This is an archive of our post on Aave governance forum. Read the full thread here.

Summary

LlamaRisk supports the proposal. The end of Renzo’s Season 2 rewards program is expected to greatly reduce the profitability of the ezETH/wstETH leveraged-looping strategy, which is currently the main driver behind ezETH’s growth on Aave. For instance, the ezETH Yield Layer product of CianProtocol currently offers 31.41% APY. Once that yield is reduced to more reasonable levels, users can benefit from the high yields offered by the sUSDe Stablecoins liquid E-mode. The proposed increase in uOptimal will help compensate for this effect and retain the ezETH supply.

Source: CianProtocol ezETH Yield Layer, December 2nd, 2024

While a higher uOptimal increases the capital efficiency of wstETH, generating more revenue for the Aave DAO, it also increases the severity of a potential bad debt scenario with wstETH as debt. However, the primary use case for wstETH on both the Core and Prime instances are LRT/wstETH leveraged-looping scenarios. Since LRTs are correlated to wstETH, the liquidation risk is reduced significantly. For this reason, we believe this change would not materialize into a higher risk for the Aave DAO. If the main use case for wstETH borrowing were to change, then such a high-uOptimal would have to be reassessed.

wstETH Core instance

For wstETH on the Core instance, we analyze a scenario where the utilization remains the same after implementing the proposed changes (max utilization sensitivity) and another where the borrowing rate remains the same (max borrow rate sensitivity). We found that the borrowing rate could decrease to a minimum of 0.29%, and the utilization could increase to 26.97%.

wstETH Prime instance

Again, for wstETH on the Prime instance, analyze a scenario where the utilization remains the same after implementing the proposed changes (max utilization sensitivity) and another where the borrow rate remains the same (max borrow rate sensitivity). We found that the borrowing rate could decrease to a minimum of 1.25% and that the utilization could increase to a maximum of 90%, corresponding to the proposed new uOptimal.

We model the Slope1 component of the Interest Rate Model (IRM) as an affine function y = ax + b, where:

• y represents the borrowing rate

• a represents the Slope1 rate

• x represents the utilization rate

• b represents the base borrow rate

Given points A = (0; b) and B = (uOptimal; Slope1 rate), we calculate a using a = (y_B - y_A)/(x_B - x_A). We can then analyze two extreme scenarios: solving for utilization rate x using current borrowing rate y (maximum borrowing sensitivity) and solving for borrowing rate y using current utilization rate x (maximum utilization sensitivity).

These projections assume rational market behavior under perfect conditions, with all other variables remaining constant. The model is valid only when utilization remains below uOptimal, above which the Slope2 component requires separate modeling.

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.