Request For Feedback: Utilize Yam's Treasury to Purchase CRV/CVX to Mine Bribe Rewards


One of the basic value propositions of Yam DAO is the ability for the DAO to put its diversified treasury to work for the purpose of growing the Treasury and accruing value to community members. @ross has already begun the process of proposing mechanics that begin to address the latter. Meanwhile the Treasury has already been used to farm various tokens for the Treasury (e.g. $INDEX, etc) on behalf of the DAO. In the same vein as @Snake’s proposal to use the Treasury’s $UMA to vote in UMA’s oracle system in order to earn revenue for doing so, I’d like to highlight another strategy for the DAO to consider.

Curve Finance’s voting system for rewards has given rise to what some have called the “Curve Wars” where protocols are competing with one another to incentivize liquidity for their own Curve pools.

As a result of this new game theory, we now have boosted CRV staking protocol, systems like, and protocols like which allow projects to directly appeal to token voters with rewards in their native token. The ecosystem is now evolving even further with protocols like building on top of these incentives using OHM style mechanics, etc.

Yam DAO does not have its own stablecoin, so there is no need for the DAO to participate in the Curve Wars as a reward provider. However, our Treasury does give us the opportunity to participate as a voter.


I believe it would be beneficial for the DAO to consider committing a portion of the Treasury to accumulating $CRV and $CVX and utilize these holdings to vote on Curve gauge weights in exchange for rewards.

I would advocate for the revenue to be split between the Treasury and a tokenholder value accrual mechanic. Mechanics to consider include, but are not limited to, “Buy Back and Burn”, an ibYAM staking solution in the vein of $xSUSHI or $sSPELL, etc.

The specifics of the mechanic can be determined through future discussion as these are just two examples that come to mind.


There are multiple benefits to this endeavor:

  1. It allows the Treasury to generate yield without the risk of impermanent loss.

  2. It generates additional revenue for the Treasury with very low labor requirements to execute.

  3. The Treasury benefits from both the yield generated from voter rewards and also the underlying tokens’ value accrual.

  4. Revenue generated helps extend the DAO’s runway.

  5. Revenue generated helps to fund additional dev resources needed to ship products.

  6. Revenue generated can be used to provide liquidity for future Yam DAO products.

  7. Voter rewards are denominated in the native token of the protocol which is incentivizing the votes, allowing the Treasury to accrue further diversified token positions over time.

  8. With enough voting influence, Yam DAO’s voting position has the potential to encourage additional B2B and DAO-to-DAO interest and partnerships with other projects that want our vote. Such relationships have been lucrative for the Treasury in the past.

  9. Revenue sharing as a result of incoming revenue would increase community morale and contribute toward a more positive metanarrative around the $YAM token.


If this idea gains momentum and alignment from other members of the DAO, the next step would be to put forward a more formal Proposal and Request-for-Feedback specifying exactly which assets the Treasury would acquire ($CRV or $CVX), how large of a position to take, and which assets would be swapped to do so. My hope is that we can begin to flesh out these points via discussion below.

If approved by governance, I believe we would need to determine whether the existing multisig has the ability to participate in the bribing/voting protocols we would want to utilize. If a new multisig is needed explicitly for this purpose, we would need to determine who among the core contributors and other DAO members to empower as the multisig signers.

We would also need to determine general guidelines and parameters for how the DAO wants our Treasury position to vote, whether to prioritize for the most valuable rewards available each vote, whether to prioritize strategic relationships, or whether to adjust on a case-by-case basis. All open for discussion.

In terms of development load, we would need the following:

  • Smart contract to handle the initial on-chain transactions when acquiring our initial token position.

  • Adjustments to the Yam frontend to reflect the new Treasury position and any resulting revenue.

  • If the DAO arrives at consensus around a revenue sharing mechanic, a smart contract would be necessary to manage this process unless the revenue share can be tied into an existing mechanic (i.e. “Buy Back and Build”).

  • If a staking or ibYAM mechanic were to gain favor, additional work would need to occur to Yam’s frontend to allow for interfacing with the staking contract.

This is not a formal proposal.

At this stage in time, I am only seeking to put this idea in front of the community to spur discussion, solicit feedback, and determine whether the DAO has an appetite for this. Any and all on-topic feedback is welcome. I invite anyone who wants to participate in the discussion to do so.

I am particularly interested in @krugman25’s POV on whether the value proposition is worth it compared to other ways we could put the Treasury to work, in addition to how this would impact our Treasury rebalancing and established risk parameters.

I look forward to hearing everyone’s thoughts on this.


I like the idea! As discussed on the discord, we would need to think about whether we buy or farm CVX/CRV. My gut says farm, but I’m open to all suggestions or other opinions. It may make sense to buy a little bit (100k) in a rebalance to get things started and then also farm it.

We will require some more robust farming/investing infrastructure to make this work (probably a multi-sig) but this is something that we are actively discussing for other proposals so it could dovetail here.

I’m not sure if you have read part 3 of my vision, but it proposes some ideas about using the treasury that would need to be considered with a plan like this. How rewards are distributed and/or re-invested is an important piece that needs to be figured out.


Working my way through your previous posts and wrote this one with the assumption that some reconciliation will need to happen between this and other conversations happening concurrently, so will bring up anything that comes to mind after I finish reading!


Def a fan of this idea! Thanks for suggesting this, Demosthenes.

I would be in favor of jumping in with a fair amount - the 100k suggested by Ross sounds like a good start.


TLDR; I believe taking a position in CRV/CVX should be based on a larger strategy to be involved in these ecosystems not to chase yield or rewards. If we were holding these tokens as part of a larger strategy (like voting on a YAM related CRV pool), it would be a no brainer to create a process to delegate and participate in these votes.

UMA is a strategic partner to the DAO as we are building out products using UMA’s financial contracts. I believe this to be the main reason YAM has UMA voting tokens in our treasury, not to earn rewards.

That said, the UMA token is an inflationary voting token and the DAO’s position is diluted if we abstain from governance of that protocol. So my proposal to vote with our tokens is not based on the earnings potential of doing so, but on the theory that if we don’t vote then we are diluting our holdings.

The only issue I have with this proposal is based on this. It’d make sense to me if there was a strategic reason to hold these tokens in the first place. So as you have mentioned, if we had a stablecoin that we were looking to boost in the curve ecosystem, it’d be a no brainer as we would be long CRV. Without that being part of the strategy, I’m not following how this would be a strategic win for the treasury.

I’m not dismissing joining the CRV wars and launching our own stable, but that seems like a completely different proposal.

Although I’m not proposing this (yet :wink:)…|

I’d like to contrast these benefits to just adding more UMA to our treasury.

  1. IL is only relevant in holding LP tokens so this would be the same.
  2. We have the ability to affect the usage of UMA with our products (in our current plan), but not the usage of CRV. Ultimately the revenue comes from the growth of the underlying protocol. I’m not sure what are you measuring as having low labor requirements though so maybe I missed the mark on this one.
  3. I’d say that this would be the same for holding more UMA as well with the difference that our product lines (ie. Yam Synths) would have a bigger impact on the UMA protocol than that of CRV/CVX.
  4. This is not unique to any asset in our treasury as they are all investments that hopefully have this outcome.
  5. Same as #4
  6. same as #4
  7. As long as there is plenty of liquidity, the denomination of the earned rewards token doesn’t change anything. Uma voting rewards would work the same way in this aspect.
  8. This makes the assumption that we would have a significant amount of the total voting power and without a strategic reason to hold those tokens (like having a crv pool), I don’t think that will ever be the case. In contrast to holding more UMA, there is a larger incentive to hold as we are an early integration partner.
  9. Revenue sharing is not inherent to this proposal and so I see this benefit as an extension to simply earning revenue and my response would be similar to #4.

I might be missing something here as I’m in no means a CVX/CRV expert.

How do you farm CVX/CRV without purchasing these tokens first? Or are you saying that there would be a case in which we’d just buy and not participate in voting?

Deposit Stables or ETH into Curve and convex to earn the yield instead of buying with said stables.

Although if the structure of the treasury changes so that voting token holders earn all the yield then that doesn’t work quite as well.

This is a good critique and brings up valid points about the strategic value of holding CVX or CRV in the treasury.

Right now there are 3 main groupings of assets in our treasury. 1. Assets used for expenses, 2. Assets that give general exposure to the Ethereum and defi ecosystems, and 3. Strategic Assets that we hold because we are aligned with the underlying project or protocol (partners/ collaborators). Risk management and yield farming are layers on top of this framework.

Expense assets are USD, ETH, and YAM.
Ecosystem Exposure assets are ETH, BTC, and DPI.
Strategic assets are UMA, SUSHI, INDEX. We also hold some GTC and are LPing MUST.

Percentages of these different assets are managed as part of the treasury management program in the DAO. Acquiring a position in CVX or CRV would require that we either increase our DeFi exposure or reduce our DeFi exposure somewhere else. Another avenue could be to look for an index that contains convex or curve to gain upside that way.

Beyond the treasury management side of things, Is there information about what the actual returns on CVX are from bribes? Are they actually that much higher than the returns made on the pools that they incentivize? Are there specific project tokens that we are looking to acquire via the bribes? How do we decide which bribes to accept? I’d like to hear a more inclusive strategy that lays out what the longer term goals are.

We can take advantage of the bribe system by depositing stablecoins into the pools that are giving out bribes without having to actually vote. This may not have the same APR, but it doesn’t require CVX exposure or the active management of voting, which is non-trivial for the DAO.

So in general, I think generating positive cashflow and yield for both the Treasury and tokenholders should be an integral part of our larger strategy as a DAO. This is why, to me, this proposal feels like it deserves a conversation on its merits.

I don’t disagree with you here, and frankly, I don’t see this as a mutually-exclusive, either/or proposition where we’re choosing between UMA or CVX. To me, connective tissue is just that your UMA proposal was another example of proactively utilizing treasury assets to generate yield vs letting them sit there and relying solely on accrued asset value (such as in the case of DPI).

That said, I do think there’s a notable difference between the two which is worth noting. With UMA, we’re leaving money on the table by not voting. As you said, our inaction is dilutive. With CRV/CVX, if we choose not to go this route, there’s no unrealized revenue we’re missing out on from existing assets. So to me, this proposal has more to do with creating and exploring a new stream of potential revenue that didn’t exist before, which is something that I think we as a DAO should always be trying to do. I think it also aligns with what Ross was saying in terms of investing in the ecosystem. I firmly believe that, right now, Curve and the projects in its orbit, is integral to the DeFi ecosystem and will be even more so in the future. So to me, this does represent an investment in the ecosystem.

Personally, I am not in favor of launching our own stable (unless we wanted to do something completely experimental, and for that, I do have ideas…). To me, it’s just a crowded space, incredibly competitive, and we’d be at a severe disadvantage. Just my personal thoughts.

That said, in lieu of a flagship product that has definitively established its PMF and is excelling, I think the only real compelling alternative model is going to be a coalition of smaller revenue generating strategies that, in aggregate, accomplish the same result in terms of revenue. So in my view, moves like this have strategic value in that sense.

Again, I don’t think there really needs to be a comparison with the UMA in our treasury, and I apologize if making the reference to your UMA proposal made it seem like the two were in competition somehow.
Keep in mind that these benefits are not meant to be contrasted with any specific asset in our treasury.

  1. That is correct, IL only applies to holding LP tokens. I would argue that we should be routinely looking at revenue strategies across the board that do not require us to hold LP tokens that would suffer IL.

  2. Regarding affecting the usage of the underlying asset (UMA vs CRV), I don’t think you’re wrong, but in practice, this is not yet the case for our products. Maybe that will be true in the future (I sure hope so), but in the meantime, I don’t think it’s very relevant. What I mean by low labor requirements is that all it requires of us would be voting and/or delegating our votes to another entity through a protocol like Votium so that it’s completely passive on our part. It doesn’t require us to build a product to generate revenue. Additionally, unlike with UMA, it doesn’t require a contributor to spend time doing research to make sure they vote correctly in order to receive any benefit. So in my view, since we are a labor constrained DAO at this time, I see a huge benefit in sustainable passive revenue streams.

  3. Certainly true, but I am wary of putting all of our eggs in the UMA basket from a diversity standpoint. And again, the revenue generated from this proposal could be accomplished passively whereas revenue generated from our UMA position must be semi-active at minimum.

  4. Agreed, although not all assets in our treasury generate revenue beyond accrued value.

  5. Agreed.

  6. Agreed.

  7. I think it depends on whether we plan to keep the accrued rewards in our treasury as denominated or if we plan to liquidate them for other assets. For example, Abracadabra and Alchemix have both incentivized pools on curves in the past. One of which is a stakeable productive asset and the other soon will be with the launch of V2 this quarter. So to me, the ability to accumulate additional productive assets in the treasury without necessarily doing a market swap with our existing assets is a strategic positive.

  8. You’re not wrong, but I think there’s a world where we could make ourselves more influential in this space. We just have to have an appetite for it.

  9. Personally, I would very much advocate for revenue sharing to be part of the proposal, if it were to go forward. It would be a nice step in that direction and something that our community would respond very well to. But you are correct that it is currently not codified in this way.

Honestly, if the more conservative approach here is to start out by farming CVX or CRV tokens, I’m OK with that and would support it. I think it would mean a slower revenue ramp up but would certainly address concerns about swapping out assets in our current portfolio.

For me, the most important thing is that we, as a DAO, are not afraid to look for revenue generating opportunities in light of our current product lineup.

So one thing that needs to be considered, and this is likely a bigger picture conversation, is the fact that our current DeFi exposure feels very reflective of DeFi as it was during DeFi Summer. But meanwhile, the DeFi ecosystem has continued to evolve, and we are not engaging with the new generation of protocols hardly at all. Yes, there is risk there due to their novelty, but not engaging with them in a meaningful way also represents risk to the DAO in terms of ossification and irrelevancy. So we should look for ways to balance this exposure between old and new DeFi, as silly as that distinction seems.

All great questions, and I can dig into them more.

So to my understanding, if we were to do this but the pools we choose don’t “win” the vote, then our rewards would be subject to however the CRV gauges shake out for that period. So our revenue wouldn’t have any relationship to whether or not there were bribes available for that pool (unless the bribes were substantial enough to make sure that the pool is well incentivized with CRV rewards). In contrast, by voting or allowing another entity to vote on our behalf, I believe we would earn rewards from the bribes regardless of whether the vote is successful or not, and depending on how active we wanted to be, we could essentially pick and choose what kind of rewards we want by voting on different pools.

Another option that we could consider is yvBoost from yearn. It should be optimized to accept bribes and earns a portion of the CRV earned by Yearn. It’s earning about 100% APR right now and there is no need to manually vote to get bribes so it does it for us. If we are holding yUSDC or YDAI then it is also beneficial for our interest bearing stables.

There have been peg issues for the last few months, but that also means that it could be a high R/R play since if it does return to peg then that is an instant appreciation.

One easy solution to this could be to swap out some of the DPI we have for indexCoop’s new GMI index that they made with Bankless. It has a bunch of DEFI 2.0 exposure.

1 Like

Speaking of which: speaking of which:

Not opposed to this either. As long as the risk is within our acceptable limits and the DAO finds it appropriate.

Please keep your comments on topic to this proposal discussion. This is not the place for random complaints that have nothing to do with this topic. If you want to air grievances, then create your own post for that purpose and suggest alternative solutions and fixes.