Given the recent volatility on stablecoins, we wanted to give the community a heads-up that Gauntlet is conducting analysis. Action could be required both from on-chain governance and Guardian via emergency because of bad debt risk. We will keep the community informed of recommendations.
Gauntlet Compound USDC Depeg Analysis
We recommend temporarily pausing borrows for all assets on Compound V2.
USDC supply on Compound V2 was recently paused, but existing USDC suppliers are still able to borrow against their collateral.
As seen below, the vast majority of USDC suppliers ($460M total) either recursively borrow USDC/USDT/DAI, or don’t borrow any assets.
The main way users currently utilize the V2 protocol is by borrowing stablecoins against blue chip tokens (as offered in V3), as opposed to borrowing volatile assets against USDC. As seen below, roughly 96% of borrows on V2 are stablecoins, either recursively borrowed, or mostly borrowed against assets available in V3.
USDC Collateral Factor
85.50% → 80.00%
DAI Collateral Factor
80.00%
USDT
70%
1 Like
Would like to see a MUCH higher top interest rate on borrowing USDT on v2. Liquidity is currently approximately zero, and likely to stay there for until the dust settles since the max APY is so low. AAVE’s much higher max rates have kept liquidity mostly open there.
1 Like
It looks like in case of recursive borrowing more concern might be the opposite. Suppliers of DAI, borrowing usdc, as while dai price depegs together with usdc, usdc price oracle is fixed to 1$, while dai is not, which might lead to potential liquidations.
Yet, still USDC and its oracle price present quite a potential problem, as it while not looks very probable, but there is a possibility of creating insolvent positions if USDC dips far enough from peg, while protocol still treat it as 1$. And in that case we might fully expect a run on a protocol with providing usdc collateral with a purpose of borrowing either stable or quality assets like wbtc or eth. Since USDT is pretty much all already borrowed eth might be of a concern. Maybe we should consider temporary lower borrow cap of eth.
1 Like
Gauntlet Update 2023-03-12
Key Metrics
Below we present some key metrics regarding the depeg event as of 2023-03-12.
| Current Loss | Current MTM Loss | Current USDC Price |
|---|---|---|
| $0 | $0 | $0.97 |
| Current Ethereum USDC Liquidity | USDC Liquidity WoW |
|---|---|
| $37.9mm* | 45.4%** |
*(2% DEX price depth)
**(pcnt of liquidty compared to 7d ago)
Key Risks
USDC price falls and stabilizes at less than $0.855
In the case that the USDC price falls and stabilizes below $0.855, existing Compound v2 USDC suppliers would still be able to utilize the hardcoded USDC oracle price of $1.00 and the USDC collateral factor of 85.5% to borrow against their supply, thereby leaving the protocol with insolvencies.
Current Gauntlet Recommendations
If the community believes that USDC price has stabilized and there’s a trivial (< 1%) chance USDC price decreases below $0.855, then Gauntlet recommends no changes to the v2 protocol. Otherwise, we recommend pausing borrows for all assets or decreasing USDC collateral factor. The protocol is still vulnerable to users borrowing against USDC collateral if the price drops below $0.855.
Current Initiatives
| Initiative | Status | ETA | Updates |
| Market Monitoring | In Progress | 3/14/23 PM* | Market montioring is part of our Risk Management service. We are monitoring all markets closely and will notfiy stakeholders as markets change. |
| Risk Recommendations | In Progress | 3/14/23 PM* | We will continue to provide recommendations to the community and Guardians as conditions change in case drastic action needs to be taken. |
| Conversations to explain risk to stakeholders | Complete | 3/11/23 AM | |
| Inital Analysis | Complete | 3/10//23 PM | Our initial analysis was complete late evening on 3/10/2023 and communicated to stakeholders |
| Initial forum post | Complete | 3/10/23 PM | Link here |
-
*This is the current best guess for when we will get clarity on USDC value - we will monitor and provide recommendations as necessary beyond that date
Appendix
USDC supply analysis
Entire supply positions of top 30 v2 USDC suppliers on 2023-03-09
Entire borrow positions of top 30 v2 USDC suppliers on 2023-03-09
Entire supply positions of top 30 v2 USDC suppliers on 2023-03-11
Entire borrow positions of top 30 v2 USDC suppliers on 2023-03-11
USDC supply in Compound v2 has decreased since the depeg, which decreases risk to the protocol, especially given new USDC supply is paused.
The largest recursive USDC borrower ($72.6M supplied, $61.6M borrowed) with address 0x46cc9435ce65206a1b55caffeb9c14e5adc355ef exited the protocol.
The second largest USDC recursive borrower with address 0x342491c093a640c7c2347c4ffa7d8b9cbc84d1eb decreased their supply from 58.4M tokens to 41.5M tokens
The largest non-recursive USDC supplier with address 0xe2773fb045e53de5344f245e03ea614af1064ce3 decreased their USDC supply from 30.2M tokens to 7.4M tokens, along with other top non-recursive suppliers.
The user with address 0xc2f61a6eeec48d686901d325cde9233b81c793f3 who supplies 20.725M USDC tokens and borrows 17.1M USDT tokens poses the most risk to the protocol in the case that USDC depegs below $0.855. Given the USDC price of $0.96, this user actually has a borrow usage of > 100% and would be liquidatable if the USDC oracle price wasn’t hardcoded at $1.00, and would be insolvent if the USDC price decreases below $0.825. One risk mitigation strategy is to decrease USDC collateral factor, to provide an extra buffer against USDC depeg insolvencies. However, in doing so, this account would become liquidatable, and might cause additional sell pressure for USDC.
In a follow-up post we will discuss how the recent changes in USDC and DAI supplies have affected the v2 → v3 migration.
5 Likes
Can Compound v2 switch to using the chainlink USDC oracle?
1 Like