The cryptocurrencies speedrun
In which we wonder at how cryptocurrencies appear to relive every type of financial fraud, instability and weakness but at an accelerated speed and what it could mean for the rest of us.
I’m the opposite of a cryptocurrency enthusiast: not somebody that hates or despises the concept but somebody that doesn’t believe in the reason for their existence (“stick it to man” decentralization) and therefore can look at them with curious detachment. Like if we stumbled upon a silicon-based biological ecology on another planet and found crabs.
Mt. Gox
Mt. Gox was a cryptocurrency exchange in Japan. It was early days and people screwed up or got hacked or stole or all of those things and lots of people lost a lot of money. There were no silicon-based crabs here, it’s more like finding out that silicon-based life had to breath. People in the presence of a lot of money will get greedy and steal some and try to hide that. Or make a stupid mistake or get hacked and the magnitude of the mistake will get amplified by the amount of money and/or people that it hurt. I wasn’t surprising, it wasn’t significant exobiology, it wasn’t really a learning experience aside from mild schadenfreude.
DAO Heist
The first thing that peaked my interest in “crypto as financial exobiology” was the first “DAO heist”. This DAO was designed to be a “crowdsourced venture fund”. People would buy into this fund with their money and receive votes that they can allocate in how the funds should be distributed to people asking for money to fund their ideas and turn them into things that could later have value.
This is the venture capital playbook but democratized.
This was interesting because it was “distributed computation on top of distributed financial plumbing”. It was my first exposure to Ethereum which is much more exobiologically generative than simple Bitcoin.
The distributed Ethereum program that constituted the DAO had a bug.
A clever person exploited it and managed to get a lot of the DAO funds transferred to his funding request despite having few official votes from the DAO investors.
The idea of a “smart contract” is that it promised to liberate from the shackles of human fallibility, bias and interpretations. A computer that managed the financial execution of the transactions. The evaluation was a human computation (done by voters who invested in the DAO in proportion to their investment) and the machine would allocate capital optimally.
If you have ever written any software, this tension between “what the program does” and “what the author wanted it to do” is an ever-present battle, the yin-yang of software engineering.
The “smart contract” was dumb…. but it couldn’t be changed and the person that managed to “hack it” in fact, just “used it”. The fact it wasn’t executing what the DAO investors wanted to execute was irrelevant.
The DAO investors were more people and had enough computing power and authority to perform a “hard fork” of the underlying ledger thus rendering the “fraudulent” transaction unrecognized by the distributed financial plumbing.
A decentralized “stick it to the central bank” financial system had managed to reinvent the central bank crab from first principles to fix the lossy side effects of a de-humanized representation of their will turned into decentralized robotic form.
Terra / Luna
Low interest rates, pandemic forceful lockdowns, lack of sports and government capital injections combined to make crypto FOMO material.
But people told that crypto was a sound “long term storage of value” didn’t really like the volatility that cryptocurrencies exhibited. This created the need for “stable coins” a way for regular people to “dip their toes” into the crypto market with (apparently) less risk.
Unlike regular cryptocurrencies that backed exclusively by the will of the people to consider them valuable, a “stable coin” is a cryptocurrency that is backed (ideally) by some real-world asset and/or force that gives it its value. It could be gold coins, or government bonds, or cash in a mattress, whatever one could sell to redeem the value digitally denominated in the “stable coin” to whatever other form of value one wants (generally, fiat currency backed by the taxation systems of existing governments).
So, Terra was the stable coin (backed by real value) and Luna was the cryptocurrency (backed by collective desire). They lived in this cozy ecosystem of delusion in which one would invest on things with Luna, make money, take out the profits into Terra and park it there until a new investment opportunity came along, rinse and repeat.
It’s effectively gambling… but the casino is unregulated and ran in South Korea by people that started it 4 years ago and had $48B of combined market capitalization.
At one point, unclear why, Terra begins to “break its peg” to the US dollar. Maybe some “whale” decided to collect his Terra casino chips and walk. It was May 9th 2022 so it’s entirely possible that somebody needed to pay real taxes on all those gambling gains to the very real IRS and its very real associated judiciary system, cops, guns, jails and all that stuff.
But you’re probably starting to see the problem: why would anyone versed in crypto take boring fiat money and invest it in boring government bonds or ancient form of stored value? Why wouldn’t we just “diversify” it some more and get some additional value of out it?
Because input and output from your casino tokens might become correlated. They are not independent random variables. And, also, they are not uniformly distributed in value, but given to “preferential attachment” where more money generates more money, the distribution of tokens in accounts emerges as a power law.
This is the second exobiological “crab” we find: “decentralization” and “clustering” are different things. Google, Amazon and Facebook didn’t end up owning their respective markets because they owned the protocol, but because “preferential attachment” made the rich get richer.
The idea that the current financial system is centralized because it’s corrupt is thermodynamically wrong: any purely decentralized system will cluster just like extremely uniform matter in empty space after the big bang created stars and galaxies and vast quantities of near-empty space in between even if gravity is a neutral party and doesn’t really care about who wins and loses.
Crypto is like looking at a star and wanting to give light to everyone by creating a form of “really decentralized gravity this time” that won’t cluster into one giant star but would give pocket-sized fusion reactors for everybody to enjoy.
So, Terra depegs, people panic, there is a bank run, Luna’s value goes to zero, then Terra’s value goes to zero and people end up with no pocket-sized fusion reactor in their pockets floating in financial empty space wondering how it is possible they lost all of their family savings in the course of a week.
Celsius
Another way to capitalize on crypto FOMO was offering a “cryptocurrency savings account”. Basically “give us your crypto and we’ll make it work for you”. They guaranteed yields of up to 6%/y.
How would they made this money go up on your behalf? There are ways to do this: you have more money than ideas and others have more ideas than money. The broker matches the two and takes a cut of each transfer. You facilitate turning somebody else’s idea into additional value, they get the additional value (for a minority cost of capital) and the financial broker gets a cut each way.
But this is what a boring old bank does and banks are heavily regulated precisely so that they don’t do shenanigans with such matches or cuts or instead of lending to low-risk ideas go to Vegas to gamble it.
The best part about the Celsius speedrun of the need for banking regulations is that Celsius ran on a campaign of denigrating regular boring banks and their abysmal savings account yields.
I mean, sure, it’s hardly surprising that for-profit financial services have their own interest ahead of yours, but claiming that one could offer 6%/y yields vs. 0.5%/y simply because they were crypto focused and not corrupt like the old financial system is the kind of pathetic snake oil marketing absurdities that puzzles us when we see it depicted in old western movies. Look at all those poor idiot folks and their gullibility buying into potions and elixirs and miracle cures and gold rushes…. we would never fall for such schemes, with the entire human knowledge literally at our fingertips and in our pockets every minute of every day.
Celsius had been using the crypto custodian Prime Trust to store some customer assets since March 2020. This relationship ended in June 2021, when Prime Trust's risk team expressed concern about Celsius's strategy of "endlessly re-hypothecating assets ... lending the same assets over and over and over again to juice yields". Prime Trust founder Scott Purcell suggested that re-hypothecating "would be destined for failure as any sharp market movement in either direction would be catastrophic to such a ridiculously leveraged business model".
At this time, Celsius at $18B dollars in assets. It had 1.7 million customers and it offered yields as high as 17%.
In the course of a week in June 2022, Celsius collapsed and declared bankruptcy leaving $1.2B hole in their balance sheet.
The “crab” reinvented here is the reason why FDIC insurance exists and why laws for the regulation of money lending are beneficial and why “free market” zealots are in fact trying to get away with risky behavior where their upside is much higher than yours.
FTX
Which brings us to the piece de resistance (well, so far, at least) of the crypto speedrun of the financial sector history of undesired behaviors.
Sam Bankman-Fried (aka SBF) went from child prodigy crypto billionaire with humble attitude and “save the world” complex to on-the-run fraudulent villain in less than a week.
First off, the guy is literally called “bank” “man” “freed” which is the most “crypto” name that I can think of. Feels like the name the Southpark writers would come up with for the cartoon version of a fake story on the risks of cryptocurrencies.
I can’t match the hilarity of other commentaries on this (for example, Patrick Boyle’s) but it’s worth explaining what happened.
FTX was a crypto exchange. Like a stock market for crypto. It matched sellers of cryptocurrencies with buyers. It spent lavishly in ads to become known as “the place to trade crypto”, like the Nasdaq or NYSE for stocks.
But again, crypto people are all if not ambitious and enterprising so getting paid to match buyers and sellers wasn’t enough. They created their own “coin” called FTT which didn’t work like stocks in FTX (because that would be a “security” and that would be heavily regulated) but more like airline miles: the more you had, the more you could have perks like free trades.
Then they had Alameda Research, a hedge fund that invested into crypto ventures (and effectively operated as the big bank that bailed out other failed crypto ventures).
But here’s the fun part. Ideally FTX and Alameda were two separate entities ran by the same people. A totally legitimate and benign thing when done properly. The impropriety was that FTX used the funds by traders during the trades to provide liquidity to Alameda Research… and the “collateral” used by Alameda to secure those funds were FTT tokens that attached to the value of FTX.
So, to recap, people put money in FTX to buy and sell crypto. FTX tells people that they don’t do anything with those funds, they just sit there. Instead, FTX gives them to Alameda to invest in risky things. Alameda gives FTX some collateral value to secure those loans and this collateral is in the form FTX airline miles.
See the problem? The value of FTX is now correlated with the value of Alameda AND of this fact remaining secret.
Which, of course, it didn’t.
All it took was Binance (the other big crypto exchange) deciding to liquidate their $500M worth of their FTT (FTX airline miles) holdings. People thought Binance knew something they didn’t and started selling their FTTs too. Which caused a bank run on FTX, which collapsed the value of the collateral of Alameda, which then made FTX go from illiquid (I have your money, just not right now) to insolvent (I lost your money) effectively overnight.
(some say this was a clever move by Binance’s CEO CZ to effectively destroy SBF who he sees as an arch-rival, but honestly, it doesn’t add or detract to the financial exobiology nature of the story).
Cherry on top, once the FTX boat was sinking leaving an estimated $8B crater of missing money in its wake, $600M of the $1B of the only liquid assets FTX had got hacked and stolen. Maybe by FTX insiders themselves? Unknown at this point.
This story has it all: benevolent billionaire living in Bermuda in flip-flops, crazy hair and driving Toyota Corollas. Secret incestuous relationships between apparently isolated businesses. Comically high risk tolerance. Pathetically low epistemic hygiene from the parts of VC investors.
I mean, hell, there are even Silicon Valley episodes about this (Keenan Feldspar, the VR prodigy, literally looks like SBF) written several years earlier about how VC has blind spots around the ability to distinguish this type of magnetic self-confidence with the competency to turn investments into solid value (similar story goes for Adam Neumann and WeWork although crypto didn’t factor into that one).
Convergent Evolution
Carlota Perez describes a framework for technology and socio-economic development which represents it not a straight line but a series of stacked booms and busts.
The real advantage is not in the boom phase, but in the “whale fall” opportunity explosion left by the over-development of the boom phase.
It’s abundantly clear that crypto is a bubble and that it’s bursting but what it’s not clear is what value this will leave behind.
Many think that the technology is what will be left behind. The things like the Ethereum virtual machine, or blockchains, or some smart contracts honed by the use of millions under constant threat of attacks… but I am starting to wonder is the most valuable long-lasting effect is this convergent evolution that shows that the ills of the current social system are not inherently due to corruption or greedy actors but that a lot of it is the side effect of the natural complexity of such systems.
It explains the need for regulation in a single lifetime and on one’s own skin, something that no educational institution in the world can match.
The future directors of central banks might have cut their teeth into financial markets directly by creating or participating as witnesses. Future bank CTOs might have written blockchain hobby projects on Github in their youth, but it feels much more significant that future legislators could be older versions of young adults that caused the loss of all of parents’ life savings in failed crypto ventures they felt significant and visionary.
The leftover perezian value of crypto’s decay might not be at all in the technology substrate or in the cryptographic innovations, but in the education in financial plumbing and systemic thinking that is providing for millions of people.
Unfortunately, it’s unclear what the societal cost of this education is and whether it was worth it.