[TEMP CHECK] - Add GHO/USDC/USDT BPT to Safety Module

Governance
1

title: [TEMP CHECK] - Add GHO/USDC/USDT BPT to Safety Module
author: @TokenLogic - @MatthewGraham and @scottincrypto
created: 2023-10-23

Summary

This publication proposes creating a new GHO/USDC/USDT BPT stable coin category within the Aave Safety Module (SM).

Motivation

There are two main benefits to adding the GHO/USDC/USDT BPT to the SM:

  1. Diversify assets within the SM
  2. Support GHO liquidity

The GHO/USDC/USDT on Balancer v2 is the main GHO liquidity pool and the current TVL is around $10.4M.

Introducing the GHO/USDC/USDT BPT category provides diversification to the SM and for each unit of GHO in the liquidity pool, the Aave DAO accures interest from GHO debt.

If the Aave DAO elects to proceed with the strategy outlined below, the B-80AAVE-20wstETH BPT can also be integrated into the SM and replace the B-80AAVE-20wETH BPT from Balancer v1.

Rewards Strategy

This publication proposed the GHO/USDC/USDT BPT be the first asset added to the SM using the new strategy shown below. The graphic details how the yield from Balancer and Aura Finance is integrated and routed to depositors.

Within the strategy, the BPT representing a claim to the liquidity within the Balancer v2 liquidity pool remains in the SM and is controlled by the Aave Protocol at all times. This limits the risk exposure to the Balancer v2 liquidity pool as the smBPT deposited into Aura Finance’s contract and Balancer’s gauge contract serves only an accounting purpose and has no economic value.

Upon implementation of the new smBPT gauge, Aave’s vlAURA position can be used to vote rewards to the smBPTgauge on Balancer. These rewards can be claimed without unstaking from the SM.

In addition, the DAO can allocate funds to the GHO Liquidity Committee (GLC) for the purpose of creating Quests/Bribe campaigns and directing additional yield to SM depositors. The new 90-day emission windows, supports creating periodic budgets with clear TVL and APR targets. This grants the DAO the ability to periodically amend the vlAURA voting strategy and Quest/Bribe budget to support onboarding additional BPTs in effort to diversify the SM whilst targeting specific APR and TVL ranges.

The implementation of the strategy requires the Balancer community to support the creation of a smBPT gauge. A forum post was shared with the Balancer community earlier this year. We have received feedback that both Balancer and Aura Finance support the proposed integration and creation of smBPT guages as needed.

Composable Stable v5 Pool Details

The GHO/USDC/USDT pool is a ComposableStablev5 pool and the general pool details can be found here.

The asset weights with ComposableStablev5 pools are dynamic and change based upon deposit and withdrawals. This means the pool can be skewed, overweight, towards one asset relative to another. The risk of being overly concentrated in one asset is highlighted by the current pool composition being >70% GHO. If USDC, USDT or GHO experiences an issue, the pool is likely to be majority the asset experiencing the issue. Further details on the pool type risks can be found here.

The amplification factor setting, currently a low 200, reflects the off balance nature of the pool. The Amplifcation setting is controlled by a Balancer Multisig and can be adjusted higher or lower. When the pool is balanced, most likely also when the GHO peg is acheive, it is expected to be >2000. The higher the amplification factor the more efficient the pool becomes.

GHO Peg

Only when GHO recovers the peg and the pool is more balanced, should the GHO/USDC/USDT BPT be added to the SM.

Given the time to create and audit the smart contracts to integrate the strategy into the SM, the GHO peg should be recovered.

Ultimately the borrow rate will support the peg and this is being gradually increased over time as mentioned here.

In effort to align this integration with broader GHO liquidity strategy, this forum publication is now being shared. It will take time to develop and prepare the contracts. Work has already began in preparation to support developing the smart contracts.

Next Steps

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

The ARFC will present the target TVL and APR range, along with other parameters for the initial implementation. Some insights have already been shared in this proposal.

All associated development work is to be performed by @bgdlabs and will require audit.

In the past Balancer and Aura Finance have expressed interest in supporting the integration by way of contributing to the audit costs. Nothing is confirmed and this will be revisited closer to the implementation date.

Disclosure

TokenLogic receives no compensation beyond Aave protocol for the creation of this proposal. TokenLogic is a delegate within the Aave ecosystem.

We would like to recognise the contributions of @Dydymoon who initially work with @MatthewGraham on this type of proposal when @Llamaxyz was proposing to upgrade the SM. We would also like to recognise @solarcurve’s contribution of coming up with the smBPT gauge idea.

This proposal has been refined and differs from past work, but incorporates the work of @Dydymoon and @solarcurve - we would like to recognise there contributions and thank them for there efforts.

Copyright

Copyright and related rights waived via CC0.

6 Likes
2

Hello @TokenLogic, and thanks for sharing this proposal.

In the previous discussion, as well as the ARFC that has been voted in, regarding GHO’s liquidity strategy, two pools were qualified as “Primary” - GHO/USDC/USDT & GHO/LUSD, both with the perspective of eventually being included in the Safety Module, as far as I understood it.

I’d be curious to hear the reasoning leading to the drop of the consideration of the LUSD/GHO pool. Let’s keep in mind that we are discussing the Safety Module, aka assets that should be available and retain their value at all times, especially if the markets are distressed.

As we have seen recently during the USDC depeg last March, this is not given for USDC, and thus, for a GHO/USDC/USDT pool. The direct and immutable redeemability of LUSD against ETH helps to ensure that the backing value is preserved at all times, even if the most dramatic events.

The diversification provided by such an LP exposure is quite limited, considering a sizeable share of the GHO minted currently uses USDC backing. If the goal is to diversify exposure, a LUSD/GHO seems much more sensible and defensible. Considering the development timeline (apparently already started), ensuring the governance aligns on the assets that should be included should be a priority before it moves any further.

Thanks for taking this proactive step. Indeed, this proposal suggests an implementation that differs sizeably from the overall already-agreed strategy; let’s ensure the community approves this significant revision before moving further.

4 Likes
3

This is a strong proposal. Introducing GHO/USDC/USDT benefits the Aave Safety Module by enhancing asset diversification, reducing incentive spend, and bootstrapping GHO liquidity. Risk diversification is particularly important for any program designed to function as a backstop–this proposal addresses prior community concerns that the SM contained too much single asset exposure, reducing frontrunning risk should AAVE need to be sold to meet a shortfall event. Adding GHO/USDC/USDT is a nice first step for the diversification process, benefiting Aave DAO through accrued interest from GHO debt and stabilization of the GHO peg.

The Aura community supports this proposal and the excellent work by Aave contributors.

4

After years of only having one asset in the SM active which is the native token and thus being very vulnerable it only makes sense to add another asset.
With the recent purchases of vlAura and the BAL assets already being held in the treasury this new asset could get fast deep liquidity, if GHO will be reaching and holding its peg of course.
With the GLC being established and this TEMP CHECK, i think we are in the right direction of sustainability for the SM and helping GHO.
I would like to hear opinions on @TokenBrice comment regarding LUSD/GHO as an asset.

Also is there any rough timeline when to expect everything will be deployed?
Will Balancer and Aura also push some incentives?

2 Likes
5

Sharing some clarifications here: I’ve been working on the SM proposal over the past months, first under Llama scope until the end of September and still doing it on my own after their contract ended.

It’s important to know that since Matthew left Llama, he’s been trying to frontrun me on that topic and he was aiming to handle this work under TokenLogic to take credit for it.

As mentioned in the update shared here last month, the ARFC following Llama TEMP CHECKs approved where not posted yet for two reasons: AIP-42 requiring to reshape the strategy, and GHO depeg which is counter productive to the GHO committee goals.

(We can also note there that TL was supposed to post this ARFC about BAL upgrade, but I posted it with my account as Matthew refused to post it under TL, before deciding to go against my post without arguments).

This event (and many others before) led me to decide & communicate on the forum that I was leaving TokenLogic, and I clearly communicated about the fact that I’d keep working on the SM upgrade proposal on my own after Llama contract ended, as it can be seen here since the goal is to publish it once the GHO peg is restored.

Really shocked and sad to see attempts to steal part of my work again, but I guess it just demonstrates I was right to leave TokenLogic.

(On a side note, the fact that you were speaking in my name under TL without my consent was one of the reasons I stopped contributing to TL, but now you’re doing the same with the GHO committee, claiming it made recommendations that came from you as the committee wasn’t consulted: This ARFC approved that was supposed to be implemented but you changed it & delayed the CL strategy implementation to include Sommelier in this comment, mentioning a committee decision)

  1. This recognition would have been great if it was there from the start, unfortunately I know you added it after receiving comments about your initial post which you tried to present as yours.

  2. You knew I was working on the SM but didn’t reach out to ask to collaborate, or didn’t even discuss this topic with the committee.

  3. This proposal is a bad copy/paste of the update I shared below the Llama proposal but with missing info & misunderstanding about the strategy.

Yes SolarCurve came up with the smBPT idea, but you can’t seriously compare this, (or even your contributions ?) to mine on this proposal as you know the truth.

Anyway, to prove that this work was stolen, let’s compare this post with the last update I shared in September here:

Note that both GHO/USDC/USDT & GHO/LUSD are part of the assets I intend to propose in this ARFC update, as it makes no sense to only add centralized assets in the SM, and it won’t be efficient until GHO peg is restored…

Important to note that my comment in this post was totally ignored by the ACI & TL who pushed the proposal without even bothering to answer.

It’s really bad to see delegates & service providers act like this (or censoring the discussion & centralizing the decision as happened here)

I also shared this many times in a group about “SM & GHO” where there are Matthew, Stani, Marc & BGD, before I was kicked out of it by Matthew a few weeks ago while this group was unrelated to TokenLogic. I understand better why now.

Anyway, I hope that’s enough for the Aave DAO to see that this TEMP CHECK was copied & should be viewed as invalid considering previous temp checks were already approved & I’m already working on the ARFC.

TokenLogic arguable methods & frequent lack of transparency should also raise concerns about potentially approving a new service provider / delegate where Matthew Graham is involved.

5 Likes
6

Hi,

Does using GHO in safety module make sense? This to me seems like a way to boost GHO peg and potentially get more revenue for Aave but does not improve risk profile. Primary use case of Aave is a lender so boosting GHO peg should not cost depositors in terms of risk. I think a proposal like this needs to be accompanied by a proper risk assessment.

Doesn’t an LP token have more risks? It has all risks of USDC, USDT, balancer hack, then all the assets that GHO is composed of, and other depeg events.

GHO is backed by assets deposited onto Aave. Why not just add other assets to SM individually? Surely that would reduce risk?

Note: I am sure I do not understand full implications. Just giving my 2 cents.

1 Like
7

Thanks @TokenLogic for this proposal.

I just wanted to reiterate that the Balancer team is committed to continue innovating on top and around Aave. Supporting GHO liquidity is key to GHO’s success and a great way of doing so is by adding GHO/USDC/USDT BPT to the SM. This will also create stickier GHO liquidity as opposed to only adding individual tokens to the SM.

This is a good first step before adding more Balancer LP tokens (BPTs) to the SM.

I also wanted to say that @MatthewGraham, @Dydymoon and others might have some divergences but hey, you all want the best for Aave: let’s try as much as possible to put our differences aside and keep shipping!

5 Likes
8

The proposal to add the GHO/Stablecoin Pool Token to the Safety Module is an exceptionally effective idea for improving the stability of GHO, and I wholeheartedly support the adoption of this idea. However, I am opposed to adding centralized stablecoins, such as USDC and USDT, to the Safety Module due to the potential reduction in the overall censorship resistance of the protocol.

I recommend conducting a thorough investigation and consideration of the following two points:

  1. The option of including assets such as DAI, FRAX, crvUSD, and others as part of the GHO/Stablecoin Pool Token.

  2. The option of including GHO/Stablecoin Pool Tokens from other decentralized exchanges (DEXs) like Curve and Uniswap.

2 Likes
9

why is it better to use GHO instead of using assets individually?

surely GHO has more risk than individual assets?

10

Which assets and why is GHO more risky?

11

other assets like ETH, stETH, LINK etc.

The assets that make up GHO. This would separate out the risk.

GHO not only holds risk of all assets it is composed of, but if you use LP token then it also has an additional protocol risk.

Yes, i understand it will help GHO peg but GHO peg should not come at cost of extra risk for depositors.
Yes, I understand it is better than just Aave token in SM, but surely there are better options than using GHO.

12

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.