GEL/ETH G-UNI Liquidity Mining Proposal

Summary: This is a proposal centered around incentivizing liquidity for the (soon to be circulating) GEL token. Below we outline a Liquidity Mining incentives program for two G-UNI Pools on the GEL/ETH token pair (0.3% fee tier). Anyone who adds liquidity into either of these G-UNI Pools may stake their G-UNI tokens to earn GEL rewards. The G-UNI Positions on the GEL/ETH pair will have dynamically adjusted ranges using an automated strategy powered by Gelato, to always maintain deep liquidity in range as prices move. If passed, the liquidity mining program would begin a few days after the GEL public sale, and distribute 350,000 GEL over 4 weeks. It is likely that the program will be renewed for subsequent months but it seems wise to start with a short first period so we can adjust the incentives and automated liquidity management if needed.

Background & Motivation: For many new tokens a goal is to bootstrap sufficient liquidity so that buying and selling the token for other assets is easy and practical for market participants (buyers and sellers can match with each other easily, prices are not overly sensitive to individual buys and sells). The best way to do this (without paying centralized middle men to act as market makers) is to incentivize liquidity provision on a DEX (decentralized exchange) with a Liquidity Mining scheme.

Uniswap V3 is a new kind of DEX which addresses some of the shortcomings of prior DEX models by introducing concentrated liquidity: each LP can provide liquidity in a custom price range of their choice. This creates a much more capital efficient DEX, allowing for much deeper liquidity in tight price ranges and thus less price slippage for buyers and sellers. However, Uniswap V3 makes LPing more complex for the average user and it makes classical liquidity mining contracts unusable since each Position on a Uniswap V3 pair is now unique rather than a fungible token representing your proportion of the liquidity in the Uniswap pool. G-UNI is Gelato’s very own framework for making Uniswap V3 Positions fungible and for automated management of these fungible Pooled Positions on Uniswap V3.

Here’s how G-UNI works: anyone can add liquidity into a G-UNI Position on a certain Uniswap V3 trading pair and receive fungible G-UNI tokens (your proportion of the total G-UNI position). These G-UNI tokens can be staked in standard Liquidity Mining contracts to incentivize liquidity on the token pair. What is particularly unique about this is a project can now incentivize liquidity in a particular target price range or with a particular automated strategy for how the liquidity will adjust its range over time. We can showcase Gelato’s G-UNI system by using it to bootstrap liquidity for our own GEL token with an automated liquidity provision strategy of GEL/ETH G-UNI Pools. This should give the GEL token much desired liquidity right after the token sale and also be a good way to further the development and adoption of G-UNI across the ecosystem.

Specification: Below are the specs of the two incentivized G-UNI Pools:

G-UNI Pool 1:

  • Uniswap V3 Pair: GEL/ETH 0.3% fee tier
  • Initial Position lower bound: 1 GEL = 0.0001235 ETH (tick -90000)
  • Initial Position upper bound: 1 GEL = 0.0004099 ETH (tick -78000)
  • Buffer length: 6000 ticks
  • Buffer orientation: upper (GEL side of pool)
  • Staking Rewards: 250k GEL over 4 weeks

G-UNI Pool 2:

  • Uniswap V3 Pair: GEL/ETH 0.3% fee tier
  • Initial Position lower bound: 1 GEL = 0.0000372 (tick -102000)
  • Initial Position upper bound: 1 GEL = 0.0001235 ETH (tick -90000)
  • Buffer length: 6000 ticks
  • Buffer orientation: lower (ETH side of pool)
  • Staking Rewards: 100k GEL over 4 weeks

Recall that the two G-UNI pools are not static, these are only the initial ranges and they will evolve over time. The ranges will be automatically adjusted by a “manager” smart contract using an automated strategy to adjust the liquidity as prices move. To understand more specific details about the automated liquidity management strategy employed see this post.

Participants/Team: Gelato Core Developers will implement, anyone can participate

Allocated Budget: 350,000 GEL (to be distributed over 4 weeks as rewards for those staking G-UNI)

Milestones and Deliverables: Implementation of the automated manager and staking rewards contracts should be done before/by the token sale. There will also be a (simple) UI for the staking of GEL/ETH G-UNI tokens. The target launch date for rewards to begin dispersal would be roughly September 15th 2021.

Risks: There is smart contract risk (though the G-UNI core contracts have been audited twice, and have 25 million under management, without incident). There is also inherent risk in providing liquidity on Uniswap (impermanent loss), especially for new tokens which can have volatile price swings. Finally, there are risks caused by diluting the circulating supply with another 350,000 GEL. Liquidity Mining schemes can promote sell pressure for the reward token, since it is not uncommon for capital to farm the rewards and immediately “dump” them. That said, the circulating supply of GEL will be extremely low in the first few months so another 350k should not create an excessive circulating supply, and most of the rewards will go to those who already have GEL to provide to as liquidity, meaning token sale participants who have alignment with Gelato and are thus more likely to hold the GEL they farm in this scenario in comparison to other projects with LM incentives.

12 Likes

Although I appreciate that we don’t yet know the volatility of the GEL token but in my experience lower mcap tokens or those with higher volatility are usually in the 1% fee tier to help better compensate for the likely IL. What’s the rationale for the 0.3% tier?

3 Likes

Great proposal!

Some potential builds on that I’ve seen other liquidity mining programs use to mitigate the farm & dumping:

  • lock rewards on a vesting schedule, which can be vested early but by taking a 50% hit which is distributed amongst those still locked in the pool
  • when staking liquidity allow the user to specify a lockup period, the longer the longer the higher the APY

Thoughts?

3 Likes

Nice suggestions Paul :+1:

A lockup sounds interesting to me. Though if I were to decide, especially at the beginning of the mining program, I’d set a hard lockup period to secure a certain amount of stability and predictability during that highly volatile period.

2 Likes

Great proposal @kassandra.eth.

What’s the rationale for the 0.3% tier?

I also would be interested why we shouldn’t use the 1% fee tier, especially for starters. Could we quickly summarise the pros and cons here?

Regarding the distribution (250k rewards to Pool 1, 100k rewards to Pool2), will that actually result in the same amount of $ value positioned in the two pools below and above the current price? Because we would of course want the ETH pool to also support GEL sell pressure the same way the GEL pool will support GEL buy pressure.

Just want to make sure both sides are equally “liquid” so people can buy and sell in a liquid market.

Regarding the comments about vesting, I think it could make sense to have a lock up of the rewards in general, however we should consider that the LM period is fairly short for this initial program (4 weeks only). I would assume that, given current gas prices, it wont be worth for most users to claim their rewards before that anyways and thus they will probably leave the rewards at least until the initial period has passed regardless of a lock up.

If we now want to lock up LPs rewards beyond the period of the LM program, then we should also consider that the risk of LPing in a volatile market is quite high and thus receiving rewards, at least after four weeks, will provide these LPs with some sort of ability to hedge their positions and remedy potential liquidity issues.

However, I am definitely open for more concrete suggestions around this, especially providing examples of projects where this resulted in a positive effect for both holders and LPs. Do you have any of those examples @btt @freshaspect ?

1 Like

I actually had missed the line about it only lasting 4 weeks, in which case I completely agree that it’s probably not worth applying lockups given gas costs plus additional complexity that it introduces.

One factor to definitely consider is judging the amount of time for people can stake liquidity before the liquidity mining actually kicks in and managing the “hype” around this given that as we know if you’re a lonely ape in a nice rewards pool you can farm up a crazy amount of tokens! (with associated dumping risk).

2 Likes

Should give them some kinda NFT too!

Although I appreciate that we don’t yet know the volatility of the GEL token but in my experience lower mcap tokens or those with higher volatility are usually in the 1% fee tier to help better compensate for the likely IL. What’s the rationale for the 0.3% tier?

Yes this is a great question. There are pros and cons but I’m on the fence as well, frankly. To put it very simply, the pro of 0.3% tier for buyers and sellers is fewer fees and thus less perceived “price slippage” on their trades. In general the trend seems to be that lower fee pools get routed through more often (for instance people expected USDC/ETH to be a 0.3% pool mainly, but 0.05% has ended up getting more trading volume recently, even with less capital in that pool) so thats a consideration as well.

On the other hand, the main pro for 1% fee tier is low mcap tokens are extremely volatile and the risks for LPs are more severe. 1% tier incentivizes LPs to try and make a return on these more illiquid tokens. The rebuttal to this is that the pools have incentives that are helping to cover the LP risks already - the majority of a staker’s return will likely come from LM rewards not from fees earned on trades.

Another thing to expect (we saw this with INST/ETH G-UNI launch) is “mercenary LPs” (i.e. people who LP on their own not through LM scheme) cropping up on the opposing fee tier, whichever we choose. Instadapp does not have the automated range rebalancing on their incentivized G-UNI pools on 0.3% tier, so when the pools would slip out of range the 1% pool with mercenary liquidity would end up taking over all the trading. We’ll be using the strategy to stay in range so we should have a better time in this regard - still its another subtle aspect to consider.

I think one major question is which is more important: providing further incentives/protections that mitigate the risks of impermanent loss for the LPs? Or giving traders a better experience (less fees to pay) and incentivizing more trading volume?

It’s also worth thinking in slightly longer time horizons (not just the proposed 4 weeks) - in the medium term what fee tier do we think liquidity would settle on? In the absence of the LM program where would the liquidity pop up and which pool would take over - the 1% or the 0.3% ? (Maybe we can take a look at some real world similar examples on Uniswap V3 to confirm).

It would be quite easy to amend the proposal to a 1% fee tier with just very minor adjustments (tick bound have to change slightly to round to the 200 tick tickSpacing of the 1% tier). Definitely open to both options here.

Great proposal! I think 4 weeks is a good test-run to see how the liquidity mining program pans out and from there we can adjust the fee tier, implement lockups, etc.

Hi guys,great LP proposal with this double pool.My question here is if it is only for Gelato devs or the community can enter as well?Because i am very interested entering the pools.

1 Like

Everyone can enter of course.

1 Like

I aksed about it because on initial proposal above it says:

Participants/Team: Gelato Core Developers

It was a bit confusing. Thanks answering Hilmar

2 Likes

Great proposal. GEL mining should be exclusive to those who get rid of the GEL they already have in order to achieve descentralization.

edited the post, to make it more explicit thanks.

1 Like

Ari can you please elaborate a bit on how this proposal works, i mean both pools work in tandem in order not to have big IL as passive rebalancing correct ?The first when GEL raises and the second when GEL drops so the buffer orientation changes.Am I understanding this right?Therefore best practice will be to provide 50-50 GEL-ETH on each one of the pools, or just one token of the two for each pool(GEL on 1st,ETH on 2nd)? Thanks.

1 Like

This proposal makes no sense for small fish. Only big token holders can participate and get some extra tokens as incentive because of high gas fees. And as we know after unsuccessful tokensale most of the tokens are in one person/group hands so they can manipulate with the market and get even more tokens by participating in these kind of events.

I support this! Perhaps go for 1% fee and adjust accordingly. If 0.3% is a good call to start/test I wouldnt mind. Nevertheless, this is good for everyone.

Revisiting Locking

Given the results of the token sale I do wonder whether lock-based APY boosting takes on increased relevance. In my mind, those who most strongly believe in the long term potential of Gelato and are therefore willing to lock up their tokens for longer should have access to enhanced rewards.

Therefore is there potential here to:

  • Drop the baseline reward rate a lot and extend it out over a significantly longer time period

  • Add a simplified CRV/Badger/Adamant booster locking mechanism such that LM rewards are boosted dependent on: (1) how long you lock GEL up for (2) The ratio of how much you have locked vs. how much you have staked - the longer you lock it up and the lower the ratio of locked:staked the better the APY that you receive

I believe this could help act as a filter to ensure that it is true Gelato valuing participants who will reap the best rewards. Arguably it could also be applied to voting power as well - in the same way as veCRV does.

2 Likes

As one of the many whitelisted users of the Dolphin pool that was not able to participate in the initial sale, I would support a locking solution that favour long-term holders.

For the ones who got screwed in the sale process and decided to buy the GEL token on the open market at 6-7x the sale price anyway, because we believe in the long term potential of the project, this would be a (very marginal) mitigation.

On the other hand, we could also hope that a non-locked liquidity mining program will result in strong dumping allowing the many who got left out to buy at a level closer to the initial sale price, but that is probably just wishful thinking (damn… I try to move forward but I am still so bitter about how that whole process has been managed).

On a separate note, it would be very nice if the program could be launched alongside a Gelato powered auto-compounding solution (i.e. automatically harvest GEL rewards and re-invest them in the liquidity pool). That would also be a way to show some support for the smaller holders for whom it would not be economically viable to do that on their own due to gas cost, while also avoiding to have to use and pay a fee to another protocol such as Harvest to do that.

1 Like

Hey guys, how about providing single sided liquidity in pool? May be a pool on Arbitrum eg. Dodo or something similar on other protocols such as bancor (for eg). Can this be executed and deployed?