Deploy liquidity pools for sETH2/ETH and rETH2/ETH on Uniswap V3

[kiriyha]

Howdy fellow stakers!

In this post, we present the team’s vision for the deployment of liquidity pools for sETH2/ETH and rETH2/ETH on Uniswap V3 and seek community feedback on some of the aspects of the deployment.

Motivation

After the setbacks with Curve and then Balancer v2, the team wants to finally nail the topic of liquidity pools and focus on the integrations with other protocols marketing the dual token system. To maximize the success of these integrations and campaigns, the bar for choosing a DEX remains very high. As a reminder, the key criteria for deployment are:

1. Ability to deploy a stable pool to achieve a strong peg between sETH2 and ETH
2. Using a top-tier DEX (out of considerations about liquidity, security and brand positioning)
3. Cheap swap costs (to not price out smaller stakers)
4. Not letting rETH2 run below 1:1 with ETH (to maintain the ability to compound)
5. Offering a good APR to LPs (to increase incentives to provide liquidity)

There are only a few DEXs that meet these criteria, and as it became clear that Balancer v2 stable pools are also being delayed indefinitely, the StakeWise team changed course to deploying on Uniswap V3. While it required developing additional functionality, we are very happy with the result and cannot wait to share it with you.

Let’s start with a quick summary of the advantages of Uniswap V3. The key innovation of V3 is the ability to concentrate liquidity in the desired price range. For example, if I choose to provide liquidity in the [$2,000; $3,000] range in ETH/USDC pool, my liquidity will be utilized (and I will earn trading fees) for as long as ETH’s price stays within this range. I can tighten or widen the range (called “ticks” in V3) around the price - the tighter it is, the more volume will pass through my liquidity, and the more fees I will earn. However, if I choose it too tight, the price may move out of my specified range – in this case my liquidity will no longer be used, and I will stop earning the trading fees. However, I can keep my liquidity in that range (or a set of ranges) and will start earning again as soon as the price returns into it.

This has several particularly important implications:

• Significantly less capital (up to 4,000x less) is required to facilitate trading, because liquidity is concentrated in specific price ranges, not across the Uniswap bonding curve (x*y=k) (example)
• LPs are incentivized to keep liquidity as tightly around the current price of an asset as possible, because that way they can earn exorbitant fees (example)
• This has an effect of concentrating liquidity around the current price of an asset, leading to a dramatic reduction in slippage and the ability to handle huge volume of trading with just a fraction of liquidity (example)
• Liquidity provision becomes significantly more profitable while slippage is dramatically reduced

While these are the most important aspects, Uniswap V3 has a ton of other cool features like limit orders. We recommend you check out this explainer video that might help with your understanding.

Applying the innovation offered by Uniswap V3 in the context of sETH2/ETH and rETH2/ETH liquidity pools, it becomes immediately clear that StakeWise i) could achieve a stable peg between ETH and sETH2/rETH2 with ii) less liquidity than it previously needed, by incentivizing liquidity provision in as tight a range around 1 as possible. This would lead to iii) significant earnings for the LPs and iv) create a strong foundation for integrations of sETH2 and rETH2 tokens in the DeFi ecosystem.

By our calculations, we will need 5x less capital in the sETH2/ETH liquidity pools than was previously required, while boosting LPs APR by 500% and freeing up sETH2 for utilization in the wider DeFi world.

So long story short, Uniswap V3 would be a great way to go for the deployment of StakeWise liquidity pools, and that is exactly what we are proposing.

The work to deploy has been completed, and we are only working on the UI to support the farming programs that are to come. We are excited about the possibilities offered by Uniswap V3 and would love to receive your support for the idea.

Specification

1. Deploy sETH2/ETH and rETH2/ETH pools on Uniswap V3
2. Deploy novel distribution systems for incentives and rewards
3. Allocate farming incentives towards both pools
4. Launch farming campaigns and UI for claiming rewards

Until now, the absence of contracts to allocate farming incentives to Uniswap V3 LPs has been a key blocker to the deployment of pools, because everything would need to be built from scratch. However, we have now developed and audited two novel mechanisms that support deployment on V3.

To incentivize the concentration of liquidity as tightly around 1 as possible, we have built one of the first incentives systems to exist for Uniswap V3. It will adjust the amount of $SWISE earned by the LPs dynamically based on how tightly their liquidity is concentrated around the current price of sETH2: the tighter the range, the more $SWISE will be farmed. It should motivate the LPs to tighten their range around the prevailing price of sETH2, creating very deep liquidity and resulting in outsized APRs.

Claiming the farming rewards would be cheap – there is no need to stake and unstake any LP tokens in order to participate in or exit farming, and LPs can collect their farmed $SWISE (and other tokens) regularly. Therefore, if we go with V3, StakeWise will be among the very few projects to incentivize liquidity there, do it dynamically and at a fraction of gas cost of interacting with other farming programs.

We also updated the staking rewards distribution model to support Uniswap V3 so that LPs will not lose rETH2 while their sETH2 is in the liquidity pool or another contract in DeFi. Uniquely, it will not require locking the LP tokens – instead, they can be reapplied elsewhere to generate more yield. This system deserves a blog post of its own, but we believe it is one of the most advanced incentives allocation systems out there. All hail Dmitri.

Drawbacks

Uniswap V3 is not the cheapest from the gas perspective when it comes to adding liquidity. It costs twice as much gas to add liquidity in V3 than in Balancer v2 (~560K units vs ~230K units) and swaps are some 50% more expensive (~200K units vs ~130K units). Still, this is better than Uniswap V2 and other DEXs out there. (Correction: Uniswap V3 seems to be ca 28% pricier than V2 in terms of gas; deployment on Optimism is supposed to fix that).

Uniswap V3 also doesn’t support adding single-sided liquidity, unless it is for the ranges that don’t capture the current price (in this case it acts like a limit order i.e. you can submit 1 ETH to the [0.9;0.95] range and swap into sETH2 only whenever it has a large discount). Hence, if you only have sETH2 and would like to provide liquidity with it, you would need to swap a half of it into ETH and add only then.

Risks

Considering recent shakiness in the market and the delay in deployment of sETH2/ETH liquidity pool until now, there exist a number of users that are planning to swap sETH2 into ETH (and then stables) as soon as the pool is deployed. The amount of ETH these users have staked is uncertain; hence, there is a risk that we as a DAO do not allocate enough SWISE incentives to the sETH2/ETH pool to attract the amount of liquidity required to handle a flurry of withdrawals without losing the peg.

To mitigate the risk of de-pegging on the first day, we kindly ask that everyone who considered exiting from staking within the first week of liquidity pool’s deployment to anonymously state the amount they would like to withdraw via this form. Your response will help the DAO to plan ahead for attracting the right amount of liquidity on day 1 and ensure that the peg is maintained.

Meanwhile, once we learn about the size of withdrawals, we as a DAO will need to discuss how much incentives to allocate to the sETH2/ETH pool. For now, the provisional amount is 0.6% of the supply per month as per this forum post.

Discussion

With a plan for deployment in place, we are now waiting for your opinions and questions. We really want everyone to be on the same page about how Uniswap V3 works and how StakeWise specifically would interact with it, so no question is off limits. We will also set up conversations about how much to allocate as incentives towards both pools in phase-2 posts of the previous forum threads, so it is a good time to start frequenting the forum.

Things that we need to discuss and decide on:

1. Size of the swap fee: 0.05% (facilitates arbitrage trades, but fewer fees for LPs), 0.30% (best of both worlds), 1% (very handsome for the LPs but prohibits exit without a “redemption fee” which is not attractive). This needs to be decided for both tokens
2. Size of liquidity mining incentives for both tokens

We hope that you support this proposal and look forward to the things to come from its successful implementation.

[mitchel]

Nice! Looking forward to this

[rustedpopcorn]

Can’t wait! I would think 0.3% for both pools would make sense

[cryptochrome]

Awesome. This has my vote. Bring it on

[TopNotchInvest]

Lot to digest, and I am like 3-4 tequilas in. Viva Mexico! However, we need to get the liquidity up and running! Want to boost my APR, not planning on exiting my stake!

Will re-read again tomorrow, and post something of more value. Salute!

[Christ]

Sounds like a good way to go, and generally, 0.3% seems very fair while 1% still seems an acceptable exit fee.
As someone holding only ~1.6 sETH I would be thankful to have some sort of guide to assess if providing liquidity myself might be profitable for me - at least as soon as we’ve decided about the LP details.

[George]

@kiriyha

Things that we need to discuss and decide on:

1. Size of the swap fee: 0.05% (facilitates arbitrage trades, but fewer fees for LPs), 0.30% (best of both worlds), 1% (very handsome for the LPs but prohibits exit without a “redemption fee” which is not attractive). This needs to be decided for both tokens

If i understand this rigth, 0.3% is the current V2 fee and 0.05% is the incentive for stable coins. Since we want a narrow price range, wouldn’t 0.05% make the most sense?

All hail Dmitri!

[stanislavkozlovski]

This is an awesome proposal!

It is hard for me to argue against using Uni v3 - it seems like the perfect fit for this. It has the largest brand name, most liquidity and its functionality for capital efficiency fits like a glove for our use case.
Unless somebody comes up with a good reason not to, I vote for choosing Uniswap v3 as the DEX of
choice.

Risk vs Reward

Stepping back, I think the larger question we need to answer with this proposal is:

“What are the incentives for me to provide liquidity on Uniswap?”

If I understand the landscape right, I believe we need to answer this question well before moving on with liquidity, as whatever we choose first will have ripple effects on the long-term sustainability of the project.

I may be naive in saying this, but I believe the LP rewards from Uniswap need to be larger than the staking rewards I would get from sETH2 in order for me to provide liquidity on Uniswap long-term.

Otherwise, I am better off just staking sETH2 in the platform and hoping others solve the chicken-and-egg problem of liquidity for sETH2 so that the project grows. Note that in this scenario, I don’t need sETH2 to have liquidity, as I can just ask the StakeWise team to withdrawal my sETH2 into ETH for me when Phase 2 happens.

I can differentiate between 3 types of allocations that have different positions on the risk spectrum. Here is an example:

NON_RISKY < ----1) --------2)----------- 3) ------- 4) → RISKY

1. All in ETH
   1. reward: ETH price appreciation
2. 50/50 (or some other weighting) ETH/sETH2; no LPing
   1. reward
      1. ETH/sETH2 price appreciation
      2. rETH2 rewards from sETH2 staking
   2. risk
      1. if sETH2 fails, you lose 50% - you’re hedged against sETH2 failure via your ETH allocation (50% remains)
3. All in sETH2
   1. reward
      1. sETH2 price appreciation
      2. rETH2 rewards from sETH2 staking
   2. risk
      1. sETH2 failing - 100% loss
4. 50/50 LP ETH/sETH2
   1. reward
      1. rETH2 rewards from sETH2 staking
      2. trading fee rewards from LPing
      3. (temporary) you gain SWISE rewards from liquidity mining incentive
      4. ETH/sETH2 price appreciation
   2. risk
      1. sETH2 failing - you are not hedged against sETH2 losing peg. If it does fail, you incur impermanent loss (actual loss in this total failure scenario) – hence you risk your original ETH allocation as well (potential 100% loss)
      2. Uniswap failing - rug pull, protocol hack, etc etc. (potential 100% loss)
      3. temporary*[1]*, sETH2 losing peg results in impermanent loss (varying %)

[1] - note that the risk of StakeWise totally failing is: it being a rug, getting hacked or experiencing devastating slashings. Otherwise, there is no reason for sETH2 to not trade equally to ETH, as it is a 1:1 backing. Therefore, we can ignore impermanent losses.

Essentially, this shows that 4) LPing is riskier than 3) HODLing – but for taking on that risk you get the additional 4.1.2 and 4.1.3 types of rewards.

As a user, if 4) gives me less rewards than 3), I will never go with 4) – taking more risk for less rewards doesn’t work out.
To start with - 4) is likely to give more rewards than 3) due to some amount of people eager to move out of sETH2 (trading volume) and 4.1.3 - the liquidity mining program.
For the longevity of the project, we need to ensure that 4) without 4.1.3 is more profitable than 3)

As a project, we want everybody to do 4). Liquidity is the name of the game - without it, we don’t have as strong of a value prop.

The job of StakeWise is simple - we need (first and foremost) a good track record of staking (no hacks, exploits, slashings, etc) and secondly - a strong peg with lots of liquidity backing it. Do that properly and you’ll attract a slew of ETH holders looking to easily make extra money.
Do the former properly and you’ve got yourself a solid product. Do the latter properly and you’ve got yourself an outstanding product – liquidity incentivizes use of sETH2 in the wider DeFi ecosystem

Liquidity is a self-fulfilling positive feedback loop where the more you have, the more confidence outsiders get in the protocol and the easier it becomes for them to jump in (& out).

What I am suggesting is that:
1) we, as a project, strongly want to have sufficient and growing LP volumes for ETH/sETH2 (as well as rETH2)
2) a solid way to achieve this desire is to ensure the risk spectrum has the same order as the profitability spectrum – i.e the riskier it is, the more profitable it is.

It’s worth noting that there are varying forces pulling from both sides of the profitability spectrum – the more interest we get, the more lucrative LPing becomes; the more people stake ETH, the less lucrative pure sETH2 staking becomes.

If all goes according to plan (happy path) with StakeWise and Ethereum as a whole, sETH2 will become less profitable as sETH2 LPing becomes more profitable as time goes by.

If we are unable to ensure identical risk/profit spectrum ordering, perhaps sprinkling liquidity mining SWISE rewards to fill that gap may work in the short term until the varying happy-path forces even it out.

Questions

With that out the way, I have a few questions:

1. Can we change Uni pool settings once deployed? (if we can, perhaps no need to dive into this in such detail just yet)
2. Have we looked into how other competitors do this?
3. e.g do we know if Lido LPing is more profitable than staking? A summary of that would be helpful in picking what our strategy should be
4. Do we have reasonable expectations that LPing will be more profitable than pure hodling/staking?
   1. If no, do we believe that liquidity mining can flip that? If so, do we believe that we can sustain that for long enough until there isn’t a need for liquidity mining?

[dreth]

the StakeWise team changed course to deploying on Uniswap V3

YES

i) could achieve a stable peg between ETH and sETH2/rETH2 with ii) less liquidity than it previously needed, by incentivizing liquidity provision in as tight a range around 1 as possible. This would lead to iii) significant earnings for the LPs and iv) create a strong foundation for integrations of sETH2 and rETH2 tokens in the DeFi ecosystem.

YES

@kiriyha

Uniswap V3 is not the cheapest from the gas perspective when it comes to adding liquidity. It costs twice as much gas to add liquidity in V3 than in Balancer v2 (~560K units vs ~230K units) and swaps are some 50% more expensive (~200K units vs ~130K units). Still, this is better than Uniswap V2 and other DEXs out there.

This is true, but I dont think it’ll be a big deal in the long term, and now fees are pretty alright after the flashbots innovation.

@kiriyha

To mitigate the risk of de-pegging on the first day, we kindly ask that everyone who considered exiting from staking within the first week of liquidity pool’s deployment to anonymously state the amount they would like to withdraw via this form . Your response will help the DAO to plan ahead for attracting the right amount of liquidity on day 1 and ensure that the peg is maintained.

Awesome idea, this link should be published everywhere and users be reminded of this, everyone participating in staking with stakewise should be a member of this forum or at the very least keep up with it.

Let’s make sure everyone that wants to exit their position know that this is ongoing.

@kiriyha

Meanwhile, once we learn about the size of withdrawals, we as a DAO will need to discuss how much incentives to allocate to the sETH2/ETH pool. For now, the provisional amount is 0.6% of the supply per month as per this forum post.

Awesome!

@kiriyha

Size of the swap fee: 0.05% (facilitates arbitrage trades, but fewer fees for LPs), 0.30% (best of both worlds), 1% (very handsome for the LPs but prohibits exit without a “redemption fee” which is not attractive). This needs to be decided for both tokens

0.3% seems perfect to me. A discussion on which options to vote on should be done here in the forum before opening a governance vote on the fee %, I personally think 1% is too much, but as someone that intends to provide liquidity, I would obviously like more fees.

However, to be fair, it’s reasonable to consider that people shouldn’t be entering staking positions through a LP, they should be using the stakewise pool site to do so.

@kiriyha

Size of liquidity mining incentives for both tokens

I think this warrants a discussion for sure, I’d say that we need to incentivize these nicely, but without giving up too much of the SWISE supply as to not dilute the price too much.

As an additional proposal I think it would be nice to have an SWISE/ETH pool on Uniswap V3, even tho I wouldn’t like to seem like I’m suggesting that we need all the pools in the same protocol, I do believe that an SWISE/ETH pool (with equal liquidity mining incentives as the 1INCH/SWISE pool) on Uniswap V3 would probably get more use.

But, at the same time, this would also cannibalize the 1inch pool, so idk. If nobody else proposes this, I’ll make a forum post in the future asking the community.

In FULL support of this proposal, I love it, I love everything about uniswap, I think it’s the one project that best aligns with the values of ethereum and as a byproduct, the values staking platforms should follow.

[George]

@dreth

0.3% seems perfect to me. If we want to benefit LPs slightly more, 0.5% is good.

I thinkin you mean 0.05%?

[dreth]

@George

I thinkin you mean 0.05%?

nop! I mean 0.5% as a sort of middle point between 0.3% and 1%

I was wrong, this isn’t possible, so I’d go with 0.3% as what I think is best

[George]

I thought you could only choose between 0.05%, 0.30% and 1% in V3.

“Pool Fees Tiers#
Uniswap v3 introduces multiple pools for each token pair, each with a different swapping fee. Liquidity providers may initially create pools at three fee levels: 0.05%, 0.30%, and 1%. More fee levels may be added by UNI governance.”

Fees | Uniswap

“We expect low volatility assets (stable coins) will likely congregate in the lowest fee tier, as the price risk for liquidity providers holding these assets is very low, and those swapping will be motivated to bear the lowest swappers will be motivated to pursue a execution price closest to 1:1 as they can get.”

[dreth]

Thanks for the correction George, I’ll correct my initial reply to the main thread.

I actually for some reason thought it was possible to customize fees beyond that.

[George]

No problem. This is what the thread is for. Discussion to make an educated vote. Was hoping you we’re rigth and we could decide on other percentages, to be honest.

[ottodv]

First I agree that 0.3% tx fee seems to be best for now. 0.05% seems a bit low to make LPing interesting, and 1% seems too high for those trading.

However the problem with both pools seems to me to be that they mostly exist for people to be able to sell sETH2 and rETH2, the question I have is what motivates people to buy those tokens… or to hold them?

Regarding sETH2, one motivation to buy them could be to not have to wait for activation, but get tokens that stake immediately. For this to be interesting the pool itself should only start rewarding new deposits once they have been activated (which is not the case right now), this additionally prevents the pool APR from dropping when there is a lot of ETH being newly deposited. This benefit stays even after the merge of the beaconchain with ETH1, when withdrawing becomes possible. It’s a way to jump the activation queue, that will always remain valuable. In theory this should make sETH2 slightly more valuable than ETH.

For rETH2 is seems to me that once the merge has happened, there is really no reason for the market to exist as it can be simply withdrawn. Until then, my thought was to make holding rETH2 rewarding by simply giving holders a reward paid in Swise that is equivalent the the staking APR of sETH2.

Holding rETH2 would thus become equally interesting from a reward point of view, which should create demand for it. At this point, whether you convert rETH2 to ETH to restake or whether you hold it would end up being equally rewarding.

It might even be better to just have an sETH2/rETH2 market (instead of sETH2/ETH). From a liquidity provider point of view, whether you have one or the other is equivalent in terms of APR. (While ETH itself doesn’t have a reward).

Coming back to the sETH2 market, that one could be sETH2/ETH, here we could apply the more traditional LP reward OR we could reward just the ETH part in the pool with a Swise reward and APR again equivalent the sETH2 staking reward (which the sETH2 part accrues). In this scenario again from a liquidity provider point of view it doesn’t matter which one you have, both return the same APR.

[jonathanstrange]

Looking forward to this! Sounds like you have done excellent work.

So, will there be people wanting to buy sETH2, not just selling? I can see the reason for selling if you staked and now need liquidity.

There might also be a few people selling half their sETH2 to be able to put liquidity into the pool.

Or does it not matter if the trading is a bit one sided?

[George]

@jonathanstrange

So, will there be people wanting to buy sETH2, not just selling? I can see the reason for selling if you staked and now need liquidity.

Arbitrage and leveraged staking would come to mind for buying pressure.

[dreth]

@ottodv

However the problem with both pools seems to me to be that they mostly exist for people to be able to sell sETH2 and rETH2, the question I have is what motivates people to buy those tokens… or to hold them?

Arbing can be profitable!

[Quijote]

Just want to say…have not been active around the discord or forum much of late because IRL work is kicking my ass, but it brings a tear to my eye to see the engagement and thoughtfulness on display here. WAGMI vibes are strong.

[Selby]

The way I see it working now, there is almost no way that sETH2 maintains a 1:1 peg. The only incentive to buy sETH2 vs. staking directly through the platform is if it is discounted.

StakeWise could build it’s own market on the website, where people could list their sETH2 for sale, so new stakers could have a choice of buying existing sETH2 at a market price (discount) or staking directly. In this way new stakers would probably absorb the supply quickly.

The discount new stakers would get from absorbing the supply would increase their APY… there could even be an automated APY calculator showing the % return based on discount plus current rewards rate.

Stakewise would be acting as an escrow service and could take a tiny fee for each transaction.

Just a thought.