It’s a 19th-century story; it’s a 21st-century story. The process takes a decade and a bit. A philosophical idea starts in Oxford, spreads rapidly, finds a believer who becomes famous and grows in fame with him, thrives despite loud criticism, and then undergoes a huge scandal as the famous person does something which confirms the charges the movement’s detractors have been making all along. In the 19th century, the relevant Oxford movement was the Oxford Movement, an Anglo-Catholic revival, which first took shape with the publication of Tracts for the Times in 1833, grew in fame along with its great polemicist John Henry Newman, was denounced for being a semi-covert form of Roman Catholicism, and then suffered a colossal scandal in 1845 when Newman became the highest profile convert to Catholicism in the history of England, thus confirming what critics of the movement had always claimed. In this century’s version, the relevant Oxford movement is Effective Altruism, which began in 2009 with the foundation of the organisation Giving What We Can. Its members agree to ‘earn to give’ – to undertake high-paid work with the intention of giving away the proceeds. Critics of the movement said that it allowed all kinds of anti-social individual choices so long as the Effective Altruist involved made them with the intention of achieving a long-term payoff to benefit humanity. By 2021 the highest profile Effective Altruist, Sam Bankman-Fried, who had started earning to give in 2012, had become the richest person under thirty in the world, with a net worth estimated at $22.5 billion. His crypto trading enterprises collapsed in late 2022 and he is currently on federal trial in New York on eight charges of fraud and theft, facing a maximum prison sentence of 115 years. Critics of this latest Oxford movement have not been slow to say they told us so. A sample observation, from Twitter: ‘A culture that is more or less custom-built to produce shameless, ruthless, notionally goal-oriented sociopaths is somehow still producing more and worse versions of that person than I thought possible.’
On the same platform, before he owned it, Elon Musk once wrote that ‘the most entertaining outcome is the most likely.’ What would be the most entertaining outcome in this case? It might well involve an admired writer of narrative non-fiction in position to tell us the before, during and after of Bankman-Fried’s vertiginous rise and terrifying – or, let’s face it, to many, deeply gratifying – fall. That’s exactly what has happened with Going Infinite, Michael Lewis’s new book. Lewis met Bankman-Fried in 2021, at the instigation of a friend who was thinking of going into business with him, and who wanted to know ‘who was this guy?’ Lewis began spending time with SBF at his company’s headquarters in the Bahamas in an attempt to answer that question.
Bankman-Fried’s parents are both Stanford law professors, Barbara Fried (public policy, ethics) and Joseph Bankman (tax). They were philosophical utilitarians who did not celebrate holidays or birthdays. SBF, as he is known online, grew up aware that his mind worked differently from most people’s. Even as a child he thought that the whole idea of Santa Claus was ridiculous, and when he encountered the fact that some people believe in God, he was incredulous. His takeaway: ‘Mass delusions are a property of the world, as it turns out.’ He ‘had to accept that there was nothing he could do about this ... He simply came to terms with the fact that the world could be completely wrong about something, and he could be completely right.’ Much of what he was taught at school he thought ridiculous, especially the humanities. Lewis observes that SBF’s parents were ‘afraid for and of’ their elder son.
In the autumn of 2012, at the age of twenty, SBF’s life began. He was a physics student at MIT when two things happened. First, he applied for an internship at the Wall Street trading firm Jane Street. A good rule of thumb for finance outsiders: if you, the outsider, have heard of a company, it’s doing something wrong. The most successful firms prefer as little publicity as possible. Jane Street is a high-frequency trading firm, a hedge fund, which employs super-smart super-numerate people to exploit angles and edges in markets. Their hiring process is unusual, and relies on making prospective interns play games with rules and odds which shift violently in the course of the game, just as financial markets shift. SBF’s try-out couldn’t have gone better. ‘By the end of the day it was clear that it was by far the best I’d ever done at anything.’ Jane Street offered him an internship, with a job on graduation.
Around the same time and out of the blue, SBF received an email from a 25-year-old Oxford academic: William MacAskill, the co-founder of Effective Altruism. ‘Sam never learned how the guy had found him – probably from the writing Sam had been doing on various utilitarian message boards.’ MacAskill exposed SBF and a group of Harvard students to the theory behind EA. The thinking was as follows. If you go to medical school and become a doctor in the developing world, fine, you save lives; but the difference you make to the world is only the difference between the lives you save and the lives saved by the other doctor who would have done the work in your place. If you become a banker, on the other hand, and give away all your money, the difference you make to the world is all that money, and all the extra lives it saves – because we can be confident that the banker who would have done your job in your place wouldn’t have used the money for that. EA estimates the cost of saving a life as between $3000 and $5000. So a banker giving away millions across a career can save tens of thousands of lives.
This made perfect sense to SBF, who bought in immediately and fully. ‘What he said sort of seemed obviously right to me,’ he told Lewis. ‘It gave me a practical course of action.’ SBF had an exceptional facility for thinking probabilistically; for experiencing events and people not as fixed entities but as a range of likely outcomes. This kind of thinking was central to both his EA beliefs and his working life. What they had in common was the idea of Expected Value: the likelihood of something happening, multiplied by its impact. An unlikely bet with a huge pay-off could well offer a higher EV than something which looked smaller and more sensible – and this applies to both ethics and finance. Say, a charitable exercise has only a one in a billion chance of succeeding, but if it does succeed, will save humanity. (This might sound like a deliberately outlandish example, but the EA movement is increasingly focused on the question of existential risks to all mankind, including people not yet born, into the distant future. See Amia Srinivasan’s LRB piece of 24 September 2015 to see the strange places this idea can lead.) You have $25,000 to give. Should you give it to a political party, to help fund your sick neighbour’s cancer care, to build wells in Africa, or to this one-in-a-billion long shot? The EA calculation is that it costs $5000 to save a life, so your long shot represents good value. The EV of your bet is $35,000: 7 billion lives x $5000 each, divided by a billion for probability. Good deal! Though bad luck for your political party, your neighbour and those Africans.
We’ll come back to the ethical ramifications of this thinking about Expected Value. In terms of financial markets, EV was extremely effective. SBF was good not just at thinking in terms of probabilities, but also at adjusting the probabilities as circumstances changed – and, crucially, he didn’t find it stressful. ‘It wasn’t that he thrived under pressure; it was that he didn’t feel it.’ He began making a lot of money at Jane Street, and gave at least half of it to EA, mainly to causes involving animal welfare. He worked at the trading firm for three years, from 2014 to 2017, and didn’t take a single day off. When he finally took a break, in the summer of 2017, he had time to think about other things he might do to make money or make an impact or both, and hit on the answer: cryptocurrency. He knew next to nothing about crypto, but could do the maths: roughly a billion dollars of crypto were traded every single day:
If he could capture 5 per cent of the entire market (a modest number, by Jane Street standards), he could make a million dollars a day or more. As the markets never closed, that meant profits of $365 million a year or more. ‘That was my ballpark estimate,’ he said. ‘It seemed crazy. So I just cut it by a factor of ten. I was thinking thirty million a year. But I felt embarrassed showing that number to anyone else. They’d say: “Fuck off, Sam.”’
He quit his job, went to live in Berkeley and began setting up a firm to trade crypto using the techniques he had learned at Jane Street. The new company was called Alameda Research and it was owned 90 per cent by SBF and 10 per cent by his first hire, Gary Wang, a friend from college. Wang was an EA, an almost entirely silent coding genius who was one of Bankman-Fried’s secret weapons in his quest to make ‘infinity dollars’. Caroline Ellison, an EA and a Jane Street colleague who was to become SBF’s off-and-on girlfriend, was another early hire. (Spoiler: Wang and Ellison have pleaded guilty, and are the prosecution’s star witnesses at SBF’s trial.) The company was staffed by EAs, who understood the mission and could be trusted, an important issue when trading crypto because anonymous untraceable electronic money is, it turns out, freakishly easy to steal.
At the time of writing, cryptoworld hates Bankman-Fried. There are a number of reasons for that. The main one is that the crypto world is touchy about how prone it is to scandal and disaster, and the collapse of SBF’s enterprises has been damaging to confidence in the entire crypto ecosystem. But a subtler reason for the loathing of SBF is that crypto believers could smell that he wasn’t truly one of them. Crypto is an ideology, an anti-government, individualistic belief system, one that SBF didn’t really share. He was a trader, and he saw in crypto a big unaddressed market with a lot of inefficiency: in plain English, he was a fox looking at a henhouse. If the market for yams had been growing at the same speed and with the same inefficiencies, he would happily have traded yams. Lewis takes a similarly brisk approach, and doesn’t go into the highly charged questions of how crypto works and whether its value is based on anything real. This works well for the SBF story, though it is frustrating for crypto sceptics, who think the answers to those two questions are ‘very badly if at all’ and ‘no’. Sceptics might like to look at Zeke Faux’s Number Go Up for an excoriating attack on crypto and all who sail in her, SBF very much in the forefront. To get a flavour of the book, consult the opening:
‘I’m not going to lie,’ Sam Bankman-Fried told me.
This was a lie.
Faux’s indignation is amplified because he focused his reporting on the ‘stablecoin’ Tether, which is effectively the on-ramp and off-ramp to crypto from the legitimate banking system. That means that Tether is crucial to the way crypto runs and if Tether is fake, or collapses, so does crypto. Tether’s pitch is that it is backed one-to-one by real US dollars residing in real-world US dollar bank accounts. There are 66 billion Tether in circulation, so that means that there are $66 billion on deposit somewhere, right? Right? Except nobody can find those deposits, and Tether – co-created by a former child actor who was in The Mighty Ducks, backed by a Bahamas bank run by the inventor of Inspector Gadget – is extremely evasive when it comes to supplying any details, about anything. Like Faux, I considered writing about Tether, because in the fifteen-plus years I’ve spent writing about the world of money, I’ve never seen anything dodgier. It’s been obviously about to blow up for a long time. And yet it hasn’t blown up. It’s enough to drive the observer slightly nuts. You can find that energy in Faux’s funny, furious book.
The fact that SBF wasn’t a believer didn’t mean that he wasn’t good at trading crypto. By January 2018, Alameda had capital of $40 million and was making half a million dollars every day. This early success allowed SBF and Wang to raise more money, and soon they had $170 million, mainly from rich members of the EA community. (You might have thought the altruists were chill and altruistic about what happened to their money. They weren’t: the donors demanded a 50 per cent rate of return.) Unfortunately, just as Alameda’s capital base went up, so did its losses, and before long, instead of making half a million dollars a day, the company was losing it. In addition, and separate from the trading losses, millions of dollars in crypto disappeared who-knew-where.
A big part of the problem, in the opinion of many of its staff, was Sam, who was primus inter pares of the five-person, all-EA leadership team at Alameda. ‘He was demanding and expecting everyone to work 18-hour days and give up anything like a normal life, while he would not show up for meetings, not shower for weeks, have a mess all around him with old food everywhere and fall asleep at his desk,’ said Tara Mac Aulay, an Australian maths whizz and a senior member of Alameda. SBF created and wanted to use an automated trading system called Modelbot, which would buy and sell five hundred cryptocurrencies on thirty mainly unregulated mainly Asian exchanges. The problem was that if Modelbot went wrong, one colleague said, ‘it was entirely within the realm of possibility that we could lose all our money in an hour.’
One evening, Tara argued heatedly with Sam until he caved and agreed to what she thought was a reasonable compromise: he could turn on Modelbot so long as at least one other person were present to watch it, but should turn it off if it started losing money. ‘I said, “OK, I’m going home and go to sleep,” and as soon as I left, Sam turned it on and fell asleep,’ recalled Tara. From that moment the entire management team gave up on ever trusting Sam.
This period, in 2018, is crucial for understanding the story of Bankman-Fried. Long before he became rich and famous, he was an intensely divisive figure inside the world of EA, about whom many close colleagues had serious doubts. ‘Tara had long since decided that he was dishonest and manipulative. Ben [West] still thought him well-intentioned – but terrible at his job ... “I had a conversation with Tara and Peter [McIntyre],” recalled Ben, “and we were talking about how to help Sam and the conversation changed to: How do we get rid of Sam?”’
A fight ensued. The story would have been different and the outcome probably better for SBF if he had lost – but he won, twice. First, the rich EA supporters, whose money was keeping Alameda in business, backed him. Faced with a he-said, they-said between SBF and the rest of the management team, they withdrew much but not all of their capital, and kept the fund in business, with him in sole charge. The SBF sceptics all quit. Second, a chunk of assets that had gone missing – $4 million of the cryptocoin Ripple – turned up on an exchange in South Korea. (‘A voice finally came on the line and said: “Are you the fucker who sent us like 20 million Ripple tokens? How the fuck are you only calling us now?” In the background, Sam heard someone shouting, “Holy fuck, we found them!”’) The result of the split and subsequent vindication was that Sam’s colleagues all came from a specific group: true believers who had chosen to take his side, despite the chaos and mismanagement and his ability to lose or misplace significant amounts of money.
He switched on Modelbot. It immediately began to make profits. In 2018, on a capital base of $40 million (down from $170 million after the rich backers reduced their loans), Alameda made $30 million. That, remember, was SBF’s ‘ballpark estimate’ when he quit Jane Street, so he should have been happy. He wasn’t, though. The trouble was the maths. The rich investors took half, so there was $15 million left. Pay and severance (for the senior team who’d quit) took another $5 million. After paying tax on the remainder, he was left with $1.5 million to donate to EA causes. Not enough, in his view.
The solution was to create a new version of the most lucrative institution in crypto, an exchange. Almost all crypto is traded through exchanges, and most crypto is also stored on them. By taking transaction fees, exchanges can make money during the wild oscillations in value typical of all the coins. SBF would set up a futures exchange, because that was what the crypto world wanted: a place where you could bet on future movements in the direction of the currency, using leverage to magnify the size of your bets. It is legal to buy crypto at a specific price in most places, but trading in crypto futures is illegal in the US. The new business would need to be incorporated somewhere its activities were legal. An exchange needs customers, so the company would need to tick the maximum publicity box to attract them. Finally, the exchange would need an edge: something existing exchanges did not do, or did not do as well.
These were the parameters for FTX, the company whose collapse has landed SBF, as I write, in the Metropolitan Detention Centre in Brooklyn. FTX was set up for futures trading and incorporated in Antigua. The secret sauce was computer code written by Gary Wang, which limited customers’ potential losses on their bets to the amount of capital they had posted on the exchange: in other words, you could set out exactly how much you were willing to lose. (This feature, a ‘stop-loss’, is common in markets of all sorts, but tends not to work: when the market moves quickly, the stop-loss lags behind and people end up losing much more money than they had thought possible, or can afford. Wang’s innovation was software that monitored bets by the second and closed them as soon as customers went into the red.) That saved everybody money, because one of the problems with anonymous untraceable electronic money was making people pay up: it was easy to disappear and make the exchange take the loss instead. These losses were passed back to other customers, in the form of higher fees. Wang’s software gave speed, protection against losses and lower fees – an attractive package. The exchange was funded by a newly minted cryptocoin, FTT. The 350 million shiny new FTT were in effect shares in FTX, and the holders of FTT were entitled to a third of the revenue from the new company. A clever trick, FTX funding itself using its own newly invented asset.
SBF’s fame was part of the business strategy. To get customers, an exchange needs publicity, and it quickly became apparent – with the help of professional advice, not least from the highly regarded PR firm M Group – that the best way to get publicity was through Sam. He started to appear on business television, to be profiled, to appear at conferences. His chaotic mass of curly hair and his cargo shorts and schlubby T-shirt were powerful branding. He was also raising money from venture capitalists: a total of $2.3 billion from 150 venture capital firms. One VC’s in-house magazine featured a hysterically adulatory profile that claimed SBF could become a trillionaire. One capital raising round, the ‘meme round’, included the crypto world’s two favourite numbers: 420, which is a marijuana reference (don’t ask) and 69 (if you need to ask, don’t). In that round FTX raised $420.69 million from 69 investors – which was puerile, but served the higher purpose of attracting attention as well as capital. The fundraising and publicity took place in 2021, which, uncoincidentally, is when FTX began making proper money: the exchange made $20 million in 2019, $100 million in 2020 and $1 billion in 2021. In October 2021, Forbes profiled Sam as ‘the world’s richest 29-year-old, with a net worth of $22.5 billion’. That was a conservative estimate, because it was based only on his share of FTX, and omitted his 90 per cent share of Alameda and the extensive but difficult to value crypto assets it owned. By January 2022, now based in the Bahamas, FTX was worth $32 billion. On 11 November, SBF signed the papers putting it into bankruptcy and by 12 December he had been extradited to America and was facing trial.
It’s as big and as quick a fall from grace as the financial world has ever seen. The question of what went wrong is currently being settled by the jury in the case of USA v. Sam Bankman-Fried. The underlying structural factor, on which all parties can agree, is that there was too little separation between Alameda, 90 per cent owned by SBF, and FTX, mostly owned by SBF. In the spring of 2022 there was a general downturn in crypto. When Alameda got into trouble as a result, FTX, without its customers’ permission, lent Alameda customer deposits to cover its capital losses, to the tune of $8.6 billion. There was a run on FTT, the token used to set up FTX; customers on the exchange panicked and wanted their money back; the money wasn’t there, because it had been lent to Alameda and then lost. SBF says this was all a misunderstanding, based on the fact that FTX customers’ money was deposited in Alameda accounts, because FTX didn’t have a US dollar bank account. This would be an easier defence to run if he knew where the $8.6 billion currently is – but he doesn’t, and neither does anyone else.
Going Infinite is wildly entertaining, surprising multiple times on pretty much every page, but it adds up to a sad story, even a tragedy, for its central character and for all the people who lost so much thanks to his actions. Lewis, whom I know, is charming and amenable to charm; he likes SBF and is amused by him. I don’t feel the same, mainly because SBF, as well as being reckless with things that don’t belong to him, and deeply arrogant about his own intellectual superiority, is unredeemably careless about people. ‘The notion that other people don’t matter as much as I do felt like a stretch,’ he once said. A worthy insight, but SBF doesn’t act on it: in Going Infinite he repeatedly, compulsively, acts as if other people don’t matter at all. He plays video games during meetings and conversations, fails to be where he’s said he’ll be and do what he’s said he’ll do, and in general does exactly whatever he feels like doing, all the time. A detail: ‘I watched as Sam entered the empty townhouse, opened a closet, and, without so much as a glance at the row of empty hangers, tossed the ball of clothes onto the closet floor. We then drove together to the airport and returned to the Bahamas.’ The person whose job it will be to pick up those clothes, as far as SBF is concerned, does not exist.
This pattern of behaviour was accompanied by a profound internal void. His inner life and cultural world are severely truncated. ‘I don’t feel pleasure,’ he wrote in his journal while at Jane Street. ‘I don’t feel happiness. Somehow my reward system never clicked. My highest highs, my proudest moments, come and pass and I feel nothing but the aching hole in my brain where happiness should be ... I don’t feel pleasure, or love, or pride, or devotion.’ Years later, after moving to Hong Kong, he was saying something very similar: ‘In a lot of ways I don’t really have a soul. This is a lot more obvious in some contexts than others. But in the end, there’s a pretty decent argument that my empathy is fake, my feelings are fake, my facial reactions are fake. I don’t feel happiness.’
I believe him. It’s sad. Sceptics have latched onto a remark that SBF, thinking he was off the record, made to a journalist about his ethical commitments in November 2022: ‘Man, all the dumb shit I said. It’s not true, not really.’ Some take that as a gotcha! revelation about SBF’s not really believing in EA. I don’t think it is that: I think it’s consistent with what Lewis quotes about his inner emptiness. I think that interview, like most things SBF says about himself, is a glimpse into the abyss. He has no moral compass, other than one on loan from EA. Many people borrow their moral compasses from religion, but all religions have a place for empathy, even if it’s selectively applied. EA has no place for empathy, and neither does SBF.
As I write , the trial is going into its third week. (Autobiographical note: I’ve reviewed a gazillion books, but I’ve never written about one whose main character is currently in the dock.) Things seem to be going about as badly as they possibly can for our protagonist. His two closest associates, Wang and Ellison, have testified in convincing detail about his guilt. Ellison had several zingers about his personality and behaviour, but Wang gave the killer detail: as long ago as 2019 SBF instructed him to insert secret code in FTX software that allowed him to move customer deposits to Alameda. If the jury believes that, SBF is going to prison for many, many years. It doesn’t help that he has broken not just the eleventh commandment – don’t get caught – but its two important subordinate clauses. Don’t get caught; if you do get caught, make sure it isn’t by the Feds; if you do get caught by the Feds, don’t piss off the judge. His Honour Lewis Kaplan, a Clinton appointee, has the reputation of being an intelligent hard-ass, and has already revoked SBF’s bail for leaking information to journalists. From the official drawings, even the courtroom artist seems to hate him.
That said, from a UK perspective, there are some odd features to the trial. Granted, we are amazingly bad at prosecuting fraud in this country, and high-profile cases tend to collapse. However, in the US last year the conviction rate in federal cases was over 99.5 per cent. It’s hard to contemplate that number without thinking injustices must occur. As a lawyer said to me, ‘even the Japanese’ – whose legal system hardly ever acquits – ‘would be embarrassed by that.’ The decades-long prospective sentence is alien too: in most of the UK the maximum penalty for fraud is ten years, and in cases of multiple fraud, the sentences are almost always concurrent.* Finally, although there is obviously a serious case against SBF, there is an oddity about the trial being held in the US. FTX.com was a company incorporated in Antigua, with headquarters in Hong Kong and then the Bahamas, on whose main exchange US citizens were legally forbidden to trade. Yes, there was a US subsidiary, one of FTX’s 120-odd subsidiary companies, but the grand jury indictment cites the parent company. The prosecution presented as its first witness a representative victim of the FTX collapse. An American victim, for an American fraud trial? No, a French cocoa trader, based in London, who had been attracted to FTX because he had seen an ad featuring Gisele Bündchen. He lost $100,000 in bitcoin. A heartrending tale of woe, obviously, but the basis for a decades-long sentence in a US prison? A non-US citizen dealing with a non-US company whose main business was off-limits to citizens of the US?
Thoughts like this aren’t popular at the moment. SBF might not have been convicted by the court, not yet anyway, but he sure as hell has been by public opinion. The modern world features a category of entitled, arrogant, oblivious billionaires, who seem to float above us and see the common mass of humanity down below as a not especially interesting kind of ant. In SBF, we have finally got our hands on one of this class. We’re going to make sure he knows how we feel. In this climate, Going Infinite has landed to a predictable backlash, since Lewis spent a lot of time with his subject and is at pains to show that SBF the person is more complicated than SBF the pariah.
Some critics have said this amounts to a whitewash. I don’t think that’s fair. Lewis tries to answer the first question he was asked about Bankman-Fried: who was this guy? The question of his guilt or innocence Lewis leaves to the criminal justice system. I think that’s good practice, given that the trial is happening right now. For what it’s worth, I see no contradiction between the person described in Going Infinite and the things SBF is accused of having done. In fact I think the book makes it easier to understand how and why he did what he allegedly did. I can easily imagine SBF seeing certain actions in terms of their Expected Value, and acting accordingly. If he had a 50 per cent chance of making $100 billion to give away, with a 50 per cent chance of being caught and losing everything and going to prison for the rest of his life, he wouldn’t have hesitated – the EV of that bet would be $50 billion, and in his value system it would be unethical not to take it. As Caroline Ellison said in court, ‘he said that he was a utilitarian, and he believed that the ways that people tried to justify rules like “don’t lie” and “don’t steal” within utilitarianism didn’t work, and he thought that the only moral rule that mattered was doing whatever would maximise utility.’ It’s as narrow and constricting a principle as it’s possible to imagine, and it has put Bankman-Fried in the narrowest, most constricted possible place.
October 20.