[Discussion] Migrate, Consolidate and Deploy Polygon Treasury


title: [Discussion] Migrate, Consolidate and Deploy Polygon Treasury
author: @llamaxyz - @MatthewGraham & @Dydymoon
shortDescription: Migrate liquidity from v2 to v3, consolidate assets and deploy a portion of the assets to earn yield.
created: 2022-12-10


Summary

@llamaxyz would like to kickstart a conversation within the community that involves redeeming, consolidating, and redeploying Aave’s assets held within the Polygon v2 Collector Contract and v3 Treasury addresses.

Abstract

The Polygon Aave v2 Collector Contract has accumulated holdings in several long tail assets over time. This publication intends to help the community restructure how Aave utilises its funds on Polygon. A high level overview of the various discussion topics are mentioned below:

  • Transfer - Assets from v2 Collector Contract to v3 Treasury
  • Transfers - BAL and CRV holdings to Ethereum v2 Collector Contract
  • Consolidation - Swap GHST, xSUSHI, and DPI for USDC
  • Deploy for Yield - Deposit DAI, USDC, and USDT funds into the newly created Aave Boosted Pool when it becomes available
  • Deploy for Yield - Deploy wMATIC 50/50 into newly created bb-a-wMATIC/MaticX & bb-a-wMATIC/stMATIC Pools when they become available

The goal of this discussion is to gauge the community’s thoughts on the ideas presented below. On instances where the community seems fairly aligned, Llama will proceed to generate an ARFC publication and progress a proposal through the governance process.

Motivation

Llama seeks to kick off a discussing relating to how Aave restructures it holdings held on Polygon network. With the migration from v2 to v3 needed and the potential launch of GHO being soon, there is an opportunity for Aave to reposition its assets to support additional revenue streams.

The below tables show the composition of the v2 and v3 revenue holdings.

Collector Contract


Address: matic:0x7734280A4337F37Fbf4651073Db7c28C80B339e9

Treasury Contract


Address: matic:0xe8599F3cc5D38a9aD6F3684cd5CEa72f10Dbc383

Background

The following provides insights and context that will help support the ideas mentioned in a later part of the publication.

  • Aave v3 offers a significant improvement in the ability to manage risk.

  • BAL and CRV voter escrow contracts are located on Ethereum. ie: veBAL and veCRV is only achievable on Ethereum.

  • Balancer v2 was launched April 2021, has over $1.2B in liquidity, and offers one of the largest bounties in the industry.

  • The Aave Boosted Pool utilising the current version of the AaveLinearPoolFactory directs around 75% to 80% of the deposited liquidity into Aave v2. ie: Deposit $100 into a Linear Pool, $75-$80 is deposited into Aave. There are three Linear Pools bb-a-USDC, bb-a-USDT and bb-a-DAI that make up the Aave Boosted Pool.

  • Balancer’s new AaveLinearPoolFactory static aToken contract developed by BGD, currently progressing through audit, will deposit tokens into Aave v3.

  • Balancer Liquidity Pools that contain >50% productive assets (Linear Pools, stMATIC, MaticX etc…) qualify for Core Pool status. This means 75% of Balancer’s revenue, 50% of the yield, from the pools is used to bribe for BAL emission rewards via Hidden Hand. ie: The pool generates $100 in yield from aTokens and swap fees, Balancer’s revenue is $50, and $37.5 is used for bribes on Hidden Hand. A minimum of $5k is needed for bribing.

Redeem Collector Contract aTokens and deposit into v3 Treasury

As has been widely discussed on the forum, Aave v3 offers enhanced risk management tools and presents a lower risk environment to users. Here, we propose that v2 aTokens be redeemed and transferred to the v3 Treasury address. From the v3 Treasury address, the tokens can be deposited into Aave v3 Reserves or deployed onto Balancer.

Utilisation across the v2 deployment is low and there is excess liquidity within each reserve that enables Aave’s aTokens to be redeemed without affecting users.

Llama proposes migrating liquidity at the first available opportunity. Once the ability to reroute v2 Collector Contract revenue to the v3 Treasury and/or migrate positions becomes readily available, this will be adopted and implemented. Currently, a migration tool is being worked on by @BgdLabs.

v3 Treasury Post Transfer

Consolidate Assets

Similar to a proposal relating to the Ethereum v2 Collector Contract consolidation, we suggest swapping all xSUSHI + DPI to USDC.

The general flow of events is redeeming xSUSHI and DPI from the v3 Treasury and then swapping the assets for USDC.

Even though these holdings are of small economic value, a swap contract can be created quickly and easily. Any premium applied will be just enough to cover gas and price impact when selling the assets to USDC.

Transfer BAL and CRV to Ethereum

Both BAL and CRV assets are considered strategic to the DAO’s future growth trajectory. Voting escrow contracts exist only on Ethereum and because of this, both BAL and CRV across all non-Ethereum networks need to be transferred back to Ethereum.

The exact method of transferring the BAL and CRV to Ethereum requires further investigation as there are non-matching smart contract addresses on each side of the Polygon bridge. These BAL and CRV holdings are small and this is more of a directional suggestion before diving deeper into the how to achieve this aspect of the proposal.

Deposit Funds into Balancer Linear Pools

In this section, we present the idea of depositing a portion of the consolidated v3 Treasury across several new Aave Boosted Pools on Balancer v2. The pools are to created by the Balancer or Llama teams prior to the funds being deposited. The pools will utilise the new AaveLinearPoolFactory that incorporates the new static aToken being developed by @bgdLabs. Once the new static aToken has passed through audit and is integrated by the Balancer team, the new liquidity pools can be created.

The three pools being proposed are as follows:

  1. Aave USD Boosted Pool consisting of three Linear Pools
    1. bb-a-USDC
    2. bb-a-DAI
    3. bb-a-USDT
  2. Aave MaticX Boosted Pool
    1. bb-a-wMATIC
    2. MaticX
  3. Aave stMATIC Boosted Pool
    1. bb-a-wMATIC
    2. stMATIC

Each pool contains greater than 50% productive assets and in this instance, 100% are yield bearing assets - and therefore qualifies for Core Pool status. 50% of the yield generated from Aave v3, staking MATIC, and swap fees will accrue as revenue to Balancer. Balancer will retain 25% of the revenue and direct 75% toward Hidden Hand bribes to attract more veBAL voting support for the pools.

Aave is expected to earn 50% of the aToken-derived yield, 50% of the staked MATIC yield from the respective receipt token and 50% of the fees + the BAL emission rewards by being a Liquidity Providers (LPs) within the pool. Llama and Balancer will work together to create the gauges and progress them through Balancer’s governance process.

Furthermore, any Liquidity Mining (LM) on Aave that is accrued by the Balancer LPs is to be redeemed by the Balancer team and will either be distributed to LPs or used to bribe on Hidden Hand to attract more BAL emissions to the pools. This is worth noting, given the potential Polygon Foundation, Stader Labs, and Lido Community LM programs.

This is most likely to be applicable to wMATIC deposits and thus, the bb-a-wMATIC pools. A proposal similar to the Whitelist Balancer Liquidity Mining Claim will be needed to support claiming the LM rewards.

The primary focus here is to be an early supporter of these pools, earn BAL rewards whilst also retaining 50% of the Aave yield, and gain exposure to stMATIC + MaticX yield. There is additional smart contract risk surface area to consider.

The bb-a-USD pool on Polygon is to become the ideal pair for the future GHO Liquidity Pool on Balancer: bb-a-USD/GHO. We hope other communities like QiDAO’s create pools such as bb-a-USD / bb-a-MAI for their MAI stablecoin. Creating this pool is a continuation of the strategy started on Ethereum to make the bb-a-USD the equivalent to the 3CRV pool across all networks within the Balancer ecosystem.

The Linear Pools (bb-a-wMATIC) utilise the AaveLinearPoolFactory that was created by the core Balancer team as well as the static aToken developed by @BGDlabs. Certora is the auditing team for both communities. Due to the involvement of these teams and the performance of the existing bb-a-USD pool on Ethereum, we are comfortable with this added smart contract risk.

The table below show compares the relative yield.

The below details the proposed allocation to be discussed:

Aave USD Boosted Pool (bb-a-USD)


Depositing approximately half of the main three stablecoin holdings on Polygon is being proposed. As the pool is equal amounts of USDC, DAI, and USDT, we are opting to limit the initial deposit based upon the smaller of the three holdings. This also enables the DAO to retain $3.96M of stablecoins for other potential uses.

As the current total stablecoin holdings of the Aave Treasury is $31.22M, this strategy aims to allocate 12.8% of the total stablecoin treasury.

Aave Matic Boosted Pool


When creating the pools, equal amounts of wMATIC and staked MATIC receipt token are to be deposited. MaticX is to be acquired by interacting with the Stader Labs contract by depositing wMATIC and receiving MaticX. For stMATIC, this will likely be via a swap contract where wMATIC is swapped with stMATIC based upon the Chainlink oracle pricing.

This will deposit the wMATIC equally across the bb-a-wMATIC / MaticX and bb-a-wMATIC / MaticX Balancer Liquidity Pools. Aave will gain exposure to MaticX, stMATIC, awMATIC, BAL incentives, and swap fee yield. The acquired BAL is strategic for the DAO as it can be converted to veBAL on Ethereum. Aave forgoes 50% of the a-wMATIC, stMATIC and MaticX yield which is revenue to the Balancer DAO, whilst gaining 50% swap fee yield and will receive BAL incentives. We expect the yield for BAL emissions to exceed the yield from Aave v3. BAL rewards have the added strategic upside for revenue generation within the Aave ecosystem.

Discussions with Lido and Stader indicate the bb-a-wMATIC pairing will become the primary pool for stMATIC and MaticX on Balancer, respectively. The Balancer community has indicated strong support for these liquidity pools. All constituents are productive and therefore supportive of revenue generation for the community. We are hopeful of strong support from prominent veBAL holders.

The larger the bb-a-wMATIC pool liquidity grows, the more wMATIC that is deposited into the Aave v3 wMATIC Reserve. Communities are building leveraged yield strategies on Aave v3 that borrow wMATIC. The more wMATIC entering the Reserve, the greater the revenue from borrowing demand as the borrowing cost of wMATIC will loosely equalise with staked MATIC returns over time. It is important to note a 50/50 bb-a-wMATIC pool will allocated around 35-40% of its liquidity into Aave v3 wMATIC Reserve.

Discussion

We are hoping to receive feedback from the broader community in the comments\ section below. This proposal is a blend of treasury management along with preparing the asset allocation for the potential GHO launch on Polygon. This may be some time away and this proposal will take time to implement. Creating and supporting Linear Pools on Balancer is a strategic directional decision for the community. There is a trade off between smart contract risk, yield from Aave and/or Balancer, plus the potential upside of positioning Aave’s assets to earn BAL that then contributes to supporting the new aToken-based AMM Aave v3 deployment being discussed.

Llama is keen to work with the community and understand the community’s risk appetite. The suggestions present here are positioning the DAO to commit funds where possible to support growth based initiatives. In essence, we seek to give GHO and the new AMM v3 deployment every chance of success. The initial capital deployment is small and leans into being conservative initially and then adding to the strategies in time based upon the feedback received from the community.

Specification

The following assets held within the v2 Collector Contract are to be redeemed and transferred to the v3 Treasury wallet.


The following assets upon being redeemed for the underlying are to be swapped for USDC via a swap contract.


The following assets upon being redeemed from v3 Treasury are to be transferred to the Ethereum v2 Collector Contract.


The following assets are proposed to be deployed in the newly created liquidity pools on Balancer.

Aave USD Boosted Pool (bb-a-USD)


Table above assumes Collector Contract funds are deposited into v3 Treasury.

Aave Matic Boosted Pool


Table above assumes Collector Contract funds are deposited into v3 Treasury.

Next Steps

There are four key aspects to this proposal.

  1. Redeem and Transfer assets v2 Collector Contract to v3 Treasury wallet
  2. Clean up the Collector Contract
  3. Transfer BAL & CRV to Ethereum Collector Contract
  4. Deploy assets to earn yield on Balancer

This is the first discussion around deploying funds to earn yield and influence in another community’s tokenomics to encourage the adoption and growth of Aave. The initial deployment values are smaller than what is achievable on Ethereum and gives the community the opportunity to start small before leaning into this philosophy with additional capital on other networks.

It is expected that point 4) above will generate the most dialog and the learnings here will shape how we all collectively position the communities assets. When the community converges towards a particular outcome, a separate ARFC will be created progressing that aspect through governance.

We would like to encourage all community contributors to express their thoughts in the comments. For those who are not comfortable sharing publicly, please do reach out via direct message.

Copyright

Copyright and related rights waived via CC0.

2 Likes

Hi @Llamaxyz,

This an exciting proposal, and hopefully the first of many for turning a % of the Aave Treasury into productive assets.

As this proposal covers only Polygon, is there a plan to extend a similar strategy on Ethereum mainnet?

If so, I believe a future proposal should be tabled to gauge the maximum exposure of the total treasury the community is willing to take towards a counterparty. For example, I am of the belief that no more than 20% of treasury assets should be deployed to Balancer strategies across all networks. Such a % ensures that if there were a failing of any kind, in Balancer or another counterparty that the risk is limited. In many other cases of treasury management, this is the norm.

In general, I am supportive of this proposal however, before proceeding I believe there should be a discussion about the maximum risk the community is willing to face towards a counterparty.

Look forward to your feedback.

2 Likes

Hi @G-Blockchain,

Thank you for the feedback and the questions on this proposal.

@Llamaxyz views the Treasuries and Collector Contract in aggregate which allows for more generous allocations on a specific network whilst retaining overall diversification at holistic level across Aave’s funds.

At the time of writing, 12.8% of Aave’s stable coins are being used to provide early liquidity into the Polygon bb-a-USD pool. This is less than 20% figure mentioned in the comment section. If Polygon’s funds are viewed in isolation then yes, this allocation is concentrated. Depending upon feedback in the comments section, we are happy to refine the allocation to the bb-a-USD to a 20% allocation.

@Llamaxyz would like to encourage the community to provide feedback on any aspect of this proposal.

Hi Everyone :wave:

We shared the following forum post which follows on from this forum discussion with some minor changes.

Changes include, swapping GHST (not in the proposal above), DPI, SUSHI and JEUR to USDT rather than USDC. The logic for USDT is that Aave has larger USDC and DAI holdings on Polygon and if the community was to deposit into a 3pool as outline in the proposal above additional USDT would be beneficial.