[ARFC] sUSD Risk Parameter Adjustment

Overview

This proposal continues on the proposed steps from this post by recommending the removal of sUSD as collateral on Aave V3’s Optimism. Additionally, we recommend reducing the LT and LTV of the Stablecoin E-Mode on Optimism, which sUSD is part of, and setting sUSD LTV to 0.

Motivation

This recommendation follows the recent proposal to reduce the supply and borrow cap of sUSD to 10m and 1, respectively; that adjustment follows a period of reduced sUSD DEX liquidity, which dropped to a buy-side low of $2M before recovering after the introduction of token incentives.

During this period, shorting activity on sUSD was observed between January 9th and January 15th. Following the implementation of token incentives on January 16th, as detailed in this announcement, these short positions were closed, and new borrowing demand primarily originated from users farming the newly introduced incentives.

Below, we present a table of the current sUSD borrows on Aave and the associated strategy exposure. The distribution of such strategies seems to be quite dispersed, with the recent incentive structure leading to demand to create perpetual positions on Synthetix coupled with liquidity pool growth rather than short positions.

User Action Size Date
0xD8b88C185e06eBF8C58B7dc8b7AFf18304CdD888 Short 350k Jan 9 to Jan 15
0x49bf093277bf4dde49c48c6aa55a3bda3eedef68 LP 150k 16th
0x407cff84eeaacda390fe302c99fa5dd32521bc53 Holding 400k 16th
0xe31deacee1770dfa56f8849724bcb7f8e5f76ef2 LP 130k 16th
0xf7ca1f0ff0995c84fef530f7c74c69fb80331e81 Perp 200k 16th
0x9644a6920bd0a1923c2c6c1dddf691b7a42e8a65 Perp+LP 750k 16th
0xfd81b27d9796a1ba7d7171ea70010c9befb2a62a Perp 380k 17th

Peg Dynamics and Mechanism Design

An analysis of the peg and utilization dynamics revealed that the price deviation of sUSD can be attributed to a combination of multiple factors, including traders profiting from the Synthetix Perpetual platform, the lack of interest rate on sUSD debt in an environment of heightened rates, and the ongoing migration to Synthetix V3, which reduces demand for sUSD. This created an imbalance as this excess supply was not met with proportional demand, putting downward pressure on the price of sUSD. This conclusion is supported by the increase in sUSD’s market cap, which closely followed the downward price deviation and the growth in interest rates.

While this behavior does not pose a risk to sUSD’s fundamental backing, the exit of sUSD’s price from its concentrated liquidity range reduced buy-side liquidity, thereby increasing Aave’s risk to sUSD positions. This liquidity constraint underpinned the recommendation to reduce caps, aiming to mitigate Aave’s risk.

Removing sUSD as Collateral

With the recent deterrence of the creation of new sUSD debt on Aave and continued minimized supply exposure, the majority of outstanding debt is fundamentally leveraged to explicitly contribute to the recent sUSD incentive programs by minimizing effective peg exposure. However, while highly overcollateralized, the current economics associated with the soft peg mechanism is fundamentally shaky; minimizing the usage of sUSD as collateral is recommended to be performed as a preventive measure. The current distribution of such sUSD-collateralized stablecoin debt positions is highly conservative, as can be seen below, thereby requiring a significant depeg to start contributing negatively to the Aave protocol, and the expectation is such that exposure will continue to be minimized over time, as has been the case with a 3M reduction in supply on Aave (25%) over the last 7 days.

Moreover, during a swift depeg event 7 months ago, we proposed an immediate LTV0 change due to the deprecation of their spot synths contributing to significant downward pressure, as can be seen here. The fundamental characteristics of the asset eventually improved once more through mechanical changes on Synthetix’s side in the redemption process, leading to the unfreezing of the asset, as seen in this post. However, we continued to reduce exposure significantly in the form of decreasing caps, coupled with decreasing LTV/LTs in both the general market and E-Mode configuration, leading to a more conservative state. As such, a significant buffer of 6% price movement already exists between the collateral value at the LT and the bad debt threshold of 1/(1+LB) in e-mode itself.

Recommendation

To further improve this safety buffer and given the conservative distribution of collateralized debt positions throughout the stablecoin E-Mode config on Optimism, we recommend decreasing the LTV/LT in Stablecoins E-Mode while minimizing induced liquidations.

Considering the current positioning within the E-Mode, we estimate the optimal LT reduction to be 87%. While we intend to deprecate the E-Mode and remove sUSD from it progressively, this change poses an initial step in that direction. Chaos Labs will monitor the market and introduce additional Liquid E-Modes if demand for a stablecoin-correlated E-Mode remains.

The change to 87% LT is expected to cause the liquidation of $95K of collateral.

The liquidations caused at the recommended LT are primarily stablecoins with significant liquidity to support the minimal liquidation size, with sUSD liquidations only representing $2.4K, or 3.2%.

Additionally, we recommend setting the LTV of sUSD outside of the Stablecoin Correlated E-Mode to 0; this measure will prevent additional borrowing from being performed using sUSD as collateral, hence limiting the future risk posed by an sUSD downward deviation.

This measure will simultaneously apply to both sUSD parameters within and outside of the E-Mode, hence limiting the opening of new positions.

Following an improvement of the sUSD peg and a reduction in sUSD-denominated collateral, Chaos Labs will reinstate the borrow cap to a defined value determined by our methodology.

Specification

For the sUSD asset on Aave V3 Optimism instance, we recommend the following:

Asset Deployment Current LTV Rec. LTV
sUSD Optimism 60.00% 0.00%

For the Stablecoin E-Mode on Aave V3 Optimism instance, we recommend the following:

Parameter Current Value Rec. Value
LTV 90.00% 0.00%
LT 93.00% 87.00%
Liquidation Penalty 1.00% 1.00%

Disclaimer

Chaos Labs has not been compensated by any third party for publishing this ARFC.

Copyright

Copyright and related rights waived via CC0

4 Likes

Summary

LlamaRisk supports the proposed changes. The decision to limit exposure to sUSD is rational; therefore, the changes to Stablecoins E-Mode are well justified. Nonetheless, considering that this E-Mode is planned to be deprecated, it may be beneficial to limit new E-Mode exposure completely and set the E-Mode’s LTV to 0 immediately. This would also achieve the objective of fully limiting originations of new sUSDe-backed loans but would still let current positions be topped up subject to LT.

sUSD Stability

Even though sUSD DEX liquidity has recovered from the lows observed in December 2024, the secondary market price remains 2% below the peg. This means the market has fundamentally shifted, with the liquidity concentration being out of bounds. The price movements have also become more volatile, with the trend failing to show signs of recovery. While the overall liquidity may be sufficient to handle unexpected liquidations, the observed trend prompts the enactment of safety measures on Aave’s Optimism market targeted to limit USD exposure.

Stablecoins E-Mode

Out of 9.3M sUSD (~$9.1M) currently supplied on Optimism’s market, $5.14M of borrows are made against the sUSD collateral. Most of these positions are opened using the Stablecoins E-Mode and are highly over-collateralized; therefore, even with the proposed LT reduction, additional borrowing of up to $2.6M would still be possible. As outlined by @ChaosLabs, forced liquidations would immediately affect only $2.6k of sUSD collateral. Nonetheless, lowered LT is expected to move the users to start unwinding the positions gradually.

Given that the intention is to deprecate the Stablecoins E-Mode and potentially replace it with new ones, we consider it beneficial to set the LTV of that E-Mode to 0 starting from this proposal. This would limit new borrows that would be done using this E-Mode but would leave the flexibility for current borrow positions, letting users top up the collateral if needed.

Disclaimer

This review was independently prepared by LlamaRisk, a community-led non-profit decentralized organization funded in part by the Aave DAO. LlamaRisk is not directly affiliated with the protocol(s) reviewed in this assessment and did not receive any compensation from the protocol(s) or their affiliated entities for this work.

The information provided should not be construed as legal, financial, tax, or professional advice.

3 Likes

The current proposal has been escalated to ARFC Snapshot. Voting will commence in 24 hours, within the following timeframe:

Start
Jan 31, 2025 · 1:20 PM UTC
End
Feb 3, 2025 · 1:20 PM UTC

1 Like

After Snapshot monitoring, the current ARFC Snapshot ended recently, reaching both Quorum and YAE as winning option, with 325.4K votes.

Therefore, [ARFC] sUSD Risk Parameter Adjustment has PASSED.

Next step will be the publication of an AIP for final confirmation and enforcement of the proposal.

3 Likes

As Michigan Blockchain, we support removing sUSD as collateral on Aave V3 Optimism and reducing its risk parameters in Stablecoin E-Mode. Given sUSD’s declining liquidity and inconsistent peg stability, we believe this is a necessary step to protect Aave from unnecessary risks. Most recent borrowing demand for sUSD has been driven by incentive farming rather than organic usage. This further reinforces our position. Lowering sUSD’s LTV to 0% will prevent new borrowing against it. It’ll further reduce Aave’s exposure to potential peg deviations. Additionally, reducing the Liquidation Threshold (LT) from 93% to 87% ensures a smoother transition while maintaining protocol stability. These changes allow Aave to phase out sUSD in a controlled manner. The reason Aave has proved itself to be a robust protocol over the years - despite all the market crashes - is risk management, so we believe that this is a responsible move that strengthens Aave’s risk management and long-term sustainability while ensuring users are protected.

-Kerem Dillice and nsks

1 Like