The Four Horsemen of the Cryptocalypse
Move over, war, famine, pestilence and death. Meet the CEOs of Terraform Labs, Three Arrows Capital, Celsius Network and Voyager Digital. For blowing up crypto, Do Kwon, Su Zhu, Alex Mashinsky and Stephen Ehrlich are four of CoinDesk’s Most Influential 2022.
An NFT of this image was sold at auction on Coinbase NFT. A percentage of the sale went to oneearth.org.
What a difference a few weeks makes.
Every year, CoinDesk plans its Most Influential list starting as early as August. Back then it seemed obvious who belonged in the rogue’s gallery of the 2022 crypto collapse (crippling an industry is, after all, a kind of influence, too). The four CEOs behind Terraform Labs, Celsius Network, Three Arrows Capital and Voyager Digital built flawed products, enticed retail investors to take huge undisclosed risks and borrowed heavily to lever themselves up to nosebleed heights. Then each of them blew up spectacularly.
Now, of course, we’re dealing with a seemingly even darker figure: Sam Bankman-Fried, who not only did all of the above but also lent funds to himself and his inner circle, tracked billions in assets on a pile of crumpled napkins and strategically posed as a beneficent public-minded messiah while doing it. (That’s why he gets a story all to himself.)
Read more: Presenting CoinDesk's Most Influential 2022
In comparison, someone like LUNA creator Do Kwon could now seem almost innocuous. And the Four Horsemen, being the Machiavellian narcissists they seemingly are, have indeed seized the opportunity of Bankman-Fried’s unmasking to try and mend their own reputations. Kwon and Su Zhu in particular have stuck their heads up from various non-extradition spider holes to humanize themselves on chummy podcasts and argue that everything was SBF’s fault, really.
So maybe it’s a good time to remind ourselves: Do Kwon, Su Zhu (with co-founder Kyle Davies), Alex Mashinsky and Stephen Ehrlich are hardly better than Sam Bankman-Fried. All of them acted out of ego, ignorance and reckless disregard for the people who trusted them with their money. None of them were actually as smart as they thought they were. Each is responsible for huge losses inflicted on more honest (if lazily trusting) players across the cryptosphere. All of them enriched themselves personally as they strode away from the destruction wrought by their failures.
And none of them deserve the redemption arcs they’re seeking.
Who killed crypto?
As we gaze over the ruined landscape of the crypto market at the end of 2022, with assets down by more than two-thirds since one year ago, it’s impossible not to ask: Whose fault is this?
The biggest, most fraudulent and most interlinked failures are easy to list. Do Kwon’s Terra ecosystem was the first keystone to crumble, in turn helping reveal weaknesses at Three Arrows, Celsius and Voyager.
Terra was also, by some measures, the largest failure. Terra’s stablecoin terraUSD (UST) began to lose its peg on May 7, 2022, and the death spiral that followed was an inevitable consequence of its flawed design.
At their height, Terra’s LUNA token and UST were valued at $60 billion, though that included a lot of illusory implied value. It’s still not entirely clear how much real money from investors was vaporized, but the amount is likely in the tens of billions.
And a lot of the fake value was on the books at entities such as Three Arrows, which had about $200 million worth of LUNA on its balance sheet right alongside UST and bitcoin (BTC). Three Arrows, in turn, had received large loans from entities that aggregated retail investor funds, including about $75 million from Celsius.
But trusting Three Arrows was simply one misstep among many at Celsius, which also lost as much as $350 million in customer funds through an apparently hapless active-trading program. The lending platform began to wobble as crypto markets declined over the course of early 2022. On June 12, Mashinsky issued his notorious demand that investor Mike Dudas find “even one person who has a problem withdrawing from Celsius.”
The platform stopped withdrawals the very next day, and filed for bankruptcy on July 13.
Three Arrows came next (though as we now know, far from last) with their disclosure of huge losses on June 17 and a bankruptcy filing on July 1. In addition to its big LUNA position, Three Arrows was also wounded by losing positions in Lido staked ETH (stETH) and the GBTC Bitcoin trust run by Grayscale (Grayscale is owned by CoinDesk parent company Digital Currency Group).
Far worse, this and other Voyager loans were, as CEO Steve Ehrlich had blithely put it in an investor call just weeks before the collapse, collateralized only in “a very small portion.” But Ehrlich assured listeners everything was fine because “the people we lend to are some of the biggest names in the industry.” “Other people trust them” is not the level of financial due diligence that should be expected of any lender – particularly one that fraudulently claimed its customer deposits were insured by the U.S. Federal Deposit Insurance Corp., of which Voyager was accused by the FDIC.
These failures and many others were linked both by a variety of loans that couldn’t be recalled and more nuanced interconnections – a case study in financial contagion.
Many smaller services acted as little more than front ends for Terra’s Anchor program, making it easier for less-crypto-savvy users to get a bit of that sweet 20% yield – and then lose 100% of their principal. Zhu and Davies’ Three Arrows was managing a reported $10 billion at the time of its collapse, a tally revealed to have been pumped up by extensive loans from entities including Blockchain.com, Genesis Trading (another Digital Currency Group company) and Voyager.
Celsius, whether by luck or foresight, managed to withdraw more than $500 million worth of assets from Terra’s Anchor system before the collapse. But Celsius was also speculating with customer funds across a seemingly huge variety of DeFi platforms and assets, and the LUNA crash drove down prices across most of that ecosystem. So even if it wasn’t a direct hit, LUNA’s blast radius still badly wounded the lending platform.
Celsius is believed to have had about $11.8 billion in deposits in May 2022, the month before its collapse. Unlike Three Arrows, though, much of that amount was deposited by individuals who trusted Mashinsky’s frequent, sneering insistence that Celsius was just like a bank, only better. Mashinsky’s rhetoric exemplified the crypto-carpetbagger’s impulse to simply parrot simpleminded slogans, while in fact contradicting one of the space’s fundamental appeals: self-custody.
Voyager had attracted fewer assets, holding about $5.8 billion in deposits at its peak. That could hold a perverse lesson, because Steve Ehrlich’s relatively restrained personal style was seemingly outperformed by Mashinsky, Kwon and Su’s Donald Trump-like declarations and belittling of critics. Voyager pursued other ruthless and damaging marketing approaches, though, such as openly lying about FDIC insurance on user accounts and running a very generous promotion with Mark Cuban’s Dallas Mavericks basketball team that wound up costing fans a bundle. So while Voyager was the smallest of the big blowups, there was still plenty of pain to go around.
Those are relatively simple measures – dates and numbers. They suggest, very broadly, that Terra was the first and biggest of the dark horses past the post, making it a flashpoint for contagion. Celsius, Voyager and Three Arrows were smaller by some measures, and came later, suggesting their failure was downstream from Terra’s.
But the causality of the crypto collapse is ultimately a lot more complicated than any straightforward timeline, or even a comparison of balance sheets. Because finance is a form of time travel.
Great expectations lead to great contagions
Do Kwon’s Terra system collapsed two weeks after CoinDesk highlighted fundamental flaws in the design of the UST “algorithmic stablecoin” (I use air quotes because there’s actually no such thing). As soon as the token began to waver from its dollar peg in early May, it was all over but the shouting.
The system had grown to such dangerous proportions for two basic reasons. First and foremost, the above-market returns promised by Terra’s Anchor system enticed many to buy the UST token for deposit there: Just a few weeks before the unwind, 72% of all UST was deposited in Anchor, meaning the 20% APR in Anchor was a major prop for the market value of both UST and the systemwide LUNA token.
It has since become clear, however, that Anchor’s returns were not actually being generated by interest on loans, but instead largely came from other sources. Those sources included venture capital investments and tokens assigned to Do Kwon’s Terraform Labs and related entities, as well as revenue generated by retail buyers of LUNA and UST. That arguably made Anchor and Terra an obfuscated Ponzi scheme.
At the same time, Do Kwon’s arrogant, venomous public persona helped keep his supporters lined up against anyone who questioned whether UST could really be magically worth $1 simply thanks to fancy math. But even as Kwon called critical analysts “retarded,” he had allegedly engaged external market makers to help prop up UST day to day. Whatever rationalization Kwon held in his own mind, UST was never actually self-balancing.
But Kwon wasn’t alone in aggressively touting his creation. A shocking number of supposed professionals across the cryptosphere invested in or speculated with his financial perpetual motion machine (including, a recent report suggests, the team at Alameda Research). Investors in turn trumpeted their faith to the public, lending the system credibility and effectively baiting individual investors to deposit into Anchor or buy LUNA. That included, most stunningly, the formerly reputable Mike Novogratz of Galaxy Digital, who proudly showed off a huge LUNA tattoo.
This is what I mean about things being more complicated than a simple timeline. Sure, Terra was the match that helped burn down Three Arrows and Celsius. But would Terra and Anchor have grown so big, or attracted so many direct retail deposits, without the apparent cosign of supposed experts like Novogratz? Or, for that matter, without the deposits Celsius itself parked there?
Do Kwon was an incredibly loud, bullying defender of his own incredibly bad ideas. But much like his biotech analogue Elizabeth Holmes, it was big-money investors who poured jet fuel onto his flaming ineptitude, creating a cataclysm.
Strong opinions, mostly wrong
“Pride goeth before destruction, and an haughty spirit before a fall.” – Proverbs 16:18
One of the things making the current downfall of Sam Bankman-Fried and FTX so unnerving is that he refrained, at least publicly, from the kind of philosophical grandstanding that is so often a red flag for bad actors. By contrast, the Four Horsemen often wore their dangerous mindsets and flawed assumptions on their sleeves.
Three Arrows CEO Su Zhu was a bit more measured and informed in his arrogance than Kwon and Mashinsky – but arguably more extreme. Zhu frequently promoted his “supercycle theory,” which posited that bitcoin in particular would continue climbing in value indefinitely. One thing Zhu seemingly missed is the role of the coronavirus pandemic in bidding up crypto, tech stocks and a variety of other financial assets in early 2020.
Three Arrows were hardly alone in this, of course – just ask anyone who lost their shirt on hot pandemic stocks like Peloton or Snapchat. But Three Arrows was so self-assured it forgot to do the “hedge” part of running a hedge fund. In fact, it did quite the opposite, levering up with all of those loans, drastically increasing the fallout when its positions went the wrong way.
Three Arrows’ downfall exemplifies the dangerous cocktail of hubris and finance. It also epitomizes a perverse and increasingly crowded pipeline from highly respected institutions into the fraudster hall of fame: Zhu and Davies first met at the hyper-elite Philips Academy in Andover, Massachusetts, before both moving on to Columbia University. Do Kwon was a graduate of Stanford.
One is tempted to wonder exactly what’s being learned on America’s most hallowed campuses.
The finance curse
Another common thread ties together our Four Horsemen: What they were doing was not really about cryptocurrency technology, but about using finance to profit from public interest in that technology.
Zhu and Davies began Three Arrows as conventional foreign exchange traders, years before they began trading crypto. Mashinsky built what was basically an unregulated but otherwise conventional bank, hoping to profit from crypto trading and speculation. Steve Ehrlich’s past experience was running E-Trade’s professional brokerage services.
Do Kwon is the arguable exception, having built an actual blockchain. But Terra didn’t bring any new decentralization technology to the table, only financial engineering flim-flam and the same artificially inflated yields that drove growth at Celsius and Voyager. (Sam Bankman-Fried was also a creature of the finance world, emerging from the trading firm Jane Street. He frequently admitted, including on an infamous “Odd Lots” episode, that he barely cared about the substance of crypto technology – only the profits to be reaped from trading it.)
It may sound strange to level criticism at speculators and finance in an industry that’s all about reinventing money. But there’s a difference between the novelty of peer-to-peer payments and smart contracts, and the all-too-mundane impulse to gamble on the future growth of a novel technology.
In a stellar profile of Zhu and Davies in New York Magazine, my former Fortune colleague Jenn Wieczner wrote, “They built social-media cred by playing the part of billionaire financial geniuses, translated that to actual financial credit, then put billions of dollars in borrowed money to work in speculative investments they could cheerlead to success with their large, influential platforms.”
You can probably think of a few other billionaires that could easily be described, from disgraced “SPAC King” Chamath Palihapitiya to no less than Elon Musk. The rise and fall of the Horsemen, then, may say less about crypto than about much broader trends at the intersection of speculative investing and social media.
What it does say about crypto is that much of the price appreciation of the 2020-2021 bull market was an illusion, based on rehypothecation and degenerate long bets at an unprecedented scale. Those who believe in the promise of crypto technology will likely spend years digging out of a pessimistic winter before the reputational damage done by speculators wears off.
Let’s remember that the next time they come knocking.
The leader in news and information on cryptocurrency, digital assets and the future of money, CoinDesk is a media outlet that strives for the highest journalistic standards and abides by a strict set of editorial policies. CoinDesk is an independent operating subsidiary of Digital Currency Group, which invests in cryptocurrencies and blockchain startups. As part of their compensation, certain CoinDesk employees, including editorial employees, may receive exposure to DCG equity in the form of stock appreciation rights, which vest over a multi-year period. CoinDesk journalists are not allowed to purchase stock outright in DCG.
Learn more about Consensus 2023, CoinDesk’s longest-running and most influential event that brings together all sides of crypto, blockchain and Web3. Head to consensus.coindesk.com to register and buy your pass now.