Centrifuge: An Update on RWA in DeFi

Dear community,

To everyone who I have had the pleasure of meeting in Paris, it’s been amazing to finally put faces to the people who feel like age old colleagues in this DAO. I can’t wait to see what’s going to happen in the time until we meet again.

I wanted to share an update on what Centrifuge has been working on and what our priorities are going forward and seek comments from MKR holders & community members. Our mission of bringing RWAs to DeFi always included Maker as a partner and a lot of what we built was heavily influenced through our work with Maker.

RWA in DeFi: What does this mean?

We all know how antiquated, inefficient, and oftentimes just completely broken today’s financial system is. This was a big part of my motivation to dive into crypto and probably for a lot of people on this forum as well. We believe the underlying protocols that power DeFi today are not just for cryptonative assets but in the long run will offer benefits to assets that come from the real world. DeFi will replace the banking infrastructure not just for novel virtual goods NFTs and DeFi governance tokens but also for “legacy” assets like real estate or trade finance. What does that mean? People who today buy and sell these assets on Wall Street will look to DeFi as the better way to create, trade and settle these assets.

We focus on one particular part of the financial infrastructure that we are trying to bring into DeFi and pioneer this with: securitized debt

Bringing it into DeFi will scale DeFi to the trillions of dollars and potentially millions of new users to Dai in the long run. It means not just printing a ton of DAI from some off chain collateral to fix the peg but also to get both investors and borrowers to start doing more and more with DAI (I’ve personally helped more than a dozen crypto newbies buy their first DAI to invest in RWA). Why is this such an exciting thing? Because we are actually building technology that helps both on the supply side and the demand side of MakerDAO: investors will want to buy and use DAI for investing in these asset classes and borrowers will bring in attractive collateral that can be used to mint DAI.

Scale fast but smart

We have been big proponents of iterating quickly: starting out small to keep risks low, getting experience and then incorporating the learnings into improved processes. By doing so we were able to launch New Silver and have already been making the first improvements with the independent director. In this post I want to go into what else we are working on already and where we believe the future of RWAs in DeFi is going to.


New Silver is now in the top 15 of of collateral types by debt

Here is how we believe we can grow RWAs in DeFi (stabilize Maker with uncorrelated collateral, reduce USDC in the PSM, generate interest revenue for Maker and increase the usage of DAI across all of DeFi).

Theme 1: Improving our Smart Contracts and process around it

We have in total had over 32 weeks of security auditing on the Tinlake protocol by two separate auditing firms (Least Authority and DappHub, a team that includes a lot of the original MakerDAO protocol engineers). DappHub has published the audit of the lender side (incl. MIP22) recently here: Preliminary Tinlake Audit Report: They are performing a second audit of the borrower side (which was originally audited by Least Authority). We will publish this report in September.

We are also continuously improving the spell crafting process together with the Protocol Engineering Core Unit and helped prepare the spells using the most recent patterns for our Executive two weeks ago. This makes it a very simple and easy process for the PE team to onboard these collateral types quickly and safely.

We are also improving the operational security for our AOs and are working on a dedicated multisig that significantly reduces the risk of loss of funds for AOs due to incorrect use of their wallets. A technical spec can be found here: Multisig Tinlake Proxy - HackMD

Theme 2: Grow our pools by helping our issuers scale

When we started working on all this, we were trying to convince traditional businesses to bet on DeFi and Maker when no one else has. Naturally this was a lot harder and I’m truly grateful to the early AOs that believed in the vision of Maker backed by RWAs and took a risk in doing so. Some of them spent over a year on their collateral governance process.

Their courage to innovate with all of us helps convince the more conservative and more established borrowers to experiment and ultimately become users of Tinlake and collateral types on Maker. Our goal is to scale the average size of our pools continuously and have the first pools reach $100M+ in value by end of the year. Scaling these pools to ever larger sizes will also involve bringing in institutional debt investors to scale out Tinlake and ensure Maker is not the only major investor in these pools. This is an important milestone for both Maker and Centrifuge as we will start to go from experimenting at small scale to adding assets to Maker that will bring significant revenue to Maker and make a dent USDC used to back DAI.

Theme 3: Improve the legal setup

We have made first improvements with installing independent directors for each pool to reduce the risk for investors in the pool. We have also been doing a lot of work in the background on researching how a trust structure could be used to further strengthen the recourse and legal standing MakerDAO has in the pool. We are excited to see the RWF CU proceeding with setting up their own trust and we are working with our counsel to find a way to give this trust control over the collateral that is used to borrow DAI from the protocol.

Theme 4: RWA in DeFi Beyond Maker

Maker relies on there being both a supply of DAI and a demand of DAI. Our pools are settled in DAI and make use of DAI for every transaction. By borrowing from Maker we create more supply but we are also working hard on bringing these assets into the hand of more people and protocols such as Aave. You could see this as competition to Maker but I would argue it’s the opposite. AOs wanting to borrow from Aave users will create demand for DAI and use DAI, it will add uncorrelated demand to DAI and it will increase the yield people can earn by lending DAI which creates DAI demand. Maker using RWA as collateral in a vacuum makes it’s peg more stable and more secure but it doesn’t scale the demand side. An overall strategy for establishing RWAs in DeFi does.

Of course we can’t do that without the help of many in the Maker community including the RWF CU, the PE CU but also others who help make this hard topic more digestible, ask difficult questions and scrutinize what the AOs do.

I hope this gives the Maker community an insight into our strategy and current areas of focus and appreciate your feedback, ideas and collaboration on these topics whether that is in the replies here or by reaching out ot us.

13 Likes

That’s wonderful news. I will start drafting a proposal to spin up a trust for this, since we’ll want something with fiduciaries. Or if anyone else has a tightly bound legal structure ready to go, have at it. I wonder if we can shoehorn this into the recently set up trust for 6s or if that’s only got to be for 6s? Really great idea and I’m sure a burden off of Centrifuge.

Agreed! Especially if a secondary market develops on Aave and it’s denominated in DAI. Really supportive of this effort. I hope it goes well, and if there’s anything substantive I or others here can do, do not hesitate to reach out.

Thank you so much for the update :+1:

1 Like

You can’t use the 6s Trust for anything besides 6s. But, all of the documentation is public so if you wanted to engage a regulated Delaware Trustee to custody and manage DROP equity on behalf of the DAO in a new Trust, I don’t see why you couldn’t.

5 Likes

Thanks for the offer and support. The RWF CU already has a MIP in the RFC phase for establishing a trust: MIP58. We are focusing our work on this for now. I believe this will be the best solution forward and is built on the early work of both 6S and the RWF Core Unit.

1 Like

So that’s actually not a trust, but a foundation. Foundations operate more like a non-profit and are run by a board or similar group of people. It has discretion to operate however it wants, within limits. I’m not super familiar with Cayman foundations specifically, but what is described in that MIP seems to follow that non-fiduciary, corporate structure.

An actual trust, however, is a legally bound, fiduciary party who holds assets for the benefit of a third party. It’s got far more protections built into it in that regard.

Since what we need is more of a parking spot with the ability to pass KYC, and the legal entity has no need to actively manage the wallet, I think we would all agree there’s no reason for it to be in an LLC owned by a foundation, where it is much less protected by legal safeguards.

DM me the contact for your counsel, and I can loop in a trust attorney. It should take about 6 weeks to draft trust documents at the moment — trusts in favorable jurisdictions are in high demand.

That way we can place the DROP under a fiduciary in an entity that can’t accidentally mismanage or use them.

So glad you had this idea for a trust! Really wonderful!

@PaperImperium, you might have the wrong understanding of what a Cayman Foundation is. I’m clueless on US Foundation, however, you can read this nice article on Foundation and DAOs from Ogier (which is written by our counsels).

the foundation company is a perfect fit as it acts like a legal trust

You can also read this one which is broader.

I wonder where you get this impression from the MIP58 and the Articles of Incorporation we posted. A significant amount of work was done by both @christiancdpetersen (which addressed the concern here) and Ogier (Tier 1 in Trust counsel).

If your comment is about the LLC, I would say the same applies. The LLC is member-managed by the Foundation so it inherits its safety.

We started with a Trust idea and ended up with a Foundation, this was not a bug but an improvement. I understand the attachment to the word trust, but I think rationality should prevail (until proven otherwise), especially for a trustless protocol. If there is a concrete argument for a trust structure, we can go back to the drawing board, but so far there is none.

Regarding the Centrifuge assets, the harder part is the smart contract side. Having a US-based trust is probably not wise as some (Nebula Capital) will be in the EU (and others in Asia). Doesn’t seem good to risk the US regulation risk when not needed.

I don’t see any major limitations on the actions that can be taken by the foundation. Can we insert an irrevocable set of bylaws that the foundation is only allowed to perform pre-defined activities?

If for some reason this structure does not function as intended or is captured by individuals or some other unforeseen circumstance, there need to be strong limitations on what it can actually do.

This looks to me like we are designing a single point of failure for already tens of millions of approved financing. We need diversity of structure and jurisdiction so that we survive a mistake in creating a brand new type of structure.

If our past experience with RWA to date is any indication, we will not arrive at the optimal design on the first try, and should do everything possible to decentralize control of both cash flows and legal claims on collateral.

Given our governance process, we also need something that requires as little active management from the DAO as possible. We need a structure that is safe from capture but also does not rely upon our Byzantine governance process to function properly.

3 Likes

It seems to me that it is the case. We can obviously narrow it more, but I’m not sure of the objective. The directors are quite limited by the need of a DAO Resolution (see Schedule A).

The Operating Agreement of the subsidiaries will be what matters, those should be narrow to the specific task needed.

So the decision-makers of the Cayman Foundation (Directors) are by design not willing to do much (as trustee-like they don’t have business taking decisions), are not allowed of much (objects of the Foundation and Schedule A), and neither have the resources to do much (as no cash is accessible at the Foundation level).

I don’t have a view on that one. But with the combination of committees and authorized signers, you can delegate as much as you want (and change your mind over time). This will need some iterations. As a DAO we don’t want to do anything but, at the same time, we don’t want to trust third parties either (we trust the trustees but they will not do the decision making). Maybe a RWA Management CU is needed, I don’t know.

For holding the DROP, the amount of recurring work will be exactly 0 I guess. Everything is on-chain. But suing the AO (or any other remedy in case of an issue) is something I would say the DAO should trigger. But we can decide that this action is to be decided by @PaperImperium .

1 Like

Appreciate the discourse on the trust and foundation, but wanted to chime in and see if we could center discussion a bit more on the thread itself, RWA in DeFi. There’s no doubt that the legal structure around how we craft these partnerships is super important, but I feel some of this discussion might be better suited on a different thread, perhaps MIP58: RWA Foundations itself.

5 Likes