Maple Finance: On-Chain Institutional Yield & Lending Markets - Sidney Powell. Ep. 595

Transcript

The mix on the borrower side has been it's shifted more towards primes. They provide margin to their clients and and then they do significant amounts of OTC trading, but also onchain hedge funds and yield funds. So they might want to borrow from us against collateral that is actually forming part of a yield strategy they're doing on chain, but they borrow from us to lever those strategies. And then the other one would be liquid token funds. So they might have a position they have high conviction in. It's too expensive for them to finance the position using per and they don't want to risk a liquidation. So instead what they do is they could purchase the position, post it to us as collateral, borrow dollars and and lever up that position or invest in other deals that they want to.

Increasingly what shifted this year is we've started taking in more family offices and high net worths. You know they might want to purchase real estate using crypto holdings that they've accumulated over a number of years from us. What they're looking for is certainty on rates, making sure that they almost never are in danger of getting liquidated. So they'll post significant overcalization and then counterparty risk. Welcome to Epicenter, the show which talks about the technologies, projects and people driving decentralization and the blockchain revolution. I'm Fria ans and today I'm speaking with Sid Powell who is the CEO and co-founder of Maple Finance which is um which does onchain lending in various flavors we'll talk about in just a bit. Um before we do that, let me tell you about our sponsors

this week. If you're looking to stake your crypto with confidence, look no further than Chorus 1. More than 150,000 delegators, including institutions like BitGo, Panta Capital, and Ledger, trust Chorus One with their assets. They support over 50 blockchains and are leaders in governance or networks like Cosmos, ensuring your stake is responsibly managed. Thanks to their advanced ME research, you can also enjoy the highest staking rewards. You can stake directly from your preferred wallet, set up a white label node, restake your assets on IGO or Symbiotic, or use the SDK for multi-chain staking in your app. Learn more at chorus.1 and start staking today. This episode is proudly brought to you by Nosis, a collective dedicated

to advancing a decentralized future. Nosis leads innovation with circles, Nosis Pay, and Metri reshaping open banking and money. With Hashi and Nosis VPN, they're building a more resilient privacy focused internet. If you're looking for an L1 to launch your project, Nosis Chain offers the same development environment as Ethereum with lower transaction fees. It's supported by over 200,000 validators, making Nosis chain a reliable and credibly neutral foundation for your applications. Nosis DAO drives Nosis governance where every voice matters. Join the Nosis community in the Nosis DAO forum today. Deploy on the EVM compatible Nosis chain or secure

the network with just one GNO and affordable hardware. Start your decentralization journey today at nosis.io. IO. Cool. Uh said it's super nice to have you on. Um maybe um before we dive into Maple, tell us about yourself kind of like before Maple um what was your journey like? Yeah, for sure. And first of all, thank you for having me on. I've I've been a longtime listener to the to the podcast. So, my uh my background was actually um banking and banking and finance before I got into crypto. So, I used to work in an area called securitization. um at one of the bigger commercial Australian banks and it uh it sounds like a mouthful but

effectively what it was was helping helping large lending companies to borrow from both banks and capital markets so that they could give smaller loans to their customers. So I was always I was always kind of very familiar and imshed in the lending business. And then after I'd been there for a few years, I could kind of see that, you know, bank banks are a very big political organization. And I'd always sort of wanted to to start something or be at a more entrepreneurial company. So I had the opportunity to go to a smaller commercial lending company. And that was actually where I met my co-founder Joe.

And so he was running finance, I was running treasury and that company would do small loans to small to medium businesses in Australia. Um so still commercial lending. And whilst there um you know we were trying to set up a funding line with with a large bank um ostensibly so that that business could grow. And it was whilst doing that that we started, you know, investing on the side in crypto and then thinking about could we tokenize what we'd been doing. So, you know, kind of institutional lending and and issuing bonds and do that using smart contracts and using using a blockchain. And so, this was like late 2018 and then started working on it kind of on the side in

2019 and into 2020. And we made just about every possible mistake that you can in terms of thinking thinking about the thinking about the idea and getting started. So, you know, this included everything from trying to do peer-to-peer lending with like anonymous people on the internet um using a blockchain and eventually we came around to the idea of doing institutional lending. And so that was kind of how we that was kind of how the the original sort of business model for Maple evolved. But those backgrounds were really more on the finance side rather than than the technical side. So what was the specific gap in the market that you were trying to fill? Uh because kind of like the even at that time there were already some some lending markets

around, right? Yeah, it was um it it was still remarkably early. So if I look at um when I when I first wrote the white paper for Maple, it was uh would have been like January 2019. And as I'd been, you know, as as as we'd been like investigating crypto and and DeFi, like the entire DeFi space at this time, you have to remember, was about five companies. It was Maker, Set Protocol, um, uh, DYDX, uh, a company called Dharma that was also trying to do peer-to-peer lending and, um, and, uh, and and compound. So the the the beginnings of kind of DeFi and onchain lending had begun, but um what we what we wanted to do that was going to be

kind of different was we wanted to focus on large tokenized bonds effectively like what I'd been doing in in banking and the the kind of the insight was you know we used to do many different layers of of bonds um uh what you would call trenches and and the insight was well if you have software could you just program it so effective ly the every dollar is a separate trench. You know, you could almost kind of scale it to infinity. And so that was what really got me interested in smart contracts and and the idea of doing this on chain to begin with. But um as we got started doing Maple, we had to kind of find a a a differentiate like a differentiated angle for doing this. And a lot of

people were already doing over collateralized lending. So what we started to look at in 2020 was could we do um institutional loans to you know to corporate borrowers but use the pool concept that had been kind of popularized by compound and and by a which had just rebranded from ELAND at the time but bring a professional underwriter who could also put in first loss capital. So it started to look almost like a credit fund on chain. But we had to separate separate ourselves from the players who already doing overc collateralized lending. So maker, a compound. And I'm a very big believer that you have to kind of differentiate and find your niche if you if you want to be successful. Otherwise, it's very

difficult to just go in and try and compete with with folks who are already quite scaled and and set up. So, so the reason why um kind of the overcolaterized um lending markets came first was because it's much easier, right? Kind of like if you do under Yeah. If you do undercolateralized lending kind of you have to deal with kind of all sorts of like risk underwriting and kind of like some sort of reputation and kind of like there's um um price oracles and I mean it it's it's kind of like it it um it just compounds the complexity of overcolateralized

um lending markets, right? Um and so kind of and then kind of you you already were thinking about kind of like targeting um Treadfi players who um by and large are much more riskaverse than kind of like people who were already in the space. So it it kind of it sounds like um you you you had kind of had to start off on on a backhand, right? So, kind of kind of taking the more complex technical route and then also pitching to um uh to players who may not be the most Yeah. who may not have the biggest risk affinity. Yeah. No, no, that's um

that that's totally correct. And and as we got started, like we we did try and approach it from a number of different angles. So, you know, initially we looked at doing peer-to-p peer loans and I would read all these white papers and studies of the returns on um the returns on Lending Club and portfolios and and on deck and but we found that it was it was it's a really difficult business to be in doing peer-to-peer consumer loans and so we tried doing that with our own savings and then we we shifted because we realized it wasn't scalable and so the idea was okay well you could lend to businesses they produce management accounts you just identify which ones are profitable whereas If you're lending to a consumer, your risks are much more idiosyncratic. Are they going to use the money to go gamble? Are they using it to

make risky investments? The types of events that trigger you, you know, pull you into default when you're lending to consumers are fraud, they get divorced, they get ill. Um, and all of those are really hard to predict. and uh you had this big adverse selection problem where the people who were going to come to you to borrow um if they were crypto individuals would often be you know adversely selected they couldn't get finance elsewhere and so we're really concerned about going into that line of business whereas with companies by 2020 you had started to see profitable crypto firms there were the beginnings of the market makers like GSR and others who were cryptonatives so they didn't have access to to banking finance but they were profitable in crypto terms And um and so you could look at their

management accounts, you could you know you could meet them, understand their business models and by that stage also remember that the macro backdrop was that rates were trending to zero around 2020. You know postco the system was flush with liquidity and so if you're able to lend to these parties at a spread of you know 9 10% versus tea bills people were extremely interested in that and and the system was flushed with liquidity. So there there were a number of factors that kind of came together in a bit of a confluence. There was the rise of crypto trading firms.

There was the the backdrop where where rates were kind of near zero on everything else. And then there was the fact that people were already very comfortable by this stage with over collateralized lending and the natural next step was to then progress to to under collateralized lending. Okay. Yeah, that that makes perfect sense. So um maybe before we kind of go into kind of like the business case again um walk us through um the core technical stack. So um I'll I'll I'll contrast what it what it was then to to kind of what it is now. Now um back then we were built kind of purely on Ethereum and and the the technical stack was solidity code where we could take in USDC deposits.

people would receive an L people would um mint an LP token that represented their their proportional share of the pool and then you would have this um manager which we called a delegate who would then determine the terms on which loans could be sent out and then the loans would be sent using a smart contract. So a um like a borrower would request a loan. this delegate this delegate um role would uh approve the loan and then the terms would then appear um uh on chain in a a smart contract loan. So interest rate term um amount and where it's evolved to over time is um our our code base has gotten much more sophisticated. So still

predominantly in solidity like most of our most of our loan originations happen on ETH mainet and um nowadays the stack is kind of a hybrid of CFI and DeFi. So when you deposit on Maple you still mint an LP token that's a 4626 uh vault token. It acrru yield for you. The pool is denominated in USDC. We now perform the role of approving all of the loans. So a borrower requests a loan and it and our contract fires off the money straight to their wallet still with the interest rate and other terms coded in but now they can post collateral and this is an important differentiating factor. They can post native BTC, Salana, XRP to

qualified custody accounts or to a Fireblocks account. And so we manage the collateral um you know more offchain without using smart contracts. And that's become kind of a core value ad for a lot of these institutional borrowers. While our loans themselves and the deposits are all onchain managed through smart contracts on Ethereum and then um we we've also used recently um layer zero of FTS to push some of these LP tokens to other chains. So we're trying to ensure that we can achieve distribution across other chains.

But is it still a poolled approach or am I pulled approach? Yes. And how how do you decide kind of which risk profiles to kind of allow into the pool? Because kind of like underwriting risk is uh it's quite a feat in itself, right? And kind of like you have to decide kind of like who qualifies for kind of like which interest uh uh which interest rate, right? And so it's it's it's partly art, partly science. Previously under the old model where it was the undercolateralized lending, it was much more of an emphasis on review of the balance sheet of the borrower. So how much equity do they have, how much debt do they have, how profitable are they, and how much money do they make because that services your interest. And um and then we would review financials on a on

a monthly basis. So we get updated financials from from the borrowers. And it was um it we also used more third parties back then. So we'd have this kind of third party platform model where we wanted other people to run pools and we thought that was much more scalable. What we found since then is we still coped the reputational risk when something went wrong. So we decided to bring that delegate role inhouse and now we are we are the primary underwriter. Um but the model of underwriting has shifted much more to collateralbased underwrite. So now it's more okay we've accepted BTC. What's the right LTV? Is it 70% or is it 80? And we could we manage the risk by having kind of two

trigger levels. So we have your initial loan to value, which is if you give me a million, I give you $700,000 loan. That's a 70% LTV. And what I would do is I would set a margin call level at a higher LTV. So let's say 80. And at that level, you have 12 hours to post me more collateral if your BTC drops in price. and then a liquidation level above that which is no more questions asked. I'm just selling your collateral as a kind of stop loss and that would be like 85. This is if we're talking about BTC but as we start to look at other forms of collateral we've underwritten loans against Salana against XRP. We would typically do a lower LTVs because we're

more those assets just have higher volatility and less liquidity in the market. And so what we're looking at um as we try and underwrite a new piece of collateral if if if a deal is proposed to us is we would look at um what's the what's the biggest draw down that assets had over the last couple of years uh over both a day and a week and that determines how long it's going to take us to exit the position if something goes wrong. That's very different for something like a cash or um you know polygon or pole I should say um or some other altcoin versus bitcoin um and then we'd look at we go and speak with the OTC desks and see

okay like how much size could you guys move in a day if we needed to sell because that's going to determine our risk level and what kind of slippage we would take and then lastly we also take collateral yield so if I lend to you and you give me Salana I'm staking it and I'm I'm collecting the collateral yield there. So, we have to look at how long does it take us to unbond it or unstake it. If we're going to do that and that determines how much of it we can stake. If there's a really good liquid staked token version of it that has a lot of liquidity, then we can stake more. If there's not, then we then we then we would stake less. But that impacts what LTV I'm going to offer the borrower and what rate. In the market today, Bitcoin

backed lending has compressed significantly in rate. So now a 70% LTV loan would fetch around 6%, 5 1/2 to 6%. Whereas in December that would have fetched 11 or 12%. So the rates can vary very dramatically. Um whereas if I were doing an altcoin backed loan, I'm probably still going to be getting north of 12% total return on that type of lending. So we've seen spreads between what you call like almost investment grade credit and and uh high yield credit shift quite significantly or widen significantly I should say.

Now that kind of like you you mainly do overcolaterized lending too. Um how do how would you say kind of um the Maple customer base kind of is differentiated from the other lending protocols that are out there? Yeah, it's it's um it's it's a good question. And I would say when we're doing more under collateralized lending, the the borrower base was primarily market makers that doesn't work anymore when we do overcolateralized lending because market makers the way to think of their business model is they have the lowest cost to shift between assets. So it's almost never a good idea for them to post collateral because if they have BTC and they want to borrow dollars, they would just sell the BTC for dollars and they're going to get less slippage than

than anyone else in the market. They can do it the most efficiently. So it would never make sense for them to post me the BTC and borrow the dollars. Um and so the the the mix on the borrower side has been it shifted more towards primes. So think you know some of these large businesses with kind of like multiple different lines. So you know there's folks out there in the market like a Galaxy or like a hidden road. This is a this is a prime type of business. they they do they provide margin to their clients and and then they do significant amounts of OTC trading but also onchain hedge funds and yield funds. So they might want to borrow from us so that against collateral that is that is actually forming part of a yield strategy they're doing on chain and this

could be on Pendle could be on Athena could be on um you know levels money it could be on any of these kind of stable coins but or onchain yield products but they borrow from us to lever those strategies. Um and then the other one would be liquid token funds. So they might have a position they have high conviction in. It's too expensive for them to finance the position using per and they don't want to risk a liquidation. So instead what they do is they could purchase the position, post it to us as collateral, borrow dollars and and lever up that position or invest in other deals that they want to. So um those are kind of the core constituents.

Increasingly what shifted this year is we've started taking in more family offices and high net worths who now you know they might want to purchase real estate using crypto holdings that they've accumulated over a number of years. And so they want from us what they're looking for is certainty on rates. Um making sure that they almost never are in danger of getting liquidated. So they'll post significant over collateralization and then counterparty risk. If you've got somebody who's got 200 million in ETH, they don't want to post it in an RV or in a maker and risk getting liquidated.

Somebody got liquidated for 100 million in in maker the other day. That person, if they'd done an OC style loan with somebody like us, they would have been able to have 12 hours to meet a margin call and we would have been, you know, alerting them that there's um that there's issues with their deposit. So that's that's the other reason that folks might be willing to pay pay a little bit more to deal with us rather than dealing with a pure smart contract. On the on the yield side, which is the other side of our of our product, folks putting cash with us to get a yield. It's high net worths and family offices are very early adopters. they're they're like a core constituent for us, but increasingly we're getting professional asset managers who might be managing money on behalf of a foundation and they

have a yield target they need to hit whilst being fairly conservative. And then also corporate treasuries are increasingly putting money with us because they view the Bitcoin um the Bitcoin and ETHbacked lending product is like akin to an onchain money market but paying you know paying a a reasonably higher yield. Quit. Yeah. No, that uh uh that was a very good overview. Um remind me what assets can be used as collateral and which ones can be borrowed. So you already named Bitcoin, Sulana, ETH, Polygon. Um is is there how long is the list?

Um it's actually not too long a list. Like I would say we have less than 12 different kinds of collateral on boarded at the moment or or that we're actively lending against. Um at the most conservative you have BTC which just by way of its market cap is generally going to be preferred um for parties and then as you go down the list you hit ETH Salana XRP all like extremely liquid and these are you know these are top five assets um so very comfortable with those but as you go further down the list some of the ones that we've underwritten have been you know P we did the PT token for SUSD so um Athena and that was because a hedge fund was running a a looping

strategy using Pendle and and Athena and um in our view it was effectively like the collateral was almost a zero coupon bond that just drags towards one over time. Um but even still we we still will spend two weeks going through due diligence on those assets. So our smart contract team will look at all of the code and and the audit reports. We will look at whether the assets are supported on the wallets that we use, whether it's Fireblocks or or Porto or Copper. And then we'll we'll speak with the OTC desks and see what kind of slippage we could get if um if we needed to liquidate at short notice. Um but the other assets that we've done, um uh as I mentioned, Pole, we've done Treehouse ETH, we've done a cash smaller exposure

there that that um that we closed out recently. And then um uh we've done uh Lombard BTC was another one that we we actually did in decent size. Um and uh and so we're I would say one of the if you look at how we compete, we don't have the scale of somebody like an ave. So what we have to compete on is we can take native BTC and other assets on other chains because we're integrated with custodians and Fireblocks and we we will also try and keep an open mind around underwriting new collateral. So somebody, you know, one of the ones we didn't uh underwrite or didn't approve was um USD0 plus. So we looked at that kind of twice and um and declined it

both times. But um you know, hedge funds and different onra onchain asset managers can always come to us with a proposal and we will look at it in in kind of good faith and see if we're comfortable underwriting it. But as I said, it's kind of like about 12 max different types of collateral at the moment. Okay. And um what I can borrow um this for is is it always fiat or do you also we we one of them one of the more interesting deals we did over the last year was uh we lent Salana and this was for firms to hedge um to hedge the locked soul deal. So effectively, you bought Locked Soul from the FTX estate at $64 and then if you could borrow it from us when Salana was $150, you would

take it, sell it, and then repay us as your locked soul vests through um through one of the custodians. And so we were hedged because we lent one unit of soul and we were expecting to get back one unit of soul and you'd pledged one, you know, locked unit of soul as collateral. Um but that was an interesting structure where we're happy to do um happy to do a collateralized loan in something other than stables and the yield for the lenders that they were getting was you know was like 6 or 7% wider than what the Salana staking rate was at the time. So it was like if you had a if you had a long-term bullish view on Soul it was a very good way to accumulate extra versus what you could get just purely staking. And the

firms for the firms who would participate in the lock deal they could hedge their exposure and for overall Salana holders it kind of smoothed the entry of those locked um of that lock soul into the market and prevented you know prevented like points of concentrated selling. Yeah absolutely. Um you already said that you're deployed on um Ethereum uh and you kind of also take um assets um from other networks from other things. Yeah. Um whi which uh which networks are you deployed on um kind of like I mean I mean I understand kind of bitcoin and so on but um kind of like any of the L2s or uh polygon or

um interestingly it's it's primarily Ethereum. In the past, we've been deployed on Salana and we're working to get back there. I think after 2022, the kind of the lending market on Salana sort of collapsed and um we've seen more recently kind of yield products evolving there, whether it's Camino, Jupiter, um uh Drift, um and then the other one we deployed on base, but we didn't get a lot of uptake there. We tried that in 2023 with the the T- bill product um when we're kind of exploring tokenized money markets. But it's interesting primarily all the liquidity kind of still resides on on um on Ethereum. And even if you look at other yield products, whether it's um Athena or others, they've bridged to other chains,

but almost all of their liquidity and almost all of their TVL is kind of still on on Ethereum. So we're we've tried to keep an open mind about going to other chains and we would like to get there but the biggest question we always come back to is like how much TVL are we going to get there for the resourcing required to support it and it's even if you've got an EVM chain it's not trivial to support it you have to create a new subgraph you have to organize all your data and then you have to create you know the web app that connects to all of that data so it's actually still pretty significant uh resourcing cost um for your engineering team but I would say then what the other ones that we are most kind of bullish on at the moment barring you know ETH1 would be bass has

a significant quantum of USDC salana as well um and those seem to be kind of standalone ecosystems that are kind of thriving despite not having you know significant mining or rewards programs unlike some of the other newer ones and then we also recently committed to be a launch partner on converge which is going to be the chain launched by uh Athena and uh and securitize cuz we're very interested we're very interested in kind of the bet there. Athena has been able to accumulate 6 billion in TVL. It's a bit under that now. And securitize was able to develop um you know strong relationships as the issuer of Bidd um Apollo's onchain credit fund. And so

they had they have good institutional relationships. Athena has shown a really good track record of actually securing TVL. And so the bet is will they port their TVL over to that chain. So I think that's a fairly safe bet that it's going to it's going to see good TVL traction whereas a lot of the other new ones you're really effectively taking a venture bet on whether that chain can can secure TVL or users and will it outlast the rewards program. Why do you think the TVL is so concentrated on uh Ethereum mainet? So in principle kind of like there's tons of activity going on on the other networks as well. Is it kind of by virtue of ticket size or people do they

have security concerns or is it just force of habit? What do you think is at play there? I think I think it's a few things. Um and it's it's probably kind of it's probably like a a combination of several of these factors. So I think it's partly the lending effect. So if you're an institution, it's easier to make the case to go to Ethereum which is which is battle tested over a number of years versus even a chain like Salana which is you know now several years old and has a very good um institutional presence. Um so I think that's one factor. There are network effects to liquidity and and liquidity begets more liquidity. Um the other one is I would say uh infrastructure. So, if you go to one of these chains, do they have a

stable coin um like a canonical stable coin like USDC or USDT rather than, you know, like a wrapped version? Um and then are they supported by like institutional wallets and custodians? That's another big factor. Um because if I'm, you know, if I'm a let's say a pension fund who wanted to get fixed income yield on a stable coin, I might not go to, you know, chain number 20 if they don't have USDC there and if it's not supported in in Fireblocks and if I can't custody that asset with Anchorage.

So I think that um that actually is like a fairly large barrier to entry for new chains. And then I'd say in the line of work we're doing with yield, people are less concerned about cost, particularly at scale because for yield products, it's not like trading or derivatives, you're not moving and and transacting every day. Often you might put in you make a deposit and then you don't look at it again for 6 months um until you need to withdraw from for some liquidity reason. So I think that's the other reason why people are not as averse to the cost of transacting on Ethereum or the the speed issues versus something like Salana.

Yeah, that makes total sense. Um what kind of Oracle solutions um do you rely on because I mean that's obviously really crucial part of your business, right? Um it is but funny funny enough the actual the onchain oracles are not really that important to us, right? because I have I have um the collateral sitting in Fireblocks or um Porto or custody account. So I'm I can actually use you know Coinbase's APIs or um you know like Coin Gecko as a backup. I'm less reliant on feeding that information into a smart contract because I keep I my my business model is like a hybrid where I have the loans are on chain but the collateral is held in these kind of

these CFI solutions. um if we shifted and did more kind of onchain liquidations which could be an interesting um business case for us then I would rely more on oracles but interestingly I've we've spoken to a number of Oracle solutions but it's actually not that integral to our business because the c because the collateral's in those other um you know those those kind of CFI solutions so we just use APIs typically that makes things much easier is there any appetite had to kind of um go back on the initial assumption um that kind of you would service um institutions first and foremost. So I mean so for

instance you could kind of um bring some sort of credit scoring on chain or u I mean is is this something that you're looking into because that would be that would be a huge market. No, I do I do think it's a large market. Um it's it's a it's it's a little bit of a different business like operating model like when when you start so you can you can either do it at scale where you have effectively no customer service and that would be the model of an ave or a compound or a spark which is you just do automated liquidations but you make your product effectively available to anyone who can post collateral and then the the CFI version of that is something more akin to Lein or to Arch. However, to do that business model, you have to spend a

lot on money transmitter licenses and and consumer lending licenses. And there's always the kind of the put option of um of just being sued by your customers, you know, like they they default. They they claim that they didn't know what they were getting into. And so you you actually need to start staffing up like a lot of customer service people and then you got to invest in, you know, in that kind of licensing. Whereas when you lend to institutions, it's largely unregulated barring a few states in the US because they're deemed to be sophisticated parties. So we've tried to go we've tried to say, okay, for a given number of hours of underwriting, we can actually earn much more in fees if we do a $10 million loan than if we do a $10,000 loan. And so it makes more sense

to for us to be a leaner operation and service those institutions. But once we start to exhaust that market, then we could look at going to more in more of a retail direction. The advantage in retail is you get better pricing power. Like if you look at players like Leen and Arch, they're out there lending at 14% to retail when for a given at a gi like today we would be lending at 6% on the same LTV to to a lot of institutions or you know maybe up to like 10% depending on depending on tenor but it prices you know very wide for retail.

Um, so I think the the answer is we we we're always kind of open to evaluating it, but it would require a significant kind of shift in in business operations and we do think that there's still a lot of juice to squeeze on the institutional side, which allows us to kind of just keep scaling by focusing on what we're doing. Like if you look at firms like Apollo and Aries, which I would say is kind of like our northstar of where we want to be in 10 years, they don't typically do kind of like retail lending. they just focus on lending to kind of larger institutions. And I would say that our our our angle of expansion is probably going to be as more assets get tokenized, we'll just start lending to more different kinds of institutions.

Yeah, that makes uh that makes sense. Um how do you guys um manage defaults? So even kind of like with the most prudent underwriting kind of like you must have had major borrower defaults especially kind of in turbulent market periods, right? Yeah, it's been interesting. I mean, so doing the overcolateralized lending, we started that business. We we launched that business model in September of 2023. In that time, we've we've had dozens of margin calls, but we've actually only ever had one partial liquidation, which was during the yen carry unwind in in 4th of August last year. So, we've had four significant periods of volatility.

What what interestingly links all of them together is those periods were kind of all Sunday nights. So if you look at it like the biggest most volatile single day periods are always a Sunday because like Asia comes online and the US is still like offline because it's Sunday and Europe's not active either. So you what happens is you get all these liquidations and margin calls come on from like 6:00 in the evening on the Sunday and then you you know me and like the entire team will be up till like 4:00 a.m. The most recent one we had was this last Sunday where as a result of the tariffs we had a bunch of margin calls but because this is because we are underwriting institutions and we'll also look at their balance sheet to make sure that they have the wherewithal to top up

their collateral. Typically, we just like our clients will just meet their margin calls and we only had one where we had to do a partial liquidation back in August, but even then we only sold like 20% of their collateral. So, that's the other reason they like us is if you do this on an Ave or a maker, you're fully liquidated. You're liquidated, you're wiped out, and they're taking a big commission. Whereas, we will only liquidate what we need to get back get you back to like a healthy ratio. and we'll be in contact with you for that period and and try and make sure that it's um that you know you're in the loop as to what's happening. But those periods, so Sundays are the worst for the liquidations. What happens is in a situation like that, just to step through that one, we had an initial LTV

with that borrower. So let's say it was 70. Um I think in that case actually wanted to go to 80. We did a margin call at let's say like 5% wider on LTV. So we contact them, we say, "Hey, we send them an email. We message them on Telegram. We are trying to get in direct contact with them. You have to post this much or you have to repay this much of the loan." Whilst that's happening, we're contacting the market makers and the OTC desks and we're trying to put out um requests for quotes on, you know, you need to liquidate 2 million bucks of BTC. And then um if the borrower posts, we we never fill the quote. If the borrower um does not or if we hit the liquidation level, then we're just

sending the BTC or the soul or the ETH to um to the OTC desk and and receiving back dollars which we use to pay down the exposure. The price alerts are automated. The the margin calls are automated, but just what's not automated is the actual the actual sale of the asset itself. And that's actually bug or feature. It tends to be viewed more as a feature by the institutional borrowers. they are deathly afraid of an auto liquidation from a smart contract. So it actually helps and it kind of helps us with like pricing power with these inst that that makes sense. Um so you're pretty relaxed kind of given the current macro backdrop. So kind of I would have assumed that kind of like um kind of tightening liquidity and tariffs and

shifting risk appetites and so kind of uh would keep you up at night but you're uh you're chill I'm I'm I'm I'm going a little bit more gray uh Ferto um but uh no it's it's interesting because we have the keep in mind we have the two sides of the business right so we have the yield um the yield that we're providing to to people who who lend with us and and then we have the borrowers who who borrow from us. And so um what we've seen is that borrowing rates have come down with the macro environment cuz Bitcoin is depressed. So funding rates and basis are depressed. There's just not as much to do with the assets. So folks are less interested in borrowing and if they want to borrow it's going to be at a lower

rate. Um like I said December we were lending Bitcoin 70% LTV at 12 to 13. Now it's down to like you know mid to high single digits. Um but uh what we are seeing on the lending side is that because Bitcoin is not you know because Bitcoin is not rapidly appreciating folks are no longer interested in market beta. So what does that mean? They're they're actually looking at holding more stables and they're very interested in putting those stables to work. So that business is is that side of the business is going very well. we're actually seeing constant inflows and um it's just that the rates that we can put the money to work at are kind of compressing and um and borrower appetite is kind of diminished. That can

always kind of flip on a dime though. You know, if if you you can have one solid week of movement in the market and suddenly suddenly, excuse me, everybody wants to borrow and um and the dynamic totally flips again. But I would say in this kind of environment that's a little bit sideways, we're actually very comfortable sitting as a yield product. As long as we control our risk underwriting and we're not extending too much credit on too loose terms, like this is a pretty good market for us. Folks are very interested in yield and we'll see a lot of growth. Yeah, that makes sense. How are you thinking about extending um the um assets kind of more the assets that you can um borrow on

more into the RWA direction? I've always I've always been a big fan of RWAS. I um I take the liberty of calling us tokenized private credit. I think private credit is the most interesting sector in Tradfi at the moment and one in one sense we are in fact you know we are in fact that right like we I do a loan onchain the loan is tokenized. I have an offchain lending agreement. That's basically tokenized private credit. And why I've been bullish on this type of tokenized private credit is it's the only one where you can't call up your Morgan Stanley banker and order it. Tokenized money market funds, tokenized stocks, tokenized bonds, any any, you know, you can just ring up Morgan Stanley and get the TRDFI version of that. Whereas there's no there's no

Wall Street banker selling exposure to Bitcoin back loans, which I think are a very interesting asset class. good yield versus the security and then you can get your money back generally within 30 days. So it's like it's actually much better than a lot of Trady products in in fixed income. Um but RWA so tokenized real estate tokenized receivables we have done lending against tokenized receivables in the past. I think that's interesting because it's high yield short duration um and so it it it turns over pretty quick. Um what we've seen though is that RWA have a buy side problem on chain and it's for the reason I mentioned where you know the the investors can just ring up Morgan

Stanley in most cases and get another version of that asset. So and the other thing is they have a much shorter time horizon on chain. So the onchain investors are thinking in 1 month to 6 months maybe 12 months at the at the wide end but um RWA investors or issuers are thinking in terms of 3 to 5 years and so that mismatch has been the key issue and why RWA have not taken off in in the way that people were hoping and that's kind of the main barrier like I we would happily extend credit against RWA if we can bring in capital from lenders and investors and sell them on that kind of you know 3 to 5 year exposure um and I think that's as the

space matures you will see more people come in who have a longer time horizon for you know for investing and um and also just if you get much more capital coming into the space so if stable coin if stable coin issuance goes up to a trillion dollars you'll have more capital come in and it will get crowded out of only lending against Bitcoin so I think if if the rates on Bitcoin back lending compress and rates on RWA back lending like receivables or real estate stay wider then more folks will um will push you know push into those exposures and we are agnostic like we would be there to underwrite that type of stuff for us the main issue is is there a market of people on the buy side who

want that and if if so like then then we'll find you know we'll find those opportunities and curate them for them so it sounds like you're working under the assumption that onchain and offchain markets will converge 100%. What's the What's the time horizon you think that will happen on? I'm terrible. I'm terrible at predictions, but I would say um I do believe that they will converge because um the natural pressure, right, is it it's going to happen through competition. And the fact is it's just undeniable that stable coins are a better way to transact and do pretty much anything in finance. Um, if I want to settle a loan, I can settle that loan at 1:00 a.m. on a Saturday on a Saturday night if I want. Um, because and we can

just send it directly to that customer's wallet. Whereas, if we were to settle a $10 million loan through the banking system, the bank might block the transfer. You've got to key in the details manually. So, that might get done. So, it's much harder than just the the the person giving you a single hash for a wallet address. And then it's and then you have crossber FX issues where if you want to fund a US dollar loan in Japan, it's very difficult and costly to do so. And so I think the inevitable trend will be it is cheaper, faster and a better user experience to to do this in stable coins. I think my bet is um that you start seeing more institutional lending and and lending against RWAs in the next 5 years and I would say by 10 years probably more than 50% of kind of

capital markets transacting or or you know kind of inst institutional lending will be done in in a in a in a stable coin in some form or another. So, I think I think it's slower than I would expect over the first five years, but between five and 10, it's probably going to be faster than most people expect. Cool. That that is actually a very um detailed uh prediction for someone who doesn't like predictions. Um I I like that. Tell me about your biggest challenge kind of from a technical um side kind of in building um a credit market on chaining.

I would say the biggest challenge we found so far um there there's a couple of different angles. So I'd say I'm always kind of thinking most about distribution. And so I'd say the biggest challenge has been on and off ramps um have been a huge one where we might meet somebody who's very keen on our product, but they're sitting on 10 million in fiat and it's a struggle for them to get on chain and if they're in New York, on-ramps won't touch them. Um they'll face limits from what they can on board through Coinbase.

So the on and off ramps I would say have been kind of the biggest probably the biggest barrier. Second to that is the prevalence of stable coins. Like I think at 150 or sorry, we're like 250 billion total stable coin issuance, we're still significantly under supplied. And if you think of that, it's kind of like that's the fish tank in which we swim, right? Like the the the stable coins are the oxygen um or if you're a fish, the water that you need. And as that expands, that gives us more ability to kind of do do commerce and and to transact. So I'd say those have been two of the bigger barriers. Um, and then the commercially I would say it's it's been buy side appetite, but technically I would technically I would say, you

know, being able to take funds from other chains has has been a bit of a a bit of a blocker as well. So, we would I would love for anyone to be able to deposit from any chain into a Maple Pool and receive the LP tokens. It would certainly help us with distribution, but um uh but that's that's like just not the case today. So that's something that we're currently working on with layer zero and um if that gets solved then I I think we'll be able to kind of scale more. Yeah, that makes sense. Other than kind of the crosschain issue um what are the technologies that you're excited about most kind of for shaping um the next wave of growth?

I' I'd be hopping on kind of the hype bandwagon, but I think AI is pretty interesting for us because we were in a we run a fairly lean team. So, we're 32 people. I have f currently five people on the um on our um sales and origination side. And I look at AI. I look at the potential for AI for monitoring risks. So in collateral giving me predictions or insights as to when we should be looking to liquidate a piece of collateral or when signs of impending volatility are coming. And then I also like it um for the ability to to underwrite new loans. So to assess like well here's you know here's the the four things you should be looking out

for with this borrower. Here's why this collateral is good. it it will be able to scan through a GitHub and uh smart contract audits faster than a human engineer can. And um so it should it should save time and allow us to do you know a much higher volume of loans at much better risk management than we are currently able to do doing everything manually. So I'd say that's that's probably like one piece of technology that I'm super excited about. Um and then you know we already kind of talked about the um the crosschain side of things. Yeah, that makes sense. So is the AI um analyst thing on your road map for this year? It is. We we actually

have a pilot in place at the moment. So we're working with a team uh smaller smaller team startup team out of uh with a couple of Australian founders um that we're pretty excited about. So hopefully we'll have an announcement on that coming soon. Um but that pilot is around assessing collateral and um we're going to have you know we're going to run a couple of test cases where we have you know our our human team analyze a piece of collateral and then we're going to have the AI assess the same piece of collateral and and kind of see where the um the points of difference are. But I think that's like something that we have in progress at the moment. And I think it's really easy call for me to make because there if it works there's like a very clear cost saving for us. Yeah,

100%. What's what else is on the road map for the coming year for you guys? So, one of the things we've been um particularly interested in is kind of like Bitcoin structured products. So, we have a partnership with um Core Dow, which is um a BTCL2 uh that does proof of stake. And so we um have a program where we can take native BTC in a custodian without slashing risk from um from lenders and investors and um produce for them a a a Bitcoin on Bitcoin yield. So I think that's been pretty interesting. And for us it's like it's an interesting hedge on the stable coins because when the market goes kind of more bullish folks are more interested in holding Bitcoin and if they can get a yield on that

Bitcoin it's it's an even more interesting product. And also, Bitcoin is kind of the dominant form of collateral that's posted by trading companies. So, I think there's a big market if we can turn that into an LST, which is one of our um product goals for this year to um to get them using that as yield bearing collateral. So, I think and and the other thing I just note is whenever I talk with Coinbase or some of these other big institutional OTC desks, the first point of call for all new institutions coming into the space is like they just want to buy Bitcoin. And so it's a very natural corlary. The the very natural next step for them is either can I get a yield on dollars lending against Bitcoin. They start to think about that. Um and then the other thing they will think about is can they

get a yield on the Bitcoin itself because otherwise it's just sitting there doing nothing. And usually because of their investment mandates, they have to keep the Bitcoin in custody. So having a yield product that's intrinsically linked to custody solutions I think is um is is um has like a really good edge and so I can see that product scaling quite a bit. So I'm I'm very excited about that segment of the market. Mhm. As someone kind of who's very in touch with traditional markets also but is clearly very cryptonative. um kind of as an ecosystem kind of we have often talked about the fact that um kind of when um trust in governments somehow falters kind of in in principle

that should kind of like be um uh in principle that should be um a boon for cryptocurrencies. But what we've seen so far in practice is that kind of like when traditional markets have gone down um cryptocurrency markets have also gone down, right? Do do you think um this is uh this is a trend that's going to break at some point or do do you think kind of will continue as we have in the last 10 years? It's a good question. So I think the Bitcoin trades like a growth asset because largely the price is actually being driven by inflows of new adoption. And so if you look um what would have to happen for

Bitcoin to trade as a as a sort of safer defensive asset is that either people need to rotate out of another asset class defensively or you just need to have people who are already in there who kind of refuse to sell as things are getting bad. But um the thing is that most people who are coming into Bitcoin, it's still in this kind of wave of of new adopters. And so the people who are coming in are those who are otherwise investing in like growth stocks. And so it trades largely as um it trades largely like a tech growth stock at the moment. You can see it you can see it in like the correlation with the NASDAQ and with stocks like Nvidia and and the MAG 7.

Um, I do think it breaks or changes over time, but that point has to be where Bitcoin is at a larger level of adoption worldwide and then the rest of, you know, the rest of the market would kind of follow that. Um but I think I feel like that's probably like five plus years away because the the the rate of adoption like it's still you know Bitcoin is still quite underpenetrated versus other um versus other asset classes you know it's still currently like if you look at the allocation crypto falls if you looked at a a pension fund they would have their alternatives allocation which might be max 20% of their book 15 to 20 and then within that bitcoin is going to be like

5 to 10% of that. So, it's still you're still talking kind of like low singledigit allocation and it's not until it's a larger um more more folks holding it as a percentage of the investable world and larger percentage of the of their portfolio that it's really going to flip and become a kind of much more defensive asset. Yeah, that that makes a lot of sense. So, Sid, where do we send people to kind of find out more about Maple? Uh, so you can find us um our website is maple.finance. finance and then on X we are Maple Finance and I am syrupsid. Um so if uh if anyone has any questions either go to the website and we have like a contact us form or um DM us on Twitter or or or

X um or or tag us on there. But we try and be pretty active. We do community calls usually once once every um two to three weeks. U but would love to hear from anyone who's interested in learning more about Maple. Cool. Super nice. Thank you so much for coming on today. No, thank you for having me, Fu Rica.