Lido v3 Whitepaper RFC

[ujenjt]

Over the past year, a group of Lido contributors has worked on a new architectural direction for the protocol. This direction is centered around stVaults, isolated staking positions that decouple how ETH is staked (e.g. node operator, fees, risk preferences, sidecars) from the liquidity layer.

Liquidity has always been one of Lido’s core strengths, and the main design challenge has been to give stakers highly customizable, individualized positions while preserving the same liquidity benefits offered by stETH. Before drafting this whitepaper, we conducted joint research with MixBytes to explore how other protocols manage individualized staking, and how those designs could be meaningfully improved.

We expect stVaults to become a highly composable building block that serves as a foundation for creating strategies, protocols, and institutional staking products. We invite contributors from other protocols, ecosystem builders, and staking researchers to engage with the RFC, provide feedback, and help refine the final design. The core architecture is in place, though we remain open to challenges on the details and improvements where they make sense.

The draft covers a broad design surface, and we anticipate continuing the refinement process over the coming weeks, with the goal of finalizing the whitepaper by mid-summer. For suggestions regarding wording, clarity, or structure, you’re welcome to leave comments and suggestions directly in the HackMD document. For more substantive or directional questions, please use this forum thread so the discussion stays visible and shared.

We’re particularly looking for feedback on the stVault design and risk management model, especially on how to balance the stETH’s risk profile and redeemability with staker-driven node operator selection. This is one of the most critical areas of the design, and your input would be highly valuable.

Thank you to everyone who contributed to the research, conversations, and writing that led to this draft. We genuinely look forward to your feedback and questions—they matter a lot to us at this stage.

Whitepaper: Lido V3 Whitepaper — Draft for RFC - HackMD

[kadmil]

Excited to see v3 shaping up, and how specifically does it shape up!

[UniteTheClans]

Amazing work. stVaults coming in hot.

[vtrush]

This is such a huge job! I can’t imagine how big your brains must be!

[Leuts]

Big congrats on shipping the Lido v3 whitepaper!

A few questions:

1. Regarding the stVault “marketplace” → the biggest question one always asks about marketplaces: what is the curation process? Am I correct in presuming that Lido will not offer any curated “marketplace” and this will simply be a permissionless, open system where each operator must market their own offering on their own?

2. “Within stVaults, stake centralization is mitigated through economic measures. Node operators with a larger share of backing face higher fees and reduced mintable collateral. The system is designed so that users who choose node operators that improve validator set diversity receive better terms. This creates a natural incentive to support decentralization through individual staking choices.”

Can you break this down further? Is there any risk of node operators “cheating” the system via external incentives or otherwise?

3. No vault can be forced to accept an upgrade, allow more than a user-defined portion of its ETH to be locked, or remain under protocol governance once the owner decides to opt out. In edge cases where governance becomes a risk, the ability to exit without unstaking offers a strong safety valve. And because validator accounting can be complex, asynchronous, and partially off-chain, stVaults are built to put risk management in the hands of the user. This reflects a core Lido principle: ensuring control and resilience stay with the user.

Just to be clear with language here (maybe worth updating) →
Protocol governance = Lido DAO governance?
Governance becomes a risk = Risks of Lido DAO governance on stVaults?
User = Owner/staker or operator?

4. Regarding the escape hatch. It’s excellent to provide “exit” but have you weighed the expected outcome of vaults exiting versus having voice. ETH as an asset is a prime example of what happens when you give no proper voice, people exit or sell. It’s challenging to determine the balance and tradeoffs. But has there been any thought put into this? The impact of mass exit versus the impact of voice.

5. What is the metagovernance within each vault. Meaning, what decisions can an owner make with their stVault? Is this flexible?

6. Final question: from a strategic perspective what are the expectations of how lido v3 increases stETH usage and reignite growth and finally increase LDO incentive alignment (if at all)?

Overall amazing work and very understandeable. It’s very clear that modularity is still a key ingredient to create flexibility and future-proofness in all that we build! At least until our market matures, which we are a long way from imo.

[Maxim_101]

Excellent job. Looking forward to the implementation.

[March]

How can a project with a market value of one billion US dollars manage assets with a market value several times its own? This is unreasonable.

[Mol_Eliza]

Hey @Letus thank you for a very important and precise questions.
I’m Max from Analytical workstream, and I’m really glad for the opportunity for a discussion around the possible designs - so would jump in, providing my vision, on at least some of them.

1. Regarding the stVault “marketplace” → the biggest question one always asks about marketplaces: what is the curation process? Am I correct in presuming that Lido will not offer any curated “marketplace” and this will simply be a permissionless, open system where each operator must market their own offering on their own?

You are correct, there wouldn’t be any straightforward curation, as the intention of stVaults is creating a permissionless market, where stakers could express their preferences on how and with whom to stake, and Node Operators and builders can provide an options for that.

However, a completely open market is subject to market failures (for example due to unpriced externalities with correlation penalty due to slashing, or imperfect information within different approaches and view on risk valuation). Therefore I see DAO as an actor, setting and additional incentivization layer via flexible fee structure and risk mitigation levels. The examples of that could be: increased Reserve Ratio (share of collateral required to mint stETH) for vaults increasing systematic risks with centralization (for example staking with already huge Node Operator) or decreased fee for actors committed to providing public good for the network (e.g. increasing censorship resistance).

So each operator must market their offering on their own, but also taking into account the incentives and limits transparently put by the DAO.

1. “Within stVaults, stake centralization is mitigated through economic measures. Node operators with a larger share of backing face higher fees and reduced mintable collateral. The system is designed so that users who choose node operators that improve validator set diversity receive better terms. This creates a natural incentive to support decentralization through individual staking choices.”
   Can you break this down further? Is there any risk of node operators “cheating” the system via external incentives or otherwise?

This is one of the prime examples of this indirect market governance, build on idea of mitigating the risks connected with the stVaults ability to mint stETH.
Details on the suggested approach could be found in a corresponding thread, but the main idea is, as in my response above, build around:

1. Stakers, Node Operators and consensus among them - is the main, and only factor of distribution within stVaults
2. If Node operators are willing to share information about their setup, valuation on risk associated with stETH minting is lowered, as within lack of information the most conservative approach is suggested (e.g. high Reserve Ratio, assuming the highest systematic centralization risk)
3. With lower risk valuation - better conditions in terms of Reserve Ratio could be provided
4. The Risk framework is built based on Ethereum consensus specification, therefore, naturally, with increased concentration risk valuation increases, providing a better conditions in terms of Reserve Ratio for Node Operators with lower stake and incentive for diversification on the market, but (see 1) at the end it’s up to users to decide.

[dmitrii_v]

Hi @Leuts!

Thank you for your questions!

Re: Marketplace. I believe Staking Marketplace should be permissionless, as stVaults are, and competition on the Marketplace should be driven mainly by its participants.

However, DAO should be able to improve terms for those participants who make the public good and, more importantly, have proven effectiveness and durability. The last is required because improving terms for the market participants should not drastically increase risks for the Lido core protocol and stETH holders.

[Leuts]

Thanks! It’s always tough having “permissionless” marketplaces due to the risk of scams, but, looking forward to seeing how things are planned and implemented!