原文：《Are L2s Complementary to ETH?》by Michael Nadeau
Compiled by: 0x11, Foresight News
This crypto winter, the growth of Ethereum scaling solutions is one of the few stories that the market continues to pay attention to. Transaction fees are falling, applications are migrating to L2, TVL is growing; at the same time, the user experience is getting better.
L2 has reached its peak, but how does it affect ETH on L1? Should ETH holders and validators consider L2 as a complement to Ethereum? Will value accumulate across the technology stack? What does this mean for other L1 public chains?
This week we have a data-driven update covering the economics of the Web3 technology stack as it relates to L2:
L2 Business Model and Recent Growth How L2 Impacts ETH Economics Investors’ Mental Models/Frameworks Do We Still Need L1 Competing Chains?
L2 business model and recent growth
L2 attempts to solve the scalability issues of Ethereum L1. They make money by compressing data and “reselling” Ethereum’s block space. L1 is slow and expensive, with limited block space. Therefore, we should consider Ethereum as the ultimate clearing and settlement layer of the technology stack. The base layer records the final state of everything that happens on the execution and application layers above it - serving as the single source of truth.
Execution services that enhance user experience for applications (lower costs, faster transactions) will be handed over to Layer 2. These services are executed on another chain outside of Ethereum and the data is then logged into L1 Below is a simplified view of the technology stack.
Most computing resources (Gas) are currently consumed in Ethereum L1, which is where applications were originally built. Keep in mind that the Web3 stack has no centralized party coordinating development activities (neither does the Internet), and everything is open source. Builders rush in, but not everyone is on the same timeline or working from the same agenda. This is not the well-coordinated agile sprint you see in well-run tech companies. L1 infrastructure and applications are under construction until scaling solutions are rolled out. As L2 explodes, we expect to see more application migrations, and new projects may be launched directly on L2 in the future.
The data exposes this trend, so here’s a quick look at total computing resource (Gas) usage on Ethereum, and how the top L2 consumes resources now.
Data source: Arcana Analytics
Arbitrum and Optimism now take up more block space than Polygon. Overall, these three L2s account for about 4.5% of the current Ethereum block space, up from about 3% at the end of 2022.
Next, let’s take a look at the other data in L2:
Note: The average single transaction fee for Ethereum L1 so far in 2023 is $4.84.
As mentioned before, L2 serves applications by providing a better and cheaper experience for users. They purchase block space on Ethereum L1, then compress the data, package transactions, and finally record the proof of data onto Ethereum.
But are all L2s created equal? Are they all complementary to Ethereum?
L2 and ETH Economics
Before L2, transaction fees would skyrocket when Ethereum became congested. At one point in 2021, a transaction even cost as much as $200. Using Ethereum during the adoption cycle is a bit like trying to hail an Uber after your flight is delayed, there is too much demand and not enough supply.
L2 is designed to solve this problem. But what exactly happens when users pay a 21-cent fee on Arbitrum today? How much fees did Arbitrum reap? How much profit does the verifier get?
How much value does the baker get when you buy a sandwich at a coffee shop? How much should meat producers be paid? How much does it cost the dealer? How much do toppings and condiments cost? What's so special about a sandwich? What brings the most value? Is everything else a supplement?
This is something we should be thinking about when it comes to the economics of L2 and the Ethereum technology stack.
The answer is that as of 2023, approximately 62% of the fees users pay on Arbitrum are paid to Ethereum validators. Since its inception in 2021, 64% of all fees have been paid to Ethereum validators.
Meanwhile, 80% of the fees generated through Optimism now go to Ethereum validators. Since its inception in 2021, 74% of all fees have been paid to Ethereum L1. The following is the data provided by Token Terminal:
It's a win-win situation. Application users get superior execution and lower fees, which further drives more use cases, more developers, and ultimately more users. Ethereum validators will benefit in the form of transaction fees, and passive holders will benefit as increased transaction volume drives demand for Ethereum. Meanwhile, Optimism and Arbitrum benefit from Ethereum’s network effects, essentially outsourcing their consensus and security costs to the base layer.
What about Polygon?
Polygon is a sidechain of Ethereum, which makes it different from Optimism and Arbitrum. As a sidechain, Polygon has its own set of validators. Therefore it has its own consensus and security that is not "consistent" with Ethereum like Optimism and Arbitrum.
As of 2023, Polygon has paid 44% of user fees to its suppliers/validators, with the remaining fees having been burned. Since Polygon fees are paid in MATIC tokens, neither Ethereum validators nor ETH holders receive this value. That is, Polygon periodically submits its own state to the Ethereum mainnet, which contains a summary of all transactions on the sidechain since the last checkpoint. The checkpoint is then stored on Ethereum as a Merkle Root, a unique hash value that represents the state of the sidechain at that time. So although Polygon is integrated with Ethereum, its economic impact on L1 is minimal compared to Arbitrum and Optimism because checkpoints require minimal data. We can understand this by looking at the main sources of Ethereum burns. Arbitrum and Optimism top the rankings, while Polygon ranks in the bottom three despite handling high transaction volumes.
In summary, not all L2s are created equal. Sidechains like Polygon operate more like their own L1 while leveraging Ethereum for ultimate security and settlement guarantees. Rollups like Optimism and Arbitrum truly rely on Ethereum for consensus, security, and data availability.
When we analyze the economic impact on Ethereum validators and ETH holders, we believe that rollups such as Optimism and Arbitrum are complementary to Ethereum. They can create net new demand for block space by providing a superior user experience to application developers. At the same time, the majority of the value processed by their solution ends up accruing to Ethereum validators in the form of transaction fees and to passive ETH holders in the form of fee burns.
Polygon seems to have little complementarity with us. In fact, it looks like Polygon holders and validators benefit more from its close ties to Ethereum.
Investment perspective: How to understand complementarity?
Every product on the market has substitutes and complements. A substitute is another product you might buy when the first product is too expensive. For example, chicken is a substitute for beef. Complementary goods are products that you typically buy with other products, think gas and cars, or bread and hot dogs. All other things being equal, when the prices of a product's complements fall, the demand for that product increases.
Now, if L2 is complementary and they continue to reduce costs to achieve a superior user experience, what does this mean for ETH?
We believe ETH will continue to capture the majority of the value generated by L2.
An analogy might help highlight our thinking: let’s compare Ethereum, Google, and the Internet. We think owning Ethereum might be like owning a part of the Internet, or the important protocols that run the Internet. Now, Google is on top of the Internet Protocol. Google is great, it solves the search problem on the internet. In this way, Google enhances the usefulness of the Internet and brings more users. Now, let's assume that every time someone searches on Google, 6 cents goes to Google and 14 cents goes to the Internet Protocol. Which one would you choose as an investor? What if Google was just one of thousands of applications running on the internet (Ethereum)?
Likewise, Arbitrum and Optimism are solving important problems. We think they will drive more users to use Ethereum. But we can’t ignore the fact that about 70% of the value today belongs to ETH validators and holders. The market appears to be valuing the security, decentralization and settlement guarantees of execution services provided by L2.
Does this mean investors should avoid L2? There are definitely some nuances here. We believe Ethereum could reach a trillion-dollar market cap in the next bull run. If L2 continues to charge 30% of user transaction fees, and all transactions (or the vast majority) go through L2 in the future, one could argue that they could capture 30% of Ethereum's valuation, which would be 300 billion Expected market capitalization in U.S. dollars.
Along those lines, L2 still has plenty of upside. Keep in mind that only about 4.5% of the gas used on Ethereum today runs through L2. Having said that, the top three L2s combined generate 6% of Ethereum’s transaction fees, but have a fully circulated market capitalization of 14% of Ethereum’s.
Investment method
Our current thinking is that a handful of L2s will likely capture most of the value in the second layer of the technology stack. This is what we have observed so far and is consistent with the power law dynamics we observe across the entire Web3 technology stack. We will also look at Coinbase’s launch of L2, Base, later this quarter.
Another way to think about it is to compare owning ETH to owning an index fund. Just as new companies rotate in the S&P 500, L2s & Apps will rotate in and out of the Ethereum ecosystem. Ethereum validators and ETH holders will benefit from this. Just as SPY holders benefit from strong new business being added to index funds.
We think investing in L2 and App is more like picking stocks than picking an ETF or index. Of course, in this case, the (ETH) index still represents a call option on the future of Web3.
Do we still need L1 competing chains?
If L2 is solving Ethereum’s scaling issues, what is the value proposition of alternatives like Solana, Avalanche, and Cosmos?
First, let’s quickly run through some numbers to compare Solana’s performance.
Solana is still over 10x more efficient than Polygon, the most scalable L2 (sidechain) on Ethereum today. Solana also dominates the developer side and has a strong user base, exceeding 380,000 daily by 2023. Having said that, TVL revealed the benefits L2 gained from its “alliance” with Ethereum. L2 does not have to channel its own TVL like Solana does. I guess this is why SOL validators get to keep 50% of transaction fees (50% of which are burned), while Arbitrum and Optimism only keep about 30% combined.
But how does Solana’s monolithic approach to scaling compare to the modular approach via L2 on Ethereum?
Solana bundles settlement and execution services into the same solution, which is unique compared to Ethereum and alternative L1s such as Cosmos and Avalanche.
The modular approach taken by Ethereum, Cosmos, and Avalanche creates a lot of additional complexity related to interoperability, security, centralization, data availability, etc. that Solana does not have to deal with.
We believe Solana is significantly different from Ethereum and other competitors because of its unique overall architecture. In fact, Ethereum’s approach to scaling via L2 is starting to look very similar to that of Avalanche and Cosmos. So if developers can deploy on Ethereum Rollup and have instant access to Ethereum’s liquidity and network effects, why would they choose the Avalanche subnet or the Cosmos application chain?
To sum up, we view Solana as a replacement for Ethereum rather than a complement to Ethereum. In our opinion, this makes Solana the most interesting alternative. L1 Opportunities in the Web3 Technology Stack - This is why we took a position in the web during the market dislocation earlier this year. We reported on this project in early January. We also recommend using the network, wallets, apps, staking services, and more. We believe Solana offers the best crypto user experience currently.
All in all, as L2 gains traction, we think the value proposition of alternatives that are not that different from Ethereum will be questioned.
Conclusion Ethereum L2 is growing and becoming more efficient. It’s still early days, but there are signs that Rollup is very complementary to ETH as validators and holders receive 70% of fees through them. This is not the case for sidechains like Polygon which run their own validators. Optimisim and Arbitrum's L2 margin needs to be monitored. As more competitors enter the space, we expect margins to decline significantly, but we will continue to monitor the data. Zero competition could ultimately commoditize the execution layer of the technology stack. Since L2 is complementary to Ethereum, we believe that as the cost of L2 decreases, demand for Ethereum-based applications will increase, bringing more value to validators and ETH holders. Note that transaction volume flowing through L2 will eventually need to grow faster than fees fall. As solutions like Arbitrum and Optimism mature, the Ethereum ecosystem’s approach to scaling appears to be converging with those of Cosmos, Polkadot, and Avalanche. For example, Optimism’s “hyperchain” looks similar to Cosmos Hub or Avalanche’s C-Chain. Investors have a range of options when configuring their portfolio of assets in the Ethereum ecosystem (ETH, L2, oracles, cross-chain bridges, applications/protocols). Having said that, ETH appears to offer more attractive returns due to the complementary nature of L2 Rollup. Additionally, investors should consider that sidechains like Polygon are less “consistent” with Ethereum and serve more as an alternative L1 while leveraging Ethereum’s network effects and EVM standards. We believe that Polygon has stronger value-add potential than Rollup. In some ways, Polygon should be compared more to Solana than to Optimism and Arbitrum. We believe that the growth and scalability of Ethereum L2 will shake up the value proposition of competing L1 chains that take similar scalability approaches – such as Cosmos, Avalanche, and Polkadot. But Solana is significantly different because it takes a single extension approach (bundling settlement and execution).