[ARFC] Deploy stataUSDC and stataUSDT GSMs on Ethereum


title: [ARFC] Deploy stataUSDC and stataUSDT GSMs on Ethereum
author: @TokenLogic
created: 2025-01-17


Summary

This publication proposes deploying two stataToken GSMs using the ERC4626 implementation developed by Aave Labs to replace the existing USDC and USDT GSMs.

Motivation

While the primary focus of the GSM is to ensure price stability, the current GSMs issue GHO backed 1:1 with USDC and USDT, which are then held unproductively. The opportunity cost of passively holding USDC and USDT increases as the GSM size grows over time.

With the recent completion of Certora’s ERC4626 GSM audit, there is an opportunity to deploy two new stataToken GSMs to replace the existing GSMs. The USDC and USDT holdings in the current GSMs will be migrated to the new stataUSDC and stataUSDT GSMs, allowing yield generated by these assets to be claimed as revenue by the Aave DAO.

The yield from stataTokens functions as a proxy Borrow Rate and accrues revenue to the DAO continuously, unlike GHO Borrow Rate interest, which is only realized when users repay loans. The chart below illustrates the native deposit yield for USDC and USDT compared to the GHO Borrow Rate over time.

Screenshot 2025-01-16 at 14.00.34

Based on current GSM holdings (1.5M USDC and 10.1M USDT) and the 6-month average deposit rates (7.12% for USDC, 5.88% for USDT) on Aave v3 Core, the Aave DAO could generate an estimated $593.8k USD in annual revenue. This figure is indicative and subject to variations in deposits and market conditions.

Screenshot 2025-01-16 at 12.19.06

Do note, this proposal supersedes this proposal.

Pricing Strategy

While the ERC4626 GSM offers both dynamic and fixed price oracles, the USDC and USDT to GHO ratio will be configured 1:1 to support a quick deployment and reflect the relatively small size of the underlying holdings.

Debt Ceiling

A potent tool for mitigating adverse effects derived from over-reliance on centralised stablecoins, this parameter can be configured relative to GHO’s circulating supply.

Evident earlier this week when USDT traded beneath $1.000 across several venues, users are able to supply the depegged asset to mint GHO. The debt ceiling is the primary means of limiting the GSM exposure to a specific asset. Furthermore, the debt ceiling can be sized based upon the lesser of the following:

  • Preferred amount of USDC exposure; or,
  • Available exit liquidity in the USDC reserve.

Either of the above options, compares favourably when determining GSM parameters relative to configuring the Exposure Capacity for peg stability.

Specification

Overview

The below provides an overview of the GSM.

Existing GSM Migration Overview

The previous versions of the GSM will be seized, and all liquidity migrated to the new version.

The below provides an overview of the methodology.

  • Deploy new stataUSDC and stataUSDT GSM4626 from the Gho-Core repository with their respective Oracles
  • Add new GSMs to the GSMRegistry
  • Update the FeeStrategy of the new GSMs to the existing FeeStrategy: 0x83896a35db4519BD8CcBAF5cF86CCA61b5cfb938
  • Grant the Executor the 'LIQUIDATOR_ROLE' to be able to seize the existing GSMs
  • Seize the current USDC and USDT GSMs by calling the seize() function.
    • This freezes actions on them and transfers the underlying tokens to the collector
  • Add the two new GSMs as facilitators on the GHO token contract
  • Add the new stataUSDC and stataUSDT GSMs as controlled facilitators on the GhoBucketSteward
  • With the withdrawn assets above, obtain stataUSDC and stataUSDT via deposit on the respective vaults
  • With the obtained stataUSDC and stataUSDT, exchange for GHO on the new respective GSMs
  • Use the obtained GHO to burnAfterSeize() on the current USDC and USDT GSMs. If there are any discrepancies in the amount of GHO, use GHO from the treasury to bring the minted GHO by the GSMs to zero
  • Remove existing GSMs as facilitators of the GHO token
  • Remove existing GSMs from the GSMRegistry
  • Remove existing USDC and USDT GSMs from being a controlled facilitator on the GhoBucketSteward

New stataUSDC and stataUSDT GSMs Parameter Configuration

The below details the configuration of the stataUSDC GSM.

Parameter Value
Bucket Capacity (GHO) 16.00M
Exposure Capacity (USDC) 8.00M
Price Strategy 1:1
Freeze Lower Bound $0.990
Freeze Upper Bound $1.010
Unfreeze Lower Bound $0.995
Unfreeze Upper Bound $1.005
Mint GHO Fee 0.00%
Burn GHO Fee 0.20%

The below details the configuration of the stataUSDT GSM.

Parameter Value
Bucket Capacity (GHO) 24.00M
Exposure Capacity (USDT) 16.00M
Price Strategy 1:1
Freeze Lower Bound $0.990
Freeze Upper Bound $1.010
Unfreeze Lower Bound $0.995
Unfreeze Upper Bound $1.005
Mint GHO Fee 0.00%
Burn GHO Fee 0.20%

Disclosure

TokenLogic does not receive any payment for this proposal.

Next Steps

  1. Gather feedback from the community.
  2. If consensus is reached on this ARFC, escalate this proposal to the Snapshot stage.
  3. If Snapshot outcome is YAE, escalate this proposal to the AIP stage.

Copyright

Copyright and related rights waived via CC0.

7 Likes

Summary

LlamaRisk supports this ARFC and the proposed parameters and believes that the change will have a material impact on GHO’s risk profile. GSM modules help reinforce GHO’s stability, and the opportunity to improve Aave’s capital efficiency makes it even more of a net positive for the DAO.

As outlined by @TokenLogic, USDC and USDT GSMs expect to retain the stablecoin buffer for longer periods, serving as an additional sell-side buffer for GHO secondary market peg. As this USDT/USDC amount sits idle, it is desirable to generate yield while retaining fast redemption properties and low associated risks. Supplying these assets to Aave’s lending pools achieves these objectives.

Impact on Aave’s Core market

Under current conditions, if an additional 8M USDC is supplied to Aave’s lending pool, this portion would represent ~0.3% of the total current USDC supply on Aave’s Core market. Therefore, the USDC borrow APY would only be lowered by ~0.1% at current borrow levels. Similarly, an additional 16M USDT would represent ~0.5% of the total current USDT supply on Aave Core. Therefore, the USDT borrow APY would only be lowered by ~0.15% at current utilization. Consequently, USDC’s and USDT’s borrow/supply utility would be impacted insignificantly. Moreover, the gain from the supply rate would outweigh the loss from the lower borrow rate, resulting in a net positive for the Aave’s DAO. A large supply of both stablecoins in Aave’s core market minimizes liquidity and instability risks.

Projected Opportunities

The GSM opportunity cost has sharply increased since the previous aUSDC GSM proposal, underlining the size of possible upside and capital efficiency increase for Aave’s DAO.

Source: TokenLogic GHO Analytics, January 19th, 2025

Recommended Parameters

We support the proposed parameters for both stataUSDC and stataUSDT GSMs. The continued GHO burn fee at 0.2% will help to ensure that the allocated USDT/USDC buffer in the GSMs remains deposited for the longer term while providing a GHO stability safeguard at a 0.2% GHO secondary market discount threshold.

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.

5 Likes

Summary

Chaos Labs supports the proposal to migrate USDC and USDT from the existing GHO Stability Module on Ethereum into new stataToken GSMs.

Analysis

We note that this proposal is largely the same as a previous ARFC to deploy an aUSDC GSM, which we supported at the time. The primary differences are that this proposal recommends shifting both the USDT and USDC GSMs, rather than just USDC, and it proposes shifting to stataTokens, which represent deposits on Aave but replace aToken’s growing balance mechanism with an accrued value mechanism.

Thus, we refer back to our previous analysis, which found that shifting the GSM to a yield-bearing token representing deposits in Aave was appropriate and presented limited additional risk. Additionally, since the analysis, the size of the USDC and USDT markets on Aave has grown significantly, adding a large amount of borrowable liquidity; the lowest amount of borrowable liquidity observed in either market since our previous recommendation was $11.3M in USDT on December 5. As anticipated in our previous recommendation, this was quickly remedied and USDT remained above 99% utilization for roughly an hour.

Given that the amount of borrowable liquidity has increased significantly for both USDC and USDT, alongside an increase in GHO’s market cap, we are comfortable with the recommended Bucket Capacity and Exposure Cap parameters.

Disclaimer

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

Copyright

Copyright and related rights waived via CC0

2 Likes

A Snapshot vote has been created for this proposal.

Link to vote: here

Start: Jan 24, 2025 · 12:51 PM
End: Jan 27, 2025 · 12:51 PM

Thank you in advance to everyone that participates in the vote.

1 Like

As Michigan Blockchain, we support this proposal to deploy stataUSDC and stataUSDT GSMs on Ethereum. The main reason for our support is that this will lead to better capital efficiency for USDT and USDC GSMs. According to the current data, it will generate around $600k yearly for the Aave treasury. One of the biggest risks that ChaosLabs also mentioned is liquidity availability, however the USDC and USDT markets on Aave have significantly grown, meaning there’s more liquidity available to support the GSM. While we are aware of stataTokens and their utility as explained here, we are curious about the specific rationale behind using stataTokens instead of aTokens in this case.
-Kerem Dillice

1 Like