How to Tell the Story of Crypto – Part I
More words than you want to read about blockchain, but the fewest I could possible write
This is the first installment of a three-part essay on the collapse of crypto in 2022–2023. A new chapter of that story is clearly being written in 2024 – the price of bitcoin has almost quadrupled since its December 2022 lows. But this is all the more reason to get our bearings on what bitcoin actually is and does – because the story of bitcoin really only makes any sense as a story of money/power.
The tale of crypto is neither simple nor easily told. But we need to tell this story, and we need to tell it well, because crypto matters – not for the reasons adduced by its true believers, but because money and finance sit at the center of capitalist societies today. In that context, first, crypto came into existence by claiming to reinvent money, to literally engineer a wholly new form of money – perhaps the most audacious claim in the entire history of money and money theories. Second, in just over a decade, crypto ostensibly created a new system of finance, one that at its peak putatively circulated (but we’ll come back to this) over $3 trillion of economic value.
Despite all this, the only figures who seem to be taking bitcoin seriously are players in the industry, and most of them are (intentionally or not) running Ponzi schemes and rug pulls. Of course, economists will sometimes weigh in on the seemingly binary question of whether bitcoin is or is not money, and political theorists have sharply criticized the political ideology of the bitcoin industry.1 But these are not deep investigations of crypto as a phenomenon of money/power.
My point of departure is to say that even though bitcoin is absolutely not money, and even though crypto is an ideological mess filled with scams and schemers, the bitcoin phenomenon remains extremely important and should be taken seriously. To show how and why, this essay engages with, jumps off from, and provides theoretical background to, three very recent books (all published within a three-month span in mid-2023) that themselves try to tell the story of crypto.2 Read carefully and creatively, these works help us to develop a nascent political theory of crypto.
This can serve as an important corrective, given that theorists of politics (I count myself one) have often proceeded as if money, money markets, finance, and “financialization” only matter when they intrude on the proper domain of (democratic) politics. This is to treat money as an excrescence. But money is the blood plasma3 of a capitalist society; neither it (society) nor we can live without it (money). Money markets drive, disrupt, and frequently determine the conditions of democratic politics under capitalism. And over the course of the past few years, those money markets have been partially remade by the technological and very much political invention of so-called cryptocurrencies. If we cannot make sense of the recent story of crypto, we limit our understanding of much else in contemporary politics. The serendipitous almost simultaneous publication of these three books, then, offers a great place to start. Individually, none of these books narrate the crypto story in terms of money/power, but read imaginatively together – and with much supplementation! – they can.
I announced the challenge each book faces in my own lede sentence – the crypto tale is tough to tell; one will be tempted to simplify by forcing it into a narrative we already know well. All three authors can be fairly read as doing just that. For Faux, everything in crypto is a Ponzi scheme, while for McKenzie, it’s all one big scam. Lewis’s text is the odd one out: for him, crypto itself proves almost beside the point, as he sticks with his own character-driven playbook by trying to to tell the story of Sam Bankman-Fried as the genius oddball who sees the world differently than everyone else (thus enabling his power to change it). My plan is not to criticize each author for their simplifications, but rather to draw out strands from each book and then weave them together to produce a more robust narrative, one that better captures the crypto phenomenon as a money/power phenomenon.
The 2023 Crypto Trilogy
Let me start with a perhaps surprising conclusion: the best by far of these three books is Easy Money, the first book written (with the support of journalist Jacob Silverman) by Ben McKenzie, better known for playing Ryan Atwood on the network television teen-drama, “The O.C.” McKenzie’s book has flown under the radar, hard to even locate among all the other crypto noise,4 including both the (deserved) attention given to Faux and the (unnecessary) “controversy” surrounding the ever-famous Lewis. I personally followed the crypto story in excruciating detail as it has played out over the past few years, yet I learned more from McKenzie than the “professionals.” He surely did not set out to write a scholarly text, but his spirit of inquiry is admirable: beginning from a place of genuine humility (i.e., he’s an actor, not an academic) McKenzie assumes he doesn’t know anything, but then works tirelessly to figure it out, and best of all, to explain it carefully and with concise clarity to the reader (who he assumes can understand it).
Number Go Up is a journalistic gem. Written by Bloomberg investigative journalist Zeke Faux, it reproduces a series of major, deep-dive articles that show us the concrete implications of crypto.5 McKenzie consistently and usefully repeats the question: what are the real-world uses of crypto? Faux’s book provides perhaps the best answer, as he goes not just to the Bahamas to track Bankman-Fried (all three authors go there), but also to the following places:
El Salvador, where in a tight chapter nicely titled “No Acceptamos [sic] Bitcoin,” he explodes the myth that the country has made Bitcoin its national currency (Faux 263–271).
The Philippines, where we learn, in vivid detail, not only that one cannot become rich playing tokenized video games, but also that many poor people can become much poorer trying (Faux 160–169).
Cambodia, where Faux shines a light on perhaps the darkest corner of crypto: systematic scams to steal crypto from unsuspecting individuals, where the laborers who carry out the scam are themselves the victims of gang-organized human trafficking (Faux 225–244).
For these chapters alone, Faux’s book should be required reading. The results of his reporting provide ample evidence that crypto cannot be “contained” to some small corner of the world of finance, or dismissed as “not political” because it only affects the rich gamblers who lose.6 No one has built a real crypto company that makes a value-added product for the real economy. I repeat: no one. Nonetheless, as I will describe below, the institutions that crypto has built have real-world impact and significance.
The final book in this trilogy comes from the pen of Michael Lewis, the well-known author of such books-that-became-blockbuster-movies as Moneyball and The Blindside. Lewis got his start writing a first-person account of the rise of the 1980s bond market, and he has kept his toes in the waters of finance ever since. McKenzie came to write his book almost out of sheer shock: if he could see clearly that crypto was one big penny-stock scam, why couldn’t everyone else? Faux thought (wrongly) that Tether was a Ponzi scheme, but along the way revealed a series of other better-fool scams, Ponzis, and rug pulls. Lewis came to write his book for a wildly different reason: a friend in traditional finance – and the subject of a previous book by Lewis – was considering doing a deal with Sam Bankman-Fried and, in essence, asked Lewis to vet this relatively unknown character.7 Lewis’s book reads nothing like the other two because it neither purports nor attempts to explain crypto; the book, rather, tries to make sense of Bankman-Fried. Despite not really being about crypto (and despite whatever Lewis’s own intentions might be) Going Infinite shades in essential elements of the wider story of crypto – a point I will elaborate in Part II.
First, however, we must ourselves address the daunting task that all three authors face: explaining the technologies that make cryptocurrencies possible. Faux largely eschews the chore, and along the way his narrative – despite consistently aiming for critique – proves far too credulous: he takes the crypto story as told by crypto believers mostly at face value. According to Faux, in 2008 some person or persons (named Satoshi Nakamoto) invented a whole new currency, bitcoin, which could be transferred anonymously through a decentralized database; significantly, transactions would be verified through the energy-intensive mining mechanism of the blockchain (Faux 27–29).8
Lewis says almost nothing about the technology of crypto until the middle of the book, where he gives a one-paragraph summary of the original Bitcoin white paper that’s even more boiled down than Faux’s. He then appends a footnote that begins with, “that’s it for crypto explanations for the moment” – though there are no other crypto explanations in the book – before closing with the suggestion that bitcoin is just too hard to understand to even justify trying (Lewis 133). Back in the main text, Lewis continues even more ingenuously: “How Bitcoin worked was interesting chiefly to technologists; what it might do was interesting to a much broader audience. It might allow ordinary people to exit the existing financial system” (Lewis 104).
This is not just lazy; it’s dangerous. If it were really possible for bitcoin to revolutionize money – the crypto claim that Lewis (and to a lesser extent, Faux) repeats – that could only be because of the way the technology of bitcoin transforms (or creates anew) the very nature of money. Clearly that means we need some understanding of both the technology of bitcoin and the being of money.
Both crypto theory and practice rest on a kind of constitutive impossibility (or negative dialectic) concerning the technology of crypto. Crypto advocates defend it as utterly paradigm changing and world transforming: crypto makes possible, they claim, a whole new form of money and a complete restructuring of both finance and society. In response, critics have been asking the obvious questions for many years now. If crypto is new money, why doesn’t it function as money? And if the crypto “industry” is a real industry, what does it actually make or do? The pro-crypto rejoinder boils down to two distinct but related replies: 1) you don’t understand blockchain, and 2) have fun staying poor. The latter seems to wield a genuine degree of affect on twitter and reddit, but is obviously a non-sequitur. The former proves much knottier. We cannot have a productive or meaningful discussion of crypto without some viable working understanding of blockchain.
What is Crypto?
Bitcoin9 is radically new only to the extent that it creatively combines three old technologies, all of which rest on one fundamental, underlying, and also old, technology – namely, the cryptographic hash function. This is an algorithm that implements a mathematical “one-way function.” A function just takes an input and produces an output. A one-way function is repeatable and irreversible: given the same input it will produce the exact same output, but if we start with the output it proves impossible to reconstruct the input (if you plugged the output back into the function, as a new input, it would just produce another unique and irreversible output).
Everyone reading this essay depends daily on hash functions to protect their internet account passwords. When you first create a password, the website does not store it on their servers (at least they shouldn’t). Instead, they immediately run your inputted password through a hash function, and then store the output on their servers. Next time you log in, they hash what you type and check it with the stored value. If you mistype your password, the hashes won’t match and you’ll get the “wrong password” message (they can know it’s the wrong password without actually knowing your password). But if you type correctly, the hashes match and you log in successfully. Most importantly, if the website gets hacked, the hackers get the hash, not your password. With the hash, they can neither log in nor derive your password. This is cryptographic security.
Note that contrary to both our instincts and to many precepts of liberalism, this form of safety does not require strict confidentiality or privacy.10 Cryptographic security differs significantly from standard notions of privacy because it depends on a degree of publicity. We’ll come back to this point as it relates to money, in Part II.
All three technologies that make bitcoin possible rely on the transparent (public) circulation of hashed outputs.
It starts with public-key cryptography, a half-century-old technology, which uses a one-way function to create a set of public and private “keys.” I can use your public key to encrypt a message that only you can decrypt with your private key. Alternatively, I can use my private key to sign an unencrypted message; you can then use my public key to verify both that I signed the message and that it has not been modified. To start, and at its heart, bitcoin is nothing more than a message sent using public-key cryptography saying, “I transfer 1BTC to you —Sam.”
We then need to record that message in an unalterable ledger, so that everyone knows the “bitcoin” I gave you is now yours, not mine (of course, a bitcoin is not actually a thing). Enter blockchain, another old technology both in conceptualization (42 years) and implementation (32 years). Blockchain makes it possible to time-stamp records so that they are unalterable, and to do this it also relies on one-way hashing functions. Each “block” of recorded transactions itself contains the hash of the previous block. In essence, this means that the new block incorporates all the prior blocks into itself, creating an indelible chain of blocks, a blockchain.
And this leads, finally, to the hardest question of all: how do we make certain no one forges the ledger (giving themself all the coins, transferring coins they had already transferred, etc.)? Once again, the answer starts with an already extant technology: proof of work, first proposed in 1997 and implemented in 2002. The simple idea says that the prover (or worker) must engage in some degree of computational effort that the verifiers can confirm through minimal effort of their own. Bitcoin implements proof of work in the form of hashing. Here the provers are named “miners” and they must do the following work: take the hash of the previous established block, add to it a summary of new transactions, and then append a random number (a “nonce”) to the end. Finally, hash that entire thing. If the output is small enough, which means if it starts with enough leading zeroes (currently 20), then the block is considered “mined” and can now be verified by all the other computers on the network (with the winning miner awarded bitcoin for their “effort”).
Mining is computationally difficult and thus energy intensive: the only way to produce a hash starting with 20 zeroes is to keep trying a different nonce and hoping the result is successful. But mining is also dumb and should not be analogized to “solving” “hard math problems” because mining is in essence a lot of completely (necessarily) random guessing. Verifying the results is easy: all the other computers on the network use the winning miner’s announced nonce and come up with the same 20-zero-beginning hash. The combined process (proof plus verification) assures the integrity of the ledger: if the lucky miner (who found a hash beginning with 20 zeroes) had started with a forged ledger (giving themselves all the BTC), then none of the other computers would verify that result, because to get the exact same hash they must start with the exact same input, which itself includes the hash of all the previously hashed blocks.11
That’s bitcoin: a computer program that maintains a decentralized database of time-stamped and signed messages that can create and transfer a specified quantity of a variable named “value” (i.e. BTC) from one address to another. “Bitcoins” – that is some quantity of the “value” variable in the bitcoin code – are generated by the protocol itself, as reward for proof of work that makes it possible to verify the indelible chain of signatures.
Bitcoin is like an automated rewards program that itself has no link whatsoever to any company or product. The bitcoin protocol produces bitcoin points and lets one user transfer them to other users running the bitcoin protocol. It’s as if your Starbucks app continued to work, even though all the Starbucks stores shut down. Now the “stars” (rewards) are generated and can be transferred within the app itself, but they would only be of any value if you could find someone to pay you real money (outside the app) for your “stars.”12
…
You might now be asking: what does any of this have to do with money or power?
Part II is just below:
There have been some broadsides against the neo-libertarian politics of crypto actors (e.g., Golumbia 2016) and some careful and clear delineation of crypto theory/ideology (e.g., Eich 2019), but to a certain extent crypto itself becomes epiphenomenal in these accounts. Frances Ferguson’s bitcoin essay is exceptional in both senses of the word.
Those books are:
Zeke Faux, Number Go Up: Inside Crypto’s Wild Rise and Staggering Fall, New York: Currency, 2023, 304pp;
Ben McKenzie, Easy Money: Cryptocurrency, Casino Capitalism, and the Golden Age of Fraud, New York: Abrams, 2023, 304pp;
Michael Lewis, Going Infinite: The Rise and Fall of a New Tycoon, New York: W.W. Norton, 2023, 288pp.
The essay also draws heavily from a fourth “work,” a single article by Bloomberg columnist Matt Levine. At almost 47,000 words, the piece comprised the entirety of the October 25th 2022 issue of Bloomberg Businessweek and was originally offered with the alternate title “The Only Crypto Story You Need.”
Within this metaphor, commodities are the blood cells; combined (cells plus plasma) we get blood, i.e., “value.”
My internet search for “review of Easy Money” returned, ironically: as its first hit, a film review (of a movie of that name) by Roger Ebert; as its second hit, a review in the Wall Street Journal of an academic work of economic history, published just a few months before McKenzie’s book.
As a Bloomberg journalist, Faux has been taken much more seriously than McKenzie, with the book widely, usually favorably, though I must suggest uncritically, reviewed. To be clear, I would not endorse the criticism that has appeared. Fortune published an apparent review that turned out to be a thin apologia for crypto. The review presents a short and vague summary of Faux’s work before offering this riposte: “crypto has given rise to a massive new industry based around blockchain technology that's being used by everyone from corner bodegas to JPMorgan.” This is sheer rubbish, but what else can one expect: the reviewer is the “Fortune Crypto” editor, who, in announcing his own hiring in Summer 2022, declared that his job would also include “developing a web3 native business.” To paraphrase Molly White, I’m sure that’s going just great.
Faux’s book follows closely the path that his reporting has taken over the years, and unfortunately, despite so many great discoveries along the way, Faux’s initial instincts were wrong. Faux thought the stablecoin-issuer Tether was the biggest scam, sure to collapse. He then spent years trying to get to the truth of the Tether scam, and along the way witnessed the implosion of so many other, so much larger scams: Terra/Luna, 3Arrows, Celsius, FTX, and NFTs. As I describe below, Tether is a shadow bank – a shady one, at that – but in the world of crypto, genuine shadow banks are relatively legitimate institutions, compared to the myriad overt scams.
The friend is Brad Katsuyama; the book is Flash Boys. This work is significant for the serious mistakes it makes in trying to depict Katsuyama as the hero who fights villainous high frequency traders. As Kovac has incisively shown, Lewis gets some very basic facts about the money markets quite wrong.
That’s it; from there Faux unfolds the story through the concrete events around exchanges and coins. Throughout the book Faux repeatedly, but always falsely, refers to crypto transactions as “anonymous,” despite the fact that at some level Faux must know that’s not true. He has a whole wonderful chapter telling the story of how the perpetrators of the largest Bitcoin theft in history were tracked down through their wallet address and eventually arrested (Faux 129–159). McKenzie, writing a bit before Faux, puts the point aptly: “Bitcoin ownership is pseudonymous (but not anonymous, as is often mistakenly claimed)” (McKenzie: 8).
I focus on the technology of bitcoin, by far the most important “cryptocurrency.” Levine provides a great explanation of “the ‘crypto’ in crypto”.
The version of safety and security it implements both allows for and depends on transparency, sharing, and publicity. If I want to communicate safely, and confidentially, I need to publish my public key. And even things I want to keep “private” can themselves remain transparent in their encrypted form. For example, here is the hashed password for my main Mac computer: 086cde400741c3502a561194a87b75e6a30a0a8fd629c299da936a8f5067abd0.
In basic terms, to forge the ledger would thus require that 51% of the computers (actually, 51% of the compute power, the hashing power), all start with the same forged ledger, one of them finds a hash, and the rest verify. At first, 51% sounds vaguely democratic, but the actual principle concerns oligarchic power, not democratic majoritarianism.
This discussion focuses only on bitcoin because all the actual technological innovation lies in the construction of a proof-of-work blockchain. All proof-of-work coins function similarly to bitcoin, though there are few: the vast majority of crypto tokens are proof-of-stake. But by their own standards, proof-of-stake coins should not even be called “crypto” because none of them are decentralized.