The horizon is not so far as we can see, but as far as we can imagine

Bitcoin Is a Bad Way to Do Something Necessary

I don’t write about crypto-currency often because its proponents are fanatical. (You’d be fanatical too if you combined rabid self interest that might make you a multi-millionaire with a social engineering project you thought was utopian.)

But more and more, I am inclined to agree with a judgement my friend made years ago: While Bitcoin does something important (creates a peer-to-peer payment network) it does it in a terrible way.

This averages out to a shocking 215 kilowatt-hours (KWh) of juice used by miners for each Bitcoin transaction (there are currently about 300,000 transactions per day). As the average American household consumes 901 KWh per month, each Bitcoin transfer represents enough energy to run a comfortable house, and everything in it, for nearly a week. On a larger scale, De Vries’ index shows that bitcoin miners worldwide could be using enough electricity to at any given time to power about 2.26 million American homes.

This is crazy. Looked at from this point of view, Bitcoin is terrible: Its actual transaction costs are far higher than those for any other form of money of which I am aware.

Bitcoin was also a libertarian project. Libertarians hate inflationary money, so Bitcoin was made deliberately deflationary. There are a limited amount of bitcoins which can be created. There is a strong first mover advantage even without the fact that bitcoin is acting more like stock in a company than money.

Deflationary money is reactionary. It rewards people for being first, not for being productive. It encourages people not to spend and not to invest in something other than money, which is bad for economies. Moderate inflation, contra-gold bugs and Austrians, is a good thing, as it devalues effort from the past. It’s great that you did something wonderful 40 years ago, but what you do today should matter more.

It shouldn’t matter completely more, I’m not saying that retired people shouldn’t be able to eat and pay rent, but it should discount and discount more and more over time.

People who won the past shouldn’t control the future for all that long. People who are winning the present should, if anyone should.

This piece will likely have people screaming in the comments, but just because Bitcoin is not government money does not mean it gets a pass from general economic principles.

And none of this is to say that blockchains are not an important innovation, or that Bitcoin isn’t important, won’t make its early movers lots of money, and so on. Just that its embedded economic assumptions and power requirements won’t produce maximum welfare.

The results of the work I do, like this article, are free, but food isn’t, so if you value my work, please DONATE or SUBSCRIBE.


Saudi Arabian Crown Consolidates Control as War Looms


China Thinks Strategically and We Don’t


  1. Emma

    Well, hooray! A post I completely agree with.
    This is why I always come back.

    I hate libertarianism more than anything (within reason). It is an already-failed political prospect advanced either by the brainless or the evil. If you take a few minutes to DuckDuck “phossy jaw,” you’ll see what wonderful rewards a lack of regulation in the marketplace will bring you. Phossy jaw and so many other lovely things, like child labor and 16-hour workdays and black lung and typhoid babies. Modern libertarians have all sorts of reasons why the Randian Utpoia of the Future won’t be like the Gilded Age, but they mostly depend on the unfettered exchange of accurate information about products and services among consumers — sorry, I meant “citizens” — which we’re already struggling with in our current onerously-regulated capitalist regime. Libertarians don’t have an answer for “how will citizens be able to sort out good information from bad information when there are no laws designed to hold anyone accountable for anything but contract violations/border disputes and no governing bodies in place to enforce them anyway,” which is why I included ‘brainless’ in with the ‘evil.’ I think most libertarians have visions of begetting incorruptible rebel media empires that will take the place of the social media sites that ban them for TOS violations. Or they’re weird, paranoid gun-humpers who have politically-incorrect opinions about lizard people.

    Sorry about that! I do go on, sometimes. I’m sure I’ll soon have lots of people calling me a fascist and and a lover of Clintons and police states and a stupid woman who just doesn’t understand complex political propositions and etc., but I ain’t care. It will be as nothing compared to the punishment you’ll receive for doubting the omnipotence of the great god Bitcoin :]

  2. John E

    Argh. I knew when this article hit that it was going to be seen and quoted all over the place. It has a very, very, very big flaw.

    What constitutes a ‘Bitcoin transaction’? Well … two things that I can think of. Mining a *new* one. Transferring an *existing* one from wallet to wallet. The former ‘Bitcoin transaction’ is the type that takes a sh*t ton of electricity. The latter is probably more akin to using at ATM machine. But does the author differentiate between the two types of ‘Bitcoin transactions’? Nope.

    His 300k ‘Bitcoin transactions per day’ figure is correct *but* it is definitely referring to the transfer of existing Bitcoins from wallet to wallet. The number of Bitcoins mined per day is way, way smaller than 300k. There’s about 17 million Bitcoins in circulation and the cap is 21 million.

    Completely misleading article headline. The devils in the details.

  3. V. Arnold

    Imo, bitcoin/crypto currencies are a ponzi scheme.
    How can it/they, be a store of value?
    If the internet goes down, no one can access their coins; where’s the value?
    If I hold gold or silver; in my hand; the internet has no sway with me.
    Crypto is at best, just another fiat currency holding no intrinsic value.
    A store of value historically, has been gold or silver; this goes back more than 5,000 years.
    It would benefit those, paying attention, to read David Graeber’s book; Debt; the First 5,000 Years.
    IMO, crypto is a sucker’s bet; a place fools rush to go…

  4. Dan Lynch

    A better, cheaper payments system is a worthy goal, but I would suggest public postal banking is the way to go there. What if Paypal could be replaced with Postal-Pal, linked to your savings account at your local post office and having trivial fees?
    Money’s value is determined by taxes because we all have to have government money to pay taxes. Since government does not accept crypto-currency for payment, the value of crypto-currency is speculative and eventually that bubble will pop.

  5. Ed

    Given how PayPal and the major credit card companies (Visa etc.) will refuse transactions for various adult businesses and products, they are de facto censors and cultural “guardians” (I say that with disgust).

    So there’s a valid demand for a peer-to-peer currency of some sort that can do the long-distance equivalent of a cash transaction. Bitcoin may not be the implementation, but hopefully a decent system will come into existence soon.

  6. Adam Eran

    Good point. Trying to tie money to a commodity, whether gold or electricity, is a non-starter. Mark Blyth says one can either have democracy or gold-backed currency, but not both. (Check out it his comments about Brexit here)

    The problem is that compound interest (which Einstein called “the most powerful force in the universe”) eventually exceeds the ability of the economy to produce enough to repay debts. Deflation worsens that tendency.

    The natural deflationary bias of such currencies is the basis of austerity, which “bravely” crushes widows, orphans and debtors. (The “Fiscal Responsibility[tm]” is always couched in terms of “bravely” making “hard choices”….You know, like when childhood poverty tripled under Thatcher.)

  7. Peter VE

    Bit-and-all-other-crypto-coins rely on a functioning power grid and internet. How did that work out for Puerto Rico? Or for the 150,000 homes in RI without power for several days last week after a moderate autumn storm? At least with fiat money or gold, you always have access to the physical asset. As to whether gold is or isn’t fiat money, that’s an argument for another day.

  8. The Randriod Dystopian is part of the function – people need to be shorn of their savings. Homo economicus 101.

  9. Alex V

    One thing I haven’t understood is what stops people from doing transactions in ever smaller fractions of bitcoins once the 21 million coins is reached. Isn’t there potential for “inflation through that mechanism, by increasing the number of arbitrary holder of value, also essentially without limit? Apologies if my somewhat limited knowledge of the system has missed a mechanism to prevent this….

  10. different clue

    This best-estimate of residential-retail customer use of electricity per unit time which Ian Welsh has brought us is very useful. Hopefully digging around in the website will unearth such estimates for every state, state by state.

    Some retail customers are single individual persons living alone by themselves with nobody else. Other retail customers are part of a whole family full of people, and they are buying the electricity for everyone else in the family as well as their own individual selves. Does anyone have a guesstimate as to how many people the “average household” is supposed to contain? If such a number existed, then we could calculate how much electricity the average single residential individual person uses.

  11. Hugh

    Interestingly, JohnE, essentially concedes most of Ian’s points. Bitcoins have an indefensibly high cost to produce. A cap on their number makes them inherently deflationary and favors first holders. And while it also discourages productive investment, let’s face it everything nowadays discourages productive investment. It is the nature of the beast and the beast is called kleptocracy. Bitcoin also seems inordinately prone to scandals and thefts. I don’t get the attraction.

    Like Dan Lynch, I favor the creation of postal banking, simple, vanilla. That taxes are needed to impart value to currency is an idea that Mosler, the godfather of Modern Monetary Theory, came up with. It is largely false or unimportantly true. Money has value because we, as a society, say it does, and because we find it convenient to use. Ultimately, the value behind money is based on us, our skills, knowledge, and resources.

    In legal terms, in the US, it is not taxes but government spending which imparts value to money, as per Section 4 of the 14th Amendment:

    Section 4. The validity of the public debt of the United States, authorized by law, including debts incurred for payment of pensions and bounties for services in suppressing insurrection or rebellion, shall not be questioned.

  12. Lemonhead

    The last time Ian posted about Bitcoin, it was sitting at around $400.

    So this new article bodes well for a continued bull run. Thank you Ian!

  13. Hugh

    Lemonhead, the essence of a bubble is that there are always a ton of suckers who think the bubble will last forever, and when things go splat and they lose their shirts, forget everyone who tried to warn them, and start whining about hoocoodanode.

    I mean look at stocks whose real value is probably somewhere in the range of a quarter of their current levels. Doesn’t matter. The casino is open, plenty of easy money sloshing about, parteee! But it can’t last because the fundamentals aren’t there, and haven’t been there for years. Keynes said markets could stay irrational longer than he could stay solvent, but that didn’t mean they weren’t going to crash. And they did.

  14. V. Arnold

    Addendum: Re: my first comment about gold and silver being a store of value; historically true.
    I would add, that if one sees gold and silver as an investment; that’s a mistake, IMO. It’s not an investment, it’s a store of value; a very important point to understand.
    The last point is; this is about physical gold, in one’s possesion; not paper held by a second party on a promise of physical gold in a vault somewhere.

  15. Mel

    So what *is* the cost of processing/validating a Bitcoin transaction? I’d believed that mining and validation were part of the same operation, but, thinking about it, I have no good reason to believe that. It’s important to know tht cost.

  16. Hugh

    Gold and silver as a store of wealth: Just think how well off you would be if you were on a desert island with 30 pounds of gold and even more in silver. You would be one really wealthy dead person. Woohoo!

  17. Erik Poupaert

    A first problem is that you can always prove everything by dividing the number of apples by the number of oranges in a system.

    Ok, take one input in the bitcoin network: electricity consumed by mining during the 10 minutes in between two blocks. Then, take another input: new transactions that go into that block. Now, divide all of that, and draw all kinds of ridiculous conclusions.

    If nobody makes new transactions during such 10-minute window, you can even conclude that electricity consumption per transaction is infinite in that time window. Fantastic. You have made a brilliant point!

    So, stop your electric car, but leave the motor running. You are now consuming an infinite amount of electricity per mile. There must be something really wrong with that electric car, don’t you think?

    The second problem is even worse. Let us look at your claim: “People who won the past shouldn’t control the future for all that long.”

    According to Immanuel Kant’s “Critique of Practical Reason”, the “isMoral(behaviour)->yes/no” predicate function can only work properly by axiomatically deriving its rulings from a fixed set of explicitly-stated categorical imperatives. From what axiomatic base exactly do you derive your clearly hypothetical imperative?

    Seriously, why don’t you try to stick to the proper form of morality? An outcome is not moral/immoral. The “isMoral(outcome)” function simply does not exist. Only an intentional behaviour can be moral/immoral.

    Therefore, we can conclude that not just your economics are utmost conjectural, but that your undocumented morality is even more so!

  18. Hugh

    Erik, you have a gift for missing the point. The first question is does bitcoin do anything useful and is it practical. The answer is pretty much no.

    As for why past winners should not control the future. The reason why we used to have things like much shorter copyright periods and very high marginal and estate tax rates was precisely because there was a general awareness that while productive individuals doing socially useful things should be rewarded , this in no way equated to the creation and maintenance of a hereditary, unproductive class of the rich.

    As I have previously pointed out with shorter copyright limits, just because some of the content of a creator might pass out of copyright during their life did not preclude the creator from producing new content which would remain under copyright later into their life. And of course, the creator would still have access to any profits that had accrued while material was under copyright. The purpose of copyright was twofold to give the creator a chance to profit from their work and to then give the rest of society unfettered and free access to that work, to build upon or use it as any and all saw fit. Both society and individuals would benefit and progress from such an arrangement. As we see with the current absurd extensions on copyright that last or effectively last longer than a century, past winners, their children, corporations gain at society’s expense. It is rent seeking. That is an inherently unproductive activity.

    As for morality depending on intentionality, that is deeply counterfactual. Again as I have often written with regard to our elites, they act in bad faith. And it is unimportant what they intend or don’t intend. As elites, they say their wealth, power, and privileges are earned because they know more and better than us. So it becomes not about what they actually intend or know, or say they know or intend, but what they, as elites, should have intended and known. To say they are off the hook just because they did not “intend” that it always just so happens that heads they win, tails we lose doesn’t wash.

  19. Robert

    Emma, what you’re describing is specifically anarcho-capitalism, which doesn’t accurately describe libertarianism as a whole. You’re being intellectually dishonest. Libertarianism runs the gamut between anarcho-capitalism and libertarian socialists. Maybe consider educating yourself on a subject before you begin blinding pontificating.

  20. Jasper

    To clarify on the “mining” vs. “validation”, the two are one and the same. If I want to transfer money from A to B, I broadcast a transaction to the network — where it roughly floats around in a large number of open transactions known as the “mempool”. This is free and not expensive. You can see the number of open transactions here:

    Eventually, a miner, out of the goodness of its heart, will grab some transactions from the mempool, combine them into a “block”, and do an expensive math process that should take roughly 10 minutes. The technical details: the hash of the block has to have a certain number of zeroes at the beginning. The miner adjusts a number in the block, called the nonce, until they find a hash that matches. The number of zeroes varies roughly every two weeks to keep the length of that process roughly 10 minutes long, regardless of how much power has been added. Once you find a correct nonce, anybody can validate that it has the same number of zeroes at the start. The miner broadcasts this block + nonce to the network, and if everybody else agrees it’s a good block, it goes onto the blockchain.

    Hashing a block by itself is cheap, but finding the correct nonce requires a lot of hashing, and a lot of power. As your reward for eventually finding a correct nonce, the rules say that you get to add a bonus transaction which lets you mint some money and put it wherever you want, known as a “coinbase” transaction — this started out as 50BTC and halves every so often, so now it’s 12.5BTC. This is the incentive for using its expensive electricity to do hashing.

    So there’s no such thing as “minting new coin” transactions and “moving money between wallet” transactions and one isn’t more expensive than the other. Transactions go into a block, and that’s the expensive part.

  21. Merasmus

    “There is a strong first mover advantage even without the fact that bitcoin is acting more like stock in a company than money.”

    Because Bitcoin, and all crypto-‘currencies’, aren’t money. They’re an asset you can invest actual money into, if you wish. And sooner or later the fact that they’re a completely fake digital asset that does absolutely nothing other than waste energy means it’ll all come crashing down. And a few early adopters are going to walk away from the rubble, laughing all the way to the bank, now millionaires (in real money).

  22. @John E. You can’t separate the two. Mining is necessary to hash the blocks with transactions. The mined coin is the reward for proof of work, and proof of work is BitCoins consensus model.

    So you actually need the energy to keep the transactions going, and the mining is only a necessary byproduct. It also hides the costs of the transaction, making it seem to be free of charge.

    BitCoin is a very evil plan, and Satoshi Nakamoto has all reasons to hide.

    The concept of peer to peer payment itself is an idea worth to pursue. But the technology and the way coins are allocated need to be vastly different.

  23. ScottA

    Is there a published version of the comment from a friend cited in paragraph 3, or a link to it online? I’d be interested in looking at the basis for the key claim made in it.

  24. Kaleberg

    I gather most bitcoin blocks are processed in Inner Mongolia, part of China, where coal is cheap. The NY Times had an article on it:

    It sounds pretty industrial and increasingly centralized to me.

    Diesel designed his engine as every man’s answer to centralized steam power where a large, capital intensive prime mover powered an entire factory. His engine was to be a small engine, affordable and efficient to give workers independent power. As it turned out, the Otto cycle engine came closer. Diesel engines power ocean liners, locomotives, power plants, and cargo ships. A lot of technologies start out as great liberating ideas and wind up doing something quite different.

    Does bitcoin really only do 300,000 transactions a day? That’s maybe 4 per second. I think my local Safeway does that the night before the Superbowl. (I’m exaggerating here.) Is this supposed to be a currency? In the US, maybe 25,000 municipal bonds are traded daily, but they are considered illiquid.

  25. Lemonhead

    As soon as commenters I’m here start buying is when I will sell all my precious coins. The fact that most of the commenters here don’t understand the value is extremely encouraging.

  26. Charles H Green

    I went to a lecture tonight, by an eminent blockchain business person, in front of 100 alumni of Columbia University. It confirmed what I’ve been thinking: there’s a lot of confusion on this topic, and I don’t think even you, Ian, have nailed it down. Here’s my take.

    Blockchain is powerful, huge stuff. Bitcoin, the most famous instantiation of it – not so much.

    Remember the early days of Bitcoin: it was pitched as a currency that libertarians would love. Free of government intervention, it was super trustworthy, and no inflationary currency-debasing gummint could mess it up by printing useless currency, inflation, blah blah.

    One of the reasons for this – a prime feature of Bitcoin – was that the supply of Bitcoin was to be strictly limited. In fact, the stated limit is something like 24 million bitcoins in the year 2040, all according to a strict formula.

    The problem is: If you fix the supply absolutely, then the only variable to affect price is demand.

    This is not abstract economics. The massive volatility and huge upside of the price of Bitcoin has turned it into an asset class.

    Here’s the thing: you cannot be at once both a currency and an asset class.

    And given the ‘feature’ of fixed supply, it turns out that is a fatal bug for a currency. A thousand percent increase in an asset is a huge return on investment. But a thousand percent increase in a currency is hyper-deflation.

    Nobody in their right mind who a) cares about their wealth and b) is operating somewhere in the right side of legality is going to want to spend Bitcoin to conduct business. Why would you, if you expect the value to continue increasing geometrically? Unless you think it’s a bubble, in which case you might want to sell it quickly.

    The point is, it encourages hoarding. Not using it as a currency for transactions.

    This is a bug, not a feature. What you want in a currency is roughy stable purchasing power. What you DON’T want in a currency is raging volatility. For a trader, the reverse is true.

    You can’t be both. And given the decision to fix the supply, the die is cast – it’s an asset, not a currency.

    Which means half the hype you hear – the half that claims Bitcoin is a currency that will set us free, blah blah – is a pure lie. Or, more accurately, a case of raging ignorance. It cannot ever be a successful currency.

    The speaker I heard tonight acknowledged some ‘dual’ aspects of bitcoin, but compared it to gold, in its fixed supply aspects. But it’s NOT like gold: the supply of gold does vary – it varies with the price, which makes it more or less profitable to mine more of it. Also, gold prices are somewhat driven by industrial applications. By contrast, Bitcoin is absolutely limited, by design, on the supply side – and there are no comparable industrial applications.

    If you cannot be both currency and asset class, and if the die is cast for asset class by design, then Bitcoin is, just as Jamie Dimon said, a fraud; it’s good for criminals who don’t mind the relative rounding error in laundering money, and good for pyramid schemes for the naive suckers who buy the idea that they’re investing in a future social good.

    There may be, in fact I don’t see why there couldn’t be, some application of blockchain technology that can bring the benefits of secure transactions, cross-border trading, and traceable ownership to currency transactions. But it can’t be an application that begins by fixing the amount of the asset, thereby destroying its utility as a currency.

    Am I missing something?

Powered by WordPress & Theme by Anders Norén