The Regression Theorem
From the standpoint of the Austrian School of economics the most puzzling aspect of the Bitcoin network is its relationship with the regression theorem, the Austrian economists' explanation of the origin of money.1 The regression theorem addresses two separate issues: the emergence of a medium of exchange, and the evolution of a medium of exchange into money. A medium of exchange is defined to be something that is bought for the purpose of trading it for something else later. Money is the medium of exchange which is pre-eminent among all others.
Bitcoin is not and never has been money, so only the first issue is relevant today. However, since the second issue may be important to Bitcoin's future, it is worth describing. The evolution of a medium of exchange into money is explained by the observation that a more marketable medium of exchange is, all things being equal, more desirable than a less marketable one. Thus, the most marketable medium of exchange among several media will have an advantage over the others. As more people choose that medium because of its greater marketability, it becomes even more marketable and the others become less marketable. Thus, it is to be expected in the long run that a single medium will tend to dominate the rest.
As to the first issue, the explanation depends upon the observation that in order for something to be sold it must be treated as an economic good by both buyer and seller—meaning that they both would want some of it, even if they each are willing to pay different prices. If the buyer did not treat it as a good, then he would not be willing to pay for it, and if the seller did not treat it as a good, then he would be willing to give it away, or even pay to have it taken away. Since a medium of exchange is something which is bought and then resold, it must be treated as an economic good by both parties in both transactions.
Thus, the first person who is able to use a good as a medium of exchange must buy it from someone who also treats it as a good—yet who did not use it as a medium of exchange. Hence it follows that a thing becomes a medium of exchange on the basis of some other demand, and no other evolution is possible.
To recap, the regression theorem says that:
- To become a medium of exchange, a thing must be demanded for other reasons.
- Among a set of competing media of exchanges, one tends to emerge as money.
Both of these are necessarily market phenomena because they take place as peoples' individual decisions as to what they wish to trade with and with whom. Institutionalized coercion can affect the process, but always from the outside: for example, a State could try to promote one media of exchange over others by outlawing the rest. This might affect which media of exchange becomes money, but they cannot cause a thing to become money or a medium of exchange directly through decree.
On the other hand, the regression theorem does not say that:
- Something must be widely demanded to become a medium of exchange. It must begin its use as a medium of exchange with people who demand it, but from there it can spread to more people.
- Something must be strongly demanded to become a medium of exchange. Praxeology could not make a statement like this because subjective value theory does not allow a distinction between a strong or a weak desire. Desires can be compared to one another, but not characterized absolutely.
- The future of a medium of exchange is dependent upon its origins. Some origins are possible for a medium of exchange and others are not, but once it exists its origins do not need to haunt it: its future demand depends on its present demand and upon other present factors affecting it, not directly upon anything that happened in the past.
- A medium of exchange not adhering to the regression theorem is possible. Rather, any medium of exchange which apparently does not adhere to the regression theorem will, under closer examination, be found to be completely in accordance with it.2
These fallacies underpin the negative reactions many Austrian economists have to the Bitcoin network's currency, bitcoins.
The Bitcoin Paradox
It is now possible to explain the paradox of Bitcoin and the reasons that it has often been analyzed inadequately by Austrians. In order to analyze bitcoins properly, it is necessary to use only the cold, hard logic of the regression theorem rather than let ones love of gold interfere.
The Bitcoin paradox is that bitcoins appear to have no use other than as a medium of exchange. Its technological features, though elegant, are barren as long as there is nothing that bitcoins can buy. Thus, it would appear to be a violation of the regression theorem if bitcoins were to become a medium of exchange.
There are two fallacies inherent in the statement of this apparent paradox. The first is an argument from lack of imagination. Just because the nature of bitcoins' original value is unclear does not mean that there isn't one. The second is a violation of subjectivism, which is fundamental to Austrian methodology. The economist need not demand to understand the reasons why people value anything—demand is proven by the fact that something has a price, not by the fact that the economist understands why people pay for it. I don't deny that a lottery ticket is an economic good even though I can't understand why anyone would buy them.
Bitcoins are known to be a medium of exchange today. This proves that the regression theorem must apply to them even if it is hard to understand the original demand.3] It is also empirically known that they were sold for dollars before ever being used as a medium of exchange.4 This confirms what must necessarily have been true according to the regression theorem.
The correct approach should have been clear to any Austrian economist, but until recently, Austrian analyses of Bitcoin have been superficial. It is not yet generally understood among Austrians that Bitcoin is fundamentally different from both gold and fiat currencies, and therefore requires a fundamental analysis going back to first principles. This may have to be reiterated a few more times before the Austrian movement is convinced.
The anti-Bitcoin Austrians are incapable of mounting a reasonable case against Bitcoin. They are convinced that something must be wrong with Bitcoin, but when they attempt to articulate it, they arrive at conclusions which are either subjective or fallacious.5 The most common reaction is that the regression theorem implies bitcoins cannot become money or are somehow unsustainable,6 but this conclusion misunderstands the nature of praxeology. Praxeological arguments are only capable of saying that certain causal relationships are possible or impossible. It is not possible for something in real life to violate a praxeological law, even momentarily.
I believe that this reaction has to do with a misattribution of the reasons that fiat currencies (and bitcoins by analogy) are unsustainable, as compared with the reasons that gold is superior. Gold has obvious productive uses; dollars and bitcoins do not. However, gold is not stable and the dollar unstable for those reasons. Rather, the dollar is unstable because the organization issuing it is presently tampering with it. If the tampering should stop and the government should provide real evidence that it will manage the dollar responsibly, then there will be no reason to expect the dollar to be unstable after that.
On the other hand, because gold has obvious and widespread productive uses, its price cannot go to zero as long as it still has these uses. The same argument cannot be made of bitcoins, but it does not follow that bitcoins' value will go to zero or is likely to go to zero: once again, the best argument is provided by Šurda, who asks "If Bitcoin fails, what would replace it?"7 As long as Bitcoin has uses which are impossible with any other currency and as long as it remains competitive against other currencies and payment systems, it will not entirely collapse in value.
Bitcoins are a puzzle to resolve, not an excuse to deny reality. They must have had an original value. The critical moment for bitcoins was when they were first sold. Why did people begin to pay dollars for additional bitcoins beyond those that could be obtained for free? This is not a praxeological question, but it is most definitely an interesting one, and it also is at the heart of Austrians' confusions about Bitcoin. It is clear that bitcoins cannot be thought of simply in analogy with gold, and some other story must be given for it before it will seem natural to people who are used to thinking of money in terms of gold and fiat.
In general it is not always possible to explain peoples' desire very well because they are so subjective. However, a good can be explained as an instrumental, as opposed to an intrinsic good. There is no explaining human fashion, but given that it exists, there is no difficulty in explaining fashion shows or designer clothes as means of advancing peoples' quest to stay ahead of the latest trends. It is not enough, therefore, say that they were "cool" or had "mystique".8 This explains nothing more than what we already know. Ordinarily, things are cool or mysterious for reasons, not because of a sudden mass delusion (although this is not impossible).
An explanation suggested by Graf is that bitcoins were originally demanded because of an appreciation of Bitcoin's engineering. A well-engineered encryption scheme "may indeed be more highly valued to some people in some contexts than certain 'real' economic objects or specific quantities of fiat money. Regardless of any potential future indirect-exchange value, one can imagine such persons expending hundreds of hours of effort in creating and breaking encryption codes, just because they like to".9
The problem with this explanation is that it does not distinguish between Bitcoin the program and bitcoins the currency. An appreciation of Bitcoin's engineering explains why someone would download the program to play with it or peruse the code, but it does not explain why someone would pay for an inert and arbitrary string of data. Any amount beyond a few satoshis would be enough to try out everything that Bitcoin could do. Why, indeed, would an engineer prefer to engage with the Bitcoin block chain when he could start his own and have the entire system to himself to experiment with?
The correct approach, I think, was hinted at by Šurda, who obliquely says, "According to my opinion, the rational expectations of the potential utility of Bitcoin for the potential buyers exceeded the price demanded by the producers, and trade emerged".10 Bitcoins would have had value to the person with the right entrepreneurial mindset.
The Parable of the Futurist
Two centuries ago, oil was less than worthless. It was a blight that people would pay to have removed from their land. However, suppose one day a mysterious stranger appeared claiming to be a time-traveler. He told everyone that one day oil would be the world's most valuable fuel. It would called "black gold". Wars would be fought over it and enormous machines would be built to collect every last drop.
Though no one knew if the purported time traveler was telling the truth or how to develop the technologies that would one day make oil valuable, some entrepreneurs believed him and began to collect oil to prepare for the day that it became black gold. They built large vats to store it and bought it up by the ton. Because these entrepreneurs established a baseline for the value of oil, other people bought smaller amounts to speculate on the price to store their wealth outside of traditional banks.
After a financial panic, the greenbacks people had been trading with became worthless. Needing something to trade with, and finding oil to be a commodity in widespread demand, people begin to trade and quote prices in it. Soon oil becomes their new money without anyone having figured out how to use it as a practical fuel. By the time the technologies to use it as fuel are understood, oil is too valuable to be burnt up.
In real life, there is no time traveler but entrepreneurs can sometimes glimpse the future. When Bitcoin was first invented, bitcoins had no exchange value and were given away free just to generate interest. However, once the right entrepreneurs began to suspect that bitcoins might actually be used as money some day, they were willing to pay dollars to have larger amounts than were available for free. They may not have understood precisely how it could happen, but actually thinking it might would certainly be enough to give bitcoins a mystique. Thank goodness there were enough people whose judgment was not clouded with an imperfect understanding of the regression theorem. A suspicion that Bitcoin might one day be a big deal explains everything about its original demand.
This explains how an appreciation of Bitcoin's engineering could lead someone to want the coins themselves. Once it was known that bitcoin could be sold, even for a pittance, new possibilities opened up. A second generation of entrepreneurs realized that the tiny baseline value provided by the first generation allowed for certain services which would be impossible by with PayPal or credit cards. For example, WikiLeaks could accept donations in bitcoin which could not be blocked by any third party. The Silk Road could be created to establish an anonymous and secure online black market.
At that point, bitcoins had developed an exchange value. Both groups of entrepreneurs were happy because both had turned a profit. Bitcoin was still not money, but its potential was far easier for others to grasp. Thus, new entrepreneurs entered the market, both as merchants and investors. This cycle repeated itself to this day, and there is no reason that it will not continue as long as there are people ready to stake money on the vision of a Bitcoin economy.
The regression theorem is credited to Ludwig von Mises in Mises, L., The Theory of Money and Credit, Yale University Press, 1953 but Carl Menger nearly had it all figured it out in Menger, C., Principles of Economics, Ludwig von Mises Institute, 2007. ↩
Please see Šurda, P., "Economics of Bitcoin: is Bitcoin an Alternative to Fiat Currencies and Gold?", 2012, pp. 38-41 for Šurda's discussion of the regression theorem, upon which I have relied quite heavily. ↩
This argument has emphasized by other pro-Bitcoin Austrians. See Šurda, 2012, Graf, K., "Bitcoins, the Regression Theorem, and that Curious but Unthreatening Empirical World", 23 Feb 2013, and Selgin, G., "Bitcoin", 22 Apr 2013. ↩
See Gertchev, N., "The Moneyness of Bitcoins", Mises Daily, 14 Apr 2013 for an article that gives subjective arguments against Bitcoin. See Shostak, F., "The Bitcoin Money Myth", Mises Daily, 17 Apr 2013 for an article which falls into numerous fallacies in an attempt to refute Bitcoin. This article is so sloppy that it could not even manage to spell "Nakamoto" correctly. Though it is an embarrassment, is useful as a compendium of fallacies and as an indication of the level of disdain and ignorance with which some Austrians discuss Bitcoin. I will briefly discuss its problems.
First, Shostak describes bitcoins as a medium of exchange but then argues that the regression theorem says that it cannot become money because it has no intrinsic value. However, the origin of a medium of exchange has no bearing on the reasons that a medium of exchange becomes money.
Second, Shostak says that bitcoins are not physical, therefore not a commodity, and therefore, since only a commodity can emerge as money, bitcoins cannot become money. This wrongly introduces physicality, which is not a praxeological category, into the discussion. The only relevant praxeological category is scarcity, which bitcoins have despite not being physical.
Third, Shostak advances the fallacy that money serves as a "yardstick". It is true that using money tends to make the exchange rates of goods obey a transitive relationship to one another, but it is not a measure of anything because it is not independent of the rest of the economy and can fluctuate in demand just like anything else.
Fourth, he then goes on to say that "Bitcoin can function only as long as individuals know that they can convert it into fiat money" and "Bitcoin is not a new form of money that replaces previous forms, but rather a new way of employing existent money in transactions". These statements are both obviously ridiculous. He apparently regards fiat currencies as somehow the real yardstick because of their historical ties to gold and therefore simply will not take Bitcoin seriously. Shostak is guilty of historicism. ↩
See Pattison, M., "Buying into Bitcoin: An Austrian Analysis of the Virtual Currency's Sustainability", 14 Dec 2011 for the most detailed article which makes this argument. ↩