Crypto Derivatives Explained

Mark:

Welcome to WTF Crypto, where we peel back the layers of the onion of the crypto universe to understand what's really going on and how it affects you and your portfolio. I'm your host, Mark Lurie. And as a caveat, nothing in this podcast is legal for investing advice. Today, we're talking about derivatives with Barney Mannerings of Vega Protocol. Welcome Barney, thank you so much for joining us.

Barney:

Thanks for having me, Mark.

Mark:

And Barney, you told me before we started that trading volume for derivatives in the traditional markets are an order of magnitude more than trading everyday spot prices. But in crypto, derivatives trading volume are an order of magnitude less than spot volume for any given asset. That's really a big difference. And I wonder what's holding it back and what impact that has on the ecosystem. So today I'm really excited to find out more. I'd love to understand what makes you such a credible expert and guide on this subject so our audience knows where your thoughts are coming from.

Barney:

Awesome. Yeah. So I'm a technologist at heart and computer scientist. But I spent most of my career up to five, six years ago in this traditional market. So I worked as a consultant in financial services. I worked with lots of tier one investment banks. I also worked with the London Stock Exchange for two or three years on various iterations of their matching engine and trading platform. I worked every stage of the trade life cycle from risk side of things, to trading and matching into clearing and settlement. And spent a lot of time around traders and trading firms and understanding what makes them tick and how the markets are structured and work. I got into Bitcoin reasonably early, so 2013, 2012. Did some mining, got into Ethereum in the presale and then ended up by another startup creating Vega, which is a decentralized derivatives protocol.

Barney:

So, I've seen a lot of the traditional world, I've seen how derivatives are used and what they're for and how they power the financial markets and how they allow risk to be moved around and shared and mitigated and managed. And I've seen how integral they are to well functioning markets. And I think hopefully I can bring some of that experience to bear on the crypto markets and the journey, I guess, that crypto is taking from a fledgling asset class with things like Bitcoin to an integral part of the global financial markets.

Mark:

Awesome. Thank you. And do you mind giving people a little bit of context for Vega Protocol?

Barney:

Absolutely. So, as I mentioned, I'd spent all that time working with traditional markets and trading firms and seen lots and lots of things, which I thought were, I guess, suboptimal, you have a small number of centralized organizations controlling the markets. Most of them are in one of four or five cities on earth, not near where the people who potentially need those products for trading are. And the barriers to entry are super high. And there's actually a lot of regulatory capture as well. So not only are the barriers to entry high for natural monopolistic reasons, maybe they're also high because of the huge cost of [inaudible 00:03:31] compliance and the fragmented nature of global markets. And what I realized was that the same revolution that the internet had brought to bear on dissemination of information and on journalism and all those kind of related topics was about to be unleashed on the world of finance through the invention of cryptocurrency, and particularly the ability to have a decentralized system of tracking and moving money around and value.

Barney:

And I realized that, that allowed a very different way of looking at the markets. So instead of a centralized entity who are gatekeepers and enablers of trading and who you have to have their permission to do any trading, which is a relatively new invention, really simp the invention of electronic trading and financial markets in the last century. Instead of that, you can go back to more the kind of traditional invention of cash as an innovation and where money and value and trading are fundamentally peer to peer activities and ones that anyone can innovate on and get involved with. And I could see that coming along and the value that doing that, but also doing so in this online global virtual marketplace would bring in the same way that kind of internet did for information.

Barney:

And that basically caused me to go deep down the rabbit hole of Bitcoin and then Ethereum, and then think about Ethereum decentralized exchanges and decentralized derivatives realized that there were a lot of things you have to get right to build a successful trading platform and protocol. And that in my mind, we, weren't going to be able to do any of those on this sort of the current generation of block chains like Ethereum and with that generation of Dexus. And so we set out in 2018 to work out if it would be possible to build a decentralized exchange for derivatives that could actually compete meaningfully with the traditional finance market. And then here we are.

Mark:

Makes sense. Okay. So we're going to dig more into that, but let's start from the beginning. This is really relevant for everyday traders because derivatives are traded so much, but I think it would be useful to understand why derivatives are traded so much and why they're so important to the centralized markets and thus why it's so relevant to have them in crypto.

Barney:

Yeah, there's a few fundamental reasons why derivatives are traded a lot. Sometimes it can be convenient. So if actually you are looking to make a directional bet on a market as a fund or something, you can potentially get access to that synthetically through derivatives. And that can be two values there. One is that you can access it potentially with more capital efficiency. So rather than having to buy a hundred million dollars worth of stock with a hundred million dollars, you can buy a position that represents the same position that you're looking to have with say 50 million dollars or 20 million dollars and use the rest for something else. So that's simply an efficiency play. Then you have derivatives that actually represent something that doesn't exist in a physical market. So there's a difference between an option on IBM stock and IBM stock. The option has different value and it's value relates to how long the option has to expire. It relates to volatility and a bunch of other things.

Barney:

So if you have certain opinions about the market, you may be more interested in expressing opinion about the prices selling in the future or the volatility or some other aspect, which can only be captured through derivatives. And then thirdly, and very interestingly is the way I like to think about derivatives are about the unbundling of risk. And you'll hear people in the startup world and tech world often talk about bundling and unbundling. And there's a lot of theories that both startups either make money by bundling together a bunch of things that people want into an easily accessible package or by unbundling things and making some aspect of something that was previously only available as part of something else available and derivatives do this for risk. And so if you think about a business that you might run where you are, let's say you're flying airplanes around and you're exposed to many different risk.

Barney:

You're exposed to currency fluctuation, you're exposed to the price of oil. Buying an airline stock expose you to all of those risks. Whereas, a derivative can be used to hedge your get exposure to just a specific risk. So you can start to say, I am actually only interested in being exposed to the price of oil over the next six months. And you can unbundle that risk from everything else and either do that because you want to be exposed to it. You want to take it on profit from being exposed to it, or you can do that because you actually want to mitigate your exposure. You have a natural exposure in your business and you actually want to hedge that risk.

Barney:

And so that's what derivatives allow you to do. And so for those three reasons, derivatives have a huge number of applicabilities to the modern financial system and are effectively used to get people the exposure they want and to move risk around to the places where it is safest and best to keep it and effectively, they sort of act as the kind of lubricant for all of these different financial transactions and activities.

Mark:

Got it. So airlines need to buy a certain amount of oil, but they need to sell tickets ahead of time. And so they may need to lock in the price of not oil, but I guess jet fuel.

Barney:

Exactly.

Mark:

And so they're buying...

Barney:

Exactly. And another, a great example, which I love is there are manufacturers of seeds who actually purchase derivatives and then include the derivative with the bag of seeds. So you'll basically get a bag of seed where if the temperature that the winter after you buy that seeds goes below some threshold temperature, you can basically send back the, I don't know, barcode or whatever from the back of the bag of seeds. And they're going to pay out some money because the manufacturer of those seeds has taken out this derivative product that will pay off based on a weather derivative on the temperature, in your region. And that means you can buy seeds that will either produce a successful crop, or you get a payout from the manufacturer of the seeds, and that's all enabled by the derivative.

Barney:

But for the purchaser of those seeds, it simply means less risk. It means I can run my farm. I can buy the seed, I can plant it. And if something goes wrong with the crop because of the temperature and that kills the crop, which maybe happens say one in 11 years or whatever. If that happens, I can send this thing back and I can still get a payout rather than be completely out of pocket that year. It's just a great example of like...

Mark:

Oh, that's a brilliant example.

Barney:

-These things really do enable particularly small businesses or very risky businesses to manage that risk and to have more certainty in terms of their financials.

Mark:

So there's a couple reasons then... That's a great example, by the way, I've heard a few examples, but that's maybe the best day to day one. I love the idea of going to Home Depot, getting some seed and having it packaged with a weather derivative. But some of this is speculation and some of this is real economic activity. And it's probably really tough to figure out what percentage is what, but I wonder if you have any views on how much of this is just hedge funds playing the market and how much is it a real business offsetting a real world risk? Not that they're mutually exclusive. Right? On one side of a trade, you might have a real world business. On the other side, you might have a trader who's taking on that risk, or you could have two businesses. I mean, a winter with cold weather is bad for the farmers, I guess, but great for the snow plow company, right?

Barney:

Yeah, absolutely. And you actually have three types of business in the market. You have the hedges and the people who are naturally long something, or naturally short something, and they want to hedge that risk. You have the speculators. So the people who want to take our position because they think they can make money. And then you also have the market neutral players, the liquidity providers, the market makers who actually don't care either way. What they want to do is make money by offering both, to buy and sell as a spread and they manage their risk and exposure and they facilitate actually, the trades. Because one thing that's true is you don't necessarily have a coincidence that exactly the time you want hedge your crop failure risk, there's not necessarily a hedge fund that wants to do exactly the same size deal on the other side.

Barney:

So there's market makers ride this useful service of basically moving the risk around in time. You want to buy now, they offer you a price now. It's not the perfect market price, but it's offset by a spread which represents their risk for inventory in your position until someone comes along the other side. So you have all three of those players. One thing that's worth mentioning, because I think a lot of people assume all speculation is gambling and just bad, is that part of the point of speculation in markets is to allocate resources. So if people speculate that something is popular and they start buying it up, that changes the price and that changes the allocation resources. So it is worth mentioning that the speculation is not just always gambling. It can actually help set prices, which helps to allocate resources in the market. And if those markets work well, that's beneficial.

Barney:

In terms of how much is which, in the traditional markets, very large amounts are non speculative. So in traditional markets, many sort of futures markets and things like that, an actual physical settlement to real products and there's very large amounts of producers and suppliers and industry purchasing things. You look at energy markets, for instance, most of there are obviously people speculating there, but there are also people fulfilling demand for say energy for homes. And there are people trying to sort all that out. And so those markets that large amounts of non spec, and I don't have the figures, but I would say probably more than 50% are your business based risks for some of those tax derivatives. Other derivatives is different and for every market's going to be different. I think in crypto, what we see right now is partly for the sort of same reasons we started Vega. There were non fully functional marketplaces. You have less derivatives and elsewhere you have more speculation because speculation is always easy to do.

Barney:

It's easy to look at a price and decide about whether that's going to be the future price. It's easy to jump in as a hedge fund, but crypto isn't fully integrated in the financial system yet. You can't use stable clients to pay as easily as you can use a Visa or a MasterCard. You can't take out a home loan quite the same way or a mortgage in quite the same way as you can. And these people are developing these products, but because all of these different products don't exist in the market yet or where they do, they're not quite as... Don't have quite as good user experience or they don't have as good capital efficiency at the moment in crypto. What we see is a lot more speculation. But really it's not the way it has to be. It's more that's where we are in the development of this technology, I think.

Mark:

And that's a great point. So let's dive into crypto. Now we have a baseline for how it works in the traditional system. In crypto, maybe it would be useful to just run through a couple of practical examples of how derivatives can be used to risk manage outside of speculation in crypto. Because so much of crypto, I mean, in general, that the noise is, there's just a lot of noise. And I guess the obvious example is the minors, right? But I wonder if you have a favorite couple examples of where derivatives actually are really necessary for [inaudible 00:14:35]

Barney:

Yeah. There are two obvious examples and one of them is mining and the other one is gas. Right? So if you look at Ethereum gas costs, if you were running a business that involved creating some number of transactions on the Ethereum chain, the number of orders of magnitude by which Ethereum gas has buried in cost over the last six months must be three orders of magnitude. You've gone from $2 transactions to $250 transactions or more. So really gas prices on Ethereum, I think were a great example because it's this kind of infrastructure cost of doing business on the Ethereum chain and it's completely volatile and unpredictable. So if you could hedge that, you're much more likely to be able to successfully do business on chain. Mining's another one. Obviously you have this kind of, you're stuck. You bought this mining equipment, you're mining you either.

Barney:

You are either using it or you're not. And if you're not, it's just going to waste entirely. And if you are creating some amount of value, which is variable out of some amount of energy, which is also variable in costs. So you have this absolute line scenario where you've got this big pile of stuff in a warehouse, you've got to pay rent. You've already bought the stuff, it's values going down as the future mining hardware gets more efficient, and the value of what you are mining and the cost of the electricity are fluctuation all the time. So all the things which you can want to hedge. And then I think the other useful thing in crypto is actually just taking the blockchain part and saying, "How do we give people products that would be better than what they get today?" If I'm a small business and I call up my commercial banker and say, "I'm exposed to the cost of the Euro." They'll offer me a product.

Barney:

And I saw some analysis a couple of years ago that looked at the average price for derivatives. And this spread is like 25 X, probably what it realistically should be given the underlying market price and spread of liquidity. So you're basically getting shafted by your commercial banker because you don't really have anywhere else to go for this product. And if we could do that, I think TransferWise did the same thing for moving money around. They kind of built this deep network of transfers, peer to peer, where you're backing transfers out against other customers rather than against the expensive banking costs and brought the cost down to everyone of transferring internationally. I think you can imagine a similar thing with crypto for normal derivatives.

Barney:

If I'm exposed to the cost of the Euro, and I can settle that in a stable coin and I have a good connectivity into a banking system or an exchange or whatever, being able to use a system like Vega to get that Euro U.S. Dollar derivative and have it at a more competitive spread and price potentially than my bank will offer me. Things like that could be a massive marketplace for crypto to come in and upend traditional finance.

Barney:

So I think there's both the crypto use cases for new crypto businesses, but also using this technology that we've built to actually make things better for things that people use every day.

Mark:

So, that's interesting. I mean, there's a whole element here of bringing derivatives into crypto and then there's element to bring crypto into TradFi. But before we get there, what do you think is actually holding back derivatives in crypto so much? Because it is quite striking. Not just... It is quite striking how low the penetration is relative to spot. The idea that derivatives should be a much actually bigger notional market than every day assets just trading, is a big thing to wrap your mind around. But the gap with crypto is big and I wonder why you think that is.

Barney:

I think it's a combination of reasons. I think there's probably no one thing. And I think we see the penetration creation increase as you get better platforms. So I think part of it comes down to the cost of trading. Part of it come down to the availability. Part of it comes down to regulatory uncertainty. And part of it comes down to the number of businesses. If let's say 50% or more of TradFi derivative trading is due to people who have big industries, big exposures. If most of the projects that are trying to do things with crypto are still quite speculative in early, and they all have relatively small user bases. That side of the market is much smaller right now, as those things grow you, as you start to see things scale to hundreds of thousands or millions of users, you could look at things like open sea or something like that.

Barney:

Maybe at some point, one of those guys will start to say, "We want to subsidize gas costs for users based on a derivative built into every transaction we do for gas." And that would look a bit like something like that bag of seeds. And actually, instead of list your NFT on OpenSea, and now someone's paying three grand in gas to buy it, maybe they could use derivatives to deal with that kind of problem. But I suspect the more and more mainstream this stuff gets, the more we'll be forced to solve these problems. Because right now you can just say to people, "Oh, it's really early connect your ledger and pay three grand in gas." And people to some extent, seem willing to do that. But I think if you imagine the next...

Mark:

But we're talking kind of on a relative basis. Right? So I guess isn't that an argument for crypto being small, not for derivatives being small relative to spot?

Barney:

I think it's related to the maturity of the marketplace. I think the crypto market is a little less mature than others. And so it is closer to being a sort of pure speculative, own it early marketplace. So early on, when you get a new asset class and people don't really know what to do with it, or something like that. When you think about businesses early on, people that were thinking who... Well, Facebook stock or Tesla stock were thinking about buying it as almost like a VC style investment about its potential future. And probably many of them were not thinking about options and derivatives and hedging and stuff for the future of that. And now as Tesla, for instance, becomes this kind of more blue chip company, making loads and loads of cars and being exposed to all the same risks as everyone else, it's market to market for the Tesla stock matures, people start trading options, people start trading futures. And so the whole thing changes.

Barney:

So I think it's about the maturity of the crypto marketplace and the different players in it. And in crypto, certainly the speculation has definitely led the use cases, but partly it's also a mystery, I guess, in a way it's. We are looking at this with our magnifying glass from above and trying to say, "Why is it like it is?" And we can ascribe these reasons, but the reality is we don't know. We don't know for sure whether as crypto matures, it will develop to look like every other market in the more traditional space or whether things will be a little different because of the differences on chain as well.

Mark:

Okay. Makes sense. So time and maturity. I presume, especially because you're building a better platform for doing this, that there's a bunch of technical reasons as well. One thing that I thought about a lot is just the processing limitations on the blockchain. I mean, I recently learned that options markets and derivative pricing is based on the Black-Scholes model, which takes volatility as a kind of input into how you're pricing. So that was a Nobel prize winning theorem. But derivatives didn't really take off in TradFi until you could do logs and exponents on a personal computer. And I think Black-Scholes came out only a few years before the first TI calculator that could do those two things came out and processing power has increased a lot since then. And derivatives has actually tracked up pretty closely because you can price derivatives more easily, unless you can trade them more easily. But on blockchain, it's just so slow and so expensive to do calculations. As far as I know, you can't do logs and exponents on the blockchain. And so there are these real technical limitations. And I wonder what you think those are in particular today.

Barney:

Yeah, that's a really good question. And actually is the heart of designing something like Vega is working around and working through and solving some of those technical issues. And the reality is that the technical issues are not necessarily inherent to the concept of a blockchain, but they are inherent to specific approaches. So when we look at Ethereum, for example, one of the things that Ethereum does is it has the Ethereum virtual machine, which runs your smart contracts. That works in a certain way, but it also is extremely flexible and extremely general purpose. And that both has advantages. "Hey, it's flexible in general purpose, you can do anything. And disadvantages, it's flexible in general purpose and needs to have this state that's entirely stored on chain. And that means that those things are slow." And like you say, maybe logs are experts and other mathematical things not only are expensive to do via something like Ethereum, but there are lots of other restrictions which make that quite difficult.

Barney:

One of the things that we have done in designing Vega for instance, is build kind of higher level abstractions. So rather than just a virtual machine and go, "Hey, it's up to you guys, do whatever you want, but we're not really helping you." We've actually built matching engines, risk systems, risk engines that are built on the bare metals. They actually run on the computer CPU just like they would if they were in a centralized exchange. And then they expose those features to the blockchain. And Ethereum does this actually with cryptographic primitive. So in Ethereum, you can call a thing to verify a hash or a signature, and that does not run it. That hash functions not run on the Ethereum EVM is actually implemented in C or C++ or something that compiles and runs on your machine.

Barney:

And that's why Ethereum can verify hashes much faster than it can do logs and exponents and do Black-Scholes. We have gone, "Hey, these are other primitives we need, not just cryptographic ones, but actually financial primitives. And we can do the same thing. We can compile them. We can make them run fast, and then we can expose them to the layer of applications and products that are being built on VA and say, now you have access to a Black-Scholes risk model, or in fact, potentially some future iteration, five generations past Black-Scholes which is even more complex."

Barney:

But the fact is because we can compile that down and run it on the pure bare metal of your machine. We can make that happen faster. We can also design systems for when that doesn't work. In banks, you'll find half of the time these risk systems run on a cluster of 20,000 machines and they report the data back in an hour. We've also designed aspects of the protocol that can deal with that so that you can have a risk model that's actually run by many participating nodes where the information then goes back through the consensus layer and comes back in a similar way. So we've really been designing Vega around those imitations and to give the ability to use these algorithms that we need to do with derivatives well, in much more closer performance characteristics to a centralized exchange compared to trying to do it all in a Ethereum contract.

Mark:

I can totally see from what you've said, that could be an order of magnitude, more efficient and handle more capacity than for DeFi and derivatives in particular, than something like Ethereum. But what about the kind of victim of your own success problem? Wherein the more people that cues a distributed system, essentially the slower it gets. These are some of the highest performance systems that exist in the traditional world, like high frequency trading, things calculating financial outcomes and derivatives. Putting it on chain is inherently going to be slower than those centralized systems because multiple computers have to agree. So almost by definition, it's going to be slower than one system. And so the more people that use it, the slower it gets, and it kind of becomes a victim of success, do you see that as a surmountable problem?

Barney:

Yeah. And there's like three different parts to this. The first one is the number of people running nodes. So if you look at how fast a system like this can be at obtaining consensus. For instance, the more nodes you add, the slower it's going to be. One of the things we can do is say, "Well, we'll make slightly different trade offs to something like Ethereum." And naturally a position or a market is much more short-lived than say an Ethereum address and potentially a balance on that address. And so you can potentially say, "We'll have slightly fewer nodes running an individual here, running a market, and therefore you trade off some amount of potential long-term security for speed." And there's a lot of research going on right now into how much of that trade off you can make and what it looks like. But even when you look at other proof of stake systems than Vega like Solana and others, the number of nodes that effectively run their network are relatively low compared to the number of nodes on Ethereum or Bitcoin.

Barney:

So there are a lot of different ways to make that trade off. The next side is to think about... Yeah, you mentioned HFT and high frequency trading and stuff. And one of the things we can think about there is actually, what's the purpose of some of this? And what's necessary versus what's what we've got. And one of the things we tend to look at is, well, if you have an order book with one microsecond latency, someone who has the fastest computer in the room is going to turn up and is going to effectively front run everyone else with their fast computer and their low latency. And they're going to make money. The question becomes in a centralized exchange where everyone trading is paying fees. And so the centralized exchange love it. They want all the HFTs because that increases the trading volume and increases their fees.

Barney:

But the rest of the market doesn't care. The rest of the market is not interested in having their values siphon off by an HFT. So one of the questions we have is actually, if you only had 500 millisecond latency on the place, rather than microsecond, there'd be less HFT but would the world be any worse off? Actually, would the world be perfectly happy? All of the speculators, the hedges, even the liquidity providers, would they all be perfectly happy to make a functioning market without the HFT siphoning off value. And we think the answer... And we've seen studies out of Chicago Booth talking about frequent batch options and things like that. We think the answer is, yeah. Actually, economically the world will be just as happy without going to the end degree of the lowest possible latency. So one thing you can potentially do is say, the market doesn't necessarily need to work exactly like it does now with HFT. So you can let yourself off some of it.

Barney:

And then the final part for the victim of your own success problem is the more people who trade, the more people who use this, the more products the harder, is to think about sharding and horizontal scalability. So if you look at Ethereum and charting, the big problem they have is every single smart contract might want to talk to every single other smart contract and every single other address on Ethereum. So effectively, if I run a shard, which is where you run multiple networks and effectively you say we process half the transactors on network A half of them on network B to try and increase throughput. The problem with Ethereum at Michigan, this horrendously difficult computer science problem of, what do I do if the program I'm running on shard A wants to talk to program on shard B and they're cross into dependencies. The nice thing about markets is once you're trading on market A there's absolutely no reason why that market needs to talk to market B.

Barney:

It's interesting. In the order book on market A and just the price on market A that the other positions, the risk on market A, the market B is irrelevant. Your position is on market A and we can manage that separately. So we can actually just do this nice thing where there's a small, fairly infrequent and very low cost, occasional settlement and moving money between shards. But each shard basically just has a subset of markets. So half of the markets are on one network, half on another. To the user, it looks like one big platform, but as we need to add more markets and increase the sort of scope to all of the world's markets, we simply add more shards. And that's a much, much simpler computer science problem because you have this nice compartmentalized situation in terms of processing and order on a market.

Mark:

I see. So it becomes easier to scale because you're designing for something more specific and the complexities aren't as much.

Barney:

Exactly.

Mark:

And then the reality is that just because the reason that we have such high performance systems running this stuff on the centralized markets today is not necessarily because that makes the experience better for everyone involved, but because there happens to be this money making opportunity that creates an arms race.

Barney:

Yeah, exactly. And the other thing to remember on that is a lot of the really high performance systems are actually held by the people doing the trading, not the platform. So the platform does need to be able to do a bit of risk analysis to calculate margins, in fact, but half, most of the platforms in traditional finance actually only do that once a day. They basically only do margin calls once a day. And they'll often give people a whole other day to send new margin in. We do do it in real time on Vega, because that helps make up for the lack of credit worthiness and credit checking and things like that helps to basically manage the risk better as we do do it in real time. But most trading platforms actually don't need to reevaluate Black-Scholes very regularly. So that's actually not too big a problem for the platform. As a trader, you might still have far more fast systems that you would never run on the blockchain, doing your modeling and analysis to send your prices into Vega.

Barney:

So if you're trading on Vega, you're still going to have all of this infrastructure to calculate the best price and to do all of that. And you are still going to be using these really, really high performance systems, but we believe we've got to a point with Vega where we can offer low enough latency. We can have good enough risk management, and we do all the things on chain that we need to. And with that scalability I was talking about, we believe we can operate a compelling platform, or we can build a compelling platform for the validates and the community to operate. And really that will be down then to the traders to run their own algorithms, which obviously will still be probably as complex as they would be in a traditional market.

Mark:

So then how do you think about order books factoring here? Because a lot of DeFi is actually not order books, it's AMMs. But if you expect the future of the derivatives world to have people making very complex calculations on their own computers and then signaling that to the market with trades, I mean, that starts to sound like an order book, right?

Barney:

Yeah. And I think there's two things. First, here with derivatives, AMMs are harder. And what I mean by that is with spot with no leverage, you can build an AMM that never goes bankrupt, right? So basically you just balance the prices. The ratio, the less of one asset is left, the more the price moves and it's designed in such a way that it never runs out of either asset. The price can get insane, but it can't run out of the asset. So, that's easy to do. If you don't have leverage, if you have leverage come into the equation, you can't do that. The pool can go bankrupt. And that is a solvable problem. And then we're working on various solutions to that. And it looks a bit more like something like Uniswap V3, where you basically have different pools and the parameters are not just the price, but also at leverage.

Barney:

So, that's a solvable problem. The problem that's harder to solve with AMMs is price discovery. AMMs are pretty good if you don't have much price discovery coming from the market. So if you have very low liquidity, there are things like the AIM, alternative investment market in London, where you look at some order books and they'll get one trade a week and there's almost no orders on it. It's very difficult to claim that is how good price discovery. Basically, no one has a clue, all the prices until someone puts something in the order book, and then it jumps by a massive amount. So in that case, I would argue an AMM might be better because you've got this sort of pool of capital and someone who put that pool of capital in can adjust the ratio if they want to change the price they're implying from it.

Barney:

And then anyone who wants to trade can access liquidity instantly. So actually for a low liquidity market, I think an AMM is better, but once you get past a certain thing where the market is trading constantly, and many different traders are sending signals from the market in, in an order book based market, those signals are coming via limit orders. And that's information coming from outside about the price. In an AMM, this curve is just algorithmically moving up and down and it always lags the true price. And the way that we think about this is the best thing is a hybrid because an AMM curve can look exactly like an order Booker. Basically, you can just turn it into orders. You basically... Every single price, what amount of volume is the AMM offering me at that pricing. It will offer me 10 at this price and then the next lowest price it'll offer me five. The next lowest price will offer me three or whatever, and that'll give you a curve.

Barney:

So we actually believe that the best type of marketing future will be a hybrid where people can deploy not just one AMM curve, but actually you gain more likely [inaudible 00:35:46] three. Liquidity providers can deploy multiple curves, but people who have more of an opinion about the price can also deploy limit orders and the Vega engine or whatever market engine you're using, will in future just look at all of these curves and all the limit orders, and it will just give you the best execution. So it will just look at it. You'll say, "I want to buy a hundred." And it will simply take from all of the liquidity curves. The AMM style curves have been deployed and all the limit orders it'll take in the ratio. That'll give you the best average price.

Barney:

And I think to me, that's the end game because that gives you the liquidity, the sort of cheap liquidity in especially low liquidity markets that you get from ANM curve. It gives you the price forming capabilities of an order book, and it gives you them all in one venue in one place where you can integrate over that whole set of data and get pricing information at any time.

Barney:

And also the final advantage of doing something like that is you can have a market which gets launched by someone who doesn't have the capability of actually say doing real time, operating a market making algorithm. They actually are more of a market designer. They see a need in the marketplace. They create product, they launch it, they stick up some liquidity and effectively in a curve as just money. People start trading that. The trading volume gets higher. Then maybe a market making fund takes a look and goes, "Hey, this is really interesting. We can price this. We can price it better than this dumb AMM curve. We're going to start providing liquidity by placing limit orders." And now you have this more liquid market with limit orders driving the price, but with this sort of backstop liquidity in the AMM curve. So I think that's sort of how we see this long term developing.

Mark:

I see. So you expect AMMs to enter... You expect there to be an AMM solution in the derivative market. You don't expect it to play out like dYdX has where you have an off chain order book in on chain settlement.

Barney:

No, I mean, no. I expect there will be an AMM solution. I think it makes a lot of sense for making it easy and cheap and quick to deploy markets and liquidity and to low liquidity markets. You could do that in an off chain or the book as well. You could do an off-chain AMM curve. There's nothing that says you couldn't. But I think for us, the reason why you want it to be on chain is that is the permissionless innovation point. So with the dYdX model, at least as they've been so far, and they sort of talked about changing this with their next version, but with the dYdX model, it's kind of like a traditional market in that there's a bunch of people in a room deciding what markets to launch and what you should be able to trade or what you shouldn't.

Barney:

And we think that actually the most exciting thing about this is not me telling you what you want to trade next, but you telling the community what you want to trade next and the community building markets and the community responding to their own demands and needs and making money doing so. The goal should be the same as when the ERC 20 standard got invented in Ethereum, for better or worse, everyone who had an idea for a new protocol, could go sell a token and build a business and try and make that protocol successful. I think if someone has an idea for a market or a derivative product that they think people want to trade, they should be able to build that and they should be able to go off and try and build a business, making that market and that's the kind of thing we want to enable. So we really want to direct that value creation back to the community of people who create it rather than to the Vega team.

Mark:

How long do you think it's going to take until this market catches up? I mean, we've talked about some of the roadblocks, we've talked about some of the solutions we've talked about what these markets will look like. When do you think they'll come into their own? Is this a couple years? Is it a decade?

Barney:

So yeah, I think in the next couple of years, we will see... I don't know whether it's generation two or three or whatever of DeFi at this point. I'm not sure I've been counting, but I think in the next couple of years, we will see a generation of DeFi products which are capable of delivering that useful quality experience where it can actually do this for the long term. So I think in the next couple of years, we will start to see these platforms emerge that are the future. I think it'll still look like the early days of the internet, where there are a bunch of people running around going, "Hey, you haven't seen how good this platform is. You haven't seen how crypto has changed in the last two years, what the new DeFi looks like. And there will be sort of small pockets of it, but it won't be as widespread as traditional finance."

Barney:

And then I think maybe in the next one to two years, something like that, maybe so coming to the end of that two year period, we'll probably start to see the first things built on top of this new wave of protocols, like Vega, where they branch out and someone actually builds something that people in the traditional finance world or people in traditional business find useful. And suddenly you find that a bunch of normal business operators have started signing up for this hedging product or something like that.

Barney:

Or so I think probably towards the end of two years, we'll start to see those kind of products. And the reason I say that is I think these platforms are just getting launched now and including Vega. And it takes a little while to go from a launch and an alpha to sort out the UX, to solve the [inaudible 00:40:51] troubles anything has, and for the world to have enough confidence to put real money in it. You wouldn't have wanted to put large amounts of cash into Bitcoin when it was still a hundred people on a forum using this software that broke every so often.

Barney:

So I think looking at where we are now, it's going to take a couple of years to get to the point where people are really trusting and where that next wave can happen. But I think in the next six months to 18 months, we'll start to see a bunch of these new generation platforms emerge that are the sort of serious contenders for that.

Mark:

Okay. Makes sense. And as this market matures, it has huge benefits insofar as it lets people manage risk, both for speculative purposes and for real "world business." Whether that's NFTs like open seeds or gaming or anything else or mining, but it also introduces risks. In the traditional markets, we have a concept of accredited investors, which at least in the U.S., and in order to trade private securities, you have to have a certain amount of wealth or expertise. Anyone can trade public securities, but they are required to make disclosures and et cetera, et cetera. And derivatives markets have an even higher bar because they're inherently more complex. And so in order to trade derivatives, you have to clear some hurdles that are perhaps even harder than being a credit investor, although I'm not totally sure on what those hurdles are.

Mark:

And I worry that as this market grows within crypto and in a permissionless way, that it can expose a lot more individuals to what Warren Buffet has referred to as financial weapons of mass destruction. How do we deal with that as a crypto community? Because we probably can't regulate our way out of it. It's permissionless, but it is a problem that I don't think is good for society.

Barney:

And this is... I mean, this makes me think of the war on drugs. If people start taking crack, it's not good for society. And the U.S. Government tried really hard to regulate their way out of that and to make that not happen. And despite that, and despite arming police and going after it, and then so much money and international coordination in the UN, despite all of that, doesn't seem to have helped. So I think the first thing is that people are going to do the things people are going to do. And certainly like you say, regulating our way out of this is not going to solve it. And perhaps there was a short period in time where enough of the infrastructure and what was needed to prevent people from doing these things was under the control of the government. The regulation maybe looked like it would help, but on the other hand, you also have laws against various laws around gambling and online gambling. And you also find that people have illegal gambling, dens and all of these other things.

Barney:

So I think a lot of people do what they're going to do. And like you say, it's actually, for a large part, is down to the community and it's both the real world community and the crypto community. And the real world community, I think it's way better to educate people on the dangers of gambling, trading drugs, whatever. Education is very key and in the crypto community, I think to the extent that you can do it self-policing is useful, but self-policing means things like if you are building the front end that connects into a decentralized protocol, something like Vega, maybe you also think about decentralized ways that people can analyze risk and give people warnings about risk. In traditional markets, if I'm a trader, I have all kinds of risk systems that tell me what type of risk I might be taking on in the position.

Barney:

Actually, most of the retail platforms for derivatives in countries that don't make them illegal. In retail in the UK, you can have retail derivatives like spread betting and things like that. And those platforms are designed like a slot machine. The flashy lights and here are some strategies get trading. You spend your money, make us some fees. We have an opportunity to do way better than that. We could be more responsible because we can build these open source front ends and go, "Here's the risk. Here's how likely you are to get wrecked in within one hour, 24 hours, one week. If you take our position of this size, here's how volatile this market is." We can show people that stuff and we can give them all the same days they would have in an investment bank. We could also find better ways to present it to people who are less financially savvy and help them understand.

Barney:

And because we are not a spread betting platform who makes money off the fees because we're open source developers, we have less incentive to screw people into wasting all their money. We actually have an incentive, I think, to build these interfaces and to give this education so that when someone starts typing large numbers into the deal ticket on a decentralized exchange, maybe that platform can start telling them about the risk they're taking on and start offering them ways to think about that. So I do think there's that side of things. And then there's also the actual, probably the worst thing in crypto, which we have to deal with is scams and sort of frauds. And that's really hard that you see things with Uniswap [inaudible 00:46:19].

Mark:

Yeah. Cause I mean, if that stuff happens with ERC 20 tokens and...

Barney:

Exactly right, you see.

Mark:

-It's going to just amplify itself.

Barney:

You see this stuff with Uniswap. People will create the completely fake token, fake Vega token or whatever. They'll deploy a Uniswap market and then they will go spam it. And then people will think they're buying some token and they're buying worthless nonsense and sending their money to a scammer. We actually think... At Vega. We actually decided that problem was so big. And it's even worse with derivatives because with derivatives you rely on Oracle to settle. So it's not even like you can actually look at the token. It's like, this is an Oracle where someone is saying, they're going to eventually one day report the price of this thing. We are doing what we can to find ways to create open source repositories of information about known Oracles and so you can start to verify all this data is actually signed by Coinbase. So we trust it or whatever.

Barney:

But the other thing we're doing, Vega, is we've said anyone ought to be able to propose a market, but actually maybe it's better if the token holders who sort of own the governance token can actually vote to close a market and vote to say, "No, this shouldn't exist." So someone comes along and says, "Hey, people should be able to trade this thing with leverage." If all the people who have an incentive for the platform to be successful and to not be full of scams, look at it and go, "We have no idea what this Oracle is. This could be a complete scam. We think the best thing to do is those people just vote for it not to exist." That's not centralized control. That's not censorship. That's the bunch of people who are building open source platforms and code having shared incentives for it to actually do something useful and not be just a net drag on society.

Barney:

And so we tend to think that something like that with that community curation is probably a slightly better option than just allowing a complete wild west. Because like you say, we're talking about this gap, this catch up. How does crypto catch up to traditional financing terms, derivatives? How does crypto catch up and displace traditional systems? It's not going to do that if the average person thinks it is dangerous and that they can't trust anything and that they need to know how to read a solidity contract to verify the market they're using isn't a scam. So we really need a solution technically, and within the community that enables the average person who would want to use this product, not to also need to do a bunch of other things that are unrelated to not get scammed.

Mark:

Makes sense. Okay. So Brave New world, happening whether we like it or not. And sounds like it's going to be a knife fight to make sure that we mitigate the bad while we allow the good to grow.

Barney:

Yeah, absolutely. I'm optimistic though. I think in general there are more good people designing good systems and making them better than there are bad people. And really you don't have to make it so bad. People can't do bad things. You just have to make it so that on average, doing bad things is incredibly difficult and frustrating. It doesn't pay off very well.

Mark:

I mean, that's a great point. Ultimately, you can make more money doing good things than you can doing bad things. It's sometimes harder, but I guess you can level the playing field by making it harder to do that.

Barney:

Yeah. And that's what communities... In face to face communities, getting shunned by people because you're just a bad person and things like that, the way that communities work is they basically make it more effort and create more friction and make it less pleasant to be a bad and leeching member of that community who brings everything down than to be a positive contributor. And I think that's a good lesson we can learn in crypto. We've got to make it a little bit easier to be a positive contributor and a little bit better to be nice than it is to be bad.

Mark:

Awesome. Okay. Well Barney, thank you so much for joining us today and sharing your thoughts with us. Such an important topic, but a very confusing one for a lot of people. So thank you for shining some light on it. Before we go, are there any parting thoughts and how can people follow you and your project and learn more?

Barney:

Yeah, I think my parting thoughts are the most exciting thing about all of this is handing the power to use these tools that have been sort of locked up in a few offices in London and New York and Hong Kong and wherever else for years. The power to use these tools to do incredible, cool things and to improve the value for communities. To people to create their own little businesses, people to do their own stuff. And to hand that power to everyone and see what they do with it. I'm super excited to see this next wave of stuff, including Vega launch, and to see what people do manage to do. If you want to follow us, Vega.XYZ is our website. I'm @Barnabee with two Es at the end on Twitter. And @Vegaprotocol is our official Twitter. Please get involved, join the Discord, ask questions, take a look at our GitHub posts and yeah. Hope to meet some of you in person.

Mark:

Great. Well thank you Barney. Really appreciate it. Super interesting.

Barney:

Awesome. Thanks Mark. Thanks for having me.

‍