[ARFC] - Chaos Labs Risk Parameter Updates - Aave V3 Optimism - 2023.03.22

Risk
1

Simple Summary

A proposal to adjust five (5) total risk parameters, including Loan-to-Value, Liquidation Threshold, and Liquidation Bonus, across two (2) Aave V3 Optimism assets.

Motivation

Chaos Labs’ Parameter Recommendation Platform runs hundreds of thousands of agent-based off-chain and on-chain simulations to examine how different Aave V3 risk parameters configurations would behave under adverse market conditions - and find the optimal values to maximize protocol borrow usage while minimizing losses from liquidations and bad debt.

Note: As a general guideline, we limit the proposed changes by ±3% for all parameters as a high/low bound for a given proposal. This ensures more controlled changes and allows us to analyze their effect on user behavior before recommending further amendments to the parameters if the optimal configuration is outside this range.

Please find more information on the parameter recommendation methodology here.

You can also view the simulation results and breakdown for the different assets by clicking on them on this page.

The output of our simulations reveals an opportunity to increase LTVs and LTs for WBTC and DAI on V3 Optimism, resulting in improved capital efficiency of the system, with no effect on the projected VaR (95th percentile of the protocol losses that will be accrued due to bad debt from under-collateralized accounts over 24 hours) and EVaR (Extreme VaR, the 99th percentile of the protocol losses that will be accrued due to bad debt from under-collateralized accounts over 24 hours)

Simulating all changes jointly yields a projected borrow increase of ~$155,000, with no increase in VaR and Extreme VaR compared to simulations with the current parameters.

Untitled (30)
Untitled (30)2686×1318 239 KB

Untitled (31)
Untitled (31)2688×1334 228 KB

Liquidity Analysis:

The TVL in Aave on Optimism (118.71M$) is concentrated primarily on three blue-chip, high-liquidity assets, which together account for ~75% of the supplied assets in $ values: USDC (34%), WETH (28%), and WBTC (14%). Based on our analysis, we recommend increasing the Loan-To-Value (LTV) and Liquidation Threshold (LT) for WBTC and DAI, as we have determined that there is adequate on-chain liquidity to support significant liquidations. Similarly, we recommend reducing the Liquidation Penalty for WBTC, which is set conservatively at 10% to a lower level (8.5%) closer to that of USDC and Ethereum (5%).

Positions Analysis

We have not identified any outsized positions that are actively affecting our recommendations. However, we should call out that ~40% of the on-chain supply of WBTC on Optimism is currently held on Aave, which calls for some caution. We have already proposed and implemented an amendment to the supply cap to limit the concentration of liquidity on Aave.

Given the blue-chip nature of BTC/WBTC, we expect significant arbitrage buying pressure to offset any slippage that may be created by future liquidations of WBTC and, therefore, recommend reducing the Liquidation Penalty even with the current significant liquidity concentration on Aave. We expect a lower LP to drive liquidators to liquidate positions in smaller portions, pacing the process and allowing the market more time to react.

Recommendations

Asset Parameter Current Recommended Change
WBTC Liquidation Threshold 75% 78% +3%
WBTC Loan-to-Value 70% 73% +3%
WBTC Liquidation Penalty 10% 8.5% -1.5%
DAI Liquidation Threshold 80% 83% +3%
DAI Loan-to-Value 75% 78% +3%

Next Steps

We invite a community discussion around the proposed parameter updates and will follow up with a Snapshot vote in 5 days, barring community objection.

1 Like
2

We have published a Snapshot for the community to vote on.

Start Date: Tuesday, Mar 28, 2023, 3:00 PM
End Date: Wednesday, Mar 31, 2023, 3:00 PM

Thank you in advance for your participation in the vote.

1 Like
3

Gauntlet Analysis

Simple Summary

For conservatism, the community may wish not to move forward with the above DAI LTV and LT increases. We flag several potential risks about increasing the DAI LTV and LT suggested in this forum post.

Analysis

Under extreme conditions, the LT increase makes it potentially cheaper to force a situation in which abnormal and/or cascading liquidations occur. We’ll outline a potential abnormal situation below.

The DAI LT increase from 80% to 83% increases the leverage multiplier from 5x to 5.9x for recursive short attacks, potentially making an attack 20% cheaper to execute. The least liquid borrowable token on Aave v3 Optimism is sUSD. In the event there is SNX volatility excess of the market, or idiosyncratic, unquantifiable Synthetix protocol abnormalities, sUSD may experience more difficulty in maintaining its peg.

If a bad actor borrows sUSD against DAI and market sells the sUSD, it can shock sUSD. In the event of exogenous, idiosyncratic sUSD volatility, this shock could further affect any existing downwards pressure on sUSD and make it more difficult for the sUSD redemption mechanism to function properly. The increased liquidations can trigger cascading liquidations or lead to similar insolvencies to the USDC non-parity in March.

2 Likes
4

The risk mentioned in the comment is not limited to the proposed action of increasing the LT for DAI, and the recommendation itself does not pose any new risk to the Optimism deployment. A malicious actor can exercise this vector by utilizing USDC with a higher LT (85%) or E-Mode, which sUSD is a part of, with an LT of 97.5%. A proper mitigation for the risk described could be to minimize the borrow cap for sUSD. However, at this time, dumping the entire borrow cap of sUSD (13M) leads to minimal price impact and we do not recommend an immediate change to the cap.

Chaos Labs considers the potential for these types of attacks in our analyses and maintain our recommendation to increase capital efficiency with no projected increase in VaR and EVar. We plan to publish the AIP for voting early next week.

Screen Shot 2023-04-12 at 12.42.55 PM
Screen Shot 2023-04-12 at 12.42.55 PM872×952 57.8 KB

Additional note: This proposal has been on the forums for 3 weeks, has been approved via a Snapshot vote, and is ready for the on-chain vote. This is not the first time Gauntlet’s concerns have been raised at the last minute. The governance framework purposely incorporates multi-day buffers to facilitate discussions and serve as a feedback period. Ignoring these periods and responding as proposals converge creates confusion and extra work for governance voters and contributors alike. There are cases in which this is warranted, specifically when major circumstance shifts occur between a Snapshot and AIP. However, this has not been the case recently, including here, and concerns, even if unfounded, could have been raised earlier. Our communication channels are open, and we are happy to assist Gauntlet or any other contributors if extra time is needed. We continue to welcome collaboration across risk protocol contributors and a streamlined communication process, using governance frameworks and timelines as a baseline for what is expected to advance the speed and efficiency of Aave’s governance process.

1 Like
5

We have published AIP-206 to update risk parameters on V3 Optimism, starting in 24h.

Thank you in advance for your participation in the vote.

1 Like
6

** :upside_down_face:*

emphasized text