KVORUM
aaveSource: Aave governor v2 · #272

Chaos Labs Risk Parameter Updates - Aave V2 Ethereum

executedJul 16, 2023, 02:29 AM UTCby320x329c…3ed4

01Summary

AISummary by Kvorumclaude-haiku-4-5

Proposal to reduce risk on 11 frozen Aave V2 Ethereum collateral assets (BAL, BAT, CVX, DPI, ENJ, KNC, MANA, REN, xSUSHI, YFI, ZRX) by lowering Liquidation Thresholds and Loan-To-Value ratios to zero, while increasing Reserve Factors by 10% and Liquidation Bonuses to 10% (except BAL and CVX). This first step aims to gradually phase down low-to-medium cap frozen markets and reduce borrowing power.

Type
parameter change

Key changes

  • Liquidation Threshold reductions across 11 assets (BAL from 70% to 55%, BAT from 70% to 52%, CVX from 45% to 40%, DPI from 70% to 42%, ENJ from 67% to 60%, KNC from 70% to 1%, MANA from 75% to 62%, REN from 60% to 40%, xSUSHI from 65% to 60%, YFI from 65% to 55%, ZRX from 65% to 45%)(high)
  • Loan-To-Value ratios set to 0% for all 11 assets to eliminate borrowing against these collaterals(high)
  • Reserve Factors increased by 10% across all assets (BAL, BAT, CVX, DPI, ENJ, KNC, REN, YFI, ZRX from 20% to 30%; MANA and xSUSHI from 35% to 45%)(medium)
  • Liquidation Bonus increased to 10% for 9 assets (BAT, DPI, ENJ, KNC, MANA, REN, xSUSHI, YFI, ZRX); BAL and CVX remain at 8% and 8.5% respectively(medium)

Notable concerns

  • LT reductions will trigger liquidations across affected accounts: KNC liquidations totaling $1,790 across 7 accounts, REN $1,800 across 6 accounts, BAT $610 across 6 accounts, MANA $500 across 5 accounts, BAL $60 across 5 accounts, ZRX $500 across 5 accounts, ENJ $300 across 4 accounts, YFI $500 across 2 accounts
  • Aggressive LT reductions particularly for KNC (70% to 1%) represent significant risk parameter changes that may disproportionately affect certain users
  • Users may be liquidated upon proposal approval; Chaos Labs will publicly communicate affected accounts before on-chain execution
ContractsAave V2 Ethereum - BAL collateralAave V2 Ethereum - BAT collateralAave V2 Ethereum - CVX collateralAave V2 Ethereum - DPI collateralAave V2 Ethereum - ENJ collateralAave V2 Ethereum - KNC collateralAave V2 Ethereum - MANA collateralAave V2 Ethereum - REN collateralAave V2 Ethereum - xSUSHI collateralAave V2 Ethereum - YFI collateralAave V2 Ethereum - ZRX collateral
Generated by Kvorum
model: claude-haiku-4-5
provider: anthropic
profile: anthropic-fast-v1
prompt: v1.0
generated: 2mo ago
Read the description ↗

02Description

751 words

Simple Summary

A proposal to reduce the Liquidation Threshold (LT) and Loan-To-Value (LTV) and increase Liquidation Bonus (LB) and Reserve Factors (RF) for eleven (11) frozen collateral assets on Aave V2 Ethereum.

Motivation

In line with our ongoing commitment to reduce the potential risks associated with V2 markets, we suggest a phased reduction of the frozen V2 markets, comprising low-to-medium cap assets with limited liquidity. The following proposal is the first step to reduce capital efficiency across V2 collateral assets, which will be followed up with more detailed plans for additional parameter updates for these assets.

LT Reductions

Implementing a gradual decrease in LTs effectively diminishes the borrowing power for the existing markets.

  • The community has aligned on the aggressive option, which suggests a LT configuration that optimizes reductions without significantly increasing the number of accounts eligible for liquidation. The proposed values are set at a margin of 8% from the closest LT figure which would trigger more substantial liquidations, as can be seen in the supporting graphs here.
AssetCurrent LTRec LTValue LiquidatedAccounts Liquidated
BAL70%55%$605
BAT70%52%$6106
CVX45%40%01
DPI70%42%00
ENJ67%60%$3004
KNC70%1%$1,7907
MANA75%62%$5005
REN60%40%$1,8006
xSUSHI65%60%00
YFI65%55%$5002
ZRX65%45%$5005

As Liquidation Threshold reductions may lead to user accounts being eligible for liquidations upon their approval, we want to clarify the full implications to the community at each step. Chaos Labs will publicly communicate the planned amendments and list of affected accounts leading to the on-chain execution.

Reduce LTV to zero

Lowering the Loan-to-Value (LTV) ratio to zero can potentially curb borrowing by less sophisticated users. It's important to note that due to V2's design, users can bypass the LTV limit to borrow up to the LT value.

RF Increase

Progressively increasing the reserve factors diminishes the interest rate for providing these assets and lowers the suppliers' incentive to supply the assets. In this proposal, we suggest an initial hike of 10% for all inactive assets.

LB Increase

We recommend increasing the LB for all proposed assets (excluding BAL and CVX) to 10% to better incentivize liquidations and accommodate for low liquidity for the given assets.

Specification

AssetCur LTRec LT*Current LTVRec LTVCurrent RFRec RFCurrent LBRec LB
BAL70%55%65%020%30%8%8%
BAT70%52%65%020%30%7.5%10%
CVX45%40%35%020%30%8.5%8.5%
DPI70%42%65%020%30%7.5%10%
ENJ67%60%60%020%30%6%10%
KNC70%1%60%020%30%10%10%
MANA75%62%61.5%035%45%7.5%10%
REN60%40%55%020%30%7.5%10%
xSUSHI65%60%50%035%45%8.5%10%
YFI65%55%50%020%30%7.5%10%
ZRX65%45%55%020%30%7.5%10%

References

Copyright

Copyright and related rights waived via CC0.

03Mismatch analysis

AIMismatch analysis by Kvorumlow confidenceclaude-sonnet-5
Consistent

The decoded calldata consists of a single opaque call to execute() on what is the standard Aave V2 governance payload contract address. This is the typical pattern for Aave risk parameter proposals: the actual per-asset LT/LTV/RF/LB updates are encoded inside the payload contract's bytecode (linked in the 'Payload' reference) rather than being visible as discrete decoded arguments in this action list. Because the decoded action provides no visible parameters (asset addresses, percentages, etc.), it is not possible to directly verify that the specific numeric values in the description (LT reductions, LTV to 0%, RF +10%, LB to 10%) match what the payload contract executes on-chain. Nothing in the decoded data contradicts the description, and the pattern (one execute() call routing to a dedicated payload) is standard and expected for this proposal type, so there is no basis to flag a discrepancy. However, since the calldata itself cannot be used to confirm the specific parameter values claimed in the prose, confidence in this comparison is low rather than high.

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
0x6aec…1a6d
Value
0 wei
Function
fallback()
Arguments
{}
Calldata

05Tally

Summed from votes

Voters0
VP participating0
Outcome (current)—
Current support—

06Voters

0 total
Filter:
VoterChoice% of totalRationale
320x329c…3ed4For242.9K0.00%—3y ago
070x0703…961bFor109.3K0.00%—3y ago
730x732a…b848For106.4K0.00%—3y ago
130x13bd…8548For82.2K0.00%—3y ago
850x8556…ab0bFor69K0.00%—3y ago
E40xe459…e666For51.9K0.00%—3y ago
B90xb933…8dd1For42K0.00%—3y ago
EA0xea17…6676For15.3K0.00%—3y ago
180x1855…98baFor3K0.00%—3y ago
A00xa06c…61f7For604.60.00%—3y ago
050x0527…8469For4030.00%—3y ago
5E0x5e09…c634For350.00%—3y ago
DF0xdf06…4977For27.30.00%—3y ago
5F0x5f38…f51fFor3.10.00%—3y ago
8B0x8bc8…7276For1.20.00%—3y ago
740x74a2…c125For10.00%—3y ago
670x671f…634fFor10.00%—3y ago
B30xb39b…8ecfFor0.20.00%—3y ago
2F0x2fda…6a16For0.10.00%—3y ago
340x3496…126bFor0.10.00%—3y ago
500x50d2…5318For0.10.00%—3y ago
240x2474…acf8For00.00%—3y ago
Showing 22

07Forum

No forum thread is confidently linked to this proposal.