancing utility, security, and its role as a store of value will be key to its long-term success. Furthermore, the emergence of new competitors, such as Solana, highlights that Ethereum’s challenges are not unique, and innovation across the blockchain space is inevitable. Ultimately, while the relationship between Ethereum and L2s may seem contentious at times, it is this dynamic interplay that will likely shape the future of decentralized technology.
Disclaimer: This research report is exactly that — a research report. It is not intended to serve as financial advice, nor should you blindly assume that any of the information is accurate without confirming through your own research. Bitcoin, cryptocurrencies, and other digital assets are incredibly risky and nothing in this report should be considered an endorsement to buy or sell any asset. Never invest more than you are willing to lose and understand the risk that you are taking. Do your own research. All information in this report is for educational purposes only and should not be the basis for any investment decisions that you make.
Every blockchain, including Ethereum, has inherent limitations on how much data and how many transactions it can process. While Ethereum has been a pioneering platform for decentralized applications, its limitations, particularly around transaction speeds and costs, have led to the rise of L2 solutions and rollups.
However, the perception of L2s as "parasitic" entities has emerged due to the belief that they siphon value away from Ethereum by handling the bulk of user activity on their networks, leaving Ethereum with fewer transactions and lower fee revenue. The fear is that Ethereum’s economic value could be diminished as L2s, which operate as independent blockchains, capture most of the user base and transaction volume.
The economic relationship between Ethereum and its L2s is complex. L2s often rely on Ethereum for security, periodically settling their transactions on the Ethereum mainnet. However, Ethereum collects only a small fraction of the fees generated by L2 activity, leading to reduced revenue for the main chain. The fear is that as more users and developers migrate to L2s, Ethereum will struggle to justify its current valuation.
The migration to L2 has led to a significant decrease in transaction fee revenue generated on L1. This reduction is a double-edged sword. On one hand, it demonstrates the effectiveness of L2s in reducing costs for users, which is positive for user adoption and network growth. On the other hand, it raises concerns about the long-term economic sustainability of the L1 network, which has traditionally relied on transaction fees as a significant revenue stream.
This report dives into the numerous debates around the central question, “Are rollups parasitic to Ethereum?”
Things We Can All (Hopefully) Agree On
1. The “Are L2s parasitic to Ethereum?” narrative only emerged because of sluggish ETH price action.
If ETH was ~$4000+, this narrative/debate would almost certainly not exist. During downtimes and lulls in price action, people become frustrated and/or apathetic and need something to latch onto (example below). Add in the fact that the SOL price surged over the last 1.5 years, and a natural SOL vs ETH debate was bound to arise. Only this time, the proliferation of rollups are the scapegoat.
Note the date, which was two years ago, during the depths of the FTX bear market. While the Polygon PoS chain is completely architecturally different from Ethereum rollups, the point is that these debates are not new, and neither is the timing.
2. It is inarguable that ETH's price has lagged behind BTC and SOL for much of the last ~ two years.
However, the performance is HIGHLY dependent on where you start the chart. Starting the comparison at the market highs in November 2021 vs. the market lows in 2022 produces a very different picture/narrative. Regardless, we are left to figure out if this recent price underperformance is BECAUSE of the rise of L2s or merely coincidental, i.e., a convenient narrative.
Comparing BTC, ETH, and SOL price performance from Nov 2021 tops paints less of an ETH emergency narrative.
3 An endless number of things drive price action in crypto, some innate to the protocol and many, many more exogenous to the protocol.
Often the least of which is anything resembling “fundamentals” like users, fees, revenue, developers, dApp usage, etc. One thing is for sure: protocol fees are NOT one of the more significant aspects driving price appreciation. Every L1 in existence (outside of maybe Ethereum) had negligible fees prior to 2023. However, that has not stopped the market cap of many protocols from ballooning to comical valuation. Celestia was the poster child for this earlier in 2024 (image below). It was “earning” ~$200 a day in revenue, yet the protocol was valued at ~$18 billion. Therefore, attributing price movements to fees/revenue is, at the absolute best, a fraction of the overall story.
While pessimism surrounding Ethereum may arise from concerns about questionable scaling solutions or competition from other blockchain ecosystems, many of these challenges are not exclusive to Ethereum. As the space evolves, competition is/was inevitable. Let’s look to Bitcoin as an example. BTC was the first cryptocurrency; because of this, it was ~100% of the crypto market for years. However, over time, new, interesting protocols were built, and with each subsequent crypto bull/bear cycle, its market cap dominance has been diminished.
Ethereum is in a similar position within the smart contract blockchain market. It was the first and, therefore, enjoyed the lion’s share of the market for years. However, as the concept of smart contracts was proven out and accepted as a legitimate use case, new chains with varying tradeoffs and advancements have been created. It is irrational to believe that one chain, be it Bitcoin or Ethereum, was never going to cede market share to competition.
Right or wrongly, ETH’s underperformance has been largely attributed to a decline in Ethereum’s total transaction fees and activity, especially following the Dencun upgrade in March 2024 (despite the fact that SOL had been outperforming ETH for months at this point). The upgrade introduced binary large object (blob) storage, which shifted much of the transaction activity to Ethereum's rollups.
However, some believe that the migration to L2s is not the sole cause of ETH's underperformance. While only base and blob fees are burned according to EIP-1559 (and L2-generated fees have been minimal), even burning all L2 fees wouldn’t prevent ETH from becoming inflationary. This suggests that factors beyond L2 growth are at play. Additionally, while on-chain activity on Ethereum's layer-1 hit a four-year low by August 2024, it has since rebounded over the course of September 2024.
One example of out-of-protocol factors that may have contributed to ETH's recent price weakness is the unwinding of the Japanese yen (JPY) carry trade in early August. While this event triggered deleveraging across multiple asset classes, ETH may have been particularly vulnerable due to another exogenous factor, disappointing ETF flows. The launch of spot ETH ETFs failed to generate strong inflows, with a net outflow of $57M in the first two weeks, in contrast to a small net inflow of $69M for U.S.-based spot BTC ETFs during the same period.
4. If Ethereum did not have the rollup roadmap and was still stuck trying to scale the L1 with $50+ fees, sentiment would be a whole lot worse.
Ethereum’s shift from a single-layer (monolithic) blockchain attempting to solve scalability issues to a more modular structure, where Layer 1 functions as a settlement layer and Layer 2s take on additional responsibilities (execution), has been technically smooth but less successful in the court of public opinion. The issue is that this scaling strategy was always going to create profound shifts in its user activity and economic dynamics for Ethereum. That was by design because, literally, the only thing crypto Twitter could all agree on for years was that Ethereum fees were too high.
These rollup developments are a deliberate part of Ethereum's strategy to address scalability issues that have long plagued its network. By offloading a substantial portion of transaction processing to L2 solutions, Ethereum aims to birth countless new L2 environments while remaining the core, central piece that enables it all to work. This transition, however, has sparked intense discussions regarding its economic impacts and the future role of ETH in this design.
5. Overall activity (L1 + L2 transactions) is way up, and fees are way down. This was the goal.
Fees/revenue were always supposed to go down. Like way down. Whether or not you like how rollups are currently constructed, they have inarguably accomplished what they set out to do: reduce the burden on the L1, offload execution, and provide a means of scaling for Ethereum.
Since the Denun upgrade in March 2024, which significantly reduced the cost of rollup transactions, the combined Ethereum+rollup daily transaction count has roughly tripled while L2 TVL has been near all-time highs.
This illustrates that demand was there, but users were cost-sensitive. If rollups did not exist or remained cost-prohibitive, where would these users have gone? Would they have moved to an alt-L1 like Solana or BSC? It’s obviously impossible to know, but it's almost inarguable that having those users and that activity on a rollup is better than moving to an alt-L1 where there is no connection to Ethereum.
6. Let’s acknowledge the paradox we have constructed for Ethereum. High fees are bad, but so are low fees.
This contradictory thinking is, IMO, just another offshoot of topic #1. People are mad that the ETH price is lagging and thus need something to blame it on. It was high fees during the previous dull periods, and now it is too low fees. Again, fees are not the primary driver of token price—they never have been. But when a token isn’t mooning, people will find anything to blame.
Obviously, there are extremes on both ends that are not productive for a blockchain. If the fees are too high, you exclude certain people and use cases. If the fees are too low, the chain becomes insecure. But Ethereum is nowhere close to succumbing to either of these, especially compared to its competitors.
7. Ethereum L1 is not making as much revenue as before Dencun, but it is still competitive with the top L1s in terms of revenue while also experiencing far less inflation.
While Ethereum’s lead in fees is not as large as it once was, even with the addition of 100+ rollups over the last ~ three years, Ethereum mainnet is still generating 2x the fees of the second-place chain, Tron.
Meanwhile, Ethereum’s inflation rate is ~0.5%, better than nearly all alt-L1 competitors. This means that despite the recent histrionics around Ethereum, it still makes more revenue and issues less supply than its competitors. This leads to the question: Why all the worry about Ethereum and not for all the other L1 darlings like Solana, Aptos, and others? It is valid to critique the Ethereum-rollup model, but it is not like there is some proven, bulletproof alternative model that we know works today.
For reference, there are still questions about Bitcoin’s revenue/fee model. Additionally, Solana issues ~$6.35 M worth of SOL daily while generating less than $1.5 M in fees. This would make it far less economically sustainable than Ethereum, yet there is little concern for Solana (if you want to know why, just check the SOL price chart since 2023).
8. NONE of the chains today are valued on anything resembling traditional cash flows, revenue, DCF models, or business-like valuations. So, why are we sweating Ethereum’s revenue so much?
The “intrinsic” value of far more mature and socially accepted assets like BTC and even gold cannot be explained by any traditional financial model. Their market capitalizations seem disproportionate when viewed through traditional financial lenses such as earnings, productivity, or utility. Instead, these assets derive value primarily from their status as stores of value (SoV) and their perceived monetary premium. This concept prompts a broader question: what if the L1 blockchain that ultimately prevails is not the one that maximizes extraction of miner extractable value (MEV) but rather one that balances utility with its role as a store of value?
Unlike conventional businesses, which generate value through earnings, L1 blockchains will increasingly see their fees and MEV trend toward zero as network efficiency improves. Relying on cash flow measures to value these networks would, therefore, lead to a valuation of zero—an outcome that disregards their broader role in the ecosystem. Therefore, there must be other factors at play. Obvious qualities (at least to me) that command some sort of premium but are hard to model or quantify are SoV, “moneyness,” decentralization and censorship-resistance, brand and reputation, acceptance from TradFi and governments, and others.
Things We’ll Probably Argue About
So, we’ve established that Ethereum L1 fees are down dramatically, but this has always been the plan. $100 L1 transaction fees without an alternative solution for users was never going to enable Ethereum’s vision of a global, neutral, permissionless economic settlement layer for all. We’ve also established that a user moving from the Ethereum mainnet to a rollup is better for Ethereum than losing that user and activity to an alt-L1. Rollups are a significant factor driving demand for ETH. Over 34 million ETH has already been transferred into various L2 ecosystems, effectively creating another liquidity drain for ETH. While the ETH moved to these L2s may not be subject to direct burning, the need for users to hold native tokens to cover transaction fees contributes to an indirect lock-up of ETH. This growing reserve of ETH, held by new wallets to manage fees, represents a soft lock on a substantial and increasing portion of the overall token supply.
Therefore, the logic looks something like this:
1. Are L2s Symbiotic with the Ethereum L1 or Parasitic?
First, a definition!
Parasite: an organism that lives in or on an organism of another species (its host) and benefits by deriving nutrients at the other's expense.
Ok, but we’re not dealing with animals…
A parasitic business relationship is one where one party gains at the expense of another, and is often characterized by an imbalance of power. An example of a parasitic business relationship would be a company that only extracts profit, and doesn't create/preserve more value than it takes.
While blockchains are not 1-to-1 analogous to companies with traditional revenue metrics (as discussed in previous sections), let’s use this framework to dissect the mainnet-rollup relationship and determine if they are, in fact, parasitic (e.g., they extract more value than they create/preserve.)
Before attempting to determine whether or not rollups are parasitic to the L1, we must first address another common debate in this narrative and that is “Are rollups even an extension of Ethereum?”
In essence, rollups—whether optimistic or zk—quite literally are extensions of Ethereum. They are smart contracts on the Ethereum mainnet. They also hold significant amounts of ETH, submit data and proofs back to the Ethereum network, and use ETH for gas fees. This is precisely the difference between a rollup and an alt-L1. There is no connection between Ethereum and an alt-L1 in terms of native token, security, consensus, fees, etc. There are for Ethereum and rollups.
An analogy often used to explain the L1-L2 relationship is that of states within a federal system. Just as Texas is part of the United States but retains its own governance and tax structures, L2s operate under Ethereum’s security and infrastructure while maintaining some level of independence. However, this relationship is not unconditional. For example, if an L2 decides to retain the EVM but uses its own gas token and settles transactions on another blockchain, it would no longer be considered a part of Ethereum.
The L1-L2 Economic Relationship
The relationship between Ethereum and rollups revolves around the economics of gas fees. As a quick refresher, Ethereum's native token, ETH, is critical to the network. As the principal medium for paying transaction fees, ETH is essential for any interaction on the network, including transfers of stablecoins, decentralized exchange (DEX) transactions, non-fungible token (NFT) minting, and on-chain gaming. This foundational utility drives persistent demand for ETH, much like the demand for oil in global markets.
In contrast to ETH, Layer 2 (L2) tokens primarily serve governance functions. While some L2 networks permit the use of their tokens for gas fees, the majority of these fees are still paid in ETH. L2s also need to pay for block space on Ethereum using ETH, indicating that L2 tokens lack an intrinsic utility that can drive demand. Rollups, which offer scalability solutions to Ethereum, generate their revenue primarily through user transaction fees but must also contend with certain costs.
The cost of producing block space is a key consideration in blockchain economics, with distinct differences between Layer-1 (L1) and Layer-2 (L2) systems. L1 Blockchains, like Ethereum, rely on token incentives to maintain Sybil resistance and achieve consensus through Proof-of-Stake (PoS). Ethereum’s block production involves transaction fees (Base Fee and Priority Fee) and Block Rewards. Ethereum's scalability challenges highlight a paradox: its supply-constrained block space increases transaction fees but contradicts the goal of providing low-latency, affordable compute.
Rollups purchase block space from Ethereum, offering higher scalability by offloading transaction execution off-chain and batching data onto Ethereum. This enables rollups like Arbitrum to scale more efficiently, with transaction throughput uncapped by Ethereum’s constraints. L1 blockchains face fixed costs (e.g., block rewards) and floating income (transaction fees), with constrained scalability and inconsistent margins. L2 rollups, by contrast, experience floating costs and income, offering scalable demand-driven throughput and consistent, higher profit margins.
One of the largest costs for L2s is the expense of storing transaction batches on Ethereum’s L1. Despite these expenses, many L2 projects have begun to demonstrate profitability, thanks in large part of the Dencun upgrade and EIP-4844, which introduced “blobs.” Blobs drastically reduce the cost of posting data to Ethereum, an aspect that was previously costing rollup teams millions per day.
Despite the dramatic cost savings offered by rollups, Ethereum’s mainnet remains integral to the overall structure, even as L2 rollups gain traction. Several activities will likely remain anchored on Ethereum's L1 for the foreseeable future. For example, protocols such as EigenLayer require restaking activities, and governance decisions for key decentralized finance (DeFi) protocols like Aave, Maker, and Uniswap remain firmly tied to the mainnet. Additionally, users with significant security concerns—often those managing large amounts of capital—are more inclined to maintain funds on L1 due to the enhanced security it offers. Despite the advances in scalability and efficiency brought about by L2s, ETH will continue to play a foundational role in these protocols, acting as both the primary unit of account and as necessary collateral for paying "rent" on Ethereum’s L1 infrastructure.
When L2 operators begin to generate substantial profits from user fees (image below), it suggests that Ethereum may be undervaluing itself. In August, L2 operator revenue increased by ~10%, which marks the first instance of rising revenue despite declining operational costs. This development points to the possibility that L2 projects can remain profitable in a sustainable manner, particularly as they benefit from new cost-saving mechanisms like data blobs.
2. What Do The Rollups Get Out of the Ethereum-Rollup Relationship?
This has been explained again and again over the years (including here and here) but the TL;DR is:
- Rollups inherit Ethereum’s security and no longer need to expend money, time, and resources on trying to bootstrap a decentralized community, validator set, and global consensus mechanism. Instead, they can deploy their chain and focus on the other aspects of building a successful blockchain.
- Once the rollup is live, it can easily tap into Ethereum’s industry-leading TVL, liquidity, dApps, users, developer tools, and network effects.
3. What Does Ethereum Get Out of This Relationship?
Retaining Fee-sensitive Users
If Ethereum had not scaled through the introduction of rollups and blobs, some portion of its activity would have shifted to other alt-L1 blockchains rather than to L2 solutions, which would have been a far worse scenario for Ethereum than just a loss in fee revenue. While there's currently a disparity between the revenue generated by rollup sequencers and what Ethereum’s layer 1 collects, this is likely only a short-term issue. Ethereum is in a phase of hyper-scaling, where it has more blockspace available than current demand can fill. However, now that numerous rollups offer sub-$.10 transactions, entirely new use cases, and users can stay within the Ethereum-rollup ecosystem.
High-frequency dApps and Use Cases
The advent of L2 technologies not only facilitates scalability and privacy enhancements but also introduces new application possibilities. By driving down fees and boosting transaction capacity, L2 solutions generate benefits across the board—for app developers, users, and ETH validators and holders. Looking forward, the ecosystem is likely to witness the emergence of numerous L2 blockchains and application-specific platforms. This diversified landscape is expected to feature a mix of large, general-purpose L2s, such as Arbitrum and Optimism, alongside others catering to specialized niches that require micro-transactions, streaming transactions, or simply numerous onchain interactions that would otherwise be too expensive on the mainnet.
Experimentation
Rollups and other Ethereum L2s are less constrained than the L1, in a technical sense but also in a social, economic, and development sense. Because rollups are not the central pillar responsible for security and consensus (that’s Ethereum mainnet) and because they are isolated environments, the stakes (while still high) are not existential to the Ethereum economy. This means they can conduct experiments from which Ethereum can learn, serving as a pseudo “testing ground.” For instance, Gnosis implemented the Dencun hard fork before Ethereum did, allowing the Ethereum Foundation to observe and potentially avoid mistakes when forking the mainnet. Others are experimenting with new virtual machines, parallelizing the EVM, using a UTXO structure rather than accounts, and more. Much of this experimentation could eventually migrate to Ethereum's L1. zkEVMs, in particular, are gaining attention as a potential integration into Ethereum’s main network.
Rollups can also present a new range of perspectives on scaling Ethereum due to their developers, priorities, and users. With the flexibility offered by L2 networks, each group can effectively design its own "state," unlike in the physical world, where land is limited. For instance, platforms like ZORA cater to artists, while Arbitrum focuses on DeFi and XAI on gaming.
4. Solana is the obvious solution
Over the past two years, Solana has emerged as the clear Ethereum alternative, experiencing tremendous growth in onchain metrics as well as offchain (project launches, developers, new tech advances). It differentiates itself with a unique focus on high-speed performance, cost efficiency, and an integrated/monolithic design. The platform's near-instantaneous transactions, ultra-low fees, and Web 2-style speed make the network intuitive and easy to navigate, enhancing its usability. This user-centric design eliminates the need for switching between networks or bridging assets, offering a more streamlined experience that mirrors traditional digital platforms.
Solana’s original vision aimed to create a decentralized marketplace with the efficiency of traditional financial systems, akin to an on-chain Nasdaq. A critical element that sets Solana apart is its integrated architecture, where all applications operate within a shared environment. This setup promotes composability, a feature that is highly valued in the DeFi space. Composability allows developers to interconnect various applications, creating a synergistic effect where protocols can be built upon one another, often referred to as "money legos." Solana’s unified state architecture facilitates this process by enabling all apps to work within the same state, eliminating the need for cross-network interoperability solutions.
One of Solana’s most significant advantages lies in its high throughput, allowing the network to process thousands of transactions per second. This scalability makes it ideal for applications requiring quick transaction times and low latency.
However, the network has not been without its challenges. Despite the inarguable performance enhancements and scalability advantages it offers, it is by no means a panacea. The chain currently handles < 4,000 TPS and has even struggled under this demand load. In early 2024, transactions regularly failed during peak congestion, raising concerns about reliability. Additionally, the chain has a long history of (short-term) downtime and outages. While a new client, Firedancer, has made real progress and looks like it has a legitimate shot at ten or 100x’ing the chains performance, it is still a work in progress and has yet be be released on mainnet. Even with Firedancer, it is a tremendous leap of faith to assume that the chain we’ve all experienced for the last ~five years has any shot of running the entire world’s financial activities atop it.
And that is, largely, the crux of many of the crypto debates: how much credit/goodwill are you willing to extend to one side and their ability to execute on their vision and not the other?
SOL bulls remain confident in the network's potential to eventually scale and handle the entirety of the world's financial activity despite numerous data points (outages) suggesting otherwise while operating at 1/1,000,000th of the needed throughput. At the same time, they often downplay the possibility that Ethereum could find an effective balance between its Layer 1 and rollup solutions. Such wildly imbalance positions reflect bias rather than a grounded assessment based on data or logic.
At present, many impartial observers would agree that Solana has matured into a more robust and decentralized platform than most Ethereum rollups. However, this was not always the case. In its early stages, Solana faced a series of challenges, including a lack of a functional fee market, frequent downtime, issues with initial token distribution, excessive reliance on a single entity (Sam Bankman-Fried and FTX), dependence on a single client, and limited scalability, capped at approximately 10,000 transactions per second. Remarkably, Solana has addressed nearly all of these issues within a three-year period. This raises the question of whether rollups should be afforded the same developmental leeway or be judged based on their current limitations.
It is true that Ethereum rollups currently face issues such as centralized sequencers, fragmented liquidity, and transaction throughput limitations. However, many leading projects have clear plans to mitigate or resolve these challenges. Ultimately, the debate centers on how much trust one is willing to place in these projects' ability to successfully execute their roadmaps and overcome their present obstacles.
Ethereum’s Solana’s Validator Revenue Problem
In the Solana ecosystem, the allocation of inflationary rewards is largely determined by the stake-weight of validators. Validators that accumulate more SOL tokens staked to their nodes experience a greater chance of being selected to produce blocks and, as a result, stand to earn more inflationary rewards than those with lower stake-weight. This relationship between a validator’s stake-weight and its probability of selection operates on a proportional basis. Specifically, a validator’s chance of being selected as the leader for a particular slot is calculated using a straightforward formula:
This linear correlation incentivizes validators to accumulate as much stake as possible since there is no diminishing return or penalty for amassing more SOL. Validators that command higher stakes directly increase their likelihood of being selected, thus allowing them to secure more rewards.
Jito has introduced a new level of sophistication to the Solana validator ecosystem by integrating a Maximum Extractable Value (MEV) marketplace directly into the validator client’s codebase. This customized validator client, a fork of Solana Labs’ official client, allows validators and their delegators to participate in MEV rewards that would otherwise be inaccessible. Jito enables validators to capture these rewards and pass them on to stakers, offering an additional 10-15% in SOL rewards.
The mechanism behind Jito's MEV marketplace operates through competition among MEV searchers. These participants identify profitable opportunities and submit a bundle of transactions to the Jito MEV marketplace, bidding for the chance to have their transactions included in the next block. The slot leader (validator) selects the highest bidder, and the associated MEV tip is distributed among the validator and its delegators. This process adds a layer of complexity to staking but offers significant benefits by increasing overall returns for stakers who delegate to validators running the Jito client.
Solana validators earn rewards not only through inflationary mechanisms but also from block rewards. These block rewards consist of two main components: base fees and priority fees. When a validator successfully proposes a block, they collect a base fee of 0.000005 SOL per signature. Priority fees, on the other hand, vary based on transaction demand and are determined by the sender of the transaction. Currently, Solana burns 50% of priority fees, base fees, and vote fees.
However, the revenue model for validators in 2024 has become increasingly constrained. A Dune dashboard reveals that ~90% of Solana validator revenue comes from SOL issuance, which is designed to only decrease over time. Fees and Maximal Extractable Value (MEV)—the additional revenue validators can earn from reordering transactions in a block—constitute the remaining ~10% of their income.
Further exacerbating the problem for smaller validators are the recent changes by the Solana Foundation’s delegation program. The Solana Foundation delegation program is designed to bolster small validators by covering some operational costs and staking SOL tokens with them to ensure the network remains decentralized. According to a report from Helius, approximately 72% of Solana validators benefit from this initiative. However, this financial support does not fully alleviate the burden on these validators. Their costs, many of which are fixed, increase the pressure as revenue depends largely on the stake they control and the value of SOL, both of which are in decline.
Profitability got even harder in 2024 due to the fact that the Solana Foundation recently reduced the maximum commission that validators participating in its delegation program can charge. The previous limit of 7% has been lowered to 5%, and a new cap of 10% has been imposed on Jito MEV, a form of validator income. Under prior conditions, validators could claim 100% of Jito MEV, but that potential has now been curbed.
For smaller validators reliant on the Foundation’s delegation program to stay operational, this policy change directly impacts their ability to generate sufficient revenue, as they can no longer charge as much for their services.
The financial implications of these changes are profound. One small validator estimated that, without Foundation support, a validator would need to control around 65,000 SOL to break even—up from 40,000 SOL previously. The relative profitability of recent months for small validators appears to be unsustainable, and the current trajectory suggests that continued reliance on the Solana Foundation's delegation program may not be viable in the long term. While the program has played a key role in decentralizing Solana’s validator set, it is unlikely to provide indefinite support. The recent reduction in allowed commissions is a signal that small validators must find ways to stand on their own. This shift presents a critical juncture for Solana’s network, as small validators must adapt or risk being left behind. Decentralization is a key strength of the network, but it remains to be seen whether Solana can sustain its decentralized nature in the face of such financial pressures.
However, with the upcoming protocol upgrade Agave 2.0 100% of the priority fees will be allocated to the block proposer. This shift in protocol design aims to further incentivize validators by ensuring they capture all priority fees, rather than burning a portion. The change is expected to be implemented in late 2024. However, it still does not address the centralizing, economies-of-scale issue at the heart of the Solana validator ecosystem. Larger validators will still be more profitable than smaller validators and over time, will consolidate down to just a select few.
Solana Extensions
Despite Solana’s current industry-leading (actual) TPS and promises of 1 million TPS in the near future, thanks to the Firedancer client, the Solana ecosystem is experiencing the beginnings of its own “L2” revolution. These new projects are known as Solana “Extensions” (L2s), and 5-10 are being developed/released in 2024.
*Side note*: Notice the careful language used: Extensions. Kudos to Solana for getting in front of the exact same future debate. No one is going to pose the question, “Are Solana Extensions an extension of Solana?” Just look at the name! Debate over.
While there are very real and distinct technical differences between Solana Extensions and a traditional Ethereum rollup, the fact that they are emerging may suggest that no matter your base layer scalability, there are valid reasons for “breaking off” from the primary L1. A promising alternative that maintains Solana’s composability while improving scalability is Magic Block’s concept of ephemeral rollups. These rollups offer a solution that allows applications to scale specific operations without disrupting the composability of the base layer. By leveraging Solana’s SVM account structure and parallelization capabilities, developers can temporarily lock one or more accounts and shift their state to an ephemeral rollup. During this period, the sequencer can modify these accounts within the rollup, while other non-delegated accounts continue to interact with programs on the base layer.
The benefits of ephemeral rollups are multifaceted. First, they preserve the composability of the base layer, allowing programs to interact with existing protocols and assets without interruption. This ensures that the foundational interconnectivity that distinguishes Solana from other blockchains remains intact. Developers also gain the advantage of using existing Solana infrastructure, such as the Anchor Framework, RPCs, and libraries, which reduces the friction of adopting new scalability solutions.
Should these new Extensions garner meaningful adoption, questions will naturally arise. Like is this an invalidation of the integrated/monolithic narrative? If these Extensions grow to be large enough, will they eventually spin off into their own chain? These questions are part of the natural progression of a successful L1. Much like the Ethereum debates around fees, MEV, governance, etc. that have already taken place, any chain that becomes significant enough with meaningful adoption and usage will have tough questions to answer across various entities with different priorities.
Conclusion
In conclusion, the rise of Layer 2 solutions, while presenting significant benefits in scalability and cost reduction, has sparked heated debates about their impact on Ethereum’s economic model. Despite concerns that L2s might be parasitic, extracting value from Ethereum without proportionally benefiting the network, the reality is more nuanced. Rollups, as extensions of Ethereum, rely on the mainnet for security and settlement, ensuring that Ethereum remains integral to the broader ecosystem. While fee revenue on Layer 1 has declined, Ethereum’s scalability strategy through rollups has helped retain users who might otherwise migrate to alternative L1 blockchains.
The reduction in fees and the shift in activity to L2s were intentional and reflected Ethereum’s vision of a modular, scalable blockchain. As Ethereum continues to evolve, balancing utility, security, and its role as a store of value will be key to its long-term success. Furthermore, the emergence of new competitors, such as Solana, highlights that Ethereum’s challenges are not unique, and innovation across the blockchain space is inevitable. Ultimately, while the relationship between Ethereum and L2s may seem contentious at times, it is this dynamic interplay that will likely shape the future of decentralized technology.
Disclaimer: This research report is exactly that — a research report. It is not intended to serve as financial advice, nor should you blindly assume that any of the information is accurate without confirming through your own research. Bitcoin, cryptocurrencies, and other digital assets are incredibly risky and nothing in this report should be considered an endorsement to buy or sell any asset. Never invest more than you are willing to lose and understand the risk that you are taking. Do your own research. All information in this report is for educational purposes only and should not be the basis for any investment decisions that you make.
Every blockchain, including Ethereum, has inherent limitations on how much data and how many transactions it can process. While Ethereum has been a pioneering platform for decentralized applications, its limitations, particularly around transaction speeds and costs, have led to the rise of L2 solutions and rollups.
However, the perception of L2s as "parasitic" entities has emerged due to the belief that they siphon value away from Ethereum by handling the bulk of user activity on their networks, leaving Ethereum with fewer transactions and lower fee revenue. The fear is that Ethereum’s economic value could be diminished as L2s, which operate as independent blockchains, capture most of the user base and transaction volume.
The economic relationship between Ethereum and its L2s is complex. L2s often rely on Ethereum for security, periodically settling their transactions on the Ethereum mainnet. However, Ethereum collects only a small fraction of the fees generated by L2 activity, leading to reduced revenue for the main chain. The fear is that as more users and developers migrate to L2s, Ethereum will struggle to justify its current valuation.
The migration to L2 has led to a significant decrease in transaction fee revenue generated on L1. This reduction is a double-edged sword. On one hand, it demonstrates the effectiveness of L2s in reducing costs for users, which is positive for user adoption and network growth. On the other hand, it raises concerns about the long-term economic sustainability of the L1 network, which has traditionally relied on transaction fees as a significant revenue stream.
This report dives into the numerous debates around the central question, “Are rollups parasitic to Ethereum?”
Things We Can All (Hopefully) Agree On
1. The “Are L2s parasitic to Ethereum?” narrative only emerged because of sluggish ETH price action.
If ETH was ~$4000+, this narrative/debate would almost certainly not exist. During downtimes and lulls in price action, people become frustrated and/or apathetic and need something to latch onto (example below). Add in the fact that the SOL price surged over the last 1.5 years, and a natural SOL vs ETH debate was bound to arise. Only this time, the proliferation of rollups are the scapegoat.
Note the date, which was two years ago, during the depths of the FTX bear market. While the Polygon PoS chain is completely architecturally different from Ethereum rollups, the point is that these debates are not new, and neither is the timing.
2. It is inarguable that ETH's price has lagged behind BTC and SOL for much of the last ~ two years.
However, the performance is HIGHLY dependent on where you start the chart. Starting the comparison at the market highs in November 2021 vs. the market lows in 2022 produces a very different picture/narrative. Regardless, we are left to figure out if this recent price underperformance is BECAUSE of the rise of L2s or merely coincidental, i.e., a convenient narrative.
Comparing BTC, ETH, and SOL price performance from Nov 2021 tops paints less of an ETH emergency narrative.
3 An endless number of things drive price action in crypto, some innate to the protocol and many, many more exogenous to the protocol.
Often the least of which is anything resembling “fundamentals” like users, fees, revenue, developers, dApp usage, etc. One thing is for sure: protocol fees are NOT one of the more significant aspects driving price appreciation. Every L1 in existence (outside of maybe Ethereum) had negligible fees prior to 2023. However, that has not stopped the market cap of many protocols from ballooning to comical valuation. Celestia was the poster child for this earlier in 2024 (image below). It was “earning” ~$200 a day in revenue, yet the protocol was valued at ~$18 billion. Therefore, attributing price movements to fees/revenue is, at the absolute best, a fraction of the overall story.
While pessimism surrounding Ethereum may arise from concerns about questionable scaling solutions or competition from other blockchain ecosystems, many of these challenges are not exclusive to Ethereum. As the space evolves, competition is/was inevitable. Let’s look to Bitcoin as an example. BTC was the first cryptocurrency; because of this, it was ~100% of the crypto market for years. However, over time, new, interesting protocols were built, and with each subsequent crypto bull/bear cycle, its market cap dominance has been diminished.
Ethereum is in a similar position within the smart contract blockchain market. It was the first and, therefore, enjoyed the lion’s share of the market for years. However, as the concept of smart contracts was proven out and accepted as a legitimate use case, new chains with varying tradeoffs and advancements have been created. It is irrational to believe that one chain, be it Bitcoin or Ethereum, was never going to cede market share to competition.
Right or wrongly, ETH’s underperformance has been largely attributed to a decline in Ethereum’s total transaction fees and activity, especially following the Dencun upgrade in March 2024 (despite the fact that SOL had been outperforming ETH for months at this point). The upgrade introduced binary large object (blob) storage, which shifted much of the transaction activity to Ethereum's rollups.
However, some believe that the migration to L2s is not the sole cause of ETH's underperformance. While only base and blob fees are burned according to EIP-1559 (and L2-generated fees have been minimal), even burning all L2 fees wouldn’t prevent ETH from becoming inflationary. This suggests that factors beyond L2 growth are at play. Additionally, while on-chain activity on Ethereum's layer-1 hit a four-year low by August 2024, it has since rebounded over the course of September 2024.
One example of out-of-protocol factors that may have contributed to ETH's recent price weakness is the unwinding of the Japanese yen (JPY) carry trade in early August. While this event triggered deleveraging across multiple asset classes, ETH may have been particularly vulnerable due to another exogenous factor, disappointing ETF flows. The launch of spot ETH ETFs failed to generate strong inflows, with a net outflow of $57M in the first two weeks, in contrast to a small net inflow of $69M for U.S.-based spot BTC ETFs during the same period.
4. If Ethereum did not have the rollup roadmap and was still stuck trying to scale the L1 with $50+ fees, sentiment would be a whole lot worse.
Ethereum’s shift from a single-layer (monolithic) blockchain attempting to solve scalability issues to a more modular structure, where Layer 1 functions as a settlement layer and Layer 2s take on additional responsibilities (execution), has been technically smooth but less successful in the court of public opinion. The issue is that this scaling strategy was always going to create profound shifts in its user activity and economic dynamics for Ethereum. That was by design because, literally, the only thing crypto Twitter could all agree on for years was that Ethereum fees were too high.
These rollup developments are a deliberate part of Ethereum's strategy to address scalability issues that have long plagued its network. By offloading a substantial portion of transaction processing to L2 solutions, Ethereum aims to birth countless new L2 environments while remaining the core, central piece that enables it all to work. This transition, however, has sparked intense discussions regarding its economic impacts and the future role of ETH in this design.
5. Overall activity (L1 + L2 transactions) is way up, and fees are way down. This was the goal.
Fees/revenue were always supposed to go down. Like way down. Whether or not you like how rollups are currently constructed, they have inarguably accomplished what they set out to do: reduce the burden on the L1, offload execution, and provide a means of scaling for Ethereum.
Since the Denun upgrade in March 2024, which significantly reduced the cost of rollup transactions, the combined Ethereum+rollup daily transaction count has roughly tripled while L2 TVL has been near all-time highs.
This illustrates that demand was there, but users were cost-sensitive. If rollups did not exist or remained cost-prohibitive, where would these users have gone? Would they have moved to an alt-L1 like Solana or BSC? It’s obviously impossible to know, but it's almost inarguable that having those users and that activity on a rollup is better than moving to an alt-L1 where there is no connection to Ethereum.
6. Let’s acknowledge the paradox we have constructed for Ethereum. High fees are bad, but so are low fees.
This contradictory thinking is, IMO, just another offshoot of topic #1. People are mad that the ETH price is lagging and thus need something to blame it on. It was high fees during the previous dull periods, and now it is too low fees. Again, fees are not the primary driver of token price—they never have been. But when a token isn’t mooning, people will find anything to blame.
Obviously, there are extremes on both ends that are not productive for a blockchain. If the fees are too high, you exclude certain people and use cases. If the fees are too low, the chain becomes insecure. But Ethereum is nowhere close to succumbing to either of these, especially compared to its competitors.
7. Ethereum L1 is not making as much revenue as before Dencun, but it is still competitive with the top L1s in terms of revenue while also experiencing far less inflation.
While Ethereum’s lead in fees is not as large as it once was, even with the addition of 100+ rollups over the last ~ three years, Ethereum mainnet is still generating 2x the fees of the second-place chain, Tron.
Meanwhile, Ethereum’s inflation rate is ~0.5%, better than nearly all alt-L1 competitors. This means that despite the recent histrionics around Ethereum, it still makes more revenue and issues less supply than its competitors. This leads to the question: Why all the worry about Ethereum and not for all the other L1 darlings like Solana, Aptos, and others? It is valid to critique the Ethereum-rollup model, but it is not like there is some proven, bulletproof alternative model that we know works today.
For reference, there are still questions about Bitcoin’s revenue/fee model. Additionally, Solana issues ~$6.35 M worth of SOL daily while generating less than $1.5 M in fees. This would make it far less economically sustainable than Ethereum, yet there is little concern for Solana (if you want to know why, just check the SOL price chart since 2023).
8. NONE of the chains today are valued on anything resembling traditional cash flows, revenue, DCF models, or business-like valuations. So, why are we sweating Ethereum’s revenue so much?
The “intrinsic” value of far more mature and socially accepted assets like BTC and even gold cannot be explained by any traditional financial model. Their market capitalizations seem disproportionate when viewed through traditional financial lenses such as earnings, productivity, or utility. Instead, these assets derive value primarily from their status as stores of value (SoV) and their perceived monetary premium. This concept prompts a broader question: what if the L1 blockchain that ultimately prevails is not the one that maximizes extraction of miner extractable value (MEV) but rather one that balances utility with its role as a store of value?
Unlike conventional businesses, which generate value through earnings, L1 blockchains will increasingly see their fees and MEV trend toward zero as network efficiency improves. Relying on cash flow measures to value these networks would, therefore, lead to a valuation of zero—an outcome that disregards their broader role in the ecosystem. Therefore, there must be other factors at play. Obvious qualities (at least to me) that command some sort of premium but are hard to model or quantify are SoV, “moneyness,” decentralization and censorship-resistance, brand and reputation, acceptance from TradFi and governments, and others.
Things We’ll Probably Argue About
So, we’ve established that Ethereum L1 fees are down dramatically, but this has always been the plan. $100 L1 transaction fees without an alternative solution for users was never going to enable Ethereum’s vision of a global, neutral, permissionless economic settlement layer for all. We’ve also established that a user moving from the Ethereum mainnet to a rollup is better for Ethereum than losing that user and activity to an alt-L1. Rollups are a significant factor driving demand for ETH. Over 34 million ETH has already been transferred into various L2 ecosystems, effectively creating another liquidity drain for ETH. While the ETH moved to these L2s may not be subject to direct burning, the need for users to hold native tokens to cover transaction fees contributes to an indirect lock-up of ETH. This growing reserve of ETH, held by new wallets to manage fees, represents a soft lock on a substantial and increasing portion of the overall token supply.
Therefore, the logic looks something like this:
1. Are L2s Symbiotic with the Ethereum L1 or Parasitic?
First, a definition!
Parasite: an organism that lives in or on an organism of another species (its host) and benefits by deriving nutrients at the other's expense.
Ok, but we’re not dealing with animals…
A parasitic business relationship is one where one party gains at the expense of another, and is often characterized by an imbalance of power. An example of a parasitic business relationship would be a company that only extracts profit, and doesn't create/preserve more value than it takes.
While blockchains are not 1-to-1 analogous to companies with traditional revenue metrics (as discussed in previous sections), let’s use this framework to dissect the mainnet-rollup relationship and determine if they are, in fact, parasitic (e.g., they extract more value than they create/preserve.)
Before attempting to determine whether or not rollups are parasitic to the L1, we must first address another common debate in this narrative and that is “Are rollups even an extension of Ethereum?”
In essence, rollups—whether optimistic or zk—quite literally are extensions of Ethereum. They are smart contracts on the Ethereum mainnet. They also hold significant amounts of ETH, submit data and proofs back to the Ethereum network, and use ETH for gas fees. This is precisely the difference between a rollup and an alt-L1. There is no connection between Ethereum and an alt-L1 in terms of native token, security, consensus, fees, etc. There are for Ethereum and rollups.
An analogy often used to explain the L1-L2 relationship is that of states within a federal system. Just as Texas is part of the United States but retains its own governance and tax structures, L2s operate under Ethereum’s security and infrastructure while maintaining some level of independence. However, this relationship is not unconditional. For example, if an L2 decides to retain the EVM but uses its own gas token and settles transactions on another blockchain, it would no longer be considered a part of Ethereum.
The L1-L2 Economic Relationship
The relationship between Ethereum and rollups revolves around the economics of gas fees. As a quick refresher, Ethereum's native token, ETH, is critical to the network. As the principal medium for paying transaction fees, ETH is essential for any interaction on the network, including transfers of stablecoins, decentralized exchange (DEX) transactions, non-fungible token (NFT) minting, and on-chain gaming. This foundational utility drives persistent demand for ETH, much like the demand for oil in global markets.
In contrast to ETH, Layer 2 (L2) tokens primarily serve governance functions. While some L2 networks permit the use of their tokens for gas fees, the majority of these fees are still paid in ETH. L2s also need to pay for block space on Ethereum using ETH, indicating that L2 tokens lack an intrinsic utility that can drive demand. Rollups, which offer scalability solutions to Ethereum, generate their revenue primarily through user transaction fees but must also contend with certain costs.
The cost of producing block space is a key consideration in blockchain economics, with distinct differences between Layer-1 (L1) and Layer-2 (L2) systems. L1 Blockchains, like Ethereum, rely on token incentives to maintain Sybil resistance and achieve consensus through Proof-of-Stake (PoS). Ethereum’s block production involves transaction fees (Base Fee and Priority Fee) and Block Rewards. Ethereum's scalability challenges highlight a paradox: its supply-constrained block space increases transaction fees but contradicts the goal of providing low-latency, affordable compute.
Rollups purchase block space from Ethereum, offering higher scalability by offloading transaction execution off-chain and batching data onto Ethereum. This enables rollups like Arbitrum to scale more efficiently, with transaction throughput uncapped by Ethereum’s constraints. L1 blockchains face fixed costs (e.g., block rewards) and floating income (transaction fees), with constrained scalability and inconsistent margins. L2 rollups, by contrast, experience floating costs and income, offering scalable demand-driven throughput and consistent, higher profit margins.
One of the largest costs for L2s is the expense of storing transaction batches on Ethereum’s L1. Despite these expenses, many L2 projects have begun to demonstrate profitability, thanks in large part of the Dencun upgrade and EIP-4844, which introduced “blobs.” Blobs drastically reduce the cost of posting data to Ethereum, an aspect that was previously costing rollup teams millions per day.
Despite the dramatic cost savings offered by rollups, Ethereum’s mainnet remains integral to the overall structure, even as L2 rollups gain traction. Several activities will likely remain anchored on Ethereum's L1 for the foreseeable future. For example, protocols such as EigenLayer require restaking activities, and governance decisions for key decentralized finance (DeFi) protocols like Aave, Maker, and Uniswap remain firmly tied to the mainnet. Additionally, users with significant security concerns—often those managing large amounts of capital—are more inclined to maintain funds on L1 due to the enhanced security it offers. Despite the advances in scalability and efficiency brought about by L2s, ETH will continue to play a foundational role in these protocols, acting as both the primary unit of account and as necessary collateral for paying "rent" on Ethereum’s L1 infrastructure.
When L2 operators begin to generate substantial profits from user fees (image below), it suggests that Ethereum may be undervaluing itself. In August, L2 operator revenue increased by ~10%, which marks the first instance of rising revenue despite declining operational costs. This development points to the possibility that L2 projects can remain profitable in a sustainable manner, particularly as they benefit from new cost-saving mechanisms like data blobs.
2. What Do The Rollups Get Out of the Ethereum-Rollup Relationship?
This has been explained again and again over the years (including here and here) but the TL;DR is:
- Rollups inherit Ethereum’s security and no longer need to expend money, time, and resources on trying to bootstrap a decentralized community, validator set, and global consensus mechanism. Instead, they can deploy their chain and focus on the other aspects of building a successful blockchain.
- Once the rollup is live, it can easily tap into Ethereum’s industry-leading TVL, liquidity, dApps, users, developer tools, and network effects.
3. What Does Ethereum Get Out of This Relationship?
Retaining Fee-sensitive Users
If Ethereum had not scaled through the introduction of rollups and blobs, some portion of its activity would have shifted to other alt-L1 blockchains rather than to L2 solutions, which would have been a far worse scenario for Ethereum than just a loss in fee revenue. While there's currently a disparity between the revenue generated by rollup sequencers and what Ethereum’s layer 1 collects, this is likely only a short-term issue. Ethereum is in a phase of hyper-scaling, where it has more blockspace available than current demand can fill. However, now that numerous rollups offer sub-$.10 transactions, entirely new use cases, and users can stay within the Ethereum-rollup ecosystem.
High-frequency dApps and Use Cases
The advent of L2 technologies not only facilitates scalability and privacy enhancements but also introduces new application possibilities. By driving down fees and boosting transaction capacity, L2 solutions generate benefits across the board—for app developers, users, and ETH validators and holders. Looking forward, the ecosystem is likely to witness the emergence of numerous L2 blockchains and application-specific platforms. This diversified landscape is expected to feature a mix of large, general-purpose L2s, such as Arbitrum and Optimism, alongside others catering to specialized niches that require micro-transactions, streaming transactions, or simply numerous onchain interactions that would otherwise be too expensive on the mainnet.
Experimentation
Rollups and other Ethereum L2s are less constrained than the L1, in a technical sense but also in a social, economic, and development sense. Because rollups are not the central pillar responsible for security and consensus (that’s Ethereum mainnet) and because they are isolated environments, the stakes (while still high) are not existential to the Ethereum economy. This means they can conduct experiments from which Ethereum can learn, serving as a pseudo “testing ground.” For instance, Gnosis implemented the Dencun hard fork before Ethereum did, allowing the Ethereum Foundation to observe and potentially avoid mistakes when forking the mainnet. Others are experimenting with new virtual machines, parallelizing the EVM, using a UTXO structure rather than accounts, and more. Much of this experimentation could eventually migrate to Ethereum's L1. zkEVMs, in particular, are gaining attention as a potential integration into Ethereum’s main network.
Rollups can also present a new range of perspectives on scaling Ethereum due to their developers, priorities, and users. With the flexibility offered by L2 networks, each group can effectively design its own "state," unlike in the physical world, where land is limited. For instance, platforms like ZORA cater to artists, while Arbitrum focuses on DeFi and XAI on gaming.
4. Solana is the obvious solution
Over the past two years, Solana has emerged as the clear Ethereum alternative, experiencing tremendous growth in onchain metrics as well as offchain (project launches, developers, new tech advances). It differentiates itself with a unique focus on high-speed performance, cost efficiency, and an integrated/monolithic design. The platform's near-instantaneous transactions, ultra-low fees, and Web 2-style speed make the network intuitive and easy to navigate, enhancing its usability. This user-centric design eliminates the need for switching between networks or bridging assets, offering a more streamlined experience that mirrors traditional digital platforms.
Solana’s original vision aimed to create a decentralized marketplace with the efficiency of traditional financial systems, akin to an on-chain Nasdaq. A critical element that sets Solana apart is its integrated architecture, where all applications operate within a shared environment. This setup promotes composability, a feature that is highly valued in the DeFi space. Composability allows developers to interconnect various applications, creating a synergistic effect where protocols can be built upon one another, often referred to as "money legos." Solana’s unified state architecture facilitates this process by enabling all apps to work within the same state, eliminating the need for cross-network interoperability solutions.
One of Solana’s most significant advantages lies in its high throughput, allowing the network to process thousands of transactions per second. This scalability makes it ideal for applications requiring quick transaction times and low latency.
However, the network has not been without its challenges. Despite the inarguable performance enhancements and scalability advantages it offers, it is by no means a panacea. The chain currently handles < 4,000 TPS and has even struggled under this demand load. In early 2024, transactions regularly failed during peak congestion, raising concerns about reliability. Additionally, the chain has a long history of (short-term) downtime and outages. While a new client, Firedancer, has made real progress and looks like it has a legitimate shot at ten or 100x’ing the chains performance, it is still a work in progress and has yet be be released on mainnet. Even with Firedancer, it is a tremendous leap of faith to assume that the chain we’ve all experienced for the last ~five years has any shot of running the entire world’s financial activities atop it.
And that is, largely, the crux of many of the crypto debates: how much credit/goodwill are you willing to extend to one side and their ability to execute on their vision and not the other?
SOL bulls remain confident in the network's potential to eventually scale and handle the entirety of the world's financial activity despite numerous data points (outages) suggesting otherwise while operating at 1/1,000,000th of the needed throughput. At the same time, they often downplay the possibility that Ethereum could find an effective balance between its Layer 1 and rollup solutions. Such wildly imbalance positions reflect bias rather than a grounded assessment based on data or logic.
At present, many impartial observers would agree that Solana has matured into a more robust and decentralized platform than most Ethereum rollups. However, this was not always the case. In its early stages, Solana faced a series of challenges, including a lack of a functional fee market, frequent downtime, issues with initial token distribution, excessive reliance on a single entity (Sam Bankman-Fried and FTX), dependence on a single client, and limited scalability, capped at approximately 10,000 transactions per second. Remarkably, Solana has addressed nearly all of these issues within a three-year period. This raises the question of whether rollups should be afforded the same developmental leeway or be judged based on their current limitations.
It is true that Ethereum rollups currently face issues such as centralized sequencers, fragmented liquidity, and transaction throughput limitations. However, many leading projects have clear plans to mitigate or resolve these challenges. Ultimately, the debate centers on how much trust one is willing to place in these projects' ability to successfully execute their roadmaps and overcome their present obstacles.
Ethereum’s Solana’s Validator Revenue Problem
In the Solana ecosystem, the allocation of inflationary rewards is largely determined by the stake-weight of validators. Validators that accumulate more SOL tokens staked to their nodes experience a greater chance of being selected to produce blocks and, as a result, stand to earn more inflationary rewards than those with lower stake-weight. This relationship between a validator’s stake-weight and its probability of selection operates on a proportional basis. Specifically, a validator’s chance of being selected as the leader for a particular slot is calculated using a straightforward formula:
This linear correlation incentivizes validators to accumulate as much stake as possible since there is no diminishing return or penalty for amassing more SOL. Validators that command higher stakes directly increase their likelihood of being selected, thus allowing them to secure more rewards.
Jito has introduced a new level of sophistication to the Solana validator ecosystem by integrating a Maximum Extractable Value (MEV) marketplace directly into the validator client’s codebase. This customized validator client, a fork of Solana Labs’ official client, allows validators and their delegators to participate in MEV rewards that would otherwise be inaccessible. Jito enables validators to capture these rewards and pass them on to stakers, offering an additional 10-15% in SOL rewards.
The mechanism behind Jito's MEV marketplace operates through competition among MEV searchers. These participants identify profitable opportunities and submit a bundle of transactions to the Jito MEV marketplace, bidding for the chance to have their transactions included in the next block. The slot leader (validator) selects the highest bidder, and the associated MEV tip is distributed among the validator and its delegators. This process adds a layer of complexity to staking but offers significant benefits by increasing overall returns for stakers who delegate to validators running the Jito client.
Solana validators earn rewards not only through inflationary mechanisms but also from block rewards. These block rewards consist of two main components: base fees and priority fees. When a validator successfully proposes a block, they collect a base fee of 0.000005 SOL per signature. Priority fees, on the other hand, vary based on transaction demand and are determined by the sender of the transaction. Currently, Solana burns 50% of priority fees, base fees, and vote fees.
However, the revenue model for validators in 2024 has become increasingly constrained. A Dune dashboard reveals that ~90% of Solana validator revenue comes from SOL issuance, which is designed to only decrease over time. Fees and Maximal Extractable Value (MEV)—the additional revenue validators can earn from reordering transactions in a block—constitute the remaining ~10% of their income.
Further exacerbating the problem for smaller validators are the recent changes by the Solana Foundation’s delegation program. The Solana Foundation delegation program is designed to bolster small validators by covering some operational costs and staking SOL tokens with them to ensure the network remains decentralized. According to a report from Helius, approximately 72% of Solana validators benefit from this initiative. However, this financial support does not fully alleviate the burden on these validators. Their costs, many of which are fixed, increase the pressure as revenue depends largely on the stake they control and the value of SOL, both of which are in decline.
Profitability got even harder in 2024 due to the fact that the Solana Foundation recently reduced the maximum commission that validators participating in its delegation program can charge. The previous limit of 7% has been lowered to 5%, and a new cap of 10% has been imposed on Jito MEV, a form of validator income. Under prior conditions, validators could claim 100% of Jito MEV, but that potential has now been curbed.
For smaller validators reliant on the Foundation’s delegation program to stay operational, this policy change directly impacts their ability to generate sufficient revenue, as they can no longer charge as much for their services.
The financial implications of these changes are profound. One small validator estimated that, without Foundation support, a validator would need to control around 65,000 SOL to break even—up from 40,000 SOL previously. The relative profitability of recent months for small validators appears to be unsustainable, and the current trajectory suggests that continued reliance on the Solana Foundation's delegation program may not be viable in the long term. While the program has played a key role in decentralizing Solana’s validator set, it is unlikely to provide indefinite support. The recent reduction in allowed commissions is a signal that small validators must find ways to stand on their own. This shift presents a critical juncture for Solana’s network, as small validators must adapt or risk being left behind. Decentralization is a key strength of the network, but it remains to be seen whether Solana can sustain its decentralized nature in the face of such financial pressures.
However, with the upcoming protocol upgrade Agave 2.0 100% of the priority fees will be allocated to the block proposer. This shift in protocol design aims to further incentivize validators by ensuring they capture all priority fees, rather than burning a portion. The change is expected to be implemented in late 2024. However, it still does not address the centralizing, economies-of-scale issue at the heart of the Solana validator ecosystem. Larger validators will still be more profitable than smaller validators and over time, will consolidate down to just a select few.
Solana Extensions
Despite Solana’s current industry-leading (actual) TPS and promises of 1 million TPS in the near future, thanks to the Firedancer client, the Solana ecosystem is experiencing the beginnings of its own “L2” revolution. These new projects are known as Solana “Extensions” (L2s), and 5-10 are being developed/released in 2024.
*Side note*: Notice the careful language used: Extensions. Kudos to Solana for getting in front of the exact same future debate. No one is going to pose the question, “Are Solana Extensions an extension of Solana?” Just look at the name! Debate over.
While there are very real and distinct technical differences between Solana Extensions and a traditional Ethereum rollup, the fact that they are emerging may suggest that no matter your base layer scalability, there are valid reasons for “breaking off” from the primary L1. A promising alternative that maintains Solana’s composability while improving scalability is Magic Block’s concept of ephemeral rollups. These rollups offer a solution that allows applications to scale specific operations without disrupting the composability of the base layer. By leveraging Solana’s SVM account structure and parallelization capabilities, developers can temporarily lock one or more accounts and shift their state to an ephemeral rollup. During this period, the sequencer can modify these accounts within the rollup, while other non-delegated accounts continue to interact with programs on the base layer.
The benefits of ephemeral rollups are multifaceted. First, they preserve the composability of the base layer, allowing programs to interact with existing protocols and assets without interruption. This ensures that the foundational interconnectivity that distinguishes Solana from other blockchains remains intact. Developers also gain the advantage of using existing Solana infrastructure, such as the Anchor Framework, RPCs, and libraries, which reduces the friction of adopting new scalability solutions.
Should these new Extensions garner meaningful adoption, questions will naturally arise. Like is this an invalidation of the integrated/monolithic narrative? If these Extensions grow to be large enough, will they eventually spin off into their own chain? These questions are part of the natural progression of a successful L1. Much like the Ethereum debates around fees, MEV, governance, etc. that have already taken place, any chain that becomes significant enough with meaningful adoption and usage will have tough questions to answer across various entities with different priorities.
Conclusion
In conclusion, the rise of Layer 2 solutions, while presenting significant benefits in scalability and cost reduction, has sparked heated debates about their impact on Ethereum’s economic model. Despite concerns that L2s might be parasitic, extracting value from Ethereum without proportionally benefiting the network, the reality is more nuanced. Rollups, as extensions of Ethereum, rely on the mainnet for security and settlement, ensuring that Ethereum remains integral to the broader ecosystem. While fee revenue on Layer 1 has declined, Ethereum’s scalability strategy through rollups has helped retain users who might otherwise migrate to alternative L1 blockchains.
The reduction in fees and the shift in activity to L2s were intentional and reflected Ethereum’s vision of a modular, scalable blockchain. As Ethereum continues to evolve, balancing utility, security, and its role as a store of value will be key to its long-term success. Furthermore, the emergence of new competitors, such as Solana, highlights that Ethereum’s challenges are not unique, and innovation across the blockchain space is inevitable. Ultimately, while the relationship between Ethereum and L2s may seem contentious at times, it is this dynamic interplay that will likely shape the future of decentralized technology.
Disclaimer: This research report is exactly that — a research report. It is not intended to serve as financial advice, nor should you blindly assume that any of the information is accurate without confirming through your own research. Bitcoin, cryptocurrencies, and other digital assets are incredibly risky and nothing in this report should be considered an endorsement to buy or sell any asset. Never invest more than you are willing to lose and understand the risk that you are taking. Do your own research. All information in this report is for educational purposes only and should not be the basis for any investment decisions that you make.
