CP103: Launch an RWA Lending Market

uses-components: cp4, cp2
authors: Lucas Vogelsang (@lucasvo, lucas@centrifuge.foundation), @ImdioR
contributors:
technical-proposal: no
wallet: 4d9knu1Wnr5mai3SJoGnfKethQE1TB997D7qPuvZq4NR93LM
impacts:
date-proposed:

Short Summary

Centrifuge should create an RWA focused lending market to give users the ability to lend and borrow against eligible pool tokens. This brings additional utility to RWAs and helps achieve our goal of bringing financial markets onchain.

High-level Objective

Composability with DeFi protocols and giving RWAs onchain utility is a vital part of what’s needed for a thriving RWA ecosystem. Centrifuge worked closely with MakerDAO to onboard RWAs as collateral to DAI and later on launched the RWA Market with Aave as the first onchain lending market.

Centrifuge Pools are a great way for primary issuers to bring products onchain. Often requested features by holders of these tokens are:

  1. Leverage in the form of being able to borrow against these assets
  2. Short-term liquidity while redemptions are processed.

We propose to deploy a lending market for Centrifuge RWA assets built on Base. We see Base as an ideal blockchain to build on because of Coinbase attestations that allow for very easy access control to the market.

Background

Market Structure for Financial Products

Financial products benefit from having a robust market infrastructure. Financial products are more useful to owners if they have access to short them, to leverage, or to get immediate liquidity. In the long run DeFi primitives can offer a lot of these functions not just to crypto native tokens but also RWAs. This proposal is a major move in this direction.

For example, while Treasuries onchain are an exciting development in DeFi becoming more mature, these instruments lack the utility they would in TradFi. Shorting Treasuries is something anyone can do from their brokerage account, or offered via ETFs. Posting Treasuries as collateral for leverage is also widely utilized as your collateral grows over time helping prevent liquidation events. Additionally, borrowing against Treasuries for instant liquidity is common, especially for those in leveraged or low liquidity positions. DeFi users also don’t have the ability to speculate on rate changes as you would in TradFi, most commonly used to hedge interest rate exposure through options or interest rate swaps. Yes, vanilla real-world assets are being bridged onchain, but they lack all of the utility that these assets come with in the traditional world. Users can’t hedge, speculate, borrow, or service their liquidity needs with or on RWAs today. If DeFi wants to expand to institutions, the key infrastructure they already enjoy needs to be here first.

Review of the RWA Market with Aave

Centrifuge DAO members and Aave jointly worked on the RWA Market, a separate deployment of the Aave lending market that accepted different Tinlake pools. The Aave Request for Comments has a lot of details on the original design. The market was the first of its kind and pioneered this structure. At the peak, the market grew close to $18M in total assets. However its use has declined for a few reasons:

  • The interest rate relative to other DeFi rates was too low
  • Onboarding to the market was cumbersome and slow
  • The market couldn’t always promise liquidity because the underlying assets had long redemption periods.

This proposal makes a few improvements over the original design of the Aave RWA Market that we believe serves the user needs of today:

  • Require highly liquid underlying collateral (nothing that has less than 7 day liquidity or liquid secondary markets that can step in when redemptions happen)
  • Require isolated markets that give lenders better control over where they want to deploy liquidity
  • Higher target interest rate
  • More user friendly onboarding

Creating a lending market for RWAs

What are the use cases for lending markets?

There are a few reasons why users would want to borrow against RWAs, they ultimately fall into two categories:

  1. Access liquidity when redeeming from the issuer: a need for immediate liquidity
  2. Borrow cash while still earning interest on the underlying asset: this could be to get extra leverage, to short the borrowed asset, or if the market is unbalanced and rates are simply too low.

There are several benefits of supplying liquidity as a lender:

  • Increased liquidity versus investing into the underlying asset
  • Creating a market for instant redemptions can have higher yields

Creating access to instant liquidity on redemption
The current redemption process for pools on Centrifuge usually looks as follows:

redemption_process

Upon triggering a redemption there usually is a waiting period which even for the most liquid pool on Centrifuge today (currently Anemoy LTF) is during US market hours roughly 8 hours and more on weekends. Having 24/7 liquidity on these tokens is a huge benefit and looking at other products, something users are willing to pay a premium for.

Generally, issuers try to offer better liquidity by keeping some cash balance in the pool which is not ideal because it reduces yield for everyone else. Or they try to maintain credit lines themselves which is also costly. An open lending market means anyone can provide liquidity in these instances and available liquidity follows supply and demand.

liquidity_on_redemption

Centrifuge should create a mechanism by which tokens can be put into redemption and used as collateral. This would allow a very simple process for any owners of pool tokens to get instant liquidity for tokens they want to redeem.

Management of Market Parameters

The DAO should work to establish risk management processes for the market. As part of the implementation of this CP, a governance process for the market should be designed and implemented resulting in a future CP.

Qualifying Collateral

Because the market needs high liquidity, the market should initially be limited to assets with high underlying liquidity that can reasonably redeem a large portion of the collateral in less than a week. The final collateral should be determined at launch.

Legal Structure

The DAO instructs k/factory to research and design an appropriate legal structure for the lending market that protects lenders and creates an operating framework for all participants ensuring institutional users are able to interface with this vault. It instructs the k/factory to set up any necessary legal entities and report on its progress. The legal structure chosen for this has a few objectives as listed below:

It is important to prevent sanctioned entities and persons from interacting with the market. Complex onboarding requirements are a big hurdle for users to use the market so ideally we can allow any users with existing KYC and verifiable credentials to use the market. With over 125M accounts on Coinbase, we can use their Attestations on Base to allow fast (one-click) and safe onboarding for a large group of potential users.

The legal entity we will set-up will be a bankruptcy remote special purpose vehicle.

Technical Design

Choice of lending protocol

After evaluating different lending protocols (Aave, Compound, Euler and Morpho), the initial recommendation is to adapt Morpho and build on Morpho Blue and Metamorpho. Morpho recently passed a proposal to deploy on Base which was a prerequisite for consideration. This proposal is to deploy with Morpho because of the following benefits:

  • Isolated markets means that it’s easy to segregate this market and not have depositors and lenders for other assets mixed.
  • Isolating this market is also required for legal reasons because most RWAs are permissioned tokens and require onboarding by counterparties.
  • The permissionless ability to launch new markets means no complicated forking has to take place and there isn’t a new codebase that has to be maintained by us. See the Morpho Blue documentation.
  • The Metamorpho Vault concept means the Centrifuge DAO could create a vault and support setting parameters and aggregating liquidity. See MetaMorpho documentation.
  • With currently over $500M in deposits and over $250M in borrows, Morpho is battle tested and has an active user base.
  • Morpho Blue is already been used for RWA assets namely IB01 and mTBILL and is particularly well suited for their custom set-up

Permissioning the market

The market requires permissioning of lenders, borrowers and liquidators. This can be achieved by wrapping the lending asset (USDC) into wrapped UDSC that enforces permissioning on wrap/unwrap and transfer.

Creating a redemption wrapper

The current implementation of Centrifuge requires the pool tokens to move into the redemption contract when an investor decides to redeem and as such using them as collateral is not possible without an additional smart contract. To launch the market, a wrapper contract has to be built that mints a token to be used as collateral in the market for every token the investor submits for redemption. Upon redemption, the wrapped token has to be able to get converted into USDC.

Deploying on Base

While not a strict requirement, deploying this market on Base has several benefits; Centrifuge is already live, it has low transaction fees and most importantly, it allows the seamless integration of Coinbase attestations.

An alternative would be to use a message bridge to make Coinbase attestations available on mainnet.

Funding Request

This proposal requests funding from the DAO treasury to implement this market and has the following initial funding request.

Name Description Budget
Creation of DAO controlled foundation Establish a Cayman foundation controlled by the DAO $60’000
Research and develop legal structure for market $150’000
Risk assessment and market management Engage DeFi and TradFi experts to come up with the appropriate initial market parameters and create an ongoing process to manage the market in the DAO. $80’000
Smart contract development Develop smart contracts to create permissioning features and integrate with a lending market. $75’000
Code Audits Security audit of the smart contracts $40’000
Total $405’000

The EMA 7 days price for CFG is 0.669, thus the request for this proposal from the treasury is 606,000 CFG. Any extra funds (unused) after converting, if the proposal passes, would be returned to the Centrifuge Treasury account in 14 days. The proposal requests these funds to be transferred to this wallet: 4d9knu1Wnr5mai3SJoGnfKethQE1TB997D7qPuvZq4NR93LM.

Alignment to the mission of Centrifuge DAO

This proposal to build lending infrastructure for LTF and other liquid RWA tokens on Centrifuge. It’s the first step in giving utility to these tokens, but also RWAs as a whole regardless of asset class. Treasury tokens are a clear first asset class to build utility around as they have the most opportunity for DeFi composability, and the most use cases due to their riskless and stable nature as well as high liquidity.

Building infrastructure around Centrifuge issued tokens is a key advantage for us over other RWA protocols as they stand today. It serves as a crucial selling point for institutions coming onchain who already have utility with their assets. Infrastructure is also what will allow our DeFi native users to utilize their RWAs as they would with their crypto native assets. All of this serves the purpose of helping grow the TVL of Centrifuge, and maintaining our position as the largest RWA protocol.


Update 05/21/2024: The proposal has been submitted on Github: cps/cps/CP103/CP103.md at main · centrifuge/cps · GitHub

16 Likes

This is a great proposal. Could less-liquid collaterals be involved but potentially have a limited market size? Say 20% of the total pool volume, or whatever is approved by issuer/community?

2 Likes

The volatility in DeFi leads to lending markets growing and shrinking quite a bit. It can very well happen that the market increases in size in an up market by 2x and then shrinks again. In the original RWA Market the market peaked at $18M IIRC and then when FTX happened withdrawals saw the market shrink by 70% in a few months. If you were to set the parameters to 20% in that scenario you’d end up suddenly with 90% of the market being less liquid tokens.

I think until there are secondary markets for these tokens it will be hard to get the parameters right.

3 Likes

Huge +1 to this. I’m quite familiar with the model that Centrifuge uses, where pool tokens represent investment in a pool, and I’ve been looking forward to the day when they are composable with the rest of DeFi. With the technical + regulatory infrastructure we now have, both on Centrifuge + with working with Coinbase and Base, I think we’re ready.

3 Likes

Overall seems like a great proposal. I am in favor of this :+1:

1 Like

Big fan of this proposal! Two points:

  • I agree to launch the lending market on an L2 as it is much cheaper to use than mainnet. Base is the perfect candidate here given the hype around the ecosystem and their close ties to Circle (USDC). Currently, Coinbase withdrawals to Base are free of charge and this could be a further catalyst to get more users on Base vs other L2s.

  • Regarding the choice of the protocol to use as the lending market, I think the best choice here is Morpho given its flexibility and low overhead. Maker recently used Morpho to create a $1B lending infrastructure for USDe (Ethena). Morpho is currently deployed only on mainnet, so need to investigate with their team if a Base deployment is possible and what would be the timeline.

Does the integration with Coinbase attestations mean that Coinbase users will not have to go through KYC again before using Centrifuge?

6 Likes

Thank you for such a detailed and high-quality proposal. I really like the idea, and if a real-world asset (RWA) market is to be established, it makes sense for Centrifuge to be the one to implement it.

“We propose to deploy a lending market for Centrifuge RWA assets built on Base. We see Base as an ideal blockchain to build on because of Coinbase attestations that allow for very easy access control to the market.”

Given the recent partnerships, the Base blockchain seems like a reasonable choice. However, I find the CELO blockchain more appealing for RWA projects considering our partners and the compatibility of our visions. Shouldn’t we consider launching on CELO? Perhaps a multi-chain structure could be beneficial.

For example:

  • Arbitrum
  • VeChain
  • Avalanche

These platforms are not yet our partners, but initiating partnerships could be a reason in itself.

In my view, the risk aspect is the most crucial. Subsequently, legal issues could expose us to many challenges.

How do you plan to handle scenarios of extreme market volatility or liquidity crises?

more user-friendly onboarding (AAVE RWA)

I’d appreciate if you elaborate on this a bit more

  • How will the proposed fast and safe onboarding process using Coinbase attestations align with privacy concerns and user control over personal data?
  • Are there alternative onboarding processes being considered to accommodate users not on Coinbase?

I’m not a tech guy but am curious about the Interest Rate Strategy.

  • How will the target interest rates be determined?
  • What mechanisms will be in place to adjust these rates in response to market conditions?

Regarding Scalability and Future Growth, I want to address that with a starting budget of over $400k, we need to plan the end as well as the beginning. Unanticipated questions can later create pain points at irreversible stages.

  • How scalable is the proposed infrastructure?
  • What are the plans for expanding the types of collateral and services offered as the market matures?

Given that the legal structure involves a special purpose vehicle, how will the DAO ensure compliance with varying international financial regulations, particularly concerning KYC and AML standards? We certainly don’t want to get the FBI on our back :slight_smile:

1 Like

The reason why we proposed to do this on Base is because of Coinbase attestations which are a key part in how this market could be launched given there needs to be a mechanism to permission the market.

How do you plan to handle scenarios of extreme market volatility or liquidity crises?

If we limit the market to only have underlying assets with a very short maturity the risk of market illiquidity is greatly reduced which is how I would suggest we launch it.

I got it, makes sense to do it with Base.

If we limit the market to only have underlying assets with a very short maturity the risk of market illiquidity is greatly reduced which is how I would suggest we launch it.

In that case you mean as a very short maturity is less than a week, right?

Hello all,

This proposal presents a well-thought-out plan to create an RWA-focused lending market on Base, addressing key issues faced by the previous Aave RWA Market. The use cases and benefits for both borrowers and lenders are clearly outlined, and the improvements over the previous design are substantial.

The choice to deploy on Base and leverage Coinbase attestations for user onboarding is a smart move, as it will help attract a large user base while maintaining necessary security measures. The legal structure and risk management processes are crucial aspects of the project, and it’s good to see that they are being given due consideration.

The technical design seems sound, with the selection of an open-source lending market, permissioning using wrapped USDC, and the creation of a redemption wrapper for pool tokens. The initial funding request is reasonable, given the large scope of the project and the various components involved.

My main point of feedback is,

Lending/borrow markets are great in the good times, users get to use the service, Centrifuge collects the fees. But the tech stack needs to be able to survive at every price, including outliner events. Which are common place in crypto markets.

I would love further information regarding the back end of the liquidation process. Generally when we’ve seen borrow markets fail. This is the reason.

  • Can you explain the process of liquidating a user position that no longer maintains margin compliance?

  • Can anyone become a liquidator of these assets? Myself and i’m sure many others would be willing to buy on chain T-bills at a discount to the fair tradFi market price. Will this be possible during liquidation events ?

Thank you for your time. Proposal is very exciting.

Regards,

TheOyster / OnlyDeFiGuy

2 Likes

I like this proposal, once announce all token holders will be pretty happy with it as it gives them the chance to borrow against their assets, offering immediate liquidity and the option for further leverage. We saw it more and more in the RWA industry, it’s a growing demand. Everybody need and want more adaptable financial tools in the market. It also proposes an innovative answer to liquidity challenges faced by RWA token holders, boosting the appeal and functionality of Centrifuge pools. I am 100% in favor!

Interesting aspect. +1

This is the main point since Satoshi. Processing yet, however people are still optimizing for the %1 in the ecosystem. Therefore we need more real life use cases, and RWA narrative is perfect for onboarding web2 people to here.

Hey all,

This is a very exciting proposal. I share the conclusion that one of the main use cases of RWA onchain is the ability to collaterize them on money markets, similar to the repo market today.

Given the bespoke nature of Centrifuge Assets requiring custom lending pools and the leaning on working with Coinbase attestations on Base, I’d like to suggest the utilization of Morpho Blue for this RWA lending market:

  • Morpho Blue is a custom lending primitive with permissionless creation of isolated lending markets for any ERC-20, very well suited to Centrifuge custom needs
  • Morpho Blue is immutable and doesn’t charge protocol fees
  • Morpho Blue’s isolated lending pool design enables higher capital utilization and a “per-market risk profile.”
  • Morpho just proposed to deploy on Base (MIP - Morpho DAO as owner of Morpho Blue on Base - Governance - Morpho Governance Forum
4 Likes

My question is who is lending to whom? Is this simply a crossing market where borrowers post requests? Who does the credit analyis? Is there a one size fits all credit structure or are credit agreements tailored to the situation? If collateral to RWA is put in place, how is the lien on the collateral perfected? Who is responsible for enforcement or restructuring in the event of default? Who services the loans? What is the loan origination process? How is credit quality assessed? What is the mechansim to respond when there is market dislocation? None of these questions are to say this is a bad idea. On the contrary, there is a fundamentally good idea here. I only pose questions to fuel the product development process. If you need any assistance, feel free to reach out.

With the news that Morpho just shared I believe we should choose their stack to build on. Thank you @0xloth_Morpho for adding some context. I have updated the CP and would move to propose this shortly with Morpho added.

Thank you everyone who shared feedback and questions on the proposal.

@theoyster You posed two good questions:

Can you explain the process of liquidating a user position that no longer maintains margin compliance?

Key to this market at least in the beginning should be to ensure the underlying liquidity of the collateral assets remains high. LTF and other similar products typically take 12-48hrs to redeem at which in an extreme market environment could lead to lenders not being able to withdraw until liquidators are able to redeem enough underlying tokens to create liquidity. I think that risk though is manageable given the short period.

Can anyone become a liquidator of these assets? Myself and i’m sure many others would be willing to buy on chain T-bills at a discount to the fair tradFi market price. Will this be possible during liquidation events ?

If there is indeed an increased withdrawal demand from the market, this is what would happen and tokens would be sold off at a discount. However LTF and the underlying tokens are permissioned and require onboarding with the issuer to be able to hold and thus participate in any onchain liquidations.

CP103 has now merged into the proposal repository on GitHub and is now final.

The OpenSquare snapshot has been published and will go live 21st May 18:00 CET and will be open until 28th May 18:00 CET (exactly 7 days).

:ballot_box: Please vote here: CP103: Launch an RWA Lending Market

1 Like

Good day
The snapshot has passed with following results:

image

Thanks everyone for participating.

CP103 is now live as Referendum 57 for all CFG token holders to vote on.

If passed, this will approve the treasury spending of 606,000 CFG to be transferred to the beneficiary wallet in the next spending period.

:ballot_box: Please vote on the referendum here:

Referendum 57 for approving the treasury proposal for CP103 passed unanimously.

The requested funds (606,000 CFG) will be paid out to the beneficiary wallet in the next spending period, which starts at block #5,745,600 (~ 4.00am CET 22nd June).

Thank you for voting everyone!