On-Chain Real Estate with Sanjay Raghavan

In this episode, Sanjay Raghavan from Roofstock onChain joins us to discuss bringing real-world assets (RWAs) onto the blockchain. Sanjay explains how Roofstock enables buying and selling of tokenized houses (real ones, not digital), discusses the legal considerations of bringing real estate, and other RWAs, on-chain, and digs into the securities debate surrounding cryptoassets. 

Transcript

Mark:
So I appreciate you coming on. I'd love to start just by brainstorming a few ideas for us to talk about today.

Sanjay Raghavan:
Absolutely.

Mark:
So I think an obvious one is real world assets and real estate in particular. But before I bias you, because of course you are on the Roofstock team, I'd love to hear if there's anything else that you've been thinking a lot about recently that you'd love to write an op-ed on or that you have a contrarian take on, and then we can pick one to dive in.

Sanjay Raghavan:
Certainly. I've actually been spending time on a lot of the securities-related issues, specifically in Web3. And so yeah, that's a topic we can definitely talk about. And I think it also intersects with RWA and what can be done with real estate, and also outside of real estate, with DeFi and stuff like that.

Mark:
That's a critically important thing for everyone to understand if they're trying to think about the future of crypto and DeFi. And so maybe before we dig into this, could you give the audience a little bit of context for what makes you such a credible expert and guide on this topic?

Sanjay Raghavan:
So my background before all of this is I have about 25 years experience, mostly in financial services, in product operations, banking and finance. And I have an MBA from the Wharton School, the University of Pennsylvania. I spent a few years as an entrepreneur, built a lot of financial services-focused products, and also spent about five to six years as a lower middle market investment banker trying to structure transactions that involve either cashflow-oriented assets or fixed income type assets that could be done as off balance sheet.
And so a lot of those are very relevant to the conversation pertaining to DeFi because essentially, what we're trying to do is create different structured products on DeFi that pertain to various real world assets. And having spent time in banking and finance, those are highly regulated. I had securities sales and structuring licenses, and so I, over time, just spent a lot of time pertaining to exempt security such as Reg D, Reg A+ and so on.
And I follow that world very closely with respect to crypto because especially in today's market when there's a lot of ambiguity about what is a security and what is not a security and different views from the SEC and CFTC on these topics, as well as broadly the sort of unfortunate turn of events where we're looking at regulation through enforcement in a lot of cases pertaining to crypto. So I try to be vocal about my thoughts and opinions on that matter and just wherever the SEC may have a reasonable argument, I would step in and say why that's reasonable. But where I feel that it's particularly not reasonable, I'm also equally happy to raise my voice about that as well.

Mark:
Great. Well, this is the place to be vocal, so I'm glad you're here.
And two other questions about your background. First, just a fun call-out that you're also an advisor to Pudgy Penguins, which is just a fun side project, it sounds like, for you.

Sanjay Raghavan:
Exactly. With NFTs, my day job also involves NFTs because at Roofstock, we are using ERC-721s to simplify how single-family central properties can be bought and sold. But there's this absolutely fun element to NFTs and through Pudgy, I get to experience that. I attend their IRL events as well and they have really good parties.
It's one of those few NFT projects where it's not something that only a few people can identify as POAPs for themselves or as PFPs for themselves, but it's actually a PFP that the entire family can identify with and enjoy, especially if you've seen Happy Feet and all these other Disney movies, and kids have an affinity to penguins. And so my kids pick out my Pudgy and they draw it.

Mark:
I always loved penguins. I had no stuffed bears as a child. I just had stuffed penguins, so I'm with you.

Sanjay Raghavan:
Exactly. That makes it a lot of fun.

Mark:
Can you explain PFP for the audience?

Sanjay Raghavan:
So yeah, it's basically proof of picture. I mean-

Mark:
It's a profile picture.

Sanjay Raghavan:
Or the profile picture, but leaving aside technically what it means, it's a way for the Web3 community to feel like they're part of something by associating the picture of that, whichever collection that they bough into, associating their identity through that picture and feeling like they're part of a community that has shared values and can discuss things in common.

Mark:
It makes sense. And what's interesting there is that's actually an NFT, right, and it's an unchained product?

Sanjay Raghavan:
That's correct, yeah.

Mark:
And it sounds like what you're doing at Roofstock is actually using the same technology, even though people associate NFTs with art and on-chain things, collectibles, they're also probably the foundational technology that's going to be used to bring a lot of real world assets, or RWAs, on-chain. And so NFTs will mean a lot more in the future.
Which brings me to my second question, which is can you briefly just explain what Roofstock does so that the audience has that context as well, and then that can lead into our conversation about real world assets.

Sanjay Raghavan:
Yeah, absolutely. So Roofstock is a venture-funded company that's been around since 2015 focusing on the single-family rental asset class. And those are essentially detached homes or duplexes or triplexes. So they're small, they're not like multifamily or apartment buildings that have 12, 24, 30 units. But these are anywhere from one to four units would classify as a single family. Most of those are, of course, detached, but you do have some duplexes and such.
But this asset class, even though you might not think of it traditionally as a rental asset class, because when you think about rentals, you typically think about either a garden apartment complex that's got 300 units or a high-rise in the city that has 10 floors or 20 floors. Actually, the single-family rental asset class is larger than apartment buildings in the US. We have about 21, 22 million units that are used as rental units and these, in aggregate, represent about $4 trillion of GMC. So it's a very large investible asset class.
And historically, about 89% of these properties are owned by very small mom-and-pop investors that might have less than 10 units. And there's really, maybe when you look at who has more than 2,000 of these units, it's probably one to 2% of institutional investors that make up more than 2,000 units. And so the reason-

Mark:
So this is like most people is your uncle or your doctor, they own a rental property or two.

Sanjay Raghavan:
Exactly, exactly.

Mark:
And they're homes on side, they manage them and it's some side income.

Sanjay Raghavan:
That's right. And historically, the challenge with this asset class, even though it's appealing as a diversification play, it's non-correlated to public markets, you can make a rental yield that goes up a little bit over year-over-year so it acts like a little bit of an inflation hedge. When you hold the asset for 10-plus years, generally there is appreciation associated with that, absent any other kind of systemic issues.
So generally, it's a good asset class to have, but the problem with it has been, both for the buyer and seller, there's a lot of friction. And what I mean by that is typically when you're selling one of these assets, the leakage from the system can be about 7 to 7.5% of the value of the house. And I'm talking generally about starter homes in the US and [inaudible 00:08:05].

Mark:
As in the transaction costs, when you say leakage?

Sanjay Raghavan:
Yeah, the transaction costs associated with it.

Mark:
This is like the escrow agent and the costs of sale and processing everything.

Sanjay Raghavan:
There's a dozen-plus intermediaries. And when you look at, typically, most people probably don't even realize this, but when you buy a house, eventually you get a HUD financing statement and there'll be like 50 line items in that and you're like, "Who are these people and why are they all taking a fee?"

Mark:
Yeah. I'm a recent first-time homeowner and I noticed that as well with confusion.

Sanjay Raghavan:
Exactly. So that's from the seller's perspective, most of the costs are born by the seller, but from the buyer's perspective, the biggest challenge is financing. And the reason for that is because this particular asset has sort of this duality. One, you can buy it as an owner occupant and use it as your sort of primary residence, or you can buy it as an investor and use it for a rental purpose.
But the entire TradFi lending infrastructure has been built for one of these use cases. And so Fannie Mae allows you to have a few rental properties, no more than 10, but you can get up to 10 rental properties and get it financed through a Fannie Mae product. But that means you go through your personal credit underwriting each time. And so even though you're going to be using the property for a commercial purpose, for generating rental income out of it, they're actually looking at your income, your taxes, your bank accounts and everything else to make a financing decision on this loan. That makes it extremely complicated. A lot of people I talk to, if they have the cash, they just buy it as a full-cash purchase because the financing aspect of it is so painful to go through.
And then generally, you can do one or two or three, but each time you add onto it, you're trying to get your combined assets, combined liabilities, and report everything and get all the financials pertaining to all of this, and it just gets complicated. Plus, you are buying all these assets with the title registered to your name, which means that as you have more and more of these, you also start taking on some personal liability if there's a slip and fall or something. Of course you have insurance on the properties and they're supposed to cover that, but that won't stop somebody from suing the owner of the asset as well. You can litigate, probably you can litigate your way through that process, but you still have to go through that painful process of litigation.
So most people will eventually want to essentially limit their personal liability by throwing these assets into an LLC structure, which a limited liability company, by definition, is supposed to provide you that liability protection. However, when you do that, the Fannie Mae program is no longer accessible to you. So now you have to find a private non-bank lender who's going to charge you an arm and leg and probably want a personal guarantee and other kinds of cross collateralization. So the financing for the buyer's incredibly hard and in the TradFi world.
So when you take these two problems and see what can Web3 do to solve these two biggest pain points for this $4 trillion investible asset class, is how can you make the purchase and sale really simple, fast, transparent, cheaper and make it an instantaneous one-click buy? And then also, how do you integrate commercial lending through asset-based financing solutions on DeFi that allow you to add leverage? Almost like when you need a... Just like if you had an ape or a pumpkin and you decided you need a 30% loan against it, based on the flow price of those assets, you can pledge it to any number of NFT lending solutions and immediately get USDC. Lending on these should be similar or at least get very close to that experience.
So that's the goal we have for our Web3 product is make it fast, cheap, and simple to sell and then make it really easy to get leverage.

Mark:
It makes sense. Okay, great. And this is really important because real estate is one of the largest asset classes in the world. I mean, it is enormous and so many, mortgage-backed securities and everything, derives from it. And so it's just an enormous asset class. So when we talk about bringing things on-chain, real estate is one of the obvious candidates.
The other interesting thing about it is that in contrast to, let's say, equities or fixed income directly, these are all unique. They're not fungible, which means they have a lot of different idiosyncrasies if you're bringing them on-chain, representing them and transacting them and financing them.
What portion of your business is Web3-oriented as opposed to targeting the traditional market? And then how do you think about bringing these assets on-chain? Can we start in the weeds with how do you represent as an NFT? How do you get it actually financed on-chain, and does it really make sense to do that on a public blockchain or should it be on its own?

Sanjay Raghavan:
Fantastic questions. So Roofstock itself obviously set up as a large Web2 proptech company in this space. What we've built is an institutional class platform for hedge funds and private equity and other institutional investors to come and say, "I want to deploy $100 million or $500 million into this asset class. Can you help me source, acquire, renovate, manage, and help me run this program?" And we're basically like a SaaS platform for making that happen for these types of investors.
But that means that we've spent a lot of time and energy creating data and analytics that we can look at the entire country and break it down into major metros and say, we know, for example, which institutional investors are buying in, not just who's buying in Atlanta, but we can narrow down into Atlanta and say, "This investor is buying properties in this part of Atlanta or these suburbs of Atlanta and this other investor's been buying here." And that kind of data is useful when somebody wants to come and decide, "Hey, I'd like to learn more about the Atlanta market. Where should I be? Where shouldn't I be? Who else is purchasing here? Et cetera, et cetera."
So that's seven years of engineering, big data, AI, ML, whatever to get to that position. And we can leverage all of that as part of Roofstock to basically offer the same to the investor who wants to buy one, five, 10 rental properties through NFTs in our Web3 marketplace.
The Web3 division itself is really tiny. It's less than 1% of the overall Roofstock. And we just... I'm sorry. You're on mute again.

Mark:
So you have a platform and you're taking that platform, which exists in the kind of TradFi economy, and you're using it to launch these assets on-chain.

Sanjay Raghavan:
That's exactly right.

Mark:
You have assets, have liquidity in that context. Positions you very well to actually do.

Sanjay Raghavan:
Exactly, exactly.

Mark:
Awesome. Totally makes sense.

Sanjay Raghavan:
And so now bringing this conversation into Web3, in essence, as you know, there's like 3,600 or so counties in the US and when you buy or sell properties, typically there's a process of recording that with the local county registrar and the deed gets transferred from the seller to the buyer. And that process, at least in the short term, we don't see a viable path for that all to come on-chain where the provenance or the title can be represented on-chain and you can look at all the leans and encumbrances there. It might happen over time and we certainly expect or hope it does, but for now, the best available compliance solution we have available to us is to see how can you abstract this entire process away? Because if you want to do something close to Web3, you cannot really straddle between Web2 and Web3. If you say, "You can buy the NFT here and then give me a week to go back and finish up all the paperwork with the local county," that's really not a Web3 solution.
So we spent a lot of time trying to engineer a solution with the single goal of we want to solve these two problems we talked about earlier, but make it as close to Web3 from an experience perspective. And so the solution we came up with was to take that single property, go through a traditional real estate close, but close it into this bespoke single purpose limited liability company structure that we have created. And when you close it into that specific structure, it has the documents pertaining to that LLC are written in a certain way that essentially refer to the Ethereum blockchain for their ownership.

Mark:
So basically, you have an LLC and it says the unit holder of this LLC is the wallet address which controls this specific NFT.

Sanjay Raghavan:
That's exactly right. And there's two things, two advantages to that. One is of course if you separately had an LLC and then separately there was an NFT, but there was no way to bridge the on-chain asset with the off-chain asset, then it's hard to have any enforceability because the whole idea is this transaction should be trustless and it should be legally enforceable.
So if I just said, "Hey, there's this NFT. Why don't you buy it? And then after you've bought it, let's figure out all the paperwork," Then what happens is there is a moment in time you have to trust the platform that once you buy that NFT, you just spent $200,000 buying this JPEG of a house, you're actually going to get legal ownership of that house eventually.

Mark:
Yeah. It doesn't actually do anything helpful or new, right?

Sanjay Raghavan:
Exactly.

Mark:
And it makes the financing aspect tougher and the financing is the biggest driver of [inaudible 00:17:57] do these deals.

Sanjay Raghavan:
That's exactly right. So the only real way to do it in a legally compliant manner is to... Because the only legal document pertaining to that LLC is the operating agreement of the LLC. So that document needs to very clearly outline that, exactly as you said, that the ownership of this LLC will be determined by that wallet on the Ethereum blockchain that has this NFT associated with this smart contract.

Mark:
So that makes sense. But this brings up one of the issues with real world assets brought on-chain, which is not only do you have to structure such that the on-chain representation has legal tie to the off-chain asset, but also, you introduce a lot of frictions that are required by law. And because it's a claim on an underlying asset, you have to respect those in order for that claim to be valid.
And so a few that come to mind are like, well, now you're the sole owner of an LLC, you're the beneficial owner, there's a KYC obligation.

Sanjay Raghavan:
That's right, that's exactly right.

Mark:
There's all these things. And if it's a rental property and it's cash flowing, then that's a security, right?

Sanjay Raghavan:
No, no. So that's where I think we can get into the nuances of what makes something a security or not a security. And these are great points and let's cover them one by one, right?

Mark:
Okay. Let's dig in because I think it's great. Now we understand the asset class, we understand why it's important, we understand how you're representing it on-chain, let's go into these frictions.

Sanjay Raghavan:
Absolutely.

Mark:
And then we can pull out into why and when real world assets can actually be demanded and successful on-chain.

Sanjay Raghavan:
Absolutely. So the first question is obviously there's an LLC involved here. There's a real property involved here. To your point, these are cash flowing assets and should not be used by either sanctioned parties or by people using these monies for money laundering or drug trafficking, et cetera. So KYC is a legal requirement for these types of things.
So the question then becomes, if you want to use the blockchain and you need to KYC the buyer, there's a couple of different choices. One, you can just run the whole thing on a private blockchain and just say, "Everybody that's participating in this has already been approved to be part of this walled garden solution." And a lot of the banks that start foraying into Web3 sort of take that tack because they don't want to be on the public blockchain. They want keep everything contained. The JP Morgan product, for example, is all done like that.
We wanted, because what we're dealing with here are SFRs and we wanted to use the NFT marketplaces as a distribution option for that and not just create something... Web2 companies typically think of solving these problems. When you need a marketplace, you build one. Web3 companies try to see what marketplaces exist that I can leverage? So in our case, we wanted to be closer to the Ethereum main net world and participate in that ecosystem. And so what we decided to do was create a permissioned architecture on top of the standard ERC-721 contract, the ERC-721B.

Mark:
Okay. I just want to flag there for a moment because that's a really interesting point. You can create your own blockchain permission in the chain itself. That's the most compliant option and that is probably the way to use the technology and "on-chain assets" to appeal to traditional institutions, right? Because from their perspective, they don't even need to know it's really a blockchain underneath the hood. It's just technology that's useful.
And a lot of people in traditional markets, they kind of say, "Oh, well I don't know how I feel about crypto, but blockchain's an innovative technology." And sure, that's true, but what you're saying is the thing that loses is composability with the rest of a public blockchain so that you can plug into marketplaces, whether they be OpenSea or someone else, you can plug into asset-backed lending protocols. This composability aspect is actually the really interesting thing about bringing assets on-chain as opposed to just using it in a chain that is a walled garden itself. That makes sense.

Sanjay Raghavan:
And look, I think there's probably... We want both use cases to flourish so that eventually people get more comfortable with the idea. If they do something with JP Morgan in a walled garden, maybe two years from now, they might be open to doing it on Ethereum with somebody else, right?

Mark:
Sure. Not mutually exclusive.

Sanjay Raghavan:
Yeah, they're not mutually exclusive.

Mark:
One can be a gateway to the other for traditional institutions.

Sanjay Raghavan:
Exactly, exactly. But for us, that composability was very important because again, we want to be able to leverage all the tooling and infrastructure that's already been built out and not try to, if we had to replicate all of that, that's a large army of engineers and a couple of years sitting in a dark room.

Mark:
Yes. You can just layer on top of the chain that's there. You don't even have to spin up a server.

Sanjay Raghavan:
Exactly, that's right. So from our perspective, that meant that how do we create a permissioned architecture on top of the Ethereum blockchain? And some of the solutions we looked at, there are different people who have different whitelisting-type solutions, but the easiest one for us was let's get everybody that's interested in buying a house, mint a soul-bound token, which is also an ERC-721. They come to our site, mint the SBT-

Mark:
And a soul-bound token is a token that you can't transfer to someone else.

Sanjay Raghavan:
Exactly. So the transfer function is disabled. If you take a standard ERC-721 contract and disable the transfer function, then it basically becomes-

Mark:
And sorry, one more flag is an ERC-721 is an NFT.

Sanjay Raghavan:
It's the NFT standard.

Mark:
When people say an NFT is a unique asset, ERC-721 is just a developer term for the most common, unique non-fungible token.

Sanjay Raghavan:
That's exactly right. So once you have that SBT, or sold-bound token, SBT in the prospective buyer's wallet, then you can actually curate all kinds of useful use cases through that. So you can use that as a membership token or as a loyalty rewards program and say, "Every time you do something with us, we can go and change the metadata of your SBT to say, 'What is your status with us?'" So if you want to have silver, gold, platinum programs and you have some SBTs have an attribute that says, "Okay, what is your status with us?" And you can mark it as silver, and when they get to gold, you can make it gold. So there's lots of interesting things you can do with it.
The thing we're doing right now is once they mint this SBT, they get a link to essentially get KYC'd. And that is a simple process. You either get it on an SMS or an email, click that link, it asks, our partner that does this, Jumio, will ask you for a front and back copy of your driver's license and a selfie. And then within a minute, they come back with a yes or no. And mostly for US, it's pretty straightforward, takes less than a minute. And when it comes back as your KYC, then the soul-bond token gets updated with a KYC flag.
And once you have that flag, then it makes you eligible to go to any NFT marketplace where one of these properties is listed. And of course, you're connecting your wallet anonymously to those marketplaces. So that's an important distinction. With KYC, the owner of the wallet, but the marketplace is not aware of who you are because you're basically just anonymously connecting your wallet there and going and completing a purchase. Our 721 smart contract or the NFT smart contract simply looks to see if the buyer's wallet has the SBT and the KYC flag has been turned out.

Mark:
I see. So if you want to buy one of these things, you first go get that KYC flag and then you can purchase the item and that it essentially unlocks and can become transferrable instead of being soul-bound.

Sanjay Raghavan:
No, the token itself, that token remains soul-bound, but the NFT that represents the house can only be transferred to a wallet that has the louse-bound.

Mark:
I see. The soul-bound token is the KYC flag.

Sanjay Raghavan:
Yeah, yeah.

Mark:
I see, okay.

Sanjay Raghavan:
I mean, it has the KYC flag on it, so you can mint the soul-bond and then opt in to get KYC. If you get KYC, your soul-bond gets marked with that, yeah.

Mark:
I see. And then that makes you eligible to actually acquire the underlying NFT.

Sanjay Raghavan:
Exactly, exactly. So by doing that, in our case, again, privacy is important. So you don't want to say that Mark bought this house necessarily, but everybody on Ethereum will know that this wallet owns this NFT. But that's fine because the KYC information is all private, it's not put on the blockchain. It's only a flag on the blockchain that says that this particular wallet is approved.

Mark:
Makes sense. So that is cool, and I can understand how you brought it, this asset, on-chain. You mentioned there's not securities implications.

Sanjay Raghavan:
That's right. So that's the next big piece.

Mark:
Can you please explain that?

Sanjay Raghavan:
Yeah. So when the legislation about securities came out, at that time, Congress knew back in the '30s or whenever they did it that they cannot foresee every possible scenario and document to say, "This is a security and this is not." So there was this one term that was left pretty broadly called an investment contract, and that, a lot of people are familiar with the Howey Test, which kind of became the landmark-

Mark:
Oh, wait. You know what? I think I know where you're going.

Sanjay Raghavan:
Okay, yeah.

Mark:
Okay, that is a great point.

Sanjay Raghavan:
Let's hear you.

Mark:
So the Howey Test, which is the test by which you determine whether something is a security, requires... It's an investment contract in which the appreciation comes from the work of others. And if you are the sole owner of an LLC, the appreciation is coming from others, it's coming from you. You're actually sole owner and manager.

Sanjay Raghavan:
You're actually making all the decisions.

Mark:
And so that makes it not a security.

Sanjay Raghavan:
That's exactly right.

Mark:
Okay, makes sense.

Sanjay Raghavan:
And in our case, once the buyer buys the NFT, we will introduce them to local property managers in that market, but besides that, we don't get involved in helping pick one or making any decisions on how much money do you spend on rehab, what should you rent it at? All of those decisions are made by the owner.
Now, the owner does not have to look at tenant applications by themselves and go through all this selection and collecting rent and all of these activities themselves. They can certainly hire a property manager to do these activities as ministerial work, but they're in control of their economic profits or losses that come from this asset.

Mark:
Yes, that makes sense. Okay, got it. All right. So this makes a lot of sense.
The issue though is it doesn't seem like real world assets structured in this way have a lot of demand in the crypto ecosystem and a lot of demand from capital. And I've tried to hypothesize why that is. And the things I can come up with are, well, a lot of crypto capital doesn't want to be KYC'd, not just... It doesn't mean they're illicit, right? Maybe they're not in the US, maybe they don't see the point, maybe they just ideologically don't like the idea of it. Maybe it's a pain and they can put their capital elsewhere. So that's like a friction.
The second is that crypto actually yields, like the baseline cost of capital and crypto is pretty high, probably higher than these real world assets. I think you can earn more in crypto through, I don't know, yield farming and stuff than you can from buying a rental property. You just look at the comparable yields. And so maybe capital just like doesn't want to chase real world assets because the growth projections in crypto are so high that the yield, the yields that come from that and the potential appreciation is so high that it means capital doesn't want to go into these real world assets.
Do you think either of those... Well, first, do you agree that there's a lack of demand? And second, do you think that either of those are the things causing it or is there something else?

Sanjay Raghavan:
So I think there's some truth to what you just went through. When you look at purely crypto-native investors, they have other options available to them. One, of course, even though there's a real world asset and all that, maybe there's a lack of education about why this is a good diversification strategy. If you go to a wealth manager in the TradFi world and you said, "Hey, I have a million dollars that I want to invest with you, what do you suggest?" Back in the day, they would've said 60% in stocks, 40% in fixed income. That was the philosophy 10-plus years back. Today, wealth managers will probably tell you look at an allocation that includes stocks, bonds, and alternative investments. And when you look at alternatives, crypto is part of that universe also, but real estate generally fares pretty highly on that alternative market.
And the reason for that is wealth generation, historically, generational wealth has been created through real estate in the US and that's why this American dream is still alive and well and because you feel that that's the asset class that's something that you can pass on to your next generation and so on.
When you look at the yields generated, yes, you're right. I mean, in today's market, maybe an unlevered net yield, you get is 5% on a property in Atlanta, and couple of years back, maybe it was closer to 6 or 7%. And there's definitely market factors that are affecting that right now, including the very rapid interest rate hikes the Fed has done in the last year and a half or last year or so, and how that's really impacted real estate. However, in the long run, there's still about 7 million unit shortage for just single family properties in the US.
So when you look at demand supply characteristics from a macro perspective, once we get past this current macroeconomic economic scenario, we expect that this asset class will continue to grow. Like as we said earlier, large institutional ownership today is like 2% or so of the market, 2 to 3%. There's estimates that when you look at multifamily, on the other hand, that's probably like 50% owned by institutional.
So there's a lot of demand from Wall Street and institutional capital to get into this space, so we think the space will continue to grow and education will be a part of getting Web3 people to understand that diversification is a good thing and that's how you preserve wealth because otherwise you're in a very self-referential ecosystem, and not only you have the impact of when Bitcoin goes down, Ethereum goes down and other L-1s and native coins might go down with it, but also when the stock market goes down, these things tend to go down with it, at least for now, they're highly correlated.

Mark:
Well, okay, so that makes sense. I agree that this stuff should be part of a well-diversified portfolio, but if you think through the... I think what most people in crypto are wondering is will the next wave of massive adoption and 100xing and users in crypto come from real world assets being brought on-chain? And so through that frame, it's this is definitely a thing. The question that I think a lot of people have is, is this the next big thing in the medium term? Is this narrative a thing that's going to drive the next wave adoption?

Sanjay Raghavan:
Yeah. Sorry, go ahead.

Mark:
And you've described a couple personas. So people in the traditional market, people who allocate their assets in the traditional market, they're going to use crypto probably to allocate their crypto portfolio and they'll buy a property maybe through your Web2 interface, or maybe the ease of this will bring them on-chain, but probably the experience of using the chain will have to be easier before they decide it's actually just easier to buy this stuff on-chain. So that's one persona.
The other persona is people who are in crypto and have almost all their wealth in crypto and they're rotating that capital into a more diversified portfolio. And I guess I just wonder if that persona actually exists today.

Sanjay Raghavan:
So here's my take on this. You're absolutely right about these two personas, and with the properties we sold so far, we actually saw that demand coming not from the Web3 crypto natives, but from Web2 people who felt that the purchase process was a lot easier, the financing was a lot easier here than what they would have to go through in the TradFi world to get it.
And so they were crypto-curious people who already had maybe a custodial wallet with Coinbase or something and they owned some Bitcoin and E and so they were comfortable creating a self-custody Coinbase wallet and going through this process of remembering seed phrase and putting it away and all that.
But to your point, the wallet process is fairly cumbersome. And if you're not into crypto, you wouldn't normally... My parents would not go through the trouble of figuring out how to create a self-custodial wallet and buy Bitcoin.
And so what I think needs to happen in terms of infrastructure and tooling to create that massive option is wallet as a service and just abstract that whole thing away. And as long as you can log in using Google or whatever other preferred login mechanism you have, the underlying thing that you're using to store your assets can still be a wallet. And then you just have a website where you can actually see what it is that you hold in that database, so to say, or that idea.

Mark:
Okay. So basically, it sounds like to unlock real world assets on-chain from the kind of traditional markets, that requires a better user experience for crypto broadly, which is coming hopefully through account abstraction and wallets as service.

Sanjay Raghavan:
Right, right.

Mark:
Okay, totally get that.

Sanjay Raghavan:
And you and I know this, we are in this field and often, we'll try to connect our ledger and it won't connect properly to OpenSea, and now you're trying to use WalletConnect. This is painful enough for people like us who know this industry. It's really painful for people who don't know anything about this industry. So that wallet infrastructure definitely has to improve to get massive option.
I think someday when on your Apple iPhone there's actually, just like you have your wallet, where you scan your credit card and suddenly you can take your phone and point it at Starbucks and it gets you a coffee, if one day all cell phones also have a built-in Ethereum wallet, it creates all kinds of interesting use cases, and I hope we get there sooner than later, right? So that's on the Web2 site, yeah.

Mark:
Okay, so that's one blocker and one big unblocker. I think I'm sold on that, yeah.

Sanjay Raghavan:
The second aspect is it's really hard today to buy... Our transactions happen on USDC. They don't happen on E because as a real estate company that's also involved in the buying and selling of real estate, we cannot take the volatility of E then have a hedging program against it and stuff. So people typically need to show up with USDC, and today it's really hard for a Web2 person who does not have E or USDC sitting otherwise, to be able to come up with a property that costs $200,000. You are showing up with 80,000 USDC to buy it and you're trying to get the rest through some asset-based DeFi lending solution.
Getting that $80,000 USDC is really, really challenging today. So I hope we have better infrastructure and I mean, these are KYC people, US residents, we can prove that they're getting this for this purpose. There's no nefarious intent there, but it's still challenging and I hope the infrastructure gets better to allow these kinds of use cases to flourish.
So these are the two impediments I see on the Web2 front that will prevent massive option. And as soon as these are solved, then a lot more use cases can come up similar to ours that can use these features.
On the Web3 side, you are absolutely right that in the past, the yields have been so attractive on other products that people have just not had to look elsewhere. But today on Compound or Aave, I don't know what the yield is today, but maybe it's a sub-2%, something like that.

Mark:
It's very low, yeah. It's less than Treasury's.

Sanjay Raghavan:
Yeah. And Treasury's are 4.5%. And so that creates this unique situation even today where the risk-free assets, so to say, are providing twice as much yield. Why would you want to do over-collateralized crypto lending and all that?
I think there's going to be a lot of companies like Ondo and others that try to bring tokenized ETF, bond ETFs to the blockchain. And I think that will create maybe the first wave of RWA adoption at a larger scale is just people that want that flight to quality. They want to access US treasuries, which are highly liquid, short-term instruments like T-bills and three-month products for example. And you make a pretty decent yield there that you don't have to look at staking as a service or potentially an earned product on USDC or something else to make that yield. So there are TradFi yield products that will make their way into Web3, and I think that will start a bigger adoption of the real world yield products for Web3 natives.
So that, I think, is a start. But beyond that, all these other kind of use cases become somewhat esoteric and people do need to get educated. I think the recent deal that Centrifuge and BlockTower did with Maker that opened up $220 million of funding for RWAs, that's a starting point, but again, you're using Web3 native distribution there and what happens after this $220 million has been exhausted? Where do you find the next $200 million?
So I think they announced it a few weeks back. It's $220 million is the total facility size. Maker is the senior capital, I think roughly at $160 or $165 million, something of that magnitude. BlockTower came in as the junior capital on that, and that's broken down into four different walls, Maker walls with different lending. I think one is for business loans, one is maybe for receivables or trade finance, and a couple other walls for different purposes.

Mark:
I see. So this actually brings up an interesting issue, which is you may need to be KYC'd in order to own the asset, but as an owner, you can potentially go get financing using that asset as collateral from the crypto crowd, which is not necessarily subject to a lot. It's all on-chain, so it's not really subject to the same constraints. So you could take the NFT into an NFT lending protocol, put it in there, and borrow against it. In fact, that's how you could get financing for the property.

Sanjay Raghavan:
I'd say yes and no because I think there are a lot of programs that today allow you to, and even Compound and Aave have programs where you don't get... There's the Treasury program where you have to get KYC'd, but there's the compound available to everybody else where you don't need to get KYC'd.
And I think we will get to a place in the future, I think, where a lot of these types of real world assets, there will probably be more regulation around because ultimately the US government does not want you to be lending or using LP money from North Korea or Russia or other sanctioned countries. So if it's going to be a US-based entity that's offering these types of DeFi products, I feel that the regulation eventually will require you to KYC the liquidity providers and make sure you're offering that in a permissioned architecture as well.

Mark:
Well, sure, but I mean, these NFTs are composable, so it's up to an individual where it secures the financing, right?

Sanjay Raghavan:
Yes.

Mark:
And there's a big range of design space between blacklisting certain addresses and some sort of AML screening or blocking, and needing to whitelist every source of capital, right?

Sanjay Raghavan:
Yeah, agree. So I don't think that's a problem that's going to find a perfect fit solution. If I'm creating a lending protocol, I would create that as something that has a permissioned architecture on it.
Now of course, as you said, the NFT owner is free to use me or free to use anybody else. We're not restricting the NFTs to only be using protocols that have been whitelisted for borrowing. If you find some other NFT lender that's willing to give you better terms, you have every right as the owner of that NFT to do what you want with it.
But I do think folks like Centrifuge, for example, are creating permissioned architecture, so they do everything as securities offerings. If we create something of our own, we'll also probably use a permissioned architecture, but there are other DeFi protocols that may choose not to do that. And if they're operating in the US, they are taking the risk that the regulators could come after them for not being compliant and offering something to US residents or from US residents. So as long as any US parties are involved, the federal government has jurisdiction. So that's a risk individual protocols will have to-

Mark:
As an individual, let's say I own a home, am I actually... Is there a regulatory or legal issue with me getting an asset-backed loan, let's say, from a foreign capital provider? Do I have an obligation to KYC that person as an individual?

Sanjay Raghavan:
I think if you are operating in the US and you're facilitating a loan of this nature, I think you're subject to US securities laws or any other US laws.

Mark:
Well, hold on. We're mixing, we're using pronouns that are too vague, I realize. So, sorry.

Sanjay Raghavan:
Okay, yeah.

Mark:
If I own a house, a rental property, and I want to go get capital, I want to get a loan from a friend in Europe, let's say, and they say, "Sure," and they're willing to lend me the money, that's peer-to-peer, right? So are there regulatory considerations there? I hear you that if there's an intermediary, they almost certainly have regulatory.

Sanjay Raghavan:
Yeah, I think these types of considerations largely come into play with intermediaries. The question of the peer-to-peer scenario that you described, it depends on what your agreement with the counterparty is because ultimately, let's say you default on this loan and this other person gave you the loan from Europe, what remedies do they have?
So generally speaking, if you're going with a protocol that has a well-defined structure for the lending process and it's complying to some laws and regulations, and there are remedies described in the legal agreements as to what happens when one or the other party does not follow through on what they're supposed to do, it's a safer alternative. So I think most people will... That's kind of how it'll evolve as you find a well-structured program that you can be a part of. Of course, you can borrow from somebody else.
In our case, again, like I said, the token itself, the home token can only be transferred to a KYC'd wallet. So if your lender says, "Yeah, I'll give you an unsecured loan and so I'll just drop $100,000 USDC to you," and we have a handshake that, if you don't pay me, this NFT's mine, that's between you and them, but it'll be unsecured because the NFT cannot also go over to their wallet.

Mark:
So you have a smart contract that serves as escrow, right?

Sanjay Raghavan:
Sure, you can do that. But again, that smart contract would also have to have... We would have to drop the membership token as well.

Mark:
I see. So the vaults, like the escrow vault of an NFT lending protocol would have to be KYC'd as well, and then that would make it-

Sanjay Raghavan:
Yeah, yeah. So that's true in our case, so we can have a little bit more control on this, but there are probably other real world assets that don't need this. And so in that case, nothing stops you from entering into these types of agreements.
Now, obviously if there's an intermediary involved, and if there are US people that are part of this intermediary as well as if either counterparty is US resident, then I think the US does have jurisdiction on those types of transactions.

Mark:
Okay. Part of the reason I'm interested in this is because this is nitty-gritty, but the nitty-gritty is the rub for bringing real world assets on-chain. So actually, that's what you have to get into to understand the future of it. But I realize there's probably a third, I don't know if it's a persona, but a third path by which real world assets become very prominent, which is... So we've described one, which is for Web2 people, it's just easier to buy things on-chain. There's a bottleneck there we've identified, which is usability. Two is people who have crypto wealth and want to access real world assets for portfolio diversification purposes. I think we've realized there's a little bit of a blocker there with yield opportunities on-chain and some user frictions, but in theory, that should happen eventually.
But there's also a third, which is if capital were cheaper on-chain than off-chain, you would pull a lot of people on-chain so that they could access cheaper financing, right?

Sanjay Raghavan:
That's right.

Mark:
That would trigger a very large capital flow. People get a mortgage broker and they shop around 10 mortgage providers or whatever, and that's a big part of the purchase.

Sanjay Raghavan:
Absolutely, absolutely.

Mark:
So if one were, oh hey, what about getting an on-chain loan or accessing on-chain capital and that actually had better rates, I mean, for sure tons of people would go deal with the chain.

Sanjay Raghavan:
That's a really interesting observation as well. So in our case, when I said the two kind of buyers wanted to use the Web3 mechanism to buy it was primarily because they got one much easier underwriting on Web3 than they would have to go through on Web2, but also the rate and terms were compelling also compared to TradFi.
Now, I would probably call these buyers as Web2.5 because they did have some exposure to crypto, they understood blockchain, generally highly educated people.

Mark:
Usability issue is still an issue, right?

Sanjay Raghavan:
It's still an issue.

Mark:
But there's a pull that's much stronger than just, "Oh, it's easier."

Sanjay Raghavan:
Absolutely. And so I do believe that that's what will drive... Ultimately, if the question were... Let's say it's really easy to get a wallet. Let's say it's really easy to get USDC. Why should people still do it this way versus doing it the other way? And that to me is an answer. If you think about, people used to use yellow cabs before and you'd Google and find out who's the cab company that you can call for an airport drop next morning at 5:00, and you go through the list and the first one says, "I don't have anybody that can do that." You call the second one and third one. Then at 5:00, you're standing outside in the cold and hoping that the cab actually shows up. It's 5:10. You get really desperate, and then you're going back to Google and calling other people.
Then Uber and Lyft came along. It required that you now install an app and you learn how to use the app and put your credit card on it, this and that. And so, I mean, sure, that's not as complex as setting up a wallet and all that, but in those days when a lot of people were not doing a lot of apps, having to set up these apps and doing all these things was still challenging to some extent. But they did it. And over time, the convenience, and initially it was also cheaper, I guess, when they launched these products, Uber and Lyft were cheaper than yellow cab, so they overcame the extra work of setting up all this for a better price. But then over time, the convenience just also won everybody over to the other side.
So I think for us, the combination will be the convenience of doing something in a much better fashion in Web3 without all the kind of traditional hassles involved in financing. But also because there are so many intermediaries and brick and mortars involved in TradFi, and you can eliminate a lot of that with smart contract and programmable money, some of that money can also be passed between the lender and the borrower. And so you do have the option to make it cheaper as well.
If you look at securitization, for example, typical securitizations are $200 million-plus. There's a commitment fee, a use fee, a non-use fee, a minimum yield threshold, and then first right of refusal on securitization, there's an exit fee if you decide to go with someone else, there's a mandatory hedging fee for an interest rate cap. You figure out all these things and then you're hiring lawyers and you know.

Mark:
I see. So what you're saying is the process of providing capital in Web2 or TradFi has a lot of frictions which create cost and that's a tax, essentially, that create...

Sanjay Raghavan:
Okay. Ah.

Mark:
Okay. Looks like we're recording again.

Sanjay Raghavan:
Awesome. All right.

Mark:
Okay. I lost my train of thought. Where were you? We should tie up just to get-

Sanjay Raghavan:
Yeah, so we were talking about, in the traditional process once-

Mark:
Right, okay. Got it.

Sanjay Raghavan:
With the lending going to securitization, that process is extremely cumbersome, very expensive, and so a lot of that can be-

Mark:
So in theory, every time you transact a property, there's a big transaction cost. You mentioned 7.5% before. And also in the financing process and the securitization process, there's a lot of operational and efficiency, which create a lot of costs there too. And so if you can do it all more efficiently, in theory, you should be able to have sustainably better and cheaper financing because the costs of providing the financing are less, and that has to enter into the rate one way or another.
Okay, that totally makes sense long term. In the short term, macro really kind of seems to drive this, right?

Sanjay Raghavan:
That's right.

Mark:
It's like, well, could the cost of capital for getting a mortgage on a home, on a second home that's going to be a rental property, be less if I borrowed on-chain than if I went to a lender off-chain? Do you think that's feasible in the short and medium terms?

Sanjay Raghavan:
It's not only feasible, I think it's happening right now, I would say, because-

Mark:
Really? Where is that capital coming from?

Sanjay Raghavan:
Yeah. I mean, so if you look at the, again, crypto LPs that are working on different DeFi protocols... So actually, let's rewind for a second.
In the traditional world, say a 30-year Fannie Mae, conforming mortgage is probably hovering around the 7% range right now. If you're trying to buy a rental property, that typically has a 75 to 150 basis point on the rate at which it's financed, so you're already looking at something that's a, call it 7.5 to 8.5% for a Fannie Mae-type product. If you try to get out of that, then either you have relationships with a very strong community bank that will create a bespoke program for you, or you're looking at non-bank lenders, which are typically north of 8%, so they're like 10 to 12%, have other cross collateralization requirements, personal guarantees, and so on.
And I would even argue that some of the community banks, if you're trying to do this outside of Fannie Mae as a portfolio lending-type scenario, they may have cash collateral requirements and other things as well before they make you eligible for a program like this. So there's, call it in the traditional world, if you want to get a rental property, you're looking at something between 8 and 12%.
When we do these on-chain, we don't have those restrictions. So it's looking at where demand meets supply, and as you pointed out earlier, in 2021, there were a lot of DeFi opportunities for people, and a lot of those have since evaporated. And on compound, you're looking at a 2% rate of return. So people are looking at where else can I invest my money to get a better yield on it? And if you, like we did, the last deal we did was at 7% interest-only loan for two years. Now, that's a significant premium. Maybe it's only a 250-basis-point premium from risk-free Treasury rates, but it's a 500-basis-point premium from what you might get in over-collateralized crypto lending.
And so that form of capital is finding its way to these types of assets.

Mark:
Wow, that's so interesting. I mean, it strikes me that if we think about the things that will... To me, out of everything you've described, if those numbers work, where on-chain capital can actually be gotten for cheaper than off-chain capital, that's probably the biggest draw should we solve the usability issue, which may be in the near term with account abstraction, that's something that could bring a huge wave of real world assets.
And so I guess my big takeaway is the usability technological process is continuing at pace. In the meantime, the macro and the cost of capital is what will really tell us whether there will be a massive shift where real world assets can be brought on-chain. And that's the 1000x that the narratives are thinking about.

Sanjay Raghavan:
That's right, that's right.

Mark:
When they think about [inaudible 00:56:55] rates.

Sanjay Raghavan:
I do think eventually the Fed will start dropping, I think, once they hit their target inflation rate. And there's a lot of hypothesis that by June, we could be at a 4% or sub-4% inflation rate. And the reason for that being that CPI is measured as a 12-month average, and of course, 12 months back, inflation was really high. So when that's a component in the 12-month simple average, that plays a role and it weights the inflation higher. But as you keep progressing and you get to your 12 months, where you start off at seven and you end up with three. And on an average basis, you might be at four or whatever.

Mark:
Well, what's interesting is like to that point, if the Fed drops rates, then crypto probably, risk assets, probably moon again, which raises the cost of capital in crypto and decreases the cost of capital in TradFi and vice versa, right?

Sanjay Raghavan:
Yeah. There's this delicate dance that's going on between TradFi and crypto, but I think eventually, the short-term arbitrage opportunities should dry up in efficient markets. And then you're basically looking at the cost of capital. What is the ultimate investor in a TradFi deal receiving versus what is that ultimate investor receiving in crypto? And in TradFi, that lending rate has to be much higher because there's 12 intermediaries to feed before that ultimate investor makes their return.

Mark:
Totally makes sense, yeah.

Sanjay Raghavan:
And so the question really is, can you give a little bit more to that investor and a little bit more to the borrower and make crypto more attractive?

Mark:
Yeah. I mean, and also in the long term, is crypto, is Bitcoin a risk asset? I mean, once it manifests its destiny, perhaps not, right? Growth slows, it's no longer a risk asset. So in the long term, totally agree with you. And I think one of the conclusions here is that in the short term, it all comes back to the Fed.

Sanjay Raghavan:
It certainly does.

Mark:
And what they do may drive the timing of the RWA narrative.

Sanjay Raghavan:
It will. And I think the timing is we expect 2024 to be the year where things start picking up again, both on the macro front, and I think largely I'm very encouraged with all the use cases I'm seeing right now that are being experimented on DeFi. There's lots going on with refi, regenerative finance, carbon credit trading, microfinancing, and obviously there's lots of protocols trying receivable factoring using Web3.
So there's lots of RWA use cases that are finding their way to Web3 and there's lots of companies trying to make things faster and more efficient for their LPs and borrowers as well. So I'm encouraged when I see all of that. And that's why I feel that when the market... Everybody that's building right now, great, keep at it. And when the market comes back up, hopefully that is the trigger where there's half-a-dozen different use cases and they've all tried out and piloted different things and there's an opportunity for some or many of them to start growing and becoming more mainstream.

Mark:
Makes sense. Well, Sanjay, thank you so much for a really interesting conversation about the future of real world assets on-chain.

Sanjay Raghavan:
My pleasure. This was fantastic. We should do it again.

Mark:
Absolutely.

Sanjay Raghavan:
In a few months.

Mark:
Yeah.

Sanjay Raghavan:
We'll see what the Fed does.

Mark:
Yeah, yeah. Before we go, is there anything you'd like to share with the audience that we haven't touched on? And how can they follow you and learn more about you and the projects you work on if they want to learn more?

Sanjay Raghavan:
Onchain.roofstock.com. We're the Web3 division of Roofstock, so definitely that's an area. And then on Twitter, we are @RSOnChain, and then I'm personally @Eth_Sanjay, S-A-N-J-A-Y.
And my goal this year is to be a spokesperson for our community, try to educate where I can, and help people sort out securities issues or other complex... I mean, some of these, we haven't even talked about taxes, which is another thing people should be worried about when they're planning these things. So if there are companies, Web3 companies, looking at real world assets and figuring out tokenization strategies and they have questions on securities issues or tax issues, while it's not legal advice or tax advice, I'm always available as a [inaudible 01:01:23]-

Mark:
Oh, I love taxes. We should do another episode on taxes involved here because especially with real estate...

Sanjay Raghavan:
Absolutely.

Mark:
I mean, that is a tax structurer's dream. You know?

Sanjay Raghavan:
That's right.

Mark:
Great. Well, thanks so much.

Sanjay Raghavan:
Thanks for your time and really enjoyed this, Chad, and thanks for everything you're doing to educate the community and help the Web3 progress forward.

Mark:
Thank you and likewise.