What is more difficult to understand is why what you say here isn't obvious to everyone. And it gets back to "what is money?" Of course, money is what everyone agrees it is. But who is everyone?
More than the rule of law and property rights, it is a 25 trillion dollar economy, the most powerful military in the world, and relative political stability for 250 years that makes most people agree on the value of dollars. Otherwise, I might be sending you a box of tulips to pay for my Slow Boring subscription.
“The Segway of currencies” is pretty apt. I think a lot of the enthusiasm for cryptocurrencies in the tech world stems from the fact that blockchains actually do solve an incredibly hard problem: distributed consensus among non-trusting parties is really difficult! If bitcoin had been someone’s CS PhD dissertation, they would have passed with honors.
And the tendency among my fellow nerds is to look at something like that and just assume that it has a million real world uses. How could it _not?_ It’s a vertical self-balancing scooter!
Except it turns out that sometimes really impressive technologies… just aren’t very useful. To my mind, the real tell here isn’t the collapse of FTX et al: it’s been screamingly obvious for a while now that the “crypto finance industry” was a Ponzi scheme about to go critical. It’s that boring companies like IBM, Amazon and Oracle are slowly winding down their blockchain projects. It turned out that while blockchains are really neat, they were strictly inferior in all respects to legacy technologies for inventory tracking and transaction clearing, primarily because the throughput limits of blockchains turned out to be much harder to engineer away than anyone initially assumed.
In the tech world this is a normal cycle and it’s why startups still hire product people to ride herd on my fellow nerds’ enthusiasms. 20 years ago XML was going to fix everything, it was a weird time. It’s just unfortunate that so many real people’s money got incinerated in the process this time.
I was in school for computer science around the time the Bitcoin paper came out. Aside from the piquancy of its anonymous authorship, everyone could see that it was an extremely clever paper. What was ingenious about it was that it solved the double-spending problem in an incentive-compatible way, and this use of economic incentives as part of system design was unusual at the time.
What was not unusual about the Bitcoin paper was the assumption of zero trust. Almost all research papers in distributed computing assumed zero trust, because otherwise you wouldn’t have a problem to solve. Few people mistook this for some kind of financial desideratum; it was an admittedly artificial assumption made for the sake of having something to write about.
The internet itself is very much not built on zero-trust systems. The way the domain name system works is that there are about a dozen root servers that have all the answers. The way SSL works is that a company called Verisign vouches that the server belongs to Bank of America, and there's a file on your laptop that instructs it to believe whatever Verisign says. The way the Internet Protocol works is that you hand off your packet of data with a destination address, and every computer along the way is just supposed to make a best-effort attempt to pass it along without any verification or compensation.
Of course, this lack of paranoia can be abused, and there are tons of proposals to solve spam or denial-of-service attacks by accounting for malicious actors. But these proposals never catch on, because the cure is worse than the disease. Just as every well-performing is a high-trust economy, the most efficient systems are ones that assume trust.
This whole bizarre cryptocurrency fever dream reads to me like everyone missing a key part of the big picture---technologists not knowing anything about economics, enthusiasts not understanding the underling technology, libertarians having no grasp of political economy. That includes me, of course. I remember that there used to be a website called the Bitcoin Faucet where you could click a button once a day and it would send you a bitcoin. If I had understood then to what weird and frankly stupid places we go, I would have clicked the button!
And you didn't even mention the absurdly high energy costs to keep the crypto machine running, which are estimated at ~0.5% of global energy consumption. That's a pretty high negative externality for something whose main purpose is mostly to enable criminal activity.
Please! More overt examples dismantling libertarian grand theories about the economy, world history, society, and systemic incentives!
You've mentioned that Peter Thiel has a very bad theory of everything, dependent on many misreadings, but I'd like to know what those are!
These guys are very influential, and yet have very specious reasoning.
I can get that occasionally they're worth endorsing because they appear like "Rationalists" who use numbers and have an aesthetic of objectivity, but they often use ambiguous language and paint themselves into a corner, all to the ends of making sure that reach people pay lower taxes.
I think Smart Contracts may be something genuinely new, but overall I agree incredibly with this post. Bitcoin and the blockchain, in particular, are just a super-energy intensive system of title with no protections for mistaken transactions.
The latter part (no protections) is mostly bad and means that as Matt says its only real applications are extralegal (arguably Silk Road was the bootstrap for getting BTC to more than $0, for example).
The first part - a system of title - is actually an incredibly valuable and important part of a system of exchange, which is *exactly why it's basically a solved problem and has been for centuries.* Title to real estate has been a concern of the state apparatus since time immemorial -- literally! "Time immemorial" in the context of (English) property law actually refers specifically to the time before the year 1189, which is the epoch for saying "this person owns this property and has sound title since they have owned it since time immemorial."
What the blockchain enables that is new, is criminal, and what the blockchain does that is not criminal, is not new.
I don't think the comparison to gold even works, as gold's properties means it can obviously be used to make jewellery as well as serving a purpose in some electronics. Crypto doesn't do anything at all
I admit to a fair amount of schadenfreude over all of crypto’s struggles. I spent 27 years in corporate finance, and have a decent understanding of how the financial system works. I’ve never been able to figure out how, exactly, cryptocurrency would be both safe and widely useful. So far, it’s neither. And with respect to gold: anyone stockpiling gold in anticipation of a complete societal collapse is delusional. Like paper dollars (and diamonds, for that matter), it only has value because lots of people agree that it does. But you can’t eat it, or shoot anything to eat with it, or grow anything to eat with it.
I used to work for an anti-money laundering NGO. We heard this argument a lot about crypto back when it was commonly used to purchase drugs on the internet, before Silk Road was shut down in 2013. We used to hear similar arguments about other money laundering methods used for crimes that people thought weren't so bad (often evading taxes, currency controls or oppressive regimes seizing assets).
Our reply was always that you can't pick what crimes a lane for money laundering is open to. If you can hide the source of funds or disguise it, any criminal can do it too. There's no fundamental difference* between money laundering to evade currency controls and money laundering to hide the proceeds of sex trafficking.
*There's some nuance here. Certain activities lend themselves to different types of money laundering. But crypto is an all-purpose workaround for our AML system.
I appreciate and agree with your conclusion here that the rule of law is an important abstract technology. I think there is some tension here with your past occasional commentary to the effect that judges, justices, courts etc. are to some degree just a pretext or front for politics. I would value your further exploration or explanation of your beliefs on the value and legitimacy of the judiciary.
This substack post - https://cryptadamus.substack.com/p/by-the-claw-satoshi-is-revealed - makes a speculative but fascinating case that the inventor of Bitcoin is... an international criminal who had a huge interest in money laundering, and has been of interest to the authorities since right about the time that Bitcoin's anonymous inventor went quiet forever.
It's a pretty compelling circumstantial case that would really benefit from some proper investigative journalism being directed at it. If you enjoy the post (which I think you will, its excellently done) it would be amazing if you could help to bring it to the attention of the type of people who might be inclined to direct resources towards digging further.
No doubt the social technology of deposit guarantees and the bank regulation necessary to support it have been very useful, especially when they were first implemented. But today I can't help worrying that they are paving the way to panopticon. Those regulations opened the door to AML and KYC, which combined with ever more powerful computing is going to give states the power to target transactions they disfavor at the individual level in real time.
As Scott Alexander said in his post - "Freedom of speech is hollow if you can’t pay the print costs for your magazine. Freedom of religion is hollow if you can’t pay the rent on your church. The freedom to protest is hollow if you can’t pay bus fare to the protest site."
Even if the US government hasn't quite crossed that line so far (and Scott cites several examples where maybe they already have), I think we need to start considering whether we might want to diversify away from the infrastructure that makes it a possibility. Just as it's very useful to have a robust clean energy industry around when you're trying to end the dependence on fossil fuels, it's going to be useful to have a robust alternative finance industry around if we want to reduce our dependence on a useful social technology that is nevertheless starting to become a big liability.
Tech people don’t understand banking. One of the worst experiences of my life was using an online bank owned by an edgelord VC type (SoFi) for a home deposit because I had to get a lot of money transferred quickly, which is just much easier if you have an actual human to work with (or at least a non-zero level of customer service). The customer service was so bad that I’m probably just going to stop using online banking entirely and just park spending money somewhere it doesn’t get interest, and keep the rest in Vanguard.
The Segway, originally intended as transportation, now mainly serves the niche 'analogy' market.
Matt,
What is more difficult to understand is why what you say here isn't obvious to everyone. And it gets back to "what is money?" Of course, money is what everyone agrees it is. But who is everyone?
More than the rule of law and property rights, it is a 25 trillion dollar economy, the most powerful military in the world, and relative political stability for 250 years that makes most people agree on the value of dollars. Otherwise, I might be sending you a box of tulips to pay for my Slow Boring subscription.
“The Segway of currencies” is pretty apt. I think a lot of the enthusiasm for cryptocurrencies in the tech world stems from the fact that blockchains actually do solve an incredibly hard problem: distributed consensus among non-trusting parties is really difficult! If bitcoin had been someone’s CS PhD dissertation, they would have passed with honors.
And the tendency among my fellow nerds is to look at something like that and just assume that it has a million real world uses. How could it _not?_ It’s a vertical self-balancing scooter!
Except it turns out that sometimes really impressive technologies… just aren’t very useful. To my mind, the real tell here isn’t the collapse of FTX et al: it’s been screamingly obvious for a while now that the “crypto finance industry” was a Ponzi scheme about to go critical. It’s that boring companies like IBM, Amazon and Oracle are slowly winding down their blockchain projects. It turned out that while blockchains are really neat, they were strictly inferior in all respects to legacy technologies for inventory tracking and transaction clearing, primarily because the throughput limits of blockchains turned out to be much harder to engineer away than anyone initially assumed.
In the tech world this is a normal cycle and it’s why startups still hire product people to ride herd on my fellow nerds’ enthusiasms. 20 years ago XML was going to fix everything, it was a weird time. It’s just unfortunate that so many real people’s money got incinerated in the process this time.
I was in school for computer science around the time the Bitcoin paper came out. Aside from the piquancy of its anonymous authorship, everyone could see that it was an extremely clever paper. What was ingenious about it was that it solved the double-spending problem in an incentive-compatible way, and this use of economic incentives as part of system design was unusual at the time.
What was not unusual about the Bitcoin paper was the assumption of zero trust. Almost all research papers in distributed computing assumed zero trust, because otherwise you wouldn’t have a problem to solve. Few people mistook this for some kind of financial desideratum; it was an admittedly artificial assumption made for the sake of having something to write about.
The internet itself is very much not built on zero-trust systems. The way the domain name system works is that there are about a dozen root servers that have all the answers. The way SSL works is that a company called Verisign vouches that the server belongs to Bank of America, and there's a file on your laptop that instructs it to believe whatever Verisign says. The way the Internet Protocol works is that you hand off your packet of data with a destination address, and every computer along the way is just supposed to make a best-effort attempt to pass it along without any verification or compensation.
Of course, this lack of paranoia can be abused, and there are tons of proposals to solve spam or denial-of-service attacks by accounting for malicious actors. But these proposals never catch on, because the cure is worse than the disease. Just as every well-performing is a high-trust economy, the most efficient systems are ones that assume trust.
This whole bizarre cryptocurrency fever dream reads to me like everyone missing a key part of the big picture---technologists not knowing anything about economics, enthusiasts not understanding the underling technology, libertarians having no grasp of political economy. That includes me, of course. I remember that there used to be a website called the Bitcoin Faucet where you could click a button once a day and it would send you a bitcoin. If I had understood then to what weird and frankly stupid places we go, I would have clicked the button!
And you didn't even mention the absurdly high energy costs to keep the crypto machine running, which are estimated at ~0.5% of global energy consumption. That's a pretty high negative externality for something whose main purpose is mostly to enable criminal activity.
I never fell for the Segway fad. I just bought them in order to flip them quickly to a bigger fool.
I had no idea they'd turn out to be such a solid long-term investment! Contact me if you'd like a piece of this lucrative action.
Please! More overt examples dismantling libertarian grand theories about the economy, world history, society, and systemic incentives!
You've mentioned that Peter Thiel has a very bad theory of everything, dependent on many misreadings, but I'd like to know what those are!
These guys are very influential, and yet have very specious reasoning.
I can get that occasionally they're worth endorsing because they appear like "Rationalists" who use numbers and have an aesthetic of objectivity, but they often use ambiguous language and paint themselves into a corner, all to the ends of making sure that reach people pay lower taxes.
I think Smart Contracts may be something genuinely new, but overall I agree incredibly with this post. Bitcoin and the blockchain, in particular, are just a super-energy intensive system of title with no protections for mistaken transactions.
The latter part (no protections) is mostly bad and means that as Matt says its only real applications are extralegal (arguably Silk Road was the bootstrap for getting BTC to more than $0, for example).
The first part - a system of title - is actually an incredibly valuable and important part of a system of exchange, which is *exactly why it's basically a solved problem and has been for centuries.* Title to real estate has been a concern of the state apparatus since time immemorial -- literally! "Time immemorial" in the context of (English) property law actually refers specifically to the time before the year 1189, which is the epoch for saying "this person owns this property and has sound title since they have owned it since time immemorial."
What the blockchain enables that is new, is criminal, and what the blockchain does that is not criminal, is not new.
I don't think the comparison to gold even works, as gold's properties means it can obviously be used to make jewellery as well as serving a purpose in some electronics. Crypto doesn't do anything at all
I admit to a fair amount of schadenfreude over all of crypto’s struggles. I spent 27 years in corporate finance, and have a decent understanding of how the financial system works. I’ve never been able to figure out how, exactly, cryptocurrency would be both safe and widely useful. So far, it’s neither. And with respect to gold: anyone stockpiling gold in anticipation of a complete societal collapse is delusional. Like paper dollars (and diamonds, for that matter), it only has value because lots of people agree that it does. But you can’t eat it, or shoot anything to eat with it, or grow anything to eat with it.
Re: Crypto as useful to commit good crimes
I used to work for an anti-money laundering NGO. We heard this argument a lot about crypto back when it was commonly used to purchase drugs on the internet, before Silk Road was shut down in 2013. We used to hear similar arguments about other money laundering methods used for crimes that people thought weren't so bad (often evading taxes, currency controls or oppressive regimes seizing assets).
Our reply was always that you can't pick what crimes a lane for money laundering is open to. If you can hide the source of funds or disguise it, any criminal can do it too. There's no fundamental difference* between money laundering to evade currency controls and money laundering to hide the proceeds of sex trafficking.
*There's some nuance here. Certain activities lend themselves to different types of money laundering. But crypto is an all-purpose workaround for our AML system.
I appreciate and agree with your conclusion here that the rule of law is an important abstract technology. I think there is some tension here with your past occasional commentary to the effect that judges, justices, courts etc. are to some degree just a pretext or front for politics. I would value your further exploration or explanation of your beliefs on the value and legitimacy of the judiciary.
Hi Matt,
This substack post - https://cryptadamus.substack.com/p/by-the-claw-satoshi-is-revealed - makes a speculative but fascinating case that the inventor of Bitcoin is... an international criminal who had a huge interest in money laundering, and has been of interest to the authorities since right about the time that Bitcoin's anonymous inventor went quiet forever.
The post rightly notes that this is one of the all time most interesting Wikipedia pages - https://en.wikipedia.org/wiki/Paul_Le_Roux
It's a pretty compelling circumstantial case that would really benefit from some proper investigative journalism being directed at it. If you enjoy the post (which I think you will, its excellently done) it would be amazing if you could help to bring it to the attention of the type of people who might be inclined to direct resources towards digging further.
In this post, Matt refers to Die Hard as a 90s cinematic classic, but Die Hard was actually released in 1988.
No doubt the social technology of deposit guarantees and the bank regulation necessary to support it have been very useful, especially when they were first implemented. But today I can't help worrying that they are paving the way to panopticon. Those regulations opened the door to AML and KYC, which combined with ever more powerful computing is going to give states the power to target transactions they disfavor at the individual level in real time.
As Scott Alexander said in his post - "Freedom of speech is hollow if you can’t pay the print costs for your magazine. Freedom of religion is hollow if you can’t pay the rent on your church. The freedom to protest is hollow if you can’t pay bus fare to the protest site."
Even if the US government hasn't quite crossed that line so far (and Scott cites several examples where maybe they already have), I think we need to start considering whether we might want to diversify away from the infrastructure that makes it a possibility. Just as it's very useful to have a robust clean energy industry around when you're trying to end the dependence on fossil fuels, it's going to be useful to have a robust alternative finance industry around if we want to reduce our dependence on a useful social technology that is nevertheless starting to become a big liability.
Tech people don’t understand banking. One of the worst experiences of my life was using an online bank owned by an edgelord VC type (SoFi) for a home deposit because I had to get a lot of money transferred quickly, which is just much easier if you have an actual human to work with (or at least a non-zero level of customer service). The customer service was so bad that I’m probably just going to stop using online banking entirely and just park spending money somewhere it doesn’t get interest, and keep the rest in Vanguard.