KVORUM
aaveSource: Aave governor v2 · #320

Chaos Labs Risk Parameter Updates _ Aave V3 Ethereum

executedSep 17, 2023, 01:20 AM UTCby320x329c…3ed4

01Summary

AISummary by Kvorumclaude-haiku-4-5

Proposal to optimize seven risk parameters across four Aave V3 Ethereum assets (DAI, wstETH, rETH, cbETH) based on Chaos Labs simulations. Changes include increasing Loan-to-Value ratios for staked ETH assets by 7.5-9.5%, adjusting Liquidation Thresholds by +1-3%, and raising DAI's Liquidation Penalty by 1%. Expected outcome: ~$2.5M increase in protocol borrowing with no increase in projected losses (VaR/EVaR).

Type
parameter change

Key changes

  • DAI Liquidation Penalty increased from 4% to 5% (+1%)(low)
  • wstETH Liquidation Threshold increased from 80% to 81% (+1%)(medium)
  • wstETH Loan-to-Value increased from 69% to 78.5% (+9.5%)(high)
  • cbETH Liquidation Threshold increased from 74% to 77% (+3%)(medium)
  • cbETH Loan-to-Value increased from 67% to 74.5% (+7.5%)(high)
  • rETH Liquidation Threshold increased from 74% to 77% (+3%)(medium)
  • rETH Loan-to-Value increased from 67% to 74.5% (+7.5%)(high)

Notable concerns

  • wstETH LTV increase of 9.5% exceeds typical ±3% change guideline mentioned in proposal
  • Parameter changes concentrated on staked ETH derivatives which may have correlated risk
ContractsAave V3 Ethereum - DAIAave V3 Ethereum - wstETHAave V3 Ethereum - rETHAave V3 Ethereum - cbETH
Generated by Kvorum
model: claude-haiku-4-5
provider: anthropic
profile: anthropic-fast-v1
prompt: v1.0
generated: 2mo ago
Read the description ↗

02Description

481 words

Simple Summary

A proposal to adjust seven (7) total risk parameters, including Loan-to-Value, Liquidation Threshold, and Liquidation Bonus, across four (4) Aave V3 Ethereum 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 parameter 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.

Please find more information on the parameter recommendation methodology here.

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.

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 optimize parameters for DAI, wstETH, rETH and cbETH on V3 Ethereum, resulting in improved capital efficiency of the system, with a negligible 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)

In addition, for wstETH, rETH, and cbETH, we recommend adjusting the LTV, as the current spread between the LTV to LT for these assets, which were originally set for the conservative V3 Ethereum launch, is unjustifiably large. The new spread will be similar to that of WETH.

Simulating all changes jointly yields a projected borrow increase of ~$2.5M, with no increase in VaR and Extreme VaR compared to simulations with the current parameters.

Liquidity Analysis:

Based on our analysis, we have determined that there is adequate on-chain liquidity to support any significant liquidations with the updated parameters.

Positions Analysis

We have not identified any outsized positions that are actively affecting our recommendations.

Recommendations

AssetParameterCurrentRecommendedChange
DAILiquidation Penalty4%5%+1%
WSTETHLiquidation Threshold80%81%+1%
WSTETHLoan-to-Value69%78.5%+9.5%
cbETHLiquidation Threshold74%77%+3%
cbETHLoan-to-Value67%74.5%+7.5%
rETHLiquidation Threshold74%77%+3%
rETHLoan-to-Value67%74.5%+7.5%

References

Copyright

Copyright and related rights waived via CC0.

03Mismatch analysis

AIMismatch analysis by Kvorumlow confidenceclaude-sonnet-5
Consistent

The proposal follows the standard Aave V3 governance pattern: a single on-chain action that calls execute() on a dedicated payload contract, with no decoded arguments exposed. The actual risk parameter changes (LTV, Liquidation Threshold, Liquidation Bonus for DAI, wstETH, cbETH, and rETH) described in the prose are expected to be hard-coded inside that payload contract's bytecode, consistent with the linked implementation contract in the References section. Because the decoded calldata only shows a generic execute() call with no parameter values, there is no way to directly verify that the seven specific percentage changes listed in the Recommendations table are the ones actually applied on-chain from this data alone. This is typical for this proposal type and not itself suspicious, but it does mean confidence in a line-by-line match is low rather than a hard confirmation. No contradictions, target mismatches, or extra/missing actions can be identified from the available decoded data, so there is no basis to flag a discrepancy — only a limitation in observability. Delegates relying solely on this decoded view should cross-check the linked implementation contract source code to confirm the exact parameter values before voting, but nothing here contradicts the description as written.

Generated by Kvorum
model: claude-sonnet-5
provider: anthropic
profile: anthropic-strong-v1
prompt: v1.0
generated: 2mo ago
Review the decoded actions ↗

04Actions

1 total
#0fallback()chain 0x1
Target
0xc52b…90d1
Value
0 wei
Function
fallback()
Arguments
{}
Calldata

05Tally

Summed from votes

Voters0
VP participating0
Outcome (current)—
Current support—

06Voters

0 total

No votes recorded yet.

07Forum

No forum thread is confidently linked to this proposal.