Base
Polygon
*This research was done with M3TA Analytics and the dashboard can be viewed at M3TA Analytics website
Rollup projects like OP-Mainnet, zkSync Era, and Polygon zkEVM generate revenue primarily from L2 transaction fees paid by users. They also incur various costs like L1 storage and operating costs.
A majority of Rollups’ costs come from storing transaction batches on L1. OP-Mainnet, zkSync Era, and Polygon zkEVM also appear profitable based on the data.
Rollup framework developers like OP Labs and Matterlabs currently offer open source frameworks, but will need to capture business opportunities to generate sustainable revenue as forks may diminish profits. Offering RaaS could optimize costs and increase revenue.
As more projects utilize rollup frameworks to launch their own rollups, RaaS will play an important role in analyzing each rollup's revenue/cost structure to find optimization paths that reduce costs and increase revenue.
Rollups have become ubiquitous. As Ethereum's popularity grows, the network has struggled to keep up with demand, resulting in network congestion and high fees. Rollups have alleviated these issues, but it is important to understand the current landscape and assess whether the future looks promising based on financial analysis.
Rollup economics is an extremely important topic that has yet to receive the attention it deserves. The concept of Rollups has been gaining traction in recent years, and as such, it is crucial to delve deeper into its intricacies. Understanding the costs and revenue for token holders, particularly investors, is paramount for Rollups. It is also important to note that Rollups are becoming more like RaaS (Rollup as a Service), which further underscores the need for builders to fully comprehend the cost structure of Rollups.
Rollups work by offloading a batch of transactions to be processed off-chain, then submitting the compressed transactions as a batch to Ethereum. This allows for much faster and cheaper transactions while still preserving the security and decentralization of the mainnet.
Many successful rollup projects are moving towards becoming a Rollup framework, creating a separate ecosystem for each chain built on the framework.
Some of the examples
OP-Stack from OP-Mainnet, vision is Superchain
ZK-Stack from zkSync Era, vision is Hyperchain
zkEVM from Polygon, vision is Polygon 2.0
Orbit by Arbitrum
Starknet Stack from Starknet, vision is Fractal Scaling
There have been several previous studies conducted to comprehend the economics of rollups, including David Crapis's article, Kofi's Dune dashboard, Blockworks Dashboard, and Tokenterminal. These sources have offered valuable insights by attempting to estimate the revenue and cost associated with operating OP-Mainnet.
Notable related works include:
Rollups are Real — Rollup Economics 2.0: analysis on the rollup economics
Modular Economics by M3TA & FourPillars: Cost and Revenue analysis on OP-Mainnet, Polygon zkEVM, and zkSync Era
Rollup Economics by 0xKofi: This dashboard shows the cost and revenue data of major rollups in Ethereum
Blockworks Data Dashboard: This is paywalled but provides a comprehensive analysis on the economics data related to Ethereum and rollups.
However, we still need a more thorough analysis of the revenue and cost structure. It's important to consider the hidden costs that are often overlooked, such as the expenses related to operating a sequencer and the revenue generated from L1 transactions sent to L2. Furthermore, formalizing this structure further will contribute to a more comprehensive understanding of the economics of rollups.
Why an income statement? In traditional finance, companies produce financial statements consisting of a Balance Sheet, Income Statement, and Cash Flow Statement. There are no standardized financial statements in the world of cryptocurrency, although great teams like Token Terminal and Blockworks Research are working to build them.
There is a need for more structured financial resources for investing and launching a rollup. I believe that the income statement is a good starting point. It should cover various aspects where rollups incur revenue and costs beyond well-known revenue sources like rollup transaction fees.
Furthermore, projects are exploring the use of rollup frameworks such as OP-Stack, Polygon-CDK, and potentially ZK-Stack to build their L2 solutions. It is crucial to comprehend the technological distinctions among these stacks, as well as the financial implications of utilizing each rollup. Moving to L2 entails developing a customized rollup specific to the project's requirements, with the aim of generating increased revenue.
Source: Snowflake SEC Filings 2023-08-31 10-Q
OP-Mainnet is an optimistic rollup solution designed to scale Ethereum. It moves computation and state storage off-chain, reducing gas costs for users. Unlike other scaling solutions like sidechains or plasma chains, OP Mainnet derives its security from the Ethereum Mainnet by publishing transaction results on-chain. This ensures that the native security of Ethereum secures the broadcasted information. Also, as the state root changes in OP-Mainnet, it updates the change to the L2OutputOracle contract in Ethereum.
This makes the input and output data available in Ethereum, allowing challengers to contest whether the computation was performed without error.
2.1.1 Income Statement
2.1.1 Revenue
L2 Transaction Revenue: In Optimism blockchain, transactions have L2 (execution) and L1 (security) fees. L2 fee is the cost of executing the transaction, calculated by multiplying the gas used by the gas price. L1 fee is the cost of submitting the transaction to Ethereum, calculated based on Ethereum's gas price. Optimistic rollups use a gas fee scheme similar to Ethereum, with fees depending on L2 operator fees and L1 data fee.
Tx fee paid by L1 users for L2 Tx: Users can make a transaction in OP-Mainnet by bridging a transaction from L1, Ethereum. This result in a upfront fee for users to pay to handle the transaction in OP-Mainnet. However, this amount is minimal
MEV Profit: MEV profit is generated by reordering the transaction. It is currently known that the operator of the sequencer, OP Labs is not generating revenue from this. Currently, OP-Mainnet is operated under private mempool where the block is produced every 4 seconds.
Ecosystem Fee: OP-Mainnet's primary codebase, "OP-Stack Framework," is currently open-sourced and can be used to launch a L2 chain. This has resulted in several projects utilizing the OP-Stack codebase. These forks are not obligated to share revenue with OP-Mainnet. However, Base has recently announced that they will share their profits with OP-Mainnet.
OP Token: OP-Mainnet has its own token, but its current utility is limited to governance and it is not used as a network gas token. Nonetheless, the token issuance has generated a new revenue source for OP-Mainnet, which is used to fund projects that can help the OP-Stack ecosystem flourish.
2.1.2 Cost
L1 Storage Cost: The storage costs for OP-Mainnet include storage batch storage cost, and StateRoot storage cost. Batch storage cost is the cost of storing transaction batches to Ethereum's BatchInbox Address. StateRoot storage cost is the cost of storing State Root data in Ethereum's L2OutputOracle Contract.
Challenge Cost: To incentivize correct computation and deter malicious behavior by sequencers, OP-Mainnet has a system for challenging. However, there hasn't been a challenge yet, and currently, this process is centralized.
2.1.3 Analysis
According to the on-chain data, OP-Mainnet has been profitable.
Based on the total “L2 Transaction Fee Revenue”, the above graph show how much cost it has been used in the recent operation. In here, it can be observed that most of the cost comes from “L1 Tx Batch Fee”
In the zkSync Era, Rollups are used to increase Ethereum scalability by consolidating transactions off-chain and sending them to the main chain in a single batch via a cryptographic proof of validity. The platform aims to preserve Ethereum's values of freedom, self-sovereignty, and decentralization at scale. Additionally, zkSync rollups are cost-efficient for most transactions compared to other scaling solutions.
The operator in zkSync Era executes the transactions and generates a zk proof. These are stored at L1.
2.2.1 Overview
2.2.1 Revenue
L2 Fee Revenue: revenue generated from transactions in L2
Profit from L1 Tx for L2 Tx initiation: Users in L1 initiate transactions to be included in a queue where the transactions wait to be initiated in L2. Users pay a fee for inclusion and execution, which is calculated as revenue. However, the cost of initiating transactions in L2 is excluded from the calculation to determine profit.
MEV Profit: In the zkSync Era, a block is produced every 2 seconds, and the operator can capture MEV from it. However, it is known that this project does not capture MEV.
2.2.2 Cost
commitBlock: checks the L2 block timestamp, processes the L2 logs, saves data for a block, and prepares data for the zk-proof.
proveBlock: validates the zk-proof.
executeBlock: finalizes the state, marks L1 -> L2 communication processing, and saves the Merkle tree with L2 logs.
2.2.3 Analysis
According to the financial analysis, zkSync Era has been profitable based on the revenue and cost data. This profitability demonstrates the viability and success of zkSync Era as a rollup solution.
For zkSync Era, there are three major costs: commit, verify, and execute. According to the data, the highest costs come from committing the transaction batch and verifying the zk proof.
Polygon zkEVM is a scaling solution that aims to improve Ethereum network's scalability without compromising its security or decentralization. The solution is designed to process transactions off-chain and collect them into a single on-chain transaction, thereby reducing computation on the base layer and improving scalability.
Polygon zkEVM is compatible with the existing Ethereum ecosystem, allowing developers to migrate their applications to it without significant changes. Polygon zkEVM inherits the security of the Ethereum L1 network and uses zero-knowledge proofs (ZK proofs) to ensure transaction validity and safeguard user funds. This provides an additional layer of security for transactions processed on Polygon zkEVM.
By using ZK proofs, Polygon zkEVM reduces transaction costs, which makes it more affordable and improves the overall user experience. It is worth noting that Polygon zkEVM Rollup is distinct from other solutions offered by Polygon and provides a different user experience while allowing users to choose the solution that suits them best.
2.3.1 Overview
In the Polygon zkEVM L2 context, both the sequencer and aggregator have specific revenue streams and cost.
2.3.2 Sequencer
Revenue
L2 transaction gas fees revenue in ETH:Sequencers collect transaction fee revenue from users in L2. This revenue is collected in ETH.
Cost
L1 Sequencer Transaction Fees Cost in ETH: Sequencers batch transactions in L2 and commit them to the Polygon zkEVM Contract deployed in Ethereum. This incurs gas fees for storage.
Batch Fees Cost in MATIC: Sequencers send MATIC to the contract to share the overall revenue with the aggregators.
2.3.3 Aggregator
Revenue
batch rewards in MATIC: Since the aggregator generates a proof based on the batch stored by the sequencer, the reward is given to the aggregator for the generation of zk proof. The MATIC tokens go to the aggregators, and it is profitable if the transaction fees are greater than the L1 call fee plus the MATIC fee.
Cost
Aggregator Operating Cost: Polygon zkEVM currently has a single entity that generates proof. However, with this aggreagtor architecture, it will provide an easy path to improve decentralization. However, aggregators need to set up RPC to listen to events in Ethereum contracts and generate proof. This can further increase operating costs.
L1 Aggregator Transaction Fees in ETH: The aggregator generates a zero-knowledge proof and stores it in a contract to be verified on-chain, resulting in a cost.
2.3.4 Analysis
Polygon zkEVM is a recently introduced rollup, and although the Sequencer is profitable, there is a need for a better economic system to incentivize the Aggregator.
When comparing revenue and cost, it is observed that the majority of the cost is attributed to storing transaction batch data to L1, while the verification cost is relatively low in comparison to the zkSync Era.
Yes, they are currently a profitable business, according to on-chain data. However, there are additional costs in operating a sequencer, as well as development and maintenance costs. These costs are currently unknown and may be high, but like with the recent Bedrock upgrade for OP-Mainnet, development is currently active and RaaS service providers can further lower the cost.
I analyzed the revenue and cost structure of major rollups that are planning to offer their codebase for other projects to fork and run their own rollup. However, the analysis is incomplete. The off-chain cost can be significant, but the information is not well-known. Additionally, it is difficult to hire developers to develop the rollup, resulting in high developer costs.
This trend will continue if the rollups are deployed by each project. Therefore, it is important for RaaS (Rollup-as-a-Service) providers to analyze the revenue and cost structure of each rollup, as they have the ability to observe the structure. By finding optimization paths, they can reduce costs and increase revenue.
This needs to be explored further. There will be benefits to operating a shared sequencer, such as decentralization and atomic cross-chain transactions, but the operating entities and infrastructure will increase. Will this lead to higher costs? This needs to be further analyzed.
L2 developers like OP Labs, Matterlabs, and PolygonDev are entities that know most aspects of operating a rollup and are building frameworks to launch and operate them. The framework itself is currently open source for OP-Stack and ZK-Stack (planned). This can diminish revenue generation, for example, like with Cosmos Hub. Therefore, it will be important to capture business opportunities while also open-sourcing the code to make the ecosystem flourish.
Thanks to Kate for designing the graphics for this article.
Astar
Bitcoin
Polygon
