This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.
Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.
We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:
-
Shaming.
-
Attempting to 'build consensus' or enforce ideological conformity.
-
Making sweeping generalizations to vilify a group you dislike.
-
Recruiting for a cause.
-
Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.
In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:
-
Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.
-
Be as precise and charitable as you can. Don't paraphrase unflatteringly.
-
Don't imply that someone said something they did not say, even if you think it follows from what they said.
-
Write like everyone is reading and you want them to be included in the discussion.
On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.
Jump in the discussion.
No email address required.
Notes -
The trial of Sam Bankman-Fried begins tomorrow.
As a person that has worked in crypto quant trading[1], I have the tiniest slice of sympathy for him. He still seems like an unsympathetic freak overall, and has done some stuff that seems pretty unethical, and some of his actions are definitely criminal. He has given EA a bad name as well.
There are certainly a lot of process crimes he's guilty of. The fact that the US has pulled his international operations into US jurisdiction means he's in for a universe of pain and if they can't fight that he's going to jail for infinity years. I consider this legal theory a bit dubious but the US has taken the position that it can prosecute crimes that happen in the rest of the world if they even marginally involve US citizens[2]. Is everyone in the world really supposed to follow US laws? That strikes me as a bad precedent; on the other hand, I also do appreciate it sometimes that the US is an international law enforcer of last resort.
That's not really where my sympathy lies though. He knows he was playing a dangerous game. Pretty much everyone who works in quant finance occupies enough legal gray area to worry that they could all be shut down at any time and end up in court. This is even worse in the crypto era, as the position taken by the SEC and friends is shameful, giving very little guidance on new forms of financial technology and telling firms years later by indictment that they were frowning on their behavior all this time.
Many tradfi firms prostrate themselves before the SEC in the hopes of maintaining a good relationship. Even still, reputable firms who were attempting to operate outside of US jurisdiction have been caught with their pants down in the crypto era e.g. Trading Firm A, B and C in the recent Binance indictment: https://www.bloomberg.com/opinion/articles/2023-03-27/the-cftc-comes-for-binance (paywall bypass: https://archive.ph/aMi5Q )
It's still not clear to me that SBF and FTX spent user funds as a matter of course, or if it effectively became spending user funds because so much of their other assets imploded. Though, again, that's not necessarily a crime if you operate outside of US jurisdiction, which is what their international arm believed it was doing. But, that's also sort of secondary.
The primary question I keep coming back to, and I come to this every time there's a large corporate fraud scandal, is: what is fraud, actually? Because it seems indistinguishable from "I thought our business was legit and every indication I had was that it was legit and then it failed and it failed really hard and lots of people lost money".
FTX was a successful business. It was a high quality crypto exchange among many exchanges where the standard at the time was "complete clown show". They were probably the last people I would have bet on imploding and disappearing user funds. The failure is shocking. It's so shocking it's hard to believe.
One thing that's common to these frauds is that people always seem to have a moment of reckoning where they know they're fucked and they can either pack up and go home and face the consequences, or they double down and hope it'll all work out. Indeed, there are some legendary stories from doubling down: FedEx for example where the CEO literally doubled down with their last remaining $5000 in Las Vegas to turn it into a much needed $27,000 to keep the business alive. In this timeline FedEx is legitimate, but if it hadn't worked out he could've possibly gone to jail.
As far as I can tell Uber was based on complete fraud. Its business plan from day one appeared to be: completely ignore taxi laws the world over and just push out a product that was so much better than calling taxis that before jurisdictions knew what was happening they would have tons of passionate users that would be furious if Uber was taken away. This seems to be a resounding success. But it was very much organized crime? If Uber had failed their founder would have definitely gone to jail. In fact he was involved in so much other generally shady stuff that he was forced out. Yet he definitely moved the needle.
Anyway, this isn't meant to be an impassioned defense of SBF, more like my continuing fascination and horror at this alien thing we call modern business. Poor fool tried to play the game of changing the world and got burned. And in this case the burning is fantastic public spectacle.
EDIT: Matt Levine's newsletter today is about SBF's trial, which hit my inbox right after I submitted this comment. Amazing, as usual. https://www.bloomberg.com/opinion/articles/2023-10-02/sbf-s-defense-will-be-tough
EDIT2: As replies have pointed out, I am probably technically wrong for calling what Uber did fraud. Sorry to distract. I should've made my case that Uber was more like a plan to openly disregard and defy taxi regulations across many jurisdictions with the excuse that this isn't a taxi it's a "carpooling app" tee hee. I think this is an insane business plan and it depended on them delivering an amazingly useful app. And if they hadn't succeeded (by delivering an amazingly useful app) they would've all been busted for something rising to the level of organized crime.
Can you elaborate on this? I know a bunch of people who work in quant finance and while it seems completely socially useless it also seems perfectly legally legitimate.
For what?
No, he stole a bunch of money and got caught.
Every trader makes mistakes: around trading on what turned out to be material non-public information, or discussing business on non-recorded channels, or their company learns they've been failing to record chats (or emails) for months without noticing before, or due to a glitch they accidentally sold a ton of shit short they didn't have the right to short, etc.
All of these things are technically illegal but if you immediately reach out to the SEC (or whomever) and fess up they'll probably just slap you on the wrist. But it's completely at their discretion and they might bring serious charges. Or threaten licenses. Or jail. It very much depends on your relationship with the regulator. Woe is you if the SEC discovers these problems before you've noticed yourself and fessed up.
More options
Context Copy link
In HFT (which, admittedly, is not "pretty much everyone who works in quant finance"), the issue is that rules about market manipulation designed for humans operating over timescales of minutes become dangerously vague when applied to computers operating over timescales of hundredths of a second. The rules against spoofing and layering are fundamentally the application to modern markets of the rules against the tape manipulation frauds of the 1920's that were brought in by Joe Kennedy's SEC after the 1929 crash. But those rules define crimes based, at least in part, on dishonest intent (which experienced practitioners can infer from behaviour). Inferring intent from behaviour is harder when one computer is ripping off another computer in a novel way, so it is harder to enforce the rules fairly, and more opportunity for unfairness. There are plenty of people who think that the distinction between what Navinder Singh Sarao was jailed for doing and what Citadel and Renaissance do on a daily basis is Who?, Whom?
The other areas of quant finance (derivatives pricing, risk modelling, quant hedge funds other than HFTs) do not operate in legal grey areas. There are also HFT strategies which do not operate in legal grey areas, such as Jane Street's ETF business.
This seems largely correct.
The who/whom thing it always felt to me like they needed to hang someone and he was the easiest to hang. Citadel supposedly has been a big backer of anti spoofing prosecutions because it negates some of their speed advantages and models. But they get sued too they just have better lawyers and do a fine job getting out of trouble. Witness them refusing to pay a WhatsApp fine.
I also fine it funny that the whole GMC/AMC trades (no idea if they made internally) they did seem to lose some money investing in the fund that blew up. I’ve always thought the GMC/AMC longs were guilty of market manipulation. Basically organizing a group to manipulate shorts into trading (puking) their position. Which would be illegal if three people with big war chests organized the move together. But done in a distributed way is tough to nail any one person for.
Do you mean GME (Gamestop) or GMC (Post-BK General Motors)?
I agree with you that the SEC could have got a conviction for market manipulation against DeepFuckingValue if they wanted it, but equity short sellers are even less popular than market manipulators, so nailing a short seller is safe politically. Both the 1920 Stutz and 2008 Volkswagen corners nailed some of the biggest, best connected players in the market, but there was no interference by the authorities. In both cases the longs ended up running out of cash because in the process of executing the corner they bought the target company for more than it was worth.
Doing that in a commodity futures market, where the shorts include non-financial end users hedging their real-world risk, will get you hauled off in handcuffs.
Ya of course.
I mean the law should be the law. I don’t know if deepvalue crossed the line. I know chamath tweeted short squeeze and encouraged people to play which might cross the line. And he’s lost a lot of people money with his promotions.
Short sellers might not be popular but the fund that blew up does provide a real service. They keep retail losses lower by shorting.
I guess that’s a who/whom case but maybe the distributed nature of it protected him. I tend to think not.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
All the libor manipulation convictions I believe got over turned a decade later. And supposedly the desks had official pressure to manipulate libor lower. A lot of the issues is there wasn’t a true market there.
Spoofing is now illegal. It wasn’t a decade ago until they invented it and started prosecuting people for it.
WhatsApp convos just got Banks fined a ton.
I don't think the Libor manipulation was being done by quants. The false numbers were being submitted by banks' treasury markets desks The pressure to submit the false numbers was coming from the interest rate derivatives desks (or, in the case of Barclays in 2008, the Bank of England). Both desks employ desk quants to babysit derivatives pricing models (and, increasingly, risk models) but aren't run by quants. (Tom Hayes has the classic quant background, but he was working as a flow trader, not a quant).
The US convictions got overturned on appeal, the UK ones did not. The difference is that the UK appeals court ruled that "What rate can your bank borrow at under the standard LIBOR terms?" is a question with (in principle) a single correct answer, and therefore changing your submission based on pressure from another trading desk implies that you were not submitting your best estimate of the true answer, hence per se fraud. The US appeals court said that there is necessary a range of reasonable answers, that choosing one answer within this range as opposed to another based on pressure from another trading desk is not fraudulent, and that the prosecution did not prove that the rigged LIBOR submissions were outside the reasonable range. I am with the Brits on this - everyone involved in LIBOR rigging knew that what they were doing was wrong (both by conventional ethics and by the situational ethics of well-run financial markets). I expect that they also knew that the market for LIBOR-linked interest rate derivatives could not survive their behaviour being exposed (because clients don't want to be ripped off).
All the crooks are now out of jail having served their sentences. Tom Hayes is still trying to clear his name, but there isn't much sympathy for him among London bankers - even if it is technically not a crime, rigging benchmark interest rates is the sort of thing we need to be seen not to do in order to retain the trust of our clients. London as a centre of interest rate derivative trading is weaker because LIBOR has been discontinued.
The people who I do have some sympathy with (and I note that they were not prosecuted) were the people who made optimistic submissions in 2008 in order to contain the financial crisis. You can argue about whether they were cheating according to conventional ethics, but their behaviour made LIBOR (which was used by the industry as a risk-free rate) closer to what everyone expected it to be, helped preserve financial stability in a crisis, and was what the regulators were fairly clear they wanted to see.
I don’t disagree with your analysis. Though flow trader versus quant are fairly adjacent plus as you said he looks like a typical quant. The reason I mentioned libor is it’s a prime example of mixed regulatory messaging and then they ended up in jail.
I remember when Zerohedge use to run articles on things like some stocks were getting so many order messages the system was breaking. That does sound like purposefully manipulating markets to break the system for edge. I would assume there are plenty of things like that of quants pushing the line to gain an edge.
There were, in effect, two separate Libor-rigging scandals. There was no regulatory mixed messaging about the Libor-rigging-for-profit that Tom Hayes et al were engaged in - even if you accept the 2nd Circuit's argument that it wasn't fraud, it was clearly a violation of market norms about treating customers fairly. The regulatory mixed messaging was about Libor-rigging-to-prevent-bank-runs in 2008.
Still think that ties into his quant concerns. No doubt many push the line for edge. So same thing mixed message.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
I agree with the commentators that dispute calling Uber a fraud. Further, you also gloss over just how much of a legal grey area they operating in when they started. New York City is famous for its highly restricted taxi licensing, but even there you've always had the ability to schedule a pre-arranged ride through private livery drivers. This was referred to as "black car service" because the market tended to favor high-end vehicles driven by dressed-up chauffeurs. The big advantage for a taxi license was the ability to pick up customers through street hails or cabstands, and so there wasn't much of a conflict or overlap between the two markets. They existed more-or-less in harmony for decades because "calling ahead" was too much of a pain in the ass and too bougie to be a threat to the taxi industry.
Obviously that changed with smartphones. Suddenly the ability to "call a car" became way way easier, and taxis had the right to worry but they couldn't do shit about it. The regulations that allowed pre-arranged rides existed for decades did not have a "unless pre-arranging gets too easy!" clause. On top of that, Uber's business model had additional layers of protection because of their defense of "We don't own or operate any 'taxis' that's crazy, we just connect you with 'drivers' who will take you places". Taxi companies desperately tried to sue Uber for numerous violations, but they lost every single lawsuit.
In pretty much every jurisdiction in the world there are rules on what you need to do to drive a car for hire and Uber frequently did not require drivers to follow every jurisdiction's rules to start driving. These jurisdictions had the option to shut Uber down for this but they either couldn't get their shit together, they looked the other way, or they did and it became a constant political issue they had to balance against how mad local taxi companies were.
In my town Uber was warned to stop or comply, they ignored the warning, the town fined them, and then Uber disappeared from town. Then the town council got shit for it for years before they changed taxi laws. Uber did some form of this everywhere. As much as I appreciate Uber and love the future we live in, this is a blatantly criminally minded business plan!
You're right, that did happen in some places. I was pushing back on the idea that it was the primary business plan, by highlighting what the law actually was in major cities when they entered the scene.
More options
Context Copy link
In Korea, Uber was kicked out of the entire country, and only foreigners complained, because domestic app companies worked with taxi companies to come up with non-shitty ride hailing services of their own. I think they might have a Korean service now, having miraculously discovered how to work within the confines of existing taxi regulations.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
That's not really modern business. Merchants and noblemen running estates always did stuff like this. Caesar was perennially in debt and a lot of what he did was an attempt to climb out of that debt. All's well if it ends well is an expression for a reason and we have courts, because it often doesn't end well. The only other option is a totalitarian surveillance state and regulatory apparatus with commissars in every business (and you better hope those commissars aren't corrupt and lawsuits aren't filed anyway).
More options
Context Copy link
How is that fraud? Fraud is when you trick someone into giving you something. Disregarding government regulation is not fraud.
Anyway, the comparison is odd. Sam Bankman-Fried, as I understand, lied to his customers about risking their deposits. He effectively stole and then lost billions of dollars. Uber broke laws - artificially constraining the supply of taxis - which probably most informed intelligent people think are unjust.
More options
Context Copy link
My priors are that the progressive technocracy are the de facto rulers of the planet at this point. That means Trump gets his business (of completed buildings, fulfilled contracts, and repaid loans) taken away for “fraud,” and SBF gets a slap on the wrist and another couple billion to play with.
If we were on Reddit, I’d ask the RemindMe Bot to remind me in a year and a half.
I'd take the other end of that action.
SBF is a rich white tech-bro with an eminently-punchable face, who appears to have stolen a lot of money. Blue Tribe benefits in no way from protecting him; he already gave them a lot of money, he's never going to have that kind of money to give again, and protecting him would be terrible optics. So they won't, and he's going away for a long time, 10 years+ easy. There's no reason to believe he has blackmail material or any other kind of leverage, and if he did, we'd see that kick in way, way earlier in the process, not in some surprise courtroom twist right out in front of God and everybody.
The system is crooked, but even crooked systems still have an internal logic and some connection to reality.
More options
Context Copy link
It depends what you consider a slap on the wrist but unless the prosecution has built a terrible case SBF is going away for 8+ years at the extreme leniency end of the spectrum. He won’t get Ulbricht tier (because he’ll likely be less obstinate at the end, and it wasn’t the drug + hitmen hiring combo) or Madoff tier (because he was ripping off VCs, pension plans and tech schmucks rather than very well-connected elderly ladies in Palm Beach), maybe there will be some leniency due to his age, but it could easily be 15 years or more if he’s unlucky.
More options
Context Copy link
I would bet you a fairly large amount of money SBF will get at least ten years in prison but we're anon, unfortunately. We'll have to settle for fake internet points (which, in the spirit of FTX, I will add to my personal balance sheet as a $10,000 asset).
I don’t see a good reason why he wouldn’t get the same years as Madoff. I guess slightly smaller crime but still in the billions. It seems like the same crime to me.
More options
Context Copy link
In all the meme coins that came out a few years ago, I'm surprised there wasn't a coin built around being the currency for personal wagers.
I'm pretty sure there is a coin for that, can't recall its name though.
More options
Context Copy link
There are surely crypto (interpersonal) betting sites, but again you’d need someone to manage escrow and judge the winner, at least.
More options
Context Copy link
Yeah, 10+ years of work and you still can't do anything useful with crypto. What were we all even doing.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
The model of 'success has many legal theories, failure has one' isn't wrong -- Levine has made it and he's pretty far from an SEC skeptic -- but I don't think it's really the big driving factor, here.
Uber, most clearly, was based on illegal conduct (with Uber/Lyft making a lot of kinda marginal arguments about whether they were 'really' unlicensed taxis/limousines), but it didn't lie about the illegal things it was doing (uh, much). FedEx's founder either had a really good winning streak at blackjack or just trafficked drugs, depending on how much you trust the official story, but it didn't tell consumers or investors something that FedEx knew was false.
((Unlike their tracking number system.))
Bankman-Fried is alleged (indictment here) to have told customers that their banked assets would be segregated and protected and not be used as investment capital, and also claimed to banks to be doing different things with different organizations than he was. Many of those customers were U.S. citizens on U.S. soil, and at least one of those banks was an American bank specifically trying to separate whether FTX was transferring coins to dollars and vice versa in a way that triggered extra overhead, or if "North Dimension" was a solely-cryptocurrency business doing just cryptocurrency things that didn't require that extra overhead.
These are frauds for regulatory purposes in a different way than 'I'm not a taxi wink-wink'. This is 'I'm not a taxi, ignore the yellow spray paint, and also I'm going to submit a bunch of advertisements and legal documents where I have "TAXI" on a sign ontop of my car'. There are good philosophical issues the argue that this law isn't Correct -- not because FTX would have been a legitimate company if only lying to customers about the safety of their funds were legal, but because it is genuinely pretty vague and, even in cases like this, it's often used as a simple way to trigger criminal liability rather than describing the actual harm (ie, I care a lot more that FTX was passing customer funds to Alameda than that he didn't put a tiny legal disclaimer).
I'm not sure it was. (and not just in the janky 'is it an exchange or a futures' bulls). It spent like a successful business, and it had an unusually competent customer interface.
But the simply rates-and-flows problem was awful: even the rosiest numbers for FTX had reasonable incomes, ridiculously high outflows, and no particularly compelling argument that it was going to bump the first half of that equation with the second. Worse, FTX was publicly promoting increasing those outflows continuously, on fairly short time frames.
It wasn't an immediate issue so long as a) it had a ton of investors eager to give it cash, b) the paper value of various coins were growing, and c) no one called Alameda's bluffs. But even for those rosy numbers, a lot of FTX's balance sheet consisted of values it have never bought or bet on. Now, Alameda's bluffs got called (maybe to the tune of 8-10 billion USD?) and that exploded the whole mess first, and maybe Alameda's bluffs were a necessary part of the whole scheme to keep pumping coin values.
But all of those bluffs were moved into 'assets' with paper values, and even at those paper values FTX's debts were growing faster. Without Alameda, or if Alameda had only broken slightly-less-than-even rather than at tremendous loss, it might have outlasted other exchanges/futures in the current crypto downturn, but it'd still be a really bad bet.
That's kinda annoying! I think there is (admittedly small) business cases for coins as a way to handle (weakly) committed transfers without a lot of conventional financial system weaknesses or abuses. But you're not going to be a business doing that and spending hundreds of millions of dollars on sponsorships.
I mostly agree with you. Though, I contend during the peak of cryptomania, banking these bluffs as assets sounded "reasonable". Unless you mean something by bluffs that I'm not understanding. Bitcoin grew from $11,000 to $67,000 in a couple of months, there were 10000% APY DeFi yield farming plays, and SBF was on a podcast with Matt Levine describing
innovative new ponzi schemesmagic box technology and instead of derailing the train it blew people's minds even harder.The party is now well over, everyone is hungover and people are regretting their life choices.
SBF was obviously high on his supply. Or he's an extremely manipulative criminal mastermind. But this seems more easily explained by untethered (ha!) speculative mania to me?
I'm not very strong at economics, but my understanding of the problem as a naive outsider :
FTX (bizarrely) didn't have possession of much actual bitcoin, and did have liabilities in bitcoin, during the rush. A lot of what they did have were weird (often self-minted) project tokens, or tokens where a lot of the value was, charitably, FTX employees buying high or making offers to buy high.
Weird self-minted project tokens aren't illegal or necessarily even fraudy, but at best they're ultimately like stock in that they operate as a bet that their majority owner will do well: if FTX's business case didn't work well, the coins would not have transaction volume or value, and if FTX's business case worked fine but their financials failed at a large enough scale it would eventually liquidate enough of each token to plummet their value. Similarly, the more that other tokens were dependent on FTX purchases to weather drops in value, the less dollar value they'd have if FTX folded (or even if it had 'merely' tourniquet them).
On its own, those bluffs aren't necessarily clearly bad decisions. If you have a lot of paper value that's probably worthless and you do nothing with it, you're fine. But both FTX and Alameda were hemorrhaging money, and selling a lot of this probably-worthless paper to other people in exchange for more valuable paper, or using it to justify loans. It's not just that it was in the red, but that FTX was spending like it wasn't in the red. The more and more your business case depends on the chance that many or most of these tokens have, if not as meteoric a rise as bitcoin, at least have a stellar result, that's not (necessarily) illegal, but it's a really dumb idea. The more and more your reserve fund depends on selling things no one's buying, the more it goes from really dumb to hilariously bad.
And that's the best-case scenario -- not that Alameda's bad bets were intentional ways for FTX to pump its sales side, no one explicitly calls anything embezzlement in e-mails, so on. There's some fun philosophical questions about whether this is 'really' criminal intent or just so hopped up on adderall that they don't know the difference between right and wrong, or honestly believed that just the next day every coin he owned was going to take off in a way that would grow them massively. For the purposes of fraud, though, he said he was going to do one thing and did another. For the purposes of business decisions, it doesn't really matter what he thought, or what some random person hearing about Magic Boxes thought.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
Disagree here. It's not fraud. They just exploited one glaring huge loophole in every law that lawmakers ignore it so often that the only reason I could think of it is purposeful. The loophole is - quantity has quality of it's own.
Think of it that way - driving people has always been legal. Carpooling is also universally legal. Being reimbursed on expenses that you incur while helping people is also legal. Posting a message on your condo message board - I am working here, so I can drive back and forth 3 people every day to this point in the morning and evening is once again legal. It is also legal if a friend calls you with - my girlfriend is here, she needs someone to pick her up at the airport, and leave her at my apartment, and I am out of town. Now come Uber - the ever helping carebears - they just provide a software that makes coordination and communication easy, and they just help people with the reimbursements. They are not a taxi company. They are saving the planet by increasing carpooling. Think of the trees. And the saved CO2. And if you have good and expensive lawyers they can explain for a long long time the stark differences between them and the taxi companies. Same with airbnb. They take some informal behavior and turn it up to 11.
The loophole doesn't work. In cities where only taxis can offer rides for hire (including San Francisco at the time Uber started), Uber is an unlicensed taxi, and therefore illegal. Every city that has litigated this point has won, and driven Uber out. The exception is the one that matters - in San Francisco, where it all started, Uber were able to delay for long enough to build up sufficient lobbying power that they could get the California legislature to change the law before the case got to trial.
In cities where pre-booked livery cars (AmE)/minicabs (BrE) are legal, Uber has generally ended up complying with the laws and operating a licensed minicab service.
More options
Context Copy link
Giving your congressman money is legal. Having your congressman vote on a bill is legal. Giving your congressman money so that he will vote a particular way on a bill is illegal.
More options
Context Copy link
This is an interesting way of seeing it. I can think of a lot of regulated activities that can simply be broken down into a series of unregulated activities. Does this ever win in court, or do judges always slap it down with something like "don't try to be cute; the greater picture is obviously that you ran an unregulated taxi"?
*EDIT: I guess this is probably why legal systems rely so much on subjective human judgement rather than applying purely mechanistic rules, as the latter would be tricked by breaking down a regulated activity into a series of unregulated ones, but humans would probably just see what happened for what it is.
The Uber thing isn’t even as clever as you’re describing. At least in New York the rule was you couldn’t do flag stops unless you were a “taxi.” But you could always call someone and arrange a pickup. Uber is that with an app instead of a catchy commercial and a memorable phone number.
There are additional rules for livery car services beyond the requirement to book in advance - notably that people driving commercially should have commercial insurance. My understanding was that Uber initially ignored these rules in NYC but eventually ended up complying.
London was a slightly unusual case in that our minicab law was sufficiently flexible that Uber decided to comply from day 1. For the same reason, there was a healthy ecosystem of minicab apps before Uber arrived. Uber outcompeted them on price and convenience by being willing to lose money hand-over-fist. Now that Uber are making a (tiny) operating profit, they have the same prices as traditional minicab firms.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
See, this is strange. Because I find it incredibly difficult to put myself in SBF's shoes and see why he does certain things, but I find the actual failure of FTX/Alameda...totally boring and banal at this point?
Maybe it's because I've seen a ton of other crypto sites go under. My model is quite simple: things were...looser in cryptoland, SBF and co. were on drugs/in their own little world and so they did dumb things (play with customer money) when they started to lose money. Simply because they could.
People will always fuck up like this given an inch of leeway. Human nature.
I guess I see it as "this is the sort of story I expect to see in a Wikipedia article about banks in the 1800s that led to X regulation"?
More options
Context Copy link
SBF had a weird double-or-nothing gambling philosophy vaguely informed by EA stuff, sure. The weird thing seems to be that he didn't think there would be another severe crypto crash when it had happened like 5 times before in the last 10 years before FTX collapsed.
As @Gillitrut suggests, the SEC (and US taxation regime under the IRS too, if you're talking about personal risk) is essentially the epitome of anarcho-tyranny in the modern age. It has unlimited global jurisdiction to ruin your life; anyone with any money or indeed any business can be fucked for something; the only "defense" is that you (a) help them enough to get other people (or in some other way) so that they let you off, (b) that they show mercy for some other reason (like political connections) or (c) that you have something that makes them look bad. I don't know that all quants operate in this area, the issue for them is that as the LME situation last year showed (where quants made hundreds of millions in losses after the LME reversed $7bn in trades to prevent the whole sector from imploding because pursuing the Chinese metals guy for his assets in China would have been legally impossible; JS etc are now suing), they're not the biggest fish and regulators and exchanges have very little sympathy for them.
Billionaires don't really have a lot of power in the West, or at least most of them don't. If you make it to billionaire status in any way that isn't the most boring, legitimate tech business where your sole pre-IPO investors are legitimate domestic VCs, diligently file everything, pay everything, and operate exclusively domestically in a way that satisfies everyone (and even then...), you can get got for something. The only reason Elon Musk is fine is because the government relies on him for the Ukraine War and for NASA, which means he has leverage. But that leverage isn't because he's 'rich', it's because he controls technology that the state has deemed useful.
That was the biggest BS ever. Trading strategies very rarely just buy and hold, they hedge their exposures. If your trades get cancelled you suddenly have exposure that you didn't account for, and it's likely going to be quite toxic too if everyone else has been handed exposure in the same direction and they all try to get rid of it at the same time.
I agree that it was toxic, but they didn’t really have a choice. Metals is a very small world and if he’d been called and the banks had to try to seize $10bn in Chinese assets (majority located in mainland China and Indonesia via networks of holding companies) they knew they’d be completely screwed. The Bloomberg article said the bankers knew everyone was losing their job and never working in the business again if they fucked it up. As it is there are likely obscure provisions in some ancient administrative bylaw of the LME that mean they can win the lawsuits against the angry traders, but we’ll see. I don’t envy them, but quants also need to know when they’re playing in markets that are a lot less fair than their historical favorites.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
Probably not fraud, since the investors have given the CEO broad discretion to spend and risk company funds in order to generate a return, with the expectation that they may lose it all. (As opposed to FTX customers who had not delegated any such discretion to FTX) Nor was the CEO was the making the gamble for any personal gain. (As opposed to FTX lending funds to SBF's other firm, Alameda).
I don't see the fraud here.
It is an interesting case -- what is going on is that Uber is facilitating the rampant commission of misdemeanor/summary violations of municipal codes. This reminds me of how Google maps will send notifications if there is a speed trap coming up. I'm not actually sure who would be responsible for trying to indict Uber, and what actual statute they would cite as Uber having violated.
I am quite sure that if you take investor’s money, claiming that you’ll use it for building a shipping business, but then lose it all in Vegas, that counts as a breach of fiduciary duty.
As a matter of company law, the objectives of the company in the articles were almost certainly "make money" and the choice of how to do it was protected by the business judgement rule. If FedEx had been a public company at the time, then it might have been securities fraud. But the rules are much laxer for private companies.
Securities law applies to private companies as well, if you take investments and issue equity. That you make a distinction between public and private companies here suggests to me that you don’t have much idea what you are talking about.
Lots of securities laws only apply to public companies - in the US this includes Reg FD and most of SarbOx. The general pattern is that public companies are required to disclose a lot more information than private ones, with making a false disclosure punishable as securities fraud. A private company that doesn't make a formal disclosure can't commit securities fraud by making a false one.
If I was trying to prosecute a company for gambling shareholders' money in Vegas, I would argue that they had made corporate disclosures which implicitly said they were not gambling shareholders' money in Vegas, and were therefore fraudulent. That is much easier to do with public company disclosures than private company ones (which are basically just the accounts).
I definitely wouldn't argue that it was a breach of fiduciary duty - under the circumstances in the Fedex story that argument would be a loser in both England and Delaware that would definitely lose in court because of the business judgement rule.
I assume, though, that the gambling was done under his own name? If he lost the money in gambling, he might have a tough time proving to a jury that he intended to give the winnings back to the company. He would have to prove that he really was gambling on behalf of the company, rather than embezzling the money to himself and gambling it on behalf of himself, that might be tough to do.
More options
Context Copy link
This is a much more reasonable comment.
Yes, but this is precisely why I phrased my comment as such:
I simply don’t see how you can argue that spending all funds on gambling in Vegas is just a business decision, anymore than you could argue that spending all investor funds on buying yourself a villa and a Lambo is just a business decision.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
You might enjoy Matt Levine's take in Everything Everywhere is Securities Fraud
He also has a take on the Bankman-Fried trial in today's issue.
In the FTX case I think the case for fraud is pretty simple.
As an exchange FTX was a custodian of customer assets. If a customer bought 10 BTC then FTX was obliged to hold that 10 BTC in case that customer wanted to sell it or transfer it or whatever.
FTX did in fact represent to customers that it was a custodian of their assets and did have their assets.
FTX did not actually have custody of the relevant assets. They sold them or lent them or sent them to Alameda or otherwise did not retain the assets they were obliged to.
Well, it's fraud if you do that on purpose. It's a mistake if you have an accounting accident. And, as per Matt Levine, I agree your accounting accident is not sympathetic if you were also spending lavishly on yourselves. Where did FTX fall here? I'm slightly inclined to believe in the accounting accident story, but only because I've made multi-million dollar accounting mistakes before, which we thankfully discovered immediately. More than once. And I wasn't in a "move fast and break things" grow to the moon kind of place that FTX was.
I realize SBF arguing "accounting accident!" doesn't sit well with people amongst all of his other horrendous behavior. Poor slob.
I think part of it was an accounting error but it's an error that's ubiquitous in crypto: valuing your shitcoins at their spot price * volume. My understanding is a bunch of the loans from FTX to Alameda were "collateralized" by billions of dollars of other coins Alameda held. The problem comes when you need to turn those coins into dollars and find out the market is not as deep or liquid as you supposed.
Levine again:
If you have one of your companies give another of your companies a loan collateralized by an asset at many times what that asset could actually be sold for, is that fraud?
But this can't possibly be a "mistake" because by the time you're even asking how much your customers' assets are worth, you're already committing fraud.
If a customer entrusts me with 2 BTC and 100 DOGE then I must keep precisely 2 BTC and 100 DOGE for them. However much the value of those assets fluctuates, I'll still be able to pay them back. But if I do some math and determine that I can trade off those BTC and DOGE at a certain ratio, the problem is not with the math. Even if my market predictions are exactly correct, it's still fraud.
It's really not any more complicated than that. It's frustrating when "asset valuation" keeps getting brought up in the context of FTX because it's obviously an attempt to muddy the waters and confuse people about why what FTX did was bad.
More options
Context Copy link
The core of this is a concept called "Mark to Market." The Enron documentary (The Smartest Guys in the Room. 10/10 would recommend) spends some time on it.
It's also the same core mechanism that fucked the whole mortgaged-backed securities market in 2008-2009.
Mark to Market was never really intended for intermediate goods with weird cashflow and long-term appreciation dynamics (like houses). It definitely was never intended for use with Magic Internet Money.
Mark to Market was originally conjured up as a way for oil extraction companies to better value their inventory (oil) as daily markets could fluctuate pretty wildly. The thing there, however, is that that oil was both (a) a thing you had on hand that had a long established market and (b) a commodity that functioned ... like a commodity! There was a spot price and ... that was kind of it. Yeah, there are futures markets, but it's not like a house that has a monthly cashflow (rent or mortgage) but also an asset appreciation profile determined by all sorts of things (mainly location, but also real improvements and hyper local supply/demand profiles). There's just so much more inherent complexity in things like houses that Mark to Market can't really be a stable valuation scheme. [:1]
For a digital currency with zero non-digital assets backing it you're marking-to-a-made-up-market with a formless thought experiment of an asset. Yes ... that's really, really, really obvious dumb as shit.
[:1] To be fair, there are people who will disagree with this and make a (good) point that as long as markets stay liquid enough, they can perform accurate price discovery. I actually think 2008-2009 strongly supports that argument. The crisis point wasn't mortgages going down in value per se, it was in the lack of overnight and short term cash to help firms shift their positions and recapitalize. Firefighting by Bernanke et al. goes into the (quite technical) details of this. To "yes, but," one last time, there are also those who would say that the sheer size of the MBS market and all of the related assets and liabilities made it impossible to "soft land", regardless of any amount of short term credit and liquidity. I can't really refute that because we decided not to test it out back in '09. If we had, and gotten in wrong, we'd be having this conversation in person beneath the rubble of Midgar.
One of the points that got bandied around in the aftermath of that crisis was that the US government could have simply made the mortgage repayments on every delinquent mortgage for less money than they actually ended up handing out - but furthermore, that this wouldn't actually be enough to prevent the crash from happening due to the precarious financial structure involved. Given that you seem to know a fair bit about what happened, is there any truth to this?
Three points worth considering.
"Speed of Money" - It's hard to overstate how fast moving the 2008 financial crisis was. It wasn't hour to hour, it was minute to minute. We actually got lucky that some of the critical events took place late in the week so that there was a weekend (markets not open) to stop, think, and re-orient. The government paying back all of the mortgages directly would've taken too long. Remember during COVID how everyone got checks? How long did it take between announcement and the check arriving. IIRC, a couple weeks (at best). Even when Congress faces a crisis and passes a bill, the machinery of government can only go so fast. The best thing to do is what they did - telling the market to help itself out as much as possible while also passing TARP and constantly repeating "the full faith and credit of the USA is behind this."
Because everything was moving so fast and the scale of the collapse was unprecedented, Congress, the Fed, and the Treasury didn't exactly know if what they were planning on doing was legal. This is also in Firefighting. It wasn't just saying, "have general counsel look at this" it was literally unknown constitutional waters with (allegedly) the potential for personal liability. You can have differences of opinion with Bernanke et al., but those guys had to make some Big Time decisions under massive stress that could've (maybe) landed them in jail no matter how noble their intent.
All of this is to say, the government blanket paying mortgages might have some legal traps within it. Even if it did come back "clean" the time taken to review it would've been catastrophic (see point number 1)
2008 is fascinating to me for tons of reasons. People experienced a lot of pain in the years following and it has a non-trivial part in the political situation we have today. But it could've been so, so, much worse, and I think we avoided it just barely.
Thank you for the well thought out reply! Of course I actually think that the guillotines not being rolled out for the people responsible for the crisis was a societal failure, but I appreciate your analysis here.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
For a more detailed discussion of how mark-to-market accounting affected the 2008 crisis, see chapter 12 of Peter Wallison's book Hidden in Plain Sight.
More options
Context Copy link
Not really. Mark to Market (in technical accountancy terminology, "fair value accounting") first comes in as a way for financial institutions (banks, trading firms, brokerages etc,) with assets whose price varied based on movements in liquid financial markets to better assess their current solvency than historical cost accounting. It was brought in after the S&L crisis - a big part the reason why the S&L crisis got so bad is that you had institutions that were accounting-solvent (because 30-year fixed rate mortgages with interest rates that were now below-market were on the books at par) but clearly could not meet their obligations over a 30-year time horizon, and so they had an incentive to double down with what was in effect other people's money. (SVB was in the same situation, but had a lot of tightly-networked, financially sophisticated depositors with balances over the FDIC limit, so they got taken out by a run).
The thing that made Enron special was that they applied the mark-to-market accounting that made sense for the energy trading division of the firm to things that would not usually be marked to market like long-term contracts for wholesale gas supply, and that this allowed them to recognise revenue and profit earlier than competitors. It also meant that the business model wasn't sustainable because the contracts didn't produce long-term profits, so Andrew Fastow ended up having to double down repeatedly until he ended up ratting on Jeffery Skilling for a light sentence.
You are correct!
I had an EPIC brain snap and swapped the Mark to Market for Master Limited Partnerships in discussing their original application to Oil and Gas. Wow, yeah, my mistake. Thanks for correction.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
With crypto, there are certainly gray areas where the customer knows the company is taking risks, knows the assets are risky, but is not fully aware of just how much risk the company is taking, and the company is happy to keep the customer unaware by burying the risks in the fine print.
But this is not the case of FTX. FTX is just run-of-the-mill embezzlement.
FTX said in its terms of service that customer deposits were the property of the customer, were fully controlled by the customer, were not the assets of FTX, and would not be loaned out. FTX was acting as a custodian. Instead, FTX took the customer funds and gambled with them. This is analogous to a bank manager who is dealing with too many failing loans and so drills into people's safe deposit boxes, takes the money and gambles it at a casino. He cannot do that, that is theft, he needs accept bankruptcy, let the those who lent and invested in the bank to take losses, but allow everyone who trusted the bank as a custodian to pick up their own assets from their safe deposit boxes.. It seems like FTX also took the funds and sent them to SBF's other business (Alambeda), which is even more obvious and blatant theft. This is classic, go-directly-to-jail do not pass go crime.
More options
Context Copy link
More options
Context Copy link