Seems like the delegation/voting incentives piece would be custom, but the rest VaultCraft can provide. Also the VaultCraft contracts are integrated on OKX, that’s pretty solid validation right there. I tried depositing/withdrawing on both VaultCraft’s dApp and OKX and no issues. I also got some ARB hehe.
1 Like
Agreed. I’m assuming this would be a 4626, which also opens the doors for it beings used as a strategy in other vaults that VaultCraft creates?
Really cool concept - support from my side 
1 Like
Uniswap is only giving yield to those who vote/delegate. Makes sense to reward active participation or at the very least not penalize it. A delay to stake/unstake also helps reduce flash loan governance attacks. The token received when you stake will likely be used to vote the same as COMP which also means there needs to be a limit incase of minting exploits (like only 25% of the vote can be made using it?).
Going forward, could I suggest that there be a separate thread where outside organizations can submit their proposals to handle staking to avoid this post being half advertisements? Will also make it easier to compare them.
2 Likes
Who’s this guy?
GM All. After some convos with various parties involved including the good chads at AlphaGrowth and Humpy. Decided it would be good to chime in. I’m new here so let me introduce myself.
The DeFi Ecosystem Builder
I come from a DevOps professional background. I joined DeFi in early 2020, starting at Badger doing governance, operations and strategy work, much of which pertained to farming yield generating governance systems such as veCRV and veBAL to generate yields on Bitcoin. That worked out a bit too well in the end, and became something a bit toxic in the Balancer ecosystem (in part due to Humpy).
As a result, in late 2022, I decided to leave Badger, and ended up at Balancer where I’m working with the Balancer Maxis, also doing Strategy, Governance and Ops. Part of my work with Balancer is helping DAOs to think about how to integrate ve tokenomics combined with Balancer 80/20 pools (more on that later). I come here both as a student of real-yield-generating governance systems, and Balancer Maxi who comes with a little twist to this whole thing to make it all better based on well proven Balancer Tech 
The whale whisperer
I’ll also add that I’ve been “dancing with humpy” since he first started accumulating Badger years ago. Over the time we have come to a good understanding, and believe that it is quite possible to build mutually beneficial relationships with this whale
As a student of many of the topics being discussed here, and a believer in DeFi DAOism, I’m excited to join this conversation and contribute some thoughts based on my experience in the DeFi space. I’m also eager to learn from all of your work building one of the greatest and most successful experiments of all times, and work together to shape a better future for Compound, while also introducing much needed new models of governance into DeFi.
Feel free to find/DM me on the Balancer Discord if you want to chat more about these things.
Ok on to the meat:
3 Topic Areas
1: Composability, Yields, and Decision Making Oh My!
2: How much is staked, how big are the yields, is there still liquidity?
3: Humpy’s just a chad who still believes in DeFi and DAOs .
On stCOMP derivatives (i.e liquid lockers) and Compatibility
DeFi is supposed to be about compatibility, and is at its best when things are composing well. The whole idea is that by building something that other people can build on, you create space for others to build products around the value a protocol has to offer. These products composed on top of something can add more value, and as more and more products compose on top of a primitive, an ecosystem starts to form, which can create powerful and far reaching network effects.
@TokenDynamics has proposed a non-transferrable, stCOMP solution and brings up a lot of interesting concerns about a token that pays both yields and provides voting power can be either used and/or abused to create a more diverse governing body(delegates get paid, there is a reason to do this work) and/or help facilitate a takeover(Liquid Locker can offer the yields and better liquidity, and potentially amass large amounts of vote weight “for free”, or for the cost of maintaining liquidity for their wrapper, which is potentially much cheaper than buying/locking governance tokens.
Perhaps I misunderstand, but it seems @VaultCraft has proposed a liquid locked vault product as a base solution or at least something that compounds rewards. I would propose that the stCOMP product proposed by TokenDynamics invites @VaultCraft and anyone else who wishes to to build a product such as proposed on top of this. A good stCOMP system would be one that:
1: Supports/enables a diverse and competitive environment of projects building stCOMP derivatives, without creating a winner take all/kingmaker type scenario.
2: Generates a more diverse and engaged delegate set, where at best a decent number of people with vested interest and deep knowledge in the Compound ecosystem have to agree to make something happen.
1: Composability, Yields, and Decision Making Oh My!
Mixing governance, security, and income has to be the biggest challenge facing DeFi DAOs. For protocol governance to work well, you need a number of active delegates who care for the protocol/ecosystem in a holistic way and also have a deep holistic understanding of it.
Smart, passionate people have to get paid for their work to remain motivated and engaged. Decision making in the TradFi world command 6 figure salaries, in DeFi we expect such work for free and wonder why governance isn’t “engaged”. On the flipside, If you pay voters too much based on their vote weight, it can create situations where amassing vote weight and delegation is more an economic game than a technocratic or political one, and that can create decision making hazards.
Derivatives can compound this. For example, using stCOMP proposed above. I could create a liquid locked vault, delegate to myself, pay all the commissions back to lockers, mint a gov token and/or spend some of my own money to fund decent liquidity/add a bit of extra yields. Before long it is easy to imagine that I could amass quite a large amount of COMP from folks who just wanted yields and don’t care how I vote or have much else to do with Compound.
Some potential dynamics to think around to pertaining to the situation described above
The Bidirectional Quadratic Dynamic
When trying to figure out how to deal with Humpy at Balancer, one of the things I thought about was some form of Quadratic voting. The usual problems with Sybil of course were blockers. I spent some time looking into gitcoin passport and other anti-sybil mechanisms, but all of them seemed reasonably easy to game.
But what about if the quadratic dynamic went both ways? What about if the more an address had staked, the less vote weight it received per unit, but the more yield? In this case, the party interested in yields would seek to build the largest pool of stCOMP available, surrendering an ever growing amount of vote weight. Meanwhile those who cared about the protocol could spit their locks between addresses, make some bags, and have a deeper influence on the protocol.
Clearly these curves would need to be modified, and thought about, but by applying a root to vote weight and an exponent to yields based on the supply an address has locked, it seems to me like you create a pluralistic system that gives various actors the room to maneuver in different ways and makes it hard for a wrapper to amass a huge amount of vote weight while still offering competitive yields.
Voting that rewards consensus
UMA has a governance system that encourages everyone to agree. Voters get paid for their votes, but only if they vote for the winning option. This means that with each vote, delegates come together, discuss, agree and signal. With Uma, 100% of yields are paid based on voting. So if you don’t take the time to be part of and vote with the pack, you earn nothing.
As the plans above already create a dynamic where stCOMP holders are obligated to vote, paid to do so, and slashed for not doing so, perhaps it makes sense to consider some of these dynamics more?
2: How much is staked, how big are the yields, is there still liquidity?
@TokenDynamics also spent some time thinking about the total supply of COMP, and the resulting yields given the expected outflow to stCOMP participants. The expectation seems to be that all COMP active in wallets over the last year will stake. This is where I make my Balancer Sales pitch 
The problem
Imagine that all the COMP were to stake. There’s no real risk/short locking times, why not? First of all, the resulting yields on COMP end up looking unexciting (single digit in 2024?). Second of all, if all the COMP is staked, then there’s probably not much sitting around in DeFi to be used for DEX liquidity, enabling COMP to be used for other things. TL;DR: If you pay all your yields to single asset comp staking that has very little additional risk, you suck up all the liquidity and end up with mediocre yields. Further, the protocol does better and revenues increase, staking sucks up more and more of the available liquidity. It’s hard for new folks to buy in, because there isn’t much inventory on the market.
Solutions
This is why people do locking. To add some risk factor, increase yields, and leave room for other applications for the token. The problem here is that if it works too well, you end up with a majority of governance long term locked, and not enough liquidity/inventory for things to shift/change much. If lock times are long, one can imagine participants moving on, and ending up with somewhat of a zombie governance.
Balancer’s offer in this is our unevenly weighted pools. With Balancer you can create for example an 80% COMP/20% ETH DEX position, which has an ERC-20 deposit receipt token. This 80/20 position allows COMP holders to enter without providing too much ETH, and also suffers about 1/5th the IL of a 50/50 pool, so it behaves closer to single asset staking than 50/50 LP. Balancer, Aave, Radiant, Alchemix and a number of others have built systems that use this LP token as the unit of governance and/or yield distribution. This has a number of effects pertaining to the problem above.
1: There is some risk of IL in the 80/20 position. As a result, one would expect a smaller portion of “active COMP” to participate. Combining the 80/20 with a lock or queue (couple weeks to a couple months of commitment and maybe not instant-withdraw, not years) you create a little more risk/commitment, reducing participation and increasing yields.
2: The more people stake, the deeper the liquidity grows. This means that it is easier for people to trade. Further, as Compound succeeds, and yields grow, liquidity grows to support new entrants, instead of shrinking to keep them out.
3: One could even consider a mixed solution, where single asset, shorter locked COMP could be used for voting without yields, and off-weight COMP LP could be locked for voting with them, perhaps using the bidirectional quadratic dynamic described above.
3: Humpy’s just a chad who still believes in DeFi and DAOs
I’ve been dancing with Humpy since the early Badger days. I’ve been scared shitless by his movements, and as I’ve come to understand him even managed to find some humor in watching people scramble when he does his thing. We talk on a reasonably regular basis, and I stay a bit involved in his Gold community and have helped connect him with reliable and trustworthy (and ethical) Web3 professionals as he launched his MemeCOIN and started trying to build some stuff.
I’ve come to see Humpy as an individual, Chaotic Good actor who bought into this whole DAO thing we were selling back in DeFi Summer(or winter, or whatever you call it), Over the Bear market he held and/or accumulated many coins of DeFi Blue Chips, with the belief that DeFi would see it’s day. The value of gov tokens has to either be security - If someone has enough they can take over the protocol so then the mcap must be worth more than TVL, or based on benefitting from the DAO treasury which is growing as a result of the protocol creating value for people and somehow capturing some.
His methods are… rough at best, but his intentions are good. In the end Balancer Governance ended up doing the thing Humpy wanted us to do that sparked our whole war with him. We would have done much better to be like the Compound delegates and make peace, but it’s pretty startling to have a total stranger suddenly show up on the cusp of a gov takeover demanding things.
Bravo for how you all handled this. I too am a believer in this whole DeFi DAO thing, and I’m excited for and proud of both Humpy and Compound for pushing this forward. Don’t stop until something happens!
Wrapping up
To summarize the suggestions above.
-
Composability is good. It’s important to think about the metagame created at the base layer. Consider something like the bidirectional quadratic dynamic to handle the balance between income and influence.
-
Liquidity makes a better base unit of yield earning, and can also make a good unit of governance. Too much income/yields to single asset staking can cannibalise DEX liquidity, which is required for other on-chain applications surrounding COMP.
-
Keep pushing, it’s time for the devs to do something. Bravo for using this to push Compound forward.
I came here more with ideas and questions than coherent plans or demands. Happy to answer any questions and/or dig more into anything here that inspires. Would love to make sure that, at the right time, the idea of using liquidity instead of single asset staking is seriously considered by the Compound community as part of all of this. 
2 Likes
It’s a 4626, which provides much more flexibility terms of functionality and addresses all business requirements. The great thing thing here is that we COULD add an auto-compounding strategy for rewards if so desired.
We created a prototype if anyone wants to play around. We’re using the same UI as our Smart Vaults to give you an idea. This of course would be adjusted for locking and claiming COMP.
I’m also attaching the UI we used for our Fraxlend Smart Vaults previously which includes custom lock and claim functionality.
2 Likes
This just sounds like a full product from someone who is making/selling a full product. How would this be monitized for VaultCraft if it were to be implemented.
Does compound want to pay/give an exlcusive deal to one provider to make a “complete” solution that can be anything anyone would ever want it ot be, and then to try to decide via ongoing governance what that is? Or does compound want to create a new primitive that is reasonably simple in it’s nature and opens up a lot of design space for others to build on top of.
If the Vaultcraft thing were 100% free and could be used in a minimal way, without integrating all the other stuff on top, then maybe it makes sense. It seems to me like overkill for the base layer, and like something that could be used to make a bunch of interesting products on top of said Base Layer if a good one existed.
Maybe the issue is that the proposal is offering too much, and there is a minimal proposal that doesn’t try to “do/monopolize the entire stack” and won’t cost Compound more than it would cost to hire some devs on a grant to build a simple staking/locking solution based on the parameters the community agrees.
3 Likes
Thanks for proposing an initial design @bryancolligan, we are aligned with the general direction of the proposal however, we do have some comments.
Most of our thoughts echo a lot of the thoughts already shared by other commenters:
Claim Mechanism
Distributing rewards in 20+ different tokens is not ideal from a UX perspective, it’s burdensom on the accounting/tax side and will become quite expensive for smaller users to consolidate their rewards during periods of high gas fees. A mechanism that handles this before reward distribution would be preferred (e.g., UniStaker)
Token Locks
We agree with @dennison here and an alternative solution like streaming rewards over longer periods which are then forfeited when a user unstakes continues to reward longer-term stakers while eliminating some of these risks. Nonetheless, we are not entirely opposed to some sort of a lock up mechanism but there should be a clear purpose/risk/benefit for doing so.
Redemptions & Atomic Redemptions
For users wanting to instantly exit the system and pay the 5% liquidation fee we should allow them to do so without needing to utilise the protocol’s treasury and forcing the protocol to hold an unwinding staked COMP position that will then have to claim rewards from itself. For example, when a user requests to instantly unstake, they receive their original COMP position minus the 5% liquidation fee with the liquidation fee going straight to the DAO treasury.
Liquidity Pools
We like this idea and think more protocols should look towards owning their own AMM curves, but in this case we are concernced that the capital outlay to achieve optimal trading conditions is greater than its estimated benefits. We wouldn’t want to force swapping through these pools if better execution can be achieved elsewhere which is likely the case given most markets on Compound require a good amount of onchain liquidity. Therefore, we are likely against such a solution at the moment and believe an auction mechanism like UniStaker’s is more efficient and requires less maintenance.
2 Likes
OpenZeppelin has recently conducted an analysis for a similar staking product proposed for the Arbitrum DAO by Tally. We’ll provide a link to our analysis in case it proves helpful in providing similar insights for Compound.
1 Like
At API3 we’re happy that Compound is exploring new ways to generate revenue that can be redirected towards COMP stakers. We’ve made a post about recapturing liquidations incentives currently being paid out, which already amount to several million dollars a year on Compound.
After some discussion we just pushed a proposal to pilot this mechanism on one comet, to prove viability. These are funds that could theoretically all be redirected to COMP stakers.
Happy to contine following this discussion and also receive some feedback on the proposal above.
Edit: Here is a dune dashboard i’ve built that summerizes how much the compv3 comets have paid out in total.
Hint: It’s a lot.
https://dune.com/ugurmersin/compoundv3-liquidation-data
2 Likes
Implementation would be free. As stated, we’ve created flexible toolkit, so anything that’s not necessary can be removed. The contracts are live and have been audited several times. Monetization is something that we’re of course open to discuss and are very flexible here.
Following up a couple of your points @tritium - it’s nice to have your experience and expertise for the situation.
Quadratic voting
When trying to figure out how to deal with Humpy at Balancer, one of the things I thought about was some form of Quadratic voting.
…
But what about if the quadratic dynamic went both ways? What about if the more an address had staked, the less vote weight it received per unit, but the more yield?
Interesting concept - have you seen a similar mechanism in practice? Either way, worth exploring.
UMA consensus mechanism
UMA has a governance system that encourages everyone to agree. Voters get paid for their votes, but only if they vote for the winning option.
This mechanism encourages everyone to agree, and it works best in situations where there is an objective truth about which to find consensus. It resembles a Keynesian beauty contest, which encourages consensus but doesn’t necessarily drive desirable outcomes.
Native LPing for additional stCOMP returns
This mechanism is also interesting - somewhat complicated, but worth discussing further. Can you link some detailed docs that describe it?
1 Like
Certainly:
Re quadratic voting (2 way):
Quadratic voting is basically the idea of applying some kind of a root to someones holdings in order to determine their vote way such that each additional token (beyond some threshold perhaps) gets incrementally less votes. It’s a way to bridge the gap between totally economic token based voting (1 token 1 vote) and something truly democratic (1 person one vote).
I think in DeFi could be good middle ground. One person with lots of vested interest should get more vote weight, but we also don’t want one or two people taking over and calling the shots (the dreaded 51% attack).
The issue with using quadratic voting in Web3 has to do with anonymity and sybil. In a pure web3 quadratic setup, a dishonest actor could simple create lots of wallets with 1 token (or the threshold if above 1) and amass more vote weight than an honest actor who maintained a single identity as a quadratic program intended. Further, if voting tokens are locked, they could do this and run governance before others had a chance to unwind their big positions into lots of smaller wallets to.
Gitcoin has been a leader in the quadratic voting space. They have continued to use it to allocate grants, where each dollar someone donates signals matching in a quadratic manner. In doing so they have spent years thinking about sybil, and have come up with Gitcoin passport and a number of other mechanisms that are interesting. These do a good enough job for distributing public goods money, but are not bombproof enough for DAO governance.
|In the case of Compound, most votes don’t have much economic benefit/consequence for the voters in the short term. It’s more an issue of security in that some votes could have catastrophic economic consequences for many, and great benefit for 1 or a few. As I described above, adding yields creates an economic side of holding this governance token, and as such should create some upward price pressure making votes more expensive.
By giving a wallet more yields but less votes as it accumulates more tokens, you create a construct where by folks who want yields can focus on that (or protocols can build ways to collectively farm the yield bonus with vaults/wrappers). Folks who are more concerned in participating in governance can keep their comp in a few wallets and delegate to a single delegate.
This is all an untested thought experiment, not pitching that it is done, only considered.
Here is a model/sheet to copy to play with. It basically alters votes by taking the some root of all vote weight over a specified x-axis shift. Yield is boosted by a formula based off the curve boost formula. You can play with the constants and see how results change. Not suggesting these formulas are used, but just something to think about. If one were to build this, I’d highly suggest you get input from more than just me, and maybe run a test before you commit.
Yeah, you may be right here. In the case of Uma voters are verifying some verifiable truth, it is not subjective. Was just spitballing :). I can see how this system could drive desirable outcomes, but there’s a big “social layer” that would have to be built, and maintained on top of it for that to work. I. don’t think that’s what we are trying to do here. Was worth a momentary think
Saving the best for last this is more proven/tested, It is the thing I am an expert in, and I can share a lot of details. Before writing more textwalls, Balancer has some decent materials already prepared. Try this and let me know if you have questions/want to discuss further:
Here is a article about the 80/20 initiative from Balancer.
Here is an article with a more maths focus around 80/20 liquidity, slippage and IL.
1 Like
I’ll just add that if we could figure out how to do this multi-chain (locked/utility BPTs representing liquidity on isolated chains), it’d be huge. Think a lot of DAOs are trying to deal with how to maintain liquidity everywhere they need it as the chain-verse expands.
Sounds like you guys are thinking in that direction which is… just exciting!
3 Likes
Conclave Proposal for the Development of Staked COMP
About Conclave
Conclave is a Web3 Research and Development firm focused on Decentralized Finance and consumer applications. We have been actively building in DeFi since 2021 and have facilitated the development of applications such as Radiant V2, as well as our own suite of products demonstrated by protocols like Ironclad (Mode network) and Aurelius (Mantle network). Pertinent to this application, we have previously developed an open source incentive distribution system called Reliquary that we believe can provide an ideal solution for staked COMP. This system is fully audited and currently deployed in our own products such as Ironclad’s $ICL staking.
Introduction
This proposal introduces Reliquary (whitepaper in work, live soon), an advanced incentive distribution system that supersedes the traditional MasterChef contract, designed to augment the economic utility and governance integration of the staked COMP (stCOMP) token. By incorporating a time-weighted reward mechanism, Reliquary incentivizes long-term staking, enhancing protocol alignment, and reinforcing governance participation. We will highlight key features here and recommend reading the Whitepaper for a detailed explanation.
-
Maturity
Reliquary implements a novel time-weighted emissions model, which rewards stakers not only based on the size of their COMP holdings but also on the temporal maturity of their staked positions. This mechanism fosters deeper liquidity and incentivizes prolonged engagement using the system. The result is that COMP becomes a more robust long-term asset.
-
Composability
Reliquary leverages financial NFTs (Relics) to tokenize staked COMP positions, offering flexibility and composability. These Relics are fully transferable and can be traded on secondary markets without forfeiting accrued maturity or unstaking underlying capital, enabling seamless liquidity management and capital efficiency. This innovative approach allows stakeholders to optimize their portfolios while preserving the integrity of their time-weighted rewards, maximizing the capital efficiency of the staked COMP ecosystem.
-
Governance Integration
Reliquary’s design ensures that staked COMP retains full governance rights, allowing participants to vote while accruing staking rewards. The mechanism’s flexibility supports robust governance participation by aligning financial incentives with governance activity, ensuring that those who are most committed to the protocol’s long-term success have the greatest influence over its future direction.
-
Fee Distribution and Treasury Management
Reliquary has built in multi-rewarders that can distribute any number of assets. COMP governance can opt to distribute a portion of reserves as base assets, or can swap assets to stablecoins or COMP tokens for distribution as needed.
Reliquary allows for finite rewards emissions, meaning that the DAO can periodically refill the rewarder as needed according to Gauntlet recommendations, yield projections, or other relevant requirements.
-
Implementation (Maturity, Liquid Staking)
Implementation warrants further discussion, however we would like to share some ideas on how we can implement Reliquary for a staked COMP solution.
At a base level, users can stake their COMP tokens in Reliquary and receive a Relic that represents their position. We propose a 1.2x yield boost applied linearly over 6 months as a way to reward long term stakers (alternative to locking). Users can then trade their NFT Relics at a perceived maturity premium without unstaking their COMP tokens. This mitigates the need for locking and enhances user experience. (The system can offer a larger or smaller yield boost over a longer or shorter time, or different boost profile, depending on your priorities).
To achieve a liquid staked solution, we can tokenize Relic #1, freeze it at base maturity, and wrap it in a vault. This mitigates unfavorable maturity exposure as users enter and exit liquid staked positions. The vault would manage the Relic and compound rewards back into COMP and redeposit them into the Relic, increasing its underlying holdings relative to vault shares. Notably, if staked COMP is not locked, then liquid staked COMP can be arbitraged rapidly with no risk of prolonged depegging.
This simplifies cross-chain expansion as the liquid staked token can be bridged freely while compounding rewards on the original chain and appreciating via arbitrage.
Conclusion
At Conclave, we are excited about the potential to contribute to Compound governance and economic incentive framework with the introduction of Reliquary for staked COMP. We believe this approach can align long-term interests, deepen community engagement, and ultimately strengthen the entire ecosystem by giving staked COMP holders maximum flexibility.
We are eager to roll up our sleeves and work alongside the Compound community to bring this vision to life. By integrating Reliquary, we can create a more sustainable and impactful governance structure, helping Compound stay on the cutting edge of DeFi for many years to come.
Note: I’ve posted here to get the discussion on the foundational base of the potential design started but want to note that there are a number of potential system upgrades possible using this system that require specific requirements to address. On ongoing development partnership would be interesting to discuss if there is appetite for continued development post launch.
4 Likes