Hello, and welcome to another episode of the Odd Lots podcast. I'm Joe Wisenthal and I'm Tracy Alloway. So, Tracy, we're continuing to talk about the implosion of f t X and Alameda the Sam Bankman Freed empire. Yesterday we talked to two people who are involved in the story in different ways, one who is a market maker who had used and got caught out on the f t X site, another one who had attempted to blow the whistle. But of course we're still just sort of like beginning to
understand the story, right a very fast moving story. But one of the things that people are talking about right now, with the benefit of hindsight, of course, is the infamous Sam Bankman Freed episode of Odd Thoughts. And in that episode we had him on with Bloomberg opinion columnist Matt Levine.
We're talking about how DeFi works, and of course he described it as the sort of magical money making box, right, and so in retrospect, and at the time, of course, people were pretty you know, in shock by those comments. But now everyone's going back and listening to that in retrospect, and it also helps explain perhaps certain aspects of the f t X balance sheet, particularly the tokens that it used to capitalize its balance sheet. Anyway, so we have
to have Matt Levine back on. He's been writing a lot about this story. It's been a must read his coverage, and of course he was there with us on that episode. So Matt Levine, thank you so much for joining us. Let me start kind of with a simple question, which is, what is your summary or how would you describe why f t X collapsed. What's the basic model of what
happened here? So it's a little bit unclear, but basically what seems to have happened is that f t X has a affiliated cryptoh called Alameda h Alameda seems to have made some bad bats somehow or other, it ran into trouble, and to prop it up at some point over the summer, probably ft X started sending a lot of ft X customer money to Alameda, which Alameda then
posted to its own lenders or otherwise lost. And then last week, uh, there were some doubts about fd X in the market, and customers started withdrawing their money from fd X, and ft X realized that didn't have their money, and it went bankrupt in about twenty minutes m um that it did so prior to recent events. What was your impression of the relationship between f t X and Alameda. I have to say I didn't really have much of
an impression one way or the other. Obviously, there are a lot of conspiracy theories about what ft X and Alimata were doing because they were run basically by the same group. Would be the same people who kind of were ball roommates and a big apartment in the Bahamas Um. They were founded by the same person. They were clearly connected, and there were sort of these vague representations about them being separate, but like you know, not a lot of
necessarily enforceability to that, so I didn't really know. I assume that alimated as a market maker on the Atactic exchange, and that you know, it may or may not have had some benefits from being the affiliated market maker. When you hear people complaining about these things, you often think that the essential complaint is Alameda has some advantage in making markets on ft X. It makes too much money by trading against FTX customers because it knows something that
everyone else in the exchange doesn't know. You know, I I never find those stories that like, that's the story in in standard equity market structure, and I don't find it like that exciting. Might be true, but it's not that exciting to be like, oh, they can see the order, but a little bit faster than everyone else. That can be an advantage. But um, but I I don't know
that those stories aren't that compelling. But it seems increasingly likely that that is not at all what was happening here, because the basic problem here was very much not that alimadd was making too much money Elamated it was losing too much money. So the actual like problematic relationship here is pretty much that Alamada lost a bunch of money in fd X just bashed over its customer money to Alamada.
There are other possible stories that are even worse, and I don't want to talk too much about them because
they're all sort of, you know, speculative. But some speculations are basically that Alamada was more or less in the business of losing money on fd X, that rather than Alimated being too good at trading on fd X, Alamated was bad at trading on ft X. And so if you went to FDx as someone else, as as a crypto hedge fund, or even as a retail customer and you traded on FX, you've got really good prices because Alimated was making really bad prices because they were losing
money on every trade. And if you believe that, then you could tell a story that's like, well, ft X became a very popular exchange and made a lot of money in trading fees because it offered this really good trade getting experience to customers by basically having its affiliated
hedge fund lose money on every trade. And you know, that's a great story, except that if the way that the affiliated hedgehund can afford to lose money on every trade is by taking the customers money, then it becomes a really bad story, becomes you know, effectively a Ponzi scheme. So I don't know that that's what happened. There is certainly since speculation about that. But in any case, the sort of standard story of like Alameda has too many advantages and makes too much money by trading on ft
X does not really seem to be true. I want to just that point is really key because when I talked to sort of like professional crypto traders prior to all this, like back up the summer or earlier this year. They really liked f t X and they you know, the liquidation engine, which I think we talked about on one of the episodes with you and SPF was something they really liked it. They really liked the professional It
seemed very professional, worked really well for them. And so this idea that again we don't really know for sure, that maybe some of this was only sustainable because of the loss making trading arm of the SPF empire. Again we don't really know, but that would sort of explain many things at once at once potentially. Yeah, Like there's
sort of two points there. One is is, yeah, the liquidation engine is one possible avenue that people have pointed to as a way that alimated could be like losing money to provide a good customer experience on ft X and then through the back door of taking customer money
like you know, that being a bad story. But the other thing is that like you know, investors like ft X because like it had good technology and it had like good products, and it was you know, there's a lot of like sort of finger pointing it, like venture capitalists and and and sort of everyone who dealt with ft X saying like, you know, they were bamboozled by this, but like there was something there, you know, like the
exchange was good. Now, part of why the exchange was good might have been, you know, sort of deeply problematic financial dealings, but part of it was, like, you know, it was a good exchange that I built it well. Like I often find myself in crypto thinking like there was so much money to be made, Like why did you have to make money in bad ways when like you could have just taken fees and gotten super rich. So I think there's there's an element of that too.
So maybe now is a good time to bring in the infamous Box episode of All Lots, where we had Sam Bankman Freed plus Matt Levin on talking about the business. Matt asked Sam to describe yield farming to him, and he famously described it as basically a magic box that you put money into and more money comes out. And I think, you know, all of us were there when
he when Sam said this, and we were all pretty shocked. Um, we were shocked to hear someone describe defy basically as a pond z. But in retrospect, the really surprising thing about all of this was that it seems like Sam was basically describing what he ended up doing with two tokens so f t T and serum or SRM. Yeah.
I mean, in some ways, it's not that shocking that the head of a centralized exchange would say a lot of stuff and defies Bonzi schemes, right Like, in some ways he was you know, he wasn't talking about crypto
as a sort of broad concept. He was talking about a specific element of defy where you know, in some loose sense, defy as a competitor to centralized exchanges like FTX, and he might not like them just you know, they might be a you know, he might he might have incentive or you know, he might have reason to talk to say bad things about them. So that's one thing.
The one thing I did take away from that was just like a sort of general level of you know, in hindsight, i'd say cynicism, but like I would have said something like agnosticism about crypto, right Like, what he said was not that, you know, every cryptocurrency is this
brilliant project that will change the world. What he said was like, look, you can abstract away from the claims that are made about these cryptocurrencies and just sort of talk about like their financial characteristics, and you know, at the time, the financial characteristics were kind of that they
went up. And so you know, as the guy running in exchange who is making money from crypto volatility and from from interest in crypto, you know, I I interpreted that as sort of like this clear eyed, like I am making money because people keep putting their money into into this asset class and I don't fully understand it, but I'm gonna keep making money for it, which is like, when I put it that way, like a cynical view.
But like, I also think that a lot of people in traditional finance are not, you know, necessarily vouching for the business model of every company who stocked their trade. Right. It's just like, you know, you're a market structure guy. You don't worry too much about the underlying business. So I sort of I was not you know, I was I was shocked that he said it, because it was great tape, But I wasn't like, oh my god, this guy is running upon the scheme, right. I sort of
was like, this is the market structure guy. But yeah, as you said there, you know, f t X had its own token, and the token you not entirely just a magic money box, right, I mean, the token is sort of like shares of stock and f t X. It's not really f t X as a company has SHOs of stocks that did raise money, but like the f t T token is kind of like a tokenized version of, you know, a bet on the future cash loose of f t X. And there's also another token
called Serum, which is sort of a weirder case. Sirium is like this decentralized finance, like a decentralized exchange protocol on the Salona blockchain that you know, it's not exactly owned by f t X, but is you know, it's kind of started by f t X and Alimta and like clearly has ties to them, and it's sort of like, you know, I had interpreted as kind of like FTX is bet on like if everyone really wants decentralized exchanges, then they won't want to trade on ft X, but
like sirium will be sort of like the f t X of decentralized exchanges. And fine, he started these companies or whatever, these you know, these entities and issued these tokens because like you know, you do that in crypto. But it turns out that when he was sort of shopping for a rescuer last week, he was circulating this balance sheet and it basically showed that, like to exaggerate slightly, basically everything at ft X they were all they had
was like these tokens. So they started these companies or whatever, these like you know, f t X and Serum are like you know, trading it to these exchanges. They issued these tokens SERUM and ft T and like that's fine,
that's sort of part of the crypto business. But then it turned out this balance sheet that circulated last week when when SPF was was shopping for rescue financing for fd X, like most of their assets are like a lot of their assets, certainly more than their net equity, were in these tokens, and at like really implausible evaluations. Basically they would issue like a little bit of these tokens. They trade a little bit in the market. It's unclear if anyone other than all I made it was actually
trading them, but they had a market price. And then you take that tiny bit you know, there's like a hundred millions circulating it, and you multiply that by the you know, millions more tokens that held in reserve, and you say, oh, we've got like two billion dollars of room token. It's very hard to imagine anyone putting a
two billion dollar valuation on all those tokens. And so in some sense, like what happened at FTX was kind of what he described to us on that add box where he got a lot of money against the box, and what was in the box was these tokens that turned out to be worthless, and now the money has come. There's a lot of people in crypto who talked about changing the world, and you sort of have like, you know, whether it's like the Bitcoin view of the world about
you know, monetary disorder and the FED. I think there's a different world. I don't know how to articulate vitalis philosophy, but like crypto itself can sort of reorganize society in a positive way. And my interpretation of the box comment
was kind of like yours that he saw. Basically, crypto is a way to make money, and of course SBF has his own vision or head or his own vision of how to change the world through the effect of altruism philosophy but that basically my read at the time was he more or less cynically saw a huge opportunity to make money from all of these de facto ponzis to filter money through his own causes. Now on the tokens Serum and F T T here's something that I'm
like a little confused about. But and you've written about this a lot, so okay, No, there they were holding these tokens of their own creation on their books at these multibillion dollar evaluations, which they almost certainly couldn't get in the market. Who was that for? Because they're not a public company, so it wasn't like they needed to say, show these two investors per se, Like what would what do you think the purpose of these this uh, these
sort of inflated marks on a private balance sheet? Where who are they trying? If the point was was it self deception about the strength or balance sheet? Was about the auditor? Like what what what purpose did this serve? Yeah, it's a good question. I don't know the answer, Um.
I mean, I think part of the answer is, like the place where we immediately see this is this excel sort of rough balance sheet that they circulated to rescue financers, Right, so that was the that's that's the audience we know of right where they were like, we need rescue financing. We actually have positive net equity, so if you rescue us, you know you'll do well. And you know, if you look at the totals on the balance sheet, it's like, yeah,
there's more assets than liabilities. And then you like look at the actual you know, entries, and you're like, oh, these are all just these weird tokens that you can never monetize. That So the immediate audience is like the people who might provide bailout financing, where the audience for
for like these inflated numbers. But you know, that doesn't really answer your question because like, you know, the whole like they had they created these tokens before last week, you know, and like presumably they were thinking before this like oh, you know, we have billions and billions of
dollar was both of these tokens. So I do think that part of the answer and I don't know here, I'm really speculating, but part of the answer seems to be that the f T T token, which is really the thing that brought this down, right, like like what triggered or crisis of confidence in f t X was basically coined us publishing a story about how much ft T tokens Alameda had, and then CZ and Finance tweeting, you know, we're going to dump our our f t
T token and that kind of collapsed everything. So, like the ft T token was was doing some work, and I think the basic work that it was doing was that presumably Alameda was using it as collateral for loans, and I think it's a reasonable guest to say that it was using it as collateral for loans from ft X, you know, possibly of f t X customer money, and that the you know, use of this inflated amount of f t t is collateral was designed to sort of
like make everyone feel a little bit better about it, right, Like you can sort of I don't know if you can necessarily tell your orders we got fair value back in collateral, but you can, like you can tell your junior employees that, you can tell yourself that you can sort of like point to that. I'm not really sure that's what going on. What's going on, but clearly like somebody is posting f tt is collateral for something and having an inflated market value of FTT made that possible.
The serum stuff, it's less clear to me that how that was done, you know what what what purpose that served, And it might be just you know, we shopped that serum around desert looking for valad financing. So, Matt, one of the things that you hear from crypto proponents even now is this idea that, well, the f t X saga, it doesn't show that there's necessarily a problem with crypto. It shows that there's a problem with bad actors in the crypto space. And f t X wasn't a truly
decentralized exchange. It was a centralized one. And not only that, but it was performing all these different market functions all wrapped up in one entity, so you know, brokerage, custodian um, all of that. Is that like a valid excuse at this point. Is the problem here a market structure one, or is the problem something fundamental to the crypto business? Well, I don't know that you can completely separate those things.
Insofar as like, I think it's true that like bitcoin as like a concept is not really affected by this, right, But I also think that like crypto is you know, sort of refers to a broad ecosystem and it's just clearly the case that big centralized exchanges are a big part of crypto, right, And certainly there are people who do not use centralized exchanges and who sort of self custody their coins and use decentralized finance and are very
boulish on decentralized finance and think everyone who uses f t X or coin based or binance or any other centralized exchanges is a chump and is not sort of true to the true meaning of crypto, And like that's fine, but like it is hard to imagine a world where crypto is really big and really mainstream and really you know, big institutional investors are putting money into crypto and there are no centralized exchanges and everything occurs on the blockchain.
I just think, like that is a world where you can do some stuff like that, right, and where where people are building stuff on that, But it's not the world of sort of like mass mainstream adoption. The world of mass stream mainstream adoption is like there's a company with a website and you you know, you have a password, and you can trade on a centralized platform that is fast and not run on the blockchain, and you have like allegible entity that a black rock can custody its
coins within all these things. And so like when one of the biggest and most respected and most sort of like compliant seeming of the big the big offshore crypto exchanges, I'll talk about that, But like when when at the extra was like sort of viewed as a good actor turns out to be very much not a good idea, Like you, I think that it is fair to say, like, you know, the crypt cryptosystem broadly needs to do some
introspection about that. And I think that like some of that is like you have these giant levered financial institutions that don't have public balance sheets and don't really have a regulator, right or like sort of have a regulator, but they have a million different entities and no one knows which one runs the exchange, and the regulator is like you know, the Bahamas, and like they're like one
of the bigger you know, contributors. You know, like there's a lot of like what you don't have is like an ecosystem where the biggest players are headquartered in like you know, the US and Europe and have like a regulator who has you know, fifty examiners in their offices every day looking over their books, and who have capital regulation and stress tests and public balance sheets and all these things. And I don't. I mean, I know, Joe,
you are. You're skeptical about the future for crypto regulation. And I am too, because I think, like the path of cryptos, that these big firms have sort of grown up in a sort of like you know, jurisdictional limbo, where like they all live offshore and are sort of you know, get to choose their own regulation a little bit.
There's an advantage to a bank of being an American bank, right, and the advantage is like the FED, but also like the sort of market confidence that comes of like you're an American bank, You're listed on the New York Stock has change, like your balance sheets are audited by American auditors, like all the like US regulatory apparatus is helpful for banks that want to raise money and trade and in
crypto like that doesn't really exist like it is. It's a little bit, right, there are some you know, crypto entities that make the choice of we're gonna be in the US, We're gonna try to be regulated by the US, we're gonna try to be US public companies because that is going to be really onerous and annoying, and I think they all find that because US crypto regulation is tough, but also it's going to in the long run lead to more confidence because we are going to be more
trustworthy than people who form their exchanges offshore and don't publish their balance. I guess it's sort of like you know, when any financial entity implodes, you know, you can just in finance, you can implode for sort of like benign reasons. You make bets, you thought they were going to go one way, they go the other way. You have to
close up shop, and then there's fraud. Like it is weird when it's an exchange because we don't think of exchanges as making bets and typically you know, they say or holding our or holding your money and they don't take positions, etcetera. Yeah, I do want to push back on that a little bit, just because like like crypto
exchange is not as a broker dealer. So I was so this is this is basically the question I was going to ask, which is like, is the version of this story that I don't know not benign because people are destroyed and like you know, they may have a million people have lost money and you know, massive damage. But is there a you know you talked earlier about like there's this sort of like worse version of the mainstream narrative. Is there like a better version of the
mainstream narrative that we perhaps aren't thinking about. It's tough. I mean, you know, as I was saying, like, the crypto exchanges are not just like people use the word exchange, and so they think of like this neutral thing that doesn't like that just matches buys themselves, right, I mean the crypto exchanges are broken dealers. They hold their customers
money for them, They provide leverage to their customers. And so if you're a futures exchange, like you're sort of taking risk, right, You're taking market risk, and so there are versions of the story where you get sort of mismatched on that risk and then you implode and it's a series of understandable but bad market judgments. I think that the reporting about this has been pretty bad. I'm sorry, I don't mean the reporting is bad. I mean, like the sort of effects that have come up seemed pretty
bad for for fd X, right. I mean I think that you know, fd X is terms of service as we can't use your crypto deposits for our own purposes, and SPF tweeted that we don't invest customer deposits even in treasuries. And then it turns out, as far as I can tell, in fact, customer deposits a ring center over to Alimata to make some sort of bets that
lost money, right, so that's bad. You could tele aversion of the story that's not so bad, right, where it's like Alimated as a customer on the exchange, we made a reason possibly biased, possibly you know, unfortunate decision to extend credit to Alamada, as you know, part of our function as a as a crypto exchange, and like you know, the crypto exchange, that exchange that extends leverage to customers, Alimada as a customer extended leverage to that customer, and
it blew up and took all of you know, took all the exchanges capital with it, and the exchanges capital is usually in some ways anonymous with the with the other customer deposits, Like I don't know, in the abstract, you can tell a story like that, but it's not
not the most likely explanation at this point. So I have one more question, and it's kind of a process question, but with your former lawyer's hat on the complicated structure of the f t X empire, and I think there was a sort of a diagram floating around recently that showed something like a hundred thirty corporate entities, and I think in the diagram they mixed up some non f t X entities with actual f t X entities. But
that just shows how confusing it all is. The fact that there's f t X U S and f t X International and you know, one of them is based in the Bahamas, and Alameda itself is based in Hong Kong. How much does that complicate I guess the resolution and potential prosecution of what we've been talking king about. It's
a mess in terms of entertinuing. Like the thing that I'm really in the dark about is f t X U S, which is like the US entity which exists because U SCRIPTA regulation is really tough and so you can't do it. Like a lot of the products that are the bread and butter of f t X, you can't trade in the US because of some you know, interaction with either securities law or like futures law. And so fd X US offers sort of a subset of ft X products and as you know, possibly more regulated
and possibly safer. And for a while the storyline was that fd X dot Com was bust and bankrupt, but that f t X dot u S was was like
you know, walled off from that. And now there's you know, there seems to be some trouble at ft X u S two and it's just unclear like how intertwined these things are financially, but like, you know, regulatorily, they were different, you know, in terms of like prosecution and in terms of like like broadly speaking, like FTX dot com is in the Bahamas because it first products that would have required like regulatory like let's say registration in the US,
like products that securities in the US and would have to get security registration and like require FTX to registers a stock exchange or products that would be like regulated futures, where the ft X would have to registers the futures exchange. There would be and and you know, the u S regulation is tough enough that it might be hard to actually do those things. And so f t X didn't trade those products in the U S. UM and ft X dot Com is not not I think, technically open
to US persons. But if the question is, can the U s. Attorney's Office for the Southern District of New York go after x ft X dot Com for wire fraud? Which is like kind of the question here, like the answer is yeah, of course, like easy, like and I don't I don't know the exact mechanism, but basically, like you know, at some point someone's moving dollars from a bank into fd X dot com and probably a lot of the people his money got lost at FTX dot
com or not, like you know, US retail invests. But are you know Cayman Islands crypto hedge funds that are run by New Yorkers or whatever? Right, Like there's enough like connection to New York that I don't think, you know, I think in general, like the the federal prosecutors are aggressive and creative about finding US jurisdiction for foreign companies doing things to foreigners abroad when they want to UM, and I don't think this is like this is like
they're one of their harder challenges. Like I think there's gonna be enough kind of US stuff here that if that, like US prosecutors are gonna feel very comfortable looking into it, which is not to say that I'm you know, I don't want to pre judge whether they'll find any crimes, but like I think that they'll find jurisdiction they did. In other words, the problem won't be it's like, oh, they committed these crimes, but during the Bahamas, we can't
do anything. If they conclude the door crimes, they'll find a way. Yeah, being in the Bahamas is a way to avoid securities registration requirements is not a way to record as criminal frog law. Matt Levine, thank you so much for your perspective. Great to have you back, and this was a very helpful conversation. Yes, thanks Matt. Obviously, Tracy talking to Matt is always helpful. And I want
to start with actually just that last point. We don't know as of recording this right now on November six, whether there is going to be criminal charges ultimately brought in the f t X case, but it's sort of useful to know that simply by virtue of technically domiciling ft X. I guess not technically because that's where they were based in the Bahamas that that is not some sort of like escape card so to speak. Yeah, absolutely,
certainly seems like it isn't. The other thing that's standing out to me as I think back to all of this is, I think the f t X LaMDA relationship. I think there was always a perception in the industry that there was something there, that there was some sort of you know, beneficial relationship as Matt described it. Maybe they were getting like early info on tokens that we're about to list on f t X and then flipping
them and things like that. But in retrospect, in retrospect, it seems like what they were doing was pretty simple and just going massively long various coins, Like there wasn't really a secret sauce. There wasn't necessarily like some amazing liquidation engine. It seems like they were just taking outside
bets on digital tokens. Yeah, you know, there's the thing I think it's it's the James Street pedigree, Right, It's not just Sam, but others who he worked with who had gone through this sort of like famous quant trading shop Jane Street, which we talked about the same the first time, and so you sort of get the impression like, okay, Alameda is going to recreate Jane Street in some way
but for crypto, but I mean maybe not. And this idea, you know, I think everyone writ is Matt was saying, and you know, one way or another, it seems like Alameda lost a ton of money. So either they were trying to do something sophisticated and they couldn't do it, or maybe they were taking bad directional bets, or they were spending badly, or it was sort of like this lost leader to make the ft X platform more attractive
to institutional investors. But whatever it is, it does not seem to have been that sort of like super brilliant quawn shop that you would sort of expect from like Jane Street alone. No, and the other thing that stands out to me, and this has been a recurring theme on the podcast recently, but this idea of truly decentralized
crypto versus centralized players and exchanges. And I've said this before on Twitter, but I very much agree with Matt this idea that like everyone is going to be running their own node and storing their keys in like cold storage, and no one's ever going to need to interact with a centralized player, I think is unrealistic because not only a not only is a lot of crypto stuff still very difficult for people to figure out and do on their own, but be people want to do stuff with
their money, right, You want to get additional credit for it, you want to trade it, you don't want it necessarily just keep it on a wallet and never actually use it. And that's what a lot of the centralized entities and exchanges actually did. Yeah, I mean, look, you can trade
and speculate on a defy exchange. But I think Matt is right, and it's I think it's a definitely true that in the last few years there is no way crypto would have gotten as big had it not been for numerous websites where you log in and you have a password and you move your money onto a website like, which is like the idiom of normal finance, right, And so maybe there is like an alternate scenario in which everyone in crypto like really adhered to like these sort
of like defy principles, and maybe that would have been better, and maybe that would have been more you know less. I don't know, it's something, but I don't think there's any way in that world it would have gotten nearly as big or like attracted this sort of like money risk seeking behavior in such volume. Then I agree, al right, shall we leave it there for now? Let's leave it there. This has been another episode of the ad Thoughts podcast. I'm Tracy Alloway. You can follow me on Twitter at
Tracy Alloway and I'm Joe wi Isn't All. You can follow me on Twitter at the Stalwart. Follow our guest Matt Levine. He's at matt Levine. Follow our producers Carmen Rodriguez at Carmen Armon and Dash Bennett at Dashbot. And check out all of the Bloomberg podcasts under the handle at podcasts, and for more Odd Lots content, go to Bloomberg dot com slash odd Lots, where Tracy and I blog to push the transcripts, and we even have a once a week newsletter that you should check out and subscribe.
Thanks for listening.