[ARFC] Update GNO Risk Parameters on Aave V3 Gnosis Pool

[MarcZeller]

Proposal edited to reflect risk service provider recommendation

---

Title: [ARFC] Update GNO Risk Parameters on Aave V3 Gnosis Pool

Author: @marczeller - Aave Chan Initiative

Date: 2023-11-20

---

Summary

This ARFC proposes to update GNO risk parameters on the Aave V3 Gnosis Pool to enable borrowing, fostering Gnosischain decentralization and use-cases for passive income and staking.

Motivation

The goal is to enhance GNO utility by allowing it to be a borrowable asset, thereby supporting Gnosischain’s decentralization. This adjustment also prepares for future GNO LSTs and a possible GNO-correlated e-mode, expanding utility and passive income opportunities for GNO holders.

Specification

The proposed changes aim to make GNO a borrowable asset outside of isolation mode with adjusted risk parameters to accommodate this new functionality

Risk Parameter	Current Value	Proposed Value
Isolation Mode	Yes	YES
Borrowable in Isolation	No	No
Stable Borrow	No	No
Enable Borrow	No	Yes
Enable Collateral	Yes	Yes
Loan To Value (LTV)	31%	31%
Liquidation Threshold	36%	36%
Liquidation Bonus	10%	10%
Reserve Factor	N/A	20%
Liquidation Protocol Fee	10%	10%
Supply Cap	30000 GNO	30000 GNO
Borrow Cap	N/A	1100 GNO
Debt Ceiling	1M$	1M$
uOptimal	N/A	80%
Base	N/A	0%
Slope1	N/A	15%
Slope2	N/A	80%

Next Steps

1. Collect feedback from the Aave community & Aave Risk service provider on the proposed changes.
2. If the community feedback is positive, advance the proposal to the ARFC Snapshot phase.
3. Upon successful Snapshot, implement the updated risk parameters via an AIP vote.

Disclaimer

This proposal is powered by Skywards.

The Aave Chan Initiative has not received any form of compensation from Gnosis for creating this proposal.

Copyright

Copyright and related rights waived under CC0.

[Gauntlet]

Gauntlet Risk Analysis - Aave V3 on GnosisChain

Summary

Gauntlet previously provided initial risk recommendations for Aave V3 on GnosisChain and concluded that the highest risk to Aave on GnosisChain was in limited wETH liquidity against USD stablecoins.

We acknowledge that the total value locked (TVL) in Gnosis Chain has almost tripled since early October, increasing from $75 million to close to $250 million, and this surge was driven by the transfer of stablecoins to Gnosis Chain.

Despite this growth, the liquidity between wETH and USD stablecoins remains low, maintaining the risks identified in our initial analysis.

Gnosis Chain experienced a temporary drop in liquidity following a Balancer V2 pool hack in late August 2023. Although the liquidity has since recovered, such incidents can have lasting effects on users’ trust and willingness to provide liquidity.

Hence, we strongly recommend adopting cautious initial parameters as per our revised analysis until a more comprehensive understanding of the market dynamics and potential risks associated with GNO borrowing is achieved.

Implementing conservative measures can serve as prudent steps towards ensuring the stability and security of the platform while minimizing exposure to adverse market conditions.

Furthermore, contemplating a phased approach for the future could be a viable consideration, where the introduction of GNO borrowing is executed incrementally, facilitating ongoing scrutiny and evaluation of its effects on the platform’s liquidity and overall functionality. However, it’s crucial to note that as of now, a phased approach is not recommended. This cautious stance will afford us, at Gauntlet, the opportunity to meticulously monitor the landscape and address any emergent issues or unforeseen hurdles promptly. This methodical approach not only underscores our commitment to maintaining a tight rein on the process but also ensures a more disciplined and safer framework for potentially integrating GNO as a borrowable asset down the line. Our vigilant oversight aims to mitigate risks and foster a conducive environment for such integration when the circumstances are deemed favorable.

Specifications

We recommend against listing GNO in isolation mode to allow for wETH borrowing, since isolation mode only allows for stablecoin borrowing. GNO-wETH liquidity is much stronger than GNO - USD stablecoin liquidity. The correlation between GNO and wETH is stronger, which may provide a more predictable and coherent market behavior, aiding in maintaining a balanced lending ecosystem. The restriction of isolation mode to only allow for stablecoin borrowing further complicates the scenario, making the inclusion of GNO less beneficial in this setting.

We recommend against listing GNO as borrowable assets. The liquidity dynamics between GNO and other assets such as wETH are more favorable compared to GNO and USD stablecoins, which is a critical factor for ensuring smooth operations on lending platforms.

Risk Parameter	Current Parameters	ACI Proposed	Gauntlet Updated Recs
Isolation Mode	YES	NO	NO
Enable Borrow	NO	YES	NO
Reserve Factor	N/A	20%	N/A
Borrow Cap	N/A	24,000 GNO	N/A
Isolation Mode Debt Ceiling	1M$	N/A	N/A
uOptimal	N/A	80%	N/A
Base	N/A	0%	N/A
Slope1	N/A	15%	N/A
Slope2	N/A	80%	N/A

[ChaosLabs]

Enabling GNO as a Borrowable Asset

We support the proposal to enable GNO as a borrowable asset on the Aave V3 Gnosis Pool with an initial conservative cap. This will provide valuable insights into the borrowing dynamics and utility of GNO, enhancing the understanding of its potential use cases before considering further potential increases.

Borrow Cap

For our recommendation for the borrow cap for GNO, we review the average daily trading volume of GNO. Over the past 180 days, this volume is approximately $2.1 million (CeFi & DeFi). Based on this data, we propose an initial borrow cap set at 10% of the average daily volume, equating to roughly 1,100 GNO.

IR Curves

We recommend setting the initial IR Curves in line with those for other governance tokens on Aave. These parameters are subject to future adjustments based on usage on the protocol and evolving market conditions

Taking GNO out of Isolation Mode

Given the recent launch and the absence of significant changes in liquidity conditions, we continue to support our earlier stance as mentioned in our previous recommendations, which is to retain GNO in Isolation Mode. We understand the consideration for a potential GNO LST E-Mode category, but currently, we advise that keeping GNO in isolation mode remains the safer and more effective approach for managing the protocol’s risk exposure. This is due to the current market dynamics characterized by low average daily volume and on-chain liquidity for GNO. We plan to keep evaluating the status of GNO in relation to isolation mode, taking into account its liquidity and usage on the protocol, and the introduction of new use cases, such as LST assets that might be listed on the protocol.

While we recommend this approach, it’s important to note that if the community opts to list GNO outside of isolation mode, given the current market conditions, platform usage, and risk parameter settings (specifically supply caps and LT), we do not foresee substantial risk to the protocol.

Additionally, we’d like to point out that currently, almost 96% of all GNO is supplied by a single account, which has not borrowed against it.

Specification

We propose these changes for the enablement of GNO as a borrowable asset with adjusted risk parameters:

Risk Parameter	GNO
Isolation Mode	YES
Enable Borrow	YES
Enable Collateral	YES
Loan To Value	31%
Liquidation Threshold	36%
Liquidation Bonus	10%
Reserve Factor	20%
Liquidation Protocol Fee	10%
Borrow Cap	1,100
Supply Cap	30,000
Debt Ceiling	$1M
uOptimal	45%
Base	0%
Slope1	7%
Slope2	300%
Stable Borrowing	Disabled
Flahloanable	YES
Siloed Borrowing	NO
Borrowed in Isolation	NO

[MarcZeller]

Hello, and thanks for these recommendations,

@ChaosLabs, the main use-case to borrow GNO as it is now is to use it to stake, a bit similar to RPL on Mainnet.

The staking yield is currently ~15%. Placing Slope1 at 7% will likely result in an equilibrium higher than desired as this is a deal “too good” for borrowers.

Would you accept working with our recommended interest rate curve while maintaining isolation mode and very conservative borrow caps?

we designed the curve with the “borrow to stake” use case in mind, these legacy “45% uOpt very high slope2” curves can’t hurt the protocol but are inefficient in nature and end up with a lot of “sleepy” liquidity that doesn’t generate yield for users and revenue for protocol.

[ChaosLabs]

We are open to launching with your recommended IR parameters, given the conservative borrow caps and other conservative parameters.

To reiterate the reason for our initial recommendation, the primary concern with a higher UOptimal threshold lies in the current liquidity constraints of GNO. At 45%, we aim to maintain a reserve level that ensures adequate liquidity for essential functions like liquidations and withdrawals. This conservative setting is more aligned with GNO’s current liquidity and market conditions, offering a safer integration route as a borrowable asset within the Aave ecosystem.
We emphasize the importance of monitoring GNO’s usage and the impact of these settings on the Aave platform. We remain open to future adjustments based on ongoing market observations and data analysis, with a continued focus on the safety and stability of the protocol.

[ACI]

This ARFC has been escalated to the ARFC snapshot stage and vote will start tomorrow.

[ACI]

The ARFC Snapshot for this proposal has ended with YAE option as winning option (541K votes and quorum reached). Next stage will be the implementation of the updated risk parameters via an AIP vote.

[ACI]

As mentioned in the previous reply, the ARFC Snapshot for this proposal ended with YAE option, therefore an AIP has been published. Voting will start tomorrow. We encourage you to participate!