Crypto collapse: the cult of personality and the normalisation of fraud in FTX and Celsius

Abstract

Purpose

This paper reviews the recent collapse of two cryptocurrency enterprises, FTX and Celsius. These two cases of institutional bankruptcy have generated criminal charges and other civil complaints, mainly alleging fraud against the CEOs of the companies. This paper aims to analyse the fraud leading to these bankruptcies, drawing on key concepts from the research literature on economic crime to provide explanations for what happened.

Design/methodology/approach

This paper uses a case study approach to the question of how large financial institutions can go off the rails. Two theoretical perspectives are applied to the cases of the FTX and Celsius collapses. These are the “normalisation of deviance” theory and the “cult of personality”.

Findings

In these two case studies, there is an interaction between the “normalisation of deviance” on the institutional level and the “cult of personality” at the level of individual leadership. The CEOs of the two companies promoted themselves as eccentric but successful examples of the visionary tech finance genius. This fostered the normalisation of deviance within their organisations. Employees, investors and regulators allowed criminal and highly financially risky practices to become normalised as they were caught up in the attractive story of the trailblazing entrepreneur making millions in the new cryptoeconomy.

Originality/value

This paper makes a contribution both to the case study literature on economic crime and to the development of general theory in economic criminology.

Citation

Mackenzie, S. (2024), "Crypto collapse: the cult of personality and the normalisation of fraud in FTX and Celsius", Journal of Financial Crime, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/JFC-01-2024-0054

Introduction

In this paper, we will consider two case studies of economic crime, in the form of recent “crypto collapses” – the FTX exchange and the “crypto bank” Celsius. We will review these cases in light of two theoretical ideas, which we will propose to be complementary in some aspects. One of those theoretical perspectives is a core part of the established cannon in sociological studies of white-collar crime: the “normalisation of deviance” thesis (Vaughan, 1996). The other theoretical orientation, which we refer to as “the cult of personality”, is a consideration of how contemporary developments in so-called celebrity culture might affect our thinking about economic crimes.

Both FTX and Celsius had inspirational and energising chief executive officers, who propelled their respective organisations to success and then onwards to disaster. The white-collar crime literature has a lot to say about wayward leaders who come undone because of an overconfidence in their own abilities and a sense that they are above the law (Arnulf and Gottschalk, 2013). The question we will explore here is whether this individual-level perspective can be usefully integrated with the organisational deviance perspective that Vaughan’s orientation involves. In other words, how do we reconcile a theoretical perspective on how organisations go off the rails with the research literature on how individuals go off the rails – in particular individuals at the highest levels of the organisations in question.

Economic crime is defined as financially motivated crime that uses deception rather than physical force to cause a loss, and is commonly linked to the workings of “the economy”, i.e. the production, distribution and consumption of goods (Button et al., 2022: 13-14). Economic crime is a much bigger problem than traditional crime whether the yardstick is number of victims, costs involved or number of offences (Button et al., 2022: 17). Fraud – the main criminal charge in both case studies reviewed here – is a key offence of interest for economic criminology, and the perspective includes the study of criminogenic cultures in organisations as well as individual motives and opportunities. Economic criminology is an interdisciplinary pursuit based in criminology but including ideas and scholars from many other disciplinary areas (Button et al., 2022: 22). As such it provides a good platform for the kind of analysis to be undertaken in this paper, since we intend to look at both organisational and individual features of the financial crimes in question, and build an interdisciplinary explanation that acknowledges a recursive structuring relationship between the one and the other; that is, the effect of the individual on the organisational culture and vice versa.

Using economic criminology as the basis of the inquiry in the paper also helps mitigate to some extent key limitations of the two theoretical perspectives used. For the “cult of personality” approach, a major limitation is that there has been a relative dearth of robust empirical research studies on the topic of personality in economic crime, and those studies that have been done present somewhat contradictory and inconclusive findings (Button et al., 2022: 198; Collins and Schmidt, 1993; Blickle et al., 2006; Nee et al., 2019). In terms of the “normalisation of deviance” approach, a key issue would be that like the differential association theory on which it is based, it can risk suggesting quite a black and white view of organisational culture, drawing attention towards what is normative and obscuring the complexities of inter-organisational conflict. In both cases, situating these perspectives in the wider disciplinary field of economic criminology can provide helpful balance and context, so that what might otherwise seem isolated and not altogether unproblematic concepts can instead be taken as interconnected strands in a larger web of explanation in economic crime.

While traditional “street” crime rates have been subject to a long and sustained fall, economic crime rates are rising and, consequently, being increasingly recognised as a major and growing threat even in the face of a complacent political discourse that would prefer to deny it (Button et al., 2023). This makes the study of the etiology of economic crimes a matter of paramount importance for contemporary criminology. Given the breadth of offending behaviours that can fall within the ambit of economic crime – a crime like fraud, for example, can refer to myriad activities – looking to notable case studies of particular examples of economic crime seems to be a productive approach.

In what follows, we will therefore use the following structure. Firstly, we will present the main elements of the two case studies in question: the collapse of FTX and the role played by its CEO Sam Bankman-Fried; and the collapse of Celsius and the role played by its CEO Alex Mashinsky. Secondly, we will briefly review the main tenets of Vaughan’s theory of the normalisation of deviance. Thirdly, we will consider some of the personality features of corporate leaders that make people want to follow and believe in them, and the ways in which that influence can help to explain the role of high-level individuals in crime committed by the organisations they lead. Finally, we will consider how to integrate these two approaches, in light of the case study data.

The collapse of FTX: “Scam Bankman-Fraud”

The cryptocurrency economy is based around trading conventional currency (“fiat”) for crypto, hoping that the value of that crypto will rise, and then trading it back for more fiat currency than you put in to begin with. This simple routine has, however, given rise to a fantastically complex array of ways to make or lose money once you have made the transition from fiat to crypto (Mackenzie, 2022). We will not attempt to touch on most of that complexity here, other than those parts of it that are relevant to understanding the histories of the two organisations reviewed. Within that ocean of hot speculation and highly convoluted trading and investment possibilities, the overall cryptoeconomy is anchored around a few large institutional bodies, known as exchanges. FTX was one of the largest cryptocurrency exchanges. In early 2022, it claimed it was handling $15bn in daily trading volume (US District Court SDNY, 2023b).

Exchanges are the “front end” of crypto: they are the way that most traders get their fiat money into and out of the system. To buy crypto, you first register for an account with one of the big exchanges, and then transfer money from your bank account on to the exchange. All of this happens over the internet: there are no branches to visit or paper forms to fill out as there might be if you were to open a conventional bank account. Once you have credited your exchange account with funds, you can use the exchange’s other main function besides the money in/out pathway. The other main function of exchanges is to provide a mechanism for trades to take place. So you can trade your fiat credit balance for a crypto of your choice, and you can also trade one type of crypto for another type. The exchange makes its money by taking a small percentage fee on each trade that is made.

Sam Bankman-Fried established the FTX exchange in 2019, and between 2019 and its collapse in 2022, he supported the exchange with fundraising rounds, taking around $1.8bn from venture capitalists who put up funding in exchange for shares. Known as SBF, he had reportedly made billions of dollars designing and using algorithms to trade crypto (Lewis, 2023). Prior to establishing FTX, he had built his trading empire into a crypto hedge fund called Alameda Research Ltd, which by 2019 was said to be one of the biggest quantitative traders and market makers in the space, moving “$600m to $1bn a day” (US District Court SDNY, 2023b: 5). Initially, from 2019 to 2021, he remained CEO of Alameda while also taking on the role of CEO of FTX but in 2021, recognising the optics of the conflict this involved, he stood down from Alameda, installing Caroline Ellison as its CEO. Ellison was SBF’s girlfriend and it has been suggested that she was in place as a puppet mainly for appearances’ sake, while he continued to run both Alameda and FTX.

SBF was found guilty on seven counts of fraud and conspiracy to launder money in November 2023, and was sentenced to 25 years in US federal prison in March 2024. At the time of writing, he is appealing that sentence. The main charges were that from 2019 through to the collapse of the exchange in 2022 he diverted FTX customers’ deposits for a variety of purposes: to Alameda to fund its risky crypto hedge business; to politicians in an attempt to attain influence through illegal campaign donations; and to himself for personal expenses.

This behind-the-scenes illegal activity was in sharp contrast to the public persona that SBF cultivated. As the indictment sets out, he “represented himself as the figurehead of a trustworthy and law-abiding segment of the cryptocurrency industry that was focused not only on profits, but also on investor and client protection” (US District Court SDNY, 2023b). Yet throughout the existence of FTX, the basics of investor protection were routinely flouted. There was no segregation of client funds from those belonging to FTX and available to SBF to use. Despite publicly claiming that on the FTX platform, Alameda was treated the same as any other client, in fact Alameda had access to a borrowing facility that allowed it to take a practically unlimited amount of the exchange’s reserves by running a negative balance. In effect, therefore, Alameda was borrowing FTX client money to use in making its high-risk crypto bets. And since SBF was in de facto control of both organisations, this amounted to him taking FTX client funds to use for his own purposes while making representations to clients that their funds were safely held in the exchange’s reserves.

As well as allowing Alameda access to FTX funds, SBF used client deposits to pay for a range of things aimed at enhancing his, and FTX’s, image. He bought celebrity endorsements for FTX, paying influencers and malleable public figures millions to speak well of the platform. In shades of what now calls to mind the sad saga of “Enron Field” in Houston, he sponsored a sports stadium: the home of the Miami Heat basketball team took on the name “FTX Arena”. He bought advertising during the US Super Bowl in 2022. He appeared on stage with Bill Clinton and Tony Blair, discussing the future of finance. All of the endorsements and adverts reinforced the central message that FTX was a safe and trusted exchange – indeed, the most safe and trusted platform in the otherwise wild-west style free-for-all that the decentralised cryptoeconomy represented.

In fact though, as well as allowing Alameda untrammelled access to FTX client funds, SBF had also confused the bank accounts of these firms, using an Alameda account to receive deposits into FTX by investors who did not know about this arrangement. FTX, which in the USA was operating through a subsidiary entity, FTX.US, had been denied banking services when it tried to open an account due to its incapacity to assure the bank that it was not dealing in securities without a licence (which it surely was). So SBF lied to US banking officials when opening a new account in the name of Alameda, saying it was intended for use in the speculative hedge trading of that firm when he actually intended to use it to receive FTX deposits from US customers. As such, SBF had direct access to FTX client deposits before they even hit the exchange, and he could easily appropriate them for use by Alameda since the bank account into which they were deposited was Alameda’s. The result was that Alameda stole billions of dollars of FTX clients’ funds, ultimately losing most of it as the positions it took on the speculative cryptomarket were wiped out in the market downturn in 2022.

In November 2022, Alameda’s balance sheet was leaked and appeared on the internet. Analysts quickly identified a serious problem: Alameda’s valuation of its assets, on which its solvency depended, included a massive holding of FTT, which is the crypto token issued by the FTX exchange. As a “share-like” offering (although without any of the traditional rights of an actual shareholding), the price of a token such as FTT is correlated to investor sentiment towards the perceived worth of FTX as a business. The valuation of Alameda’s “assets” on its accounts was therefore problematically insecure given the overly close relationship between Alameda and FTX. More worrying still, allegations were made that Alameda had been using its FTT holdings as collateral to borrow money from other crypto lenders while making automated purchases of FTT on the market to artificially inflate the token’s value, thereby essentially faking the value of the collateral its lenders thought they had as security for their loans (SEC, 2022).

Another large holder of the FTT token was the CEO of FTX’s main competitor exchange, Binance. Once he learned the true nature of the nexus between Alameda and FTX in November 2022, he threatened to sell all his FTT tokens. Knowing that such a large sale would “dump” the price of the token, other holders began selling their FTT tokens en masse, setting off a death spiral for Alameda as its aspirational valuation of the token, and therefore its own solvency, was undermined. Sensing trouble ahead, FTX users began a flight to safety, withdrawing their funds from the FTX platform in droves. This “run on the bank” scenario presented more liquidity outflow than FTX could handle, since of course due to the prior appropriations of client funds it did not actually hold the capital necessary to support high numbers of clients asking to be paid out. This rapidly led to withdrawals being frozen and, in fairly short order, bankruptcy, liquidation proceedings and the arrest of SBF and several of his colleagues, with civil lawsuits being filed against him by the Securities Exchange Commission and the Commodity Futures Trading Commission in parallel to the criminal action (SEC, 2022; CTFC, 2022).

The collapse of Celsius: “banks are not your friends”

Established in 2018, Celsius Network LLC was a crypto lender and “bank”. The scare quotes around “bank” are in recognition of the fact that the main pitch of its founder, Alex Mashinsky, was that conventional banks rip off their customers by keeping too much of the margin they make in lending out depositors’ funds. Mashinsky, on the other hand, promised depositors fantastic rates of interest on the crypto they put into Celsius. The rates varied according to the type of cryptocurrency deposited, but ran up to 17% per annum in some cases. Mashinsky said he could afford this simply by taking less profit off the spread than is the norm in the mainstream banking economy. Following the model of a regular bank, Celsius would receive customer deposits, lend the money out to borrowers who wanted to use it for their own purposes, charge those borrowers a rate of interest, and then pay most of that interest to the customers whose funds were being lent out. Mashinsky was adamant this was a solid business model for Celsius and a good deal for customers, as he put it “either Celsius is lying or the banks are lying” (Levine, 2022). He regularly made his public appearances wearing a t-shirt that said “banks are not your friends”.

Unfortunately, it turned out to be neither a solid business model for Celsius nor a good deal for customers, and it has now become clear that Celsius was lying. Celsius undertook only to make fully collateralised loans which were therefore safe bets: a borrower of a given crypto token would have to deposit with Celsius an alternative crypto token worth no less than the value of the loan. If they defaulted on the loan, Celsius could sell their security and would lose no money in the transaction. The only partially-collateralised loans made would be to large, reliable institutional borrowers who would be sure to pay them back, Maskinsky said. However, Celsius was also making uncollateralised loans, although it did not make this known to its depositors. In 2021, for example, some 39% of Celsius’s loan portfolio was, according to its own internal analysis, uncollateralised (US District Court SDNY, 2023a: 19). Even the loans which were technically collateralised were still, in some cases, risky. There is an overlap here with the FTX bankruptcy detailed above. Between 2020 and 2022, Celsius made around $1bn of loans to Alameda Research Ltd, SBF’s crypto hedge fund. Those loans were collateralised using the FTX platform’s FTT token as security. When FTX collapsed and both it and Alameda filed for bankruptcy in November 2022, the value of the FTT token collapsed too, leaving Celsius holding near worthless collateral and joining the queue of bankruptcy claims in the hope of recovering the value of the loans it had made to FTX.

Celsius was also making risky bets on other crypto ventures and directional bets on Bitcoin and other crypto, contrary to its express public statements of being market neutral and simply making money from the yield on loans issued. Some of these risky bets failed and lost Celsius considerable amounts of money. One of the key problems with the Celsius business model is the compounding nature of risk in crypto markets. The main reason people borrow crypto is to gamble with it. Crypto investors with a sharp eye for opportunity may think they have spotted a way to make a good return on investment, but they may not have the initial capital to make an investment sizeable enough to make the payoff worth it. So they borrow crypto to make these investments, which are in fact very high risk bets. If they manage to do so via an uncollateralised loan, and if they lose the money they are gambling, there is unlikely to be a way for a lender like Celsius to claw back the money.

Having initially raised $1bn in venture capital and a further estimated $19bn in customer deposits, by 2022 Celsius found itself in deep trouble. In mid-2022, the crypto market was collapsing. Several high-profile market participants had been liquidated, and Celsius had been lending them millions of uncollateralised dollars on the assumption that they were safe borrowers. In a classic Ponzi scheme routine, Celsius was now depending on new customers making deposits to allow it to live up to its commitments to existing customers at the interest rates it had promised. Mashinsky is accused of misleading investors by pumping up Celsius’ undeserved image of popularity and reliability in his public discourse. He said it had 2 million customers when in fact two-thirds of the 1.7 million users who had created an online account on the platform held less than one dollar’s worth of crypto in their accounts (Supreme Court of the State of New York and NY, 2023: 11). He also gave customers false assurances that their assets on the platform were insured when they were not, and intimated that Celsius had been given the green light by regulators when no such thing had happened.

Through all of this, Mashinsky cut a very public figure, holding weekly open video call “AMAs” (short for “ask me anything”, or in this case as he would have it “ask Mashinsky anything”). As Celsius was collapsing, Mashinsky was on Twitter fighting those naysayers he considered trolls. On 11 June 2022, in response to a question from a twitter user about rumours Celsius was restricting withdrawals, Mashinsky said “do you know even one person who has a problem withdrawing from Celsius?” (Fanelli and Ge Huang, 2022). Celsius “paused” investor withdrawals the next day, 12 June 2022, purportedly “in order to stabilise liquidity and operations while we take steps to preserve and protect assets”.

As well as the public misrepresentations of Celsius’s business practices and its financial position, Mashinsky is alleged to have been perpetrating another kind of fraud. When it launched in 2018, Celsius introduced the CEL crypto token. It did this by way of an ICO, an “initial coin offering”, comparable in many respects to the initial public offering of a mainstream company when it offers shares to the public for the first time. Mashinsky said the ICO had sold all of the available CEL, at a total price of $50m. However, only $32m of the CEL had been sold, leaving one-third of the ICO tokens still in the hands of Celsius. Celsius had said these tokens would be destroyed, but they never were. Had all of this been widely known, it would have deflated the price of CEL, showing the ICO to have been a lot less popularly subscribed than the market imagined. Mashinsky and colleagues then, over the following years, used Celsius’ money (including of course client depositor funds) to purchase massive amounts of the CEL that had been placed onto the market through the ICO, without disclosing their identity as buyers. They did this to inflate its price well beyond what they knew it was worth, to allow them to profit personally by selling their own CEL at these higher prices. Mashinsky is said to have made approximately $42m from his own personal sales of CEL (US District Court SDNY, 2023a: 25, 35).

Mashinsky is currently subject to several lawsuits alleging misconduct in his running of Celsius. In 2023, as well as being criminally indicted (with trial scheduled to follow in the second half of 2024), he has been sued in civil lawsuits by the New York State Attorney, the US Securities and Exchange Commission (SEC), the US Commodity Futures Trading Commission (CTFC) and the US Federal Trade Commission (FTC). All of these lawsuits allege variations on the same themes: a history of fraud, deceiving the tens of thousands of customers of Celsius by promising high returns while investing their funds in high risk ventures that were never disclosed, and misusing Celsius/client funds to prop up the price of the CEL token by buying it through Celsius while Mashinsky and his colleagues were selling their personal holdings.

The normalisation of deviance

Mashinsky has now been arrested but even after Celsius had declared bankruptcy he remained on Twitter sounding off against critics he considered to be trolls seeking to spread “false claims” about him and Celsius. Similarly, SBF continued his public discourse after the FTX collapse through numerous interviews with journalists while under house arrest and by continuing to tweet about the bankruptcy and writing reams of excuses and justifications for public consumption (Yaffe-Bellany, 2023). Both Mashinsky and SBF seem to think that they are victim rather than perpetrator. Their enemies were out to get them, they say, and therefore the true cause of the misfortune of all the customers who have lost their money is not the fraudulent misrepresentations and criminally reckless business models of these renegade CEOs. In SBF’s version of reality, it is the competing firm Binance which started the public run on FTX by threatening to sell off its large holdings of the FTT token, and which subsequently considered saving the day by buying out FTX in a takeover but then withdrew from those negotiations. In Mashinsky’s case, it is sellers and shorters of his CEL token, which he has acknowledged is an asset that retains value only if everyone thinks it will be worth more tomorrow than it is today – and such assets with no underlying “real” value can quickly fall into a death spiral if mass selling gathers momentum. The exculpatory “word salad” produced by SBF (Weiss, 2023) also looks internally to allot blame to supposedly incompetent colleagues. He complains that Ellison should have hedged Alameda’s positions more effectively and that it was the fact that she was out of her depth that brought the problems to bear upon his corporate empire although, as other colleagues pointed out, “Alameda’s hedging would have been irrelevant if FTX hadn’t misused customer money” (Yaffe-Bellany, 2023).

This attitude – that corporate malfeasance leading to financial catastrophe is someone else’s fault – is not uncommon in the white-collar crime literature. It is part of the well-observed pattern of economic criminals contesting the criminal label: often these criminals acknowledge technical law breaking but deny criminal intent (Benson, 1985; Conklin, 1977). Salient to both the FTX and Celsius cases, a common rationalisation in economic crime is that the offender was “just borrowing” the money, with no intent to steal it (Button et al., 2022: 192; Cressey, 1953). The prevailing sentiment is that mistakes were made, lessons have been learned, but “I’m no criminal”. Croall found similar attitudes among business criminals she observed in court some 35 years ago, their explanations for their crimes being “mistakes, accidents and someone else’s fault” on her recounting (Croall, 1988).

On an organisational, rather than an individual, level, blaming others outside the organisation can be an indicator of the normalisation of deviance within the organisation. The normalisation of deviance is Vaughan’s term for what she also calls “history as cause” (Vaughan, 2003): the personnel working in the organisation becoming used, over time, to incrementally pushing more boundaries and taking more risk. This leaves the organisation carrying much more risk in the end than anyone within it would have thought acceptable at the beginning. The taking on of more extreme risks is not recognised at the time as problematic, because each small step forward is but one minor extension of what the organisation is already doing at the time (Vaughan, 2004). The normalisation of deviance is an example of differential association; Vaughan describes it as a “social psychological product of institutional and organisational forces” (Vaughan, 2004: 34).

As with Vaughan’s case studies of the organisational failures in NASA which were responsible for causing the deaths of several astronauts in catastrophic explosions (Vaughan, 1996), when disaster strikes people outside the organisation ask “how could this have been allowed to happen”? This is exactly the question that is being asked now of FTX and Celsius: both large established financial fixtures of the cryptoeconomy which, with variations on the theme, misused billions of dollars of client funds to gamble with and lost. Interestingly, aside from the two reckless CEOs we have focussed on in the case studies here, key personnel within FTX and Celsius knew that all was not right with the conduct of the businesses. Evidence has emerged in both cases of employees raising concerns, sometimes extremely bluntly. In Celsius, internal messages reveal an open discourse of frank concern among employees, such as that “the main problem [with the CEL token] was that the value was fake […] the issue is that people are selling and no one is buying except for us” (Dugan, 2023); one employee called Celsius “a sinking ship”, while another wrote that “there is no hope […] there is no plan”. On 21 May 2022, a Celsius executive candidly acknowledged in an internal message: “we don’t have any profitable services” (SEC, 2023).

So, these are not organisations that were “sleepwalking into failure” – they were heading for failure in a way that well-placed individuals within the businesses could quite clearly see. The normalisation of deviance here made the deviance routine but not unrecognised. Did workers “become so insensitive to deviant practice that it no longer feels wrong” (Price and Williams, 2018), subject to a “mindlessness induced by institutionalisation [that] may cause individuals to not even notice what might arouse outrage under other circumstances” (Ashforth and Anand, 2003: 14)? Both of those states of mind – insensitivity and mindlessness – are different ways of describing a passive routinisation of wrongdoing that is one interpretation of the normalisation of deviance theory as it might apply in some contexts. In such situations, as with Vaughan’s NASA case studies, the boundaries of what the organisation perceives to be “acceptable risk” are subject to an expansion over time to the point that the groups involved acclimatise to dangerously high levels of risk. Yet while this acclimatisation to extremely high-risk environments does seem to characterise some aspects of the FTX and Celsius cases, it does not capture the entire picture. As with some of the NASA staff in Vaughan’s studies – those who sought to postpone the launch, objecting to the risk being taken, but who ultimately fell in line and implemented the launch order (Vaughan, 1998) – instead of mindless routinisation it seems that some workers in FTX and Celsius recognised that what they were being asked to do felt wrong, but they did it anyway.

This institutionalised corruption went along with the rationalisations associated with a drift into doing things that are known to be wrong (Matza, 1964). Various types of such rationalisations have been well rehearsed in previous studies of white-collar and corporate crime: “just following orders”; “it’s not my call”; “I’m just a cog in the wheel” (Huisman, 2020: 145–6). In moving further and further away from law-abiding conformity and good judgement in risk management over time, red flags that were raised within the organisations simply failed to gain enough traction. That these individuals rang alarm bells that went unheeded is testament to the overwhelming power of the CEOs, Mashinsky and SBF, in running their outfits. These are two powerful personalities, regarded by employees as well as outsiders with cult-like awe that manifested in a misplaced faith in their judgement and abilities. Employees of FTX and Celsius chose loyalty to their CEOs when they should have chosen otherwise (cf. the role of loyalty in “crimes of obedience”: Kelman and Hamilton, 1989; Hamilton and Sanders, 1992).

SBF, in his autopsy of the FTX collapse on Twitter days after the event said:

Once upon a time – a month ago – FTX was a valuable enterprise. FTX had $10-15b of daily volume, and roughly $1b of annual revenue. $40b of equity value. And we were held as paragons of running an effective company. I was on the cover of every magazine, and FTX was the darling of Silicon Valley. We got overconfident and careless (Bankman-Fried, 2022).

SBF did not take seriously enough the risks involved in his unlawful approach to running the finances of FTX, and had come to treat the absence of consequences for his mishandling of the organisation as normal. This pattern has been seen before, including among inside traders on Wall Street in the 1980s: “the atmosphere grew relaxed. There really was a deterioration of caution” (Sethi and Steidlmeier, 1991: 112).

As well as his own deterioration of caution, SBF brought others along with him in that normalisation of deviance with his (over)confident approach to what the future might bring. It is not unusual for lower-ranking employees in organisations to resist exercising their own moral judgements about right and wrong, seeing these “calls’ as ‘above their paygrade’: ‘let the people making the high salaries tackle the difficult ethical decisions’ seems to be a widely held view among occupants of lower echelon corporate positions” (Jones and Ryan, 1998: 440). These rationalisations and others (see Cressey, 1953; Benson, 1985), take place in a context where regulation is lax, tending towards the laissez faire attitude that allows wrongful and harmful behaviour to occupy a regulatory “grey area”. This point has been well made in relation to tax avoidance schemes developed by corporations which are similarly “grey area” – de facto legal because they are not prosecuted as tax evasion, but still clearly designed in the spirit of depleting the public purse (Hock, 2022).

Economic grey zones like tax avoidance schemes − and much of the crypto industry (Mackenzie, 2022) − contribute to the social-psychological normalisation processes that Vaughan discusses, and the rationalisation processes discussed by economic criminologists (Button et al., 2022). A lack of effective regulation or prosecution can suggest to corporate employees that the ostensibly illegal actions they are instructed to be part of are actually condoned by those with the power to make the official judgements about legal right and wrong. The assumption can follow that the CEO knows how to maximise opportunities for profit while navigating the edges of the grey zone of legality. Regulation is weak in crypto for many reasons, including the sense that scams and bad business behaviour are the growing pains of a new industry, and that the crimes if not victimless are predations upon victims who have willingly taken on significant risk by participating in the scene as an attempt to get rich quick: gamblers, in other words, who know they will lose some of their bets to crime but still proceed anyway (Mackenzie, 2022). There is, of course, little justification in treating these victims differently from the victims of other more “traditional” financial crimes, not least because the gambler caricature overlooks all manner of more conventionally innocent and naïve investors, and the scale of the losses suffered can be huge. Nonetheless, the point remains that the differential association processes of the normalisation of deviance within the crypto industry are group processes that happen in a wider socio-economic context that supports, rather than interrupts, those normalisation and rationalisation patterns.

The cult of personality among corporate leaders

The idea of the “cult of personality” looms large in the tech start-ups of Silicon Valley. Perhaps the signal example is Steve Jobs, the now deceased CEO of Apple. Jobs was revered by his peers and many now try to emulate his perceived leadership style. He was imbued with the supposed ability to “see round corners”, in other words a kind of profit-oriented prescience, an ability to predict success and drive others to help him achieve it. He was driven and imaginative but also fairly eccentric. He “branded” himself by sporting a trademark approach to smart-casual workwear. Behind the scenes, he was sometimes perceived as arrogant and intolerant, being erratically rude and dismissive towards critics and colleagues alike (Isaacson, 2011). Many of these traits displayed by Jobs, even the negative ones, have been latched on to by acolytes as signifiers of genius, and this public readiness to see signs of the successful entrepreneur in oddball and, in some cases, unlikeable features of personality can be seen in the coverage of both SBF and Mashinsky, both slated as difficult people whose problematic lack of the usual social graces was supposed to be part of their gift. SBF was even accused of faking his “tousled hair and cargo shorts” approach to office-wear precisely to trade off the perceptions of Einstein-level genius they encourage, to which his defence has been that, no, he is just genuinely scruffy by nature (Montgomery and Bekiempis, 2023).

The celebration of this kind of cult of personality leads organisations and the many people working in them to be overshadowed by the public-facing image of their charismatic CEOs, and at the same time for the success of the organisation to be attributed disproportionately to those CEOs. We can see this pattern currently with Elon Musk (Tesla/X), Mark Zuckerberg (Facebook/Meta), Larry Page and Sergey Brin (Google) and more. The social effects of the cult of personality are to elevate certain business people to quasi-celebrity status, attributing to them a market-leading competence and, to an extent, putting faith in them to deliver successful outcomes. These social effects of the cult of personality can be counter-productive for employees, investors and the public, when the skills and attributes the leader is thought to possess turn out to be absent. Most relevantly for our purposes, we can find examples of white-collar fraudsters who have benefitted from the veil of competence the cult of personality has given them, and the confidence that it has inspired among others.

One recent example would be Elizabeth Holmes, the CEO of the health tech company Theranos that emerged from Stanford University’s campus, took Silicon Valley by storm, but ultimately came to a halt when it transpired that the transformative new product she had been touting – a finger-prick rapid blood test technology – did not work and never had. Holmes benefitted from “the myth of the brilliant founder” (Carreyrou, 2018), persuading investors to back her: “sophisticated politicians and investors fawned over Elizabeth Holmes” (Kilby, 2023). Similarly, “FTX had no board of directors, unknown auditors and there was both incompetence and embezzlement, but it did hire amazing celebrities to state they were investing in it […] FTX is the marriage of various fraud schemes in the age of the Kardashians” (Kilby, 2023).

In other words, investors are sometimes so captivated by celebrity nowadays that the glitz and glamour of wow-factor personalities can overwhelm the usual need for demonstrable financial performance or legal probity. In such a context, entrepreneurs are counselled to “fake it till you make it”, and there is an expectation that new tech start-ups are operating in a context where some disrespect for conventional legal standards may prove beneficial, especially where there is a grey area that can be exploited around what precisely the law requires in any given circumstance. This sort of law-bending entrepreneurialism has long been noted by sociologists of economic crime. Vaughan suggested that leaders of small start-ups operate with a view that “new organisations only rise rapidly if they have some disrespect for traditional standards” (Vaughan, 1983: 60). McBarnet, in her studies of so-called “creative compliance”, noted lawyers employing euphemisms aimed at reducing impressions of culpability where they were overstepping the boundaries of the law, calling this merely “sailing close to the wind” (McBarnet, 2006). In the contemporary Silicon Valley start-ups, the watchword is to “move fast and break things” (Taplin, 2017), which broadly means it is better to be pioneering and make some mistakes along the way than to be stultified by a risk-averse approach to business innovation that minimises errors but slows progress in so doing. Crypto, as a socio-economic sphere where everything is seemingly cutting edge and moving well ahead of the regulators, encapsulates that tech mantra of moving fast and breaking things in some of the purest forms yet seen by economists. “Things” in crypto are breaking all the time! This includes the technology, which despite in some cases being worth billions of dollars still regularly stops working or is exploited by hackers. It also includes laws, which tech entrepreneurs like SBF and Mashinsky have been critiqued as thinking simply “don’t apply to them in the same way they do to everyone else” (Katersky, 2023).

Apparently, it is not only investors who can be swayed in their judgement by innovative and seemingly dependable CEOs. Regulators can be influenced by the cult of personality too. Bernie Madoff, probably the most famous Ponzi schemer in living memory, ran an investment portfolio that made suspiciously reliable and impressive returns in good markets and bad. Despite this, he remained unscrutinised by regulators in significant part because of his good name, and his connections:

[…] he helped create the Nasdaq stock market, and became a senior member of its self-regulatory organisation. He and his firm partners worked in various capacities in the financial industry’s main lobbying group. Madoff and his family gave money to connected politicians with financial-system oversight responsibilities. His compliance officer, a niece, was married to an SEC official (Gasparino, 2022).

Comparably, SBF invited members of the CTFC to work in compliance roles in FTX: an example of the so-called “revolving door” between markets and regulators that is part of the overall problem of regulatory capture (Makkai and Braithwaite, 1998) and fiscal corruption (Hock, 2022). Like Madoff, SBF made political donations with the goal of achieving influence, and lobbied politicians responsible for regulation which served the ends both of making his views known in regulatory circles and painting him with the “good guy” brush as, unlike many other figures in crypto who preferred the anonymity of the shadows, he was at least prepared to front up and talk.

Other examples of white-collar criminals who benefitted from somewhat mythologised personalities can be found in the literature. For example, “the cult of personality that arose around Michael Milkin enabled him to assemble a force of like-minded disciples at Drexel Burnham Lambert (Stone, 1990)” (Ashforth and Anand, 2003: 7). Ashforth and Anand, following other commentators, consider “the personal power of charisma” to be important here: “the more charismatic the authority figure(s), the greater the identification, trust and reflexive obedience that he or she is likely to engender among subordinates (Conger and Kanungo, 1998)” (Ashforth and Anand, 2003: 7). Both SBF and Mashinsky are figures who crypto enthusiasts would recognise as having benefitted from the cult of personality – both were charismatic in their own way.

As well as the benefits charisma can deliver to the charismatic leader, however, there is also a clear sense in which one can be afflicted by the social processes that develop around being the object of the cult of personality. Terms used to describe SBF and Mashinsky have included genius, boy wonder, master-of-the-universe, tech savant, math whizz and visionary. These plaudits are reminiscent of other white-collar crime scandals, such as where Enron executives lauded themselves as “the smartest guys in the room” (McLean and Elkind, 2004). Such images of greatness can wreak havoc with one’s moral compass. When everyone around you is telling you that they have untrammelled belief in your capacity to make great decisions, and where your day-to-day activities in terms of how you run your organisation are not scrutinised or questioned, the main system of moral checks-and-balances of your action becomes yourself. And if you come to believe what everyone is saying about you – that you are a business genius and you make great decisions and you are destined to lead your organisation to success – there is something of a circular rationale at play whereby if you want to do something a certain way it must be the right decision because it is your decision and you are someone who makes great decisions.

Conclusion: the influence of the cult of personality on the normalisation of deviance

The entrepreneurial risk taker in business cuts a quixotic figure in contemporary late-capitalist socio-economic discourse: celebrated in success and damned in failure. Both SBF and Mashinsky were lauded as captains of the crypto industry while the going was good. However, a moral philosophy that considers that the end justifies the means – a utilitarian consequentialism – which SBF explicitly espoused (Kolhatkar, 2023) and Mashinsky perhaps more implicitly, is only as good as the ends it produces. That philosophy is exposed here as a form of gambling, for if the beneficial ends do not eventuate, all you are left with are immoral means. For high rollers of the capitalist dice, this may well be the perfect philosophical accompaniment to extreme financial risk taking, discarding conservativism and good sense for the thrill of “going big or going home”.

The two theoretical reference points we have looked to here – the normalisation of deviance and the cult of personality – come together in contemporary investment culture as a form of wilful blindness among those who are carried along in the excitement of being part of, or investing in, the crooked enterprise. Vaughan proposed history as the cause of organisational deviance and in these case studies we can see that, up to the point of institutional collapse and the ensuing criminal charges, history had provided an apparent record of successful entrepreneurship for both SBF and Mashinsky. Investors wanted to believe this version of the story because if it were true it would make them money. Employees wanted to believe it too; the idea of following an inspirational trail-blazing tech-finance guru is an attractive one. In these two case studies, we can see the silencing effect on prospective critics of the personas SBF and Mashinsky had constructed. Key employees within FTX and Celsius knew of the flaws in the operating models that would prove deadly to the businesses. Some others did not know such details but were reticent to inquire or investigate, papering over any concerns they might have with a misplaced trust in the individual leading them. They became unreflective and therefore irresponsible followers of a leader they credited with unrealistic expectations of prowess.

We can therefore see that normalisation – habituation to a way of doing things in which escalating risk is accepted where it should be increasingly critically scrutinised – is in these crypto examples tied closely to a particular cultural manifestation of the tech entrepreneur as idol. Driven by these unethical and dangerously self-assured visionaries, FTX and Celsius had ineffective or absent compliance systems both internally and in terms of external controls or audits. There were no “capable guardians” in place sufficient to prevent the economic wrongdoing (Levi, 2008; Felson, 1995). Crypto is currently a model of unconstrained capitalism, and unfortunately this looks much as we might expect. The recognition of the propensity in crypto towards the fetishization of caricatures of entrepreneurial genius − that has made celebrities out of irresponsible and incompetent “wunderkind” in SBF and Mashinsky − brings the normalisation of deviance theory into this new era where celebrity culture clashes with laissez faire tech-finance to produce, predictably, white-collar financial crime on a globally significant scale.

What generalisable lessons can we draw from these case studies for the continuing development of the field of economic criminology? “Features of offender personality” is an explanatory category that is normally glossed over quickly in coverage of theories and perspectives that can help us to understand economic crime. Whatever the impacts on the offender themselves, however, personality can be an important factor influencing others, such as those who work for, invest in, or regulate, the offender and their enterprise. The externally influential nature of that “cult of personality” is one aspect of the context of opportunity that presents itself to the offender. Opportunity is of course regularly suggested to be an important causal influence for economic crimes (Benson and Simpson, 2009), and in the cases reviewed here we can see that the aura of personality effects such as charisma, charm and confidence can influence the opportunity landscape for prospective economic criminals by diminishing regulation (“I’m one of the good guys, you don’t need to worry about me”), reducing surveillance (“just trust me, I know what I’m doing, you don’t need to know the details”) and providing a willing workforce ready to support the enterprise even where it seems to be doing illegal things (“the rules don’t apply to us; we’re breaking new ground here; the ends justify the means”).

In the introduction, we referred to an observed recursive structuring effect between the cult of personality and the normalisation of deviance. As the case study data reveal, this appears to be a “structuration” type of recursive relationship (Giddens, 1984): the cult of personality of a strong charismatic visionary leader supports the normalisation of deviance within the organisation, and concomitantly a context in which deviance is normalised gives the leader the opportunity to engage strategies that break rules, adding (for a time at least) to the general impression that they are indeed a successful, visionary, pioneering, no-holds-barred leader. Furthermore, the receptive audience that the leader finds for their vision supports their ability to rationalise the legally problematic elements of it; after all, the rule-breaking approach seems to be achieving good results, and investors and employees appear to be impressed. Breaking rules to succeed in business – especially in a new economic arena like crypto, characterised by a sense of anomie and a “new frontier” spirit – is rationalised as smart business, and in a classic economic crime rationalisation does not seem like “real crime” (Button et al., 2022: 190; Karstedt and Farrall, 2006).

In the end, the relationship between the influential personality of the leader and the normalised deviance in the organisation becomes too destructive to continue. In the two case studies, an intervening force occurs in the form of a general loss of confidence in the leader, a run-on-the-bank scenario, insolvency and then a long period of autopsy and re-evaluation. In the re-evaluation stage, experts like liquidators and the courts, those involved like employees, regulators and accomplices, and onlookers like the media and the general public, all seek answers to the question how did things go so wrong? Ex post facto, the answer to that can be seen more clearly than it could when the pattern of offending was in full swing, for the reasons that our analysis has suggested here, using some of the tools of economic criminology.