[ARFC] Aave Deployment on Celo

[PennBlockchain]

Title: [ARFC] Aave Deployment on Celo

Author: @pennblockchain, @Isha - Isha Varshney (Celo Foundation)

Date: 2024-05-09

ARFC updated by ACI on 2025-03-05 to update & change latest Risk Parameters

---

Summary:

This ARFC proposes the deployment of Aave on Celo to expand Aave real-world lending and green asset collateral to users around the world, with the provision of rewards to bootstrap liquidity.

Abstract:

Upon approval by the Aave community, Aave will be deployed on Celo with a selection of suitable assets for collateral, including CELO, USDC, USDT, cUSD and cEUR. This deployment aims to enhance the growth and collaboration between the Aave and Celo communities, providing new opportunities for both ecosystems. This deployment will be incentivized by the tune of $3M over 3 years (with option to extend) by Celo Foundation directly or through ecosystem partners.

Motivation:

Deploying Aave on Celo presents potential benefits for both Aave and Celo ecosystems. The Celo ecosystem offers a unique value proposition featuring real-world usage and ReFi. As Aave will be given a prominent role on Celo, this deployment offers Aave differentiated expansion and revenue opportunities. For more details on Celo’s real-world usage and key features, please refer to the temp check.

1. A deployment on Celo expands AAVE’s utility and user base to a unique population of Celo’s mobile-first community which has grown significantly 2 over the past few months led by key launches that have brought in returning real-world users in emerging markets.
2. Aave will be heavily utilized for real-world DeFi and ReFi use cases, particularly in emerging markets given Celo’s growing presence in these regions led by protocols like GoodDollar, Impact Market and the Opera Minipay wallet.
3. The Celo Foundation directly or through ecosystem partners will allocate $3M over 3 years (with option to extend) as well as the option for Retroactive Public Goods Funding as an incentive for Aave users to support the adoption and bootstrap growth on Celo. This deployment will bring new users to the Aave platform and reach more users from the Celo community.
4. The Mento protocol on Celo proposed returning 120M CELO to the Celo Community Fund, which the Celo community has expressed interest in placing in Aave to boost on-chain liquidity for CELO (pending a governance process). More details here 3.
5. In the long-term, this Celo x AAVE deployment has the potential opportunity to expand into RWA tokens, natural capital-backed assets such as tokenized carbon credit and other ReFi assets (conditional upon rigorous liquidity and credit assessments by Chaos Labs and others) etc.

Specification:

The ARFC suggests deploying Aave on Celo with the following assets for collateral: CELO, USDC, USDT, cUSD and cEUR.

The table below combines both risk managers’ suggested risk parameters for each asset previously submitted. The highlighted values are Gauntlet’s more liberal recommendations. Over time, we expect we can continue to increase caps as the Celo blockchain grows through the development of more real-world use cases and the transition to an L2 on Ethereum.

ARFC updated 2025-03-05 with latest Risk Parameters

Asset	CELO	USDC	USDT	cUSD	cEUR
Isolation Mode	Yes	No	No	No	No
Enable Borrow	Yes	Yes	Yes	Yes	Yes
Enable Collateral	Yes	Yes	Yes	No	No
Loan To Value	55%	75%	75%	-	-
Liquidation Threshold	61%	78%	78%	-	-
Liquidation Bonus	10%	5%	5%	-	-
Reserve Factor	20%	10%	10%	15%	15%
Liquidation Protocol Fee	10%	10%	10%	-	-
Supply Cap	1,000,000	3,500,000	2,000,000	1,300,000	80,000
Borrow Cap	100,000	3,150,000	1,800,000	1,170,000	72,000
Debt Ceiling	500,000$	-	-	-	-
UOptimal	45%	90%	90%	90%	90%
Base	0	0	0	0	0
Slope1	10%	7.5%	7.5%	7.5%	7.5%
Slope2	150%	40%	40%	75%	75%
Stable Borrowing	No	No	No	No	No
Flashloanable	Yes	Yes	Yes	Yes	Yes
Siloed Borrowing	No	No	No	No	No
Borrowable in Isolation	No	Yes	Yes	Yes	Yes
E-Mode Category	N/A	N/A	N/A	N/A	N/A

Deployment Checklist:

• Feedback from both risk service providers
• BGD Labs Infrastructure/technical evaluation Report
• Oracle recommendation (Chainlink 5)
• The graph support
• Cross-chain messaging infrastructure for governance

Disclaimer

Franklin DAO is not compensated in any form by Celo or creating this ARFC proposal. This ARFC has been prepared solely to facilitate community discussion. At the time of writing, the author has been delegated $0 worth of assets on the Celo chain.

Next Steps

1. Refine the ARFC based on community feedback and risk service provider recommendations.
2. If consensus is reached and the proposal is refined, submit the ARFC for a snapshot vote for final approval.
3. If consensus is reached, submit a deployment AIP to activate Aave on Celo.

Copyright

Copyright and related rights waived via CC0.

[EzR3aL]

So accrodingly to Defilama the current TVL on CELO is sitting at around 123m with Moola being the biggest lending protocol with around 4m TVL. The biggest protocol Mento holds about 70% of all TVL.
My main concern here is that in the last 30 days there have only been outflows of assets from that chain. TVL is basically sitting in the same area for more than a year already.
I don’t know if it makes sense to use resources and funds to deploy an market there where the potential return might be the same we had with Fantom back then.

Also is there still the risk with third-party bridges as BGDLabs outlined in their technical evaluation?
How is Celo attracting user to use Aave? What will they use Aave for? Can the proposed assets be used on that chain for yield, thus making a lending market important?

[Tor_GAINS]

I also see very little upside in deploying on Celo

Let’s try to avoid another fiasco similar to what happened with Fantom and Harmony

A new cycle has just begun and we need to be even more pragmatic about how resources are allocated

[MarkusBerlin]

Great initiative to deploy Aave on Celo.

Celo has strong user traction, DAUs and MAUs are all growing strong and Celo is for daily active users consistently in the top 10 chains.

Wallets like MiniPay, Valora, Kotani Pay etc. show fast growth, especially across Africa, unlocking a new user base for Aave.

Mento, a local-currency stablecoin project on Celo is launching long-tail stablecoins for different markets, especially for markets with a larger credit-gap, which can bring new users into Aave. The stablecoin treasury is fully on-chain, lending markets are important for this one as well.

[rachblondon]

Very much in favour for this. With MiniPay, Celo is gaining fast traction across Nigeria, Kenya and other African countries.

With the rise of stable token usage across emerging markets, it would be great to have Aave as a yield source and lending platform on Celo

[MiniWhale]

Fully support this initiative! Great momentum behind Celo which Aave could also benefit from!

[daveytea]

Besides technical concerns which i’m sure more relevant parties can give their opinion on (i.e @bgdlabs), I have noticed that the Celo community has been quite active in emerging markets, mostly in Africa AFAIK. Projects such as HaloFi (previously GoodGhosting, which was an OG Aave Grantee) has deployed there with success. However they had to integrate with other lending providers since Aave was not yet deployed there.

TVL may be relatively low, but real world usage is legitimate, albeit low volume due to the target markets. I support this as it’d help bring Aave to new markets it doesn’t currently serve well, and may create new use cases to built in these emerging markets.

[ACI]

The current proposal has been escalated to ARFC Snapshot.

Vote will start tomorrow, we encourage everyone to participate.

[ChaosLabs]

Overview

Following BGD’s technical evaluation of Celo and its suitability to deploy an instance of the Aave protocol, Chaos Labs has provided a framework and recommendations for the rest of the risk parameters for community visibility (found at the bottom of this post).

General Overview

• TVL: $111M
  • TVL fell sharply in April, coinciding with a broader pullback in markets.

• Volume has been consistently higher since February, with multiple days with over $2M in daily volume.
  • Of the DEXes on Celo, Uniswap facilitates more than 90% of volumes.
  • There is an instance of Curve Finance on Celo, though volumes are relatively low.
  • Ubeswap — a Uniswap V2 fork — was previously the largest DEX by volume on Celo, but has since fallen to a distant second place.

• Overall, the largest protocol on Celo is Mento.
  • Mento allows users to mint stablecoins against CELO and other collateral; stablecoin reserves are backed 50/50 by USDC and DAI.
  • $20M of Curve’s TVL comes from Mento providing liquidity to the cUSDC/USDC pool.

Liquidity

The majority of liquidity for the assets in question — CELO, USDT, USDC, WETH — is on Uniswap. The largest pool is CELO/WETH, with $24.87M TVL and $176.99M monthly volume.

Pools for native USDT ($70M circulating supply, $30M of which is held by user 0x5754284f345afc66a98fbB0a0Afe71e0F007B949, the Tether Treasury) and newly added native USDC ($16.33M circulating supply) are still relatively small.

Given their relatively new addition and the prevalence of native and bridged stablecoins on the chain, native USDC and USDT have high slippage.

Echoing BGD’s analysis, we do not recommend listing bridged assets — including Wormhole WETH — because of inherent risks associated with bridging.

Recommendations

Liquidation Threshold

For CELO, the only volatile asset we recommend listing, a 30-day annualized volatility of 200.78 leads us to recommend a liquidation threshold of 60% and an LTV of 55%.

For USDT and USDC, we recommend aligning the LT and LTV with similar parameters on other deployments.

Enable Borrow

We support listing all assets as borrowable under reasonable limits as we do not observe a significant risk to the protocol by allowing this.

Enable Collateral

We recommend CELO, USDT, and USDC be enabled as collateral assets, with CELO subject to a conservative debt ceiling to first observe user behavior. cUSD and cEUR are not recommended to be listed as collateral assets because of their mixed backing, including assets held on Ethereum.

Borrowable in Isolation Mode

We recommend enabling cUSD, cEUR, USDT, and USDC as borrowable in Isolation Mode.

Supply Cap, Borrow Cap, and Liquidation Bonus

As described in our Metis and BNB Chain deployment recommendations, we recommend setting the Supply Cap at 2x the liquidity available under the Liquidation Bonus price impact. The Liquidation Bonus is aligned with similar assets on different deployments.

Other

We recommend aligning the Reserve Factor, Liquidation Protocol Fee, and Interest Rate curves with other deployments.

Full Recommendations

Following the above analysis, we recommend the following parameters:

	CELO	cUSD	cEUR	USDT	USDC
Isolation Mode	YES	NO	NO	NO	NO
Enable Borrow	NO	YES	YES	YES	YES
Enable Collateral	YES	NO	NO	YES	YES
Emode Category	N/A	N/A	N/A	N/A	N/A
Loan To Value	55%	-	-	75%	75%
Liquidation Threshold	61%	-	-	78%	78%
Liquidation Bonus	10%	-	-	5%	5%
Reserve Factor	20%	15%	15%	10%	10%
Liquidation Protocol Fee	10%	-	-	10%	10%
Supply Cap	1,000,000	2,000,000	800,000	700,000	2,000,000
Borrow Cap	500,000	1,800,000	700,000	650,000	1,800,000
Debt Ceiling	$500K	-	-	-	-
uOptimal	45%	90%	90%	90%	90%
Base	0%	0%	0%	0%	0%
Slope1	10%	7%	7%	7%	7%
Slope2	150%	75%	75%	75%	75%
Stable Borrowing	Disabled	Disabled	Disabled	Disabled	Disabled
Flashloanable	YES	YES	YES	YES	YES
Siloed Borrowing	NO	NO	NO	NO	NO
Borrowable in Isolation	NO	YES	YES	YES	YES

[Isha]

Thank you for your comment @EzR3aL, really appreciate the time you have taken on reviewing this proposal. Let me add some additional information that should help address some of the issues you have raised:

1. Growing Ecosystem -

• Top 10 DAU Chain: Over the last 12 months, Celo’s daily active users have grown dramatically, averaging over 370,000 DAUs in recent weeks, placing Celo among the top 10 blockchains in terms of daily active users (Source: Token Terminal).
• Opera MiniPay Wallet: Opera has integrated the MiniPay Stablecoin Wallet, built on Celo, within their mobile browser. The Opera browser is accessed by over 100 million monthly active users, primarily in Africa, providing a robust channel for seamless and secure mobile payments across the continent. MiniPay wallet launched in September 2023, and has exceeded 2M + users across Nigeria, Kenya, and Ghana. Since the users hold value in dollar stablecoins in the wallet, there is a large opportunity to add the Aave markets as a yield source to the growing MiniPay user base of 2 million+.
• We expect Aave on Celo to be heavily utilized for real-world DeFi and ReFi use cases in emerging markets. Most of the users on Celo are retail consumers in markets such as Africa and Latam, many of whom were onboarded to Web3 through Celo. This makes Celo a unique distribution channel for Aave to expand its geographic presence as most of our user base is net new to Web3. This also means there should not be any overlap between Aave’s existing user base.
• Given the above, an Aave deployment on Celo, with potential integrations of the lending market into projects such as the MiniPay wallet with its 2M+ users would likely 10x the DAUs on Aave, far exceeding the DAUs numbers on some of the other L1s and L2s

2. Stablecoins
   There are multiple stablecoins in the Celo ecosystem. Some key aspects:

• Native USDT and USDC Launch: Recent launches of native USDT and USDC have boosted liquidity and streamlined transactions within Celo’s ecosystem. There is already pent-up demand for yield on these stables which is clearly demonstrated by the fact that the only lending market on Celo (Moola Market) is tracking a borrow rate of ~26% on cUSD, one of the stable assets on the Mento Protocol that tracks the value of US Dollar. Since USDC and USDT are not listed on Moola, there is a clear opportunity for Aave to be the first lending market in the Celo ecosystem to list these stablecoins.
• Gas Fee: Unique to Celo, users can pay gas fees in multiple ERC-20 tokens, including native USDT and USDC.
• Forex Stablecoins: Celo has positioned itself as a major player in the stablecoin market, particularly focusing on emerging markets where stable assets are crucial for economic stability amidst hyperinflation and financial growth. Celo supports regional stablecoins such as cEUR, cREAL, and eXOF (tracking the Euro, Brazilian Real and West African Franc, respectively) and continues to expand its stablecoin offerings. This diversity extends the utility and reach of Aave markets on Celo as there is incredible potential for foreign currency exchange pairs on the lending market.

3. Real World Assets

Celo is a top 5 chain for RWAs: With the focus on real-world use cases, Celo has steadily emerged as the chain of choice for RWA projects including Centrifuge, Huma Finance, Jia, Untangled Finance, Zoth, and is constantly welcoming RWA protocols to the ecosystem. Aave 2030s increased focus on RWAs presents an opportunity for the Aave deployment on Celo to expand into RWA tokens, natural capital-backed assets such as tokenized carbon credit, and other ReFi assets, aligning with both organizations’ strategic focus.

4. Bridged Assets

Considering the recommendation of BGD Labs, you will notice that we have not added any bridged assets to our ARFC proposal. Our proposal is that Aave deploys on Celo with a selection of native assets, including CELO, USDC, USDT, cUSD, and cEUR.

Happy to jump on a call to discuss these further in detail. Thanks!

[Isha]

Thanks @ChaosLabs for posting the assessment !

[EzR3aL]

Hi @Isha and thank you for your detailed answer.
It may be that I underestimated Celo as a chain. It may not be the biggest chain but it is getting traction. Given that the proposed assets seem very safe for the protocol and its user and that the Celo foundation is offering liquidity bootstrapping it may be worth to explore an Aave deployment there.

[Isha]

Thanks @EzR3aL for taking the time to review the proposal and all the additional information. Really appreciate the support!

[Isha]

As a signal of our continued commitment to the Aave community, we are proposing an additional one-time grant of $15,000 from the Celo Foundation to absorb the initial cost of deployment as detailed in this proposal.

[SaucyBlock]

We are supportive of deploying Aave V3 to Celo for the following three reasons:

• Given that Celo targets mobile users, Aave is likely to attract a new user base by deploying Aave V3 on Celo. Additionally, the Celo ecosystem is highly diverse, and the integration with Aave V3 is expected to create many positive-sum.

• The Celo Foundation is positive about a partnership with Aave, as evidenced by their commitment to a $3 million incentive program* and grants of $15,000 (as deployment costs).

• As Aave V3 Fork does not exist on Celo, the deployment of Aave V3 can offer Celo users a new lending market experience

*We believe that committing to a three-year incentive is not a wise decision in this industry, and we recommend to the Celo Foundation that they concentrate the majority of the incentives in the first year instead.

[ACI]

After Snapshot monitoring, the current ARFC Snapshot ended yesterday, reaching both Quorum and YAE as winning option, with 689K votes.

Therefore the ARFC has passed, and next step will be the publication of an AIP for final confirmation to deploy Aave on Celo.

[sid_areta]

Hi @ACI, is there any update on this deployment?

We recently saw that Uniswap trading volumes have increased significantly on Celo, and according to our rationale for voting in favour of this proposal, we think there are numerous benefits to deploying Aave on Celo. It’d be great to know where we are in the process as it has been 5+ months since this ARFC was passed.

[sakulstra]

@sid_areta probably it makes sense to wait for the Celo L2 migration (as i understand, they are “close” L2 Testnet Update and Upcoming Code Audit - Protocol - Celo Forum).

Imo having some patience here makes sense as:

• such a massive upgrade does not come without risk.
• one a L2 with a proper native bridge, I think it’s reasonable to believe that the current bridged tokens will lose relevance.

[sid_areta]

Hey @sakulstra thanks for the update. So to confirm, do you think the nature of the deployment would change once Celo is a native L2? And hence the business value of the deployment would change too?

[sakulstra]

So to confirm, do you think the nature of the deployment would change once Celo is a native L2? And hence the business value of the deployment would change too?

Not sure if i understand the question. I think, no? My concern is 100% on the security/stability side of things.

My point is that in a few weeks Celo will undergo a massive upgrade. Even networks who started as a rollup had some growing pains (just check how often the sequencer was down on metis, check how often zkEVM still is down etc pp) - so deploying today, hoping for the best and then fighting with sequencer downtime and similar in a few weeks seems reckless and not worth the risk.

Therefore I think it’s more reasonable to just wait the last few weeks until the hardfork and then re-evaluate afterwards.

To be clear: This is just my personal opinion. Not representing anyone but myself with this opinion.