5 Models of L2 Staking: From Revenue Sharing to Sequencer Security
What Is L2 Staking Really About? Breaking Down the Hype, Models, and Misconceptions
Most people in crypto have heard about staking, a foundational mechanism in Proof-of-Stake (PoS) blockchains.. By locking (staking) tokens to secure the network, participants are rewarded for contributing to consensus and economic security. This mechanism underpins the operation of chains such as Ethereum and Solana, offering a scalable and energy-efficient alternative to Proof-of-Work (PoW).
Recently, the concept of staking has expanded beyond its original scope. New approaches now enable token holders on Proof-of-Work chains like Bitcoin to use their assets to support the security or functioning of external PoS networks. One example is Babylon, which has accumulated over $5 billion in Total Value Locked by enabling Bitcoin holders to stake BTC in support of other networks.
Parallel to this evolution, the concept of staking is also being redefined in the context of Ethereum Layer-2 (L2) networks. While rollups do not run their own consensus mechanism, several L2 protocols have begun to explore staking models. These mechanisms are distinct from traditional PoS in that they do not secure consensus directly, but instead aim to distribute economic value, incentivize ecosystem growth, or eventually support decentralized sequencer designs.
This is where the term "staking" becomes nuanced.
How Rollups Work
Rollups, layer-2 blockchains, including Arbitrum, Optimism, ZKsync, Starknet, Zircuit, rely on an Ethereum Layer-1 blockchain for data availability and final settlement.
Rollups bundle transactions and post them to a base Layer-1 like Ethereum. They use different validity models — optimistic or zero-knowledge — but they all rely on Ethereum, and its PoS consensus, for security.
Rollups do not have their own consensus mechanism but rely on Ethereum Proof-of-Stake for security and data integrity.
The core operational actor in this process is the sequencer, which is responsible for transaction ordering and batch submission. Each rollup has a sequencer responsible for:
Ordering transactions
Compressing and submitting batches to L1
Paying L1 fees for calldata and posting
Sequencers collect L2 gas fees, and because most rollups today use a single centralized sequencer, this can be an extremely profitable operation.
As decentralization pressure grows, some rollups are exploring how to open this role and redistribute value. This is where staking enters the conversation.
What Does "L2 Staking" Mean?
Staking on L2 does not mean participating in consensus — at least not today. Instead, it falls into two categories:
Revenue-sharing or yield participation
Users stake tokens and receive a portion of the protocol’s economic activity — often tied to sequencer revenue or off-chain yields.Sequencer security (future-oriented)
In models where sequencers are decentralized, staking could be used to bond honest behavior and enable slashing — similar to PoS designs.
Let's explore some real-world implementations.
L2 Staking: Emerging Categories
1. Revenue Sharing from Centralized Sequencers
Example: Unichain
Staking is used to distribute a portion of sequencer revenue to token holders. The sequencer remains centralized, but a share of the profits (e.g., gas fees) is paid to those who stake native tokens at sequencers. These tokens act as a security guarantee that the sequencer behaves honestly, similar to staking at validators in the Proof-of-Stake blockchains.
Today, most rollups run on one (centralized) sequencer. Once the sequencers become decentralized, this mechanism will be a must-have. However, so far, only Unichains (L2 from Uniswap) is planning to offer such L2 staking, with 65% revenue from the sequencer committed to stakers.
The redistribution of sequencers revenues to L2 stakers can be implemented in one two ways:
Option 1 - Key Features:
No impact on protocol security
Token holders gain passive revenue exposure
Often tied to governance token utility
Option 2 - Key Features:
Staked tokens act as a guarantor for honest behavior of rollup sequencer.
Token holders gain passive rewards exposure for participating in sequential revenues.
Often tied to governance token utility.
2. Staking for Decentralized Sequencer Security (Future-Oriented)
Example: Starknet (planned), Espresso (design stage)
In future architectures, staking will be required to secure decentralized sequencers. Token holders bond assets that can be slashed if the sequencer misbehaves (e.g., reordering, censorship). This introduces a PoS-like model to rollups.
Although it remains a question whether the entire implementation of PoS consensus mechanism is required for decentralized sequencers, or whether the security-oriented L2 staking decision in the previous model will be enough.
Key Features:
Enables slashing and accountability
Essential for credible neutrality and trust minimization
Currently only in design or pilot phases
3. Liquidity Staking for Ecosystem Bootstrapping
Example: Zircuit, Linea
Users stake ETH, stablecoins, or native tokens, which are redirected to DeFi protocols on the L2 to support lending, AMMs, or other economic activity. This isn’t staking for validation, but for capital formation within the L2 economy.
Key Features:
Supports DeFi growth and activity
Yield may be indirect (e.g., via incentive programs)
No involvement in infrastructure or ordering logic
4. Protocol-Native Yield Distribution
Example: Blast
Yield is sourced from external protocols (like Lido, RWA platforms) and distributed natively within the L2 environment. Users automatically earn staking-like returns just by holding assets on the network. This is a user-side abstraction of staking, not protocol-side.
This approach is implemented by the native rollup bridge that, for every wETH (on L2), it stakes Ether on Ethereum and, for every stablecoin (on L2), it acquires a yield-paying stablecoin (on Ethereum).
Key Features:
Yield integrated into the L2’s UX
No user-side delegation or bonding
Yield may vary based on off-chain protocol performance
5. Governance-Linked or Speculative Staking
Examples: Tokens locked in anticipation of airdrops, governance rights, or protocol upgrades
Most of the L2s operate in this model today: Arbitrum, Optimism, ZKsync.
Some L2s, like BASE, do not have even a token.
Users stake or lock tokens for non-economic utility, such as voting power, governance delegation, or to gain access to future features or distributions. While not tied to sequencer or yield mechanics, these mechanisms mimic staking behavior.
Key Features:
Non-revenue, non-security purpose
Often speculative or reputational
Used in DAO voting or long-term alignment
Most leading rollups do not yet offer staking mechanisms linked to infrastructure-level participation. Arbitrum, Optimism, and ZKsync have not introduced staking models that reward users based on network usage or revenue. Base, developed by Coinbase, does not currently have a token or associated staking system.
Final Thoughts
L2 staking today is not really about consensus. It’s about aligning incentives and distributing protocol value.
The models are still early, but the direction is clear: L2s will increasingly look like economically independent platforms with new models of staking emerging: designed to distribute value, enable protocol participation, and prepare for a future of decentralized infrastructure.
L2 Staking is about to redefine staking beyond PoS consensus, with significant implications for token utility, protocol design, and user incentives.