Governance Overhaul and Future Rewards

Authors

@banteg

Summary

Move fully to off-chain voting hosted by Snapshot. Enable voting with your YFI from everywhere, be it wallet, staking contract or providing liquidity on Uniswap. Distribute the rewards based on voting participation without vesting. Open up the way for yUSD rewards and gYFI proposals.

Abstract

Signal and YIP votes migrate fully to off-chain governance on Snapshot. Voting results are submitted on-chain.

Leverage the new Snapshot strategies feature to enable users to vote with YFI locked in other protocols. Centralized exchanges and malicious staking contracts will be barred from voting with users’ funds.

Vote results are persisted on-chain along with a merkle root in the new Governance v3 contract. The upcoming Treasury Vault v2 contract attaches the rewards to the finished votes.

Governance stakers can claim the rewards from multiple votes they’ve participated in by providing merkle proofs.

Motivation

Pressed by the gas price surge we have already de facto deviated from the initially outlined YIP process. I believe it’s an opportunity for a complete governance overhaul.

This proposal addresses many common complaints like:

  1. not being able to vote with your YFI sitting in a vault or some protocol
  2. or fears of that protocol voting with your YFI
  3. the rewards not covering the gas costs for smaller voters

Given the increased participation in other DAOs, we’ll have more proposals from them flowing in. For example, YFI holders will soon be able to cast a combined vote in Curve DAO.

Specification

TBD

Vote

For: Move forward with the proposal

Against: Reject the proposal

Poll: (verysoon)

66 Likes

Good to see YFI holders will be participating in Curve DAO. We need to voice our opinions on Curve since it’s tied to how our vaults work.

11 Likes

So one open question.

yYFI vault, crYFI, and aYFI.

Can choose whether you want to enable these. To fix the double counting problem, could only count YFI reserves. So if 25% of aYFI was lent out, aYFI would be worth 75% voting power.

Would increase complexity by quite a bit. And you’d might also want to include people providing liquidity using these tokens too.

3 Likes

Would aYFI have voting rights? If so, how would we handle someone borrowing the actual YFI and trying to vote with it (resulting in “double voting”)?

Otherwise, I’m in favor of off-chain voting (as long as it can be made simple for users / YFI holders), and barring CEX’es etc. from voting with a user’s YFI.

5 Likes

Yes, that’s something made possible with snapshot, it can read underlying YFI balances. I’ve pulled some data from Nansen and I think the initial list could include yGov, Aave, Uniswap, Balancer, Cream, yYFI vault and exclude Binance and Huobi. Not sure what Harvest is, need to look into that.

6 Likes

Good point, you can’t prevent voting with borrowed YFI with the proposed strategy. But I think the answer is educating the users that it’s never okay to lend more than 1/3 or your votes.

3 Likes

As long as it’s not a “flash vote” it shouldn’t be a problem. I mean If you give your YFI to a lending protocol and someone borrows it, it’s not your YFI anymore

3 Likes

This is a point of concern for me, but I’d like to hear what other users say.

But a follow-up question: would something like aYFI be limited to “1/365” voting power? Versus a user who has locked it in gYFI who would have n/365 voting power?

Also, I’m assuming that fee income would be commensurate with voting power.

6 Likes

In the proposal, if I understand it correctly, the aYFI could vote ALONG WITH the actual YFI (if someone were to borrow it from Aave).

@DCinvestor You could account for double voting by making aYFI have less voting power if some of it is lent out. Only count the YFI reserves on AAVE.

So if 25% of aYFI was lent out, aYFI would be worth 75% voting power.

6 Likes

Assuming this is feasible with Snapshot I feel like this solution should work nicely

This is what I’m thinking as well. Though if gYFI is implemented, it’s an open question how much voting power any of these YFI deposited in other protocols would really have. gYFI would be n/365 * # of YFI. Would these others be capped at 1/365 * # of YFI?

If it’s effectively de minimis, I’m not sure there’s much point to the added complexity of setting it up to allow this. It will also require the devs to constantly be adding new forms of deposited YFI across protocols.

Is the accruing technical debt really going to be worth it?

4 Likes

Can the dodo lp pool token be included as well (there’s 300+ YFI in there): 0xe2852c572fc42c9e2ec03197defa42c647e89291

These are things I want to clarify before moving it to vote:

  • Should we ban centralized exchanges? What if, say, Coinbase makes a voting portal for its users? Wouldn’t it be the same as we do with Curve DAO, albeit less transparent?
  • Who should receive the rewards? Should they be exclusive to the governance stakers or to everyone? If so, should there be some coefficients which disincentivize YFI sitting in riskier pools?
  • How does community feel about vote lock (gYFI)? I was thinking about specifying it for ygov v3, so we can enable it if the holders decide so. On the other hand, it implies a very different reward structure and both of these proposals are a ton of work implementation-wise. Should we amend this to have a choice between “everything goes” and “ygov vote lock” in the poll?
10 Likes
  1. Yes ban centralized exchanges.
  2. Only those staking in governance contract should receive rewards.
  3. gYFI is a great idea and should be implemented.
10 Likes

Imo, if gYFI is a transferable ERC20 we should keep the rewards limited to gYFI. If it is not, I think all voting YFI should receive them. This second option would be a lot more complex considering all the different accounting we could be doing for YFI in different staking & lending protocols

I share your opinion about 1 and 2 and a feel bit on a fence about 3.

Andre says we can do it and have an escape hatch in case it doesn’t work out.

I disagree with him about gYFI being transferrable, it goes against the idea of it being locked and makes no sense.

8 Likes
  • Should we ban centralized exchanges? What if, say, Coinbase makes a voting portal for its users? Wouldn’t it be the same as we do with Curve DAO, albeit less transparent?

I think we should until they implement their portals

  • Who should receive the rewards? Should they be exclusive to the governance stakers or to everyone? If so, should there be some coefficients which disincentivize YFI sitting in riskier pools?

Ideally, I think it should be distributed every time a snapshot is taken to all eligible YFI. But it will be difficult to implement. (storage will be a problem if yCRV is stored with a list of addresses they belong to, gas will be a problem if yCRV is sent out on every snapshot)

I now think only YFI staked in governance should get rewards as YFI staked elsewhere have vested interest elsewhere that doesn’t always align with YFI

  • How does community feel about vote lock (gYFI)? I was thinking about specifying it for ygov v3, so we can enable it if the holders decide so. On the other hand, it implies a very different reward structure and both of these proposals are a ton of work implementation-wise. Should we amend this to have a choice between “everything goes” and “ygov vote lock” in the poll?

I’m neutral on combining or separating these into different votes. I don’t think either are urgent. Though separating them might improve approval since some people might like one but not the other.

I’ll have to read the gyfi proposal again. I’m more in favor of the % of rewards increasing based on length in rewards contract. I guess there is a lot more to gyfi I need to review.

Risk is too high that an unscrupulous exchange might vote with their users’ funds. It may also be difficult for them to route governance rewards to their customers. This could be revisited in the future if an exchange expressed a desire for it.

IMO, the owners staking actual YFI in the governance contract should be the only ones eligible for rewards. There is an opportunity cost to keeping YFI in the gov contract, versus on Aave, etc., and they should be compensated for this.

I like the idea of gYFI, though I’m not sure if n/365 is too restrictive. n/60 or n/90 might have enough of an effect. Or some sort of function which gives you “most” of the voting power in 60-90 days, then giving you diminishingly more voting power as you approach 365 days. I wonder if there are other protocols we may be able to learn from here.

Personally, I think all of this is more important than trying to figure out how users can vote with YFI deposited in other protocols. I’m open to that idea, too, but only if they don’t receive governance / fee rewards. The downside is the Yearn team will constantly get requests to add voting functionality for new forms of YFI. I’m not really sure it’s worth it.

13 Likes