The time has arrived for Sushi to discuss new tokenomics. An optimal token model should incentivize liquidity and promote decentralization. In addition, it should provide more equitable governance and sustainable economics. We propose a new token model that will help Sushi achieve these goals and, in addition, help bolster Treasury reserves to ensure continual operation and development.
The proposed token model is robust and offers several tried and true methods for promoting value and utility. In addition, we institute novel concepts to help promote maximum value for all stakeholders. Below is a brief tl;dr of the proposal for your convenience. We welcome your questions and feedback to help us discuss the model over the coming weeks as we plan for a productive 2023.
Liquidity Providers (LPs)
LPs receive fee share from the 0.05% swap fee, with the bulk of the fees going to the pools producing the most volume.
LPs can lock their liquidity to earn boosted, emissions-based rewards. But, they lose the rewards if removed before maturity.
xSushi
xSushi receives emissions-based rewards in time-locked tiers. Longer locks receive more rewards.
Time locks are “soft locks”; you can remove your collateral before maturity but lose all rewards.
Time locks auto-renew and compound in a new time lock.
xSushi receives no fee share, only emission-based rewards.
Burns
We use a variable percentage of the 0.05% swap fee to buy back Sushi and burn it, permanently removing it from circulation. The percentage changes based on the total time-lock tiers selected.
If xSushi or LPs seeking emissions-based rewards prematurely withdraw their collateral from their time locks, the rewards get forfeited and burned.
Because time locks get paid after maturity, but burns happen in “real-time” when a large amount of collateral gets unstaked before maturity, it has a sizeable deflationary effect on supply.
Locked Liquidity
We will use a portion of the 0.05% swap fee to lock liquidity for price support.
Emissions
We introduce a nominal 1-3% APY for emissions.
Our goal is to have a supply within a predictable margin that balances with buy/burns/locks.
Although we provided some of the highlights of the token model we’re proposing, please read the complete proposal that we put together, linked below. It is a comprehensive overview and will likely answer many questions. Please post any unanswered questions in the comments below.
Ultimately, our goal is to help provide long-term value for token holders and liquidity providers without extractive qualities and detriment to either party. This new model promotes a holistic and sustainable value that scales with the DEX while boosting deeper liquidity to help revitalize the Sushi ecosystem.
Technically, if xSushi rewards become emission-based, the accrued xSushi/Sushi bar reward burn for withdrawing collateral before stake maturity isn’t removing any circulating supply. Or did I get it wrong?
Also, for the burns stemming from collected fees, since they occur before rewards payout, how would you mitigate front-running?
Key challenges that Sushi faces, even with the revised tokenomics:
(1) Competition against DEXes like Uniswap that have 0 protocol take rate.
(2) DeFi TVL exiting for juicier yields in TradFi.
(3) > 90% of treasury assets are in [$SUSHI]
This proposal definitely improves upon the current unsustainable emissions and incentive alignment.
A few points to add:
The reward burn for liquidity providers is “punitive” and would encourage LPers to migrate to other DEXes. The reward burn also opens up space for a secondary market for max-timelocked LP positions, which probably leads to value extraction by some third-party who buys locked LP positions at discount to dump the rewards at maturity rather than the intended SUSHI token deflation.
I’m not a fan of “punitive” mechanisms but open to linear or exponential time-decay for how much the pending rewards are burned. Alternatively, a similar outcome could be achieved where we assign reward weighting based on TVL*time (akin to GMX’s multiplier points or TraderJoe’s veJOE), so that LPers are incentivized for long-term participation without it feeling “punitive”.
To protocols that list their tokens on SS, there should be a more direct way for them to incentivize for more liquidity. Of course, the protocols can create their own farm where users stake Sushi LP tokens, but perhaps a better way would be to use a ve-token/gauge/bribe system where protocols can bribe veSUSHI holders to increase emissions into the protocols’ own pools. In this gauge system, veSUSHI holders benefit from receiving the bribes, which increases the yield of SUSHI tokens.
One of the mentioned stats is that few xSUSHI holders are LPers. I do not think this is necessarily a bad thing. But if the goal is to increase the overlap between xSUSHI and LPers, one mechanism to explore would be giving increased emission weights to LPers who also stake xSUSHI (akin to Camelot DEX and VVS Finance).
disclosure: I am an user of Sushiswap as well as a contributor to several defi projects, most relevant one for this discussion is Redacted Cartel, which created the Hidden Hand marketplace for bribes and incentive alignment. My opinions are my own.
Hi Carnation, I appreciate your thoughtful post & suggestions. I hoped that our model proposal would promote contributions of ideas to make it the best it can be, & we’ll discuss the ideas you presented. Also, if you’d like to chat via DM, don’t hesitate to contact me on Discord or in forum DM. Thanks!
This post highlights why Tokenomics needs to link to Sushi growth strategy.
Sushi LPs, xSushi, Sushi holders and Users are all interlinked.
perhaps a better way would be to use a ve-token/gauge/bribe system where protocols can bribe veSUSHI holders to increase emissions into the protocols’ own pools. In this gauge system, veSUSHI holders benefit from receiving the bribes, which increases the yield of SUSHI tokens.
Personally I like the ve-token/gauge/bribe system as it can be can align incentives for governance voting, locked sushi staking as well as encourage third party LP/voting participation. Main downside is complexity and technical effort to tailor to specifics of Sushi project.
Disclosure - xSushi holder and have created and participated in Sushi LP pools.
I probably don’t get it, but currently LPs got a 0.25% fee and 0.05% goes to xSUSHI (well, treasury after Kanpai). LPs also get some emissions. Are we now giving some of the 0.05% to the LPs for locking their liquidity and taking away their emissions?
LPs still get 15% of emissions. Please note the column in the sushi bar called LPs, which will going forward be the emissions distro. The other categories for emissions are Boosts (targeted LPs, and Swappers), and burn/LLP.
I think it’s important to note, volume makes the monies. We are addressing this by both tokenomics and product stack improvements. The goal is for them to work hand in hand to increase our attractiveness for liquidity providers and encourage behavior that generates both income, and a positive impact on the ecosystem.
Sorry, you totally lost me there… let’s make this a bit simpler…
From a $1000 swap, there’s a $3.00 fee charged to customer making the swap. Currently $2.50 goes to the LP and $0.45 goes to xSUSHI and $0.05 goes to the treasury. The LPs also receive SUSHI in some pairs (farms).
The biggest problem that Sushi has is emissions and the lack of volume to support it. Here are some of my thoughts of the new tokenomics and some additions.
“Less than 2% of xSushi stakers provide liquidity in any Sushi pool.”
Why must token holders of SUSHI be punished for not providing liquidity? Sushi token holder are doing their part by buying into the token. Must an apple stock holder buy apple products? The sushi holder and the liquidity providers are two different users and shouldn’t be clumped into one. If LPs are not be rewarded fairly thats a different problem to be solved
“Sushi’s new token model rewards LPs more for locking their stake longer.”
This is a great idea. LP need to be providing actual value. This reduced farmers from jumping from one protocol to another.
Right now 0.35% is given to LPs and 0.05% is given to xSUSHI stakers, the new tokenomics mentions this will be dynamic based on the pools making it more “competitive”. Reducing fees will make the pools much more competitive and generate more volume.
What I don’t see why the 1/6 ratio should be changed. If we don’t think the 1/6 is fair to LPs? Sure lets change the ratio but I would like to see it fixed rather than kept dynamic.
Remove non-sticky staking terms; introduce time-locked tiers for rewards.
Unlocking before maturity burn the rewards accumulated.
All this is great and I’m all for this.
Substitute trading fee revenue rewards with emission-based rewards.
Not a fan. Emission should be based on buybacks. No more sushi diluted. Thats how we got here.
Vol distro is a reward for swappers who meet a specific volume minimum and target to maximize a refund of the fees paid, which allows high-volume traders to negate fees.
Why do this? Lets the swappers pay the fees. Protocols like 1inch handle this. We will get fees for the capital effiency we have.
locked liquidity get added to a Sushi LP pool by converting half of the amount to Sushi and the other half to the base pair selected for the pool.
This does create a buyback model but this reduces liquidity from a single pool.
Proposed Additions:
Sushi has exposure and brand value. Projects create farms using their own reward tokens on their own website. We should have a Build-Your-Own-Farm feature which will charge protocol to create their own farm with their own token. This will generate revenue, reduces inflation, generates TVL, bring new tokens to have their base liquidity on Sushi.
Only Sushi-based pairs should be eligible to receive sushi emission outside of bluechip LPs (ETH-USDC, ETH-BTC etc). xSushi holders and vote to give emissions to those pools. Including bribes can also add to protocol revenue.
Sushi should also reach a specific liquidity as decided by them for blue chip LPs. We should start out with farming, slowly buy the liquidity (Protocol owned Liquidity) while reducing the block rate and once reached set the block rate to 0. No point in rent when we can buy.
Burns don’t provide anyvalue instead of pumping price. It is the same than Apple stock buyback program in which shareholder are rewarded of keeping stock (Pancake Swap is also buying back CAKE , price of CAKE is similar than Sushi).
The cash could be used in a ration 60% for Sushi DAO (development, project) and 40% in a form of dividend to token stackers in veSUSHI (1 to 4 year lock).
veSUSHI give LP provider increased yield by more fees for you less to the governance…
I prefer dividend and cash from Sushi rather than pumping price. ( I need cash no a higher token price, which I have to sell)
Hi Cryptoinvest. Our model’s buyback and burn mechanism contributes to a stable supply or focuses less on price because that is driven as much by macro sentiment and environment as the project’s fundamentals. However, it should produce a positive effect, yes. But, the goal is to target a predictable token supply by burning an amount that scales with volume increase alongside a static emissions APY.
I’m afraid I have to disagree. Sushi’s problem is how poorly it allocated and spent its emissions. We’ve produced dashboards from data mining, with Flipside’s assistance, that showcase how Sushi spent its emissions to attract TVL, which occurred at a high cost to Sushi’s holders (dilution) and the protocol’s sustainability. IMO, Sushi can manage a low APY on emissions if the incentives mechanisms properly align with the business goals of the DEX. However, Sushi is no longer in the liquidity bootstrapping phase and needs to focus on a sustainable business model, which the team has put together and currently building the stack to support.
The punitive/punishment argument addresses the problem with an incorrect mindset. For instance, what is the goal of holding a token? What is the purpose of incentives? What is the preferred behavioral outcome of the game theory you promote in your token economy? For Sushi, we need to attract LPs to allow traders to swap, producing volume and fees. These two metrics, volume and fees are the most important to an exchange, whether a DEX/CEX. Sushi holders should get something that gives them utility and ROI but not at the expense of the principal contributors carrying the most risk in the business model, which is an economy of scale. When you begin to think of what Sushi’s token economy could look like, our model makes the most sense to address the current model’s shortcomings and how it prevents longevity to the business model now that Sushi is no longer bootstrapping growth.
Glad to see you agree with the LP focus of our model. We aim to give LPs preferential treatment, scale TVL, and introduce AMM feature parity, aggregation routes for increased volume and fees, and dynamically priced pools to help scale and increase competitiveness.
Why fixed ratio vs. dynamic? Dynamic allows us to address changes in the marketplace quicker. Glad you like the non-stick term/time-locks and the burn mechanism.
Emissions-based rewards give single-sided staking to Sushi holders. But your points are received, and we can discuss further internally to see how our model can improve with your suggestions.
I like your proposed additions. I’m happy to add these to our internal discussion and see if we can accommodate these kinds of suggestions where they can improve this proposed model.
3$ total fee, 2.5 to the LP, and .5 split along the distro schedule for LPs.
The emissions is split in the distro schedule for sushi bar.
Everything is included in the attached image. Each category gets a %, based on the locked tier chosen.
