[ARFC] Onboard GHO and Migrate Streams to Lido Instance


title: [ARFC] Onboard GHO and Migrate Streams to Lido Instance
author: @karpatkey_TokenLogic
created: 2024-11-04


Summary

This publication proposes the following:

  • Onboarding GHO to the Lido instance of Aave v3;
  • Providing initial GHO liquidity from the Treasury; and,
  • Migrating Service Providers streams to GHO from Lido instance.

Motivation

Onboard GHO Lido Instance

With the circulating supply nearing 180M and nearly 50M in DEX liquidity on Ethereum, this publication proposes adding GHO to the Lido instance of Aave v3.

Adding GHO to the Lido instance of Aave v3 would provide the DAO with several opportunities:

  • Mint GHO through a Facilitator and deposit it into the Reserve to generate revenue;
  • Enable Aave DAO to earn yield on existing GHO holdings; and,
  • Establish a GHO reserve that provides an organic deposit yield.

Each of these options offers clear advantages for the Aave DAO and GHO users. GHO is to be onboarded as a borrow-only asset, similar to the current configurations of USDC and USDS.

Deploy GHO from Treasury

From observing the USDC reserve on the Lido instance, it is apparent that GHO would benefit from some initial liquidity. With passive GHO held in the Treasury, this publication proposes depositing available GHO into the new Reserve. With the GHO expected to earn a yield the DAO will benefit from a new revenue source.

A separate proposal will recommend minting and depositing GHO into the new reserve via a new Facilitator on Ethereum to futher help bootstrap the reserve. The addition of sUSDe to the Lido instance is expected to provide strong demand for stablecoin debt.

Migrate Service Provider Streams

With the GHO being deployed into Lido, several Aave DAO service provider streams are to be amended to draw GHO from Lido instance.

  • Chaos Labs;
  • Aave Labs;
  • LlamaRisk; and,
  • Aave Chan Initiative.

When the AIP is submitted, new streams will be created to replace the streams being cancelled. With several streams are soon to expire during December, these streams are not migrated to Lido instance and sufficient GHO will remain in the treasury to facilitate payment to these teams.

Within this publication, the Aave Grants DAO GHO Allowance is to be cancelled.

Specification

The below details the parameter configuration of the GHO Reserve on the Lido instance of Aave v3.

Parameter Value
Isolation Mode No
Borrowable Yes
Collateral Enabled No
Supply Cap 20M
Borrow Cap 2.5M
Debt Ceiling -
LTV -
LT -
Liquidation Bonus -
Liquidation Protocol Fee 10.00%
Variable Base 4.50%
Variable Slope1 3.00%
Variable Slope2 50.00%
Uoptimal 92.00%
Reserve Factor 10.00%
Flashloanable Yes
Siloed Borrowing No
Borrowable in Isolation No

The below table shows the GHO balance at time of writing for reference only:

Treasury Asset Holding
GHO 7,295,678.65

Note: The September funding update is to acquire GHO and transfer 5M GHO to Arbitrum to be deployed into the GHO reserve to earn yield.

Sufficient GHO shall remain in the Treasury to support streams with a 2024 expiry date, with the balance to be deposited into the Lido instance of Aave v3. This will be confirmed and detailed in the comments just prior to submission of the AIP.

The following Service Provider streams are to be migrated to using GHO from the Lido instance.

Service Provider StreamID
Aave Chan Initiative 100034
Aave Labs 100039
Chaos Labs 100046
LlamaRisk 100048

Cancel Aave Grants DAO GHO Allowance.

Disclosure

TokenLogic and karpatkey receive no 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, an AIP will be submitted to implement the proposal.

Copyright

Copyright and related rights waived via CC0.

2 Likes

Very interesting proposal. I do think that GHO can greatly benefit from this and allowing more user to use it.
Are there risks for the protocol and its user if GHO can be minted and deposited to be borrowed again?

Summary

LlamaRisk supports the propositions and parameters listed in this ARFC and believes that onboarding GHO to Aave Lido instance is a rational decision that will further promote the growth of Aave’s stablecoin. Moreover, providing initial GHO liquidity from the Treasury will help optimally bootstrap the borrow interest ensuring more stable borrow conditions. Further analysis also indicates that this measure would not impact the risk profile of GHO stablecoin or Aave’s Lido market.

Lido Market

After a successful period of growth, Lido market is now the second largest Aave market with a total size of $1.55B. This market is particularly suitable for ETH LST looping use case where most of the borrow volume is consequently attributed to wstETH and ETH assets.


Source: Aave, 8th November, 2024

Therefore, while stablecoin demand is not high at the moment, as outlined in a related proposal onboarding yield-bearing stablecoins to this market would increase stablecoin utility in this market. It can also be observed that GHO has high borrow interest using wstETH and stablecoins like USDS and USDC. Given the availability of these same assets on Lido market, this interest would be replicated, especially if the borrow rate of GHO is lower than of other stablecoins.

Source: TokenLogic GHO Analytics, 8th November, 2024

Moreover, Lido market is deployed on the Ethereum Mainnet and therefore benefits from the 28.4M GHO liquidity available there. Currently, GHO liquidity levels are sufficient and continue to grow. Additionally, the liquidity has remained stable, even amid recent market volatility.

Source: TokenLogic GHO Analytics, 8th November, 2024

It is important to note that all collateral assets in the Lido market already back GHO, meaning that this new GHO market deployment would not provide additional collateral diversification for GHO. Therefore, GHO’s risk profile would remain unchanged, aside from potential shifts in collateral composition. However, GHO could add ezETH as a new collateral type if a Liquid e-Mode is established or if the previously mentioned yield-bearing stablecoins are onboarded.

GHO Borrowing Patterns

It is important to distinguish two current setups where GHO is available:

  1. On-demand GHO minting on Aave Mainnet market
  2. Standard supply-and-borrow setup on Arbitrum (where GHO cannot be used as a collateral)

Until the recent expiration of GHO incentives on Arbitrum market, incentivized GHO borrowing resulted in extensive looping which made the borrow behaviors differ between these two setups. Quickly after the incentives expired, the borrows have traced back to usual levels. Currently, the risk levels of GHO loans on both markets are similar and no extensive risk taking is observed.

These observations indicate that optimal conditions to profitably loop GHO can only be established by using incentives. It is important to underline that if GHO looping would re-emerge, it would not necessarily impact the GHO risk profile since:

  • GHO looping is more capital intensive because GHO cannot be directly used as a collateral backing GHO loans. This poses a limit to the extent of possible looping.
  • Sufficient GHO liquidity and a supply/borrow buffer for Flashloans are maintained to cover the liquidations in case of unexpected health reduction.

Taking into account the learnings from this recent situation on Arbitrum, we believe that it is rational to onboard GHO with a supply-and-borrow setup where GHO collateralized loans are not allowed. This setup would bring more opportunities to the end user while ensuring minimal risk.

Recommended Parameters

The parameters for GHO have been aligned internally in collaboration with different DAO service providers and are presented as part of the original proposal.

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

Overview

Chaos Labs supports the proposal to onboard GHO to the Lido instance of Aave v3, but we provide alternative parameters for the supply/borrow cap and interest rate curve. Our analysis considers the rapid growth in GHO’s liquidity, the strong demand driven by staking and liquidity incentives, and the potential for increased revenue and efficiency for the Aave DAO.

GHO Liquidity and Demand

The primary driver of GHO demand has been stkGHO, currently holding over 115 million GHO. This growth is largely attributed to the attractive combined deposit APY, which reaches 17% when including Merit rewards.

The second major demand factor is GHO’s growing DEX liquidity, which recently reached approximately $50M TVL and $10M of buy liquidity within a 1% range. This growth is supported by competitive incentives across major DEX platforms, including Curve, Balancer, and Maverick, with APYs ranging from 7% to 15%.

With these high APYs being among the most consistent on-chain yields, we believe they will continue driving GHO borrow demand.

Borrow Distribution

wstETH currently is the collateral used to borrow approximately 22% of the GHO from the Main Ethereum Facilitator, amounting to around 38 million GHO. Three of the users using wstETH collateral rank among the top 10 GHO borrowers.

wstETH has been the primary collateral for wstETH since the launch of the GHO Ethereum Facilitator.

The average collateralization rate for GHO borrowers on the Main Ethereum Facilitator is 220%. Notably, among the top 10 borrowers, only those with highly correlated collateral maintain health scores below 1.2. This reflects prudent borrowing behavior, supporting the case for elevated initial listing parameters.

Given the high amount of wstETH collateral used to borrow GHO and because of the higher supply APY for wstETH in the Lido instance, we expect borrowers who use wstETH collateral to transition to the Lido instance to take advantage of the more efficient borrowing experience and lower resulting rates.

Benefits

Implementing the recommended parameters for listing GHO in the Lido instance is expected to generate multiple benefits, such as increased DAO revenue, improved capital efficiency, and further growth of the Lido instance.

One key advantage is the new Facilitator, which enables the DAO to initialize GHO liquidity within the Lido instance. This initial GHO liquidity will be minted solely for depositing in the Lido instance and backed by the collateral used to borrow it.

The liquidity in the GHO market will enable WETH in the Lido pool to be utilized as collateral, fostering further market growth and enhancing DAO revenue. Additionally, we anticipate that a portion of the wstETH collateral currently used in the facilitator will migrate to the Lido instance. This shift will also contribute to increased revenue, as the GHO borrow rate will be partially offset by the wstETH supply rate, maximizing yield potential in the Lido market.

Using Merit to optimize incentives will help reduce inefficient allocations, benefiting users who utilize the pool as intended. This will enhance supply liquidity and improve the overall user experience for the pool’s intended audience.

Interest Rate Curve and Borrowing Costs

To ensure that GHO borrowing in the Lido instance remains consistent with the Ethereum Facilitator, we recommend adjustments to the proposed Base and Slope 1 parameters. As the Lido instance’s GHO pool will initially function similarly to a Facilitator but is expected to evolve into a standard market with user deposits, we suggest implementing an interest rate curve that reflects both the initial and future use cases. Specifically, we propose a Base rate of 3.5% to ensure that GHO minted through the Lido Facilitator isn’t borrowed below the minimum rate used by the Ethereum Facilitator. Additionally, a Slope 1 of 3% aligns with the initial target rate of 6% at UOptimal, keeping borrowing attractive when factoring in the yield on wstETH and WETH collateral.

We also recommend maintaining a Reserve Factor of 10%, in line with similar stablecoin markets in the Lido instance, given the future likelihood of user deposits.

Migrate Service Provider Streams

Chaos Labs supports the proposed migration of Aave DAO service provider streams to the Lido instance. This migration enhances treasury efficiency and improves the liquidity available on the market hence generating additional revenue for the DAO.

Supply and Borrow Caps

To accommodate the initial 10,000,000 GHO liquidity minted via the Lido Facilitator, as outlined in this proposal, along with approximately 8,300,000 GHO for the service provider streams that are to be migrated, we recommend an initial supply cap of 20,000,000 GHO. This cap provides a buffer for additional user deposits, although we don’t expect the additional buffer to gain significant traction in the initial period after listing.

For the initial borrow cap, we suggest 18,500,000 GHO to support the UOptimal of 92%. This aligns with other stablecoins in the Lido instance, allowing slight overutilization if borrow demand surpasses expectations.

Recommendation

Chaos Labs supports the proposal to onboard GHO to the Lido instance of Aave v3 with the following listing parameters:

Parameter Value
Chain Ethereum Lido
Isolation Mode No
Borrowable Yes
Collateral Enabled No
Supply Cap 20,000,000
Borrow Cap 18,500,000
Debt Ceiling -
LTV -
LT -
Liquidation Bonus -
Liquidation Protocol Fee -
Variable Base 3.50%
Variable Slope1 3.00%
Variable Slope2 50.00%
Uoptimal 92.00%
Reserve Factor 10.00%
Flashloanable Yes
Siloed Borrowing No
Borrowed in Isolation No
3 Likes

After discussing this proposal with @ChaosLabs, the following amendments have been made to reflect prevailing market conditions.

Parameter Value
Isolation Mode No
Borrowable Yes
Collateral Enabled No
Supply Cap 20M
Borrow Cap 2.5M
Debt Ceiling -
LTV -
LT -
Liquidation Bonus -
Liquidation Protocol Fee 10.00%
Variable Base 4.50%
Variable Slope1 3.00%
Variable Slope2 50.00%
Uoptimal 92.00%
Reserve Factor 10.00%
Flashloanable Yes
Siloed Borrowing No
Borrowable in Isolation No

A Snapshot vote has been created:

Start date Nov 19, 2024, 5:38 PM
End date Nov 22, 2024, 5:38 PM

1 Like