On New Years Eve, two people were arrested for stealing soap from the Walgreens in my neighborhood. If you think that’s a pretty minor crime, consider this — one of the perpetrators was the exact same height as the Unabomber!
Now wait a minute, you might say. Height is not a measure of criminal impact. If that strikes you as obvious, you are smarter than the readership of the fellow who calls him “Digiconomist” and tweets stuff like this.
Now, if you chortled at the notion of measuring the serious of a crime in units of feet and inches, I trust you are chortling doubly at the notion of measuring economic impact in units of terawatt-hours. Bitcoin miners tend to gravitate (either physically or via the cloud) to where power is cheapest. Argentinians tend to use whatever power is available in Argentina. There’s no reason to think that the cost of a Terawatt-hour generated by a hydroelectric dam in Turkey is comparable to the cost of a Terawatt-hour generated by burning fossil fuels in Buenos Aires.
So why does Digiconomist report this kind of claptrap? Maybe he’s a smart guy who doesn’t always think before he posts. Maybe he’s an idiot. Or — and this is where I’d put my money if I had to — he figures his readers are idiots who are easily dazzled by shiny nonsense.
134 Terawatt Hours is a lot of power. But with not even a guess as to the (internal and external) costs of generating that power, we are left with absolutely no basis for even beginning to think about whether those costs are acceptable. And the comparison to Argentina, where costs are likely to be very different, does not help even a whit. Digiconomist appears to hope his readers won’t notice that.
I’m going on a bit about this because I happen to have just come from a conversation with a pack of idiots of just the sort who I’m sure the digiconomist finds useful. The tweet above happened to come across our desks; I pointed out that it was pretty stupid, and the general reaction was that this just goes to show that economists are living in some crazy fantasyland, where they deny obvious truths like “Everything anybody cares about can be measured in Terawatt Hours”.
Suddenly we were back in first grade where I was explaining that the reason we care about energy use is because it consumes resources and therefore denies us other things, like haircuts and tractors and fresh produce and clean air — and that if you want to think about whether the tradeoffs are worth it, you have to measure everything in a common unit, of which the most readily available is the dollar. That was taken as more evidence of how out of touch economists are. Nobody even suggested an alternative way to think about tradeoffs, having decided, apparently, that thought serves no purpose.
Usually I’m pretty good at ignoring this kind of stuff, but this one really pissed me off. Possibly that was because I was in kind of a bad mood to begin with. But I think it was also this: The idiots in this pack are not full-time idiots. They’re professional people who do their jobs well. I’m sure that if I had to take over any of their jobs, I’d be laughably incompetent. Surely they know how to process information. Surely they’re capable of spotting an obvious fallacy, or an attempt to pull the wool over their eyes. But given an opportunity to hoot and jeer at a simple and obvious point that they happened not to think of themselves, they shut down all of their critical thinking skills in order to grab that opportunity. That’s partly a reflection of how human beings are wired, and partly, I think, a reflection of the times we’re living in, where tribalism and mockery seem to perpetually trump reason and thought. And what it means is that in these times, even many of the best of us have chosen, at least intermittently, to join the pack of idiots. It’s pretty scary. In fact, I’d say it’s even scarier than the number of animals in the National Zoo.
Lol. You do have a way with words (and logic – a powerful combination, which is why i keep checking in here)
And happy new year to you, too.
You raise fair points. I acknowledge shortcomings with various economic conventions, including the convention of trying to reduce all kinds of disparate things to dollars for purposes of comparison. But I also acknowledge the (typically larger) shortcoming of NOT doing so. Many econ conventions seem to offer second-best solutions; many critics of econ conventions offer NO solutions.
That said, I share some frustration with the idea of bitcoin mining. Historically, people got rich by working hard to do something USEFUL TO OTHERS. Cryptocurrency management seems to bypass this tradition entirely.
As far as I understand it (which isn’t far), bitcoin mining is an entirely arbitrary, man-made process for managing the expansion and allocation of bitcoin currency. Since it’s arbitrary, it would be nice to imagine some different method for managing this currency that didn’t consume so many resources–or, at least, that had more salutary effects than just burning electricity. Allocate bitcoins to people who pack additional capacity onto a silicone chip, maybe? Or cure the next disease on a list of most lethal diseases produced by the UN? Or something like that. So the cost people complain about may be the foregone opportunity of harnessing all this effort for some useful purpose.
While it’s hard to know for sure, at least one source concludes that most bitcoin mining occurs in China. While China has a lot of hydropower, it also burns a lot of coal (imported from Australia). And since China is experiencing electricity shortages, it’s not crazy to imagine that the marginal watt-hour is being generated by coal, not hydro–that is, that bitcoin mining is likely contributing to climate change. Of course, China is huge, so it’s hard to say anything about the locational marginal cost of electricity at any given place, or the associated CO2 emissions.
Finally, what significance should we give to the comment that bitcoin mining consumed as much electricity as Argentina in specific? I interpreted it like the sentence “Covid has killed five million people–more than the population of Alabama.” It’s not really a statement about Alabama; it’s just an effort to provide a sense of proportion to an abstract statistic. The concept of terawatts is not very meaningful to me; the idea of the electricity consumption of Argentina is slightly more so.
Re: “But with not even a guess as to the (internal and external) costs of generating that power, we are left with absolutely no basis for even beginning to think about whether those costs are acceptable.”
Actually, we do have a basis — however much bitcoin miners spend on energy to power their computers, it’s worth it to them in their pursuit of profit, and they’re spending their own money, so it’s acceptable.
Daniel R. Grayson,
However much energy producers spend of their own money on mining coal to use to produce power is worth it to them in their pursuit of profit, but it doesn’t follow that it’s acceptable. Their private internal costs may be less than the external costs to society, such as more diseases from pollution or contributing to global warming, which wouldn’t be acceptable.
nobody.really (#2):
That said, I share some frustration with the idea of bitcoin mining. Historically, people got rich by working hard to do something USEFUL TO OTHERS. Cryptocurrency management seems to bypass this tradition entirely.
Bitcoin mining is certainly useful to others; the question is whether it is useful enough to justify its costs. One could and should ask this both in total and at the margin. At the margin, I suspect, but do not know, that mining is way over-rewarded. Of course the same is true of many things — writing books, for example.
Since it’s arbitrary, it would be nice to imagine some different method for managing this currency that didn’t consume so many resources–or, at least, that had more salutary effects than just burning electricity. Allocate bitcoins to people who pack additional capacity onto a silicone chip, maybe?
This would obviously be awesome, and I think I had a blogpost a decade or more ago speculating about the same thing. But what’s lacking, of course, is a concrete proposal. You need a problem that is parameterized by the contents of the current block and the hash of the last block. “How do you add capacity to a silicon chip?” does not likely have a variation that meets that criterion.
I am pretty close to sure that only purely mathematical problems fit the bill, which means you maybe want miners to be looking for near-counterexamples to the ABC conjecture, for example. But it’s very hard for me to see how you’d turn that vague notion into a specific task that would serve the purposes of crypto.
Another option is to abandon proof-of-work entirely but as far as I understand the state of affairs, there is nothing else on the horizon that promises as much security.
The way to reduce the costs of Bitcoin mining is to tax Bitcoin mining, which can be done entirely within the system with smaller (or even negative) block rewards. To get costs down to a point where they’re socially justified might be impossible (without sacrificing so much security that there’s no point in going on), or it might be entirely feasible. I have no idea how many miners are needed to keep the system acceptably secure. Probably someone has thought hard about this, but I’m not sure who.
Z,
As far as the miners’ contribution to global warming is concerned, it’s the government’s job to tax the production of CO2 appropriately, and there’s nothing I can do to convince the government to do it. So I find it acceptable that miners consider only their own actual costs, just as everyone else does.
Oh? Do tell!
nobody.really (#7): Suppose a million people are on the verge of reading “The Great Gatsby”, generating a million dollars in revenue for the heirs of F. Scott Fitzgerald. Now I come along and write a book which is just the teeniest bit better than The Great Gatsby, so all those people write my book instead. I’ve just earned a million dollars as a reward for an extremely tiny increase in the enjoyment those people will get from reading.
If I believe I can write that ever-so-slightly better book, I will definitely do that and take the million dollars, whereas I might have been a lot more socially useful driving an Uber.
Daniel R. Grayson,
It’s the police’s job to arrest and deter extortionists, and there’s nothing you can do to convince them to do it. Do you find it acceptable that extortionists do what they do if they find the benefits outweigh their costs?
“The idiots in this pack are not full-time idiots. They’re professional people who do their jobs well.”
Alas, around New Year’s Eve, I was giving an explanation of drug pricing to professional people, well doctors, gathered around my dinner table. My utterances were more about general incentives than anything I know specifically. As time marched on, the reactions to my diatribe became increasingly hostile.
My perceptive daughter, a biologist, who was present, opined to me afterwards that
a) People thought my utterances were freely formed opinions, not anchored in anything; and
b) People did not want to know anything contradicting or questioning their own fiercely held views.
These are doctors! Don’t get sick with a new disease.
@Steve #7:
I don’t agree with the book-writing v uber driving comparison.
Here’s why with same made up numbers.
If I become an uber driver then this will likely lead to a marginal decline in uber fares. Say I give 10,000 rides and my presence in the market reduces the average fare by $1. I have increased other’s utilty by $10,000.
Say I write a book that everyone agrees is marginally better than the Great Gastby (lets say they would only pay 1 millionth of a cent extra to read my my book over TGG). It costs the same to read my book as TGG so everyone reads mine. My book appears to have added only 1 cent in utility.
But assuming most people read multiple books a year then my book would not stop people paying to read both TGG and my book. The real comparison is between my book and the marginal book they don’t read as a result of my book appearing and not my book and TGG. As long as they would pay on average at least 1 cent more for my book than this marginal book then I am providing more utility as a writer than as an uber driver.
Z,
Extortion is evil, but bitcoin mining is not. There’s the difference.
Daniel R. Grayson,
Begging the question.
@Steve #8
Video games are an even better example. Every year, dozens of large-budget video games are published which are fundamentally the same in terms of their mechanics and story – and they are fundamentally the same as the dozens of games in the same genre that were published the year before. It’s very rare for games to offer something truly new or objectively better than the other 100+ (or 500+, depending on the genre) very, very similar games that were published in the last two or three years.
As regards bitcoin and its costs and benefits – if the overall social utility is negative, then any cost is too large.
Bitcoin currently serves no legitimate purposes other than speculation. Insofar as it is used for legitimate payments, traditional means of payment are superior. Tax evasion and black-market payments have negative social utility.
Rob Rawlings (#11):
But assuming most people read multiple books a year then my book would not stop people paying to read both TGG and my book. The real comparison is between my book and the marginal book they don’t read as a result of my book appearing and not my book and TGG.
Yes, this is absolutely true, but I’m not sure it makes much difference. There are millions of books out there, and I’m pretty confident that most people read no more than, say, ten per year. Those are, out of the millions available, the ten books they expect to be the best. Of course they’ll often be mistaken, but at least in expectation, that tenth book is still pretty darn good, and that’s the one that the new book is displacing.
Advo (#14):
Bitcoin currently serves no legitimate purposes other than speculation.
If you wanted to send, say $300,000 to somebody halfway around the world, have him receive it almost immediately, and have him know with certainty within an hour that you’re not going to somehow revoke this payment (so that he’s confident about shipping you your new $300,000 sports car), what payment system would you use?
@Steve (#15):
Would the following be a valid generalization of your point?:
At any point of time there is a huge quantity of “literary capital” in existence that can generate utility to consumers of books at low cost. A newly produced book adds a trivial amount to this stock of “literary capital” and by extension to the utility provided to consumers from this stock. The resources used to produce the new book could almost certainly add greater utility if applied to another area of production.
@Steve #17
Transferwise, probably. Alternatively, Azimo. I guess it works with Western Union as well now (I’ve never used them). Although I got to admit I’ve never sent more than several thousand dollars at once through Transferwise or Azimo and if I suddenly wire 300k it might trip some kind of anti-money laundering/fraud investigation.
I don’t think the lack of such safety mechanisms is a selling point for Bitcoin, though, at least not insofar as law-abiding citizens are concerned.
Advo #18:
Transferwise will cost you over $1000 in fees. Transaction fees for Bitcoin right now are well under $20. And of course irreversibility is every bit as much a safety mechanism for the seller as reversibility is for the buyer. It’s also a safety mechanism for both, against the possibility of Transferwise walking away with your funds.
It’s plain ridiculous to think that there’s no value in being able to bind yourself to a payment. The economics literature is full of examples of situtations where people would be better off if only they could trust each other to keep their promises. Sometimes it’s more important to bind one party and sometimes the other. Bitcoin doesn’t solve every problem, but it solves a lot of problems.
Whether those solutions are worth the social cost is an interesting question. If the answer is yes, then whether we could get the same services cheaper is another interesting question. But to deny that those services have some real value is, I think, just silly.
Rob Rawlings (#17)
At any point of time there is a huge quantity of “literary capital” in existence that can generate utility to consumers of books at low cost. A newly produced book adds a trivial amount to this stock of “literary capital” and by extension to the utility provided to consumers from this stock. The resources used to produce the new book could almost certainly add greater utility if applied to another area of production.
This is maybe a slight exaggeration of my point. I’m not sure I’d go as far as “trivial” and not sure I’d go as far as “almost certainly”, and would prefer to say something like “There’s no particular reason to think that the resources in question are best devoted to book-writing, and my best guess is that they’re probably not”.
@Steve (#20):
Thanks for the clarification !
*Thank you* This has been driving me crazy.
Why do people insist on posting “could power an entire city for a year!” instead of “has a current market value of $X and consumed $Y resources to produce” which could then be compared to, oh, [Visa Electronic Transactions for an Hour], which has a current market value of $A and consumed $B resources to produce.” (Ignoring all other market factors.)
But really, why do we care? Until someone takes the conversation forward to competition and scarcity, it’s just a weird fact hanging out in mid air. Does that use of power cause problems in other market sectors? Is Argentina experiencing power scarcity at point of delivery due to competition between Bitcoin and household demand?
I have a similar example about SpaceX. “The two Solid Rocket Boosters consume 11,000 pounds of fuel per second. That’s two million times the rate at which fuel is burned by the average family car.” (quoted from NASA). That’s great for some sense of consumption speed/scale, but until it is linked to some other impact, such as driving up prices at the pump for the average family car owner, it’s just a nifty-but-useless fact.
@Steve #20
What’s your source for that transaction fee? Also, you still have to convert Bitcoin into a local currency. What’s the cost of that?
And then there’s the empirical question – are people actually using Bitcoin for this kind of transaction? If not, why not?
As I understand it, Transferwise and Azimo are irreversible as well. Azimo provides for cash payouts; those are certainly irreversible.
Advo: You can see transaction fees by clicking on individual transactions at blockchain.com.
As for the notion that Transferwise and Azimo are irreversible, this is an astounding claim. The reason Bitcoin was a significant technological advance is that it solves the previously unsolved problem of making digital transactions irreversible by anyone. If someone has found an alternative solution to that problem, it should have been big news. What is that solution, and why haven’t I heard of it?
Also: Azimo will deliver $300,000 cash within an hour? Call me skeptical.
Now, if you chortled at the notion of measuring the serious of a crime in units of feet and inches, I trust you are chortling doubly at the notion of measuring economic impact in units of terawatt-hours…
134 Terawatt Hours is a lot of power.
(134 Terawatt-hours is a lot of energy. 134 Terawatts is a lot of power.)
Z,
Re: “Begging the question.”
Okay — why is bitcoin mining evil? You tell me.
Bitcoin miners consume something of intrinsic worth – energy – to manufacture something intrinsically worthless – money. For every watt spent, the world is one watt poorer.
Your tirade shot a rabbit and missed the elephant. The real issue is: WHY does Bitcoin exist, wherefore the demand, why would anyone spend real resources in its production?
“What and why is money?” Surely that’s one of the Big Questions. And it goes to the heart of the Bitcoin debate. The world is awash in currencies. The dollar is universally accepted. Why another?
Cast back to the events of 2008, and the fiscal and monetary policies which followed. By coincidence, the Bitcoin paper was published then. What is not coincidence, is the reaction of many, who recoiled at those policies, hence jumped on the Bitcoin wagon.
An exercise for the student: consider the stratospheric rise in Bitcoin, high end art, Tesla, and Gamestop prices. What do these have in common, what’s the underlying driver?
Capital which could have gone to new enterprises, instead flowed into Van Gogh. Because in a funny money environment, money unmoors from reality, the capital markets cannot allocate it judiciously. A vast amount of wealth has been destroyed.
The first act of the first U.S. Congress mandated the death penalty for counterfeiting. Two hundred years ago, we executed gold pluggers. Today, we call them “bankers”, and remunerate them handsomely for banquet speeches. Isn’t progress wonderful?
A tax on Bitcoin mining treats the symptom, not the disease.
Daniel Grayson: “Why is bitcoin mining evil?”
tolerate vs. condone vs. endorse
These are distinct concepts, and the distinctions are important.
“WHY does Bitcoin exist, wherefore the demand, why would anyone spend real resources in its production?”
For most of history money was gold or silver, which also took a lot of energy and effort to obtain.
A rival source suggests that China (among other nations) opposes bitcoin mining–but that Texas welcomes, or will welcome, it.
Why would Texas, a place where energy shortages killed hundreds last February, welcome a big drag on their grid? The premise is that the arrival of bitcoin miners would provide the demand (and consequently the revenues) to justify building more generators–and that these miners could be persuaded to discontinue consuming electricity during emergencies. Ergo you add more supply during most of the year without adding more demand during the crunch periods.
Sounds like a plan to me–assuming that miners really can economically suspend operations on short notice. It’s just computers, right? So is there any reason you can’t power down computers and power them up again later? For example, would the contraction and expansion of electrical components–resulting from the lost of heat when electricity stops flowing, and the resumption of heat when it resumes–cause expensive wear-and-tear to the physical plant or something?
But if this deal would be appealing in Texas, I expect many jurisdictions would be willing to make a similar plan. So Texas might have to offer really big subsidies to lure the miners away from other jurisdictions. And the fact that I haven’t heard about jurisdictions competing for bitcoin miners makes me suspect that I’ve overlooked some big problem with this plan.