Unqualified Opinions: Bitcoin Lending with Zac Prince of BlockFi

Zac Prince, founder and CEO of BlockFi, joins host Ryan Selkis to discuss Bitcoin lending and the future of BlockFi.

Unqualified Opinions S2E1 Zac Prince Bitcoin Lending Blockfi.png

Ryan Selkis
All right everyone. This is Ryan Selkis and you're listening to Messari's Unqualified Opinions. Each week I interview cryptos' top builders, investors and personalities and today I'm speaking with Zac Prince, CEO and co-founder of lending platform BlockFi. Should be a good one so let's dive right in.

Ryan Selkis
This podcast is presented by BlockWorks Group, one of the top blockchain events and media production companies I've worked with for exclusive content and events that could help you with insight into the crypto and blockchain space. Check them out at blockworksgroup.io and you will not be disappointed.

Ryan Selkis
All right everyone. Welcome back. I am Ryan Selkis @twobitidiots and this is hosted by Messari, Messari Crypto. Want to thank our sponsors, TokenSoft and TokenTax. You might be wondering why? Why the plug for TokenTax still? Well, tax day is over, but if you had to file an extension, it's never too late to do the right thing and stay out of jail by paying your taxes. Thank you again to TokenTax and TokenSoft.

Ryan Selkis
Today, I am very excited to have a conversation that has been on everybody's mind right now around crypto lending, how to actually earn interest on these otherwise dead assets. And today I'm joined by Zac Prince who's the co-founder and CEO of BlockFi here in New York. We're off to already an amazing start because he and I are wearing exactly the same shoes. Impeccably dressed

Zac Prince
Or basic.

Ryan Selkis
Or basic depending on, depending on your perspective and generally on the right path and working in a really hot sector of the market that I'm sure a lot of people are going to have questions about. Zach, why don't we start with the origin story. Your background, how you came to form BlockFi and a little bit about the basics of the product we're going to go pretty deep into but just the one-on-one to start.

Zac Prince
Yeah, sure. I've always worked in venture backed technology companies. I was originally in the advertising technology sector and part of two companies there that were acquired. One by Google, one by a Dunnhumby, which is actually where David from Flipside Crypto used to work. And he was a part of the company that acquired the company that I worked at and we've now, recrossed paths.

Ryan Selkis
Interesting.

Zac Prince
I know you had him on a podcast recently too. And then more recently and more relevantly for BlockFi, been in the fintech world, specifically the online lending side of fintech at two different companies. One that was a middleware layer for the online lending sector, that aggregated data and we had some technology services. we built a few regulated entities and we worked with lending club, SoFi, Prosper, all the largest online lenders and then institutional investors who were either buying those loans or lending to those platforms.

Zac Prince
And then I also worked at a consumer lender that integrated with retailers at the point of sale to finance large ticket purchases for people that had low FICO scores or no FICO score because they were international, for example. I started investing personally in crypto in early 2015. I whiffed it, I bought bitcoin around 300 and sold it at 600 it's not a complete whiff, but it was a whiff based on...

Ryan Selkis
In the grand scheme of things.

Zac Prince
Based on the price that I had to buy bitcoin back at. And I just went progressively further and further down the rabbit hole throughout 2015 and 2016. and at a certain point my wife said, "You're talking about this way too much. You should go find some other people to talk to about this because I don't want to talk about it this much." And in early 2017 after getting more involved in the community and seeing what was going on, I just decided I had to do something full-time.

Zac Prince
Given that I had been working in the online lending sector, debt and credit was kind of the first thing that popped into my mind. And so the original idea for BlockFi was to build debt and credit products for the crypto market. That vision has expanded a little bit since then, but it's where we started. And so in 2018 we brought our first product to market, which is the ability to get a US dollar loan secured by the value of your cryptocurrency.

Zac Prince
I actually had a funny experience with a bank in 2017 where I listed bitcoin and ether on my financial statement that I submitted to them and they not only didn't value it, but they freaked out a little bit. I thought that being able to lend against these assets was something that needed to happen and a logical place to start.

Zac Prince
2018, we raised 60 million across debt and equity. We were the first company to get an institutional credit facility to support crypto backed loans. And then at the beginning of this year we launched our second product, which is the interest account. Providing the ability for crypto owners to earn interest on their crypto in crypto. And that interest that we're providing to depositors is created by institutional lending activities that we're conducting, where we're facing large financial institutions that want to borrow crypto for various reasons.

Zac Prince
And then longer term, what we're planning to do is continue on these things, but build out a diversified suite of products that fall under a wealth management or retail banking umbrella. You'll see us launching a additional functionality around wealth management for your crypto assets. We're planning on launching a credit card where you can earn bitcoin rewards instead of airline miles or cash back. And then longer term, we think there's some really interesting opportunities with dollar backed coins to offer just normal banking services, SAM crypto to parts of the world where they haven't necessarily been available before in dollars.

Ryan Selkis
One or two things. Let's take those piece by piece though. You say 60 million was raised. For starters, we should break that out because SoFi, a number of other peer-to-peer lenders, you see the big numbers. You're still series A company, right?

Zac Prince
Correct.

Ryan Selkis
The 60 million is, have you disclosed how much of that is the credit facility?

Zac Prince
Yeah, 50 is the credit facility.

Ryan Selkis
And then 10 was the equity slot?

Zac Prince
Correct.

Ryan Selkis
Okay. And so that 50 million, what's the duration on that? Because they're, and the lending markets within crypto, it's basically overnight rates. There's perpetuals on BitMax, which is the lion share of the market. There's perpetual lending on Bitfinex. There are only a few OTC desks that now offer lending services that might have different tenders for these loans. But how do you think about that 50 and and the gradual development of this lending market? Because going from overnight into structured debt instruments, two very different things. And how have you thought about that and tackled it?

Zac Prince
Sure. That facility in particular, is committed for two years to purchase loans that meet the eligibility criteria of the facility. Every US dollar loan that we've made at BlockFi since inception, we write to a one year term, but our borrowers have the ability to prepay without a penalty at any point along the way. I think that over time we will extend that initial term. We elected not to so far because prices are coming down rapidly. The cost of that first facility that BlockFi has, and in general, the cost to borrow against crypto assets has come down since we started. When we first started making loans, we were lending it at 14% interest rate. We now make loans and interest rates as low as 4.5% so we, we will extend the duration in the future. But at the outset it wasn't smart to extend a two, three, four, five year loan and lock someone into a price that was inevitably going to come down.

Ryan Selkis
Yeah. And it's come down rapidly. And so, there is, I think more and more talk about coming up with reference rates for what's actually happening in the lending market. In part because the early opacity, but now just because there's more competition. You were able to lend at 14% because there wasn't anyone else in town, right?

Zac Prince
That's right.

Ryan Selkis
And so that's going to come into some type of equilibrium. How do you think about pricing risk in this market in particular? Because the existing credit markets are pretty healthy, generally speaking, and interest rates are low. You can envision a scenario where many early lenders get into trouble by mis-pricing risk and end up blowing up. And with the spread coming down between what you're able to do in day one versus what's available now, that risk management becomes much more important and the target customer here that's using cryptos collateral by definition has out sized risk exposure maybe versus a traditional borrower. What tools do you have? Are you working with third parties to assess that individually? Is that part of the special sauce?

Zac Prince
There's a couple of different things to segment there in terms of how to think about that risk. There's on the US dollar lending side, when we're lending dollars to largely retail or smaller corporate, secured by their crypto, there are a few things that we need to check from an underwriting perspective. They're not FICO, score related, but it's a very low risk type of lending. Basically, the only thing you need to be worried about is, do they have this bitcoin because they stole it from someone and therefore it's not theirs? And, are you able to properly source liquidity in the event that you need it? If prices are falling and you need to sell some of their crypto because they're not responding to a margin call. The risk management on that side, our system has performed perfectly since we started lending and it's pretty clear cut.

Zac Prince
Additionally, when you're lending crypto, if you're getting collateral above the value of the crypto that you're lending, it's the same risk management but just in reverse. Now, if you want to start developing rate curves for bitcoin, there's kind of two ways that you can do it. One, is you can look at the futures market and if you go back maybe five or six years, there used to be what was called a gold forward rate and it was derived from the implied borrow cost in the gold futures market. Now right now, the bitcoin futures market activity is concentrated in the in the front month. There's no, there's very limited duration.

Ryan Selkis
And it's basically just CME.

Zac Prince
In space. Well, there's CME and there's BitMEX and bitFlyer and others. Dramatically different counterparty risks in terms of, CME versus anyone else that's trading crypto futures right now on their platform. Kraken also has futures available.

Ryan Selkis
You're talking about the rate curve. Do they have longer dated debt instruments? Or lending products?

Zac Prince
You can look at the futures price relative to the current spot price. And when futures are in backwardation, you can calculate a rate by comparing the one month to the three month to the six month to the 12 month. Now the CME I believe, goes out to a year, but there's no volume. All the volume is concentrated in the front month.

Zac Prince
That's one way you can build a rate curve.

Ryan Selkis
That's all anybody thinks about in crypto anyway. You can make or lose money in the next 30 days.

Zac Prince
Yeah, so that's one way you can think about it. Another way that you could create a rate curve is if you can build up a large number of credit worthy borrowers who are willing to borrow in crypto. They're willing to borrow bitcoin, have a liability in bitcoin. And Arthur Hayes was actually talking about this recently on another podcast and I think he's kind of leaning that direction in terms of where the bitcoin rate curve might come from. I'm leaning a bit more towards the futures market angle right now, just in terms of my guess of where we'll start to see one be derived from.

Zac Prince
But what that will look like if you're doing it unsecured, is you have, BitMEX for example, or Coinbase or other kind of logical crypto borrowers, putting a bond of sorts out there at a certain duration and paying a rate on it. And then you could start to see a rate, a rate curve based on credit quality. You could see a rate curve for a triple A type borrower versus a rate curve for maybe a less credit worthy borrower, again in bitcoin.

Zac Prince
And so in terms of how BlockFi handles the risk management and making sure that we're not one of the groups that you think will inevitably blow up, you have to be sophisticated at managing both types of risk. You have to be sophisticated at managing the collateralized risk. Having access to liquidity and a system that's able to source that liquidity very quickly if and when you need it. And then you also need to have the ability to underwrite counterparty risk from a credit perspective. And so we have people on our team, most notably our chief risk officer who was previously a managing director in prime brokerage at Bamel in our director of operations who built out the sec lending desk at Guggenheim, that have built a counterparty risk credit framework for BlockFi. And right now, in any situation where we're lending to someone and they're not posting collateral above the value of the crypto that they're borrowing from us, we're insanely conservative. We're more conservative than...

Ryan Selkis
You can be too, right? Because BlockFi's just developing and there is still, I think, probably going to be some rate compression. And as people figure out the the risk management component of this.

Zac Prince
That's right.

Ryan Selkis
Before we get too much further, I'm finance degree and worked in venture capital, we just threw a lot of terms out that may be I'm sure are not as well known because folks in crypto are not necessarily as financial savvy in terms of financial engineering or the debt markets in particular. Easy one, duration. Just length to the loan or more or less. The rate curves are you're talking about are just the delta between rates for an overnight borrow versus 30 days versus 90s, so on. Define backwardation.

Zac Prince
Backwardation is when the price to buy bitcoin in the futures market is cheaper than the current spot price. If for example, bitcoin is trading at 5,000 on Coinbase on the USD pair, but you can buy a April bitcoin future for 4,900, then that would be a market that's in backwardation. The way that you can calculate an implied yield from that is by taking that $100 and amortizing it over a month relative to the $5,000 current price of bitcoin and then you have an implied yield.

Zac Prince
Right now, if you can...

Ryan Selkis
And the important thing there is that's not necessarily an indication of an unhealthy market. This happens in traditional debt markets.

Zac Prince
Commodities as a general rule trade in backwardation. And that especially makes sense when there is a rate on the dollar because there's a carrying cost that you're foregoing by not just sitting in cash. You would expect bitcoin if the use case for bitcoin moves more over time towards the digital gold commodity angle and not the money angle, you would expect that all other things equal, it would trade in backwardation in the futures market like other commodities like gold and silver and others.

Ryan Selkis
The last one we talked about just to help people understand the difference between what you are offering versus what a BitMEX is offering, correct me if I'm wrong, but the majority of BitMEX's business is perpetuals, right?

Zac Prince
Correct.

Ryan Selkis
They sell every eight hours, I believe.

Zac Prince
I think it's, yes, I think it's six.

Ryan Selkis
Is it six narrow?

Zac Prince
I could be wrong. it's six or eight, I'm not sure.

Ryan Selkis
Oh Man. I thought, I thought I just looked at it. I thought it eight, maybe it's six.

Zac Prince
I could be wrong. I haven't traded on BitMEX in a while.

Ryan Selkis
Can you explain the difference between the perpetual design? They might be taking different risks in terms of unsecured borrowers and they're handling that a little bit differently, but in particular, the perpetual design versus the BlockFi model.

Zac Prince
Sure. The biggest difference is that BlockFi is lending physical bitcoin when we're lending it to someone. We're moving bitcoin over the bitcoin blockchain, sending it to someone else, whereas on BitMEX's system, it's all closed and they're able to, via perpetual swaps, facilitate, leverage both long and short. You can get more exposure to market movements than the actual equity that you've put up. And all of the equity is denominated in bitcoin because that's the only currency you can put in and out on BitMEX. There's also a lot more leverage available on BitMEX than any amount of leverage that working with BlockFi could ever facilitate just based on the ratios and how...

Ryan Selkis
And so this brings up callable and and non-callable. If you get margin called on BitMEX because the price moves against you, then your position could get liquidated. In terms of the loans that you are facilitating, are they callable in that regard or non callable? What's the segmentation right now?

Zac Prince
When we're lending USD, there is a defined structure for margin calls and then max loan to value ratios. And there's different things that happen in each one of those ratios. At the margin call ratio, what happens is our borrowers have 72 hours to either post more collateral as security or pay down their loan or take no action. And if they take no action and at the end of 72 hours the price is not recovered, we will sell some of their collateral to rebalance their LTV to a healthy ratio.

Zac Prince
Additionally, if during that 72 hour margin call window, the price falls further and the max LTV ratio is hit, our system will automatically sell enough of the crypto that's been posted as collateral to bring their LTV back down to the margin call level. You've got one thing that happens at the margin call and then another thing that happens at a more accelerated LTV, and if that accelerated LTV is hit, you go back to the margin call level. And our risk management system just does that 24/7. we have a similar function on the crypto lending side when we're lending crypto secured by dollars, except that our system, rather than selling bitcoin to rebalance the LTV is buying bitcoin when bitcoin prices are moving up. But it's the exact same design as the exact same type of mechanism. So yes.

Ryan Selkis
There's a couple of important things here to unpack. One, where are you getting the reference data to trigger these events? Well, maybe I'll start with the second question, which is, the market has not really moved against you. You started at the company in 20, early 2018, correct?

Zac Prince
Mm-hmm. Summer of 2017 we started. Made our first loan, made our first USD loan January 2018.

Ryan Selkis
Okay. The majority of the loans that you facilitated outside of maybe the first year or so testing, which will probably yield some pretty interesting insights with this question, but the majority of the loans that you've facilitated have been originated while the market is moving back in the right direction, positive territory. You would expect your borrowers that have posted collateral in crypto have done well and have not really been subject to margin calls yet because everything has stayed above water. When the market turns south, have you seen, have you gotten to any evidence in terms of how that's going to impact the borrowers and default rates? And actually trigger your risk management system to start liquidating some of these positions? Because I remember, previous lending products in the last bear cycle were not nearly as sophisticated, but it was still difficult to figure out why someone would, how that system would work at scale if there was a price correction of say 10% a week. And some folks were levered too highly and began to face margin calls.

Zac Prince
Sure. There's a lot that goes into, on a layer above those LTV ratio levels, just looking at total exposure relative to liquidity at certain price bands that basically acts as a, as a maximum exposure that we're willing to have. We have kicked the tires on our risk management system because when we started lending in January of 2018 we were lending USD secured by crypto. In that type of loan, your risk management system gets its tires kicked when crypto prices are falling and we, we're down a fair amount in 2018. the most, the largest scale event that we experienced was actually during Thanksgiving on the 6k to 3k bitcoin drop. I think there was a 24 hour period there where it was down well over 10% and our system performed as designed. Every time that max LTV was hit, we were able to get liquidity and sell crypto.

Zac Prince
But one of the things that we learned coming out of that, was that flexibility is important, not necessarily in the design and implementation of the system but in what you do afterwards. We had a couple of clients who hit the margin call level around 7:00 PM on Thanksgiving night and the max LTV level around 6:00 AM the day after Thanksgiving and you look back at that and you go, well that, that just doesn't feel like an appropriate amount of time to be able to respond to a margin call. And one of the benefits of operating the company in the way that we are is that we were able to make some relationship based decisions in terms of how we would handle that. It's not a smart contract where everything is a hard line in terms of what you have to do after your risk management system takes certain actions.

Zac Prince
And as a result we were able to do things for some of our clients that wouldn't be an option if it were a decentralized or smart contract only platform or even an exchange that doesn't have customer service that's available to speak with you. On the crypto lending sides, when we're lending crypto and it's secured by dollars, where our risk management system tires are kicked when the market is going up, we have also tested that a little bit because we started scaling that side of the operation alongside the launch of the interest account in January. And we had a fun little 20% pop pretty recently, which was great.

Zac Prince
We've kicked the tires on both. There's a layer above those different cutoff thresholds that you have to always keep in mind, which is your max exposure relative to liquidity availability in the market over time. And we've been happy with the results so far. We've never had a late payment, we've never had a loss, we've never had anything go even remotely close to wrong.

Ryan Selkis
So far.

Zac Prince
So far.

Ryan Selkis
You touched on, well first, who are you using for a reference rate? Do you have your own reference rate or which index provider are you using?

Zac Prince
I think we use two now. We used to have only, we used to have only one. Now we used two. I don't want to say the name because I might say it wrong, but I can tell you what it is later. I didn't pick them.

Ryan Selkis
Got It. Okay.

Zac Prince
But we switched from so we...

Ryan Selkis
We'll catch up on that offline.

Zac Prince
We evolved from a, we evolved from building our own, which is what we're doing on day one and it was a combination of three exchanges with the USD pairs, to one index provider plus what we had built to now two index providers. And the key thing there is...

Ryan Selkis
Just for redundancies.

Zac Prince
Exactly, exactly.

Ryan Selkis
How do you think about the decentralized lending market? We touched on this a little bit in your last response. There's a human discretion element that's probably pretty important when you're trying not to come up with system wide outages or catastrophic, black swan events that are just triggered by smart contracts. You would expect the decentralized lending rate to be higher long term than centralized, correct?

Zac Prince
I would say it depends. It's unclear right now what to expect. I think that centralized systems will have the ability to price things above or below where an actual market would be if they want to. Which can be a valuable tool to have if you're thinking about delivering a better or worse rate to a certain type of customer that is more or less valuable to you for some reason. We're still so early in the evolution of debt and credit markets for crypto and certainly in the evolution of the decentralized lending platforms that it's hard to say. And the most kind of top of mind example for me right now at least is MakerDAO who, when I originally learned about that platform, I was like, well, how the hell are we going to compete with this decentralized credit facility that's lending a dollar that they're printing out of thin air and charging 50 basis points a year for it?

Zac Prince
And I spent three days. I was just like, we might just need to shut this thing down. It's not going to work. But what BlockFi is doing in terms of our strategy is that we don't, we don't believe that lending in a silo is enough to build a company as big as the company that we strive to become. As a result, we are aggressively creating new products and the next two products that we're creating don't fall under a bucket that a decentralized lending platform would even be able to create because you need to have compliance, you need to have certain licenses to be able to offer them. You need to have very strong banking relationships. And so that's how we're planning on differentiating ourselves.

Zac Prince
But we do think of DeFi as a potential channel for us to generate yield from the pool of crypto that we have on deposit with us. We have not done any lending on DeFi to date, but we're monitoring it very, very closely. We're certainly testing it personally all the time and overtime as it scales and is more and more tested. We'll potentially be an active participant in those markets as well.

Ryan Selkis
Just really depends on liquidity as well. I think the Di ecosystem, it's still what? 75 million Di. Something like that. You have both retail and institutional customers and users. Both on the lending and borrowing side. Is there a split either direction? Is it pretty mixed? How do you think about getting retail money in? And how that juxtaposes with the institutional business, because you've seen some of the quote unquote peer-to-peer lenders over time, they ended up just being excellent business funnels for the major investment banks. At first it was, oh, I'm Joe Sixpack, I'm going to lend this money out to be this platform. I'm going to make a nice yield. And over time it was like, no, the banks just took their risk teams and found all the gems in these peer to peer platforms and then they gobbled it all up for themselves.

Zac Prince
Or just built one themselves. We're just basically Goldman and Marcus.

Ryan Selkis
Yeah, exactly. Exactly. Have you seen that similar dynamic materializing with your book of business? Or is it skewed? Is it pretty evenly balanced? What's the skew?

Zac Prince
We think of, we don't think of institutions as being our client in the same way as retail is our client. We work with them, we have a relationship with them that necessitates a high degree of communication and service from BlockFi. But if you look at a map of our products, and I'm happy to share this if we have show notes or something, but we basically have products that we create for retail, which is who we believe is our core customer. And then activity that we conduct with institutional markets to facilitate the delivery of those products to retail. To stick with the lending club or peer to peer lending example, what that market learned if you were someone that was building a platform to lend, was that having capital in a peer to peer construct was wildly less efficient than just being able to raise hundreds of millions or billions of dollars in one shot from an institution and you could get a lower cost of capital by working with institutions.

Zac Prince
As soon as institutional capital started coming into that space, the model completely evolved away from being peer-to-peer. And we started with that mindset. We felt that our USD borrowers are going to be largely retail or smaller corporate, and then our capital should come from institutions. That's how we think about it. We don't really, there's not a number of 50% of our clients are institutional and 50% are retail. We think all of our clients are retail. We bucket some smaller corporate for maybe really smaller funds into a retail category. Call it some, 20 million AUM type of size. And then we work with retail to facilitate the, or we work with institutional rather sorry, to facilitate the delivery of the products that we're creating with our core retail customer in mind.

Ryan Selkis
You mentioned a few additional products that you're going to be rolling out in the coming quarters years. Where, what's next on the product roadmap? You've had a busy 2019 already, anything coming up in Q2? Q3? What's next?

Zac Prince
Yeah, a couple things. In Q2 we're doing a lot of, kind of feature releases and optimization of our current user experience largely around security. 2FA, white listing, the ability to list a beneficiary in your account, those types of things. And then at the end of Q2 we're going to have our third party API available so that other companies in the ecosystem could connect to BlockFi, offer access to the interest account or the loan products via their environment, whether that's a mall at or maybe an exchange in an emerging market.

Zac Prince
And then in Q3 we're going to be launching a fiat on an off ramp for buying and selling crypto. And we think there's a couple of kind of novel angles that we can bring to that. Most notably the ability to create a US dollar income stream from your crypto holdings. Earning interest on your bitcoin in bitcoin is great, but some of our users have expressed an interest in taking that bitcoin that was paid to them as interest and turning it into dollars and sending it to their bank account. We're going to have that coming out in Q3 and then the credit card with bitcoin rewards is slotted for Q1. We're going to keep working hard.

Ryan Selkis
A lot of good stuff coming out. Zac from BlockFi. You can check him out @therealblockfi, right?

Zac Prince
That's right.

Ryan Selkis
Not to be confused with the fake BlockFi. And the website is?

Zac Prince
Blockfi.com.

Ryan Selkis
It's a dotcom. They sprung for the dotcom. We could learn something from you man. All right guys, this is, this has been fun. We'll have more in the show notes and a I'm sure we'll do an update again soon, but in the meantime, thanks for tuning in. We will catch you soon. Til next one, peace.

Zac Prince
Thanks for having me man.

Ryan Selkis
That's a wrap. Thanks for listening. New episodes of Unqualified Opinions go live weekdays at noon eastern time. You can follow me in the meantime on Twitter @twobitidiot if you want to continue the conversation or troll me. Otherwise, I'll see you next week.