February 27, 2024
Sperax DAO launched Demeter in November 2022 intending to become the liquidity engine for the entire Arbitrum ecosystem. The motivation was simple, Demeter was to become a one-stop platform for both liquidity providers (LPs) and protocols looking to incentivise liquidity. We launched Genesis Farms in partnership with GMX, FOREX, and PLS to name a few. Within 2 months of the launch, the TVL of all the farms combined shot up to a whopping USD 2 Mn, most of it being stablecoins.
We later launched the Sperax Gauge in February 2023 to decentralize the reward distribution process. But we had to roll it back in and close the Gauge in November 2023 based on SIP 58. This made us go back to the drawing board and talk to our consumers. We identified the major pain points that both liquidity providers and protocols were facing in terms of using Demeter as a platform for liquidity management. After putting in a lot of thought, we will soon be launching Demeter 2.0. It will have some major upgrades to our previous design and user experience. We will be launching a brand new reward distribution mechanism with Demeter 2.0.
One of the major concerns that resonated with most of the protocols who wanted to incentivize liquidity on Demeter is that they were unable to match liquidity with their incentives. Simply speaking there wasn’t any guarantee that 1% increase in incentives would result in more than 1% increase in liquidity. To solve this problem we came up with the idea for a different kind of Demeter farm. We are currently calling it the fixed APR farm, however, the APR is not really guaranteed or fixed but the APR is going to be more consistent than the current model. The reward token managers will be able to set an APR for their reward token based on the TVL of one or more assets in the farm. The reward admins would be free to update the APR.
Let us consider the example of the GMX/USDs farm. In this case, the Sperax protocol would like to incentivise USDs liquidity with SPA tokens. Whereas the GMX protocol would like to incentivise increased liquidity for the GMX token with some other token. It is in the interest of the protocols to only incentivize liquidity for their own tokens. You can also think of other examples like SPA/ETH where the Sperax protocol may want to incentivize liquidity based on the value of the paired asset ETH and that too considering ETH as the base asset for calculating the APR. This kind of farm would work very well for assets that are correlated to ETH.
Key features of the new Fixed APR Farms:
Protocols can incentivise farms based on the TVL contained in one or more assets in the farm. The base TVL value for calculating the APR can be in USD terms or other assets like ETH.
The reward distribution rate would be configurable and can be updated by reward token managers. Reward token managers can set up an upper limit on the total number of rewards they wish to distribute daily.
Liquidity providers (LPs) can view the rewards budget left in the farm and how long it will last.
The base tokens upon which the rewards can be distributed will be a select list of tokens in the beginning, these tokens would be whitelisted by the Sperax DAO. The primary criterion is that the tokens should have a trusted oracle. (e.g. USDC, WETH, USDT, USDs etc.)
Reward tokens can be added by the protocols as long as the reward tokens have an oracle. It will be left to the users to decide whether they trust the oracle or not. Certain tokens which have a good track record will be highlighted on the Demeter front end.
Some key updates to the previous versions of Demeter:
Integrations with Camelot V3.
Introducing a farm expiry date and a subscription model for protocols to keep maintaining the farms. We found that in a lot of cases protocols and LPs lose interest in some farms and it makes sense to keep an updated list. Please note that the farm contracts will always be accessible for claiming of reward tokens. The front end will continue to support expired farms for 30 days after their expiration.
Special status for SPA or USDs has been removed to make the design more scalable and compatible with other chains where SPA and USDs are not present currently.
LPs will be able to add or remove liquidity while staked in the farms. LPs can benefit by not having to do multiple transactions to add or remove liquidity to already existing staked positions.
An active liquidity check has been added to farms to make sure that only active liquidity is being incentivised. If the liquidity goes out of range, the farm will acts similarly to a paused farm. The farm will not emit any rewards and locked LPs are free to remove their position without having to go through the whole cooldown process.
Apart from this we also interacted with a lot of our users to understand the pain points faced by them while interacting with the Demeter frontend. The majority of the users complained of slow front-end data loading. To counter this issue we will launch dedicated servers where the data will be cached. This will go live in the next version of the front end. Apart from that, we would also be introducing a dark mode for the website. We will bring in advanced analytics, like impermanent loss detection into the portfolio section of the front end.
With these updates, we would be targeting Arbitrum’s Long-Term Incentive Pilot Program (LTIPP). All protocols launching farms and pairing with USDs will receive a fixed rate of ARB tokens as additional incentives in case the LTIPP goes through. Demeter is already permissionless and anyone can launch farms, but to receive the additional emissions the protocols will have to go through the governance process. For the LTIPP we would propose to distribute ARB tokens against the base liquidity value of USDs on Camelot v3. Any protocol can open a farm and incentivize liquidity for their token against USDs. Once approved by governance, ARB will be distributed at a fixed rate of 5% on the TVL of USDs in that farm.
We want to build deeper relationships with other DEXs on Arbitrum first but we do hope to have the protocol up and running on multiple chains in the coming future. We are actively testing our contracts on Polygon, zkSync Era, Linea, and Base.