Improving stETH Distribution on Other Platforms

[monet-supply]

Background

stETH/wstETH has limited availability outside of ETH L1. Currently, the primary liquidity sources for stETH are:

ETH L1 -

• Lido deposit contract (buy only)
• Curve
• Balancer
• Sushiswap
• 1inch

Terra -

• Terraswap

Solana -

• Orca

While Ethereum is suitable for buying or selling large amounts of stETH with the deepest liquidity pools, transaction fees can significantly eat into returns for small purchases. The figure below gives estimates on the impact of fees at various trade sizes.

Source: Google Sheet

Solana and Terra both offer very low transaction fees, but they are less familiar to many users than EVM based chains. Solana and Terra also use ETH far less within their respective defi ecosystems, which could negatively impact adoption of stETH on these chains.

EVM Chain Options

There are several L1 and L2 platforms where stETH could be integrated and potentially increase access for retail users. ETH is already a central asset across each of these chains so stETH could see an easier path to adoption.

L1 / Sidechain:

• Polygon
• xDAI
• Avalanche
• Fantom
• many others

L2:

• Arbitrum
• Optimism
• Boba
• zkSync (?)
• Starkware (?)

It probably doesn’t make sense to encourage liquidity on all of these platforms (at least initially), but choosing a few of the highest impact EVM L1/L2s could improve adoption and help cement Lido network effects.

Liquidity Strategy

There are a few key decisions to make about how to encourage cross chain stETH liquidity.

(1) Where should Lido support stETH liquidity (which L1/L2s, which DEXes)?

Curve is available on most of the chains listed above. This is probably the simplest and most capital efficient platform for stETH liquidity. Liquidity on Optimism could use Uniswap v3 concentrated liquidity, which is somewhat more complicated than Curve but still capital efficient. Other platforms would need to use XYK liquidity pools which requires significantly more capital.

IMO the best initial platforms to support stETH would be Arbitrum, Polygon, Avalanche, Fantom, and xDAI. Each of these chains has Curve support, and they all have active lending markets which could become additional integration targets for stETH in the future.

Optimism and zkrollups should also be a high priority over the medium term but have maybe a bit less opportunity in the short term.

(2) How much liquidity is needed?

One way to gauge this is by setting a maximum target transaction cost across various trade sizes, and ensuring enough liquidity to support trades up to the size where ETH L1 becomes more efficient.

So for example if we wanted to target less than 1% transaction cost across trade sizes, we can see this is already well supported for sizes above $10,000 on ETH L1. So in this case we would target enough liquidity for less than 1% market impact on a $10k trade on each chain we choose to support. For typical xyk AMM like Uniswap v2, this would require about $2 million in liquidity. But for stableswap AMM like Curve, this could likely be reduced to $100-200k.

(3) How to build liquidity?

There are two primary options: liquidity mining or PCV (protocol controlled value).

PCV is capital intensive, but Lido has significant ETH resources which it could deploy for this purpose. Liquidity mining can have a larger short term impact on market depth, but increases operating cost with no certainty that liquidity would become self sustaining over time.

Assuming $200k target liquidity per chain across 5 chains, this would be $1 million in ETH capital committed to own liquidity as PCV (slightly more than 1% of Lido’s ETH reserves). Alternatively, if we run liquidity incentives with 10% target yield, total liquidity mining costs would be $100k per year across supported chains, or roughly ~2,500 LDO per month at current prices.

[PatrickOD]

I support the idea of putting resources towards increasing stETH availability on platforms with lower fees. And because Avalanche, Polygon, and Arbitrum are the current leaders in bridge TVL these are natural places to look first.

Polygon bridge already supports stETH, but (on the main bridge interface) Arbitrum and Avalanche do not. Are there any technical hurdles to using bridges to move stETH onto these other platforms or would there just need to be some coordination with those projects?

Is offering incentives for providing trading liquidity for stETH sufficient to bring stETH onto these other platforms or would it be more effective to incentivize bridging activity also?

Re protocol owned liquidity vs LP incentives, in general I think it would make sense to move away from directly paying for liquidity for the reasons OP stated or to at least or have a mix of both incentives (esp for stETH/ETH pools rather than stETH/Dai). It might make more sense to offer LP incentives where they can be combined with existing rewards programs offered by the DEXs themselves (on Avalanche for ex). $1m in ETH capital certainly seems manageable from the DAOs perspective to own liquidity on these platforms, however.

[jbeezy]

First of all, @monet-supply thank you for posting this. Well articulated and tackles a lot of key points I have been thinking about.

I generally support this. Curve doesn’t have the best UI in the game for newer people, but the efficiency is helpful. The secondary effect is that Curve LP tokens have high utility with other protocols. This increases stETH utility and offers a potential solution to longer term liquidity mining.

I do still think Lido should explore PCV, but not for this specific use case. For this, it should focus on smaller trades while exploring these new ecosystems for opportunities to incentivize downstream use outside of simple liquidity. Something that will make use of the Curve LP tokens while potentially building long term partnerships. This combined with longer term PCV might be a way to shift off of pure LM making it more tactical.

Incentivizing the liquidity makes sense in the short term to get things started quickly. Most bridges support ERC20s but might have listing requirements. stETH/ETH and stETH/USDC(T)(DAI) are important initial pairs, but we need to look at 7 day volumes for (W)ETH pairs.

[vsh]

Lido has to place LP incentives strategically. I can see a few strategies here, with a few different goals:

• provide a large liquidity pool to handle potential liquidations from protocols where staked assets are collateral. This better has to happen in the same execution environment where we expect the majority of e.g. stETH used as collateral. Atm that’s ETH L1 - the majority of stETH to ETH liquidity has to be there;
• Provide stETH (or wstETH) as a viable trading asset on L2s. I’m not sure how much liquidity should it be, but my hunch it should be >$10000 trade for 1% shift. That probably depends on an average trade size on the network;
• Make stETH a bridge currency/unit of account/a main trading pair vs. something in a particluar execution environment.

My guess is that the proposal up there solves the viable trading asset stuff - if so, reasonable, but I’d prefer to have metrics for viability.

[PatrickOD]

Agree that this proposal is more suitable for addressing general distribution of stETH to other platforms and and probably not enough to create a market capable of supporting liquidations in the short term. But I think that allowing smaller holders to participate in staking ETH with Lido is important. Other than directly interacting with the Lido protocol, the only way to do that is by holding stETH and right now it’s not widely available or liquid outside of ETH L1.

It may not be realistic for Lido to either own or incentivize enough LP participation to support collateral liquidations on these secondary platforms in the short term. But as a step towards increasing adoption and utility on other protocols besides ETH L1, it seems reasonable to start by building liquidity for smaller spot trading. And then progress from there towards having a robust enough market to support use of stETH as collateral.

Just to put numbers to point on slippage… Using the Uniswap model as an upper bound, required liquidity to achieve 1% slippage on trade sizes of $50k / $100k would be $10mm / $20mm in total pool value.
This would be smaller in the Curve stableswap model.