In Traditional Financial Circles This Would be Considered "Nonsense"
Says Kyle Kistner, who wants to bring fixed rate, indefinite term loans to DeFi
|Sep 18, 2019|| 1|
Loans, with fixed interest, that you can pay back whenever (as long as you keep paying interest and you top up your collateral if it loses value). Sounds too good to be true, if you can lock in low rates.
Lending protocols bZx and Dharma recently said they plan to offer just this, indefinite term, fixed rate loans. bZx is building a platform specifically for these types of loans called Torque, which was slated to launch Sept. 14 but is delayed. It’s being tested in a private version right now, and you can get in touch with the team to try it. I spoke with bZx co-founder Kyle Kistner to understand how something like this could work. Dharma, which announced indefinite term, fixed rate loans as part of their version two (still in beta), said it’s not ready to disclose details.
In a fixed rate loan model, the obvious question is what happens to borrowers when market rates decline and they’re stuck with a more expensive loan, and what happens to lenders when market rates rise, and they’re stuck earning worse returns than at other platforms.
In bZx’s model, if interest rates decline, borrowers can just close out their loans, Kyle said. The platform will then have to adjust and offer more competitive rates to attract borrowers again. If interest rates increase, there will be more liquidity from lenders, which will go into a pool to lend to borrowers who got lower rates and also new borrowers who will have to pay higher rates. That disparity between borrowers will mean lenders will get lower interest payments than if the platform offered variable rates. The end goal though, is to resolve this problem with interest rate swaps, but the industry isn’t there yet, Kyle said.
Image source: bZx Medium post
It will be an interesting experiment to watch. The interview has been edited for brevity and clarity and I’ve bolded my favorite quotes.
Kyle Kistner: We're starting off offering about 16 percent fixed rate, but once our reserve utilization [the percentage of the system’s reserves that are lent out] goes above 80 percent, then that rate starts scaling up with demand. When reserves are at 90 percent, it starts scaling up even faster. Between 90 and a hundred, it actually gets to 100% APR to dissuade people from taking all the liquidity out.
Lenders are going to want to flood in if their rates get too high, relieving the existing lenders with liquidity. Also loans are tokenized so they can sell them on the secondary market, hopefully not at a discount to pay a liquidity premium, given that our rate model scales like that.
But we see the big drawback of what we've done is if the market rate goes down. If market rates sink, what will happen is that everyone will flood out of our platform and go to the other platforms that are offering better rates. So lenders are not really jeopardized because they’re getting higher rates, and borrowers are in a pretty good position because they can leave, but we ourselves as a protocol, we are taking on some amount of risk by where we set these parameters. And that risk is that people don't use the protocol.
Camila Russo: So what happens if rates go down, and everyone leaves?
KK: If borrowers leave there will be a problem with lender liquidity and borrower liquidity and at that point, people will probably take neither side of the market and then we'll have a chance to tune our interest rate parameters downwards, more in line with the market. But we're going to have to wait for basically most people to abandon their positions, go to other protocols, and then we can readjust the parameters.
CR: If rates are lower and it's not attractive for borrowers anymore, can they just close out their loans or would they have to pay like their full loan and 16 percent?
KK: Interest rates are paid per block pro-rata. So if you've only taken out your loan for a minute, you only pay a minute worth of interest. That interest is continuously being paid as the loan is open. So if you want to close your loan, close it and you can leave right away. There's nothing stopping you from leaving.
CR: In the inverse, if rates are higher, then it's not attractive for borrowers but attractive for lenders. How would your model work if rates went a lot higher?
KK: Our interest rate model is very responsive in the upwards direction. So if say our reserve is 90-something percent full, interest rates might be up to 25 or 30 percent. So the reserves are very able to deal with interest rates going up and it's attractive for lenders. One drawback is that because there are some people who got a lower interest rate that causes the spread to be wider, which can make the protocol less competitive in some ways. But we try to mitigate that by having by having rates targeted at a high percent of utilization so that we're already getting competitive spreads before this effect takes over.
CR: So if rates do increase and you already have people paying 16 percent fixed rates, there's not much you can do, right?
KK: There's not much we can do. Those people, they got lucky, they got a 16 percent. As a result, the spreads are a little bit wider. Lenders’ money is in a pool, much like Compound and dYdX. Lenders are compensated by the interest borrowers pay. If the market clearing interest rate goes up, and there are borrowers at a lower rate that have secured a loan, they still pay their lower rate. However, new borrowers pay a higher rate. The undesirable consequence of this is that the spread between lending and borrowing rates is wider than it would be than if borrowers were all paying the same variable rate. Another way of looking at it is, the lending rate would end up being lower than it could have been if all borrowers were paying a variable rate.
With these reserve models at a hundred percent utilization, usually there's no spread. The lending rate and the borrowing rate are equal. With ours, if we got to a hundred percent utilization, the lending and borrowing rate would not be equal, but that's okay.
CR: How does your fixed rate model compare with other protocols? What are you doing differently?
KK: I believe Nuo does like 60 days fixed rate. When I asked Dharma on their channel they said they're acting as a central counterparty. It's not really clear how they're going to hedge their risk, but they're just going to go ahead and guarantee people certain interest rates. They're going to basically set the lending rate higher and take a premium for themselves and set the borrowing rates higher and take the premium for themselves. I think the hope is that they set the premium high enough to be self sustaining, maybe capture some profit. That might be part of their business model. But Dharma is very nimble. So what they’re thinking one minute might not be what they do down the line.
CR: What about Alice finance? Aren’t they also doing indefinite term fixed rate loans? I haven't heard about many people using them.
KK: Alice Finance is not an Ethereum project is one thing. They're on a Loom side chain and whereas Loom was maybe doing Plasma, now they're just their own independent, delegated proof-of-stake side chain that is trying to connect with EOS, Libra, Ethereum. I suspect part of the reason they're not brought up that much is because in some real ways, they're not really connected to the rest of the ecosystem. There's no composability with Alice, nobody's going to be building on Alice or integrating Alice.
CR: Was just checking and it looks like they're doing fixed rate, not indefinite term. So I guess you guys and Dharma are the only ones doing indefinite term, right?
KK: I think the catch on ours on why it can be indefinite term is people will leave on their own. Assuming market rates go down, eventually people are going to terminate it. We don't need to enforce that on our end. Right now what we're doing with Torque we think is workable. We think it fills this very, sorely needed spot in the market. We've been talking about it for a while. We don't like how rates on Compound and Maker move from under us and we didn't like how the old Dharma made us settle up in three months. We wanted the best of both worlds, and we could see that the rest of the market wanted it too.
But what we would like to do ultimately is interest rate swaps, proper interest rate swaps. We see this being the initial go-to-market release but at some point we would like to get people on board with doing this with interest rate swaps. It's probably the more elegant way to do it. You don't have a drawback of if market rates fall, then you have an exodus. We don't really want an exodus.
CR: When do you think you'll be able to do this?
KK: I think we need to have a governance procedure in place that gives people who have loans a voice. There's no point in saying it's indefinite term if you're going to switch the model up months later. We want to have a governance procedure in place and we'd like to see some rate swap protocols actually be online and audited.
We're hoping to have governance online, maybe by the end of the year, though that that might be too ambitious. Maybe near the beginning of next year, we'd like to have really decentralized governance. Once we do, we'll turn it over to them and I think they'd be open to moving to proper rate swaps.
CR: That's interesting. And how about an interest rate swap protocol, how much would that take to be up and running?
KK: There's been some pretty viable rate swaps that have been hacked together just in a few days during a hackathon. Kind of piecing together existing pieces. Whether those are really viable approaches that will be liquid, I think time will tell. From what I can tell LSDai they're using Market Protocol and that's relying on order books. I think that there needs to be a Compound or a Uniswap style rate swap. We need an automated market maker that can algorithmically adjust based on supply and demand to bootstrap liquidity because order books have traditionally been very difficult to bootstrap. I suspect that LSDai, if they went on the market, the Market Protocol order book for it might just stay empty unless, you'd have to hire market makers and then where are they going to hedge? They're going to tell you they can't hedge anywhere. So I think that it's very possible the existing implementations as they are, are not going to really take off liquidity wise. So I think we're probably going to see some more approaches.
In the last few months before ETHBerlin, I said maybe we need to put interest rate swaps on our roadmap. It seems like it's really needed. But when I went to Berlin, with the plot of trying to get people to do interest rate swaps, everyone was already doing it. It was sufficiently in the air at that point. I've been surprised at how people in the industry arrived at very similar thoughts and conclusions and moves.
CR: Speaking of liquidity, can you tell me the number of users and volume you’re seeing on your platform?
KK: In our first month we were top 10 Dapp on Kyber. We did $490,000 worth of volume. Our second month we were the second-highest Dapp by volume on Kyber. We did $1.4 million, or about 5,400 ETH. This last month we were the number two Dapp on Kyber now. The other Dapps and the trading side specifically have been falling in volume quite a bit. I think lending has been very healthy, but trading specifically has become kind of anemic, but we still had growth. We went from like 5,400 ETH to about 8,000 ETH, that's about $1.6 million in volume. Just as a benchmark Uniswap in their second month, they did about a $1 million.
CR: Will be interesting to see how these loans work out once you launch.
KK: We definitely view it somewhat as being an experiment. We think, we think it will work and, and we hope we're correct about it. But yeah, I think certainly in traditional financial circles, the notion of indefinite term and fixed rate is probably considered nonsense. Maybe things are different in DeFi.
Sign up to get the best and only daily newsletter focusing on decentralized finance news, complete with analysis, exclusive interviews, scoops, and a weekly recap. All content is free for now. Those who become paid subscribers now, before the paywall goes up, get a big discount :)
About the author: I’m Camila Russo, a financial journalist writing a book on Ethereum with Harper Collins. I was previously at Bloomberg News in New York, Madrid and Buenos Aires covering markets. I’ve extensively covered crypto and finance, and now I’m diving into DeFi, the intersection of the two.