This report builds upon Gauntlet’s first Monthly Risk Review, with the goal of keeping the community informed of market risk pertaining to Compound. So far, Gauntlet has executed 5 sets of parameter recommendations as part of our Dynamic Risk Parameters engagement. This month, Gauntlet has executed 3 sets of recs and has implemented nine (9) parameter suggestions across eight (8) assets.
| Old Collateral Factor | New Collateral Factor |
|---|---|
| ETH (75%) | 80% |
| DAI (75%) | 80% |
| WBTC2 (65%) | 70% |
| UNI (60%) | 70% |
| BAT (65%) | 65%** |
| ZRX (65%) | 60% |
| COMP (60%) | 65% |
| LINK (65%) | 70% |
**Represents parameters where Gauntlet has changed collateral factors two times.
In October, Gauntlet conducted a poll to gauge the Compound community’s risk tolerance. The community chose a Moderate risk level and since then, Gauntlet’s parameter updates have been designed to align with this risk level.
Gauntlet has launched the Compound Risk Dashboard and has invested in the infrastructure to enable daily updates. The community should use the dashboard to better understand the updated parameter suggestions and general market risk in Compound. Gauntlet is keen on improving the dashboard and has been conducting user studies with members of this community and others to inform the next iteration.
Gauntlet is currently backtesting its elasticity assumptions by analyzing on-chain changes in borrow behavior in response to our passed collateral factor changes. Gauntlet is also formulating a reserve factor recommendation methodology. This methodology incorporates several important considerations including the tradeoffs between covering insolvencies and promoting protocol growth.
Parameter Update Forum Posts
Risk Parameter Updates 2021-11-17
Risk Parameter Updates 2021-11-30
Risk Parameter Updates 2021-12-11
Compound Governance Proposals
C071: Risk Parameter Updates for DAI, BAT, ZRX, and ETH
- Passed with 1,006.1K COMP
C072: Risk Parameter Updates for WBTC, UNI, and COMP
- Passed with 1,045.6K COMP
C074: Risk Parameter Updates for WBTC, BAT, and LINK
- Passed with 907.2k COMP
Highlights
- Unlocked $605.25M of additional borrow, leading to $1.843M increase in annual income to the Compound protocol
Over the past month, Gauntlet has continued to drive meaningful value to the Compound community. This month, Gauntlet has increased collateral factors for ETH, DAI, WBTC2, UNI, COMP, and LINK. Our analysis showed that we can safely increase collateral factors for these assets and thus improve capital efficiency for the protocol. Assuming elasticity (if collateral factor increases by x percentage points, borrow amount increases by x percentage points as well), we expect there to be a $605.25M increase in total borrows driven by our collateral factor increases. This increase in total borrows will drive an additional $1.843M of annual income towards reserves. We calculate the $1.843M conservatively by assuming borrowers will use their collateral to borrow USDC, which has the lowest Borrow APY of the 3 stablecoins which account for 94% of total borrows. We also conservatively assume the USDC borrow rate will not increase, despite utilization increasing as a result of increased borrows. See the below figure for more detail. In addition, the reserves act as insurance against borrower default, thus, the increase in reserves also makes protocol safer. As previously mentioned, Gauntlet is refining its methodology of adjusting reserve factors, which will drive more value to the Compound protocol.
Reserve Impact Calculations:
Assumes elasticity of borrow in relation with collateral factor increases.
The recent market crash played a role in increased liquidations and higher VaR. Since publishing our previous risk review last month (11/23 to 12/15), there have been $23.849M in Compound liquidations, compared to $2.236M in liquidations during the same time period prior (11/2 to 11/20). Additionally, VaR increased to over $1B on December 10. Falling asset prices resulted in a lower aggregate protocol collateralization ratio (driven by riskier ETH positions) and increased volatility.
Liquidation breakdown by collateral (11/23 to 12/15):
Time series of collateral-specific volatility:
Time series of collateral-specific protocol collateralization ratios:
Breakdown of the largest non-recursive borrow accounts:
Time series of % of total borrows which are non-recursive:
While improving capital efficiency and unlocking borrows by an additional $605.25M, Gauntlet’s recommendations have maintained risk levels in the protocol. On average, Gauntlet’s recommendations have only increased VaR by 8.0%, with absolute risk amounts maintained at reasonable levels. Our simulations most recently observed 5.1% of Compound’s TVL being at risk on that day.