There's a Pipeline of ETF-Like Tokens About to Drop: Synthetix's Kain Warwick
The next one is a DeFi-based token basket.
|Aug 26, 2019|| 5|
Hello defiers! I spoke with Kain Warwich, founder of Synthetix, a platform for trading and minting synthetic digital assets. Last week, Synthetix launched its first index token, a digital assets that replicates holding a basket of tokens, in this case, a basket of exchange tokens, like Binance’s BNB and Huobi’s HT. Kain said that’s just the start. More index tokens are coming and the next one, to launch in two to three weeks, will replicate holding a basket of DeFi tokens.
He also talked about new leverage tokens launching soon, which will replicate going 2x and 3x long and short bitcoin, and about a major upgrade to the platform, which will introduce futures trading. Kain also said he believes Synthetix is just at 10% of where it needs to be and the steps he’s taking to get there, including an oracle upgrade. He also drops some advice to other DeFi platforms: Decentralization is a scale. Consider starting out keeping some control to be able to quickly iterate, before moving to full decentralization.
Image source: nulltx.com
The interview was edited for clarity and length and I bolded my favorite quotes.
Kain Warwick: I used to be in online retail, so I was kind of aware of Bitcoin for a while and then I started blueshyft, which is a payment gateway in Australia. We process cash deposits for a bunch of crypto exchanges. A bunch of Australian ones as well as Binance and a couple of other overseas exchanges. And so that's kind of how I initially got into the space, I guess more deeply. Then in 2016 we were working with a lot of exchanges and we saw something similar to what happened in Korea where the spreads on crypto, bitcoin and ether and other cryptocurrencies started to really creep up and there wasn't really any way to move funds quickly enough to arb those spreads and so I had an idea to build a stablecoin that was able to help to solve that problem for the crypto exchanges, then the user experience in Australia would be much better.
At the time really Tether was the only option as a stablecoin, but it wasn’t enough to solve the problem because a lot of people were concerned about its solvency and there was probably an issue of moving money from the normal banking system and Tether had a lot of issues with its banking relationships and so there was a view that if you had something that had less systemic risk, potentially more people would be willing to use it.
In the end it turned out that that was not really the market opportunity that it seemed, because eventually we had a regulated stablecoins like Paxos and Gemini, TrueSD and Circle pop up. So after we had launched Haven, which was the original name of the project, in April of 2018, by mid 2018 there were like 10 different regulated stablecoins. So the idea of using a decentralized stable coin was kind of marginalized. The use case wasn't as strong.
Camila Russo: So what did you do to adapt?
KW: We got to July of 2018 and it became pretty obvious that it was going to be hard to scale the network up, especially because the bear market was making everything much harder than it had been before and so we were finding it difficult to talk to exchanges about integrations because they were all busily integrating Gemini and Paxos, TrueUSD. So we said, okay, we have this mechanism which allows us to issue synthetic assets. The main one that we were going to issue was synthetic U.S. dollars. But we thought about launching other currencies, and so we decided let's push forward with that; launch a bunch of other currencies. And so in December of 2018 we launched seven different currencies and two commodities, gold and silver. And what we found once we launched was that no one was interested in the other currencies like GBP and CNY, but people were interested in gold. They wanted to hold and trade gold and we realized it wasn't so much about the utility. People wanted speculative assets, they wanted volatility. So it was kind of the opposite of what we had that we had set out to build.
CR: Can you explain how these synthetic assets work?
KW: We use crypto collateral. There's no real world assets. There's no custody, so no need to trust a custodian. The big difference is the mechanism for providing the value for the collateral is activity in the network, or exchange fees. So when people convert from one synthetic asset to another, when they reprice the debt in the system, they pay a fee for that service and those fees are paid to the token holders. And so that's where the token derives its value.
CR: And so it's not like Maker where you put up collateral to create an asset. It's the system fees what's creating that collateral?
KW: You still put up collateral. It's just instead of using ETH as collateral, we use a token which is tied to the value of exchanges in the network. So the SNX token, the value of that collateral comes from the activity in the network as opposed to just using ETH, which has its own intrinsic value, which is outside the system.
CR: So the collateral used in the system is SNX, and SNX gets value from the fees in the system. So do those fees have to be pretty high for it to sustain its value?
KW: Of course. Yeah. So we had a similar problem to any layer one which is trying to survive on transaction fees, you kind of need inflation. You need block rewards of some kind, some inflationary reward. And so when we launched the network, originally we had a fixed supply of 100 million tokens and in March of 2019, we decided to pretty radically change the token economics and move to an inflationary supply, where we paid the inflation to people who stake the network as a way of bootstrapping activity. And once we did that, that's when the network growth really started to take off because we had the ability to incentivize the behavior.
CR: Synthetix is not-for profit right?
KW: Yes, Synthetix is a not-for-profit foundation, so it's not-for-profit business,
CR: Do you plan to make Synthetix a for-profit at some point?
KW: No, not at all. The idea is that it's purely governed by token holders and there's no value accrual to any entity other than the token holders. So at the moment the foundation holds tokens but at some point in the future it'll just be a dispersed group of people holding tokens, providing collateral, and anyone will be able to join or leave the system and provide collateral and get the rewards from offering the service and staking.
CR: So the only profit you're making from like developing Synthetix is the eventual appreciation of SNX.
KW: That's right. Yeah.
CR: And so you found out that people really wanted to speculate, and that gold was the favorite asset to do that. So what were the next products that you decided to launch?
KW: We went for the most volatiles thing we could think of that had some liquidity, which was Bitcoin and so we launched the synthetic Bitcoin. But we not only launched a synthetic bitcoin, we also launched an inverse Bitcoin index as well, so you could have long and short exposure to Bitcoin. And when we launched that, the network growth accelerated even faster. It was probably 10x the activity that we had for gold.
CR: The inverse tokens are really interesting, just a straightforward way to short an asset. I don't know if I had seen that before. Do you have any competitors in that aspect?
KW: We don't have any direct competitors, but probably the closest, like, analog in the traditional finance system are direction ETFs. Those inverse ETFs that are rebalancing daily, that kind of target effectively a short exposure. So an ETF that you can hold, that's the closest mechanism. But in crypto space, not really, there isn't anything that’s similar.
CR: In general, the direction you've taken is pretty similar to ETFs. The inverse tokens and then these basket tokens. It's interesting, the idea of making the investment very simple, just holding one token, and you can recreate a more complex strategy. Is that the direction that you're going for?
KW: Yeah, absolutely. We want things that have differential utility, right? That you can't just get. People can short bitcoin, maybe not with one click, but you can short it, and you can certainly hold bitcoin and go long. So in order for us to really have some value for users that's differentiated, we need new assets. And so that's why we decided to launch the centralized exchange token index. Most of those tokens are ERC20, but, for example, the Binance token isn’t. So to try and hold those tokens, you would need to hold them you would have to have multiple exchange accounts on different chains. You wouldn't be able to do it non-custodially. It's a hard thing to replicate. And now you can just go to Ethereum and with a few clicks have access to that whole basket, or if you want, you can have the inverse access if you think it's overvalued, you can short it, which is even harder to do.
CR: So the basket of exchange tokens was your first token basket. Do you already have a pipeline of other basket tokens?
KW: Yeah. So people are interested in a few different categories. But one thing that we want to do as supporters of DeFi, is a DeFi token basket. We've got like eight different DeFi tokens that we're going to put into a basket as well. So if someone wants to just have one click access to the whole DeFi space, then they can get access to that. That hopefully will be launched in the next, like, two to three weeks. And obviously you will be able to short it too if you hate DeFi and you don’t think it’s real.
CR: Besides basket tokens, what else is coming up or is that the main thing that you're focusing on right now?
KW: Short term, that's the most immediate asset, but we are also looking at leverage tokens as well. So probably 2x, and 3x leverage BTC token, both inverse and long. Hopefully in the next two to three weeks we'll have that launch. Depending on what sort of response the market has, those will be the next couple of assets. And then we'll have a big upgrade of the platform where we'll support synthetic positions, which will be a decentralized futures market. So you'll be able to take direct long or short leveraged positions in maybe 20 or 30 different assets.
CR: And how is that different from, from the tokens that you’re already offering?
KW: So they actually won't be a token, it'll be a contract where you'll have a long or short leveraged position in that contract, but it won't be a tradable token.
CR: Got it, so like a regular futures market. Also wanted to ask you about your oracle. You had that oracle hack and it’s a topic of discussion in DeFi, how oracles can be a centralized point of failure. So what are your thoughts on that? I know you've talked about upgrading your oracle, where are you with that?
KW: So one of the teams that we've met a couple of times in Berlin in the ChainLink team. They've been doing a lot of work in the background to support the specific use cases that we have and there are many teams working on oracle solutions. So we're talking to a number of teams, but ChainLink is the most advanced one that we've been speaking to and I think there's a lot of interest in supporting us because it really validates their value proposition as an oracle team. If you can support our use case, which is pretty complex, then you can probably support almost any use case.
CR: Do all these different oracle solutions ease concerns about centralization?
KW: I don't think we have a perfect solution yet. There are trade offs with all of these solutions right now. But one of the nice things about DeFi and about the Ethereum ecosystem is composability. So it may be that it's some combination of different approaches that is sufficient. And ultimately, I guess what it comes down to is having sufficient security for the value of the assets that are being secured. If you've got a few million dollars worth of assets, you can potentially have an oracle that is less robust. But as you scale up and you've got hundreds of millions of dollars worth of assets, then you need a much more robust oracle systems that might need multiple fail safes and other mechanisms. So I think we won't really know what the solution looks like for another six to 12 months. And the other thing is also speed. We've got extremely decentralized oracle mechanisms like Augur, but they're not fast enough for the use case of trading. So there's tradeoffs between speed and decentralization as well.
CR: And the about speed, long term, do you see yourself competing directly with centralized exchanges? Will you be able to match them in speed?
KW: As someone who is extremely bullish on the future of decentralized services, I think all centralized services will be consumed eventually. These are technical problems. There's some fundamental, research issues that need to be resolved. But I think eventually we can get to a point where decentralized systems are more than capable of delivering outcomes that are equal to, if not better than centralized services, and particularly when you're talking about security of funds, and self sovereignty, and censorship resistance, centralized services can't compete with that. Ultimately you need to provide the best user experience as well and there's still a ways to go in that regard, but I'm pretty confident that we can catch up with the centralized services.
CR: In that sense, talking about user experience and catching up with centralized exchanges, what specific gaps are you seeing and what steps are you taking to breach them?
KW: About the gaps? All of them. There's so many gaps at the moment. There's issues on layer one, there's issues with layer two, there’s the oracle issues. My sense is that we are maybe 10% of the way there, in terms of Synthetix, in terms of where the platform could eventually get to. So the vast majority of the work is still ahead of us, and that includes UX, UI, onboarding, identity, basically every single hard problem in the space, Synthetix is a consumer of, and so we're actively talking to almost every project in the space that's working on these hard problems to try and work with them to understand how close they are and to be ready to integrate those solutions when they're live and accessible.
CR: Specifically on layer two, are you thinking of using any of those like scaling solutions?
KW: Yeah, so we've talked to Skale, and Connext, , which are both very different approaches, but we see them potentially solving different problems in our stack. Again, almost every project that you can think of we're talking to, to try and help us bridge the gap between the existing solution and where we want to get to.
CR: Also wanted to ask you about the difference between your basket tokens and Set Protocol, which is doing something similar. How do you differentiate with them?
KW: So interestingly, we're actually talking to Set Protocol as well, and one of the advantages that we do have as a purely synthetic derivatives platform is it's actually easier for Set Protocol to rebalance on our system than it is in the spot market. So we've spoken to them a couple of times now, and they're interested in potentially working with us to provide the same sort of rebalancing mechanisms. They've built the existing platform to support ERC20 tokens with liquidity pools that exist now, but it's something that hopefully in the future we'll be able to integrate. So I don't see us as competitors. I see them as augmenting the existing trading platform.
CR: So you mean they're looking to integrate with Synthetix to help rebalance their own trading strategy tokens?
KW: So the way that we see it possibly working is that the same rebalancing strategy, the algorithms that they've built, could actually be applied to a synthetic asset, so to a Synth.
CR: So for example, using their rebalancing algorithm for your gold token.
KW: Correct. You could use the same mechanism and apply it to as synthetic asset.
CR: Cool. You already touched on this, but want to learn more about your long-term vision, where you see Synthetix in like in 10 years. Do you expect to be doing more volume than Coinbase? What assets will people be trading?
KW: Yeah, it's interesting that if you look at a traditional markets and you look at spot markets for commodities versus futures, there's a lot more trading activity in the futures markets for gold than there is in the spot market. Significantly more. We see synthetic derivatives as additive to trading tools that people want to access, for hedging and for all kinds of reasons. But I think maybe more centralized services will struggle supporting equities, traditional equities indices, and other more traditional assets and making those assets accessible to anyone who's got an Ethereum wallet. That's not something that Coinbase will be able to do in the short term, probably. I see us as having a pretty strong competitive advantage there in the types of asset classes that we can support.
CR: What's your volume right now and number of users?
KW: We have about a thousand active users monthly on the exchange, and roughly 50 wallets –it's hard to say users, right?– so there's 50 wallets daily that are trading. And we're doing somewhere between $300,000 and $1 million a day. It's come down a little bit from June and July when we had some fairly active trading bots that were attacking the system and the volume was pretty high, but we've implemented some mechanisms to prevent front running. So that caused the volume to drop a little bit, but still doing about half a million dollars a day on average.
CR: So if it's 50 wallets trading half a million every day, they're trading pretty large amounts,
KW: Yes, and that's one of the things that I think differentiates us from other decentralized exchanges because within the platform, there's no liquidity restrictions. Once you're in the Synthetix exchange, a lot of the trades are $25,000 to $30,000 trades. You can trade much larger, block trades than you see on the average Dex where the order sizes might be $500 to $1,000.
CR: What do you mean there are no liquidity constraints?
KW: The liquidity constraint is on the on-ramp and off-ramp. We use Uniswap as one of the major on-ramps and off-ramps into the ecosystem. So think of it as when you're depositing to Coinbase, you need an on-ramp to get your fiat into crypto. Likewise, once you have purchased a Synth, let's say you bought $10,000 worth of synthetic USD, you can start trading on Synthetix, and you can reprice that synthetic USD into synthetic bitcoin without a counterparty. So there's no liquidity limitation. You could have $10,000, you could have $100,000, you could have $1 million, and the contracts would just allow you to reprice that debt as much as you have can be repriced. There's no limitation.
CR: Because the token is synthetically created and you don’t need to have someone to trade it with.
KW: That's it. So in that example, you have 10,000 worth of synthetic U.S. dollars, and that debt that you hold is tracking the U.S. dollar. And then you decide you want to reprice that debt and have it track the Australian dollar, hypothetically. So the system will burn the USD debt, and it will mint new AUD debt, so if you've got 10,000 USD, in that case, it'd be about 15,000 Australian. So if the Australian dollar drops in price against the U.S. dollar, and then you trade it back to USD, you might have 9,000 USD, , after you convert it back. But you can reprice that debt as much as you want. You don't need to find someone who wants to go from USD to AUD. The contract just does it for you.
CR: And then if you wanted more, then you'd have to go back to Uniswap and buy more tokens there.
KW: Yeah, that's right. So if you wanted to do a $50,000 trade, you need to go and buy another $40,000 worth of debt, but once you have that, you could start trading $50,000 at a time.
CR: Got it. So what are like some of the key lessons that you've learned since starting this platform that maybe other DeFi or Ethereum-based projects could learn from.
KW: I'm a startup person, right? I've been working in startups since the early 2000s. So my view is that, while we are trying to build decentralized structures, we also need to be able to iterate. And if we build things in a very sort of waterfall approach, where we have very long time horizons before the things that we build meet the market, there's a high risk that by the time the thing that you have built actually hits the market, the market may not respond in the way that you expect. The approach that we've taken is to be very iterative, and to run small experiments, play small bets, and see what the response is.
So we expected that euro was going to be much more interesting than gold and we were wrong, so we, we responded to that and said, okay, we need more volatility. How about bitcoin? And I think that there is a tradeoff between trying to be as decentralized as possible, and the utility that you can deliver, and how fast you can iterate and we've tried to find that balance. Eventually, if you want the system to scale, it needs to be fully decentralized because there's only so much risk that users will be willing to take from a funds perspective, if there are elements of centralization. My view is that you need to find that balance between decentralization and speed of iteration, with the view long-term view of eventually cutting old bridges and being as decentralized as possible.