Supporting the Compound Ecosystem

Last week I wrote a post about an idea for Compound Governance to support the growth of the PoolTogether protocol. A strong consensus emerged against that specific implementation but many seemed supportive of the general principle of helping support PoolTogether and thereby also help the Compound protocol grow.

After talking to members of the community, I have a new proposal. This proposal achieves the same end goal of supporting growth in a way that doesn’t use a limited resource (COMP) and costs the Compound protocol much less.

The proposal is for the Compound protocol to “sponsor” the USDC prize savings pool on PoolTogether with $10 million of USDC.

“Sponsorship” is money deposited into PoolTogether savings pools that contributes interest to the savings prizes but is NOT eligible to win. This raises the expected value for all depositors and helps kick-start a reflexive growth loop for PoolTogether (larger prizes create more deposits, more deposits create larger prizes). It’s important to note the sponsorship can be removed at any time! So the value is not locked into PoolTogether. At the end of the day, all this capital is going into Compound so it’s good for Compound as well.

Execution
The actual proposal would do the following:

  1. Send COMP from the Comptroller to the Timelock
  2. Mint cCOMP with the COMP
  3. Borrow $10 million USDC using cCOMP as collateral
  4. Deposit USDC into sponsorship of USDC savings pool

Risks

  1. The USDC deposited into the PoolTogether would be subject to the smart contract risk of the PoolTogether contracts. This risk is very low based on a few factors. 1) The PoolTogether smart contracts are non-upgradeable. 2) The USDC prize pool contract has been live on main net since January 11th 2021, securing over $80 million dollars with no exploits of any kind. Contracts using the same code but different asset types have been live for even longer. 3) the prize pool contract has been professionally audited by Consensus Diligence. 4) The prize pool has no external dependencies (no oracle risk). 5) The team has an excellent security track record and no smart contracts related to PoolTogether have ever been exploited.

  2. The only additional risk is that the loan could be liquidated if COMP price majorly decreased. This risk can be well managed by ensuring the liquidation price is very low.

Benefits
Reiterating the introduction. This is a low cost and low risk way for the Compound protocol to incentive growth via integration partners. As a protocol, making sure strong third party interfaces exist is crucial to the long term health of the Compound ecosystem.

8 Likes

I would like to highlight that the $10 million USDC will be redeposited back into Compound. I think this is a good proposal.

This is a creative approach to supporting PoolTogether, and could become a template for supporting other applications or use cases with liquidity. As an experiment for the Compound protocol & stake-holders, I’m a huge fan of this approach; it puts the protocol/treasury/community resources to work directly.

A few questions/considerations:

  • The “cost” of this proposal to COMP token-holders would include the USDC interest paid to the protocol; this isn’t free. For $10M, at a current interest rate of 4.6%, it would cost the protocol/token-holders 460k USDC/yr. Not that this is necessarily an issue, but should be paid attention to; the Timelock would have to repay additional USDC before being able to withdraw the COMP.
  • How would this impact the cCOMP market, or USDC market? Given the size, it’s not necessarily material, but should be paid attention to.
  • When would the sponsorship USDC be returned to the protocol? Does the Compound protocol receive POOL? How does Compound enter into an agreement with another protocol? Who administers this? Is there any recourse, if the USDC isn’t returned?
  • What proposal actions would need to occur? I believe the Timelock would need to ERC-20 approve COMP to cCOMP, etc.
  • What alternatives exist, from the Compound protocol’s perspective? Selling COMP for USDC, to sponsor PT? Withdrawing USDC reserves, to sponsor PT? Are these more attractive?

Bootstrapping partner liquidity is certainly an interesting experiment to explore. I’m a fan of trying something along these lines.

6 Likes

I love this idea, and I think it could have many other awesome applications. You have my support!

1 Like

The deposited COMP as collateral would earn interest and COMP too, this might compensate the USDC interest paid in value.

The question is that should this process claim COMP after the deposited COMP or leave it.

I think the idea is great.

1 Like

Any COMP earned by the protocol effectively “lowers” the total COMP Distribution / Day.

I love the creative idea. Certainly needs and economic analysis. I can do it with a grant. (
Let me know if I should apply for this or I can help out. Possibly needs also needs a legal look? I’m just thinking on the top of my head. But I love this creativity. Maybe a dedicated committee to deal with “investments” like this. I know governance is right for the self driving banks but we need “third party” due diligence.

Great questions! Here are my thoughts:

The “cost” of this proposal to COMP token-holders would include the USDC interest paid to the protocol; this isn’t free. For $10M, at a current interest rate of 4.6%, it would cost the protocol/token-holders 460k USDC/yr. Not that this is necessarily an issue, but should be paid attention to; the Timelock would have to repay additional USDC before being able to withdraw the COMP.

Yes, good to acknowledge this isn’t free. As @blck noted it would be somewhat offset by yield on COMP and also high probability it would be completely covered by accrued POOL (more on that later).

How would this impact the cCOMP market, or USDC market? Given the size, it’s not necessarily material, but should be paid attention to.

The thing to pay attention to here is the supply side on the COMP market. That is currently $411 million and to do this would likely want to supply around ~30 million to have a nice buffer. So that is a decent size. On the USDC market it will be immaterial

When would the sponsorship USDC be returned to the protocol? Does the Compound protocol receive POOL? How does Compound enter into an agreement with another protocol? Who administers this? Is there any recourse, if the USDC isn’t returned?

To clarify my proposal. Compound governance would maintain complete control of these funds and could unwind the position at anytime. I would advocate for a one year timeline as the assumed deposit length which could be unwind early or extended if desired.

Although currently “sponsors” do not accrue POOL I do think that will change and Compound governance would accrue POOL. There has been a consensus in the PoolTogether community to start rewarding sponsors with POOL and although this hasn’t been implemented yet I expect it will be.

I do not believe any type of agreement would be necessary between the protocols. We could do something but I think it would be unneeded overhead.

Finally, in terms recourse if USDC isn’t returned. PoolTogether governance would have no way to stop the funds from being withdrawn so this would only happen if PoolTogether protocol was hacked and the funds were stolen. If this were to happen Compound governance should assume there is no recourse to get funds back.

What alternatives exist, from the Compound protocol’s perspective? Selling COMP for USDC, to sponsor PT? Withdrawing USDC reserves, to sponsor PT? Are these more attractive?

Withdrawing USDC reserves or selling COMP to USDC would both be viable options. Personally I don’t think withdrawing USDC reserves is the best though. We should keep those in the protocol at least for now. Selling COMP to diversify might be the right option but is a very different discussion.

Overall, I would like to think about how we could get ~$250 million USDC “Compound Liquidity” fund and deploy it across multiple protocols building on Compound.

2 Likes

I’ll do a quick presentation on this at this week’s developer call. I don’t think it’s needs a grant at this stage, better to let it develop a bit more. Let’s wait for the call to determine next steps.

1 Like

I think you killed it at the presentation. Compound liquiduty

Also, sweep sai and rep reserves

On the point around alternatives - would it be helpful to consider kicking off a model where Pooltogether (or any partner protocols) could work on setting up ‘limited/whitelisted’ native token pools as a basis for setting up borrowing markets?

The intention would be to allow for:
i) Compound to run essentially a pilot program to kick-start new money-markets with partner protocols and build up reserve proceeds before opening these up to the broader market
ii) Enable partner protocols to tap into a line of diversification (beyond outright sales) and allow them to start thinking about filling in operational/working capital needs
iii) Potentially provide a good use for staked COMP to serve as a backstop for higher-risk/reward yields

Could have limits imposed pertaining to collateral and reserve factors as well as limits on the money-market sizes for these.

2 Likes