YIP-69: Reduce and cap fees through yRates

[0xJiji]

Authors: banteg, flashfish, jiji, jmonteer, newmickymousse, saltyfacu, wavey

Summary

Establish a new yTeam, yRates, and transfer to it the “Set Fees” power, with the requirement that earned fees must never exceed the yield earned by vault depositors.

Status

Discussion
This proposal is currently in the discussion phase. As per our voting rules outlined in YIP-55, it will be in discussion for at least 3 days with a non-binding forum poll to gauge sentiment before it can be assigned a YIP number and move to Snapshot for a binding vote.

Background

In the previous bull market, opportunities were readily available, and Yearn vaults could generate high yields. Since the height of DeFi Summer those yields have shrunk, and it is now common for many crypto assets to return in the lower single-digit APRs.

Yearn vaults never deploy funds unless vault depositors profit from this, as this would add unnecessary risk to depositor funds.

In current low APR environments, this rule forces Yearn vaults to have large amounts of capital uninvested, earning no income for depositors or the Yearn treasury.

Current state of fees

As per YIP-51[1], the current fee structure is:

• 2% management fee
• 20% performance fee

Fees are automatically levied on deployed capital only, and is the primary income source for the Yearn treasury. It finances protocol expenses including grants, gas costs, infrastructure, YFI buybacks, and more.

If a vault strategy is earning 2.5% APR or less, depositors stand to realize no profit on harvest (after the current mgmt + performance fees), which means that funds do not get deployed to this strategy. In such a scenario, some of the implications are that:

• All strategies which earn between 0.5% - 2.5% APR become non-viable.
• Yearn treasury earns no fees.
• Yearn vaults require additional monitoring to react to situations where APR dips too low capital needs to be unallocated.

yTeams

• As per YIP-61[2], yTeams…

  …are small, autonomous groups of yearn contributors empowered by YFI holders to act independently in the best interest of Yearn within a constrained domain of action and with enumerated, discrete decision-making powers.

  Examples of existing yTeams are yBudget, yOps, and yPeople, each having their own distinct domain and decision-making powers.

• “Set Fees” power is currently with YFI holders, as per YIP-61[2].

• YFI holders have the power to ratify new yTeams.

• The yOps team has the power to ratify new signers for yTeams.

Motivation

Efforts are underway to evolve protocol operations across several areas to improve performance during the bear market.[3] As part of this, a dedicated team that is responsible for product fee structures and can react on short notice will allow vaults to stay competitive with higher yields that draw more TVL.

This benefits all stakeholders: vault depositors earn more yield, yearn treasury accrues more fees, and YFI holders see more capital allocated for YFI buybacks.

By establishing strict fee guidelines, yTeam powers become better restricted. Vault depositors also get an understanding of the worst case impact of fees.

Future possibilities

• A fee dashboard to visualize current fee structures across vaults to depositors and improve transparency.
• Delegation to veYFI[4] lockers to dynamically adjust fees of vaults.

Specification

1. Ratify a new yTeam, yRates. yOps is tasked with ratifying initial signers and quorum, but it must consist of at least four individual signers. This does not imply increasing full time grants, signers can be taken from the existing contributor pool.
2. Transfer the “Set Fees” power to yRates from YFI holders.
3. Enforce a strict condition that Yearn fees must never exceed the yield paid out to its vault depositors, instructing yRates team members in their work to optimize fees to benefit both vault depositors and the Yearn treasury.
4. Moving forward, the published quarterly financial reports[5] should include an update prepared by yRates on past fee structures performance and the future outlook. These updates should not block the publication of the report.

Vote

Non-binding signalling poll

Proceed with this proposal in its current form?

• Yes
• No

References

1. http://gov.yearn.fi/t/yip-51-set-vault-v2-fee-structure/
2. http://gov.yearn.fi/t/yip-61-governance-2-0/
3. Building during bera. Haven’t you heard, anon? DeFi is dead… | by banteg | Yearn | Medium
4. https://github.com/yearn/yearn-pm/tree/master/financials/reports
5. http://gov.yearn.fi/t/yip-65-evolving-yfi-tokenomics/

[0x7d54]

I support the idea of a yRates team that is empowered to dynamically adjust rates, now less constrained than before with the passing of YIP-66 and with other necessary/reasonable compensation adjustments given the current environment, that have been discussed by yearn team members in the referenced blog post and on twitter.

That said, I would like to see the incorporation of more discussion of the broader related present issues/ goals for this team/ mandate, beyond merely the points here concerning fees in a lower-yield environment. i.e. this should not be only about the bear, but about increasing competitiveness and maximizing growth both at the current moment as a leader in this space and also looking forward to the next growth phase as the landscape continues to evolve. Specifically:

• Yearn fees need to be more competitive relative to other opportunities for DeFi users/ other protocols.
• There is criticism/ understandable frustration that the curve vault strategy fees, especially, are too high.
• Fees could/should dynamically reduce as vaults scale, in a win-win way for users and yearn (i.e. yearn will earn more total net profit, rather than less, and more sustainably, by charging lower fees as vault deposits scale and for more straightforward strategies, to encourage even further deposits and growth).

[0xa9a]

3. Enforce a strict condition that Yearn fees must never exceed the yield paid out to its vault depositors, instructing yRates team members in their work to optimize fees to benefit both vault depositors and the Yearn treasury.

Why not enforce this on-chain by modifying this condition in Vault.vy?

if total_fee > gain / 2:
    total_fee = gain / 2

[mhdempsey]

Very much agree with this and have been trying to beat this drum for awhile now ([Working Proposal] Innovating on Fee Structures)

Think this is a good first step but would think that as we enter different yield environment (both in terms of lending/borrowing but also perhaps with the slow death of liquidity mining) we should think about reducing fee structures based on competitive dynamics and yield bands.

[robzer]

supporting this. fees has been a topic for a while not regretting the wait becase this proposal is very comprehensive.
can we imagine a world where the yRate team also take into more inputs (overall yield levels, tvl of vaults, COMPLEXITY of strategy, etc…)?
I think it makese sense to charge more fees for strategies with an inherant edge: i expect to be charged less if its jsut farming and dumping a convex/curve pool because I can do it myself as a pleb, however, something more esoteric requiring smart yield routing to different protocol am happy to pay more fees.
ty

[Crolev]

How about a simpler approach like a rule that all the fees shouldn’t take more than 50% of total yield?
Both could be implemented in the future vault versions and in the meanwhile yTreasury could donate the fees earned above 50% of yield back to the vault.

[mmohz]

I am surprised this is not already a factor of the original proposal. It should be a graduated fee based on the yield. Akin to a tax bracket in the US.

At the current 0.2% or 0.01% rates for Stable Coins, we are un-attractive to most projects. And this will certainly help. The fact that its not hardcoded is a positive in some factors… Since they can take a look at the market to see what APY people are expecting for that asset class and use that when determining fees as well…

I think a reasonable base goal should be at least 1% net apy per-asset (stablecoin) that is and hopefully near aave so yearn is the easier choice with more exposure to different baskets but a similiar yield to aave etc…

[miguel567]

Thanks for posting this and specially about rates!

Question: how often do you see that rates would change? just wondering how often the APY would be affected by changing rates and the user experience thereof.

Thanks!

[Juanma]

I guess it can be applied for next deployed vaults, all the existing ones would require migration which is painful

[banteg]

Voting is live Snapshot

[miguel567]

Since the vote is already happening I want to share:

• I don’t see the current governance process as a friction element in changing rates. My argument is that current governance process is 3 days discussion in the forums + 7 day snapshot vote = 10 days to change rates. I don’t see yRates changing rates every 2-3 days (less than 10 days). I feel rates would change once a month or quarter. So I don’t see why changing rates through governance is a pain for the protocol operations.
• Yearn must move towards on-chain governance, automating changes through on-chain voting which give power to governance (bigger group of ppl). yRates is completely the opposite: giving power to a small group of ppl compared to now.
• yRates team will be formed by members who get their salaries through the protocol revenue = rates. Hence, they are incentivized naturally to keep rates high to keep their salaries and protocol expenses at the same level. Governance has an incentive to make the protocol profitable, but that is not necessarily meaning salaries and running costs need to remain at the same level for profitability to be achieved. I think keeping rates on governance is a fairer, more transparent and better game theory setup than giving it to smaller team, and far more decentralized. We need Yearn to be more decentralized, let’s not take steps back.