Has COMP been considered to be added as an asset? Obviously with a collateral factor of 0%. Personally I would love this, but I also think it may be a bad idea for the protocol.
Are there any guidelines on requirements to add an asset? If there isn’t it may be worthwhile to discuss what these should be to ease the many ‘what about token X’ posts that are surely coming…
Why do you say obviously 0% collateral? The biggest issue when adding new tokens are:
Once compound transitions to the open oracle, we will have a price feed for COMP. The rest is up to the community.
Ah, because of low liquidity you’d want to prevent liquidations as much as possible. I figured 0% collateral prevents liquidations due to the price going down. But now I see this isn’t really an issue. The real issue is a price rise and liquidators having to source COMP in a market with low liquidity. So you’re right… collateral factor isn’t so important. Now I see why sufficient liquidity is key. Thanks.
“Rule Number 4, I know you heard this before
Never get high on your own supply” -The Notorious B.I.G. from the Ten Commandments
Adding support for COMP is in line with the mission of the protocol; as with any asset, it requires a price feed (coming soon), sufficient market liquidity 
, and community interest (which, given the repeated chatter in Discord, seems extremely high).
The primary question that the community should ask, is how borrowed COMP could interfere (or, improve) Governance:
If I had to guess, Bart’s motivation is enabling new mechanisms to short COMP. That’s not a bad thing (efficient markets are strong markets), and a borrowing market for COMP could unlock new use-cases that haven’t been considered, too.
I think it might be good time to come back to discussing addition of COMP market, though i would like to look at it at different angle.
First i would like to address market liquidity. That’s a moot argument. It was presented with WBTC first AND proven wrong. There’s not sufficient market liquidity BECAUSE there’s no market implementation. So as long as there’s no market, liquidity WILL be low obviously. Was same with WBTC for month with liquidity never coming. Compound COULD fix it itself with enabling WBTC as collateral, but that was actually addressed first by MAKER team, by adding the WBTC vault, and as soon as there appeared usability, capital went there creating liquidity, hitting their vault caps multiply times.
Concern about WHAT if someone would be able to borrow COMP from market and use it for creating proposals/ voting. I guess we never know before it comes, AND the thing is IT DOESNT ALREADY MATTER what Compound do. Since CREAM.finance already implemented that market and it already can be used for that, and while liquidity might not be enough yet, it will get there eventually no matter if Compound itself create that market or not. Besides you don’t really need to borrow COMP, you can borrow some stable coin and temporarily buy COMP with that. And that not necessarily would be more expensive than borrowing COMP, likely even cheaper, might even end in profit if market buying push price high enough, and then you can slowly sell it back higher. So that possibillity kind of always was there, and probability will just increase with more COMP created and entereng market daily.
But that’s not my main point. I want to higlight the benefits of having COMP market.
Many people mentioned that it would be nice if Compound protocol would reward COMP holders. And i think this is great way how. Compound could adapt some of the methods “food tokens” utilise and benefit from them. If someone not aware how it’s done, the idea is quite simple: User provide COMP to supply side of COMP market and recieve cCOMP tokens. User than “stake” (timelock) that tokens and recieve reward in COMP distributed to the pool of that “staked” tokens. For visual example how it could work, you can check cream.finance and their native CREAM token. By that initiative basically several things are achieved, creating liquidity in supply side, locking that liquidity (user can’t unstake it prematurely and thus can’t remove COMP from supply pool) and rewarding of long term holding of COMP.
But more than that, i believe the system could be improved. Compound could have longer timelock on staked pool tokens. Like 3 months, 6 month, 1 year which could recieve bigger rewards for longer lock of tokens. Maybe even locked tokens could give weighted votes in governance. For example, 1 year locked COMP tokens could have 2x voting weight or something like that, ensuring that long standing commitment have a bigger voice out there.
Now from where COMP tokens to reward that initiative should come? Well that’s quite simple, from that same pool of tokens which are now distributing via liquidity mining initiative. I see couple of ways to do that, we can either reduce emission speed OR “cut the tail”, keep emission speed same but shorten the time, like not 4 years of distribution, but like 3 years, 9 months. Either way is fine and could be voted on details.
Maybe some of you noticed that i was strongly against reducing emission. Well i’m not against reduction itself, i’m against “just reduce it so we could possible use it for something else”. I believe we should FIRST have that something else, and then we could reduce emission to secure funding. But if there’s nothing else existing yet, than emission shouldn’t be touched yet as well.
And from financial stand point ability to “stake” cCOMP tokens timelocking them in staking pool will remove a portion of tokens from circulation on market likely helping the price. But that’s not really main point, rather an additional benefit for COMP community.
Last it’s a bit strange that Compound don’t support market for it’s own governance token. Is it not enough confidence in own token?
I’d like to get this done soon. I think we should deploy the current development branch version of CCompLikeDelegate and CErc20Delegator using this commit. It uses previously deployed code for the CErc20Delegator for reasons stated in this PR.
Things we need to discuss:
My opinion:
For the most part, I am basing these off existing non-stable cTokens.
Arr00,
This is exciting, thanks for stepping up to make a new cToken & proposal!
For Collateral Factors, historically this has been an analysis based on the liquidity and volatility of the asset. Gauntlet has done extensive work analyzing these risks for the Compound protocol (cc @tarun), and their input here could be very useful.
A Borrow Cap makes a lot of sense, and is a great way to test the new technology! This could start low and increase as the market scales, and there is more information on the market’s impact on Governance. Throwing a simple poll in to gauge the community’s thoughts:
0 voters
The USDT Interest rate model is a good place to start; it’s already deployed, and becoming canonical. That being said, more work is needed to analyze the optimal rate models for collateral assets.
In terms of Reserve Factors (always a contentious topic), 20% is lower than the RFs on collateral assets, but seems to be in-line with where the community wants to take RFs going forward for collateral.
I’m starting to think that might be a bit more tricky for governance token. To the point that we might be possibly interested in having 2 different markets for COMP tokens. In some way it applies to other governance tokens too, but since COMP is native governance token of Compound that should be our primary concern. My primary focus on having COMP market isn’t really about providing borrow opportunity, but more of:
And here we come to a problem if we are having a borrow side of COMP pool. We can’t preserve voting power for original owner, because the very same tokens could be borrowed from pool and borrower would expect voting power to come with them.
That could be addressed if we not allow cTokens from that pool to be borrowed. So basically they would be able to serve as collateral AND preserve voting power to original Supplier, as well as be timelocked in that pool, but by doing so we lose abililty for them to be borrowed.
Or we could allow them to be borrowed like traditional markets, but then supplier have to forfeit their voting rights for the duration of his supply.
Or maybe there could be 2 types of cComp, one with preservation of voting power and another with not. What is more important for future of protocol, preserving voting power for suppliers, or ability to borrow COMP as i don’t think we could have both in one cToken. Or am i missing something here?
Yes, I agree with your concerns; however, due to implementation constraints, enabling COMP as collateral while still using it for voting would be difficult and require a lot of new code.
The cCOMP market will have the ability to delegate its liquidity. If the community would like, we can setup the delegation to the community multisig and have off-chain voting based on a cCOMP snapshot. This will allow cCOMP holders to still have a say in governance. This is a solution to the issue you identified: maintaining voting ability while allowing borrowing.
In the future, I hope to see additional ways to use COMP in the Compound ecosystem such as staking COMP to receive more COMP. In this scenario, we will setup a system in which the user will maintain their COMP voting/delegation rights. I don’t know if it will be possible to use the staked COMP as collateral, but at the moment I assume not.
I have deployed the cCOMP contract using here. A new JumpRateModelV2 with the following parameters:
The implementation for this cCOMP contract is the same contract as is used by cUNI.