Not getting crypto exposure or borrowing money to buy things in the real world.
|Aug 13||Public post|| 3|
Good morning defiers! I spoke with Varun Deshpande, co-founder of Nuo Network, a decentralized lending platform which consistently shows up as offering the best interest rates for lenders on LoanScan. But that comes with a caveat: While rates are fixed for borrowers, they’re volatile for lenders. That’s created both confusion and a loyal user base.
Varun explains his platform’s unique model, talks about why he thinks DeFi’s biggest use case is enabling people to earn interest, not getting them to borrow for non-speculative use cases. He also goes head on against the idea that ETH is money.
Image source: LinkedIn
The interview has been edited for length and clarity and I’ve bolded what I think are the best quotes.
Varun Deshpande: My partners and I wanted to create a digital bank after selling our credit analytics platform, and we needed a bank partner, so we went to the banks and we were roaming around with every bank to get their access to their APIs but no one gave us access. As soon as we started building on top of Ethereum and using crypto assets, we didn't need anyone's permission to create and deploy these products. People appreciated it. People started using it and that's how we got that traction. So just the ideological reason behind permisionless is fascinating, but the fact that we actually experienced it is even better.
Camila Russo: Tell me more about the specifics behind Nuo, what were the main things you wanted to achieve with the platform?
VD: The idea is to connect lenders and borrowers from around the world, which means that instead of being dependent on the local conditions or in the local interest rates, you create a global crypto economy, where lenders from developed markets like Japan or the U.S. and actually lend to borrowers in Thailand or in India or Brazil where the interest rates are super high. Lenders from developed markets can earn higher interest on their deposits, while the rate for developing nations’ users will be lower.
We started off with the peer to peer loans platform that we launched around eight to nine months back. You had to essentially put up collateral and create a loan and then someone lends and fulfils that particular loan. The first thing that people experience was the fact that they had to actually install MetaMask, they had to do like 10 things before they could actually, you know, start using the products.
We completely revamped it in such a way that any normal average user who's not been exposed to the idea of crypto, should also be able to use it. And that's how we started building out Nuo, the version that you see currently. We ensured that as soon as a user requests for a loan, they automatically get it, thanks to a pool-based mechanism, with a fixed interest rate. On the lenders’ side, when you deposit money in a bank all you care about earning interest daily or weekly or monthly.
We heavily used two things to improve UX. One is instead of being dependent on MetaMask, when you create a new account, a smart contract wallet is created and you can transfer money from MetaMask, or Binance, or any other wallets, or in any exchange that you're using. The second thing is, the whole idea of gas is complicated for users to understand and that's why we meta-transactions extensively on the product. It's basically a very magical experience of putting in money from your Coinbase account into Nuo, which is just a couple of clicks and that money starts earning interest.
CR: What differences are there with, with the other lending platforms?
VD: Dharma is completely peer to peer, while we have a pool-based system. Compared to Compound, their rates are variable, based on the utilization and demand and supply. But we have a fixed interest rate market for borrowers. The user asks for the duration of the loan, and when the user pays it, we distribute to all the lenders. It doesn’t matter if they pay the loan ahead of time, they still have to pay the full amount. On Compound, when the loan is issued, that's when the interest distribution starts, on a per block, daily basis. But we wait until the loan gets repaired and then distribute the interest. Essentially lenders are incentivized to be a part of the pools for as long as possible, which helps us reduce interest rates for the borrowers.
Also our mechanism of liquidating collateral is different from Compound, Dharma or dYdX. We liquidate the collateral by sending it to Uniswap and Kyber through smart contracts in a decentralized way. On Compound or dYdX, you can see all the loans which are under liquidation and then some market maker comes and gets that 5% or 10% fee for liquidating that particular collateral. We don't rely on market makers for liquidations. We just use decentralized exchanges for that. The majority of the time what Compound and Dharma do is, because there's not enough market makers, they must be using their own funds to liquidate and that is, we believe, not a very scalable model. We've tried to decentralized that part completely. We never raised any outside funding and we didn't have enough tokens pooled with us to help liquidate our own collateral.
CR: I noticed on LoanScan noise, Nuo is the only one that has kind of an asterix next to it saying, Nuo's borrow interest rates are estimates only and may significantly vary from actual interest rate paid or earned. Wondering why it's the only one with that kind of warning.
VD: That's because of our model I mentioned. For example, when you put money into a lending pool, based on the loans which are going to get repaid on that particular day, we'll distribute that interest rate. But a lot of times what happens is that because we also do margin trades, if the market is suddenly going up, or suddenly going down, a lot of shorts or longs are hit. When that happens, the loans get paid, you know, much earlier than expected, and that increases the interest drastically for the lender.
So basically because we fixed the interest rate for the borrower, that means the lending experience is highly volatile for the lender, in terms of some days the interest rate will be very high, and sometimes it's going to be much lower. And that's the reason these interest rates that we mentioned on the platform are essentially estimates and it may vary based on market conditions and how these loans get repaid eventually.
CR: How are you finding people react to this kind of risk of not knowing exactly what interest rate they'll be getting?
VD: Initially there was some issues with that. I think the more you understand the model and the more lenders actually start using it, the more they love it and we have a very loyal user base. Basically the best part about this particular models is that market makers can use the platform in a very interesting way wherein they can analyze when these loans are supposed to get repaid.
Like for example, if a lot of loans are supposed to be repaid in let's say one week. So what they will do is they'll put a large amount into the pool reserve of let's say Dai or USDC, based on the pool where the money is getting repaid, and increase their stake in the pool. So maybe initially their sake was like 5%, they'll suddenly push a lot of money in and now their stake has gone up to like 50%. So when these loans get repaid in the next one week, they'll get 50% of all the interest which is going to get generated. So it's a completely a gamified system, which actually lenders who spend a lot of time on the platform understand.
A user who comes expecting a Compound-like experience might get disappointed. But what happens is when a new user comes and puts money, that means that, the reserve pool size increases, but that reduces the interest rate and that attracts a lot of other borrowers. But then these lenders are not satisfied with their returns and they just probably go away. But these loans when they get repaid, that entire interest, which was supposed to go to all these different lenders actually goes to the ones who are patient and stay invested in the platform for the long term.
CR: Wanted to ask you about the, the recent changes you made to the platform. Can you walk me through them?
VD: Definitely. We were one of the first platforms to introduce margin trading in a decentralized way. And the way it works is that as you put up collateral you borrow with as much as 3x leverage. And because we were using Uniswap and Kyber for the first time, we didn't expect some issues, which eventually happened, which was when the market crashes, the problem with Ethereum is that it gets completely congested. So that's the reason that with margin calls, it takes a lot of time to get confirmation and by the time you do, the value will have dropped drastically. And this happened a few times especially with BAT and REP, where there was a lot of price slippage on Uniswap and Kyber because the volume itself in general is much lower compared with a centralized exchange. This happened a few times for these tokens and we realized that we have to do two things. One, we have to increase the buffer, essentially the collateral ratio for these tokens because there's not enough liquidity for them.
And second, we have to also launch our own insurance fund so that in the case of a flash crash we can insure whatever money is lost by the lenders. This insurance pool is created essentially by charging the lenders 5% to 10% of all the interest that they're generating on the platform so that if something like this happens that amount is used to repay lenders who might have lost money.
We have also introduced lower limits for large orders because slippage depends on order size. So we made all these changes to safeguard lenders.
CR: I've noticed on LoanScan that you're able to, to offer very tight lending and borrow rates. Tighter than other platforms and sometimes the lending rate is even higher than the borrow rate. How can that happen?
VD: Again, it goes back to the difference in our model. Say you have a loan and you need to close it out because you get a margin call, or you repay ahead of time. Because we have fixed rates, it doesn't matter when you pay, you have to pay the entire amount. So if you are reaping the loan in one week or one month or three months, it doesn't matter. So for example if you were given a 12% interest for three months, but you end up paying it in one month, that means the lender got paid three times as much.
Image source: LoanScan
So the lending rates is a function of when the loan was repaired, how the distribution happened to different lenders, and it can be much higher based on when the borrower is repaying because the rate will be fixed. I think that gets a lot of people confused.
CR: I noticed that on DeFi Pulse the amount of Dai on Nuo suddenly dropped. Why do you think that happened?
VD: I'm not sure exactly what might be the reason, but I've seen USDC has been increasing these past few months and Dai has been declining and. I think the sentiment is moving towards stablecoins which are fiat-backed, instead of Dai which is based on its CDP mechanism.
I think also a lot of Dai was moving to Compound because a lot of CDPs are going to Compound. InstaDApp recently launched a feature so that they could allow people to move their CDPs from Maker to Compound.
CR: Why do you think there’s more interest in USDC?
VD: When people talk about Dai specifically and CDPs, every Dai is actually worth 1.2 Dai, because you have to earn 20% of the fees to pay for the Dai that you've minted. Dai isn't free so you have to ensure that you're doing something with that Dai. So if the CDP's fee is 20, 25%, now you have to ensure that you actually earn that much amount so that you can pay off that loan and still make some profits. So what lenders end up doing is that people lock their Dai and earn some interest, but they also go long on ether, and do other things, so that you can earn more than the CDP fees.
Right now on Compound, you're earning like 14 percentage on Dai, but that is not going to be enough of you want to repay the CDP tomorrow because then you essentially lost money. So people will stop creating CDPs eventually. In a bull market Dai takes a huge hit because people want to sell Dai and hold ETH, that's why Dai value keeps going down and the stability fee keeps going up. So that's why many traders prefer fiat-backed USDC and USDT.
CR: I wanted to ask you about the new use cases that you said you wanted to introduce. In what timeline do you think you'll add them and what's the next one that you think you'll see in your platform?
VD: Currently we are revamping our existing margin trading product to make the user experience much better. For other use cases we're working on the version two of the product. So that is going to go live probably in the next few months and we will be announcing in the version two of the product very soon. That is going to be drastically different than the existing product. The whole idea behind the version two is essentially, all these issues that lenders and borrowers might've faced from the platform we want to solve that and also take a step further in terms of how we can take DeFi to the masses and get some retail adoption.
The biggest criticism for DeFi, and that includes me as well, is the fact that they are focused a lot on AUM, how much money is actually locked in DeFi, compared to actually how many people are using these finance applications. Compound I think has not more than 5,000 users, the number of people who have created a CDP was less than 3,000 or 4,000 before the Coinbase campaign. DeFi doesn't have more than 10,000 users. That is something that you want to change so that we can not just focus on how much money or how much tokens are locked onto these DeFi platforms, but how many people are actually genuinely using these platforms.
You don't want to focus on the hedge fund guy who's putting in $1 million. Rather on the user from around the world who's putting in $10, $100 and actually earning interest. We believe earning interest is the biggest use case for DeFi. Less people are going to be excited about converting their fiat into bitcoin or ether because of the price volatility. If you want to go to a mass mass market and target an Uber driver or a shop keeper who actually wants to use crypto eventually, I think this is the right use case. They're storing that money into a stable coin asset and earning higher interest.
There are less than, you know, 10,000 users and it is completely driven by large hodlers and whales pumping money into these platforms. DeFi platforms are essentially taking money from each other. These platforms are essentially currently not growing beyond a certain point. They are just moving money from one platform to the other. The lenders and borrowers are probably making the most of it by getting better rates but we are really moving people around and not taking this any further. So there's a fundamental shift that needs to happen to create a platform which is ready for market adoption.
CR: How do you hope to attract, these are retail users?
VD: A few ways. 60% of all our users are from Asia and we will continue to expand our services and acquired users in Asia. The second is providing a Web2 experience even if you are using decentralized lending or borrowing. Cracking the UX extremely important. That's why we create all the products in house so that we are able to control the user experience really well. Third is ensuring we have the right partnerships in terms of on-ramps. We announced a Wyre partnership in the U.S. so anyone with a U.S. debit card can get money into their account in a completely seamless way without understanding crypto.
CR: How many users do you have right now?
VD: We have 4,000 users with active accounts, who have actually funded these accounts. We limit the amounts that can be used, so growth has been steady and it's not a large market maker coming in, dumping a few million dollars into the platform. We’re completely driven by retail users.
CR: One big criticism for DeFi is that these collateralized loans are being used to basically continue speculating on the market. Like you said, you need to gain a high return to be able to repay your CDP. so it ends up being very niche and specific to traders and speculators and not many users are taking out loans to, you know, buy a car or do things in the real world. Do you still need to make that leap?
VD: I think that criticism is actually kind of flawed. We don't expect borrowers to use the platform for real world applications. There might be some, but the real-world use case is actually earning interest. So the user acquisition and retail adoption is actually going to happen on the lending side and not on the borrowing side. The real use case is that instead of me getting 1% or 0% on my bank account, I'm earning 10%. And that is the biggest real-world use case possible, compared to giving out uncollateralized or under-collateralized loans to buy cars. I don't believe that is the genuine use case. The genuine use cases is interest. Earning interest.
CR: That’s a very interesting point. In terms of your business model, like how, how are you making money?
VD: We are not currently charging our customers That is something that we plan to do in the new version. We are also a very unique in terms of DeFi because of the fact that we are one of the only DeFi products based out of Asia, in Singapore jurisdiction, we will be approaching the regulators to get us as regulated in Singapore itself. And that can happen in the next three to six months. And once we get some clarity on that, and that is when it would probably be the right time to start charging customers.
CR: When you start charging, where will those costs be?
VD: They'll be mainly on the lending side. The fees will be baked into the interest rate itself.
CR: We talked about your biggest criticisms to DeFi. What are you most excited about?
VD: One thing that I'm super excited about is the fact that we certainly have with ETH a programmable asset which can create very interesting financial products. You can put up some collateral in the form of ETH and create very different opportunities and financial products. The whole criticism about collateralization is again not warranted because that collateralizing is exactly what needs to happen, what was needed to create financial assets. That is something which is very powerful.
Something else we’re excited about is bitcoin actually. Some of the people in the community are actually very critical of of bitcoin, which is kind of sad as well. Bitcoin found a very good product market market fit in terms of being digital gold. It’s 10x better than gold and it can disrupt that $7 trillion market. Ethereum has yet to find that product market fit.
CR: So what do you think Ethereum can represent in the world?
VD: I was reading the column Ryan did in your newsletter he talked about how ETH is money, and maybe this is controversial but I think that’s probably the laziest description of what is formally ETH. I do not believe ETH is money at all. For me, it’s a different programmable paradigm. The fact that you can create applications on top of it, which were never possible before. The fact that we can create these trustless financial applications is so fascinating. ETH is bad at many things, but the fact that it can create financial applications which were never possible before, I think that’s very untreated. Ethereum as money is just probably going to attract regulators. Instead we should look at Ethereum as a programmable store of value, as a programmable asset, as a new programmable paradigm. That is something which is exciting for me. It’s almost like an internet in itself.