243 comments and no one has mentioned Gresham's Law? ("Bad money drives out good.") That's not the only reason so few people use btc as normal money, but you can glen interesting insights by thinking that through.
Can anyone offer an explanation about the policy/impact differences between restricting capital inflows vs outflows? I’ve realized I’ve sort of digested the leftish take that outflow restrictions are important for restricting capital flight, but haven’t really thought that much about it. Why would you want to restrict inflows vs outflows?
Restricting inflows can help you control your economy - like controlling who you allow to invest in your company. Restricting outflows can be useful, too, in theory, but in practice it almost always just politicians covering up poor management - like not allowing investors to sell out when they no longer want to be invested in your company.
I'd point out that Tabarrok's problem doesn't exist in country's with far less wealth than ours that have actually invested a bit in their financial systems. Places like Brazil have instant transactions. Blame the federal reserve or the treasury department that our money moving system is so clunky, but it's a solved problem globally and crypto fixing it in the USA (where people are among the least worried about instant deposits) seems absurdly niche.
I have been employed full-time at a blockchain/cryptocurrency startup as a researcher/engineer for 18 months, moving from a job at a bank after I discovered (thanks to the pandemic) that I didn’t mind remote work. I have also been a big fan of Matt’s writing since his Slate days, and politically find myself very much in alignment with him (this is not a coincidence, but a measure of his influence on my thinking). I thus think I am pretty well-positioned to provide a counterpoint. I am going to give my perspective on the reasons to be optimistic about blockchain/cryptocurrency in this post, and respond to Matt’s points in a second post.
The concise and high-level reason I’m optimistic about blockchain finding many use cases, is that blockchains allow the transfer of value over the internet. I will try to explain, but I want to emphasize that I’m as skeptical as many critics about many of today’s blockchain/cryptocurrency projects; in the long run, though, I’m very optimistic.
I will first briefly describe what a blockchain is, before going into a near-term use case I could see for a blockchain. After that, I’ll step back and argue for the assertion above. I should note up-front that I am much less optimistic in the specific use-case I discuss than I am that blockchains will find some use-case, I just think it’s helpful to go through the possible benefits using a concrete example.
(I should preface this with a disclaimer that my background/work is as a mathematician, and to a lesser extent on the modeling/economics side; my expertise is not actually in blockchain engineering per se. I may very well get some details wrong below, but I am pretty confident that my high-level explanation is correct.)
Blockchains are protocols which allow a decentralized network of computers to come to consensus on a state. You can think of the state as a spreadsheet. On the Bitcoin blockchain, the spreadsheet just records which accounts have how much Bitcoin. On Ethereum, the spreadsheet holds numbers which actually define the state of the “Ethereum Virtual Machine”, which we can think of as a decentralized computer. This is possible because everything stored on a computer – including the code that makes up an application, including the operating system itself – is ultimately stored as numbers in addresses. By coming to a consensus on what numbers are stored in which addresses, the Ethereum network can come to a consensus on the state of the Ethereum Virtual Machine. Although Bitcoin was the technological innovation that kicked it all off, my optimism entirely derives from what Ethereum shows is possible.
All blockchains I’m familiar with require a cryptocurrency to function, because the protocols must incentivize decentralized networks of participants via systems of rewards and punishments. The cryptocurrency must be desirable for the participants (for the incentives to be meaningful) and attainable for new participants (if you want the network to be able to grow, effectively a requirement for decentralization). Though this does not automatically mean the cryptocurrency must be possible to exchange for money, but in reality this is difficult to avoid. For most newer blockchains (if they use proof-of-stake), it must also be very difficult for anyone to accrue too high a percentage of the cryptocurrency, meaning that the more money a blockchain’s cryptocurrency is worth, the more secure the network. But while cryptocurrencies are necessary for the operations of blockchains, the most exciting potential lies not in the cryptocurrencies themselves but in the blockchains they enable.
The most interesting near-term use-case, in my view, is the surplus of recent examples in which software companies develop network products with the intent of then sitting atop those networks extracting rent.
Take Doordash, for example. Doordash is an app which operates a 3-sided market of restaurants, delivery drivers, and customers. Doordash, obviously, is not a monopoly (I have no idea if it’s even the largest food delivery app), and it’s easy for all participants in the network to use multiple apps. In theory, this should mean that the food delivery apps are competing to give the best experience and pass on the most profit to delivery drivers and restaurants, and to give the best experience to users.
It’s not clear that this happens in practice, although I’ll confess I don’t know much about the details. Local governments (like mine in Minneapolis) have found it necessary to pass regulations preventing some sketchy practices by these delivery apps, to which delivery apps have responded with punitive fees which they tell customers are direct results of decisions made by their local governments.
But what if the restaurants and delivery drivers had direct *decentralized* control over the network, and dictated to the software developers how the app should work? This could be a compelling use-case for a blockchain. Restaurants/drivers would earn the network currency by participating in the network (by making/delivering food). Holders of the network cryptocurrency (primarily or exclusively restaurants/delivery drivers) would then get a say in decisions made about the network. Note that although the money transferred from customers to drivers and restaurants would likely be recorded on this blockchain as evidence of participation in the network, the use of the blockchain here isn’t actually about getting any of the actors involved out of the banking system or anything like that. It’s about changing the economics and power dynamics of a network product.
To dig into one particular detail which I think may be helpful, note that a blockchain in this application would *still* have “middlemen” (this is somewhat a heterodox framing of mine, that most crypto proponents would disagree with, but hear me out). The centralized middleman (Doordash) would be gone, but the computers that would be taking transactions and using them to update the decentralized network state would each be under the control of someone. Perhaps the restaurants and delivery drivers would be running these computers, but more likely they would be run by specialists (this is what we have seen in similar existing blockchain networks). These specialists can be seen as the middlemen in the *new* system. However while existing middlemen like Doordash can be restricted only by regulation, the new middlemen who specialize in running the blockchain can have the scope of their actions restricted *by code*. This opens up a huge design space which we as blockchain researchers are only beginning to explore.
The technical details of the specific food delivery use-case are not really the point, and certainly there are arguments for why this might not work. But it seems very, very premature to discard the idea that blockchain *might* work in some similar use case.
(The U.S. Dollar is also a network product, and many arguments about crypto center on whether it can or should disrupt the Dollar. But I believe that both critics and proponents of crypto who focus on this particular network product are thinking too small.)
To take a step back from particular use-cases, my basic reason for optimism is that blockchains allow the transfer of value over the internet. This isn’t about believing that the dollar value of this or that cryptocurrency won’t go to zero. The internet allows frictionless transfer of information, but cannot provide any guarantees about scarcity, as it relies on this information being trivial to duplicate. By allowing the possibility of decentralized consensus on state, blockchains allow for credible scarcity on the internet, and thus for the transfer of value.
The internet has transformed the world, I think I can claim without much controversy. The ability to transfer information instantly and frictionlessly around the world has found use cases that were unimaginable to the technological innovators who set us on this path, much less anyone who thought such innovations would lead to *no* real use cases. Given this recent history, I have to believe that the ability to transfer value in addition to information will be a very big deal, the “Value Superhighway” rather than the “Segway of Finance”.
*Cisco & UN have been working for awhile to make cheap cell phones accessible in nations with little banking structure to allow mobile banking over sock-money. Crypto is only an extension of that program, as folks wouldn’t have access to crypto without the cell phones in the first place. It’s really not the solution itself, just an extra layer of computation put on top of the UN solution.
Evading-bad-policies has a little more credence but is becoming moot in the US since that vast majority of retailers are using digital financial exchanges or crypto-exchanges that are monitored. I am rather salty about it because I had my preferred name on Venmo when it started requiring ID verification for transaction monitoring, and my account got permanently locked. Staff claims there is no way to change account names. Seems odd considering there are, if I recall correctly, >15 million women in the US whom have changed their legal names.
In all the countless times I have watched Die Hard, I have never thought once until this moment about what on earth Hans Gruber was stealing! And Rickman lingers on the pronunciation of "bearer bonds" in such a memorable way, odd that it never clicked. I just accepted it was a valuable thing to steal and moved on.
You're aside regarding Alex Tabbarok is a nice little reminder that as much as he does write some interesting Marginal Revolution posts (and op-eds) and is a smart guy that I find myself sometimes agreeing with. He's at his heart a crank. Like "end the Fed", full blow Austrian economics crank.
Revealing couple pieces to me. Especially the NYtimes op-ed (he's not 100% wrong, long term investment in real estate was a good bet. But yeah he's basically Larry Kudlow 2.0 here).
"I’m actually pretty bullish on Bitcoin partially displacing gold as a paranoiac’s hedge over time."
This has already struck me as hilarious. Society is falling into total chaos or Big Brother will totally take over and control your every waking minute but don't worry, because there will always be a working and totally independent Internet.
These paranoids are such trusting babes in the woods.
> Former President Donald Trump on Thursday revealed a new foray into NFT sales, sharing a website that offers a series of what he called “digital trading cards” for $99 each.
> On a website called CollectTrumpCards.com, the 45th president is selling a series of artworks featuring likenesses of the 45th president depicted as a superhero with laser eyes, an astronaut, a John Wayne-like cowboy and, of course, a dark-suited Trump surrounded by gold bars and one of him golfing.
> But Trump’s launch comes at a time when the NFT market has stagnated, with prices dropping 97% since the start of the year.
> “Remember, Christmas is coming, and this makes a great Christmas gift,” the 45th president said in a video accompanying the launch.
Is this true? Not a techno nerd myself so I have no opinion but I'd appreciate the smart SB folks weighing in. Crypto is certainly impressive in the vast amounts of computer power (and planet-killing energy requirements) it consumes, but is it impressive technology? And if so, why? Say, compared to AI, VR, mRNA vaccines, fusion power, current chip technology, and on and on.
Most newer blockchains do not use proof-of-work, and therefore do not have energy consumption downsides you refer to; these days pretty much only Bitcoin uses proof-of-work (Ethereum recently switched away from it).
Blockchain tech is only impressive at the intersection of computer science and social coordination. There are plenty of very good computer scientists who don't get it because the problems it attempts to solve aren't purely computer science problems, they are also social coordination problems. There are plenty of people who focus on social issues who don't get it because they see crypto/blockchain enthusiasts with whom they have political disagreements, and they fail to understand that the underlying technological/computer science innovation isn't about any one particular political persuasion but in fact can be applied to different social coordination problems in different ways. So there are a lot of legitimate experts who don't get it.
This also makes it hard to say it's "technologically" impressive specifically. But it's an interesting computer science solution to problems of social coordination, and my own guess is it's not a question of whether we'll use it, but how.
Bitcoin is still a pretty big deal but sure we seem to be moving towards a day when the most evil part of crypto is in the rearview mirror and its contribution to heating the planet, while locked on, is no longer moving apace.
As for it being impressive because of its combining computer science and social coordination, I read that to mean it's still all vaporware -- certainly not an argument for it being impressive in any form. I find the idea that (mostly libertarian) computer scientists being the ones whose vision of how to improve social coordination rib-tickling when it's not downright scary.
I'm not a fan of Bitcoin personally, because I think the libertarian/monetary policy argument is a bit silly, and I also don't buy the idea that proof-of-work (which is the source of the energy usage) is better than proof-of-stake, but I will very slightly defend it by saying: We each believe that the things we think are important are worth using energy for. Bitcoin uses electricity, which can easily be generated in clean ways. So to me, arguments about the energy usage of Bitcoin are pretty much beside the point.
As a progressive who works for a blockchain startup, I can tell you that it's not just -- or even mostly -- libertarian computer scientists. You don't have to be impressed by anything, of course, I suspect that someday the apps you use will simply switch to using blockchains on the back-end without you knowing, and you won't have to worry about it :-)
It's the visionaries, residing safe and happy in their liberal democratic nations, believing that the blockchain will liberate us from the suffocating hand of government who would worry me if I thought there was much to this.
But if it turns out that blockchain technology does some useful technical things in the background that common people like me will unknowingly benefit from, then great. Like html, TCP/IP, and all the other million things that are the backbone of our world that attract no attention, don't make billionaires out of speculators (if not worse) etc. Perhaps the first rule of blockchain success is that no one needs to talk or think about blockchain.
"Perhaps the first rule of blockchain success is that no one needs to talk or think about blockchain." - Most of us actually building in the industry basically agree with this (as opposed to people using crypto to gamble, who do very little except talk about crypto, which then unfortunately means that most of the general population associate crypto only with them).
A cryptocurrency is *required* for a blockchain to operate, but many users who will benefit from it won't need to worry about it at all, I hope. For example, one possible use-case would be something like a version of Doordash that is owned not by the VCs who funded the app, but by the restaurants and drivers themselves. The restaurants and drivers would need to be aware on some level that the network was operating as a blockchain, only because the power and responsibility for making decisions would belong to them. But the customers would have no reason to care or know about those details, or deal with any cryptocurrency at all.
A core principle of nerds is that we get annoyed when normies discover something that we're into. Usually they just suck all the fun out of it and ruin it for everyone. But this time I got a free car out of the deal.
Bitcoin was a fun way to put to use the pile of old graphics cards and motherboards I had in my closet. All the core developers hung out in one place and were fun to talk to about cryptography and computer science stuff. Getting their code to compile and optimizing hash rates was fun. Downloading the blockchain was fun. I didn't even know or care what cryptocurrency was supposed to be for, it was just a fun thing for a computer nerd to do on weekends, like ripping DVDs you have no interest in watching or hacking Dish Network cards when you don't even have a satellite dish. And there were some weirdos out there who would trade you something of value if you figured out how to send them some of your Bitcoin.
Then, suddenly, you could trade Bitcoin for money, which was totally weird but also a cool way to squeeze some value out of old computer hardware (and free electricity at work). Fast-forward a decade and I cashed out for enough to buy a car outright and cover all the taxes. Pretty neat! And I still use Bitcoin for transactions in foreign currencies and to pay for various online services that are a pain or whose purpose is defeated by involving a middleman that has to Know Your Customer, like VPNs or server instances.
I have no point to make, I just thought I'd share my experience with crypto since Matt was kind enough to point out that Bitcoin is to FTX what BBSes are to Twitter. Most people who write about crypto focus exclusively on the speculation and facilitation of crime.
I've had a lot of conversations with various levels of laser eyed people "on here" and in real life. They seemed to never have a good answer to the life changing use case question, the answers they gave vs the commitment in time and money ratio never lined up. "Oh, cool, you're worried about instant settlement of large international transactions - Is this a problem you have frequently in your life as a normie data analyst?" "You were just mocking African countries, and haven't shown any interest in the welfare of people outside your own family, am I to believe you're really excited about Nigerians using BTC to get better prices on imported goods, and that's why you spend so much time and money on this technology?" It was all just weird.
Yeah, this is a lot of people's experience and I think it's because fundamentally blockchain/crypto is still at a very early stage and a lot of different people in the industry have a lot of different use cases in mind (source: I work in a blockchain startup). It doesn't help that a whole lot of people who like crypto like it because they like gambling, rather than because they are interested in the actual tech, and they tend to be noisier than those of us who are building/researching the tech.
I have been surprised by the lack of pro-crypto takes in these replies, and I tried to give my version of that in a comment directly on the post (I'm not familiar enough with Substack to figure out how to link to my comment), which I'll try to verrry briefly summarize...
Basically, in the near run I think blockchain might be an appealing technology for products with network effects where having a single owner/middleman in the network is not ideal. Currencies are network products and you'll hear a lot of crypto enthusiasts who want to e.g. replace the U.S. Dollar, but this isn't the most obvious use-case to me. I'm much more interested in e.g. building a food delivery app that is owned and controlled by the restaurants and delivery drivers instead of owned by some centralized software company who, once they attain network dominance, can just extract rents.
In the long run, it's seems clear to me that while the internet allows frictionless transfer of information around the world, blockchain technology allows the transfer of *value* which is a really big deal.
Your aside on the early Web is a bit misleading. It was invented as an iteration of scholarship tools, then adopted for business use. Recommend Berners-Lee's Weaving the Web. My summary is here: http://whatarecomputersfor.net/the-worlds-knowledge/
Where did you see Punisher: War Zone? I'm hoping you saw it in the US the day before you flew to Berlin through Copenhagen, and not that you spent precious Copenhagen or Berlin time watching Punisher: War Zone.
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.
They did manage to catch and prosecute criminals before the KYC/AML system was created. It's never been clear to me why we tolerate so much financial surveillance and reporting in ways we would never tolerate analogous physical surveillance. The system just seems like a way to circumvent 4th amendment protections to me.
Then again, physical surveillance is getting much more powerful these days too and is generating minimal outrage. Still, as far as I know, most places operating security cameras aren't sending automatically generated mandatory reports to the government about the comings and goings of all people on their premises including gait and facial recognition data.
I don't have much to say other than you are wrong about the importance KYC/AML system. I mostly worked on policy in countries that didn't have an effective AML system. The alternative is financial anarchy, massive corruption, tax evasion and crime on a scale unimaginable in the U.S. If you think Nigeria is a good place to head toward, then kill the KYC/AML system. There's a reason why every developed country in the world has some version of it.
Just one example of what banks do when KYC/AML isn't enforced:
As another KYC/AML advocate, I just want to chime in with supportive analogy.
You need a driver's license, car registration, and insurance to drive. In a narrow framing, that is a restriction of privacy and freedom. Yet these restrictions are necessary because cars are dangerous. We want to ensure that cars are used responsibly so therefore the privilege of driving requires these restrictions, including documentation of ownership and active driver certification. One can always walk or take the bus if they find these restrictions overly invasive.
Similarly, the modern finance system is incredibly powerful. Anyone can move arbitrarily large amounts of money with trivial effort, yet solid guarantees of security and fidelity. Even at a small scale, electronic funds transfers can greatly facilitate crimes. E.g., selling child pornography over the internet with electronic payments. Therefore we have instituted numerous regulations and restrictions around banking, including KYC/AML. One can always use cash if they find these restrictions overly invasive.
Regulations also include strong protections against misuse of KYC/AML information by either financial institutions or the government. That includes narrow access controls and permanent audit records of all access as well as criminal penalties for misuse. This is similar to how a police officer can be charged with a crime if they attempt to misuse DMV records. Our government is therefore guarding the privacy of individuals who comply with its regulations because that is recognized as important.
That analogy is one way to look at things, but I don't think it quite fits. The roads are arguably government owned infrastructure so it has a pretty clear justification for making rules about their use, but the financial system is really more of a network of agreements and communication channels between private banks. As you said, if you don't want government intrusion into your life, you can still take the bus to travel long distances on the roads. It's not clear what the equivalent outlet is for financial transactions. Cash is too local to fit the bill.
Perhaps a better analogy would be the internet. It too is very powerful and useful, but also dangerous. Still, so far we have avoided doing things like banning end to end encryption and VPNs and TOR or requiring anyone providing internet access from ISPs to coffee shops to KYC. There are people pushing for these things, and such things are already the norm in China. Despite not having that level of regulation in the US, I don't think the internet has brought us to an era of unprecedented crime and danger. Every few weeks there is a big story about the government arresting cyber criminals or shutting down a black market, so internet criminals are being arrested with our current tools. But if we move in the direction of a heavily regulated internet, I guarantee you that when I make the point in 50 years that maybe end to end encryption and anonymous internet usage is okay actually, I'm going to be pounced on by people outraged that I'm suggesting something that "could greatly facilitate crimes".
There is a balance to be struck between liberty and safety, but when a government can cut protestors or strikers off from the financial system, you know that you've gone too far. The vaunted "protections" that should exist in a developed nation's AML/KYC system didn't do much to protect the Canadian protestors from being targeted earlier this year. Governments shouldn't have those tools. We shouldn't be giving it the authority to do such things, but in case that fails, we also shouldn't be giving it tons of information that can help it piece together which accounts it wants to target. That's AML/KYC.
I’m not much of a political theorist, so I don’t care a lot about the justification. I’ll only add that the alternatives to the current AML regime are much more intrusive than what we have. For example, right now lots of activity that is kind of sketchy, such as depositing $100,000 in bulk cash at a bank branch, triggers requirements that banks start knowing something about the source of funds and a suspicious activity report. Obviously, this regulation prevents laundering money from business that generate bulk cash like drug cartels. Some legitimate businesses generate that much cash, but not many.
One alternative would be to ban depositing bulk cash over a certain amount entirely, or to require permission from a government regulator that requires that you prove where the money came from. That’s basically what happens when you try to bring cash through customs. It’s intrusive as hell. Better let the intermediary bank do a risk based assessment and let a person deposit their money in the mean time. The bank only has to outright refuse to do business if it knows for sure that the source of the money is crime.
The problem with crypto is that there often isn’t an intermediary. I can transfer infinite amounts of crypto from A to B without an institution subject to AML regulation being involved. This isn’t a huge problem right now for a lot of reasons, but mostly because pretty much no one buys normal goods and services with crypto, so a criminal has to eventually go to a real bank and exchange it for real currency. Banks are for now required to treat those transactions as bulk cash, but the crypto industry doesn’t like that. If crypto were used to buy things, criminals could just live their life paying in crypto. That’s real bad and would likely prompt very intrusive regulation like banning merchants taking crypto over Z dollars.
Once again, that can't possibly be right. The US was not suffering from massive corruption, tax evasion and crime on an unimaginable scale before the KYC/AML system was set up in 1970.
I'm not going to bother educating you here. You're wrong. Do your research. If you think that a libertarian vision of civil liberties is worth the cost of rampant crime, that's you're decision. But don't confuse political theory with policy analysis.
Our old reports at Global Financial Integrity were pretty good. I’ve never seen any real explainers on the Bank Secrecy Act, but that’s what I would Google. I learned the details from the people around me, but that was my job. Maybe go see if you can find some old Carl Levin hearings on money laundering.
243 comments and no one has mentioned Gresham's Law? ("Bad money drives out good.") That's not the only reason so few people use btc as normal money, but you can glen interesting insights by thinking that through.
Can anyone offer an explanation about the policy/impact differences between restricting capital inflows vs outflows? I’ve realized I’ve sort of digested the leftish take that outflow restrictions are important for restricting capital flight, but haven’t really thought that much about it. Why would you want to restrict inflows vs outflows?
Restricting inflows can help you control your economy - like controlling who you allow to invest in your company. Restricting outflows can be useful, too, in theory, but in practice it almost always just politicians covering up poor management - like not allowing investors to sell out when they no longer want to be invested in your company.
I'd point out that Tabarrok's problem doesn't exist in country's with far less wealth than ours that have actually invested a bit in their financial systems. Places like Brazil have instant transactions. Blame the federal reserve or the treasury department that our money moving system is so clunky, but it's a solved problem globally and crypto fixing it in the USA (where people are among the least worried about instant deposits) seems absurdly niche.
I have been employed full-time at a blockchain/cryptocurrency startup as a researcher/engineer for 18 months, moving from a job at a bank after I discovered (thanks to the pandemic) that I didn’t mind remote work. I have also been a big fan of Matt’s writing since his Slate days, and politically find myself very much in alignment with him (this is not a coincidence, but a measure of his influence on my thinking). I thus think I am pretty well-positioned to provide a counterpoint. I am going to give my perspective on the reasons to be optimistic about blockchain/cryptocurrency in this post, and respond to Matt’s points in a second post.
The concise and high-level reason I’m optimistic about blockchain finding many use cases, is that blockchains allow the transfer of value over the internet. I will try to explain, but I want to emphasize that I’m as skeptical as many critics about many of today’s blockchain/cryptocurrency projects; in the long run, though, I’m very optimistic.
I will first briefly describe what a blockchain is, before going into a near-term use case I could see for a blockchain. After that, I’ll step back and argue for the assertion above. I should note up-front that I am much less optimistic in the specific use-case I discuss than I am that blockchains will find some use-case, I just think it’s helpful to go through the possible benefits using a concrete example.
(I should preface this with a disclaimer that my background/work is as a mathematician, and to a lesser extent on the modeling/economics side; my expertise is not actually in blockchain engineering per se. I may very well get some details wrong below, but I am pretty confident that my high-level explanation is correct.)
Blockchains are protocols which allow a decentralized network of computers to come to consensus on a state. You can think of the state as a spreadsheet. On the Bitcoin blockchain, the spreadsheet just records which accounts have how much Bitcoin. On Ethereum, the spreadsheet holds numbers which actually define the state of the “Ethereum Virtual Machine”, which we can think of as a decentralized computer. This is possible because everything stored on a computer – including the code that makes up an application, including the operating system itself – is ultimately stored as numbers in addresses. By coming to a consensus on what numbers are stored in which addresses, the Ethereum network can come to a consensus on the state of the Ethereum Virtual Machine. Although Bitcoin was the technological innovation that kicked it all off, my optimism entirely derives from what Ethereum shows is possible.
All blockchains I’m familiar with require a cryptocurrency to function, because the protocols must incentivize decentralized networks of participants via systems of rewards and punishments. The cryptocurrency must be desirable for the participants (for the incentives to be meaningful) and attainable for new participants (if you want the network to be able to grow, effectively a requirement for decentralization). Though this does not automatically mean the cryptocurrency must be possible to exchange for money, but in reality this is difficult to avoid. For most newer blockchains (if they use proof-of-stake), it must also be very difficult for anyone to accrue too high a percentage of the cryptocurrency, meaning that the more money a blockchain’s cryptocurrency is worth, the more secure the network. But while cryptocurrencies are necessary for the operations of blockchains, the most exciting potential lies not in the cryptocurrencies themselves but in the blockchains they enable.
The most interesting near-term use-case, in my view, is the surplus of recent examples in which software companies develop network products with the intent of then sitting atop those networks extracting rent.
Take Doordash, for example. Doordash is an app which operates a 3-sided market of restaurants, delivery drivers, and customers. Doordash, obviously, is not a monopoly (I have no idea if it’s even the largest food delivery app), and it’s easy for all participants in the network to use multiple apps. In theory, this should mean that the food delivery apps are competing to give the best experience and pass on the most profit to delivery drivers and restaurants, and to give the best experience to users.
It’s not clear that this happens in practice, although I’ll confess I don’t know much about the details. Local governments (like mine in Minneapolis) have found it necessary to pass regulations preventing some sketchy practices by these delivery apps, to which delivery apps have responded with punitive fees which they tell customers are direct results of decisions made by their local governments.
But what if the restaurants and delivery drivers had direct *decentralized* control over the network, and dictated to the software developers how the app should work? This could be a compelling use-case for a blockchain. Restaurants/drivers would earn the network currency by participating in the network (by making/delivering food). Holders of the network cryptocurrency (primarily or exclusively restaurants/delivery drivers) would then get a say in decisions made about the network. Note that although the money transferred from customers to drivers and restaurants would likely be recorded on this blockchain as evidence of participation in the network, the use of the blockchain here isn’t actually about getting any of the actors involved out of the banking system or anything like that. It’s about changing the economics and power dynamics of a network product.
To dig into one particular detail which I think may be helpful, note that a blockchain in this application would *still* have “middlemen” (this is somewhat a heterodox framing of mine, that most crypto proponents would disagree with, but hear me out). The centralized middleman (Doordash) would be gone, but the computers that would be taking transactions and using them to update the decentralized network state would each be under the control of someone. Perhaps the restaurants and delivery drivers would be running these computers, but more likely they would be run by specialists (this is what we have seen in similar existing blockchain networks). These specialists can be seen as the middlemen in the *new* system. However while existing middlemen like Doordash can be restricted only by regulation, the new middlemen who specialize in running the blockchain can have the scope of their actions restricted *by code*. This opens up a huge design space which we as blockchain researchers are only beginning to explore.
The technical details of the specific food delivery use-case are not really the point, and certainly there are arguments for why this might not work. But it seems very, very premature to discard the idea that blockchain *might* work in some similar use case.
(The U.S. Dollar is also a network product, and many arguments about crypto center on whether it can or should disrupt the Dollar. But I believe that both critics and proponents of crypto who focus on this particular network product are thinking too small.)
To take a step back from particular use-cases, my basic reason for optimism is that blockchains allow the transfer of value over the internet. This isn’t about believing that the dollar value of this or that cryptocurrency won’t go to zero. The internet allows frictionless transfer of information, but cannot provide any guarantees about scarcity, as it relies on this information being trivial to duplicate. By allowing the possibility of decentralized consensus on state, blockchains allow for credible scarcity on the internet, and thus for the transfer of value.
The internet has transformed the world, I think I can claim without much controversy. The ability to transfer information instantly and frictionlessly around the world has found use cases that were unimaginable to the technological innovators who set us on this path, much less anyone who thought such innovations would lead to *no* real use cases. Given this recent history, I have to believe that the ability to transfer value in addition to information will be a very big deal, the “Value Superhighway” rather than the “Segway of Finance”.
*Cisco & UN have been working for awhile to make cheap cell phones accessible in nations with little banking structure to allow mobile banking over sock-money. Crypto is only an extension of that program, as folks wouldn’t have access to crypto without the cell phones in the first place. It’s really not the solution itself, just an extra layer of computation put on top of the UN solution.
Evading-bad-policies has a little more credence but is becoming moot in the US since that vast majority of retailers are using digital financial exchanges or crypto-exchanges that are monitored. I am rather salty about it because I had my preferred name on Venmo when it started requiring ID verification for transaction monitoring, and my account got permanently locked. Staff claims there is no way to change account names. Seems odd considering there are, if I recall correctly, >15 million women in the US whom have changed their legal names.
Pardon, keeping with the analogy, mobile banking is the road, crypto is merely a segway on the road.
I think there is a typo, the proposal is likely for "long-unused mineral rights" to *revert* -- not "refer" -- to landowners.
If there is better way to share notes like this, rather than seeing them buried in the comments, please advise. Thanks.
In all the countless times I have watched Die Hard, I have never thought once until this moment about what on earth Hans Gruber was stealing! And Rickman lingers on the pronunciation of "bearer bonds" in such a memorable way, odd that it never clicked. I just accepted it was a valuable thing to steal and moved on.
You're aside regarding Alex Tabbarok is a nice little reminder that as much as he does write some interesting Marginal Revolution posts (and op-eds) and is a smart guy that I find myself sometimes agreeing with. He's at his heart a crank. Like "end the Fed", full blow Austrian economics crank.
Revealing couple pieces to me. Especially the NYtimes op-ed (he's not 100% wrong, long term investment in real estate was a good bet. But yeah he's basically Larry Kudlow 2.0 here).
https://www.nytimes.com/2008/03/18/opinion/18tabarrok.html
https://marginalrevolution.com/marginalrevolution/2010/11/has-the-fed-been-a-failure.html
"I’m actually pretty bullish on Bitcoin partially displacing gold as a paranoiac’s hedge over time."
This has already struck me as hilarious. Society is falling into total chaos or Big Brother will totally take over and control your every waking minute but don't worry, because there will always be a working and totally independent Internet.
These paranoids are such trusting babes in the woods.
Perfect timing! Today's news in NFTs, “Donald Trump’s Big Announcement: He’s Selling $99 Digital Trading Cards in His Likeness”, https://www.yahoo.com/entertainment/donald-trump-big-announcement-selling-161804314.html
> Former President Donald Trump on Thursday revealed a new foray into NFT sales, sharing a website that offers a series of what he called “digital trading cards” for $99 each.
> On a website called CollectTrumpCards.com, the 45th president is selling a series of artworks featuring likenesses of the 45th president depicted as a superhero with laser eyes, an astronaut, a John Wayne-like cowboy and, of course, a dark-suited Trump surrounded by gold bars and one of him golfing.
> But Trump’s launch comes at a time when the NFT market has stagnated, with prices dropping 97% since the start of the year.
> “Remember, Christmas is coming, and this makes a great Christmas gift,” the 45th president said in a video accompanying the launch.
God, this is somehow much more pathetic than I expected when I heard he had a "major announcement" for today.
"while crypto is technologically impressive"
Is this true? Not a techno nerd myself so I have no opinion but I'd appreciate the smart SB folks weighing in. Crypto is certainly impressive in the vast amounts of computer power (and planet-killing energy requirements) it consumes, but is it impressive technology? And if so, why? Say, compared to AI, VR, mRNA vaccines, fusion power, current chip technology, and on and on.
Most newer blockchains do not use proof-of-work, and therefore do not have energy consumption downsides you refer to; these days pretty much only Bitcoin uses proof-of-work (Ethereum recently switched away from it).
Blockchain tech is only impressive at the intersection of computer science and social coordination. There are plenty of very good computer scientists who don't get it because the problems it attempts to solve aren't purely computer science problems, they are also social coordination problems. There are plenty of people who focus on social issues who don't get it because they see crypto/blockchain enthusiasts with whom they have political disagreements, and they fail to understand that the underlying technological/computer science innovation isn't about any one particular political persuasion but in fact can be applied to different social coordination problems in different ways. So there are a lot of legitimate experts who don't get it.
This also makes it hard to say it's "technologically" impressive specifically. But it's an interesting computer science solution to problems of social coordination, and my own guess is it's not a question of whether we'll use it, but how.
Bitcoin is still a pretty big deal but sure we seem to be moving towards a day when the most evil part of crypto is in the rearview mirror and its contribution to heating the planet, while locked on, is no longer moving apace.
As for it being impressive because of its combining computer science and social coordination, I read that to mean it's still all vaporware -- certainly not an argument for it being impressive in any form. I find the idea that (mostly libertarian) computer scientists being the ones whose vision of how to improve social coordination rib-tickling when it's not downright scary.
I'm not a fan of Bitcoin personally, because I think the libertarian/monetary policy argument is a bit silly, and I also don't buy the idea that proof-of-work (which is the source of the energy usage) is better than proof-of-stake, but I will very slightly defend it by saying: We each believe that the things we think are important are worth using energy for. Bitcoin uses electricity, which can easily be generated in clean ways. So to me, arguments about the energy usage of Bitcoin are pretty much beside the point.
As a progressive who works for a blockchain startup, I can tell you that it's not just -- or even mostly -- libertarian computer scientists. You don't have to be impressed by anything, of course, I suspect that someday the apps you use will simply switch to using blockchains on the back-end without you knowing, and you won't have to worry about it :-)
It's the visionaries, residing safe and happy in their liberal democratic nations, believing that the blockchain will liberate us from the suffocating hand of government who would worry me if I thought there was much to this.
But if it turns out that blockchain technology does some useful technical things in the background that common people like me will unknowingly benefit from, then great. Like html, TCP/IP, and all the other million things that are the backbone of our world that attract no attention, don't make billionaires out of speculators (if not worse) etc. Perhaps the first rule of blockchain success is that no one needs to talk or think about blockchain.
"Perhaps the first rule of blockchain success is that no one needs to talk or think about blockchain." - Most of us actually building in the industry basically agree with this (as opposed to people using crypto to gamble, who do very little except talk about crypto, which then unfortunately means that most of the general population associate crypto only with them).
A cryptocurrency is *required* for a blockchain to operate, but many users who will benefit from it won't need to worry about it at all, I hope. For example, one possible use-case would be something like a version of Doordash that is owned not by the VCs who funded the app, but by the restaurants and drivers themselves. The restaurants and drivers would need to be aware on some level that the network was operating as a blockchain, only because the power and responsibility for making decisions would belong to them. But the customers would have no reason to care or know about those details, or deal with any cryptocurrency at all.
A core principle of nerds is that we get annoyed when normies discover something that we're into. Usually they just suck all the fun out of it and ruin it for everyone. But this time I got a free car out of the deal.
Bitcoin was a fun way to put to use the pile of old graphics cards and motherboards I had in my closet. All the core developers hung out in one place and were fun to talk to about cryptography and computer science stuff. Getting their code to compile and optimizing hash rates was fun. Downloading the blockchain was fun. I didn't even know or care what cryptocurrency was supposed to be for, it was just a fun thing for a computer nerd to do on weekends, like ripping DVDs you have no interest in watching or hacking Dish Network cards when you don't even have a satellite dish. And there were some weirdos out there who would trade you something of value if you figured out how to send them some of your Bitcoin.
Then, suddenly, you could trade Bitcoin for money, which was totally weird but also a cool way to squeeze some value out of old computer hardware (and free electricity at work). Fast-forward a decade and I cashed out for enough to buy a car outright and cover all the taxes. Pretty neat! And I still use Bitcoin for transactions in foreign currencies and to pay for various online services that are a pain or whose purpose is defeated by involving a middleman that has to Know Your Customer, like VPNs or server instances.
I have no point to make, I just thought I'd share my experience with crypto since Matt was kind enough to point out that Bitcoin is to FTX what BBSes are to Twitter. Most people who write about crypto focus exclusively on the speculation and facilitation of crime.
I've had a lot of conversations with various levels of laser eyed people "on here" and in real life. They seemed to never have a good answer to the life changing use case question, the answers they gave vs the commitment in time and money ratio never lined up. "Oh, cool, you're worried about instant settlement of large international transactions - Is this a problem you have frequently in your life as a normie data analyst?" "You were just mocking African countries, and haven't shown any interest in the welfare of people outside your own family, am I to believe you're really excited about Nigerians using BTC to get better prices on imported goods, and that's why you spend so much time and money on this technology?" It was all just weird.
Yeah, this is a lot of people's experience and I think it's because fundamentally blockchain/crypto is still at a very early stage and a lot of different people in the industry have a lot of different use cases in mind (source: I work in a blockchain startup). It doesn't help that a whole lot of people who like crypto like it because they like gambling, rather than because they are interested in the actual tech, and they tend to be noisier than those of us who are building/researching the tech.
I have been surprised by the lack of pro-crypto takes in these replies, and I tried to give my version of that in a comment directly on the post (I'm not familiar enough with Substack to figure out how to link to my comment), which I'll try to verrry briefly summarize...
Basically, in the near run I think blockchain might be an appealing technology for products with network effects where having a single owner/middleman in the network is not ideal. Currencies are network products and you'll hear a lot of crypto enthusiasts who want to e.g. replace the U.S. Dollar, but this isn't the most obvious use-case to me. I'm much more interested in e.g. building a food delivery app that is owned and controlled by the restaurants and delivery drivers instead of owned by some centralized software company who, once they attain network dominance, can just extract rents.
In the long run, it's seems clear to me that while the internet allows frictionless transfer of information around the world, blockchain technology allows the transfer of *value* which is a really big deal.
EDIT: ah, figured out how to link: https://www.slowboring.com/p/cryptocurrency-is-the-segway-of-finance/comment/11179445
Excellent piece on an important topic, thank you.
Your aside on the early Web is a bit misleading. It was invented as an iteration of scholarship tools, then adopted for business use. Recommend Berners-Lee's Weaving the Web. My summary is here: http://whatarecomputersfor.net/the-worlds-knowledge/
Where did you see Punisher: War Zone? I'm hoping you saw it in the US the day before you flew to Berlin through Copenhagen, and not that you spent precious Copenhagen or Berlin time watching Punisher: War Zone.
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.
They did manage to catch and prosecute criminals before the KYC/AML system was created. It's never been clear to me why we tolerate so much financial surveillance and reporting in ways we would never tolerate analogous physical surveillance. The system just seems like a way to circumvent 4th amendment protections to me.
Then again, physical surveillance is getting much more powerful these days too and is generating minimal outrage. Still, as far as I know, most places operating security cameras aren't sending automatically generated mandatory reports to the government about the comings and goings of all people on their premises including gait and facial recognition data.
I don't have much to say other than you are wrong about the importance KYC/AML system. I mostly worked on policy in countries that didn't have an effective AML system. The alternative is financial anarchy, massive corruption, tax evasion and crime on a scale unimaginable in the U.S. If you think Nigeria is a good place to head toward, then kill the KYC/AML system. There's a reason why every developed country in the world has some version of it.
Just one example of what banks do when KYC/AML isn't enforced:
https://www.justice.gov/opa/pr/hsbc-holdings-plc-and-hsbc-bank-usa-na-admit-anti-money-laundering-and-sanctions-violations
As another KYC/AML advocate, I just want to chime in with supportive analogy.
You need a driver's license, car registration, and insurance to drive. In a narrow framing, that is a restriction of privacy and freedom. Yet these restrictions are necessary because cars are dangerous. We want to ensure that cars are used responsibly so therefore the privilege of driving requires these restrictions, including documentation of ownership and active driver certification. One can always walk or take the bus if they find these restrictions overly invasive.
Similarly, the modern finance system is incredibly powerful. Anyone can move arbitrarily large amounts of money with trivial effort, yet solid guarantees of security and fidelity. Even at a small scale, electronic funds transfers can greatly facilitate crimes. E.g., selling child pornography over the internet with electronic payments. Therefore we have instituted numerous regulations and restrictions around banking, including KYC/AML. One can always use cash if they find these restrictions overly invasive.
Regulations also include strong protections against misuse of KYC/AML information by either financial institutions or the government. That includes narrow access controls and permanent audit records of all access as well as criminal penalties for misuse. This is similar to how a police officer can be charged with a crime if they attempt to misuse DMV records. Our government is therefore guarding the privacy of individuals who comply with its regulations because that is recognized as important.
That analogy is one way to look at things, but I don't think it quite fits. The roads are arguably government owned infrastructure so it has a pretty clear justification for making rules about their use, but the financial system is really more of a network of agreements and communication channels between private banks. As you said, if you don't want government intrusion into your life, you can still take the bus to travel long distances on the roads. It's not clear what the equivalent outlet is for financial transactions. Cash is too local to fit the bill.
Perhaps a better analogy would be the internet. It too is very powerful and useful, but also dangerous. Still, so far we have avoided doing things like banning end to end encryption and VPNs and TOR or requiring anyone providing internet access from ISPs to coffee shops to KYC. There are people pushing for these things, and such things are already the norm in China. Despite not having that level of regulation in the US, I don't think the internet has brought us to an era of unprecedented crime and danger. Every few weeks there is a big story about the government arresting cyber criminals or shutting down a black market, so internet criminals are being arrested with our current tools. But if we move in the direction of a heavily regulated internet, I guarantee you that when I make the point in 50 years that maybe end to end encryption and anonymous internet usage is okay actually, I'm going to be pounced on by people outraged that I'm suggesting something that "could greatly facilitate crimes".
There is a balance to be struck between liberty and safety, but when a government can cut protestors or strikers off from the financial system, you know that you've gone too far. The vaunted "protections" that should exist in a developed nation's AML/KYC system didn't do much to protect the Canadian protestors from being targeted earlier this year. Governments shouldn't have those tools. We shouldn't be giving it the authority to do such things, but in case that fails, we also shouldn't be giving it tons of information that can help it piece together which accounts it wants to target. That's AML/KYC.
I’m not much of a political theorist, so I don’t care a lot about the justification. I’ll only add that the alternatives to the current AML regime are much more intrusive than what we have. For example, right now lots of activity that is kind of sketchy, such as depositing $100,000 in bulk cash at a bank branch, triggers requirements that banks start knowing something about the source of funds and a suspicious activity report. Obviously, this regulation prevents laundering money from business that generate bulk cash like drug cartels. Some legitimate businesses generate that much cash, but not many.
One alternative would be to ban depositing bulk cash over a certain amount entirely, or to require permission from a government regulator that requires that you prove where the money came from. That’s basically what happens when you try to bring cash through customs. It’s intrusive as hell. Better let the intermediary bank do a risk based assessment and let a person deposit their money in the mean time. The bank only has to outright refuse to do business if it knows for sure that the source of the money is crime.
The problem with crypto is that there often isn’t an intermediary. I can transfer infinite amounts of crypto from A to B without an institution subject to AML regulation being involved. This isn’t a huge problem right now for a lot of reasons, but mostly because pretty much no one buys normal goods and services with crypto, so a criminal has to eventually go to a real bank and exchange it for real currency. Banks are for now required to treat those transactions as bulk cash, but the crypto industry doesn’t like that. If crypto were used to buy things, criminals could just live their life paying in crypto. That’s real bad and would likely prompt very intrusive regulation like banning merchants taking crypto over Z dollars.
Once again, that can't possibly be right. The US was not suffering from massive corruption, tax evasion and crime on an unimaginable scale before the KYC/AML system was set up in 1970.
I'm not going to bother educating you here. You're wrong. Do your research. If you think that a libertarian vision of civil liberties is worth the cost of rampant crime, that's you're decision. But don't confuse political theory with policy analysis.
Anyone have suggestions on how to learn more about the impact of KYC/AML systems?
Our old reports at Global Financial Integrity were pretty good. I’ve never seen any real explainers on the Bank Secrecy Act, but that’s what I would Google. I learned the details from the people around me, but that was my job. Maybe go see if you can find some old Carl Levin hearings on money laundering.