Hello, and welcome to another episode of the Odd Lots Podcast. I'm Joe Wisntal and I'm Tracy Alloway. Tracy, do you remember the conversation that we recently had with Doug Seafood, the Virtue CEO. Yeah, that was a great conversation. Looks funny, Yeah, no, that was that was a really fun one. Um. Obviously, that was a discussion about high frequency trading, electronic market making and listed equities and what's striking is ultimately, I mean, these are like profitable businesses, but you know they're in that business just scraping for like pennies or fractions of appennies, like these are pretty uh efficient, you know, pretty efficient markets. It's it's hard to ring more profit out of out of them. It's definitely a volume business. You're making like a penny on the trade aid, um from a tiny tiny like difficult to see spread, right, there's so many different players and as everyone's trying to get their margin, and so you know, I've said it before, if you know US listed equities in particular and probably elsewhere too, it's like they're pretty efficient markets. Yeah, I think that's right. I mean I think, um, well, we also just had an episode on the U. S. Treasury market, and I think you'd probably say stocks and maybe US treasuries are the most liquid markets out there. Um corporate bonds not so liquid. We've talked a lot about that on the show. So one thing we haven't actually talked about, I realized is we've done a few episodes about like crypto and bitcoin, but we've never really talked about like market structure in the space as far as I know, or at least not in depth. And my impression is even though the sort of like industry of this ecosystem that's been built around it several years old now, I think it's safe to say completely still on the other side of the efficiency spectrum from the space that like the for twos of the world are the citadel securities of the world are playing it. Yeah, I think that's a fair assessment. I mean, for one thing, you have just tons of platforms and the market is completely fragmented, and then you also have these big regional variations where like the price of bitcoin can be one thing in Korea and then a different thing on a platform in the States. There are a lot of arbitrage opportunities, I think, but Also, I can't imagine what it's like trading in that environment, and I think you probably have to think about costs and platform risk and other things a lot more than you would for certainly for equities. Right, So you have these sort of like massive different you know, all these different platforms, huge spreads potentially at times between what an asset trades at one place elsewhere, different regulatory regimes, different currencies, different banking systems. In theory though, those costs create big opportunities for savvy traders. Yeah, and we've heard some stories about people making millions um, possibly billions, just by arbitraging like a simple regional spread. Yeah. So today we're going to talk about crypto market structure with someone who I think is basically the perfect guest to discuss it because he previously was in the world closer to the sort of market making uh world that we've talked about with Doug. We're gonna be speaking with Sam Bankman Freed. He is the CEO and co founder of the crypto exchange f t X. He's also the CEO and co founder of the crypto hedge fund alimt Research, and previously to that, he was in the in the market making business. I think trading E t F at James Street, which is one of these trading shops that compete in that space of of you know, high tech market making for regulated listed assets. So has really seen both both ends of the spectrum, the perfect guest to talk about crypto market structure with us. Sam, thank you so much for joining us. Yeah, thanks for having me so, Sam, I've you know, read a little bit about your work. I take it you got you know, not this boom, but the last sort of like crazy cycle for cryptov I think is when you sort of made that switch over. I think you're at Jane Street at the time. Talk to us first about sort of like your background. How did you first get into finance and how did you find your way into the world of before we even get into crypto and make that John jump the world of the world of trading. Yeah, I it's interesting, you know, sorry for in retrospect, looking back on it, it seems like so weird it in contingent how some of these things happened, but also like kind of inevitable isn't quite the right word, but like you know, it worked out like kind of remarkably neatly given that, but you know, the context is, so I went to have I t is, you know, server math nerd, albeit a math nerd who was sort of coming, you know, into the middle of their college experience, having kind of reckoned finally with the fact that like I wasn't going to become a math professor. And I mean, they're probably pretty compelling arguments that that wasn't gonna be true. Um, but realizing that I really like doing academic research was maybe a little bit of a you know, a hole in the plan. So sol it's sort of like, didn't didn't really know it's gonna do with my life. The one thing that I sort of know is that I wanted to find out how I could have the most positive impact on the world. Um. I've been into utilitarianism for a very long time and and recently started getting into effective alters, which is basically this movement of like, you know, if you're trying to figure out how to impact the world, like try and quantify things, trying try and figure out what the most efficient way of doing that is, you know, what the ways that you know gives you the most bank for the book, and was kind of playing around with a lot of possible careers that were kind of all over the place. I had some conversations to people, and basically, so, you know, you go to work for for one of these charities or organizations that that you think are good or you could kind of donate to them, and you know, frankly, given like your strengths and weaknesses, you know, maybe maybe you're gonna be able to donate more to them than you'd be able to contribute working directly for them. Um and certain I thought about that. It seems like a pretty plausible argument, and so I sort of started looking into for the first time, well, okay, like if that's my goal, if my goal to figure out or at least maybe kind of like short term goal, how how can I make as much money and donate it as possible? Like where would that lead me? And sort of one of the obvious, you know, obvious things to look at was, well, how about finance. You know, I kind of heard that's what happens there. I didn't really know a whole lot else about it, but like, frankly, really was just like I know, it's sort of like in my junior didn't know what to do, and like just like applied for some internships. You know, I don't know, some friends headered in into James Street. They said, like, you know a lot of really good things. So it's just like, you know, sent in her resume, you know, really low conviction, and then she's like interview there and really liked the interviews and she found them really engaging, and then sort of one thing led to another. Interned at James Street Capital, you know after my junior year, really liked it and and then went back full time when I graduated. So what were you doing at Jane Street exactly? And I guess, um, what was the opportunity that you spotted in crypto because you know, it took a while for a lot of financed people to jump into that market, but um, you did, did it fairly early on. So what did you see and how did your Jane Street experience, um in form that decision? Yeah, I mean I think a Jeans Street really the goal was buying low and selling high. I know, um, but but but you know, I think that there is sort of like something there of like people think of like to use these you know, quant prompt firms on Wall Street and either they think through this old style you know, gigantic men in suits talking really loud, being grabbing shairs of stock from each other on you know, the stock exchange floor, or they think of sort of a bunch of nerds with like, you know, equations you could not fathom, you know, I mean, Jeezu is much more on the second end than the first end. You know, we were a bunch of a bunch of nerds basically trading trading stocks. But you know the truth is that like a lot of it is not sort of like this incomprehensible girbledy book. A lot of it is like really straightforward simple trade ideas, just don really carefully with a lot of supporting you know, evidence and ringing everything you can out of the trade, optimizing it as much as you can, thinking hard about it. You know. So I was treading internationally ETFs there um, which are you know, ETFs are funds which contain other other equities um, and the funds themselves trade on exchanges, and there's this classic arbitrage of well, you can sort of price the e t F based on you know, the sum of all the stocks that it don't you know, and and you can sort of you know, great and rogeem back and forth between them. And you know, if the e t F is trading below the stocks, you can buy the et F and sell the stocks. And I was training both the stock team buy the stocks, you know, merge them into the e t F and sell the e t F and do an arbitrage. So I was trying internationally ts, which are those types of products, but where the stocks don't trade on you know, on on NISY where they trade on foreign stock exchanges on you know Green or or you know Indian or or you know the UK stock exchanges. And it's basically the same thing, except incredibly messy and complicated and intricate because get all these nice properties that fall out when you're training U S ETFs that you don't with foreign ones and just want to start. Is the et F itself is trading you know what PM New York time five days a week. You know, you take a n e TF on Korean stocks, right, and there's Korean stocks are trading, you know, I don't know whatever it is, like a A M two the three PM or whatever Korean time five days a weeken, those don't really even overlap. So whenever you're training d TF, right, it's it's it's this fund of stocks that haven't traded for twelve hours, and you can't just say, well, through the arbitrage stocks aren't trading, no price for them, And so that sort of opens this gigantic set you know, can of worms of like how do you think about these e t s? How to price them? And how do you understand exactly what they hold? Uh? And how do you understand the liquidity of the underlying stocks, how do you understand the fund itself the mechanisms behind it, and basically just be Hans is sort of like you know, somewhat intricate modeling problem and and I and that does sort of you know, what what we did was basically provided liquidity in those international ETF and and particularly kind of specialized in providing liquidity when it wasn't trivial, you know, not doing the single easiest trade because anyone can do that and that's super crowded, but trying to do even just the second easiest trade through the hardest trade, but being you know, trying to be out there twenty you know, not going for something. But whenever US markets broken, have as much liquidity as we could in those products given the constraints and the uncertainty of you know, what they represented, And yeah, that's that's sort of like what that desk and you know, which is the death stares on, you know, specialized in. So when you say that like the second easiest trade, what do you what exactly do you mean by that? And I'm glad you know it's funny. I've actually wondered about this exact question before, Like, Okay, a Korean ETF is trading during American hours when the underlying shares have stopped trading, So like what is this based on, you know, price relationships? Like you know, there's probably like some historical correlation between what the SMP does and what the COSPY does the following day, and sort of like imputing what those stocks would theoretically be doing during that time based on what the trading of other equities are happening. Like talk about that a little bit more. Yeah, totally. And I think that you're you're you're you're talking about really good things there. And I think that's sort of another takeaway there is like how do you start It's not by you know, waving magic wand and using native words. You start with literally the most sort of like intuitive, straightforward things you can think of. You know, it's like if you kind of think hard about this and have a lot of context, and then she like, I don't know, maybe I tried it this way you want. Honestly, that's probably a good start. You know, you're probably gonna do a lot better than someone who doesn't do that thing. It doesn't do anything at all, But you know whatever, twenty people can do that who have the setup and the liquidity and the capital and everything like that, and so and you have to do better. You know, that's okay, you know you've you've done better than than just no no modeling. But but there's there's enormous amounts of extensions that you can build on that. And it's all about acknowledging you're never going to be perfect, but trying to get you know, as good as you can. So, can you talk a little bit more about the crypto transition, Like what was the opportunity that you spotted just to circle back and what was um, I guess the shared skill set that you thought could be applied to the crypto market. You know, I love King Street and and by the way, I really really liked it there maybe the most notable thing about it was how much and how much I liked it, how much a lot of the people they're liked it. It was just a really good place, and it probably still is in just a number of different dimensions. And we just look at things like, you know, like employee satisfaction compared to competitors. You know, my sense is that's just off the charts. And again, you in in many cases just for kind of straightforward reason, it's you know, it's just like it does a good job of treating its employees well, giving them responsibility when it makes sense, and you know, guidance when they need it, and and you know all that. So so anyway, you know, I really liked it there, but after after about three and a half years, basically just felt like, um, like it was time to try my own thing. And I think the core of that came from just thinking about, you know, going back to to why I start in the first place, which was, how can you figure out how to have you know, as much positive impact as I can on the world, and you know, as much as more than just like a good amount. And so I really started thinking very clear about like, all right, is this the best thing I can be doing. And basically it came away feeling like I have a lot of things I want to try it with my life. I don't know what's what's going to end up being the right thing, but at least one of them might go extremely well and maybe not. There's there's a lot of risk in that, um, but so be it. And so you know, left and sort of like started trying things out. Um. And one of the first things that that that I looked at was like, you know, all right, like it's not the most imaginative thing ever, but have a crypto. And this was late It's when Crypto was really going through that that first gigantic public boom, and you know, this was the time when when I had sort of the mass retail appeal more so than ever before or or after. You know, to this point where you passed two people on the street and you see them talking with each other and you're like, they're probably talking about things. You know, She's is everywhere right, and it wasn't clear why she's completely captured the public's attention and imagination. In seventeen I mean, you know, the North Bricing Creas obviously was like a large part of that, it had a lot of the hallmarks of something that might be, you know, really an efficient system with you know, a big need for liquidity, which is basically gigantic demand all of a sudden growing really rapidly, you know, want the volume, lots of retail users, and not a lot of time to build up institutions, not a lot of time to build up liquidity, not a lot of time to build up systems that worked seamlessly. You know, it sort of felt like the thing that was decently likely to have very large volume and price discrepancies and just like not not really an a fliquidy prior to priding to that. So like, can you give a little comparison between a seventeen what you were seeing in trading internationally t F versus the spreads that you were seeing crypto and then also like, you know how twenty seventeen spreads and crypto compared to spreads and crypto. It's obviously whatever, there's there's you know, always whatever. Different situations have different spreads, blah blah blah. But but okay, fine ball parking it. I think you look at sort of like that, the like lowest margin trades you find in finance that the really purest h f T trades and the most competitive ones what to spread on those. So I've never really done pure h f T, but my sense is that the answer to this is, you know, you're talking about like a tenth of a penny to a hundredth of a penny or maybe even a thousandth of a penny, where a penny is is like about a hundredth of a percent on on the average stock, and so you know you're talking about like something even getting to like the one one million of the price level presents it's really really slim margins, right, And it's the kind of things that like seem like how can you make any real amount of money on that? Well, if you create you know, five percent of all volume of everything in the world every day, that adds up. Then you get to sort of like scale like quant trades with you know, what I was doing sort of being you know, an example where the right unit to talk about it and it's like a basis point and uh, and that that the hundredth of a percent. That's not that everything had exactly that amount of margin, but that was that was sort of the order of magnitude that like the industry talking there, sort of talking and like, are you making like one hundredth of a percent? Are you making half a hundreds of a percentery making like three percent? And you know, and in a hundreds of percent, you can think of it approximately is something like a penny on the average stone. So okay, now that's sort of like roughly the sort of margins that you're thinking about, right, And you know, occasionally you see things a bit bigger. Occasionally you'd be trying to scrape out anything you could, but but whatever that that's sort of like the level at which things at which competition tended to get. And then you get take Crypto, and it just had things I've never seen before or after, things that were just completely bananas, and I'll give you They're all over the place. It was sort of like a whole industry filled with implausibly big spreads. But to give you the biggest that I that I've ever seen, basically and biggest in terms of obviously, you know, you can make ten percent on a trade if you don't mind the total size available being seven dollars ever, right, and that's like sort of not interesting, right, like bias stick of gum and sell if someone kind of wants to stick a gum right now. But the real thing is finding traits are really good and really scalable. And so from that perspective for late early there's this really highly publicized thread called it and it's often called the Kimsey premium um And what was it was the price of bitcoins on creating scanatings. And at its heart there is huge man for crypto all over the world, but it wasn't equal everywhere, and there's even huger demand in a number of countries including Korea, then in in in many others to such a massive net bling of crypto from Koream users. And the koream one is also restricted currency. It means you can't just freely sell it. You can't freely get it out of the country, trade it for USD. You and all these Crean citizens trying to sell their Korean wand for bitcoins, and um, no one who could do much with the cream one. You know, it's not like other people just like selling bitcoins and deal with the current, get stuck with it. That spread got up to like at the peak, which say that pre and big wins started trading at you know, fifteen dollars, while every else in the world it was ten dollars. So that is a completely insane spread. But and this was the big catch. You get back to this restricting currency thing where sure you could turn you know, ten U S dollars into fifteen dollars worth of grand wand doing this, but then it's it's stuck. You're not regular clearly allowed to just really turn that grand wand back into US dollars. And so you know, if you really wanted to buy a lot of consumer goods in Korea, you could do it, but you can't sort of easily loop that arbitrage through. You get stuck at one end. And a lot of people tried to do this trade. Many found a way to do it for small size, very very hard to do it for big size, even though there are billions of dollars a day volume trading in it, because you couldn't all float the grand wand easily um for for non crypto and and so does serve a complication to it. And no one is known to practice for massive size. You know, there are a lot of people who are known to have practiced for small size. No one is known to have really been able to maximize the hell out of that one because of the regulatory concerns. And and this was sort of like the interesting thing down the road. There are worth these Japanese exchanges, and on the Japanese exchanges, like on the Krea exchanges, they're huge inclose, huge neckline. Crypto is praying at a really big premium. But the Japanese yen, unlike the Korean wand is not a restricted currency. It's a free currency. You can you can sell it, you can trade it, you can do whatever you want with it. Um and that it wasn't trading it quite the same premium, but it was trading at a fifteen percent premium or so at the peak instead of fifty okay, fifteen, it's not fifty, it's a lot. And there is you know, billion dollars of volume trading in in in big point against Japan's yeen each day, and you could actually do that trade. You can actually buy bitcoin for ten thou send it to Japanese exchange, send it, sell up for eleventh. That was in five hundred worth of Japanese yen, and then sort of you know, turned that end back into dollars, send it back to the States and and cycle anew um. It took about a day to do that trade, um, given the wire transfers involved, but it was doable, and you could scale it, making literally ten percent per weekday, which is just absolutely insane, right. That is that that's a thousand times what you make on you know, traditional trade. But you're able to do it for like hundreds of millions of dollars of volume per day. There was you could actually scale it. And so you can start doing that map. And I'm sure some you know you may be doing that as I speak in multiplying those numbers, and yeah, that's the right map. I don't know what to tell you. And son, you can ask like, how could that possibly be? Right? How could you possibly see a trade that inefficient? Why wouldn't that get competed away? Right? And the answer is, well, think about what you actually realistically need to do. I told you could do this trade. I can tell you how what you need to do to do it? What what goes into this? Right? What's the recipe? And well, okay, first of all, step one is you start out with a US bank account and you wire money to point base and then your bank acount gets shut down because JP Morgan Chase doesn't want to deal with Krypto. Um. Okay, So now you're back to the drying board, right. And as you try again and somehow you solve that step I don't know how. You send it to point Base and then you realize you have a thousand dollars a day withdrawal from Point Base and you just sent your entire hedge funds capital there and it's gonna be thirty decades before you can get any money out, and their support ticket to is three months long. So okay, So now you solve that problem, right, You buy your bitcoins, you send them to a Japanese exchange, and you realize that in order to get verified there you need to be Japanese, but you're not Japanese. In fact, maybe you couldn't be Japanese to get some of the other things you needed. So okay, now you have to somehow beat Japanese and not Japanese. And you solve that, and then you try and send it to the Oh god, now you realize you need a Japanese bank account send it to and needs to be a domestic bank account in Japan. Um. And you need to again be a Japanese resident to get one of those. So as you get a bank account, and then you realize they're not They don't like crypto either, so you need to solve that. And then you finally get all these pieces and you go to them you're like, yes, I would like to do a wire transfer please, Yes, it would be you whever, like fifteen million US dollars to this random bank account in America from this random bank account in Japan, different owners, same wire transferated yesterday. Yes, that's correct. Yes, it's always in the same direction. I acknowledge that. Yes, it's international. And there's sort of like this is literally the sketches that you can possibly do, like like think about like every day this random guy walks in sends it to an seemingly unrelated bank account across the world, always in the same direction, always changing currencies, right, and sort of like that no business has that right there, like it's obviously just money laundering, and like no, it's arbitrage. You're talking like it's color at Japanese bank, you know, just be Japanese, right, So okay, Son, you're trying to explain, like no, you buy it this point, you have to transfer it on the block chain, and they're like, sorry you you said you trade stocks but in Japanese, so okay, so whatever. So you solve all these problems, right, and then you realize that you have two thousand dollars to your name, and so you're gonna be making two hundred dollars a day and that's pretty cool, but it wasn't the fortune you thought. And now you're trying to raise two hundred million dollars of capital, and you know this trade won't last out long. So you've got about a week to build a two million dollar hedge fund that's doing random international cryptocurrency trades in twenty seventeen. And everyone's like, yes, obviously will not find your hedge fund. That sounds incredibly risky um, And you're like, no, it's arbitrage, don't worry. And they're like, sure, I'm hot ten percent per day returns. Yeah, I've heard that one before, but okay, fine. Somehow you then get your million dollars in capital, and now now what like, now do you make twenty million dollars a day? The answer is yes, you do. So I mean you just laid out the very fragmented nature of the market with the different platforms and also with its interaction with the traditional banking system. You laid that out really well, but I I have a slightly weird question, which is about like the nature of crypto trading itself and how it maybe differs to stocks or bonds or more traditional financial assets. So, I mean, it seems like when bitcoin really takes off, it's usually because the price is going up quite a lot, and people are talking a lot about holding and waiting for it to go up even more So I'm curious, like, how do you reconcile the hoodal or huddle concept with having a liquid two way market, Like how do you get that activity going? It's really interesting and and there's a lot of really weird dynamic and cryptom You've hit on one of them, which is how incredibly I mean, there are lots of ways to phrase it, but one of them is, like you can phrase it as how incredibly bullish the crypto ecosystem is. On the crypto ecosystem, everyone's bullished on the thing they believe in, the think they're spending their top time on. Generally, but not to this extent. This is sort of like much more so than you find with with stuff. And then also like how about south side liquidity? Right? If people want to buy bitcoins and no one who holds bitcoins is willing to sell their bitcoins, who are you buying it from? And and you know, both are sort of like a how do you ever find people buy it from? But also like a day to day where where the offers coming from? Whereas where the market makers? If all the bitcoins a ruined by people who are just toddling and not you know, then I don't know you you actually if you trace that down and start to uncover some interesting things about crypto and the ecosystem. So one thing is that everyone is bullish inside of crypto on crypto and actually massive impact on the ecosystem because any way that people can find to get longer, they will. And so you see these weird things where features are always trading it not always, but on average training at big premiums in crypto where people are willing to pay a year to borrow dollars in crypto, um generally chooses to buy more crypto, and anything else that serves a synthetic interest rate or forward type product reflecting will generally reflect this, you know, forward looking premium. So it's that's sort of one thing, and that plays into these concepts of yield quite strongly. But another thing is, as you said, like a lot of the real supply here is not circulating. It's not really circulating, right, It's it's it's sort of some guy who's just not going to sell. And so you look at crypto and the the trading volume divide by market cap is actually quite high compared to other assets, and the market cap is also even lower than it looks like. It's even more extreme than that, because a pretty large number of holders of crypto are just not trading it. They're not selling it. They're they're going to keep it no matter why. And so you actually have this really large amount of volume traded relative to the effective liquidity in the products on both by side and sell side, and you get these really giant market moves. Firstly, as as a product of that, there's a video of you online that's really interesting where it shows you acquiring a bunch of bitcoin that someone had dumped on the exchange binance, and you like walk through this trade where like, I guess someone made a terrible trade or they were not very sophisticated did, and they sold a bunch of bitcoin on that exchange and it depressed the price, and you recognize this. I thought it was an interesting video. One question I have is, so a how much um more efficient does the market these days than it was when you first started trading? But be there's a part where you talk about um. Part of the slowness that you had in scooping up some of those bitcoins that were just dumped was you were waiting for your tether to arrive at finance and uh, that took a little while because you didn't have cash on hand at that exchange. How much also, you know, talk about that trade, but also like how much of the premium that you can collect has to do with to put it better better ways, like sort of like counterparty risk. And I'm not I'm not one of you know, sending aside all the sort of conspiracy theories about tether. You're dealing with like a lot of sort of like either unregulated or underregulated entities that are not at all like say the exchanges that a Korean et f for a Korean stock is. So you have to rely on these entities in order to complete the trade. Talk to us about those calculations and how they play into both the risks but also the profit opportunities. Absolutely, and I think what I would say is basically um that that I think those things do play huge, huge roles in the ecosystem for the reasons you outline, which are you see these big premiums sometimes open up, and the reason is like, you know, how can there be like a three arbitrage for a hundred million dollars of size all of a sudden pop up? Right that seems like a lot. And part of the reason, part of the part of the answer there is roughly, you know, in order to clothes that you need to have a hundred million dollars of tether sitting on bindance. How many people are doing that right now? You know? It's like and and you there's no cross marketing between exchanges, there's no central clearing firms or brokers, and so you can it's not just you don't even just need a hundred million dollars in crypto, a hundred million dollars in every single exchange if you're gonna wait for you know, any one of them to go crazy. And so it's really capital intensive. And also you you have to worry about counterparty risk. And the thing, big thing I'd say about counterparty risk is it's not so much that, like with all good trades, they're actually not nearly as good to say, look because secretly you're being paid three percent in order to assume two and a half percent worth of counterparty risk. Rather, what's going on is, if you take a really outside point of view, right at a point of a sort of relatively uninformed one of this space, you're sort of like, look, there's counterparty risk all over the place. I don't know what sketchy and what's not. Like you sort of like using your instincts, but all of these trades have these weird forms of counterparty risk, and you have just sort of, like, you know, from a low information perspective, doing the math and decaying that it is all counterparty risks, but there's no edge after that. From a high information perspective, if you really know this space deeply, then it's a bit different because then you can sort of feel like, Okay, but I actually know that in this particular case, the county party risk is very close to the room, and the edge is still quite high, and other people aren't doing it because they can't distinguish this from the cases where the counterparty risk is real. And so I think really the answer to that is the specter of counterparty risk holds a lot of liquidity out of the ecosystem um, and then there's a lot of money to be made to be made if you can really figure out and pinpoint when there isn't isn't a ton of edge and whether it isn't isn't a ton of actual counterparty risk? Um. I want to go back to that liquidity point, UM, because I think you yourself have been described as one of the biggest whales in crypto. I think someone called you like the moby Dick of crypto whales. And I saw one estimate that you you can move like ten of the market when you make a trade. Um. How does well first of all, how big do you think you are? Um? And how that happens? How does that liquidity risk affect your trading strategy? Like how do you offset that or how do you manage around it? Yeah? So, I you know, the first thing that I'll say, I mean, you know first things that I you know, I was I was a full time trader a few years ago in crypto, building out Alameda. You know, I'm now spending you know, my time focusing on building out fts the crypto exchange I founded, um so I'm I'm you know, have been um a lot less involved in that, but but recently. But the other thing is that you know, Alaminus core core model is acting as sort of a liquidy fighter, an arbitrage firm. And what that means is that like if if you trade large volume, the core trade and this didn't say every trade, but the average core trade is sort of just adulta neutral hedged position. You know, it's it's whatever it is, you know, selling a future against buying spot or something like that, something where you're keeping your book hedged in neutral uh and and so so that's sort of one of the big things there, which is like, I you know, if the risk is in sometimes a lot lower if you don't have a net position on right, But that's not trivially true, and in particular here, you know, the thing that I would add to that is roughly like, you know, well, the nature of crypto that's sort of like cross exchange and not the exchanges don't talk to each other, adds this interesting new twist of risk to it, which is you have to manage your position on each platform, on each wallet, and you see the global view, but no one else does of your book, right, and so you have to deal with exchange specific liquidity risk um and you have to worry about cascading liquidations on a particular product, blowing that one out and not anything else. And you have to make sure that your book and your cackle in your positions can withstand those moves. And it's one of these, like you know, you have to make sure that that you know you can hold your position for you know, longer than the market can stay crazy, so to speak. If you're doing an arbitrage and eventually right you can just lock in your gains and and there's nothing anyone can say about that as long as you can make it to that point to expery or whatever. But you can sometimes seem gigantic dislocations and sort of the biggest one the industry scene was on you know, the probably the bloodiest day in any market, maybe in the history of anything, which is market twelfth of last year, when all global markets burned free fall, you know, COVID was causing and just panic and equities were down or whatever. Crypto was down like SI that day. It was a massive move. You know, you've gone from like nine thousand to four thousand. You know. One way to think about what happened was that it went from nine thousand to sixty or something like that. And then it triggered two billion dollars of liquidations of people who have leveraged long positions in the futures products that were now close to going under water. And so the exchanges began closing down the customers because they put on margin positions they didn't have the margin for it. And as they started closing these positions down dead to sell off the leverage longs and that caused more selling pressure that drove markets down more, which meant even more people had to be liquidated. There's this danger that can happen. She's really weird and wacky, and it's when this ratio gets above one, which is like the impact of liquidations. You sort of look at like how much you move price when you when you you have a logation of a certain size, and how much how large of liquidations at triggers as it jus larger size than it took to get there in the first place, you have this exponentially diverging growing set of liquidations and crashing and rather than like things calming down, things seroically just crashed his rown, you know, in an exponentially increasing way. And obviously it's not exactly what happens, but but it is sort of what happened. And you saw them they playing crash from some exchanges as just cascading liquidations did their thing. So I'm really glad you brought up this day last March because this leads into a sort of bigger question I've been having. Um. So obviously, like that was just this insane selling across all markets everywhere, and everyone wanted liquidity, everyone wanted wanted cash, and it clearly sort of on that day, it was very clear how the global demand for liquidity was intersected with both crypto and traditional traditional financial instruments, and you know, this was sort of seems like it's what's kind of a new phenomenon, because I have to imagine that like back in the old like ten years ago, during the Mountagox days, when it was just a handful of sort of like weirdos who were trading bitcoin, They're like it was probably like pretty disconnected from anything that was happening in the quote real world, so to speak. So it's clear that like and you know on on f t X and I want to get into this in a minute, but you also have actually your trade tokenized forms of equity. So I'm curious about, like how much these linkages are emerging between what's happening in say the NASDAC and what's happening in bitcoin. It feels like it's happening. Um you can see on days lately, for example, there have been some really uh ugly days in the NASDAC textell off, and it kind of looks like on those days, even like bitcoin gets caught up in the down draft. What is your perspective and view right now on a sort of these like cross crypto asset linkage is and the degree to which there is this sort of emerging correlation and feedback between the two that's probably didn't used to exist. Yeah, totally. And you if you go Rewindah, you had big arguments about whether stocks in corrictly were positively or negatively for it, and a lot of people like you know bitcoin is is that the flight to safety asset, like when stocks crash and people don't trust the financial system anymore, they're all going to turn to bitcoin with like a thing you heard a lot, and then other people would say, no, it's sort of like a risky asset. It's gonna be risk on asset and positively correlated. And you look at historical data and you know, you could maybe try and draw conclusion from it, but but really there's sort of like summary was that I don't know, it wasn't totally clear. You know, there certainly wasn't extremely strong or consistent positive or negative correlation. It just like it wasn't clear what what the answer was going to be. So it's sort of like, really the answer historically and then after March Paul Feforn is like, Okay, no, we know the answer now. It's positive. That's fine, positive relation. And that's what happened there was you know, basically when when when people are training crypto, there's a question of what do they see it as? What is the real trade that that most people are doing when they're buying bitcoin, And at least recently, the answer hasn't been that they're betting against etonomy exactly, or or companies or things like that. It's first of all, it's been much more intertwined with currency and with worries about inflation and money printing and and and so so, and in fact, like you can see some of the positive correlation coming from that, people feeling like, well, you know, if it's sort of like you know, the currency hyperinflates, then stops will go up in crypto will go up, and gold will go up, and everything will go up measured into hyperinflating crypto ther currency. So that's a part of what's going on here is just sort of like expectations of future inflation basically, but there are other things too, And you know, it's clear that at least right now, people see crypto as something with huge upside potential and huge volatility, and that's the sort of thing that does well in risk on markets, not risk off markets, because you know, it's sort of like when you've got a lot of money to play around with and you're looking for something to do with it, something with huge upside sounds really appealing, and you're willing to take the risk that it goes to zero. But you know, when everything's crashing, you flaw stall your money. You know, a lot of people are just sort of gonna put in what they're they just think is gonna be safe. So so I think that's sort of like why my best guess about why it seems like that correlation has become just really robustly positive. There's one other thing I would say too, though. In the last year, equities have started to look a little bit more like crypto and I mean you look at game Stop, right, and I mean I think for a lot of people like this is nuts. I've never seen something like this before. Um, there's like these short squeezes, like people are in liquidated and like the markets can't handle it. Liquidity has fallen apart. And what's driving it is not rampant disagreement on the business model. It's like people ship posting on social media and then getting together and then Elon Musk sort of like acting as a standing standard bearer for it. And you're sort of like, this is this has like nothing to do with like analysts publishing reports on on Apple's future dividend stream. You know, from someone who's been in the crypto ecosystem, as they're like, I don't know if there's a word for that. It's a it's a checkpoint. There's thousands of them. We see it every day. That's normal. That's how the market works for all these assets. Like GameStop was behaving like a s Illoqually cryptence the ship colinization of equity markets exactly. And it's like this moment, and I think the beautiful moment of this was the moment that Robin Hood banned buying a game Stop. Do you guys know what happened next? What happened? Um, game Stop crashed because you could only sell it, not buy it. Okay, not talking on Robin Hood. So everyone was either you know, start selling their game Stop or being forced to or couldn't buy it and had money, and so they stopped buying game Stop and instead they bought what he's in retrospect the only possible answer to this question, like as soon as you hear like, oh, of course that's what they bought. Um they bought so as soon as game Stop started crashing, dodge quintin act and and it's like absolutely beautiful of like, yes, that is exactly what this is. Like game Stop and dodge Coin are like very very similar products. This is the idea that like a stock can basically be a token for gambling, right, Like you're just speculating at that point whether it's going to go up or down based on popular interest. Right, you're you know, you're you're speculing whether it's gonna go up or down based on whether you think people think it's gonna go up or down. You know, and and and I think that one of the key insights here, which is really powerful and spooky at the same time, is um let's say the Apple, you know, the Apples, no news and Apple, but someone sells a lot whatever liquidations. Who knows something? It doesn't have any bearing on the company directly, um an Apple's market cat cap crashes from what is it now, like one or two trillion dollars, and it crashes down to a billion dollars tomorrow. Right, what happens next? Apple crushes to a billion dollars. I'll buy it because their cash flows are so high that I'm not worried about. I'm not worried about. Right, you should google like Apple earnings, then they'll take themselves private. At exactly I'm googling it right now, your Apple earnings. And you know Google is claiming that their revenue was a hundred billion dollars. You know, let me see if I can figure out what their earnings are. Here Apple was whatever, you know, revenue of a hundred billion and guessing their earnings were, you know, fifty billion or something, right, So, so so you're sort of like, okay, so you're telling me that for a billion dollars, I can buy an entire company and that company is going to make what is this like fifty times that in the next year, like I'll be sailed by right Warren Buffeties comes in. It's like I'll take it all of it now. And so there's just sort of this lower bound what equities can trade at, which is like, you know, they're discount time, discountry future cash flow adjusted for risk, and you know how long it's gonna take to realize it, and you know whatever else you want to kind of tack on there. But now flick script around, right and instead of Apple, you're talking about game Stop, and instead of going from a trillion to a billion, you're talking about going from a hundred million to right, so it's market cap, you know if thousand excess in a day, and it's clearly not because game Stop is going to be selling a lot of games this year. And now you can ask like what you do, right, Like what's the trade there? And the problem is, well, it's not it's not actually really clear, is it, right. I mean, you could sell it, right, you'd be like, this thing will make seventeen dollars from the next year, and like it's never gonna make more than seventeen dollars. Ever, it doesn't even really have much of a business left. And now it's got a hundred billion dollar market. Yet that's the best short sale ever. Right, you can short sell it, but it's not like it's not like you can short sell and then be like ha ha, earnings were low, Please give me money now, everyone, Like that's not how it works. There's no strict arbitrage in the selling direction. Like if it just never goes down, it just always trades at this insane premium to earnings. You're never gonna make money. You're just gonna slowly be bleeding to borrow costs and and the dividence duo each year as you slowly look sillier and sillier, and and so while there's you know, things can't trade in theory that much below their future cash flow, there is no upper limit to what they can trade at. And if things start to diverge on the upside, you know, the global liquidity priors. Can't you step in and fix it? Like it's not to slew it? That means, so I have a question about f t X, and you know we're talking about equity, So this is a good you have not everything on ft X is actually strictly crypto, and you actually have what you call token ized equity, and there's listed companies sort of like I don't I think it's like like Tesla futures trading and Square Future trading and and this is sort of the even weirder part you have had. And I think you have now equity trading of companies that aren't trading. So I understand, like say like Tesla futures, because you could you know, there's a link between the price of what you could get it on. But you have you have coin based equity trading on f t X. But there's no actual venue right now for coin based trading yet because they haven't gone public. How does that work? And how do you head your exposure in that kind of situation? I don't really get how like you can make a book, so so um, the first thing to how it works there's two different models for it, but though they're ultimately economically very similar. And and the way to think about is basically one version of this. Let's pretend for now that you happen to know for sure coin based was going to go public sometime in the next three months. Right, you know, it's not public yet, but it will be, you know, for whatever it is, and you're a hundred percent confident in that, then what you could do is you can say, okay, here's a coin based feature. It's gonna cash settle to the price of one coin based share um on June one or on you know, August first, or whatever whatever point is safe, right, and and and the point is that like, even if it's not trading now, if you know it will be trading, then you can always do something like cash expiring too, you know, to whatever price it ends up trading. Act. That's sort of one way to think about what you can do with this. And another thing, which is, you know, economically very similar, is you could say this converts into one actual share of coin based stock once it lists. Either way, it's something between a stock and a future on stock, and query what the difference even is if you can't actually do anything with stock today. So that's the way to think about things like the coin based market. Um. And you know, we had an Airbnb market, and you know, I think we're gonna have another one of these coming out soon. So anyway, you know, that's sort of like, you know, that's how you can think about um, these like pre I p O markets, so to speak. And and then you ask like the second part of your question, right, which is like, how the hell do you hedge that? Because it's it's not trading, right, It's not like like someone wants to go buy you know, if someone goes on on robin Hood Right to buy a share of Apple, I mean, fine, you know, robin Hood whoever hands it to some you know pe fops it out, um, some h f T firm that then goes and buys it on on nasdacs. That that's all fine, you know that that works. But then you say, okay, but but point based is not trading anywhere. There's no you know, even if you've got sated out there, where are they gonna buy this? And part of the answer, you know, there's two possible answers to this. Neither of them are perfect, but both of them theoretically work, and one of them is, well, um, you know these sickly in order service you need to actually own equity, right. That would be one model is that, you know, you you have to be like an existing equity shareholder to sell this to short, you know short this que quote and then you're basically converting your equity into the tokenized equity and you know that will sort of settle out once um, once at list and and so and you're you're start backing the shares by actual equity. But of course then in order to get any liquidity in this, you need to like, you know, find someone who happened to be you know, a big VC in point base. The other thing that you do is is the cash subtle futures model. And again this is all playing off of you know, either treating this as a cast subtle future or training it as a uh, you know, as a physically settled thing. Um. Once it lists with the cast settle futures model, what you say is you know, whatever, it's fine, Like, um, you know this is just gonna settle to whatever it'll settle to, and you need to post collateral, right, you know, you need to post if you want to go along or short this. You have some dollars in your account, and you know you'll make or lose money equal to the difference between what it trades that when it lists, and what you buy or sell it at. Now as then anyone can be a liquid e in fighter. They just need to post collateral. And now that doesn't answer the question how they could do this in a risk this way, and the answer is probably they can't. They can't just arbitrage out of this um, and that makes it harder to get liquidity. But they can still trade it. They can do what they think is a good trade. You know, if they see these trading at ten trillion dollar valuation, they just short it. They're pretty sure they're gonna make money. And so it's it's not the most liquid thing in the world, but you can do it. And those are basically the true models I have. What I think in normal circumstances or for traditional financial assets would probably be a very boring question, but in relation to the crypto market, I'm hoping it's more interesting. But like, how do you manage taxes? I'm really curious, Like, yeah, both for f t X and for UM. The hedge fund when you are running it. Like if you think that it's difficult to even open a bank account that's dealing in crypto, I can't imagine what it's like to actually try to, you know, work with the government on crypto profits or losses. Yeah, and and of course you know then you know, and I think the answer is it's about what you think, which is to say, a complete total mess. And and so what you do, well, you know you can get part of the way they're easily right, Like it's sort of like you know, if you like in the middle of your you buy a bitcoin for ten thousand, you sell it for eleven thousand, you just clearly got a thousand dollars of gain somehow, right, So to some extent you can just treat is trading. Whatever trading is trading. And similarly, if you've got an exchange, right, and you know you make a thousand dollars of revenue from fees earned on your exchange, that's that's okay. You know, I've got thousand dollars of revenue. You know, if you don't have expenses you booked as profits. So to first or some of this is easy something it's just a straightforward answer. But once you dig under the hood to some of the weirder cases, what you learn is the same thing you've seen in crypto regulation, which is, you know, the answer isn't known, not in a lot of cases. There's there's a lot of cases where it's just like world isn't ready for that yet. You know, here's one case which I think is pretty clearly something the world needs to think through a lot harder than it has um, which is, let's say that you get a locked token from some like defy staking or air dropped type system. Right, you aren't allowed to sell this token for years, and you sort of forcibly got this. You didn't purchase it, you got it for free. Okay, what's that mean for taxes? And did you just did you just realize a lot of game, you know, equal to the price of an unlock token mark to market when when it was forcibly air dropped on you. Maybe I don't know. I mean you can't you can't not get that game and you can't sell it out for dollars to pay the taxes locked you actually kind of forcibly bankrupt someone if you really think this, right, you just like forcibly air dropped them like a billion dollars marked to market if to pins that the blockchain won't let themselves for two years and then let them have fun with their tax bill, right, So okay, So as soon as you start to get into the weirder cases is when it becomes clear that you know, actually this this is really complicated, and like there's going to have to be a lot of new thinking about how to treat these things. And the world isn't there yet. I know, we're like sort of running short on time. I have like a few can we do like a little like short lightning round questions where I just ask you a few quick things. And so here's one question. There's a lot of stable coins, Now, what can you explain the persistent dominance of tether in this market? Given uh, you know, there's all kinds of always concerns about it and legal questions. What is it about their role in the crypto ecosystem that's so hard to dislodge despite the existence of many competitors. It's a good question, and a lot of people have this question. And and if I understand your question right, you're not saying you know them to be a fraud erant I don't, and I actually think most of the fraud arguments aren't compelling. But I do know also that there's other more regulated ones, right exactly. It's like even if you don't think, if you think it's basically totally fine, it's like okay, but there's also something that no one even has ever thought was a fraud, like like where's the bar here? Really, like, so part of this is just historical. Part of the answer here is, well, it was first and it was the first stable pining crypto at least first one traction, and what that means is that a lot of people started using it. It became listed on all the exchanges. It wasn't just listed in the way that like random tokens are listed on the exchanges. It was the only way that these exchanges made sense. I think it's like a crucial part of the history of tethers back in if you want to have an exchange, a spot exchanging crypto, right, and you want to have like a quote unquote bicklin usd market, how would you do that if you didn't have a bank account, um, and and of course you probably didn't have a bank account because it was really hard to get a bank account. If you're in crypto, no one wants if you want, and it's certainly no one want to get you want if you said you're gonna be, you know, putting customer deposits in it from you know, and all of a sudden you start talking about all the messy things in running exchange and and it just sounds so much worse to a bank than it did when you were a trading firm. What you do then, right, Like, the answer is you listed tether. You got a big poin tether pair. And if people want to get dollars on you're like, go talk together, put your dollars in together, and then come here. And so because of a lot of the industry just grew up with tether, is the answer to how you got dollars and that that gives a huge built in advantage. That's one piece of it. But there's another piece to which is quite important. There is a big East West divide here and in the West you'll generally hear people saying, like, all right, we got one regulated stable point held in the US bank account like audits from US auditing firms, and like got another one which like it's not clear it's been audited. Really, it's probably not regulated. There's like rumors of fraud. It's offshore maybe something Bahamas. I don't think any government can can can really regulate it or or control it, like I don't know which of those sounds like like like like you know, lower risk and and sver. It's like, okay, no, obviously it goes like you know USDC or whatever. Right, But then you go to China and and you say, okay, yet two different stable points, one of which is how is in an untouchable offshore bank account that there's sort of like uncertainty surrounding, and another of which is a highly government regulated bank account, you know, with with like lots of people looking at it and auditing it. Which is those sounds safer? And they're like, jeez, you're telling me that in one of these cases the government has access to this bank account and the other they don't, Like, obviously, I'll take the one the government doesn't have access to. Please And and this is sort of a weird. Big part of it is that this notion of being government regulated is a good thing. And as astanding safety is, you know, so omnipresent has to be obvious in the West in a lot of countries, like government regulated doesn't mean state necessarily, right, Like sometimes that that that's even maybe the opposite right right, And and so I think that's another piece of this is that a lot of people are like, oh, geez, like you know, give me the tether pees, like you're telling me that, Like no one, not even a government or auditor, can confirm for sure what's in this bank account. That is so much safer, Like the worst thing I could do would be with this in a bank acount ever knows about this is a very short question. There's a report that f t X is going to sponsor the name of the Miami Heat Arena. Is that true? So, um, I think there's a sorry, I'm just thinking about what can be said now? Are that there's there's a board vote scheduled for tomorrow. It's a basic answer, um, and you know after that, um, there will be uh, you know, actual final confirmation on this. But but yeah, I think it is now public that, um, that that is the intention. Cool with the toe on ft X, with the Since you trade crypto and equity, do you see people doing cross asset trades themselves? They're like arbitraging square against bitcoin, stuff like that. We do see some of it. It's super cool. It's not large size. I'd be super excity if it were huge size. But it's one of my favorite things to see. And you can even use them as merchant for each other. You can buy micro strategy and you see this flateral for bitcoin future. You can buy bigcoin says flateral for micro strategy or micro thregy short um. They're all we cross margin everything, including the currency, crypto and equities. We do see some of that. You know, we see bitcoin run up after hours and and then sometimes we'll see you know, micro strategy move, We'll see buyers of it, and you know, the actual stock won't be trading on other venues because most places are in point four seven. But Sam, uh, so great to have you. This is like such a fascinating here. I think this is one of the most fascinating crypto discussions I've ever had. So really appreciate you joining us. And well, uh, maybe in a year or something we'll talk about how much more the industry is matured, because there's still just like a million things I want to know take care of. Sam, I really did think that was probably one of the most interesting crypto conversations I've ever had. I mean, you know, I guess I'll say this. I like talking to a lot of people in bitcoin. I find it to be a very fascinated world. But you know, I'm the sort of typical conversations. The government's printing all this money and they're gonna want to like buy fiat. It gets a little repetitive to actually like talking about market structure and how to exploit these seemingly huge obvious arbitrarge trades, but how difficult that is in practice. Super interest think topic. I totally agree, Like I am legitimately excited about crypto market structure now in a way that I didn't necessarily think I ever would be. But the other thing that I thought was really interesting was Sam's point about the equity market resembling crypto more and more like Bitcoin sort of like open the floodgates to people becoming comfortable with trading, you know, basically a token um that's not attached to anything specifically. And I think like we've seen a few stocks like game Stop, like AMC that started to take on that characteristic as well, Like people were just trading a number on the screen. Um. And that's really new for the market, I think, and it's a really difficult one for the existing stock market structure the way it's organized actually deal with, which is why game Stop was such a huge deal in was it January February? I can't remember January. Yeah. And I also I also feel like, um, this like the existence of exchanges like f t X, and we didn't even really get into like its own situation that much, but like f t X as a token that itself now has like an eleven billion dollar market cap um. And so this is like a you know, pretty substantial platform that Sam Sam is built and the ability to do like these sort of like you know, he said a small still, but the ability to do like cross crypto equity trades. It just feels like this like melding of what we call like the crypto world versus the traditional world is only going to get more gonna get more notable as time goes on. And it really started last March, like in that huge crash. But it just feels like they're getting more and more interlinked. Yeah. Well that's the other thing I was thinking about. So if you're doing that sort of cross trading, it feels like you're adding a lot of leverage and perhaps unappreciated linkages into the market. Like all these coins basically come from nothing, and so if they're going to be tied to traditional assets, like again that's something that I don't think has really happened that much before. No, But it also seems like how can you know that there has been a lot of selling of bitcoin as a diversifier and it's hard to see how that persists as they get more linked because you know, again going back to that last March, when you need liquidity, Like sure, intry day whatever, things will be different, but people want diversification for the big moment, and in the big moments, the question is often comes down to like do you have liquidity or do you need liquidity? And so as you sort of like get these into linkages where it's like the world's falling apart, we gotta gross down and you're gonna sell some SMP and you're gonna sell some other stuff and you're gonna sell some Bitcoin, it feels like we're going to have that those linkages are allowed to get even tighter than they are. You know, it would be interesting is if we got a big market sell off because of like some massive inflation shock or scare, and then see what bitcoin did. Yeah, that's like the ultimate question for bitcoin. That would be like the real test. I think of the thesis because yeah, that would be that would be interesting. Okay, shall we leave it there, Let's see it there. This has been another episode of the All Thoughts podcast. I'm Tracy Alloway. You can follow me on Twitter at Tracy Alloway, and I'm Joe wi isn't Thal. You can follow me on Twitter at the Stalwart. Follow our guests Sam Bank and Fried he Is at s b F Underscore Alameda. Follow our producer Laura Carlson. She's at Laura M. Carlson. Follow the Bloomberg head of podcast, Francesca Levi at Francesca Today, and check out all of our podcast at Bloomberg under the handle at podcasts. Thanks for listening to to