Tuesday, October 30, 2012

The European Central Bank Talks Bitcoins and the Misesian Regression Theorem

Check this out.

The European Central Bank has put out a 54 page paper discussing Bitcoin and other virtual currencies, titled, The Relevance of Virtual Currency Schemes for Central Bankers. They are getting nervous.

The paper even brings up Ludwig von Mises and his Regression Theorem.

You can find the entire report here.

As soon as I have a chance to read the paper, I will post comments at EPJ.

(ht Trace Mayer)


  1. Here's what the article has to say about bitcoin and the Regression Theorem:

    "Although the theoretical roots of the scheme can be found in the Austrian School of economics,
    Bitcoin has raised serious concerns among some of today’s Austrian economists. Their criticism
    covers two general aspects:4 a) Bitcoins have no intrinsic value like gold; they are mere bits
    stored in a computer; and b) the system fails to satisfy the “Misean Regression Theorem”, which
    explains that money becomes accepted not because of a government decree or social convention,
    but because it has its roots in a commodity expressing a certain purchasing power."

    These statements are easily answered here:

    a) No thing has "intrinsic value" - not even gold. Things are "valued" by external actors/individuals for the useful properties that they provide.

    b) Gold is valued for what it is and what it enables one to do. It is a unique commodity that is valued. Bitcoin is valued in the same way.

    For more information on Bitcoin and the "Regression Theorem" see:

    Bitcoin: A New Commodity Created To Serve

    1. Corrected link:

      Bitcoin: A New Commodity Created To Serve Market Demand

    2. I agree with your A, but as for your B, I think the important insight that comes out of the money regression theorem is that a monetary commodity must have at one time been valued by individual human beings 100% for DIRECT USE--before any other individual human beings could subjectively assign any component of EXCHANGE value to the commodity themselves for the first time. So, I don't think your B explains how bitcoins were ever valued by anybody for 100% DIRECT USE.

    3. In short, the Mises Regression Theorem is wrong (I know this is heresy to the Austrian Religionists) but think through this a little:

      Bitcoin was primarily created to be used (100% DIRECT USE) as a medium of exchange (although it has many other uses ie., title registries and smart contracts). There was never the intent for it to have a prior use.

      It was created to serve the market demand for a better medium of exchange (specifically in the hampered market).

      Another way to look at bitcoin is that it is a publicly auditable accounting mechanism which works extremely well for hawala-type transactions.

      Two trading partners can voluntarily agree on an exchange rate (price discovery) and then trade using this trusted accounting mechanism (medium).

      In another sense, you can look at bitcoin as an "opt-in" community currency (much like Ithaca Hours or any other community currency) where trading partners can opt-in or out at any time. People are "opting in" because it is trusted, decentralized, and better than gold in many respects (you can smuggle bitcoin across borders much easier than you can gold).

    4. Mises' definition of money is not synonymous with medium of exchange. He defined money as 'the most marketable commodity'.

      Therefore he is not wrong. Bitcoin is NOT a money yet. It is a commodity with exchange value based on another cause than its monetary function.

      Right now a econ geek toy, a means of indirect exchange for people to buy drugs over the internet and many other quirks, but it is NOT a money.

      Once it is, it will fit nicely with Mises' regression theorem.

  2. Does anybody have any links to libertarians or Austrian schoolers analyzing bitcoins with regard to the Money Regression theorem? When I first heard about bitcoins, my first instinct was to dismiss them because it seemed as though bitcoins were never valued for direct use purposes, so how could they ever acquire exchange value?

    I searched the ECB bitcoin .pdf for the word "regress" and I got 1 hit. After attributing the emergence of bitcoins to Austrian schoolers, they say:

    "Although the theoretical roots of the scheme can be found in the Austrian School of economics, Bitcoin has raised serious concerns among some of today's Austrian economists.

    Their criticism covers two general aspects: a) Bitcoins have no intrinsic value like gold; they ar mere bits stored in a computer; and b) the system fails to satisfy the "Misean Regression Theorem", which explains that money becomes accepted not because of a government decree or social convention, but because it has its roots in a commodity expressing a certain purchasing power."

    The "intrinsic value" aspect doesn't make any sense, but does anybody have a link to any actual Austrian "criticisms" based on bitcoins never being valued 100% for direct use?

    1. See

      Mises' Regression Theorem Is Wrong

    2. He isn't wrong. At least get your basic Mises right before trying to refute the greatest economy in the 20th century with a blog post.

      To Mises money isn't the same thing as 'medium of exchange', to him 'money is the most marketable commodity'. Is Bitcoin that? No.

      It is an econ geek toy, an indirect exchange means to buying illegal shit and many other things.

      Once and if it becomes a money, it will fit Mises' regression theorem nicely.

  3. Thank you for this find! Key reason why I visit your site! Now if I can find the time to actually read the document...

  4. Thank you! Sources like this is a key reason why I visit this site!

  5. Here's another bankster paper sponsored by Citi: http://eprint.iacr.org/2012/584.pdf

  6. Here's an interview on EconTalk with the Principal of Bitcoin, Gavin Andresen:


    1. Correction: Gavin is technically a lead developer of this open source project. He only has as much influence as the greater community will allow.

  7. Wally,

    The Regression Theorem is a possible explanation but not a requirement. Robert and I discussed it on the Show as linked to above. At the end of the day, Bitcoin is functioning as money and currency. If it is incompatible with the Regression Theorem then that Theorem needs to be modified due to a change in observation. And those who did not dismiss Bitcoin earlier but instead embraced the new technology have reaped substantial financial effects as a result of their superior foresight and economic and financial understanding coupled with risk taking.

    But I am not sure that any modifications need to be done to the theorem. Acting man likes to collect things including digital cows, crops, ebooks, awards, etc. Bitcoins were collectible, therefore having Direct Use value, before the pizza exchange which is when bitcoins slinked out of the primordial monetary ooze and begin imparting Exchange Use value. But why an individual values something is actually fairly irrelevant. All that matters is that an individual does value the thing.

    And Bitcoins are things; they are organized information. And the further we move into the Information Age the more wealth is transitioning from the physical to digital. For example, while these questions confound many issues: is the value in silicon and metal or the organized information used to construct and sell iPhones and iPads? What about the Google Search Algorithm? The network effects of Paypal, Visa or Mastercard?

    You may want to watch this interview:


    That Citi sponsored paper is pretty pathetic. Surprised S let his name be put on it. Embarrassing.

  8. Wenzel's recent interview with Trace Mayer (Oct 7, 2012) had some discussion of this. I find it absurd that one would reject Bitcoins as money because they never were never a consumption or production good, as were most forms of money. The Yap Islanders used giant stone coins, yet they never were a useable good, at least I think not.

    Just because most forms of money arose from this path does not mean it is a necessary condition for every form. Bitcoins apparently satisfy all the usual necessary conditions of scarcity, transportability (not so important to those Yap Islanders, though!), divisibility, durability, etc, and now the question is how much adoption will take place in competition with other monies.

    Some adoption has taken place, but there's a long way to go. To the extent the other monies of the world fail at the "store of value" role, and there is enough demand for anonymous, untraceable transactions, Bitcoins will make inroads.

  9. FTA:

    "The total supply of Bitcoins is expected to grow geometrically until it reaches a finite limit of 21 million. If, however, the number of Bitcoin users starts growing exponentially for any reason, and assuming that the velocity of money does not increase proportionally, a long term appreciation of the currency can be expected or, in other words, a depreciation of the prices of goods and services quoted in Bitcoins. People would have a great incentive to hold Bitcoins and delay their consumption, thereby exacerbating the deflationary spiral."

    While falling prices MAY lead people to seek to increase their Bitcoin balances by delaying their consumption, it does not lead to perpetually increasing Bitcoin balances (this is even apart from the fact that the supply of Bitcoins will eventually be fixed).

    For example, a change from 2% annual price inflation to a -2% annual price deflation may lead people to seek to hold, say 10% more cash relative to their assets as valued in Bitcoins, but in no case can it be assumed that people would seek perpetually increasing cash relative to assets. Sure, people may delay their consumption from an average of $X per year to $(X - 10%) per year, but this would be a one time, delimited increase in Bitcoin "hoarding". Once people's Bitcoin to asset ratios are satisfactory, in accordance with the annual -2% price deflation, there is no more reason to seek to hold more Bitcoins.

    This whole "deflationary spiral" idea is a fear mongering assertion, based on either stupidity, or a conscious agenda to politically agitate for retaining fiat hegemony of the state, or both.

    It's remarkable how even "prestigious" institutions like the IMF (don't laugh!) are subject to the same old fallacies that continually rear their ugly heads in economic discourse.

  10. Wow central bankers are already calling for regulation of virtual currencies. They must be scared.

  11. A few thoughts about the regression theorem and bitcoins:

    1. Above, Anthony Freeman says "Bitcoin was primarily created to be used (100% DIRECT USE) as a medium of exchange..." I cite this to show how the gist of Mises' idea can be completely misunderstood. Direct Use value and Exchange value are mutually exclusive. If I like to drink whiskey and also know that for whatever reason I can exchange whiskey for other goods, then I might value a bottle of whiskey partially for direct use and partially for its exchange value. But to say that something is "used (100% Direct Use) as a medium of exchange" shows a complete non-understanding of the meaning of these 2 terms.

    2. I listened to the interview of Trace Mayer by Bob Wenzel. I was pleasantly surprised to hear Wenzel say that the emergence of bitcoins as a medium of exchange seemed to violate the regression theorem. But the discussion veered off on the non-issue of the costs of bitcoin production which has nothing to do with human beings subjectively valuing bitcoins either for direct use or exchange value. So, although I think Bob's instincts are correct, the discussion really didn't get to the heart of the matter.

    3. I think the answer might be found by thinking about what Scott O says above: "The Yap Islanders used giant stone coins, yet they never were a useable good, at least I think not" and then thinking more about the direct use value of gold back at the time when gold transitioned from being valued 100% for direct use to being valued partially for its exchange value. Back at the time of it first being valued for exchange purposes, was gold really any more of a "useable good" or qualitatively different from the Yap Islanders stones or these unique pieces of data sitting at these ip addresses?

    Maybe not. The act of consuming any of these 3 super-durable consumption goods might be the same for all 3: The subjective ends of individual human beings are accomplished merely by possessing or having access to all 3 goods and contemplating their ornamentality--that's all that's necessary.

    So, perhaps that first transaction of 10,000 bitcoins for 1 pizza was not the moment when bitcoins emerged from the primordial ooze of consumption goods and evolved to an exchange commodity as Trace Mayer says in the interview. Maybe that was a 100% barter transaction where the pizza maker valued the 10,000 bitcoins completely for their ornamental direct use value.

    In short, what I'm saying is that perhaps its not so clear how gold was valued 100% for direct use either.