Reconciling the Regression Theorem with Bitcoin

June 6th, 2013   Submitted by Michael Hendricks


When Austrians argue about Bitcoin there seems to be two basic camps. On the one hand there are those who use the regression theorem to argue that Bitcoin isn’t money, and on the other there are Bitcoiners who don’t see the regression theorem as relevant to Bitcoin at all. The regression theorem applies to Bitcoin in the same way that modern evolution theory applies to Darwinism. The regression theorem explains the origins of money, but it does not explain the current state of the money market.

It is important to keep in mind that the evenly rotating economy is merely a construction, and that life is always changing. Dogma is not a part of science. It is not a question of if, but a matter of when the regression theorem will be obsolete. Does this degrade Mises? No it does not. Newtonian physics is obsolete. Modern computer technology also renders obsolete the technology of the past. All of these things were necessary steps on the road to modern achievement.

The regression theorem seeks to explain the origin of money by saying that at some point a commodity used in barter became widely popular and individuals began to trade and use it as a medium of exchange. As far as it goes the regression theorem makes sense. The regression theorem explains why commodity money exists today.

The author of Smiling Dave’s Blog is a proponent of the regression theorem and is a critic of Bitcoin. In his article, Bitcoin Takes a Beating he writes:

“And this is where bitcoins achieve their fail. Unlike gold, no matter how far back you go, bitcoins were never worth anything intrinsically. There was never a reason for bitcoins to suddenly become worth a dime, or any other price. They are totally useless as jewellery, or anything else. So, that there is no reason people should accept them as being worth a dime, much less $33.

One may ask, then how come they were traded on certain websites at $33 a bitcoin? P.T. Barnum provided the answer. There’s a sucker born every minute. A small handful of people decided to speculate in bitcoins and bought them at whatever silly price they thought was worth it. But bitcoins were never generally accepted at any price but zero. They have no reason to be.”

The regression theorem does a good job of explaining the creation of money, however it does not necessarily apply to all forms of money. The first thing used as money had to have been a commodity that had value in trade in a barter system. Does this mean that all monetary institutions must have the same characteristics? I think not. The market is ever changing.

Bitcoin is not backed by a commodity and thus violates the regression theorem, according to some sects of the Austrian school, so what does this mean for Bitcoiners? There are some who have gone so far as to reject the regression theorem in total. I think this reaction is a little over the top and is unnecessary. The theorem explains commodity money, its genesis if you will, but the only bearing the genesis of money has on the current iteration of money is the cause and effect relation of the history therein. Bitcoin is money because people see it as such and it has utility that commodities don’t have.

The anonymity and decentralized design of the blockchain and the entirety of the digital nature of the currency give it advantages over a centralized, controllable commodity backed currency. Bitcoin has features that will remain valuable even after the fall of the State. Bitcoin, or something very much like it, is the future of exchange.

While Bitcoin doesn’t follow the regression theorem and is not based on a commodity, that doesn’t mean that the accuracy of the regression theorem and the exchange value of Bitcoin are mutually exclusive. Life is ever changing. The fact that money has now evolved passed a point in which the Austrians of old had considered does not mean that this current form is doomed to fail. Back when Mises wrote The Theory of Money and Credit computers didn’t even exist. Bitcoin was inconceivable when Mises first contemplated the origin of money. He had no way to take Bitcoin into  account.

The regression theorem concerns the development of money, it does not necessarily concern all adaptations of money. Applying this theorem to Bitcoin is stretching it beyond its scope. Just as physics has evolved over the centuries and just as our understanding of these things increases so too shall our understanding of money change. Change happens. It is foolish to fear it.

Clinging to the old forms of money simply because of a theory put down by one of the founders of the Austrian tradition is dogmatism and is indeed a critical error. Capitalism and the idea that competition and innovation are good for society are the antithesis of dogma. Any capitalist worth their salt is not a dogmatist. While education may be conservative by nature that does not mean that our way of thinking must too be conservative and dogmatic.

Tags: , , ,

82 Responses to “Reconciling the Regression Theorem with Bitcoin”

  1. Smiling DaveNo Gravatar says:

    I see a few arguments here. 1. The regression theorem is only talking about a barter economy. 2. The regression theorem never talked about money in a computer age. 3. The regression theorem is a history lesson only, and is not even talking about what a money must be, only what happened historically.
    4. Bitcoin has an advantage no other money has, its security and secrecy. 5. The regression theorem is over a hundred years old, and times change.

    The common mistake in all the arguments above is that they do not grasp that the regression theorem is a statement backed by a step by step, logical proof. As such, it can only be refuted in one of two ways. First way is by showing exactly where the proof breaks down because computers exist, or because we are in a money economy, or because times change. It is not enough to say that when Mises wrote his book there were no computers, for example. The article here fails to do this. The second way is to show that reality itself refutes the theorem, which the article fails to do.

    To help get what I an talking about, let’s give an example. Pythagoras wrote that for every right triangle, A squared plus B squared equals C squared. Along comes someone and says that Pythagoras was not talking about green triangles, that he lived long ago, that times change, that computers did not exist in his time and so the theorem does not apply to computer generated triangles, that he was talking about a barter economy not a money economy, that my triangle has great new features no other triangle has, like security and secrecy, and that there are plenty of right triangles in existence right now that refute his theorem.

    I hope the reader sees the flaws in such an argument against Pythagoras’ Theorem. The exact same flaws apply to the arguments in this article.

    I’ve dealt with all of this before in my humble article Bitcoin All in One Place. ne-place/

    • Seth KingNo Gravatar says:

      This whole battle over whether or not Bitcoin is money all comes down to definitions. The people clinging to the Regression Theorem have defined money as a commodity. If Bitcoin isn’t a commodity then it isn’t money.

      Others define money as a unit of account. Therefore, Bitcoin is money.

      If I could go back in time and ask somebody to define what an automobile is, what would they say? They would define it as something with an internal combustion engine with four wheels, etc.

      That’s all they knew back then. But is an electric powered bus with ten wheels not an automobile? Would Henry Ford refuse to ride a bus because he doesn’t think it’s an automobile? That would be absurd.

      Bitcoin is a wonderful invention that is going to liberate us from state controlled units of account. Any libertarian who fails to embrace that because of some definition of money from over 100 years ago is just going to beclown themselves.

  2. Michael HendricksNo Gravatar says:

    The regression theorem doesn’t even talk about tally sticks which have existed for millennia as well when I get the time probably this weekend I’m going to do a write up about tally sticks and Bitcoin.

    Comparing a mathematical theorem to a social science theorem is a non sequiter btw. Honestly I’m not sure why you think that a mathematics theorem compares to a social science. One of the main points I got out of Human Action was that with Praxeology Mises wanted to remove the idea of rigorous modelling from the science of economics (a soft science).

    The proof doesn’t apply to Bitcoin not because it breaks down but because it doesn’t make sense within the context of language to say that the reason money is money is because purple. Mises does a good job explaining the origins of commodity based money, but commodity based money wasn’t even the only money in medieval Europe, as I said Tally Sticks were also used.

    The point of my article is NOT to refute the Regression Theorem but to simply say it doesn’t apply to Bitcoin so stop trying to say that the regression theorem shows that Bitcoin is inadequate as money.

    • Martin BrockNo Gravatar says:

      As a mathematician, “regression theorem” makes me cringe. Mises’ account of the role of money is not a “theorem” at all in my way of thinking, and calling it a theorem is akin to Hayekian scientism. It’s mathematicism.

      And as you note, the “theorem” relies upon a Mengerian theory of the origin of money that doesn’t square well with historical accounts anyway. Tally sticks are not commodity money of the sort imagined by Menger at all. They are credit money, like the promissory notes of a fractional reserve bank.

      That said, I remain a skeptic of Bitcoin 1.0, any my skepticism has something to do with theoretical critiques of fractional reserve banking under a statutory gold standard, but it has nothing to do with the regression theorem.

      I expect Bitcoin 1.0 ultimately to fail as a monetary scheme largely because Bitcoin is not enough like a system of tally sticks. In the free competition to create stateless, digital money, I expect something more like tally sticks ultimately to prevail. Bitcoin works as a stateless medium of exchange only until something better comes along.

      When something better comes along, Bitcoin either morphs into a different system more like tally sticks (or becomes coupled to such a system) or it enters its Ponzi Scheme mode permanently.

      By “coupled to such a system”, I mean that Bitcoins could be used as something other than money in a hybrid system. The blockchain is all about solving the double spending problem (virtual counterfeiting) without a centralized registry of ownership. The original Bitcoin paper addresses this problem specifically, not digital money more generally.

      A system of electronic promissory notes must also solve the double spending problem. Conventional bank notes solve the problem with serial numbers and a central repository. People can validate a note’s serial number against a central registry. If two notes with the same serial number appear, one of them must be counterfeit.

      Typically, this registry is the issuing bank. Circulating banknotes quickly return to the issuing bank, because the bank pays interest on the notes only upon deposit, so a bank typically is holding a particular note and can therefore detect the existence of counterfeit note with the same serial number.

      So here’s how the Bitcoin system could contribute to a system of electronic promissory notes in a decentralized system of money and credit (the two being fundamentally inseparable). A Bitcoin (or a Satoshi) is not the unit of currency. It’s the serial number.

      • MAMNo Gravatar says:

        Yeah I was studying basic algebra with a friend of mine today and he was saying the same thing that calling the Regression “Theorem” a theorem was inaccurate and really way too generous.

        Bitcoin solves the double spending problem, I believe it does this with the Proof of Work. I’ve got to run but if I’m wrong about that and if you don’t know how Bitcoin has solved it I’ll find it. I just don’t have time to search right now.

        • Martin BrockNo Gravatar says:

          I grant that Bitcoin effectively solves the double spending problem. My point is that a system of promissory notes (promising a tangible good like silver rather than bitcoins) can use the same solution of the double spending problem.

          An electronic note promising silver is a digitally signed document. The document specifies a quantity of silver promised and corresponding Satoshis, say a hundred Satoshis per gram of silver. The promissory note incorporates the Satoshis themselves, i.e. the note is a Bitcoin wallet.

          The person (or persons) promising the silver owns the Satoshis, and he signs the note promising the silver with a private key. This person now promises me the silver by giving me the note including the Satoshis. My possession of these Satoshis proves that promisor owes me the specified quantity of silver.

          I may similarly transfer the note/Satoshis to anyone else trusting the promisor, so if the promisor is widely trusted, the note is money. I may also divide the note. From a note promising a gram of silver associated with a hundred Satoshis, I may create two notes each promising half a gram of silver and each associated with fifty Satoshis.

        • Martin BrockNo Gravatar says:

          To be clear, I prefer a system of notes promising common labor rather than silver or another precious metal. Extending credit by accepting notes promising a precious metal, or any good with a supply that is not sufficiently elastic relative to demand for money, is problematic for reasons we can discuss. Critics of a fractional reserve banking system have a point, but they often misunderstand the system.

          Unfortunately, many libertarians, particularly an-caps, hear “common labor” and think of Marxism and a Marxist central bank. To be very clear, I’m not any sort of Marxist and have no sympathy with Marxists or with any sort of central banking regime.

          In the system I imagine, your promissory note may promise anything you want to promise. I believe that notes promising common labor best serve the interests of common laborers and also create a more stable system of money and credit, but I have no interest in compelling anyone to use any particular form of money.

          • MAMNo Gravatar says:

            I find the Labour Note to be interesting to think about. I don’t really have a problem with a Labour Note, I’d give it a try if I had the opportunity.

  3. DimitriNo Gravatar says:

    the regression theorem as i understood it money must first have a utility along with the other properties of proper money(divisbility. fungiblity, etc). Its just so happens commodities had the most utility back than for obvious reasons. Gold and silver, utility being mainly beauty and jewelery in the beiginning become the prefered commodity for medium of exchange. Bitcoin has better utility then any other commodity cause of the fact that its a ‘synthetic intangible commodity’. Its utility is a payment system, value transfer system, and can be use for much more like smart contracts, wills, and can be hidden if one so wish.

  4. AforAtmosphereNo Gravatar says:

    I don’t understand why the 2 monies can’t coexist peacefully. Gold and Bitcoins both have properties the other does not have that are beneficial. So therefore wouldn’t they both have reason to exist and be used as money?

    • Smiling DaveNo Gravatar says:

      It’s not that gold and bitcoin are fighting. Even if gold did not exist, bitcoin, by it’s very nature, is not and can never be money, as proven by the regression theorem.

      If you reread the article Bitcoin Takes a Beating, you will see that you are asking something like this: Why do women not use chewing gum in their hair? There’s shampoo and there’s chewing gum, and they need not fight, especially since chewing gum has so many good properties shampoo doesn’t have.

      The answer is, of course, that chewing gum has a fatal flaw that outweighs all its useful properties and will never be used by women in their hair. The regression theorem is all about explaining what the fatal flaw is in bitcoin.

    • MAMNo Gravatar says:

      I use silver and Bitcoin they aren’t mutually exclusive for people who don’t cling to dogma.

  5. Smiling DaveNo Gravatar says:


    1. Tally sticks were fiat money imposed on the populace as the only way to pay taxes, which gave them their value, as Wikipedia explains. I explained this concept in detail in my article Bitcoin and Bitclothing: lothing/

    2. The whole point of Human Action is that Economics can be established on a logical foundation, built from self evident axioms, with every step logically deduced. He did not consider it a soft science. He did mock mathematical modelling by differential equations and the like, which is another kettle of fish.

    3. Your summary of the theorem, that money is money because of purple, is not accurate. It claims that nothing can ever become money unless it first has a non monetary value which makes it widely desired and coveted. It offers a rigorous proff of this claim. Bitcoin has no non monetary value, thus violates the theorem. In your OP, you very wisely linked to my article Bitcoin takes a Beating. [ -beating/]. Anyone interested can reread that article and see what the theorem actually says, and the rigorous proof offered in support of it, in simple language.

    BTW I am a tad surprised at the tone you adopt in your comment.


    Saying bitcoin’s utility is a payment system, value transfer system, doesn’t help at all, because the theorem states the initial utility must be NOT as a medium of exchange. See Bitcoin takes a Beating, linked to in previous paragraph.

    Saying can be use for much more like smart contracts, wills, and can be hidden if one so wish, also does not help at all. And the reason is because although it CAN be used for that, there is nobody actually using it for that. Mises says a potential money must start off as something that people actually wanted and actually used for some use besides money. See Bitcoin takes a Beating, linked to in previous paragraph.

    If and when people start buying bitcoins to write their wills, we’ll rethink the issue.

    • Michael HendricksNo Gravatar says:

      That money is money because purple was not my summary of the Regression Theorem it was my summary of why applying the regression theorem to Bitcoin doesn’t make sense.

    • DimitriNo Gravatar says:

      ‘Saying bitcoin’s utility is a payment system, value transfer system, doesn’t help at all, because the theorem states the initial utility must be NOT as a medium of exchange’… There is nothing saying that the initial utility canNOT be as a medium of exchange. It just explains how money comes to be and how the medium of exchange transform (as economies become more complex and sophisticated) in relation to there utility

      • Smiling DaveNo Gravatar says:

        “There is nothing saying that the initial utility canNOT be as a medium of exchange.”

        Of course there is. That is the whole point of the theorem. I refer you to my article that Mike linked to in his OP

        • ShawnNo Gravatar says:

          The initial utility was cryptography. Please see my post at the bottom.

          • Say “cryptography have value” is to beg the question “for who?”

            The initial utility is not known. What can you do with bitcoin for yourself?
            Was you the first to exchange BTCs for good or service?
            Someone could have just exchange BTC for money for pizza just to do something funny and interesting or he could have exchanged it by mistake.
            IT DO NOT MATTER.
            As an initial exchange value was determined, the market forces started to work on it.

            • ShawnNo Gravatar says:

              Cryptography has been around as long as civilization. And if you want to go that route, for whom did gold first hold value? This has really devolved into a bunch of pedantic silliness.

  6. Michael HendricksNo Gravatar says:

    If you read my last article “Applying Diax’s Rake to Thoeries of Objective Value” found here ies-of-objective-value/

    The last case that I talk about applies to the intrinsic value claim in The Theory of Money and Credit so just read that.

    Whilst I am interested in reading your other articles about the Regression Theorem and Bitcoin, I don’t know when I’m going to have the time. I’m running behind schedule and I think it’s going to take a week or more for me to catch up.

    Peace be with you.

    PS I’m not sure what offended you in my previous comment but I’m not trying to be hostile so don’t take it as such. I’m just in a little bit of a hurray and doing about a dozen things right now.

  7. Smiling DaveNo Gravatar says:

    The Diax’s Rake article has a logical error in its analysis of that section by Mises. You write:

    “Just because it makes sense for someone to take the utility of purchased items into account when determining what it is worth doesn’t mean that they must.”

    All Mises is saying is that money cannot be used for anything else but to buy things. So that if anyone based his subjective valuation on money on anything other than its ability to buy things, he would give it a subjective valuation of zero, because he cannot do anything else with it.

    Thus any subjective valuation money has must stem from it’s ability to buy. And the more it can buy, the more subjective valuation people will place on it, all other things being equal.

    In any case, none of this is relevant to Bitcoin.

    Mike, I understand if you are busy, as am I. So I refer you and all interested readers to my articles on bitcoin.

    Seth’s article I was Wrong About Bitcoin also has at least two serious errors [one of which is misunderstanding the concept of “backed by”].

  8. Michael HendricksNo Gravatar says:

    Suppose that Bitcoin is NOT money and that you’re right and Bitcoin violates the Regression Therom, does that invalidate Bitcoin as a medium of exchange?

    Or is there more to your argument than semantics?

    Also I’ve heard the same claim made about gold as money that you just made about tally sticks as money. Which is correct, are both?

    It makes sense that governments would want a standardized currency such as gold or silver because its more desirable to to take 5 schillings in tax than half a watermelon.

    Gold is more consistent than a tally stick as a tally stick represents a contract and is subject to change on a case by case basis. Gold does not enjoy this indpendence as its value is linked to its mass. I suppose however that tally sticks are more desirable than fractions of fruit so it’s possible that they were also used to pay taxes. If the government doesn’t care how they are paid then fruit could be used to pay taxes.

    Could it be said that Bitcoin has utility and that utility is in the way it’s designed? I think Seth may have touched on this in his article but I value Bitcoin because I can use it and its reasonably hard to track this is why I don’t use banks I have trust issues.

    I define money as a medium of indirect exchange.

    You seem to define money to include the restriction that it must have value beyond its value as a medium of indirect exchange.

    Given your definition Bitcoin is not money.

    Given my definition it is.

    In either case Bittcoin has value and is traded and is used as a medium of indirect exchange. It has the potential long term to free the market and thus the world and at the very least to cut banks out of the equation and potentially reduce the size and power of the Banking Cartel. It can undermine the finicial base of the State the more individuals use it and aviod the system. Do you contest these facts? They make Bitcoin worth using and promoting in my opinion.

    • Don DuncanNo Gravatar says:

      Regression theorem uses a historical approach to money. But it does not follow that because the origin of money was from intrinsic worth to abstract worth, i.e., as an indirect medium of exchange, all money must follow this pattern. Without intrinsic worth to restrain growth money may lose its value, but not necessarily if another control mechanism is found. BTC does have that. It follows that BTC can work as money.

      • MAMNo Gravatar says:

        I really think the hang up is the word “intrinsic”.

      • Did I cited Mises words for nothing?
        He do not used any historic explanation, it used logic.
        Every time the conditions described by Mises are met, the same things happen: a scarce commodity is used as money and the initial value is dependant on other uses.

        OTHER USES do not imply uses in the past or present. They could be future uses.

  9. Both Smiling Dave and Michael Hendricks are wrong: Mises Regression Theorem and Bitcoin are not mutually exclusive.
    The correct interpretation of Mises Regression Theorem do not say Bitcoin is not money and fit perfectly with the (short) history of Bitcoin.

    Bitcoin is a (digital) commodity. The value of bitcoin (or anything else) is subjective, so different persons will have different ideas about the value of one BTC.
    Some could believe it worthless and others could believe it have some value. The same is true for gold (I, for example, give gold no value at all apart as a medium of exchange).
    What we know is Bitcoin have by design some features very useful to anything used as money. Some features are so good, no other type of money have it.
    It just need for a subset of people to give it a value (a market price in any good or service) to allow the market forces to start to see it as a form of money (or quasi-money).
    When it start to be used for indirect exchanges (for example a programmer buy pizza with bitcoin, the pizza-maker pay for a month VPN subscription with bitcoin and the VPN provider pay the programmer service with bitcoin), it just enter in competition with other forms of money/currencies.
    When this happen, the superior qualities of Bitcoin allow it to be preferred as a medium of exchange and a store of value to fiat (easy) and gold and silver (not so easy).

    Gold and silver as a material form of money have already failed, because their use need to be done in person or with money substitutes (fiat, payment services, wire transfers, banks, etc.).
    If gold and silver were not unsuited to be used as money in our daily life, fiat would not be used, VISA and MasterCard would not be in business and so on.
    The modern world need a form of money that can be exchanged over long distances, fast and easily. Gold and silver can not be exchange over long distances with ease.

    This do not imply gold and silver are useless as a mean of exchange and a store of value or worthless over their industrial use.
    Their are necessary in our fiat world because they are the ultimate store of value, as they can not be easily forged or produced at will.
    They will be useful also if and when Bitcoin become the preferred currency available in all the world.
    They will be just another very liquid asset people with too much money (aka Bitcoin) will buy to diversify their holdings.

  10. ShawnNo Gravatar says:

    Bitcoins are cryptography.

    Cryptography has long had value.

    Therefore, the regression theorem still holds true, and Bitcoins are money.

    • Michael HendricksNo Gravatar says:

      Yeah this occurred to me a couple days ago.

    • You can also trace the Regression Theorem through the value which has always been placed on transferring money —

    • Bob RobertsonNo Gravatar says:

      Nicely said. The “service” of anonymity certainly does have value.

      • Smiling DaveNo Gravatar says:

        All these commenters have not understood, or perhaps not even read, the regression theorem. Because the theorem insists on a non monetary use for what is going to be money. Anonymity when buying is not a non monetary use. Think about it.

        As for cryptography, who ever used bitcoin as cryptography? Have spies sent secret messages to each other by sending bitcoins back and forth? Of course not. And if they did, they are a trivial amount of people. And if we are talking about cryptography when buying stuff, see previous paragraph.

        • On this I have to agree with you.
          They don’t understand Mises’ RT, so their explanations are wrong.
          But Mises RT talk about gold value before becoming an indirect mean of exchange. Gold was used as jewelry, for vanity reasons.
          We exactly do not know how much it was valued just for this reason, just it was valued.
          In the same way, bitcoin had a subjective value for the people mining and interested in the project. Some one else of them could attribute a subjective value to Bitcoin just as a vanity or just as a n object of curiosity.

          In fact, before the first exchange (the pizza exchange) Bitcoin was used to pay for services like VPN. The pizza exchange is interesting because the miner offered 20.000 BTC in exchange for two pizzas (in New York), someone in Florida thinked it cool to buy them for 40 US$ and agreed to use its credit card to pay the store in New York to deliver the pizza.
          This moment is when BTC start to move from barter and start becoming a mean of indirect exchange.

          It must be noted that the network of BTC was already 15 months old when the pizza exchange happened. So a network of trust in its ability to last was created. If it lasted 15 months before without being really used, the actors could trust it would last other 15 months more even if used.

          What exactly was in the minds of the people exchanging bitcoin for goods, services or fiat we could not know for sure, but we do not need to know it, as we do not need to know what was in the mind of the people using gold no more for barter but as a mean of indirect exchange.
          We know it happened, because logically happened.

          From that point, the market gave to a bitcoin a price other actors in the market could agree and disagree to form a new price.

  11. Don DuncanNo Gravatar says:

    I see no problem with BTC as money. But one change may kill it. If someone can find a way to do what govt. did to hard money, counterfeit it, that would ruin it. Time will tell. I’m waiting to use BTC because it is so limited. When I can use it at Trader Joe’s, WalMart, restaurants or movies, I will use it exclusively. It should be proven safe by that time. I don’t want to wake up and find my net liquid worth has just vanished .

    Dave missed the point of intrinsic worth. It is necessary to stop counterfeiting. BTC does that without value outside being a very, very convenient medium of indirect exchange. In fact, it has more value than metal in that it is safer. BTC cannot be taken or tracked by the state (yet). Let them outlaw it. It can’t be enforced. Once again, monetary transactions would be private. It could mean the end of the state. At least, it will weaken it substantially.

    Eventually, machines may sell gold/silver coins for BTC for those rare times when cash is needed.

    This is an excellent example of the market using the ‘net to expand freedom.

  12. Don DuncanNo Gravatar says:

    All the reasons Dave gave for BTC not being money can be applied to fiat money. But fiat has been used as money by billions for over four centuries. It will turn out to be disastrous but that is because govt. can’t do anything right. If the state had not gotten greedy their scam could have gone on forever. They have no one to blame but themselves. It’s the goose/golden egg parable.

  13. Smiling DaveNo Gravatar says:

    Well! I did not mean to raise up such a storm. I was informed that someone linked to an article of mine, and so posted what I thought would be one short reply.

    In any case, all the objections raised here have been dealt with in my article Bicoin All in One Place [ ne-place/ ]

    It’s not a mere question of semantics, or definition, or subjective value. Cryptography is irrelevant here, as is fiat money. When a programmer buys a pizza with a bitcoin, that does nothing. Whether you are a libertarian or a statist is also not relevant. All these things are explained in the article.

    I know I owe this honored group a detailed explanation and proof of my assertions. The thing is, if you check the forum over at you will see I have written hundreds of posts on this subject, and am weary of it. All I have to say is in the article I linked to. It is comprehensive. All the issues raised here have been talked about endlessly in the forum, so it’s not like I’m being taken by surprise. So, with apologies, I refer you to my article, and will reluctantly not post here in the comments.


  14. Peter ŠurdaNo Gravatar says:

    I treat the regression theorem and Bitcoin in my master’s thesis, and I was also interviewed about it in one of the University of Texas Mises Circle sessions.

    I also addressed Smiling Dave several times, including a more formal point by point refutation on my blog.

  15. Peter ŠurdaNo Gravatar says:

    Click on my name and then click on “Publications, lectures and interviews”, it’s all there.

  16. Michael HendricksNo Gravatar says:

    “When a programmer buys a pizza with a bitcoin, that does nothing. ”

    I really have no idea what to make of this.

    • Smiling DaveNo Gravatar says:

      You haven’t had time to read the articles, have you? About how a medium of exchange and a money have to be in widespread use?

      Am only going to answer from now when someone writes they read Bitcoin All in One Place and the links there.

      I mean, where’s the thirst for knowledge?

      • Peter ŠurdaNo Gravatar says:

        Medium of exchange does not have to “be in widespread use”. That is something you invented yourself, just like several other arbitrary assumptions, and I thoroughly documented these throughout our debates.

      • Michael HendricksNo Gravatar says:

        It takes time for things to be established. Not even the State could mandate a currency overnight.

        • Smiling DaveNo Gravatar says:

          So you agree that one guy buying a pizza is nothing. Thank you. You add, however that more time is needed.

          Yes, it takes time. But the question is, will that time ever come for bitcoin? The regression theorem proves it never will, and thus the one guy with his pizza, admittedly nothing in itself, will also never blossom into widespread use.

          • Michael HendricksNo Gravatar says:

            That one guy with a Pizza has lead to quite a few other Pizzas which is also nothing eh?

  17. Michael HendricksNo Gravatar says:

    So I looked up Tally Sticks on the Wikipedia page and I wonder if this Dave guy even bothered to read it. It opens with a statement about how Tally Sticks were used in the Upper Paleolithic era. Upon investigation into the Upper Paleolithic era it turns out that that era was before agriculture was invented. That means that Tally Sticks were used !!!!BEFORE!!!!!! there could have been a central government to force Tally Sticks on people…

    • Smiling DaveNo Gravatar says:

      Could you quote me the line that says they were used as money in the U. P. era?

      • Michael HendricksNo Gravatar says:

        “Tally sticks first appear as animal bones carved with notches, in the Upper Paleolithic; a notable example is the Ishango Bone.”


        • Smiling DaveNo Gravatar says:

          From that same source:
          “…the Ishango bone is a bone tool…” Tool, not money.

          Please quote me the line that says they were money in the UP.

          • Michael HendricksNo Gravatar says:

            Money is a tool.

            • Smiling DaveNo Gravatar says:

              Ishango sticks are tools.

              Money is a tool.

              Therefore Ishango sticks are money.

              My last reply here, sorry guys

              • Michael HendricksNo Gravatar says:

                You keep threatening that this is going to be your last reply…

                • Michael HendricksNo Gravatar says:

                  My point about money being a tool is that you can’t say this is a tool therefore it can’t be money because money is a tool I didn’t say a tool was automatically money. It could be though. A shovel standard might be interesting.

                  What purpose does gold have other than to look pretty? I prefer to look at silver.

                  • Michael HendricksNo Gravatar says:

                    Maybe your arguments on your blog hold water I don’t know. Maybe at some point I’ll read the rest of your entries.

                    I may have already read that Bitcoin in One Place article you keep referencing it seems familiar but I don’t really remember anything about it.

                    Meh. I don’t know. You don’t have to use Bitcoin but I will use it. If I am wrong well then I’ll pay the consequences for that mistake. Obviously I’m confident that you are wrong.

                    Good Luck!

                    Peace be with you!

  18. MacianoNo Gravatar says:

    Gold can not bring down central banks, bitcoin can. It is untamable, uncrontrollable and unstoppable: bitcoin makes use of crypthography and P2P, both systems have proven themselves to be antifragile.

    Gold is also antifragile, because it will always have some value after a collapse. But gold can be confiscated more easily and people (the masses) tend not to use it as a currency, when there are fiat alternatives.

    Also, bitcoin — through its OS nature — can be improved continuously if that would be necessary; there can also be addons (such as zerocoin), innovations (hardware wallets), mediums (BitPay) or complements to the system (such as Ripple & open transaction servers) to make the electronic money ecology function even better.

    It’s also not true bitcoin had no value in the beginning. Bitcoin was a price, status symbol, for gamers after succeeding certain levels in a game. While that is not high monetary value, it is still value. But, if people want to load up on gold and silver, they’re welcome, we’ll see how this argument will be play out the coming years..

    • Smiling DaveNo Gravatar says:

      You haven’t read my articles, have you?

      • Maciano Van der LaanNo Gravatar says:

        SD, i did not really respond to you. More to the phenomenon of people denying btc’s viability while it keeps growing in all directions. After your comment i felt challenged and read some of your articles. No offense, they boil down to: btc is not money and gold is; gold had some value (jewellery) and btc does not. And lots and lots of mises. Tl; dr

        When social media started i heard a lot of people arguing it was not going to work: why voluntatily put yourself, privacy online. No surely that could never happen. Looking at twitter, facebook, linkedin, pinterest, etc., it is clear people did see some value in these services — they get something out of it. Quite similar to regression theory reasoning.

        Austrian economics is science, right? It can be adjusted to new realities.

  19. Bob RobertsonNo Gravatar says:

    The Regression Theorem only has to explain the origin of MONEY, not the origin of any particular currency.

    Money is a social function that is and has been filled by many different commodity currencies, fiat currencies, mind-share in so-called “gift” economies, etc.

    So once people have the habit of “money”, once that social function has been defined, anything can act as currency and serve the function of “money”.

    Thus, Bitcoin does not violate the Regression Theorem.

    • Smiling DaveNo Gravatar says:

      “The Regression Theorem only has to explain the origin of MONEY, not the origin of any particular currency.”

      No. The regression theorem explains the origin of every and any particular money. Proof: Show which line of the proof of the regression theorem applies to MONEY, and not to any particular money. You won’t be able to. The proof works perfectly for a given particular money.

      Could not find “mind share” in wikipedia as having anything to do with money. Link, please.

      “Gift” economies are irrelevant here, as they, by definition, do not operate with money. We are talking about money, not economies that do not use money.

      Mises explained how fiat currencies satisfy the theorem, and bitcoin cannot satisfy the theorem the way they do. Look it up. It’s right there in his discussion of the regression theorem.

      Having the habit of money, having that social function defined, is not enough to make anything a viable currency. The thing still has to satisfy the regression theorem, as proven above.

  20. ShawnNo Gravatar says:

    There is, of course, also the argument that Mises was wrong in the first place: n-challenging-mises-regression-theorem/

    Heresy!! How dare you defy the God of Austrian Economics!! [/sarc]

    • Smiling DaveNo Gravatar says:

      I wrote about that very article over a year and a half ago, laying out its blunders for all to see. Link: es-regression-theorem/

      It’s not about defying or not defying the God of Austrian Economics. It’s about examining the ideas presented, without regard to who presented them.

      Now that we know what we should be talking about, the ideas, what have you to contribute? Have you found an error in my rebuttal that I just linked to? Please tell us about it if you have, and precisely what the mistake is.

      • ShawnNo Gravatar says:

        You will note I did not put that comment at the end of any of your active threads – it was not actually directed to you, personally, but just a general sarcastic remark regarding the tendency of some folks to not even entertain the idea that Mises may have been wrong on some things.

        As for the “Now that we know what we should be talking about…” I did not claim those views to be my own, so why do you feel I should be defending them to you? Tell you what, if I get the time, I’ll look over your “rebuttal” and respond. I did take a look at your blog last night, but you tend to be a bit long winded and I just haven’t had the time.

  21. ShawnNo Gravatar says:

    Ok, I’ve taken the time to read Smiling Dave’s “Bitcooins take a Beating” blog post, and considered it along with all the other points and in light of the Regression Theorem, and here’s what I’ve come up with after very careful consideration:

    I concede that Bitcoins are not money.

    Something I’ve noticed absent from this debate is the word “currency.” That’s possibly because most people today believe the words currency and money to be interchangeable, and that’s not actually the case. Currency is something that can be used as an intermediary in exchange, whereas money has that property, but the additional properties we all know and love: fungible, divisible, etc, etc. Examples include Berkshire Bucks (those familiar with Berkshire, MA know what that means), Amazon credits, trading cards, and so on. Of course, this includes all forms of money. That is, all money is currency, but not all currency is money.

    Bitcoins, then, are a form of currency. Their value, however, is not that they perform a singular purpose, ie, give you discounted access to the shops, restaurants and bars in Berkshire, but rather, serve as a rather broad medium of exchange in an extremely decentralized realm; moreover, they do this by having certain innate qualities that make them favorable to many people: very low transaction costs, a high level of security through cryptography, and a certain amount of anonymity. What this really does is allow Bitcoins to be a superior currency, so much so that its true value is as a payment system. One can exchange one’s fiat currency for Bitcoins and then send those coins to anyone, anywhere in the world for whatever one wants.

    These facts have led Bitcoins to be tested by the market in spades. We all know recently that Bitcoins had a recent spike in trade value, all the way up to ~$270.00 USD, due in large part to formal recognition by world governments along with its acceptance by some major businesses. Of course, that was a bubble, and, like all bubbles, it broke and the price dropped dramatically. But here’s the thing: the price didn’t go to zero, or for that matter, anywhere close. As I write this, Mt.Gox shows them trading at $110.10/BTC. In Dave’s article, he states: “One may ask, then how come they were traded on certain websites at $33 a bitcoin? P.T. Barnum provided the answer. There’s a sucker born every minute. A small handful of people decided to speculate in bitcoins and bought them at whatever silly price they thought was worth it. But bitcoins were never generally accepted at any price but zero. They have no reason to be.” But hold on: we’re talking about way too many rational actors for that to be the case any more. A value still exists, obviously, for Bitcoins, in the praxeological sense.

    So, it is here that I move on to Dave’s other bold statement: Bitcoins can NEVER be money, as proven by the Regression Theorem. This is patently FALSE.

    I think the flaw in Dave’s thinking lies here: “Unlike gold, no matter how far back you go, bitcoins were never worth anything intrinsically.” (Emphasize “intrinsically.”) There is, simply put, no such things as “intrinsic” value. The value of any and all things exist in the minds of human beings, to include gold. However ridiculously improbable it is, it is a non-impossible event that humans could discard the value of gold. The value of anything lies in the minds of human actors; for something to be currency relies on at a minimum a party of three; for something to be money require ubiquitous acceptance of a certain currency. Gold, silver, wheat, salt, animal pelts and whatever else would’ve had to have gone through this “currency phase” prior to becoming money.

    What will it take for Bitcoins to become money? The general adoption of them as a currency of choice. Essentially, a “Bitcoin economy” (or economies!) will need to develop. This means that Bitcoins could be earned through regular earning methods, eg, through your wage labor (such as you do now for your particular nation’s fiat system), and readily exchanged for your needs and wants, eg, pay your rent, buy your groceries, and so on. If/when this happens, Bitcoins will have truly emerged as a money, because one could then apply the Regression Theorem and trace their value back to the time when they simply served as an intermediate payment system – a value, in fact, that they could return to should they fall out of favor as a “money.”

    In conclusion: Smiling Dave was half right in that Bitcoins are not currently money, but wrong in his interpretation of the Regression Theorem stating they never can be.

    Thanks for reading.

    • Michael HendricksNo Gravatar says:

      I debunked the intrinsic argument in my Diax’s Rake article. Glad someone also came to that conclusion.

    • MajamaluNo Gravatar says:

      “This means that Bitcoins could be earned through regular earning methods, eg, through your wage labor (such as you do now for your particular nation’s fiat system), and readily exchanged for your needs and wants, eg, pay your rent, buy your groceries, and so on. If/when this happens, Bitcoins will have truly emerged as a money”

      Gold does not meet all those requirements. Should we conclude that it isn’t a form of money?

      It is not possible to draw a line between what is money and what is not money. Given its attributes, some goods are better as money than others. That’s it.

      • ShawnNo Gravatar says:

        Gold *was* used like that at one point. Though it isn’t any longer (thanks mostly to legal tender laws) it has retained its status as money. I think Bitcoins will need to go through a similar phase before they can really be considered money.

        • MajamaluNo Gravatar says:

          So gold is money because it retained its status as money, not because it is being used as money. What about salt, sea shells, big rocks, etc?

          • ShawnNo Gravatar says:

            Ok, Majamalu, here’s some money: MONEY. Now go spend that. My point is that for something to be money needs more than just a few people willing to use it, it needs to be adequate to stand alone in an economy. Gold retained its monetary status because, despite being pushed to the side by fiat, it’s still considered valuable and people will trade for it. Bitcoins are not there (yet). Just because you can trade for a few items with them or gamble with them does not itself make them money – as I pointed out, that makes them currency. Feel free to disagree, but until Bitcoins get out of the “fiat to bitcoins back to fiat” paradigm that they currently exist in, it’s really hard to actually consider them more than a currency.

            • MajamaluNo Gravatar says:

              Bitcoin is right now facilitating thousands of instant long distance – and almost free – transactions per hour, without involving a central point (something you can only do with Bitcoin). Is that enough to consider it a form of money? Which is the threshold?

              I’m arguing that we are better off thinking in terms of moneyness (Hayek) than trying to define the exact moment when a particular good becomes money.

              • ShawnNo Gravatar says:

                True…I just got into the whole seeking an answer thing about if it is or isn’t money…bottom line, I will continue to use them and let their natural evolution take care of itself.

  22. Mises wrote:
    “To explain a phenomenon theoretically means to trace back its appearance to the operation of general rules which are already comprised in the theoretical system. The regression theorem complies with this requirement. It traces the specific exchange value of a medium of exchange back to its function as such a medium and to the theorems concerning the process of valuing and pricing as developed by the general catallactic theory. It deduces a more special case from the rules of a more universal theory. It shows how the special phenomenon necessarily emerges out of the operation of the rules generally valid for all phenomena. It does not say: This happened at that time and at that place. It says: This always happens when the conditions appear; whenever a good which has not been demanded previously for the employment as a medium of exchange begins to be demanded for this employment, the same effects must appear again; no good can be employed for the function of a medium of exchange which at the very beginning of its use for this purpose did not have exchange value on account of other employments. And all these statements implied in the regression theorem are enounced apodictically as implied in the apriorism of praxeology. It must happen this way. Nobody can ever succeed in construction a hypothetical case in which things were to occur in a different way.”

    All confutations from SD, in my opinion, can be traced to this statement of Mises: “have exchange value on account of other employments”.

    SD misunderstand this because his interpretation of “other employments” is too reductive; as he believe gold is money and money can only be a material commodity, he think only about other immediate employments in the realm of material commodities.

    But “have exchange value on account of other employments” do not talk about these “other employments”. Other employments could be religious, ornamental, vanity, if anchored at the present time of the users, but they could also be in the future. Someone could foresee the employment of gold as a mean of indirect exchange as someone could foresee the use of Bitcoin as a mean of indirect exchange.

    Before the € was introduced to supplant the national currencies of the EU, it was printed but it was not widely diffused and was not legal tender.
    If I had robbed the first container full of banknotes and coins six months before the starting of the exchange, I could have used the banknotes and the coins to buy something if I was able to persuade the sellers to accept them.

    Surely I should have to pay much more than the current market price, because the owner could not take the euros to the bank immediately and could not circulate the banknotes freely. But, as the day of the bank would start to exchange the national currencies for the € would draw near, the discount would be smaller and smaller.

    In this hypothetical case, € would be treated as money (a means of indirect exchange) even if they were not (at the time) money and in widespread use, because people would had see them as valuable to have in the future as money in widespread use.

  23. PrimeNo Gravatar says:

    “While Bitcoin doesn’t follow the regression theorem and is not based on a commodity”

    Yes, Bitcoin is a commodity. It has value to people as a token – as a medium of exchange.

    If it’s a commodity, we should also be able to calculate its value against other commodities.
    How do we economically calculate the price of a Bitcoin as a token? At the very least, there is the bare minimum cost it imposes, in terms of equipment, labor, and electricity, on the miners who keep the system running. Contrary to what some have argued before, a Bitcoin is not backed by these other things, but they do show how people are calculating this commodity’s value against other goods and services.

    The only thing to keep it from changing from mere commodity into commodity money is widespread adoption. It is not (one of) the most saleable goods… yet.

    • You can not calculate value. Value is subjective.

      You can calculate costs of production and see if they are more or less than the price you sell them:

      If you are able to produce BTC (mining) and after selling the BTC mined and paid for the costs of production you are left with a profit, you know it is valuable to produce BTC mining.

      The profit obtained is, at least, a part of the value of the Bitcoins for you.

      • PrimeNo Gravatar says:

        I think you misunderstood. Costs are prices, prices are ratios of value. If you are calculating costs, you’re calculating value. Yes, it’s subjective. But people express their values as prices, which again are ratios of values.

        • I stick with Rothbard (and Mises) when they state that you can not make rations of values, because they are on a ordinal scale and not a cardinal scale.
          There is no way to make a ratio between First and Second and you can between one and two.

          • PrimeNo Gravatar says:

            You’ve misunderstood Mises and Rothbard. Prices cannot be treated as discrete units of some objective measurement of value. This is true. But values can be represented as ratios. You just can’t express the ratio as some non-subjective scale of value.


            • I do not understand how you could arrive to this conclusion reading the chapters cited.

              • PrimeNo Gravatar says:

                Let’s look at this another way… When Mises says that socialism fails because, lacking prices and lacking ownership, it cannot economically calculate in a rational way – what do you think he means when he is discussing “economic calculation?”


                “…monetary calculation fulfils all the requirements of economic calculation. It affords us a guide through the oppressive plenitude of economic potentialities. It enables us to extend to all goods of a higher order the judgment of value, which is bound up with and clearly evident in, the case of goods ready for consumption, or at best of production goods of the lowest order. It renders their value capable of computation and thereby gives us the primary basis for all economic operations with goods of a higher order. ”

                Measuring value in an ordinal rather than cardinal fashion, we can still calculate peoples’ ratios of value. Prices can tell us that people have valued a pound of bananas more than two pounds of potatoes, but less than six pounds of potatoes, for example. These are ratios, and when people express their preference ratios via market prices, it enable entrepreneurs to calculate these ratios to determine the most valuable ways to allocate resources for producing bananas and potatoes.

  24. MoniqueNo Gravatar says:

    Have only come across Bitcoin recently- am very interested to track its progress and see how it develops into the market.