IDLE staking update

[Biaf]

Summary

A proposal to discuss an update of the current IDLE staking module.

References

• Staking + Tranche Strategies Combined governance post by @Falcone - Jan 2021
• Idle Staking and Voting governance post by @tom - Feb 2021
• Staking implementation governance post by @Salome - Feb 2021
• Single $IDLE staking governance post by @Salome - May 2021
• Staking Reward Token governance post by @8bitporkchop - Jun 2021

Background

In 2021, the Idle DAO community jointly shaped the structure behind IDLE staking.

Staking is the process of locking up cryptocurrency tokens to help secure a network or smart contract in exchange for rewards and passive income.

Multiple ideas were discussed, mainly:

@Salome

1. Stake $IDLE for a reduction in performance fees. […]
2. Stake $IDLE to adjust the liquidity mining distribution of idleToken pools via voting and get boosted $IDLE rewards. […]
3. Stake $IDLE to receive a share of the protocol fees. […]
4. Staking a certain amount of IDLE in order to boost voting power on snapshots. […]
5. A combination of all the previous 4. […]

And following Leagues’ suggestion, the DAO approved the fee-sharing mechanism (option 3) first and later the Gauges system (option 2).

@Salome

Leagues Recommendation

Following all the research summarized earlier, the League Committee, therefore, suggests that Staking might start with option 3 (‘Stake $IDLE to receive a share of the protocol fees’) using Curves’ time-weighted staking model and later add option 1 and 2. This would allow Idle to use a battle-tested staking framework in a time-efficient manner, with added innovative features later to build an even bigger utility for $IDLE after.

Recently, Leagues, spurred by the post of @matus.eth, started brainstorming again on the topic with the goal to increase our governance token use cases

@matus.eth

Here is a couple of ideas as to what the IDLE token utilities could be that came to my mind:

1. Staking IDLE tokens in exchange for access to premium features on Best Yield, e.g. users who have staked at least x IDLE tokens or at least y $ value in IDLE tokens get priority access to new pools, get priority when redeeming idleTokens during congested periods, etc., this could be further brainstormed with someone who knows more about the product and its roadmap.

2. Safety Module backstopping e.g. the senior tranche on Yield Tranches, i.e. users stake IDLE tokens in a smart contract called Safety Module and earn yield (coming from fees and/or newly minted tokens), hereby increasing security of the deposits in senior tranches in cases of black swan events/hacks, etc.

3. Staking for gas fee rebates, e.g. users who have at least x IDLE tokens staked get their gas fees rebated when interacting with the protocol’s smart contracts.

Rationale

At the time of writing, the number of staked IDLE is 1.6m (source) vs a circulating supply of 5.4m (source), i.e. ~30% of IDLE staked. The ratio of staked IDLE vs the total IDLE supply is ~12%, locked for an average of 1.8 years.

Leagues are aware that the utility of IDLE staking should be revamped. It’s then worth discussing some old and new ideas coming from the community to have a clear view of the available options to update the current model.

Proposals

1. Pause the staking reward program for stkIDLE holders

Currently, stakers are rewarded with a mechanism that distributes half of the fees generated (in IDLE) by the Senior Best Yield strategy. This mechanism has given a value-capturing feature to IDLE tokens, but on the other hand, it has fallen short of sustaining the token price through the buyback mechanism setup.

@Biaf

Since January 2021, Idle is distributing $IDLE rewards to stakers with IIP-10. Based on @8bitporkchop’s work on $IDLE staking, the protocol fees have been split as follows

• 50% to Idle’s stakers
• 30% to the Idle’s Rebalancer
• 10% to the Fee Treasury
• 10% to the Smart Treasury

With IIP-18 of January 2022, Idle DAO voted to deprecate Idle Smart Treasury and increase the percentage of fees flowing to the Fee Treasury. The January 2022 split currently in place is

• 50% to Idle’s stakers
• 30% to Idle’s Rebalancer
• 20% to the DAO Treasury

We should also note that the fee-sharing contract is immutable and based on Sushiswap v2, which has no liquidity, hence buying IDLE there, is sub-optimal due to high slippage costs.

Additionally, due to regulatory uncertainty, the current staking rewards do not align with the not-for-profit structure of the soon-to-be Idle DAO LLC, thereby increasing the complexity.

For the above reasons, Leagues suggest pausing the current staking rewards and instead, introducing new use cases for IDLE stakers.

2. Introduce a performance fee discount for stkIDLE holders

@Salome

Stake $IDLE for a reduction in performance fees. A possible method might be to use a logarithmic formula where staking a percentage of all $IDLE (to be determined) would result in a fee reduction (to be determined)

@matus.eth

Staking for gas fee rebates, e.g. users who have at least x IDLE tokens staked get their gas fees rebated when interacting with the protocol’s smart contracts.

Idle’s user base is various and it could be strategic to consider both retail and institutional LPs’ interests when setting up the discount for IDLE stakers. @Teo already suggested an interesting solution

@Teo

Fee rebate: that’s the option I like more as it’s well-connected to the usage of the product suite and adds utility to it. As @william suggested, we can add a plain fee rebate/discount on the performance fee, so could be a good fit both for large LPs with fee discounts and for small LPs with gas fee rebates.

This fee discount feature fits with the current staking structure and shouldn’t imply big changes. Only new deposits, after the ratification of these changes, would be applicable. We may start to manually manage the discount and later move to an automated infrastructure. The discount would come in the form of backward distribution of IDLE or underlying tokens.

We could link the discount percentage to the amount of staked IDLE at the time of deposit conditional also to the size of the deposit itself. The table below shows, for illustrative purposes, different tiers examples with the corresponding discount applied (not yet considering deposit size).

stkIDLE linearly decreases from the lockup date to the end date. The maximum locking period is 4 years.

Tier	stkIDLE balance	Fee discount	Performance fee
I	≥ 10,000	10%	14%
II	≥ 50,000	20%	12%
III	≥ 100,000	30%	11%
IV	≥ 500,000	40%	9%
V	≥ 1,000,000	50%	8%

3. Introduce early access to new pools for stkIDLE holders

@matus.eth

Staking IDLE tokens in exchange for access to premium features on Best Yield, e.g. users who have staked at least x IDLE tokens or at least y $ value in IDLE tokens get priority access to new pools, get priority when redeeming idleTokens during congested periods, etc., this could be further brainstormed with someone who knows more about the product and its roadmap.

As suggested by @Teo and @Davide this gating feature can be more applicable to the Yield Tranches that live on the equilibrium between Senior and Junior liquidity to achieve the best efficiency

@Teo

stkIDLE for premium features: this is interesting and could work well for new Yield Tranches that get released over time – would be useful to test if there’s interest for this kind of feature among the current customer base

@Davide

Access exclusive pools via IDLE staking. Idle released 10+ tranches in 2023, and many others are in the pipeline. These products target sophisticated investors (e.g. Hedge Funds and HNWI), providing outperforming results compared to other market solutions. The priority access would amplify economic benefits for LPs, thus staking-based access would reward LPs that are also protocol owners.

This approach would particularly fit leveraged strategies, where the max return is usually achieved once a few m$ is deployed.

The stakers’ addresses are available on-chain and can be used to grant access to new Yield Tranches launched. We will set a specific ratio of IDLE staked to deposit 1 underlying token in the strategy, i.e. 0.1 stkIDLE to deposit 1 USDC. The gating condition would be time-related, i.e. clearing after two/four weeks since the pool’s release.

4. Update Snapshot’s voting power for stkIDLE holders

Currently, 1 stkIDLE is equal to 1 IDLE on temperature check voting. This ratio though, does not adjust to the stkIDLE linear decrease, hence considers stkIDLE as staked always for the maximum amount of time, i.e. 4 years. We could then set the ratio differently on Snapshot say for example 1 stkIDLE equal to 4 IDLE or 2 IDLE.

Next steps

We would like to gather the sentiment of Idle DAO toward all the proposals listed:

1. Pause the staking reward program for stkIDLE holders
2. Introduce a performance fee discount for stkIDLE holders
3. Introduce early access to new pools for stkIDLE holders
4. Update Snapshot’s voting power for stkIDLE holders

This post will be open for discussion to collect any feedback and adjustment to the proposals outlined, and only after it will be formalized in a new post and move to the temperature voting phase.

[william]

Thanks for the detailed post @Biaf

Regarding the four points:

1. I agree with this with no reservation, especially considering that Sushi v2 has very low liquidity causing all the buybacks to lose on slippage. It can of course be reactivated in the future but the program should be redesigned from the ground up imo.
2. This is a nice addon, but there should be some limits, as you mentioned, on the deposit size probably otherwise the discount could cause a big ‘loss’ in fees accumulated (and in runway) for the DAO and its contributors, especially for the lower tiers. Eg if some instituitional depositors is willing to deposit 5-10m it would only need a relative small amount of IDLE to have a big (in $ value) discount for fees. It would be interesting to give back the fees in IDLE but will the IDLE be taken from the ecosystem fund or acquired from the market?
3. That’s good especially for some strategies that can have an high apy but only with a limited TVL
4. I agree on giving stkIDLE holders more voting power off-chain, but I’m not sure I get the exact details of this

[Teo]

Great proposal @Biaf – about the next steps:

1. All things considered (current stage of Idle DAO, compliance uncertainty, and high slippage for buybacks), I agree with it – it’d allow to build up and diversify the treasury for Idle DAO and open the possibility to invest in activities to grow the ecosystem
2. I really like this as it’s directly tied to the product suite and creates a ‘premium’ version of it for stakers – agree with @william that we should think more about the discount model to make the fee rebate proportionate to the staking cost (and I’d expect that we would review it in the future with different market conditions); regarding the reward token, if we look at other examples in the space, dYdX is distributing directly DYDX tokens as fee discount rewards – I think we can decide once the program is up and running if we take the rewards from the Ecosystem Fund or with market buybacks (likely depending on the Idle DAO’s MRR)
3. Nice addition and adds more value to the ‘premium’ feeling of being a stkIDLE holder – definitely a good fit for strategies that have an intrinsic max capacity in terms of TVL (leveraged/option-based starts)
4. Agree with this, but we should discuss more the actual change in the voting power, it’s something already counted in the approved calculator for off-chain polls, and starting from there we might think about the multiplier for stkIDLE voting power.

[Davide]

Thanks @Biaf, great recap of potential follow-up plans to increase token utility.

@Biaf

4. Update Snapshot’s voting power for stkIDLE holders

Currently, 1 stkIDLE is equal to 1 IDLE on temperature check voting. This ratio though, does not adjust to the stkIDLE linear decrease, hence considers stkIDLE as staked always for the maximum amount of time, i.e. 4 years. We could then set the ratio differently on Snapshot say for example 1 stkIDLE equal to 4 IDLE or 2 IDLE.

Regarding this point, stkIDLE tokens are not equal to 1 IDLE. The formula implemented in the approved framework is a proportion between how many IDLE tokens are staked and the number of circulating stkIDLE.
For example, 1m IDLE locked and 500k stkIDLE means the average locking period is 2 years. As a result, each stkIDLE is backed by 2 IDLE. With 1m IDLE locked and 250k stkIDLE, the average locking period becomes 1 year and each stkIDLE is backed by 4 IDLE.
The ratio 1 stkIDLE = 1 IDLE can be reached only if all tokens are staked for 4 years, and represent the minimum theoretical exchange rate.

Despite that, we can eventually consider increasing the weight associated with stkIDLE by adding an ad-hoc multiplier. That bonus weight would enhance governance power for those that stake the tokens.

In parallel, the latest Barnbridge proposal about xBOND is worth to be mentioned.
The proposed model uses the industry-standard veBAL to merge staking with AMM LPing, and combines features in a novel way - like multi-AMM support and a GMX-like time-weighted multiplier.
By routing funds across different DEXs, the DAO is able to execute a more dynamic liquidity strategy and allocate assets in pools with different logics (Uni v3 ranges, 80/20 Balancer pool, 50/50 Sushi pools…).

[Biaf]

Cool guys! Feedback appreciated @william, @Teo.

There is still probably some work to fine-tune the 2nd and 4th proposals:

• Introduce a performance fee discount for stkIDLE holders
• Update Snapshot’s voting power for stkIDLE holders

Definitely worth exploring more the hints coming from @Davide regarding the Barnbridge example.

Regarding the 1st proposal, the DAO already knows the setup of the current stkIDLE distribution program and should be ready to take a decision on it. I would start then by opening a temperature vote on

• Pause the staking reward program for stkIDLE holders

After sharing further details, we will follow the same voting approach with the other proposals.

---

Temperature check voting

The voting phase to pause the staking reward program for stkIDLE holders starts now

• For - Agree with pausing the staking reward program for stkIDLE holders
• Against - Disagree with pausing the staking reward program for stkIDLE holders
• Discuss more - Discuss more the pause of the staking reward program for stkIDLE holders

Poll for $IDLE holders: HERE
Poll for $IDLE stakers (stkIDLE holders): HERE

Polls will close on 2023-06-15T14:00:00Z

The final $IDLE voting weights will be calculated using the approved calculator .

Looking to delegate your vote on Snapshot? Follow the official Snapshot guide.

Let your vote count!

[Alexintosh]

Hey everyone,

First time poster while long time Idle enjoyooor and user, I wanted to share my 2 cents on the subject.

I agree with the statement that utility of IDLE staking should be revamped, however Idle is designed as a governance token for the Idle DAO, as such the discussion in my humble opinion needs to focus on how to improve his main function, governance, instead of looking for new type of utility.

Looking at snapshot, it’s easy to notice that the vast majority of votes have between 3 and 5 voters out 3770 holders. While thinking about a revamp it would beneficial to take that into account and make that number much larger.

The proposed ideas of a performance fee discount and early access for stkIDLE holders are interesting but the underlying assumption that users of the idle products are also idle holders/stakers does not seem to be true for now.

I understand the proposal aims to change that but it could also be perceived as a bit of a stretch from a user prospective, do we have reasons to believe it will be well received at this time?

An alternative way of thinking about it would be to allow early access to new products and fee discount to those who use the products the most (leaderboard?), which would also build an incentive to deploy more capital to such products in order to unlock those benefits.

Regarding Idle Staking, I’d like to share my person opinion on what I would like to see:

• Don’t pause the staking reward program but drop the Idle rewards via buyback and distribute the real revenues (maybe in the form of tranches?)
• Incentivise governance participation
• Consider making stkIDLE the only voting asset (better align for long-term vision)

Thanks for taking the time to read the post. Ciao

[william]

Ciao @Alexintosh! Thanks for sharing your feedback here!

@Alexintosh

I agree with the statement that utility of IDLE staking should be revamped, however Idle is designed as a governance token for the Idle DAO, as such the discussion in my humble opinion needs to focus on how to improve his main function, governance, instead of looking for new type of utility

I think we need for sure focus on improving the governance part, as it’s the main one, but having also other type of utilities could help thriving a more cohesive ecosystem for IDLE where users can not only partecipate in such governance but also enjoy (with benefits) the products that they are governing. In the end the DAO is needed to govern Idle related products so it should make sense to tie the token also to the products in some way.

@Alexintosh

I understand the proposal aims to change that but it could also be perceived as a bit of a stretch from a user prospective, do we have reasons to believe it will be well received at this time?

An alternative way of thinking about it would be to allow early access to new products and fee discount to those who use the products the most (leaderboard?), which would also build an incentive to deploy more capital to such products in order to unlock those benefits.

Having discounts for those who use the product the most could be an idea, of course it should be carefully weighted to avoid losing too much fees for the DAO, but in this way we are still leaving IDLE and users of Idle products separated, while the idea here I think it’s to try to converge those users, or at least give more benefit to the ones that embrace both IDLE and its products.

@Alexintosh

Regarding Idle Staking, I’d like to share my person opinion on what I would like to see:

• Don’t pause the staking reward program but drop the Idle rewards via buyback and distribute the real revenues (maybe in the form of tranches?)

That’s for sure something worth considering for the future, especially if we link fees to tranches directly. We are voting right now to pause IDLE buybacks as there were issues with the current implementation see the initial @Biaf comment on this thread.

@Alexintosh

Incentivise governance participation

This is something that was done I think until January of this year but was deemed not really worth it in that form. This is for reference the initial post Idle DAO Governance Mining and this was the analysis that concluded that it was not very effective Governance Mining Program - Retrospective Analysis. Of course if you have other ideas or insights for this, happy to discuss further given your expertise on this!

@Alexintosh

• Consider making stkIDLE the only voting asset (better align for long-term vision)

I find this the most compelling point in the long run, but it means completely redoing the governance from the ground up as we cannot use Governor Bravo for this

[CambridgeBlockchain]

We are in support of Proposal 1 as we believe pausing the staking rewards program to be the most sustainable action for the token at present.

The current staking reward program has not been effective in sustaining the token price through the buyback mechanism. Despite distributing rewards to stakers, the token price has not been adequately supported.

Additionally, the fee-sharing contract based on Sushiswap v2 lacks liquidity, resulting in high slippage costs when buying back IDLE tokens. By temporarily pausing the staking rewards, the focus can shift towards improving liquidity provisions and exploring more efficient ways for IDLE token buybacks, thereby addressing the slippage concerns.

We also hope for the introduction of alternative incentive programs for IDLE staking in the future, particularly ones that enhance governance participation.

[Biaf]

Temperature check results

Temperature Check passed

The staking reward program pausing will be included in the next IIP.

[Alexintosh]

Ciao @william and thanks for your reply!

I was suggesting to try to get one thing right (governance) instead of multiple at once (fee discount, etc, etc), in the end it’s a matter of strategy and preference, I get the desire for a cohesive ecosystem.

@william

but in this way we are still leaving IDLE and users of Idle products separated, while the idea here I think it’s to try to converge those users

Do you have an example of a protocol / product which managed to do that successfully in your view? I’d be interested to learn more since the implementations I saw so far did not seems particularly effective.

@william

This is for reference the initial post Idle DAO Governance Mining and this was the analysis that concluded that it was not very effective Governance Mining Program - Retrospective Analysis.

Happy to discuss in another venue the subject, in general I would be interested to see a new iteration with simpler rules, removing the 5 address rules and other type of rewards besides Idle.

@william

I find this the most compelling point in the long run, but it means completely redoing the governance from the ground up as we cannot use Governor Bravo for this

You definitely can but you will need to deploy a new Bravo and deprecate the previous one. We are doing that in Auxo if you want to have a look.

[william]

@Alexintosh

Do you have an example of a protocol / product which managed to do that successfully in your view? I’d be interested to learn more since the implementations I saw so far did not seems particularly effective.

I think DYDX managed to do this successfully, although their product is a bit different ofc https://docs.dydx.community/dydx-governance/start-here/trading-fee-discount

@Alexintosh

Happy to discuss in another venue the subject, in general I would be interested to see a new iteration with simpler rules, removing the 5 address rules and other type of rewards besides Idle.

I agree on this, it should be a simplified version of the last program ideally, but still with an eye on having a non-gameable system

@Alexintosh

You definitely can but you will need to deploy a new Bravo and deprecate the previous one. We are doing that in Auxo if you want to have a look.

I’m not sure a Governor Bravo can support the decaying nature of a veToken like stkIDLE, probably with a custom implementation but I don’t think it’s that easy to implement actually. Anyway if you have some resources to share I’m happy to take a look!