This is Defi Decoded, the conversation show with the leaders, builders, and investors pushing the internet's next economic frontier, hosted by Alex Tapscott. This podcast is for informational and educational purposes only and does not constitute investment advice. Nothing discussed on this show should be considered a recommendation to buy, sell, or hold any securities or digital assets. The hosts and guests may own positions in some of the assets mentioned. Listeners should conduct their own research and consult with qualified financial, legal, and tax professionals before making any investment decisions.
Hello and welcome back to Defi Decoded. I'm Alex Tapscott. Today's conversation is one that I've been looking forward to for quite a while because my guest has spent the better part of the last two decades really pushing the boundaries of financial services in many respects, kind of dragging an often-staid industry into the future. Mike Cagney is in my view the very embodiment of a serial entrepreneur. Today he's the co-founder and executive chairman of Figure. Before Figure, he co-founded and led SoFi, which is a financial technology company that has done, I think, more than just about any to help bring money and banking into the future and into the internet era. And before that, he ran a global macro hedge fund, built and sold a wealth management software company, and headed proprietary trading at Wells Fargo. So he has, I think, insights from both inside and outside of the traditional world of financial services. So let's talk about Figure. With around $24 billion in home equity loans originated, Figure is the largest non-bank HELOC provider in the country. But Figure is much more than a fintech with a slick app. It's a blockchain-native capital marketplace. The loans, the ownership registry, the custody, the secondary trading, all of it lives on-chain, and that includes Democratized Prime, their on-chain private credit venue, and several other products and services. And so this is the topic that we want to talk about today, which is we often talk about DeFi and CeFi being sort of separate, but in many respects they are related and on a collision course, or maybe more appropriately, a convergence course, where the technology that enables crypto assets like Bitcoin, Ethereum, and so forth will become foundational to the next era of financial services. There is quite literally no better person to be speaking to about that, a practitioner who's built many businesses over different eras and is now pushing the boundaries with blockchain. So Mike, thank you very much and welcome to Defi Decoded.
Oh, thanks for having me. I appreciate it.
Yeah. All right. So, I'm not sure where to start exactly. You know, you've had many successes in life. And that's not to pump your tires. It's more to ask you, what was the decision to launch Figure? What did you see in the market that needed to be solved?
Yeah, the whole premise behind Figure is really this aspect of connecting sources and uses of capital directly. And you know, if you go back all the way to when I was running my macro fund and I was looking at what Chris Larsen was doing with a company called Prosper, which was one of the first, in fact, was the first peer-to-peer lending platform, kind of created fintech 1.0. And I just, you know, having been in a bank, running financial products and prop trading at Wells in the late 90s, had an appreciation for both the value and the limitations of a bank, and in particular, the inefficiency of the bank as a capital allocator. And so I really liked the idea of being able to connect sources and uses directly. And so, you know, the idea of Prosper peer-to-peer lending, I invest in your loan, that to me seemed really interesting and disruptive. And so in 2010, I had a chance to go down to Stanford and do a fellowship at the Graduate School of Business. And when I was there, I was fortunate to meet up with my co-founders at SoFi. And that was the original premise behind SoFi, was that we were going to get alumni to invest in students in their loans, both on primary and refinancing, and that there was a natural affinity relationship there and other externalities beyond the financial relationship that we could tap into. And you know, the both fortunate and unfortunate thing about that was the loans were so popular they far outstripped our ability to raise capital, and you know, we had to go and resort to what I would consider the fintech 2.0 model, which was traditional warehouse financing and asset-backed securitization. And never really built into that sources and uses construct. And so with blockchain, you know, I left SoFi in 2017 and I started thinking a lot about what I was going to do next. And I had this aha moment with blockchain where in late '17 I said, well look, if you can put an asset native on-chain, where you know the provenance, the ownership, the composition, the history, you can get true digital perfection to it, you can not only transform how that asset's traded because it can trade bilaterally, but you can also change how it's financed because you can finance it on an asset-based construct and have direct perfection to the collateral. And that reopened this idea of sources and uses again. And that's really the genesis behind Figure. At the end of the day, what we were looking to do was effectively disintermediate capital allocation and bring it directly to the sources and uses, and that's what we've been able to effectively do with Figure.
So from the outset was Figure blockchain-enabled or is this something that you've integrated more with time as the technology itself has also matured?
No, blockchain was paramount to being able to do it. But the initial thesis on Figure was that we could bring the technology into the financial ecosystem and the banks would adopt it. And so I pitched a dozen banks early on about the virtues of doing asset origination and aggregation and ultimately securitization or sales on-chain. And you know, universally every one of those banks is like, ah, we love it, it's great, we'd like to be the 10th bank to do this, so call us when you get the first nine. And so it just became apparent to me that we had to be first mover. And so we stood up a lending platform specifically to be first mover and force the buy side to start engaging on blockchain.
Interesting. And so in terms of the loans that you've originated then, is that capital from your balance sheet or is that money that you're sourcing from investors who want access to that kind of borrower?
Yeah. So in the beginning, we originated loans and we did it, you know, again kind of a classic fintech 2.0 model, which is we did direct-to-consumer acquisition. We originated those assets, but pretty quickly we evolved into a B2B construct where, you know, today we have over 387 partners that originate assets in the Figure ecosystem and they're tapping effectively a capital market that we built out, a marketplace, and that's really what our core business is, we're a marketplace operator. And so what we're doing is those entities are both accessing initial capital through DeFi and then downstream things like security-backed leverage through DeFi as well. So we think about blockchain as having three value props. There's transactional efficiency we get up in the front, that's reducing audit and QC expenses and warehouse frequency time and so forth. There's liquidity, which is often misunderstood in blockchain because there's this notion that you put something on blockchain it becomes liquid and obviously that's not true. But what we've been able to do is build a liquid capital market for our loans that doesn't exist anywhere else. And you can go on CoinGecko and you can see on any given day how many loans trade on Figure Connect, and it's between 30 and 300 million a day that are on there that trade, and you know, I think TVL there's probably 18 or 19 billion. And then the key is the DeFi side, your ability to finance. And so what we did early on, I would look at some of these DeFi protocols and say, geez, people are borrowing against their Ethereum at 3%. I'm paying Goldman 6.5%, why is there this dichotomy and why can't I get access to that capital? And it underscored a couple of really important premises. And I think a lot of the tradfi community that's looking towards DeFi doesn't fully grasp or understand this, which is it's not that there's this panacea of dumb money in DeFi that will take sub-market returns. It's that they're lending against collateral that is very, very liquid. And I draw the analogy, I've been lending against crypto since 2020 through Figure. We've never lost a dollar lending against crypto in any circumstance or any market environment, whereas I can't say that about my mortgages.
Can I ask, like the crypto sleeve, is that all on-chain? Is that like DeFi, like in basically lending pools?
Yeah. So what we're doing in particular with the loans is we drop the loans into a container, and we call this process Forge, but we effectively drop the whole loans into a container. We create single-dollar unit participation interest to those loans, and then we can concurrently stand up those units as both collateral and as liquidity. So, you know, if like with Agora, our partner that we're now doing auto on DeFi, we drop the loans into this Forge bucket. We create these participation units and there's a natural reason for liquidity in those units because if they get expensive, originators are going to sell loans into the bucket, get the tokens, and sell them onto the DeFi protocol. If they get cheap, entities like ourselves will go in and buy them, turn them into loans, and securitize them. And so there's always a natural two-way market and that liquidity is a necessary condition to drive any real economic efficiency out of the DeFi side because the cost of your capital in DeFi is going to be directly commensurate with the amount of liquidity that capital has.
Right. But also the loans in the DeFi side would be like traditional over-collateralized loans, right? With crypto assets like pledged as collateral in...
Well, but they so yeah, so certainly on the crypto-backed loans they are, and even on the mortgages they are, right, because you have property as collateral and over-collateralization. On auto, you hope you have over-collateralization, it's not always the case, but this works for unsecured consumer, it works for all credit. It's just you have to be able to break that credit down into some kind of structure that has a natural liquidity to it.
Yeah. I mean, the reason I asked that question is because a lot of the lending in DeFi, I think all of it pretty much, is collateralized lending, right? And but within personal finance and consumer fintech, there's so much opportunity for unsecured lending. But one of the challenges with DeFi is you don't know who your counterparty is on the other side. So you basically have, well I should say in most cases, right? And so you have collateral, you have for every dollar you're lending 50 cents or whatever it may be. But I always felt that uncollateralized loans, reputation-backed loans could be an enormous opportunity. And I'm just thinking about your original introduction regarding SoFi which was based on this sort of collective community, right, where because there were Stanford grads looking for loans there were people more willing to lend to them. I mean, part of that is about reputation and it's about a sense of connection between the borrower and the lender. So, yeah, just I'm just curious like if you think at all about that, about sort of reputation-based lending beyond what's currently possible.
Yeah. And I think you need to look at it in two constructs. There's the collateral itself. So, are those loans over-collateralized? Are they unsecured? And it doesn't really matter other than the premise of how much underlying liquidity they're going to create. So I can drop unsecured consumer into a Forge pool the same way I can drop auto or mortgage in there. And then what's coming out of that are these participation units. Now those participation units are being lent against on an over-collateralized basis very similar to how a warehouse would work with the street. Right? So when I pledge loans into Goldman, if I'm pledging unsecured consumer loans into Goldman, they're going to give me 90 cents on the dollar. And that's analogous to what's happening in the DeFi side is you're putting up a dollar's worth of collateral to borrow 90 cents against that. And what you're asking the lender to do is look at the volatility, the liquidity, and the advance rate to decide whether or not it makes sense for them to lend into that ecosystem. And so it truly is an asset-based lending market where the actual collateral is second order. It has obviously a significant impact on how volatile the collateral could be and how liquid it could be. But the first order conditions are volatility and liquidity.
Yeah. Absolutely. You use a word that we like to use a lot too which is the container. Now typically when we think about a container like a token, people think of tokens as cryptocurrencies and they get a bit confused but ultimately like a token is a container for value in the same way a shipping container is a container for goods and so it can be programmed to kind of contain anything that you may want to represent. So tell us a little bit about the ways in which you're expanding that thinking, like you mentioned home loans like HELOCs, car loans, DeFi lending. What else can we fit into this container?
Yeah. And I used to call it a vault, but the industry hijacked that term and now I'm not allowed to use that anymore. So I have to call it container. So the Forge process, this container is critical on assets, on-chain native assets that have some friction in terms of trading and liquidity. So, you know, if I were to put whole auto loans up as collateral on a DeFi protocol and you're lending against it, well, if that LTV is broken, what do you do as a lender? Let's say I lend $5,000 to the protocol. The smallest loan balance is $50,000. So, how do I take one-tenth of that loan? And even if I could take it, where am I going to go sell it? And so what we need to do is we need to transform that relatively coarse collateral construct into something that's finite, granular, and liquid. And there's certain things you don't have to do that for. So crypto, for example, already has liquidity and is infinitely divisible. Equity, so we did the first blockchain-native equity with Figure Stock. So we have public equity trading on-chain and there's a market to borrow against that and that doesn't need the container because it's also infinitely divisible and it also has intrinsic liquidity to it. But things like a mortgage or a consumer loan absolutely needs this abstraction layer to create that granularity and that liquidity to it. And then there's certain things that just are never going to be able to fit in the bucket. So, you know, I get people all the time pitching me concepts like the shopping center in Miami that kicks off a yield, like why can't we put it on blockchain and finance it? And in a lot of those situations, the perception is there's cheaper money on blockchain than there is in tradfi marketplaces, which is not necessarily true and almost exclusively not true. And in other circumstances, it just isn't a starting point because you can't create liquidity out of that collateral.
Mhm. I mean, in the case of the shopping center, the idea would be that this could be a way for an owner of a shopping center to syndicate the ownership of that shopping center, like to divide it up into like equity units or...
Yeah. But you can do that through LLC interest. And this is what I always push back on people, that's not a great use case. I use the example, and my friends at Apollo get mad at me when I do this, but there was a lot of fanfare when they put their Arcred, right? That effectively BDC wrapper on-chain. And I'm like, this is horrible collateral because if I ever foreclose and take possession, one, I can't take possession unless Apollo approves me and I need to be a qualified purchaser to do it, right? Two, there's no secondary liquidity for it. And so, which is why it's had relatively limited uptake. I think there's about $125 million of TVL of Arcred, primarily bought by a couple of foundations, right? It just, funds, for example, aren't great liquidity for this purpose. So, you need assets that you can decompose down to something that supports a liquid marketplace where there's a natural bid.
Interesting. And so, I mean, I think of something like mortgages for example, like there wasn't a mortgage bond market and then they created it and that created liquidity in an asset that previously was illiquid. So like do you just take the MBS and turn it into a token or can you like tokenize real estate that way where it's if not one shopping center like dozens of or hundreds of them or hundreds of thousands of houses and tokenize it in the same way you would an MBS?
Yeah. Yeah, I mean, you start to get to a position where you can do that with critical mass and there's certainly a path for that. I think the first step has been getting loans on-chain and then breaking them down into more granular, finite, tradable units. The second thing, and thing we're starting to do right now, is do our securitizations on-chain. So, the bonds that come off of those securitizations, in particular, the investment grade securities, they will be native on-chain as blockchain-native digital assets. What I'd really like to do is a what's called a Reg A securitization, which means retail can participate and so there's no limitation. Traditionally, since the financial crisis, securitizations are done as 144As, which limit to institutions to participate in them. Doing a Reg A means that we could have investment grade debt as collateral that anyone can take possession of just like they can a public equity, right? And so this year my plan is to do at least one Reg A deal, but before that I'll do a traditional 144A securitization on-chain and get that AAA collateral on there.
