Proposal to subsidize Concentrator's harvesting transaction cost via Fundraising Gauge (Harvest-for-All)

[0xConcentrator]

Summary:
Concentrator would like to propose the addition of the Harvest-for-All gauge with a max. emission of 25,000 CRV, so that a broader range of Curve community can sustainably benefit from more options and better services provided by Concentrator.

References/Useful links:
Website: AladdinDAO Project
Twitter: https://twitter.com/0xconcentrator
Audit report: https://aladdin.club/audits/AladdinDao_V3_Report_Secbit.pdf

Protocol Description:
Concentrator is an entirely new type of auto-compounding vault system, based on a philosophy of farm-and-hold. Rather than selling rewards constantly and buying more of the deposited token, Concentrator helps farmers concentrate yields from multiple Convex vaults into cvxCRV, then leverages auto-compounding to grow those returns even more. When it’s time to withdraw, users can choose from cvxCRV, CRV, CVX or ETH.

Motivation:
Concentrator provides a highly cost and time efficient way for Curve community to benefit from the enhanced yield. Although, it also comes with a bottleneck to further expanding our service for an even broader range of the community.

In order to achieve the auto-compounding, the system needs to periodically call the keeper function to harvest the reward from Convex vaults. The most logical timing to do so is when the admin fee that goes to the protocol (0.5% or 2.5% depending on the vault) is at least no less than the transaction fee for harvesting.

Over time, we have noticed that it is highly cost intense for the protocol in order to harvest for the users regularly, especially so when it comes to low yield Convex vaults, which would require
a huge amount of TVL and / or excessive waiting time for the vault to have accumulated enough reward before harvesting becomes economically worthwhile. Below is a list of the latest harvestings. As indicated from the table, the system already has to subsidize the cost even with high yield vaults sometimes, and it’s ever more evident after we added 3pool

Since the launch of Concentrator, we have received numerous requests from Curve community that ask us to add more low yield vaults. We truly believe that every community member deserves the benefits offered by Concentrator, regardless of their LP size or in which pool they have their LPs. However, as a very fresh protocol with limited resources, we can only subsidize the cost for our users to a certain extent. Therefore, we would like to receive funding via CRV emission that is going to be utilized as the transaction cost subsidy when Concentrator can no longer sustainably do so.

Goals and Expectations:
If approved,

• Concentrator will be able to add more low yield vaults that can be used by an even broader range of Curve community
• Concentrator at the meantime can better focus on operation optimization with its own resources to reduce the reliance on the fund
• Both Curve and Convex can see an increase in ecosystem participants that aim to make use of the broad range of options Concentrator offers

[skellet0r]

So is this proposal saying that the product is not net profitable?
If the product is net profitable, why not decrease the harvest bounty and increase the withdrawl fee / platform fee as a counter balance? These are all parameters that I see in the source code.

Are there any other things that could be implemented / modified to effectively eliminate the need of a subsidy?

Also, this is very odd proposal to submit to Curve DAO, since 0xconcentrator is built on top of Convex, wouldn’t it make more sense to request a grant from Convex DAO.

How is subsidizing 0xConcentrator beneficial to veCRV holders?
Getting CRV would require 0xconcentrator to sell the CRV for ETH in order to offset tx costs?
A fundraising gauge requires veCRV holders to allocate gauge weight each epoch (which requires a tx fee to submit on-chain), wouldn’t this cost in the long run more to veCRV holders?

Overall, in my own personal opinion, this doesn’t sound like something Curve DAO should subsidize. Doesn’t Aladdin DAO have a treasury which could be used to fund it’s own product in the short term, especially if the expectation is the long term revenue generated by the product’s fees will be greater than the subsidy?

[0xConcentrator]

Thanks for the comment.

Concentrator is profitable, as long as we only focus on high yield Convex vaults. We bootstrapped by starting only with the highest TVL / APR vaults that can be profitable in a reasonable timeframe. The intention of this proposal is to raise fund to subsidize harvest for low yield vaults that we would not otherwise even add at this time, but have been requested to do so by the community persistently, so that more Curve/Convex community members can benefit from our service. Our experience with 3pool vault has demonstrated that. In the end, the service Concentrator provides is a significant plus to CRV since we both directly and indirectly prompt more CRV to be locked.

[skellet0r]

If you could, please answer the following questions:

1. If the product is net profitable, why not decrease the harvest bounty and increase the withdrawl fee / platform fee to subsidize the harvest transactions?
2. Are there any other things that could be implemented / modified to effectively eliminate the need of a subsidy? For example, increasing the time between harvests, harvesting only on withdrawls and collecting rewards on deposits, harvesting when baseFee < 25gwei.
3. How is subsidizing 0xConcentrator beneficial to veCRV holders?
4. Why not submit a proposal to use Aladdin DAO treasury funds?

The intention of this proposal is to raise fund to subsidize harvest for low yield vaults that we would not otherwise even add at this time, but have been requested to do so by the community persistently, so that more Curve/Convex community members can benefit from our service.

Again if the service is not net profitable … why even release it?

In the end, the service Concentrator provides is a significant plus to CRV since we both directly and indirectly prompt more CRV to be locked.

Correction, if you use the CRV/cvxCRV factory pool to auto-compound (which I imagine you are since it provides a better value than direct locking CRV → cvxCRV), then you are not “locking” CRV, you are just helping the peg between cvxCRV and CRV. Secondly, since 0xConcentrator has the ability to charge a platform fee / withdrawl fee, you’re already being subsidized by Curve DAO via the fee you take.

Note: At the end of the day, it’s up to veCRV holders to decide whether to vote in a gauge, but I think this is a joke of a proposal. My opinion stays the same until you can provide some serious metrics as to how a revenue generating product being subsidized by Curve DAO is beneficial to veCRV holders. Saying “the service Concentrator provides is a significant plus to CRV since we both directly and indirectly prompt more CRV to be locked.”, means nothing. You’d need to show the CRV lock rate before and after 0xConcentrator launched to validate this statement.

*my opinion is my own, and may not be shared by other members of the curve team

[haowi]

What Concentrator is doing does help CVXCRV pegged and more CRV locked into veCRV. This is good for Curve token prices and the ecosystem overall. It will be great if Concentrator can do an airdrop to veCRV holders to make it mutual beneficial.