The following (v1.1) can be implemented using off-chain voting (snapshot)
Abbreviations (of terms) used in the following description:
‘Owner’ refers to a YFI token holder
‘SC’ refers to a Smart Contract
‘FTU’ refers to a users First Time Use
‘UI’ refers to User Interface
‘Tx’ refers to an on-chain transaction OR an off-chain signature. (Assumes snapshot can be used)
‘NZS’ refers to a non-zero-state, meaning an action that is not the users first time
DEPOSITING
Owners deposits YFI into a new YFI governance vault,
On deposit of YFI, Vault mints gYFI
SC deposits gYGI into Governance contract
gYGI will NOT begin to receive governance rewards on deposit, (read on for rewards trigger)
SC deposits YFI into Maker Vault
SC mints DAI
SC undertakes a DAI Strategy
VOTING
On YIP release: ( Vote: FTU )
The owner navigates to yearns off-chain governance portal,
The owners deposited gYFI will be visible, (represented as YFI in the UI)
The owners rewards in yUSD will be visible, (As it is the owners FTU, their rewards will be displayed as 0.00)
The owner participates in first (FTU) vote,
On vote Tx (signature) confirmation:
The owner will become eligible to receive rewards
Rewards will begin to accumulate against that owners wallet address
Rewards will unlock and ‘Claim’ CTA will become active
The owner may now claim all rewards earned
The owner will maintain access to rewards and
The owner will continue to be able to claim rewards until,
On a new YIP release: ( Vote: NZS )
Rewards will lock and ‘Claim’ CTA will become inactive,
The owner navigates to yearns off-chain governance portal,
The owners deposited gYFI will be visible, (represented as YFI in the UI)
The owners rewards in yUSD will be visible,
Rewards will be locked until vote Tx (or signature) has confirmed,
On vote Tx (signature) confirmation:
Rewards will unlock and ‘Claim’ CTA will become active
The owner may now claim all rewards earned
The owner will maintain access to rewards and
The owner will continue to be able to claim reward until a new YIP is released.
UNSTAKING
When the owner wishes to reclaim their YFI, they will have two routes:
Route A:
The owner will navigate back to the YFI vault
On selecting the vault, as it expands, in the withdrawal section:
A withdrawal warning will be visible
This warning will explain to the user that by withdrawing their YFI they will also:
Withdraw from governance any and all accumulated yUSD rewards
If the user wishes to continue participating in YIP’s and/or receiving rewards they will need to either:
Redeposit their YFI into the YFI vault or
Deposit their YFI directly into the governance contract (as existing)
Route B:
The owner navigates to yearns off-chain governance portal,
The owners deposited gYFI will be visible, (represented as YFI in the UI)
The owners accumulated rewards in yUSD will be visible,
An ‘UnStake’ Panel will be visible,
Selecting the ‘UnStake’ Panel will reveal the following:
A withdrawal warning will be visible
This warning will explain to the user that by withdrawing their YFI they will also:
Withdraw from governance any and all accumulated yUSD rewards
If the user wishes to continue participating in YIP’s and/or receiving rewards they will need to either:
Redeposit their YFI into the YFI vault or
Deposit their YFI directly into the governance contract (as existing)
On [%] withdrawal:
SC will withdraw all accumulated yUSD rewards and
SC will burn the [%] gYFI and
SC will withdraw the [%] YFI from the maker vault and
SC will balance the YFI maker vault and
SC will deposit to the owners wallet; the claimed yUSD and [%] YFI.
Motivation:
(1)
It has been our observation that an APY of ~20% (or greater) will draw the greatest number of participants, below this APY participants are more likely to be actively looking for alternatives.
The goal therefore is to align with ‘owner’ desires and community security.
a. By creating a non-transferrable token (gYFI) that is minted and burned by a governance vault at a 1:1 ratio to the YFI token, on deposit and withdrawal (respectfully) from the vault and;
b. The assumption that Maker will not lend, loan, collateralise or similar, any deposited YFI tokens in their vaults and that the only intended activity that may be imposed upon deposited YFI tokens is that of Liquidations in line with their existing models, then:
Should all yield from the implemented DAI strategy be paid out in yUSD? If yes:
Should all yield from the implemented DAI strategy be included in Rewards paid via governance? If yes:
Should uses be restricted to only one access and withdrawal point for YFI in governance and for that access and withdrawal point to be the Governance vault? (2)
(2) This is because, the above proposal currently would allow for ‘owners’ to:
A. Deposit directly into the governance contract to receive rewards BUT would NOT receive yield from the DAI strategy OR
B. Deposit into the Governance vault and be exposed to both rewards and yield.
The above questions only seek to streamline the above proposal.
I look forward to everyone’s thoughts.
I think this brilliantly solves the issue of YFI that is lent out being able to used in double-voting situations. Kudos for figuring out a way around that (assuming Maker passes YFI as an allowed DAI collateral).
I think our biggest debate is going to be outlined in what you mentioned in your comment– should governance staking be allowed as a separate option, and if so, should governance rewards continue to only flow to those stakers?
In my mind, the main argument would be that staking is safer than placing the YFI into Maker, because there are simply far fewer points of failure.
I’m not sure where I fall on this issue– I’ll need to think about it more. But at the very least, I think you’ve given the YFI vault a strategy that will prevent malicious voting of Vault YFI (assuming we trust Maker), and that will likely at least open up vault YFI to voting, if not also receiving rewards.
Could we implement a check at the snapshot block to confirm that Maker has not done anything with the YFI? We would at least need to be sure that none had been liquidated.
Agreed this is an issue that should be given ample room for debate and discussion before implemented. I am against multiple pools for voting or rewards, but it is good to see that if implemented, it would be safer than some other alternatives.
If I have a vote, I would prefer to have all the options in a single stake/vault. There is no reason, in my understanding, as to why it would be beneficial to have both options. Why not turn the yfi vault into THE gov staking. So yfi holders gain all the benefits. Since we have the power to choose as a community, we should do the best for the entire community. Just look all the benefits into a single strategy.
Some people will still not vote, but I’m pretty sure the ones that do vote aren’t keen on excluding themselves from benefits for that right…
Trusting Maker is certainly a strong consideration.
The OPs motivation stated that yield will encourage more participation. Is very true. People don’t stake for free. They want to see, at the least, the gas costs refunded overtime.
Yearn.finance really should be the best place for yfi token. And it should perform at comparable levels to other options.
Most people just want yield, and may participate in gov. Not the other way. But when they are “invested” they tend to care a bit more, maybe enough to read though a YIP like this, understanding what they may, and providing ideas and feedback.
I think the gYFI is a succinct solution.
Rewards should be cycled back into the vault as suggested. It loweres the level of entry drastically. If you only have about 500usd of yfi, you can still compound interest without worrying about gas. Pooling funds is certainly the best way to utilize lump sum gas fees.
el_dan0Today at 12:00 AM
Is the YFI governance staking pool that will receive yUSD rewards now going to be combined with the YFI vault that earns YFI rewards? Or are they remaining two separate places to park tokens? Would be cool to combine and have YFI hodlers get both rewards, strategy gains, and voting. @banteg
bantegToday at 5:28 AM @el_dan0 do you understand they are the same strategy? yYFI only automates claiming and selling the rewards for YFI
So stake in governance for voting + rewards. Deposit in yYFI = same rewards but no voting + automated re-staking
Yeah I don’t think this is clear to a lot of users. I hope once YFI is approved at collateral over at Maker we can get some strategy yield aside from the rewards (only 5% APY and continues dropping). Once we have strategy gains+rewards+voting in the same vault we will be golden
I’d note that the APY on rewards should be expected to drop in the interim as YFI returns home. This is bc the reward APY is based on yield / total YFI locked. Target here would be to increase yield as YFI returns. This will occur as strategies evolve and attract more users.
Should Maker accept YFI and a suitable DAI strategy implemented, I’d expect the APY to ramp up considerably. My guess would be 20-30%. Maybe peaking out at ~40% before trailing down towards single digits, to mid teens. Though I’d expect this to take many years as YFI comes home.
30%-40% yield on governance sounds great. But with higher yields we will see, as you suggest, more YFI moving to governance that is currently being used to generate yield elsewhere. That could lead to two things: (1) more YFI in governance held by those interested only in the yield and not governance, and therefore a more difficult time in reaching quorum and (2) a misalignment of interests as some will be more interested in maintaining yield from the strategy rather than growth of the protocol.
Good points, especially regarding reaching quorum for votes. I’m not sure how to remedy this (maybe vote delegation/“protocol politicians”), but I don’t think YFI hodlers should necessarily be punished for not actively participating in governance and just wanting an income stream. We don’t really want uninformed people voting on big proposals just for the sake of participating and reaching quorum anyway. As for point 2, I’m trying to envision a scenario but having trouble - can you elaborate on what you have in mind that could go wrong there? Isn’t it just going to be a MakerDAO-based DAI debt strategy (once it’s approved by collateral)?
Let’s assume there are two ways to stake for voting: the current classic governance pool and a new gYFI MakerDAO-based DAI debt strategy. There will, of course, be some identity of interest among these two groups because both groups hold YFI and have an interest in seeing the value of their YFI holdings increase. But those groups’ interests could also diverge.
Let’s assume rewards are shared between the two groups to attempt to keep interests aligned. Because the MakerDAO strategy also mints DAI and farms that DAI, it will always have a higher yield than the classic governance pool, perhaps significantly higher. Instead, we can assume that all rewards go to those in the classic governance pool and only voting is shared with the MarkerDAO strategy. In that case, we should expect the two pools to reach a relative equilibrium as YFI moves between pools in search of the highest risk adjusted yield. But then, there is still a divergence of interests. For example, those in the MakerDAO strategy would be more concerned about YFI price volatility than would those in the classic governance pool because of the risk of liquidation in the MakerDAO strategy if the value of YFI decreases (even temporarily). Those in the classic governance pool, may be more willing to take actions that have a short term decrease in the value of YFI because they have no risk of liquidation.
So let’s assume that to ensure we do not create additional divergence, we merge all governance to a single pool – the MakerDAO strategy – for a higher yield for all YFI holders participating in governance. At least in that case, the pools themselves to not contribute to a divergence of interest between YFI holders. But is that in the interest of the protocol?
It certainly changes things. Currently, you have to be particularly interested in governance to stake in the classic governance pool rather than look for higher yield elsewhere. So this change would likely increase staking by those that would not otherwise be interested in participating. Also, if the yield from a DAI farming strategy outweighs the yield from protocol rewards, the health of the protocol generating those rewards becomes relatively less important. In the extreme, YFI holders could become more interested in the DAI farming strategies and uninterested in the Yearn system. Granted, if YFI holders ignore the Yearn system the value of those YFI holdings will eventually decrease; however, that does not necessarily ensure that YFI holders will put in the effort for long-term improvement when they are making significant short term yield outside the protocol. Under the classic governance pool, all yield now and in the future is derived from fees generated by the protocol and increasing that yield requires directly thinking about the short and long term effects of changes to the protocol.
As I review what I wrote, I realize it is lengthy. Thank you for reading all the way through. Regardless whether we go the route suggested by this proposal or not, we need to consider how it will affect YFI holders’ incentives and how it will affect the protocol in both the short and long term.
I really appreciate you taking the time to spell this out. That makes a lot of sense. For the combo pool, even if we increased the fees collected by introducing a performance fee, it probably still wouldn’t be enough to compete with the returns provided by the DAI strategy.
It doesn’t make a lot of sense that those who are taking the time to learn proposals, stake in gov and vote will effectively be punished for doing so. It should be the other way around, if anything, but I’m not sure how to accomplish that other than introducing a 5-10% performance fee while simultaneously holding the protocol rewards back from YFIvault and having them only flow to the gov pool. As AUM grows, it might compete with the DAI strategy. Not sure
In the short run, maybe not. In the longer run, it certainly could. Also, if there are two pools and only classic governance received rewards (but those rewards are significantly less than from the DAI strategy), YFI will generally flow from classic governance to the DAI strategy until those left in classic governance get a higher yield because they are sharing rewards with fewer other YFI.
If we want to expand voting or enhance YFI yield from sources other than rewards, we should consider merging the two pools. This would make it easier to understand and put all YFI holders with voting power in the same boat.
Perhaps we should not be extending rewards or voting power outside the current governance pool. Anyone staking there will be choosing governance over higher yields elsewhere and will share rewards with others also interested in governance. Those staked in governance would also have interests in increasing those rewards over time, which can happen if the protocol stays strong and grows.
Nothing is stopping a YFI holder from choosing higher yields in lieu of voting and system rewards.
I agree that if pools are kept separate, rewards should only flow to governance pool. As AUM grows significantly, it very well could eclipse the YFIvault DAI strategy in terms of ROI (especially if we rework performance fee) and then this problem would solve itself as people migrate over to governance.
I think it’s important to keep the above in mind. In the combined vault scenario, YFI rewards may outpace the DAI strategy gains, but that still wouldn’t be until we have significantly more AUM (and/or we add performance fees). We need to give some incentive for the prioritization of growing the YFI ecosystem (and therefore protocol fees) over the DAI strategy somehow. It’s looking more and more like we need the performance fee for this, but perhaps we won’t need anything besides the current harvest fee as AUM grows.
I have to think about it more, but I am leaning towards the combined option despite this risk, personally.
Retain things as they are with all voting power and rewards with the current governance pool. We could still have a MaketDao Dai strategy for YFI, but it would not have voting or rewards.
Have a single governance pool that offers voting, rewards, and the MakerDao Dai strategy. This would help keep all YFI governance stakers’ interests aligned.
Have two pools for governance, but keep system rewards only for the current classic governance pool. Hopfully the yields on these two will reach an equilibrium.