Proposal: Adjust ETH Interest Rate Model

Thanks to everyone who discussed this proposal on today’s developer community call!

Some highlights from the discussion:

  • @jmo suggested that the proposal could be submitted in multiple parts (eg split up the borrow cap from interest rate model change), which could potentially allow for faster action by the DAO
  • @rleshner discussed timing of when proposal would be submitted, and timing of when to submit follow up proposal to set rates back into a more normal range

I’m not great a paraphrasing so suggest listening to the recording if you’re interested :smiley:

4 Likes

There’s one more factor to consider, following the launch of Compound III, which might increase the need to adjust the v2 ETH interest rate model. More on this in a bit…

The logic for adjusting the ETH rate model is the expectation that:

  1. Users will want to hold ETH in their wallet in the moments before and during a fork in order to receive POWETH tokens.
  2. Compound v2 will likely not exist (or will instantaneously be liquidated) on a fork chain in which stablecoins are valueless. A miner, in the first block, will find a way to extract the POWETH.
  3. ETH supplied to Compound v2, logically, will be withdrawable on Mainnet, but will likely be frozen on a POWETH fork.
  4. Knowing this, during the Merge, users might want to withdraw ETH from Compound v2, in order to receive POWETH.

Now, back to Compound III. One of the unique advantages of the upgrade is that collateral is isolated; it remains your property, and can never be withdrawn by other users (except during liquidation).

  1. ETH supplied to Compound III likely will be withdrawable on a POWETH fork; USDC collapsing won’t liquidate borrowers (in fact, the opposite).
  2. Knowing this, during the Merge, users might want to supply ETH to Compound III, in order to simultaneously borrow USDC and receive POWETH.
  3. This could further exacerbate an ETH liquidity crisis on Compound v2.

Given this new possibility, I feel more confident acting earlier, and with more conviction, in updating the Compound v2 ETH interest rate models & borrowing caps. There’s a little over two weeks until the Merge (at the shortest end)–excited for this to come to a formal proposal soon.

3 Likes

Hi all,

With the merge fast approaching, and the new cETH interest rate model deployed, I’m planning to submit the on chain proposal for cETH merge risk mitigations later today. Below is a preview of proposal actions and text.

Proposal Actions

Set ETH borrow cap of 100,000 ETH

Comptroller > _setMarketBorrowCaps > cETH > 100,000E18

Update ETH interest rate model with the following parameters

  • Borrow rate at 0% utilization: 2%
  • Optimal utilization (kink): 80%
  • Borrow rate at optimal utilization: 20%
  • Borrow rate at 100% utilization: 1000% APR

Note that 1000% borrow APR yields a maximum compounded APY of ~2.2 million %, or roughly 2.8% borrow cost per day. The block time assumed by the interest rate model has also been changed to 12 seconds (from 15 seconds previously).

cETH > _setInterestRateModel > 0xF9583618169920c544Ec89795a346F487cB5a227

Transfer 100 COMP to Maker governance timelock at 0xBE8E3e3618f7474F8cB1d074A26afFef007E98FB

COMP > Transfer > 0xBE8E3e3618f7474F8cB1d074A26afFef007E98FB > 100E18

Transfer 45 COMP to @arr00 at 0x2b384212edc04ae8bb41738d05ba20e33277bf33

COMP > Transfer > 0x2b384212edc04ae8bb41738d05ba20e33277bf33 > 45E18

Summary

This proposal will make several changes to the cETH market on Compound v2 in preparation for the upcoming Ethereum merge and switch to POS consensus. This includes setting a borrow cap of 100,000 ETH, as well as updating the interest rate model to a jump rate model with much higher rates after exceeding 80% borrow utilization, up to a maximum of 1000% APR.

Reasoning

The upcoming merge has the potential to cause disruption to ETH lending markets due to the possibility of receiving airdrops of ETH fork tokens. This may incentivize excessive borrowing from ETH lending pools, which leads to negative user experience for depositors who cannot withdraw funds when utilization reaches 100%, as well as safety concerns due to potential to interfere with liquidations of ETH collateralized positions. The proposed changes should help reduce risk of the cETH market reaching 100% utilization. See the forum discussion for further details.

The proposal includes payments to proposal contributors, including 100 COMP to MakerDAO governance timelock for Block Analitica (Maker risk) team’s analysis, as well as 45 COMP to Arr00 for development work.

3 Likes