Aave is one of the most popular lending and borrowing protocols (aka “liquidity protocols”) within the decentralized finance (DeFi) space, allowing users to borrow cryptocurrency by putting down another cryptocurrency as collateral or to loan cryptocurrency while earning passive income on their committed funds. In order to remain operational while still making money, Aave DAO must consistently manage its risk by adjusting certain risk parameters.
Gauntlet and Chaos Labs are two organizations hired by Aave DAO to give recommendations on how to manage the protocol’s risk. Both are professional firms that service other DeFi DAOs and are widely trusted by the community. In addition, both Gauntlet and Chaos Labs were voted by the DAO as risk stewards for the Aave V3 protocol, allowing them to make limited quick changes to the protocol. Often, their proposals are passed and executed with little discussion or controversy, given their expertise.
However, a recent Gauntlet proposal aimed at mitigating the risk of CRV collateral - the token associated with Curve, an exchange liquidity pool - failed to pass. The proposal stemmed from concerns about the riskiness of a single wallet that had put up $185 million worth of CRV as collateral to borrow around $68 million worth of stablecoins on the Aave V2 Ethereum protocol. It was Gauntlet’s view that such a large position was risky for the protocol because, since the CRV token is relatively illiquid, if the collateral was to become liquidable there was a possibility that the Aave protocol would lose a substantial amount of money.
Gauntlet’s proposal faced pushback in the forums, on both technical and ideological grounds, with some community members deeming Gauntlet’s approach insufficient to handle the risk and others objecting to a perceived breach of protocol neutrality.
"[ARFC] - Chaos Labs Risk Parameter Updates - CRV Aave V2 Ethereum" is a separate proposal from Chaos Labs to address the risk to the protocol posed by this large CRV position, following Gauntlet’s proposal failing to pass.
The Proposal
This proposal from Chaos Labs, which is in the Aave Request for Comment stage (more on the specifics of Aave DAO’s governance below), seeks to reduce the Liquidation Threshold (LT) and Loan-to-Value ratio (LTV) by 3 percentage points each (from 58 to 55, and 52 to 49, respectively). This suggestion is in accordance with the previously voted on Risk-Off Framework, which limits the reduction of the Liquidation Threshold to a maximum of 3 percentage points per proposal. The execution of this proposal, according to Chaos Labs, would only make liquidable ~$3 worth of collateral, from 2 accounts. Chaos Labs plans to gradually reduce the LT and LTV over a series of similar proposals to reduce the risk of CRV.
Stakeholders
Proposer
Chaos Labs is an “automated, on-chain economic security system enabling crypto protocols to optimize risk management and capital efficiency while protecting user funds.” They provide frequent recommendations to adjust Aave’s risk parameters. Chaos Labs is paid by Aave DAO to provide this service.
Supporters
No community members have joined the discussion as vocal supporters in the forum at this time.
Opposition
Although Gauntlet and Chaos Labs are in agreement that the CRV position presents a risk to the protocol that should be addressed, Gauntlet stands by their original suggestion: to freeze CRV on Aave V2 Ethereum and reduce LTV to 0. They do not believe that Chaos Labs’ proposal is enough to properly mitigate the risk.
Gauntlet offers a similar service as Chaos Labs, using analysis and simulation to make recommendations to Aave’s risk parameters. They are also a paid by the Aave DAO to provide this service. Usually Gauntlet and Chaos Labs are in agreement but this proposal discussion shows a rare exception to this fact.
Governance Process
For proposals to pass within the Aave governance system, they must start in the forum, move to off-chain voting for a sentiment check (i.e. TEMP Check) on the idea, come back to the forum as an official Aave Request for Comment (ARFC), pass off-chain voting again, and then move to an official on-chain vote, if necessary, as an AIP (Aave Improvement Proposal). This proposal is currently in the ARFC phase.
A voter can vote For, Against, or Abstain with AAVE or stkAAVE (staked AAVE) delegated to them. The quorum is 320K AAVE (or stkAAVE). At the time of writing, 255K AAVE (or stkAAVE) have been placed in the vote. The weight of the vote in each option (For, Against, Abstain) is currently hidden until the vote is over. If this proposal passes in off-chain voting, it will be put on-chain for a vote that, upon success, will automatically set adjust the risk parameters according.
The snapshot vote will close on June 23 at 6:06PM UTC.
The Big Picture
The context behind Chaos Labs’ proposal - specifically, the failed Gauntlet proposal that preceded it and the discussion thereof - exemplifies an engaged and diverse community governing and servicing a DeFi protocol.
Typically, Gauntlet’s proposals to deal with perceived risk to the protocol are trusted and passed without discussion. However, the rejection of Gauntlet’s proposal shows that community governance takes precedent over the expert opinion of any single service provider. Further, the fact that Aave has multiple service providers, including Chaos Labs, allows the community to have different options in dealing with important matters. On the other hand, having multiple service providers with different opinions on risk adjustments can create confusion and tension within the community. Multiple providers of similar/redundant services may compete for influence rather than collaborate. So far, however, it appears that the relationship between Gauntlet, Chaos Labs, and the community is that of healthy collaboration towards a common cause of proper risk management, as opposed to disagreement for the sake of self-interest.
The discussion around this proposal also highlights Aave DAO’s (or any other DAO’s) responsibility in safeguarding the protocol, with some community members highlighting the importance of neutrality given that Gauntlet’s original proposal predominantly affected a single large borrower. Although seemingly neutral, risk parameter updates, like any proposal to change the protocol, are likely to affect some users positively and others negatively. Ultimately, Gauntlet and Chaos Labs, as service providers, are charged with a specific responsibility on which they will rightly focus but the community plays an important role in oversight and balancing objectives.
It is unclear whether Chaos Labs’ proposal will pass or not, due to Aave’s private voting during voting periods. Whichever way the vote goes, however, the circumstances of this proposal are an important case study.