Hacker News new | past | comments | ask | show | jobs | submit login
The Sam Bankman-Fried empire crumbled. What happened? (mollywhite.net)
261 points by williamstein 82 days ago | hide | past | favorite | 198 comments

Michael Lewis, author of The Big Short, was in the process of writing a book about the success of FTX and SBF. He personally went to the Super Bowl with them and visited them in the Bahamas multiple times.

Someone on Twitter said, "he just got his perfect ending."

It's gonna be a fantastic book.

Reminds me of the documentary Weiner, which was supposed to be a comeback film about Anthony Weiner's run for NYC mayor but instead became an intimate portrait of a man's life (as well as that of his unfortunate wife) utterly collapsing.

True, only that Weiner was surprisingly endearing in that documentary. Despite at the time being a staunch republican, it's hard not to like the guy, and when the fall comes you're almost rooting for him.

But SBF? The last interview I saw him give was I think with Levine; it was shocking because SBF was describing whence crypto derives its value. And Levine remarked that SBF was describing a Ponzi scheme.

For me, that interview showed two things. First, that SBF quite obviously has no morals whatsoever. And second, he's not nearly as smart as he's portrayed to be.

His whole cred derives from his MIT degree combined with his experience at Jane Street. Jane Street is a great firm, but what they do is really specific, and not necessarily the "coolest" way to make money. It's a lot of tech and infrastructure, less finance and trading chops. (The real badasses are RenTech, for example.)

Anyway, I watched a few videos that he made while working at his crypto hedge fund, Alameda. He described the strategies, and they sounded like the typical market maker stuff - arbing exchange discrepancies yadda yadda. Not the most inspired stuff.

I don't think he lacks morals, but he seems to have a Lex Luthor "ends justify the means" way of thinking. If you're whole moral framework is effective altruism, then as long as the good you do with the money you accumulate is greater than the harm caused in its accumulation by a ratio that exceeds your alternatives, you're all good.

It doesn't even have to do that in an objective, measurable sense. It just has to do that from your perspective—meaning that any harms you can ignore don't count, and any potential good you can convince yourself you could do in the future because of it do.

Yeah, I'm glad to see someone saying this. Something that bugs me about EA is it pattern matches to a lot of the same performative behavior I grew up around with evangelical extremists. Your hypothetical good forgives your very real misbehavior in fact.

There's also just the fundamental stupidity of thinking you can reduce something as complex as human morals and ethics to a karmic checking account balance...

One remark I have on anti-"ends justify the means" is they always ignore the ends. The "ends" might literally be saving real people's lives, whose lives have value. I'm sure there are good arguments against the idea, but you must engage with the ends that you propose we forsake, or you're just not serious.

Isn't this a fundamental problem with consequentialism? If you can convince yourself of a serious enough consequence, you can do whatever you want. Steal, hurt, kill. As long as your actions lead to a .000001% chance of preventing Skynet from taking over in the year 50000 AD, any means justify those ends.

How is that something that can be seriously engaged with? Other than trying to talk some sense into these people about limiting their EA calculations to a reasonably observable time-space continuum?

I always felt the problem with consequentialism is that outcomes are uncertain, whereas actions are not. Therefore you are not guaranteed that an immoral action (means) will result in a beneficial outcome (end).

For instance, you push someone in front of a trolley to save 5 lives, but the trolley isn’t stopped and you end up with 6 dead.

That's essentially Kant's argument as to why you should have universal morals (see the classic not lying to an axe murderer).

If curing cancer for the rest of humanity now and forever required knowingly experimenting on and absolutely killing a certain number of people now, how many people could you justify with an Effective-Altruism ethical framework?

Not meaning to be inflammatory but the same scenario exists in flipped terms: how many people would you let die of cancer rather than, idk, eat the downside of running riskier experiments.

There’s no privileged neutral really. Consequentialism just owns that. Now, that might be a separate question from whether society should act as if there’s a privileged neutral (ie be nonconsequentialist).

Isn’t the inverse more, there will never be a cure for cancer, so how many people can you justify killing to confirm that?

All I know is that when humanity faced the decision with respect to WWII medical experiments , the decision was made to not utilize the data, regardless of potential benefits.

Depends on your expected remaining lifetime of the human race? (since from that you might compute the expected cancer deaths over that time and then you can do a simple comparison)

Longtermists simply assume we will harvest Hawking radiation for hundreds of billions of years after the heat death of the universe.

I wouldn't say they ignore it, but rather they categorically have decided that is an unethical justification. E.g. it doesn't matter if experimenting on patients without their consent saves more lives in the future, it is still wrong. Perhaps you might argue that this is an absolute position they can't really hold to for everything. But I think people prefer to find other ways to justify such cases.

You can take 10 healthy organs from one person without any social connections and use the organs to save 10 terminal patients.

See the problem with utilitarianism?

Unlike other hypothetical examples given here (Skynet, cancer cure), this scenario is possible today.

Sometimes when you see a guy like SBF who fits the MIT hacker/ disheveled smart guy quant stereotype a little too perfectly you have to ask yourself “how much of this is an act?” Most real geniuses I know look unremarkable or boring. One place where people are masters of dressing and acting the part to get rich is Wallstreet though.

Rentech is definitely bad [0], not sure about bad-ass but if evading taxes for years and ending up paying $7B is bad-ass, I guess they are.

[0] https://www.reuters.com/business/finance/renaissance-executi...

That article doesn’t remotely suggest that they are “bad”.

The article says they were advised by Deutsche Bank that they could classify certain trades in a way as to pay long-term instead of short-term capital gains taxes.

The IRS disagreed, RenTech went through an appeals process and eventually decided to settle instead of continuing the process.

There’s nothing evil about trying to legally lower your tax obligations and then being told by the IRS “Nope, you can’t do it like that.”

I've always said their tax attorneys contribute the most alpha!

Evidently he wasn't a "true believer"

he's just one player in the ecosystem, trying milking as much as possible

good job building FTX into one of the largest CEX but.... :)

which videos please

Sorry but this reads to me like someone who doesn't really know what they're talking about. First, I think you have totally mischaracterized or misunderstood that interview. Second, he is obviously a smart guy; I'm not claiming he's a unique generational genius but there are plenty of third party validations of intelligence (MIT, Jane Street) even if you can't tell from the way he speaks. Finally the comment "the real badasses are at RenTech" is laughable insofar as they don't hire undergrads. I would say more accurately that many of the real badasses went to PhD programs but that besides that Jane Street was a highly desired career choice for this graduating cohort.

A Phd is a liability not an asset on wall street. Source: my brother was convinced to abandon his PhD ABD by two friends who did the exact same thing a year or two ahead of him.

The real "badasses" if you insist on calling them that, work about a dozen years for a fund, hit their number, and promptly retire to do whatever the hell they want for the rest of their life. You don't read much about them vs the obsessive people who keep playing the game for the high score contest though, because there's no drama to cover.

There is also a documentary about a timeshare property mogul building the one of the biggest/most expensive personal home ever, I believe costs were over $100 million.

Filming was in 2007, the 2008 financial crisis hit, and the documentary devolves into watching this mogul go bankrupt.

I remember a quote from the mogul, something like "I have nothing but the business. I've re-invested nearly every dollar I've ever earned."


It is almost as if someone has a Twitter bot...


Wow ... so I'm assuming Lewis could smell this a mile away, given his history? (Liar's Poker, etc.)

It's funny that they seemingly weren't in on it -- "No it's not at all like the other things you wrote about".

ML wrote very favorably about IEX team in flash boys so maybe they thought it was like that

I don't understand why anyone takes Lewis seriously after Flash Boys. He seems to just make things up from thin air.

That book was entertaining, but so frustrating to read

These big corporate traders are being paid millions to be the best at what they do (managing stock portfolios) and then it turns out they're actually awful at it and a bunch of genius math kids are running circles around them in the dark. And somehow I'm supposed to feel bad for the lazy, clueless, suit-wearing, trash-talking traders?

Because he was superb since before then, for four decades now. I read Liar's Poker and was hooked from the first sentence.

Look, Michael Lewis is like the Stephen Gaghan of (semi-fictionalized) socioeconomic zeitgeist: he chronicles concepts paradigmatic of an entire generation, people and personalities that affect or transform an entire decade or generation (usually in or about the US, and often Wall Street). And he nods towards the societal implications with a sense of urgency and inevitability but also dark theater. Sure he Hollywoodizes people and quotes, not dissimilar to how the screen adaptation of Mark Bowden's superb 'Black Hawk Down' merged characters and their story arcs, or how loosely 'War Dogs' (2016) was based on Guy Lawson's 2011 Rolling Stone article [0] and 2015 book [1]. [As to 'War Dogs' and Lawson's writings, there's an obvious similar moral fable to three potheads from Miami Beach becoming in 2007 one of the Pentagon’s largest weapons suppliers, by reselling Chinese bullets to US-occupied Afghanistan, and noone at the Pentagon actually wanted to know where private contractors were getting their supply, as the scale of the contracts became increasingly insane.] If you want to more accurately label Lewis' genre "semi-fictionalized socioeconomic zeitgeist", then do.

To be blunt, people have short attention spans, people in the US only read average 15.6 books/yr or median 4 books/yr [2], and moral ambiguity seems to sell worse than clean-cut tales casting characters as good guys and bad guys, certainly on the big screen. But these writers are still fulfilling an important service. Noone actually sits down and reads Inspector-General reports, CSPAN is occasionally great but usually dull to watch, and what now passes for US mainstream media is hopelessly coopted and long ago abandoned investigative reporting [3]. Writers like Lewis, Bowden, Lawson(, Crichton, Grisham [4]) fill a very real cultural void in interpreting what's going on.

[0]: https://www.rollingstone.com/feature/the-stoner-arms-dealers... [1]: https://www.rollingstone.com/culture/culture-news/the-comple... [2]: https://katiecouric.com/culture/book-guide/how-many-books-do... [3]: https://www.fastcompany.com/90646413/why-breaking-points-wit... [4]: https://news.ycombinator.com/item?id=33391995

Speaking of the Super Bowl, their ad aged really poorly. Someone posted a clip from it[1] that has Larry David being pitched on FTX as being a safe way to get into crypto, and he turns it down, saying he's never wrong about these things.

The joke is supposed to be that he's being comically clueless to do so, because it follows a series of shorts[2] of him not "getting" the point of advancements like the wheel, the fork, the toilet, the Walkman, etc. But now he just looks wise!

[1] https://twitter.com/DegenNFTs/status/1590451000523227136

[2] Full ad: https://www.youtube.com/watch?v=_-FQqo46CJQ

He should call it The Big Short 2: Get Shorted.

Maybe "Coming up Short".

I'll pre-order it the day it's available. The big fry.

I'm having a hard time understanding why organizations like the Ontario Teachers’ Pension Plan has money in uninsured crypto funds. Somebody has some explaining to do.

Sequoia Writes Off Entire $210MM FTX Investment; Here Are All The Other Funds That Are Losing Billions In FTX


Edit: 25 billion dollars up in smoke

Ontario Teachers’ Pension Plan faces a hit on investment in crypto trading platform FTX


No mention of it on the website


  He said the investment was part of Teachers’ strategy to learn about the crypto business and whether it gives the right balance of risks and returns. “I don’t think we have the answer to that question yet,” Mr. Taylor said.
Holy shit.. they were openly investing in something they don't understand??

Nobody understands crypto from a financial perspective. That's why all the big crypto companies keep dying in new and interesting ways.

Recently they seem to be dying in the exact same way: reinvesting customer funds.

Sure, but lots of other kinds of financial institutions reinvest customer funds and don't die.

They have hundreds of billions, of which they set aside around 3% for risky bets. One of those risky bets failing isn't a big deal.

As others have said, those who understand crypto either chose to be conniving and make money off of the scam, or they avoid it.

Mr. Taylor wanted to test if he could be conniving enough to profit and he turned out not to be (very few will be in the long run).

Investing a small amount in an industry to learn about it is pretty common with institutional investors.

But no more than $100

OTPP is surprisingly big and very successful overall. And they invested at a $25B valuation, which is very different from investing $25B

> Teachers first bought its FTX stake in October, 2021, as part of a US$420-million funding round. It was one of 69 investors, but FTX listed it first in its announcement of the financing. Teachers has never disclosed exactly how much it invested.

> The pension plan housed the investment ... a portion of the portfolio dedicated to high-growth, yet high-risk, investments. As of June 30, the $8.2-billion portfolio represented just 3 per cent of Teachers’ $242.5-billion in assets.

Teachers ... reported a 1.2-per-cent return for the six months ended June 30. By way of comparison, Royal Bank of Canada’s RBC I&TS All Plan Universe saw defined benefit pension plan assets ... shrink 14.7 per cent over that period.

The Ontario Teachers Pension Plan invests in tons of things, it's like CALPERS. They're present on the cap table of nearly every company you've heard of because they have a hundreds of billions of assets and are chasing yield. They didn't have money on FTX, but had invested in FTX like they do with tons of other VC investments.

Not particularly fond of Zero Hedge, but "Scam Bankrupt Fraud" is kind of catchy.

No good investments are insured and FTX isn't a fund.

Plenty of good investments are insured. There are whole classes of investment that simply would not be sane to do without insurance. Taking an example from every day life: People buy investment properties and most of the time they insure them.

Investment properties not generally insured against loss in market value, no matter how drastic.

Nobody ever claimed they did, but it is an investment, with insurance.

If you want something more financial a short vertical spread [0] is a strategy in which you are purchasing insurance on the risk (investment) you just took (made). There are many such investments of this form.

[0] https://www.optionsplaybook.com/option-strategies/short-call...

What? Shorting call spreads is not how most people would protect a position.

Perhaps you're thinking of a collar? (short calls to fund puts?)

I’m not saying people sell these for insurance. One of the two options in this spread IS insurance.

A short call spread is someone selling calls (option A in the link) and buying insurance on said sale (option B in the link).

Because it's a tiny, infinitesimal part of their porfolio, a fraction of a percent, but one that has a many orders-of-magnitude potential gain.

When dealing with as much money as the OTPP has, you have to form a very broad risk profile: for every super-secure, government-guaranteed dollar, spend a fraction of a penny on a super-risky but absolutely humongous payoff.

I think this should be a warning to anyone still invested in other crypto companies or exchanges. Everything seems like it's going well, until the music stops and the exchange is insolvent.

FTX is not unique. A lot of crypto exchanges are doing just as much shady stuff. When you put your money with these companies, you're betting that "no really, this one is above board!" That's an extremely risky game to play when the crypto space operates with no transparency and no consumer protections.

In a lot of ways, crypto is just relearning the basic lessons of finance that banks and corporations learned decades ago.

Things like "don't allow customers to use your company's assets as collateral" and "using a bunch of leverage on volatile crypto" are much less frequent in traditional finance, specifically due to regulation or the fact that these strategies have tried and failed before.

This isn't to say that decentralized finance is exactly the same as traditional finance, because it isn't. But I think the lesson here is that when you're offering products similar to traditional finance, you should assume that many of the same rules/best practices of traditional finance should apply.

> ...crypto is just relearning the basic lessons of finance that banks and corporations learned decades ago.

To my way of thinking, they’re not relearning anything—they’re specifically using lessons learned from history as a playbook to make, well, illicit gains.

>>This isn't to say that decentralized finance is exactly the same as traditional finance, because it isn't.

This has nothing to do with decentralized finance. FTX was a traditional centralized financial institution catering to crypto.

The leading DeFi apps have had no problems this cycle, and that's because they're fully on-chain, with every transaction validated atomically/in-real-time to ensure its business logic integrity. DeFi's transparency and accountability is what the best CeFi firms strive for.

The big risk with DeFi is the possibility of a smart contract vulnerability, but this diminishes over time as these apps have their source code reviewed by more people and go through more real world battle-testing.

Everyone keeps saying regulation is the key, however in todays thread about 'not getting rich trading options' [0], all I see is people talking about how they lost money or that the game is rigged by the biggest players... and that is in one of the most regulated and well known markets on the planet.

I don't see how regulation fixes anything.

[0] https://news.ycombinator.com/item?id=33547658

While I could say banks are equally guilty of doing shady stuff, the truth is banking is hundreds of years old and as such, has regulations and insurance behind it. Not regulating is how we get messes like the Housing Bubble in the 2000s in the US culminating in the crash in 2011. Crypto being a completely new space has no regulation and no checks and balances which leads to messes like this with companies playing around with their assets and unable to repay their customers when they collapse. I suppose the draw of crypto is inherently that it is a wild, freeform space free from government interference for some people I guess

"US culminating in the crash in 2011"

The official date given by NBER says that the recession began in December of 2007, which put the economy into a fragile state, which was intelligently managed for another 9 months, before circumstances went beyond anything that could be managed. Then the financial meltdown began in September of 2008.

And to amend a bit, to give the whole perspective: And one of the triggers started when around 2005 the FED started raising interest rates.

Which it had lowered more dramatically due to 9/11.


Let that sync in, is my 2c.

(Extra:The FED interest rates takes 2 years to kick in to Adjustable Rate Mortgages which where the first to fall.)

Back then large percent of mortgages were adjustable rate, the amount of those now is a single digit.


It was even worse than that. I worked for Wells Fargo doing credit checks at the time. The entire system was set up to rubber stamp loans without looking too closely. That'd be a problem for whatever downstream sucker bought it. At the time I had a bit of a robin hood attitude about it: if it's not illegal and follows policy that these subprime folks historically discriminated against get a bigger loan so be it. Now I have a more complex view on it, and see it as much more brazenly predatory.


Question, though, and coming back on topic: Did the rising interest rates trigger this (FXT)? There might be a path here, and it would not have the two-year delay of adjustable rate mortgages.

Rising rates usually ends up showing us who was swimming naked.

The business models that relied on a perpetual supply of new capital, & other overly risky gambles become less attractive to investors when you can go get 5-7% on some boring bond with no risk.

You also have liquidation spirals as speculative investors need to sell asset X because of margin calls on asset Y. Which then causes a margin call on customer2 who owns asset X which has now gone down so they go sell some asset Z.. and so on.

Cause, no. Make it more likely, along with the world situation we are in and the supply dysrhythmia, yes.

Recall FTX used a bunch of cash to save failing companies which would create systemic shock (ironic), acquire and invest lately, including in the Silicon Valley. My guess here: "1-2/3-4 Billion" is enough for liquidity, nobody is going to run on us, we are trusted, especially after us saving the world.

There was "bad event rolls" in the system and when they came, nobody can/will/would give them that amount of money.

I am sure in great times, they would also find liquidity/loans more easily -- which is what they are still trying to do. ("Everyone" is keeping cash to the chest, waiting to see what happens right now.)

> And one of the triggers started when around 2005 the FED started raising interest rates

The fed did not cause the crypto crash. It didn't force people to gamble on cryptocurrencies. It didn't force them to not sell, at the top of the naked ponzi schemes they participated in.

Unfortunaely the draw for many isn't that they want a libertarian dream society with no rules, but rather they think they can get rich quick. The "banks are bad" is part of the appeal also. I have a cousin who works in the food service industry (think something like managing a Chipotle) and has put all his liquid assets into crypto and still thinks they will "pay off so he can retire".

As for banks - as soon as these crypto scams fail, everyone will find out just why the SEC and FDIC are there. They were not created just because some politician had an idea for more bureaucracy.

My "hot take" if you will:

SEC and FDIC aren't usually, to the general public, negatively seen due to their mission. Its the fact they don't prevent criminal / unethical behaviors beforehand, and banks get away with alot before anyone intervenes.

I actually don't think the average US citizen is "anti regulation" with this. I think - and I include myself in this - that we see agencies like the SEC (more so than the FDIC) don't act fast enough or often enough, and when they do, the fines are a slap on the wrist vs the profit made by the bad behavior.

In a nutshell, they just aren't acting effectively, and thats why they get a bad wrap.

> Its the fact they don't prevent criminal / unethical behaviors beforehand, and banks get away with alot before anyone intervenes.

A better example is the IRS with a double standard of overzealous enforcement against individuals and small business who cannot defend themselves and complete non-enforcement against higher tiers of wealthy.

In the case of the SEC or FDIC, this has resulted in the near elimination of smaller fish aside from protective carveouts like the Durbin Exempt Interchange.

Yeah, people like effective regulation, it's just regulatory capture is worse than no regulation.

Unfortunately for your cousin, he's the bag holder that the people getting rich quick are going to take for a ride.

As far as I can tell.

yes - he absolutely is. I'm not sure he knows that though.

a) They are similar products banks/certain financial institutions utilize right now, just with corporate debt -- oh what can go wrong in a recession, right?

b) Bank runs happened in 2008 (U.S.), 2010-12 etc. A lot of crypto companies buy insurance similar to what the government says, but lacking regulation and the government behind you makes it as you said problematic.

But: banks are years old and keep making the same mistakes and failures. Big banks (not your corner side community bank that acts as intermediate and its job is community welfare and good relations) haven't changed. I'd be surprised if we avoid another bank related incident, if we enter a recession.

> your corner side community bank that acts as intermediate and its job is community welfare and good relations

Spain used to have a lot of these, and the financial crisis blew almost all of them up: https://en.wikipedia.org/wiki/Savings_bank_(Spain)

Centralized financial exchange markets are at least somewhat regulated and notably FTX.US is supposedly A-OK. Just because some stuff in crypto doesn't have regulations doesn't mean it is a free-for-all.

>Just because some stuff in crypto doesn't have regulations doesn't mean it is a free-for-all.

It pretty much does. The free-for-all disregarding of laws (including laws against fraud) is the stated purpose of crypto. They've been pitching this as Fight Club For Finance since the very beginning. The exchanges with a legitimate appearance are just fronts to get you to the back room with all the unregulated tokens offering insane interest rates. If you didn't come for a scam-or-be-scammed fight to the death, there isn't any reason to use crypto. The crypto bros just missed the part where Fight Club falls apart if you tell everyone about it. Or maybe they got too excited and forgot about that rule.

In light of recent events this appears to be an inaccurate statement

What does A-OK mean? No exposure at all?

As I understand it, FTX outside the US offered many exotic financial products: e.g. 24/7 traded Tesla stock tokens[1], leverage, etc., which I don't believe US based crypto exchanges are allowed to offer(Coinbase stopped offering leverage in 2020, for example). The acquisition offer yesterday did not include FTX.us[2]

This may mean that the US part of the exchange and the associated customer funds were not lent out, but we shall find out.

[1]https://coinmarketcap.com/currencies/tesla-tokenized-stock-f... [2]https://www.cnbc.com/2022/11/08/binance-offers-to-buy-ftxs-n...

I guess we now know that A-OK just means one day away from declaring bankruptcy.

> ...has regulations and insurance behind it.

And billion dollar bailouts. With a few billions every decade or so, FTX probably survives such crises multiple times over.

I mean, banking was already hundreds of years old in the 2000s. So recent bubbles and crashes in traditional markets seem like evidence that regulation, at least the kind that ends up being implemented, isn't doing much to prevent very bad outcomes.

There are two kinds of risk that are fundamentally unavoidable in being a "bank" - that is, any institution that borrows short and lends long. One is liquidity as a result of that duration mismatch: you have enough money owed to you that you've lent out, but it's not due yet. The other is loan writeoffs: you've lent money that you cannot get back, often because it's secured against an asset whose value has fallen significantly.

These can be reduced but not entirely eliminated by setting reserve requirements, LTV ratios, etc., which is what https://www.bis.org/bcbs/basel3.htm is all about.

You can't really eliminate the duration mismatch without eliminating anything recognisable as a "bank", and it becomes much more expensive to get credit and to do basic financial operations.

The problem is that once rates hit zero people will just hoard cash like Keynes suggested with his liquidity trap. The only "funds" available are short term deposits. Nobody is buying certificates of deposits. When the bank issues a loan without a CD it has to create new liquid money, that is the only way.

The answer is quite simple, just keep going down with interest rates even if they are negative. That is fully consistent with the loanable funds model.

Thanks for the explanation, but I'm not sure how it relates to my comment. It sounds like you're saying regulations help because they force banks to keep reserves. I think this helps in the short term, but in the long run creates incentives to deceptively package assets into AAA ratings (as in 2008), and that things like FDIC also just create incentives to get too big to fail. So I stand by my statement.

This is a common mantra - if it's not your keys it's not yours. Leaving your crypto in an exchange is a huge convenience but also a huge risk for exactly this reason.

It's completely the opposite thinking where cash is more dangerous than leaving your money in a checking account for normal people.

I think one thing to think about is that I'm sure someone with integrity is running a crypto exchange and they genuinely are just running a crypto exchange that works. But those guys aren't creating exponential valuations. We can ground this all in real expectations- the CME group is worth 62Bn market cap, Deutche Borse is worth ~30Bn.

Why, why, why would you think that a niche crypto exchange is worth more than the biggest European exchange? We know what the comparables are people!


"5) The full story here is one I'm still fleshing out every detail of, but as a very high level, I fucked up twice.

The first time, a poor internal labeling of bank-related accounts meant that I was substantially off on my sense of users' margin. I thought it was way lower."


"4) FTX International currently has a total market value of assets/collateral higher than client deposits (moves with prices!). But that's different from liquidity for delivery--as you can tell from the state of withdrawals. The liquidity varies widely, from very to very little. 5) The full story here is one I'm still fleshing out every detail of, but as a very high level, I fucked up twice. The first time, a poor internal labeling of bank-related accounts meant that I was substantially off on my sense of users' margin. I thought it was way lower. 6) My sense before: Leverage: 0x USD liquidity ready to deliver: 24x average daily withdrawals Actual: Leverage: 1.7x Liquidity: 0.8x Sunday's withdrawals. Because, of course, when it rains, it pours. We saw roughly $5b of withdrawals on Sunday--the largest by a huge margin." https://twitter.com/SBF_FTX/status/1590709172936798208

I don't understand this. In FTX International terms of service ( https://help.ftx.com/hc/article_attachments/9719619779348/FT... ) they say that users have full title to digital assets, that they are the property of the user, and shall not be loaned to FTX trading and are treated as they belong to FTX trading, and that users control the assets in the account.

FTX was never claiming to run a bank where they loan out users deposits for interest. They were claiming to be running a 100% allocated, straight up custodial vault. Therefore they should be able to simply return whatever assets customers owned on their service, even if every customer wants their asset back at the same time.

If they were actually loaning out customer's assets and running a fractional-reserve bank, that is just straight up fraud, full-stop.

What does “poor internal labeling of bank-related accounts” mean? They didn’t know which of their accounts were fiat vs crypto? Asset vs liability? Something else?

My thoughts exactly. He managed to say precisely nothing in 27 (or whatever it was) tweets apart from the obvious...that he wasn't up to running the company.

WSJ is reporting that money was lent to Alameda based on a conversation that SBF had with an investor (saying that the gap was $10bn) but this isn't what he saying publicly...maybe this is obvious because, if this is true, then it is hard to see how he avoids jail.

At the end of the thread:



That’s quite the (ad-hoc) disclaimer…

"In the event anything I've said is used against me in a court of law I didn't mean whatever it is in that way."

Lol I don't think that would hold up in court.

yeah, that disclaimer is as valid as "a cop can't lie to you"

If you have to slap such a poorly-written disclaimer at the end, maybe you shouldn't have said anything...

Understandable mistake. Many of us who have been bank tellers know how easy it is to accidentally empty a client's account and spend it all on gambling and hookers. It is an honest mistake for sure

Blockchain solves this.

Can someone ELI5 what margin has to do with the bank run?

It doesn’t really, none of this is adding up. But he’s explicitly lied about this stuff in the past couple days (and since deleted the tweets) so that’s probably what’s happening here

Much of the assets traded on FTX were leveraged sidebets e.g. "perpetuals", the underlying assets didn't exist.

Does he has absolute no good counselors? Does he get advice from his roomies in the Bahamas?

This twitter thread, plus some others tweets he has deleted now, could get him into some real trouble.

His legal and compliance team quit

There is a part two available here with more opinion/analysis: https://newsletter.mollywhite.net/p/ftx-analysis

With potential for discussion here: https://news.ycombinator.com/item?id=33547135

That's a link to nowhere, not a 'potential discussion'. On HN it's better to keep the discussion of closely related things in one place otherwise it just gets repetitive. The splits, when they happen, are just more work for the moderators who end up having to merge this this stuff.

This is such an excellent write-up - showcases a veteran Wikipedia editor's ability to synthesize something comprehensive from a huge array of rapidly changing sources.

Seems like a tangent

SBF just made a tweet thread where he tried to apologize, while also distancing himself from FTX.US. Just more lies. Why anyone would believe him is beyond me. My guess, is he's desperately trying to avoid a run on FTX.US to keep the SEC from coming in like a bull in a china shop.

I feel bad for him, he's either going to jail for a very long time, or he'll have an "accident" after losing so many people so much money.

Don’t hold your breath.

Newman, Karpelès and others are alive and free.

Sequoia did a profile on SBF/FTX less than 2 months ago [1], only to mark down their investment to 0 [2]

Some of the passage from that article reads so far removed from reality that I can't even know if it's coherent. Really wild stuff.

[1] https://www.sequoiacap.com/article/sam-bankman-fried-spotlig...

[2] https://twitter.com/sequoia/status/1590522718650499073

They seem to have replaced that page with a disclaimer about the failure. Here was the archived:


> Something of the sort must happen eventually, as the current system, with its layers upon layers of intermediaries, is antiquated and prone to crashing—the global financial crisis of 2008 was just the latest in a long line of failures that occurred because banks didn’t actually know what was on their balance sheets. Crypto is money that can audit itself, no accountant or bookkeeper needed, and thus a financial system with the blockchain built in can, in theory, cut out most of the financial middlemen, to the advantage of all. Of course, that’s the pitch of every crypto company out there. The FTX competitive advantage? Ethical behavior. SBF is a Peter Singer–inspired utilitarian in a sea of Robert Nozick–inspired libertarians. He’s an ethical maximalist in an industry that’s overwhelmingly populated with ethical minimalists. I’m a Nozick man myself, but I know who I’d rather trust my money with: SBF, hands-down. And if he does end up saving the world as a side effect of being my banker, all the better.


The original dotcom implosion was caused by all these "new Internet" companies selling advertising to each other and recognizing that as revenue, even though it was just a swap of ad space that no else wanted. When people figured that out, we had the crash of 2000.

SBF maybe isn't the best quant, but he clearly knows his history.

And it went beyond the internet companies in no small part because those companies were loading up on hardware and software from all the big tech companies. (Oracle, EMC, Sun, Cisco were explicitly called out as the four horsemen of the internet but the effects were felt much more broadly.)

Alternate/additional explanation… the dotcom crash happened after the first quarter of 2000 when the quarterly financials of the big internet hardware/server vendors (Sun but also EMC, Cisco, Oracle, etc) came out and sales were way down because nobody had quite anticipated that their clients would take the servers used for y2k testing and repurpose them, rather than buying new stuff. Then the voice of skeptics had credence to raise other concerns (ad swap, etc) which snowballed.

I still don't get who is actually trading crypto on these exchanges. I have trouble believing it is not 99% wash trades.

I was in the local Co-op supermarket in North London buying some bread & milk the other day. The staff in the shop were talking amongst themselves about how "they were into crypto". I doubt they were comparing their elliptic curve coefficients.

SBF, CZ and his ilk have taken money these people can ill afford to lose in this financial climate and spaffed it up the wall on sports stadiums and billboards of themselves all over the place.

The fan-spread-shit hasn't splattered onto CZ yet.

Do Kwon, 3AC guys, Celsius guys, Voyager guys, yes. Binance / CZ no(t yet).

Binance has sponsored the Alpine Formula 1 team although the logo on the car is quite small.

yeah but even if you're into crypto, you're not selling and buying stuff all day, trading between different coins. So what are all those transactions about?

The part that is not wash trading is what I would imagine "veteran" "crypto traders" just trading the same tokens back and forth looking for just one more hit.

The first time I ever even heard of SBF was when he testified at a Senate hearing as a "billionaire crypto exchange founder." I didn't think much of it, but it was a bit weird - I thought I was pretty familiar with the space and the people in it but apparently not.

That low?

The same story as always, SBF thought he was smarter than everyone else but like everyone in 07-08' he misjudged his risk exposure and it turns out when the market stops going up the music stops.

To me its extraordinary that its just a few years old. Started in 2019, valued at $32 billion just a few years later. For an exchange?

Hence the Crypto bubble. Nothing should be valued at 32 billion in less than 3 years when all it produces is speculative stuff. Nothing screams more "get rich quick" than these things.


Thanks fixed. (I originally put 1999, I'm getting my bubbles crossed.)

The name Sam Bankman-Fried came up quite a lot in Twitter v. Musk. I wonder if he ended up with a stake in the acquisition deal. If SBF was indeed a sizable backer of the Twitter buyout, this more than anything could take Twitter out at the knees.

SBF passed on the Twitter deal[0]

CZ invested $500 million.

[0] https://mobile.twitter.com/sbf_ftx/status/158821335713146470...

Not a fan of Musk but a fan of truth.

Elon was the one who passed on them.

Even if he did invest, why would that weaken Twitter at this point? The deal is done, SBF wouldn't be able to pull his money out of it if he wanted to.

How did it happen? Sequoia invested on a personality cult who thought he was too important to stop playing games while in an investor meeting https://twitter.com/firstadopter/status/1590453539813621761

Matt Levine's talk with SBF is definitely the most bizarrely depressing (or funny, depending on your mood) thing you'll read today. Hell, this week? https://www.bloomberg.com/news/articles/2022-04-25/sam-bankm...

This is a great quick read and it's interesting because it's an abstraction of many other markets (not just DeFi).

Reminds me of Damodaran's[0] focus on price vs. value when evaluating equities:

>I have long drawn a distinction between price and value, two terms that get used interchangeably in both academia and practice, but with very different drivers and implications. As we watch stock indices around the world gain and lose trillions each day, it is worth remembering that markets are pricing mechanisms, not value mechanisms, or as Ben Graham would put it, they are voting machines, not weighting machines, at least in the short term.

[0](Well-respected finance/valuation professor at NYU for anyone not familiar) https://pages.stern.nyu.edu/~adamodar/

"Users' margin" is now apparently code for "how fictional are the assets we're selling users". Could we say that "We don't invest client assets (even in Treasuries)" was technically true on the basis that there weren't any assets to invest? :P

I guess "internal labeling of bank-related accounts" is one of those "dumb" activities that they were counting on amphetamines to take care of for them: https://twitter.com/carolinecapital/status/13790363463003054...

is this nominative determinism kicking in?

Definitely a bit of "Crowley if you think I am holy, Crowley if you think of me foully" too.

What happened was Binance squeezed a competitor, by dumping their token. Then announced stuff on Twitter that made people suddenly withdraw $6B out of it.

This is an unregualated market — but even if it was regulated, what would the regular do after the fact to make it right for everyone inside an exchange or crypto that went to zero?

Regulated fractional reserve banks are usually insured by the regulators.

If the bank goes bust, the government covers the customers' losses up to a predetermined amount (FDIC covers $250k per account in the US).

This helps to prevent bank runs, because people know that they'll get their money even if the bank collapses.

Naive question: % chance SBF goes to jail?

A prediction market on whether he's convicted of a felony in the next three years is at 47%: https://manifold.markets/mr22222222/sbf-convicted-of-a-felon...

Good family with connections, good school, white, male, didnt directly scam anyone important.

Sounds like a guy with ADHD built two banks doing bad things made illegal in the early 1900s

Cghfhguyyrret rfxz z z yuyhtty


Can someone explain to me why they can't solve this by moving the money back from Alameda to FTX? Assuming Alameda is not that bad at investing money, most of the loan should still be there no?

The noise is that they used it for the summer bailouts. I suspect a lot of this $8 billion will turn out to be illiquidity rather than vanished into the ether.

This is exactly what I was wondering.

It's possible (likely) that Alameda has invested that money in iliquid assets or assets that have since lost value. Also, since the loans of FTX to Alameda were backed by FTT, which has since collapsed, even if FTX makes a margin call, the collateral has a fraction of the original value.

I presume this individual cashed out at least some of his net worth in hard assets. Just curious how one could structure a vehicle for those assets that would be resilient to asset forfeiture? Also, what portion of the fortune would be most likely to have been squirreled away?

IF his goal was personal fraud and enrichment (and to be clear, I think this is VERY unproven) it would be far easier to stash some away in a mixed/clean bitcoin or monero account and memorize the mnemonic phrase. No paper evidence, no financial trail.

I would of expected the vast majority of his net worth would of been the same FTT tokens which are now heading to 0.

How much he cashed out (And the timing of this) will be interesting to learn.

He got busted with his hand in the cookiejar, that's what happened

I'm so tired of people who are lauded as geniuses because their lottery ticket bet paid off. Why are people so convinced that being rich means you're smart?

The just-world hypothesis/fallacy: https://en.m.wikipedia.org/wiki/Just-world_hypothesis

Because getting rich is one of the hardest things to do in society. It’s getting power + stability in one fell swoop and is everything humans are biologically built to value.

Well... not really right? Cause lots of people are rich through inheriance. And lots through the lottery.

SBY appears to have gotten very rich through a simple arbitrage trade. Props to him for exploring the space and making the moves... but it doesn't take a genius to see that if you can buy something for cheap in one place and sell it in another for more that you can make a bunch of money if you do it over and over.

So a handful of people. I think that pretty much aligns with what I was saying.

I'd say much more than a handful. Most of the super rich who made it through "smarts" had very rich parents.

See Bill Gates, Elon Musk. Etc.

See, all of the royalty of the world that still exist.

See all of the oligarchs.

It's no shocker the richest people in the world all seem to have gone to the same universities, come from the same families, etc etc.

This comment came a few times but it really doesn’t make sense to me.

How many people actively try to become billionaires like he did in a systematic way?

And contrary to many entrepreneurs, I don’t see where he got especially lucky.

Why does fucking up mean you're not smart? You have to be smart by some metric to get a job at Jane street.

Lots of smart people fuck up in finance, and "dumb" people even strike gold sometimes.

In these discussions there usually isn't a consistent definition of what being smart means to everyone. People with high IQs, academic success, and even strong histories of successfully applying their intelligence can make one off major mistakes, and also engage in long term systematic error patterns.

I love the veritasium video on luck. I highly recommend anyone who is interested in this concept watch it.

ftt is still trading at $3. It's down but not out. this is the same price as late 2020.

Should I buy some? What can I do with FTT?

The crypto community is wild. Literally in a thread about how the wheels are coming off the cart, the project frozen withdraws and possibly mishandled user funds there is the question "Should I buy in?" Also I love that the question of "What value does this provide?" comes after the question about buying it. At least the question came at all.

Warren Buffet puts it this way: "be fearful when others are greedy and to be greedy only when others are fearful". So, honestly, it's a very reasonable question.

I personally wouldn't touch it, but I'm at least curious to hear the heterodox argument. Somebody out there is still buying it at $3, I wonder why.

There's still an opportunity IF you know what you're doing. I wouldn't recommend it.

It's traded elsewhere, and there are derivatives that pay premium. There's also the chance that Binance, Tron, or someone else accept conversions of FTT at some rate that makes the trade profitable. Facts are still coming out. We can only say it's not yet worth $0.

The question people have to ask is, "If I'm considering buying now that it's low, how many people will I be having that thought ahead of?"

It's not dumb or even irrational. Or no more than any thought involving crypto is irrational.

Lots of people have made money catching a falling knife right before a big bounce.

It’s all from that buy the dip meme. I saw something similar where people were asking if they should buy that LUNA coin thing during its meteoric collapse. I think that coin is still in the dumpster so..

$LUNA zeroed, hit a 10x bump, then zeroed again.

it's a common trend, and a lot of people ride that bump for significant profit.

I thought this was obvious bait but go off

No, you should not. It's an exchange token. It's meant to be used as collateral for your trading, instead of USD, BTC, etc. It incentivizes this by giving you reduced exchange fees based on how much FTT you own.

Huh. Why would the exchange want to encourage me to hold their token instead of non-associated crypto?

I honestly can’t tell if you’re being sarcastic…

In the case of FTX: fraud. They were giving people a useless token for tokens that had value elsewhere.

With Binance, you could make the case you own equity in the exchange, but that could have the same outcome as FTX.

It’s worth noting these types of tokens aren’t permitted on US-based exchanges. Likely because they are the most obvious unregulated securities (more so than other tokens).

> What can I do with FTT?

Sell it to someone else.

Subsidize the accounts that have lost everything, I guess.

You can turn $3 into $0 very quickly.

Find some other sucker to hold the bag.


Please do. This type of low effort crap seeping in from Reddit is starting to get annoying. Take it back there, save your words here for when you actually have something substantial to say.

Very informative and clearly written, kudos to Molly!

The irony is that DeFi and blockchain protocols fixes this.

Unfortunately, it does not. Likes of FTX and Binance was created in the first place because on chain transaction is slow and expensive, and people don't like slow and expensive. Just because FTX blew up doesn't mean DeFi is at a better place than it was in the past.

"DeFi fixes this" is like saying "moving back to horse drawn carriages will solve drunk driving fatalities!" Maybe, but no one wants to deal with horse shit.

The lack of transparency in financial services - banks - was the whole reason Satoshi created Bitcoin. The people who choose to use FTX over a blockchain are either ignorant, not understanding their coins are insecure, or willing to take on additional risk to save a few dollars and trade a little faster. They are probably regretting it now that FTX is blowing up.

And speed and costs are already a solved problem - see L2 rollups.

Blockchain may fix it, but couldn't we also argue that blockchain created these problems in the first place?

I will fully admit I have only casually followed the whole blockchain/crypto scheme, so maybe I'm wildly off on something, and would definitely appreciate informed responses.

Without blockchain, there is no "crypto", without crypto, there are no exchanges and related coins, and without those, none of this happens or exists.

How would blockchain be used to fix this?

>How would blockchain be used to fix this?

All of the things that are being reported on this situation would not have been possible on a DEX.

Decentralized exchanges (DEX) exist on-chain. They incorporate automatic market making (AMM), not a third-party market maker with an association to the exchange. The DEX does not have custody of user funds, and it would be impossible to hide a multi-billion dollar hole in an exchange's balance sheet.

Do they work in practice?

Kind of..

Solend is a major one that's currently falling over.. and apparently just disappeared $6M worth of users' assets;


Solend is not a DEX. It is for borrowing against assets. https://aave.com/ is a better lending protocol. Solend has a questionable history, and is closely associated with FTX, which is why the SOL token is down so much.

Patrick gives a good description of the problems in that tweet thread, and why it is affected.


https://uniswap.org/ and https://curve.fi/ are the most popular, and have similar 24 hour volume to established exchanges like Coinbase.

The birth of the stock market also created new avenues for fraud that did not exist prior. This is a new asset class, poorly understood and poorly regulated, and still ripe for abuse.

DeFi would solve some of these problems by shifting the counterparty risk from a man named Sam to an immutable protocol that we can all audit and verify. Look at aave, compound, Uniswap.

The birth of the stock market also created new avenues for fraud that did not exist prior.

Right, but I've never heard anyone say in response to stock market fraud, "You know what would fix this? A stock market!"

The logic path on fixing what is ultimately a blockchain-related problem with more blockchain just seemed a bit circular to me. The blockchain/crypto portion does not really seem to be solving any real-life problems, or if it has, anything it has solved is being outstripped by the problems/losses.

Did the blockchain fail here? Blockchain and DeFi was created to solve these very problems of centralized trust and human financial fraud. Uniswap and Aave withdrawals are working just fine.

Leaving your coins on a centralized exchange kills the reason for a blockchain and DeFi. It’s crypto by name only.

> centralized trust and human financial fraud. Uniswap and Aave withdrawals are working just fine.

Finance always follows commerce. It’s commerce which ultimately creates an organic price floor for assets.

Given that no commerce exists in the cryptospace there is no such floor.

You are describing a decentralized casino, but it’s taboo to say it and hence the word “asset” is being thrown around

Aave withdrawals of DAI stablecoin has been unaffected by the greed of SBF, and unaffected by the speculative price swings of Eth or Bitcoin. You personally may not consider these assets, but many do.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact