Talk Series with AirSwap’s Michael Oved

 

Michael Oved, co-founder of AirSwap, met with Profit & Loss to discuss his vision for the potential of a decentralised trading platform based on the Ethereum blockchain. Previously an algo trader at Virtu Financial, Oved explains how he took lessons from the FX market and applied them to crypto trading to create a peer-to-peer platform.

Profit & Loss: AirSwap was designed around an Indexer, which acts as a search engine, matching up buyers and sellers, rather than using an order book model. With your background as an algo trader at Virtu Financial, did you look to the FX market for ideas when developing AirSwap?

Michael Oved: I view order books like a solved problem – that’s how most assets are essentially getting treated, except in the FX markets, which are a little bit different because they’re more of a peer-to-peer broker network – they’re communicating value through Swift, where the settlement layer happens. Blockchain itself is basically just a settlement layer – it’s a global ledger that you can just store transactions on. The whole concept of an order book was designed so that people trading assets like stocks, futures and bonds could come together to this one central location where they can find prices, execute trades, and then clear on some other mechanism and use whatever custody solution is appropriate. But in the case of FX – something that’s more decentralised – you don’t really need an order book to communicate value – you can communicate in this other mechanism, which is peer-to-peer.

If you look at how advanced the financial markets are in classic assets, there’s many different ways to trade. There’s auction mechanisms, there’s continuous, there’s discrete time execution, there’s peer-to-peer, there’s dark pools. If you look at crypto, there’s only one way and it’s order books. Everything is on order books – so what they’ve basically done is just taken the most active way to trade on Wall Street and kind of copied it for Blockchain. There are a couple problems with doing this in crypto. The first is that, in order to actually trade on these centralised exchanges, you store all your funds on the exchange. In the case of crypto, that has a lot of hacking risk. So essentially, the crypto exchanges have become banks – uninsured, unregulated, honeypots.

About a year-and-a-half ago, everyone was trying to move towards a decentralised way to trade. Since blockchain is essentially a settlement layer – the idea was to figure out how to communicate value on the settlement layer. But when you try to introduce an order book to that settlement layer, the problem is that the miners essentially can see the transactions before they get executed.

P&L: Which can lead to front running?

MO: Right. Let’s say you have an order book scenario and you have a big order come in – the miner can see that and decide to insert their own order in front of that. So inherently, once you start to introduce a centralised order book and try to turn it into a blockchain scenario, you introduce front running issues. So basically, the idea behind AirSwap – or the Swap protocol – was to introduce some mechanism for trading that was designed for blockchain – and it’s basically based off of the FX markets.

So instead of a central, public order book where everyone can see prices, with AirSwap, you join the network, connect to the Indexer (which is our discovery mechanism), and connect to all the peers on the network that want to send orders to you. They all send you orders and then you get to pick one of those orders and execute it, so it’s more like a local auction where you’re feeding out requests and receiving prices back and you pick out of that auction which one you like.

P&L: It sounds like the multi-contributor platforms in the FX world?

MO: Exactly. It’s very similar to an OTC FX platform.

P&L: Do you see fiat trading moving onto blockchain for this same reason?

MO: It can, and I think that will start to happen as the introduction of stable coins increases. Stable coins are a way to track the value of USD or track the value of euros by pegging a cryptoasset to a ‘stable’ asset like the US dollar or gold. The thing is, in order to interact with fiat, you need to have the on-ramp and off-ramp to a bank, which is doable and very useful, but the problem then is that it’s no longer decentralised at that point – although that’s not a huge problem.

If we’re talking about replacing Swift or something like that, you can represent an asset on the Ethereum blockchain that tracks USD, move it around the world and just make sure that you have the gateways in the different countries to convert it back from crypto to USD. So the rails are there and you can move everything much cheaper and much faster than through Swift. I’ve lost wires on Swift before – it can be a nightmare. You can replace all the rails definitely, but you need the end points to exist to actually convert it back to fiat.

P&L: AirSwap did a million dollars trading on the first day. How much is it averaging today?

MO: We did have a very strong first day. After that we dropped off, but we started to slowly grow again. I think we’re looking at around $100,000 to $200,000 a day currently.

P&L: Who is trading on AirSwap?

MO: So far, it’s mostly small sized trades, although some big trades are coming in. We’ve designed the platform such that, right now, there’s automated traders on one side and manual traders on the other side – so basically, it’s just automated flow coming through the platform. We’re eventually going to design a mechanism for institutions to communicate with each other. That’s the next version of the platform.

P&L: Are you seeing interest from the large nonbanks?

MO: We focused on medium sized firms at the beginning rather than the biggest traders. I see the large nonbank players getting in maybe 12 months from now, but I don’t think that many of them are trading ERC20 yet. We are designing our platform to service them when they’re ready – we’re definitely thinking about scalability. I’m trying to get all the big players involved.

P&L: That’s the approach Satoshi initially wanted to take with the rollout of Bitcoin, because it lets you develop the system at a slower pace with less intensity.

MO: Yeah, we needed time to be able to handle things like the message traffic.

P&L: What is the advantage to an established crypto trading firm that already has a sizable network for distributing liquidity?

MO: There’s a few different advantages. The first is that we’re going to automate a lot of their back-end process and also, they would potentially get the ability to trade with a lot more counterparties. If you think about AirSwap like an interbank tool, it’s another avenue for distribution.

P&L: Can you explain how the Oracle works to assess pricing?

MO: Basically, what an Oracle is in blockchain context is the way to take outside world information and put it into your system. We have a few different iterations that we’ve gone through in implementing it. Currently, the way the Oracle works is, it tells you when a price is bad – so it’s like a consumer protection tool. If you are getting a price that’s off market, it’s going to tell you it’s a bad price. Once you have three or four market makers on the platform, the consumer protection tool starts to become a little bit less necessary, which is where we are at now.  

P&L: Where does the Oracle get its prices from?

MO: People can decide how it pulls prices, but currently the two references we’re using are previous prices that have executed on the smart contracts – so kind of like trade prints – and prices that are coming from more centralised exchanges.

P&L: Which browsers can people use to access AirSwap and which wallets are compatible?

MO: Right now, MetaMask, Ledger and Trezor are the Web-based wallets, then we’re supported by a few different mobile wallets as well. So basically, the idea is that anything that can communicate and sign transactions on the Ethereum network we’re going to be compatible with. We’ve just prioritised getting the bigger ones in place, but we’ll support everything. I think we already do almost everything or we’re pretty close.

P&L: You launched with 25 tokens, has that changed?

MO: We launched with 25, but we’re up to 50 now and will be at 60 or 75 very soon. The market makers decide which tokens they want to trade.

P&L: How do you guard against malicious coins?

MO: We have a vetting process for tokens that’s been memorialised, so when someone adds a token, we can essentially just do a quick check on it about whether we need to blacklist it or allow it to be added to the UI.

P&L: So how does Air Swap make money?

MO: In October 2017, we sold memberships to the platform, which is the AirSwap token, and the proceeds are what we’re using to maintain the operation of the platform. It’s basically like a CME membership.

P&L: Do you see transaction fees in the future or you’re trying to avoid that?

MO: The essence of peer-to-peer trading is peer-to-peer, right? It’s on two people trading with each other and once you introduce a third party in that system, extracting rent, it’s not in our eyes a peer-to-peer trade. So we’re trying to introduce as least friction as possible in the AirSwap platform.

We could theoretically build other business models on top of it. If the goal is for this to become a foundational layer that people are building on top of, and in that case, if you think about Google, they don’t charge anything to search and discover something, right? That’s kind of what AirSwap is. It’s just a search and discovery platform – it just connects peers. So if that search and discovery mechanism becomes widely used, you can have different services, for example, prioritising certain counterparties over others through some sort of advertisement, a model like a bidding model.

P&L: What’s the plan for growing users?

MO: The token’s been a good incentive – having a community behind the project is very powerful. We also have an integration strategy that involves a widget. Any project that wants to launch, has launched a native token to be used in that ecosystem. If someone wants to go into a Dapp [decentralised app] and needs the native token, we basically have a widget that’ll pop up while you’re still in their ecosystem so you can buy it. So it’s just a way to seamlessly buy native tokens without having to go to centralised exchanges, deposit money, trade, withdraw your money, plus it’s just a much safer way to trade.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s