Friday 15 February 2013

Bitcoin Conference 2013 (San Jose)

The list of panelists for the Bitcoin Conference 2013 (the one in San Jose, California, there's also going to be a second one in November in Vienna) has been announced and I'm one of them. If you're in San Jose between May 17th and 19th, you can meet a lot of people involved in Bitcoin.

Tuesday 12 February 2013

Smiling Dave responds, or not

A short while ago, I made a post explaining the contradictions of Smiling Dave. It didn't take long and Smiling Dave responded. Or at least he seems to think that he responded. Regrettably, his reply fails to address the points that I made in a depth I consider sufficient. He concentrated on portraying Bitcoin in a negative light rather than investigating the issues. He complained that he is "besmirched and ridiculed" and that he's being "attacked".
I would like to address his reply. As the question of Bitcoin is not entirely unrelated to the issues, I will also involve Bitcoin in my response.

Trade between two non-consumption goods

According to Smiling Dave, a trade between two non-consumption goods is:
It’s a direct exchange of a peso for a bitcoin [or for a dollar]. It’s not indirect exchange, because the peso was directly exchanged for the bitcoin [or the dollar], with no intermediary.
So it is a direct exchange then? Mises defines direct exchange in Theory of Money and Credit as following:
Suppose that A and B exchange with each other a number of units of the commodities m and n. A acquires the commodity n because of the use-value that it has for him. He intends to consume it. The same is true of B, who acquires the commodity m for his immediate use. This is a case of direct exchange. [emphasis added]
Mises makes it clear that direct exchange involves both parties to the trade to consume the respective goods of the counterparty. Clearly, this is not the case between peso, Bitcoin or a dollar (only in very rare circumstances, e.g. numismatics). So Mises disagrees with Dave.

In Theory of Money and Credit, Mises differentiates between direct exchange and indirect exchange:
Indirect exchange is distinguished from direct exchange according as a medium is involved or not.
Clearly, in a forex trade, a medium of exchange is involved, so it can't be a direct exchange. Dave argues that the intended purpose of the definition was only for one medium of exchange and does not apply to situations where multiple media of exchange are involved. So that now leaves him with the issue of simultaneously ignoring such transactions, and classifying them as direct exchange, thus contradicting himself. Furthermore, in Human Action Mises writes that:
Interpersonal exchange is called indirect exchange if, between the commodities and services the reciprocal exchange of which is the ultimate end of exchanging, one or several media of exchange are interposed. [emphasis added]
Mises here makes the same argument as I made earlier, that multi-stage exchange using multiple media of exchange is also indirect exchange. Furthemore, the use of phrase "is called" indicates that this is a definition. While Mises' earlier definition from Theory of Money and Credit leaves some room open for interpretation, in Human Action he closes this gap and presents a definition that contradicts Dave's. Whereas Dave only provided the choices of ignoring such exchanges or classifying them as direct exchange, Mises classifies them as indirect exchange.

Non-money medium of exchange

Smiling Dave finally did what I asked him and provided an example of a non-money medium of exchange:
Non money medium of exchange: cigarettes in a prison. You can get pretty much everything the prison community has to offer for cigarettes.
Unfortunately, this is fuzzy. Assuming his assumption ("you can get pretty much everything...") is correct, then the conclusion that most Austrians would probably make is not merely that cigarettes are a medium of exchange, but they are money. If cash circulates alongside cigarettes, then both are money (even though some purists such as Hoppe might disagree and argue that in that case the prison is still in a state of partial barter and money does not exist).
Dave continues:
If 30% of the inmates are cigarette haters, then cigarettes are not yet money. What exact number is the make or break magic number of haters for cigarettes to be or not to be money? Mises says it’s not clear, meaning not known, meaning debatable.
But Dave was earlier adamant that a medium of exchange must be accepted by the whole community. So which is it now? If I had to extract his actual definition, it appears that it would be that

  1. a medium of exchange must be accepted "for anything and everything"
  2. money must be also be accepted "by everyone"

But this does not correspond to the distinction made by Mises. I don't know of any reference by any Austrian who claims that "acceptance for anything and everything" is a defining factor of a medium of exchange. The issue of the definition of indirect exchange mentioned in the previous section emphasises the problem. Multihoming with forex markets makes it possible to use media of exchange indirectly even though the buyer might use a medium of exchange that the seller does not accept. Due to multihoming, I can use, for example, a debit card denominated in euros to buy things from merchants that only accept dollars. Similarly, okpay allows their debit cards to be populated with Bitcoin, thus if considering multihoming, it is already possible to buy almost anything anywhere with Bitcoin.

The requirement of being accepted "for anything and everything" appears to be a pure fabrication of Dave, so I kindly ask him to provide a reference.

In addition to that, I expressed a wish if Dave could provide not only an example of a medium of exchange, but of one that he considers sustainable. With respect to cigarettes in prisons, I found this article from last year:
Though the 2004 ban on smoking in federal prisons has rendered cigarettes virtually worthless as a parallel currency. They’ve largely been replaced by stamps in that capacity.
It's not that important, I just thought I'd mention it for the sake of completeness.

Classes of goods

Dave writes:
There is a fourth class Mises didn’t bother to mention, because it usually has no influence on an economy. I call this class “Stuff Only Fools and Idiots will Bother With.” No offense. Bitcoin is right in there, at the head of the class.
While I can't find a proper quote by Mises, I found this argument by Rothbard:

It is clear that the things that must be exchanged are goods, which will be useful to the receiving party.
Rothbard therefore affirms that if an exchange occurs, that is a proof that the underlying "things" are goods. He explains the conditions for a good in a way that is practically the same as Menger (scarcity, utility, knowledge and control). Which of these utilities does Bitcoin not fulfill? Smiling Dave does not say.

Furthermore, Dave claims that Bitcoin "has no influence on an economy". But this is contradicted by empirical data showing that there are profitable businesses based on Bitcoin. So at best, Dave can say that it is small enough to be ignored. But this again is an arbitrary judgement, with no discernible reasoning or references behind the it. I do not recall any of the Austrian scholars arguing that if a good has a negligible effect on the economy, it should be ignored.

Dave also proposes a challenge for me:
How would he classify the invisible non existent clothing the Emperor paid good money for in the classic story “The Emperor’s new Clothes”, by H. C. Anderson? Wherever he puts those, bitcoin goes right with them.
Coincidentally, I analysed this question in an unpublished draft for my thesis. I propose a fourth class of goods, immaterial goods, as there are such goods that do not conform to the definition of any of the three categories. I define immaterial goods as goods that even not consumed in the physical sense or exchanged, alter our perception of value of goods. The emperor's clothes fit exactly into this category: he "used" them, and this altered the value of his reputation (which is coincidentally also an immaterial good).

However, one thing that is important is that a good can belong to more than one category, depending on the context. Gold, for example, can simultaneously be a consumer good and a medium of exchange. Similarly, Bitcoin can be simultaneously an immaterial good and a medium of exchange, and one possibility of phrasing the regression theorem with respect to Bitcoin is that before Bitcoin was a medium of exchange, it was an immaterial good. But I've already now deviated too much from the flow of the debate. Hopefully at least this demonstrates that unlike Smiling Dave, I do not dodge questions, rather confront them.

Collapse of Bitcoin

Bitcoin will collapse the same way the Ithaca Hour and all the other phony moneys through the years collapsed.
This is a complete non-answer. For starters, Dave argued that Ithaca Hours was a medium of exchange, whereas he argues that Bitcoin cannot become a medium of exchange. Why is it analogous then? Second of all, he does not address my argument (that the proper analogy would be that demand, rather than price, of Bitcoin, would collapse). Last but not least, he did not explain why the demand would sink.

As I tried to explain before (and go into more detail about this in my master's thesis), the demand for a complementary currency is driven by inflation. But as all Austrians know (the best example of this being the Austrian Business Cycle Theory), the increase in demand created by inflation is temporary and must always in the end return to equilibrium. So long term, the only source of sustainability they have is ideology, and that might or might not work well. Unlike the state, complementary currencies can't rely on force, so this factor also falls away.

However, Bitcoin has none of this. The demand for Bitcoin is driven by a reduction of transaction costs. There is nothing unsustainable about a reduction of transaction costs. It can go on forever, or at least until a better contender appears. Which is this better contender? Dave does not say.

Conclusion

Smiling Dave seems to argue backwards. He first assumes that Bitcoin is somehow bad, and then he goes on and builds his definitions based on that assumption. He can't avoid interposing negative remarks on Bitcoin instead of addressing the consistency of his arguments. However, upon a careful analysis, the definitions he uses are not supported by Austrian scholars, and in some cases, directly contradict him. Many of the issues I pointed out to him (e.g. secondary media of exchange) are left unaddressed entirely. Furthermore, he takes the exposing of his contradictions as a personal attack on him. But I don't care about him. I care about arguments. I don't know if Dave will address his contradictions, but that's at best secondary to me. What is of primary importance to me is that I formulate my arguments in a coherent manner. Yes, I'm selfish.
But history has shown that occasionally, this has a beneficial effect on others.

Monday 11 February 2013

The contradictions of Smiling Dave

For quite some time, I have been involved in online arguments with a guy who calls himself Smiling Dave. He blogs on his blog, and posts on Mises.org community forums, and a large part of those posts are about Bitcoin. While I disagree with his conclusions, that's not the reason why I'm interested in the debate. I'm interested because he is unable to provide a coherent reflection of his arguments. He continues to contradict himself, and does not want to address these contradictions. I decided to summarise the problems with his arguments and present them in a compact form.

The point of this article isn't to express my agreement or disagreement with Dave's conclusions or to express my opinion about Bitcoin. I might do that at a separate occasion. Here I only concentrate on explaining the contradictions Dave presents, in order to expose that he has a long way before his conclusions can be given any merit. What follows is an analysis of his contradictions.

Market share

His first contradiction revolves around the market share of Bitcoin (even though he rejects the label). His argument is that a medium of exchange must be, by definition, widely demanded for its medium of exchange functionality. He however also quotes Mises' Human Action:
Money is the thing which serves as the generally accepted and commonly used medium of exchange.
which implies that media of exchange that are not generally accepted and/or commonly used, exist. The best example of Dave's conflation is this passage:
Because money has to be something accepted
1. by a whole community
2. in exchange for anything and everything.
That’s what medium of exchange means.
Here he explicitly equates money and medium of exchange. Afterwards, he analyses these requirements with respect to Bitcoin:
So the day you can be given a paycheck in bitcoins, and then you can use bitcoins to buy everything your heart desires, that’s when bitcoin will be a medium of exchange. Until then it just isn’t.
...
Until the housewife goes out to the mall to shop and can buy everything offered there with bitcoin, bitcoin is not a medium of exchange.

On the other hand, Dave admits that a non-common medium of exchange exists and Bitcoin might fit into this category. He writes:
So bitcoin has been used as a medium of exchange a handful of times. Still not enough to give it the title Medium of Exchange, even if it was used technically a few times in a few isolated transactions.
Dave quotes an article by Timothy D. Terrell:
One of the consequences of the regression theorem is that money must arise from a commodity already in general use. If there is no nonmonetary use for the good, it will not develop the widespread demand that must precede its use as a medium of exchange.
and then writes:
Widespread demand before it can even be a medium of exchange? All it takes is three people willing to use it one time, a la Bob Murphy. Whatever can this adjunct professor with the Mises Institute possibly mean?
Obviously, he understands “medium of exchange” to mean “non trivial medium of exchange”. Which is the way I use it, as well.
Dave here mixes two things. He mixes the definition of a medium of exchange with the application of the regression theorem. But the regression theorem is not a part of the definition of a medium of exchange. Presenting it in the way of Smiling Dave is a circular argument: he assumes the validity of the regression theorem, and from that derives the definition of a medium of exchange. Furthermore, even if we agree with the "hard" interpretation of Terrell, he does not argue that a medium of exchange must be widely demanded. He argues that a good must be widely demanded before it can be used as a medium of exchange. It does not imply that the demand qua medium of exchange is "wide".

Dave simultaneously claims that a medium of exchange is a distinct category from money, yet he cannot explain what distinguishes them. The closest he comes to explaining himself in this respect is:
As you will recall, Mises in HA writes in Chapter 17 that the difference between money and medium of exchange is one of degree [of universality of use], with the possibility of borderline cases.
But Mises does not use the terms "degree" or "universality", at least not in Chapter 17 of Human Action. Rather, he uses the terms "generally accepted" and "commonly used" to distinguish between a medium of exchange and money. Precisely the terms that Dave uses to define a medium of exchange. Smiling Dave's case might be more understandable if he provided an example of non-money a medium of exchange. But he does not. The closest he comes to an example is his article about complementary currencies:
Let’s remember, the Ithaca Hour and all those other phony moneys were media of exchange for a little while, until they collapsed to zero.
The Bitcoin ecosystem dwarfs all the complementary currencies, apart maybe from the WIR Bank. You can probably buy a wider variety of products with Bitcoin than with the majority of the complementary currencies.Yet, for Dave, complementary currencies rank higher on the "medium of exchange scale" than Bitcoin. Again an inexplicable conclusion.

Indirect exchange

The definition of a medium of exchange used by Mises does not conform to the definition of a medium of exchange used by Smiling Dave. Dave does not see a trade of two non-consumable goods against each other as indirect exchange. He writes:
Oh, and one last thing, and a very important one. All the buying and selling of bitcoins for dollars or pesos or other currencies over at mtgox.com and other places are not, repeat not, transactions where bitcoins are media of exchange. Only instances where a person sells his apples in exchange for a bitcoin, and then buys oranges with the bitcoins, count as bitcoin being a medium of exchange. [Look this up if you don't believe me].
[emphasis added] 
What sort of exchange is it then? Mises only provides two options of classifying trades: direct exchange and indirect exchange. As none of the components of the buying and selling referred to by Dave are consumable, it can't be direct exchange either. What it is for Dave is a mystery. Dave does not explain what sort of transaction a trade between two non-consumable goods is. This is not Bitcoin specific, the forex markets are highly liquid and have a large transaction volume. Wikipedia article on Forex claims that:
The average daily turnover in the global foreign exchange and related markets is continuously growing. According to the 2010 Triennial Central Bank Survey, coordinated by the Bank for International Settlements, average daily turnover was US$3.98 trillion in April 2010 (vs $1.7 trillion in 1998).
The forex volume is about 20 times the world GDP. Yet Dave does not provide a method of classifying this huge transaction volume. For him, they can safely be ignored and can be dismissed.

When I argued that a non-negligible proportion of online gambling industry uses Bitcoin, Dave counterargued that:
Even if all gambling is done in bitcoin that does not even make bitcoin a medium of exchange for gamblers, because when they leave the gaming table and go to the mall, they have to use dollars, and cannot use bitcoins.
Yet none of the secondary media of exchange mentioned my Mises or Rothbard conform to this criterion. Mises presents the following list:
  1. Claims against banks, bankers, and savings banks which--although not money-substitutes --are daily maturing or can withdrawn on short notice.
  2. Bonds whose volume and popularity are so great that it is, as a rule, possible to sell moderate quantities of them without depressing the market.
  3. Finally, sometimes even certain especially marketable stocks or even commodities.
Rothbard calls media of exchange that are highly liquid, but not liquid enough to be money, "quasi monies". He writes:
In Oriental countries jewels have traditionally been held as quasi moneys. In advanced countries quasi moneys are usually short-term debts or securities that have a broad market and are readily salable at the highest price the market will yield. Quasi moneys include high-grade debentures, some stocks, and some wholesale commodities. Debentures used as quasi moneys have a higher price than otherwise and therefore a lower interest yield than will accrue on other investments.
Apart from certain commodities, none of the examples listed by Mises or Rothbard are accepted as a payment instrument, and in many cases it is not even technologically possible to use them that way. Yet Dave is adamant:
What counts is only one thing. Can I buy everything I want with bitcoin, or not?
When I pressed Dave to address the secondary media of exchange (as they are rarely, if ever, used to pay for anything), his conclusion was that secondary media of exchange are not really media of exchange. This emphasises his inability to distinguish between media of exchange and money (which I mentioned in the previous section), as according to Mises and Rothbard, these highly liquid goods (secondary media of exchange/quasi monies) are just one step removed from money and they do not envision something in between. You might think that if Dave was right, Mises and Rothbard would present some example that conforms to his interpretation. I do not know of any such example.

Sustainability

Dave simultaneously argues that the regression theorem's conclusion is that Bitcoin is unsustainable, and that Bitcoin cannot be a medium of exchange. But these are two entirely different propositions. On one hand, he quotes several Austrians claiming that a medium of exchange must evolve as described in the theorem, in other words that it is logically impossible for it to evolve in a different manner. But Bitcoin exists. Even if Dave denies it is a medium of exchange, it exists. So, Dave made a twist on the theorem by starting to claim that it doesn't mean that Bitcoin couldn't have evolved, but that it is not sustainable.

He writes that Bitcoin:
It’s not a medium of exchange now.
...
The day it collapses into being valueless, like the Ithaca Hour, it will confirm the regression theorem.
[emphasis added]
Nowhere does the regression theorem say that things that do not conform to the regression theorem must collapse. This interpretation was invented by some "junior" Austrians, such as David Kramer or Melissa Pattison.

Dave also does not understand either complementary currencies, nor how the regression theorem applies to them. Complementary currencies are technically money substitutes, not money in the narrower sense. Their market price is pegged onto money (e.g. USD, EUR, CHF), and liquidity is achieved by a method similar to credit expansion performed by the banks (e.g. loans without interest). Having a price and liquidity, it is logically possible for them to act as media of exchange. Dave admits that complementary currencies are media of exchange (see above), and rather argues that they are unsustainable.

He argues that many of these complementary currencies (e.g. Ithaca Hour, BerkShares, ...) "collapsed" and are "worthless". But this is an erroneous portrayal of their history. Complementary currencies lose market share alright, but they usually don't lose their value. This is also confirmed by the articles about complementary currencies Dave himself references, as none of them indicate that the complementary currencies lost their value. Their value is still pegged to the underlying money. I'm not saying that it is logically impossible for them to lose value, I'm merely pointing out that Dave ascribes this possibility to situations where it didn't happen, and that even if they were somehow analogous to Bitcoin, he still has no evidence that the value of Bitcoin will collapse. At best he argues that the demand for them will collapse. And why should it collapse? He does not explain.

Dave writes:
So bitcoin has been used as a medium of exchange a handful of times. Still not enough to give it the title Medium of Exchange, even if it was used technically a few times in a few isolated transactions.
Similarly, on his article about secondary media of exchange, he argues on one hand that
In other words, if we agree that bitcoin is a secondary medium of exchange, it is not money, and never will be, either.
So here he argues that it the regression theorem's conclusion is that it is logically impossible for Bitcoin to become money. However, just a couple of days prior to that, he argued that:
So there exist secondary media of exchange. So what?
The regression theorem applies to them, as well.
They pass the requirements of the theorem.
and
Sure bitcoin is a secondary medium of exchange in almost all the transactions it’s used in.
...
Do not satisfy my reasons but satisfy regression theorem = secondary medium of exchange.
I.e. he simultaneously claims that Bitcoin is a secondary medium of exchange, that it does not conform to the regression theorem, and that secondary media of exchange do conform to the regression theorem. Out of these three statements, only two can be simultaneously valid. Furthermore, if it already is a secondary medium of exchange, then why must it collapse? It does not follow irrespective of whether secondary media of exchange do, or do not, conform to the regression theorem.

Conclusion

Smiling Dave presents a hopelessly confused "theory". He simultaneously presents three pairs of contradictory claims:
  1. Medium of exchange is distinct from money, but on the other hand there is no discernible differentiating factor between them
  2. Medium of exchange must be used as a payment mechanism, yet he presents no example of such a medium of exchange (technically he contradicts Mises and Rothbard rather than himself)
  3. He alternates between arguing that according to the regression theorem, Bitcoin cannot be a medium of exchange, and that Bitcoin's value is unsustainable
It is clear that the problems of Dave's argument are deep and fundamental. His goal, apparently, is to somehow "disprove" Bitcoin. But the goal is irrelevant. Even if his intention was exactly the opposite, or on the other hand, if I agreed with his conclusions, that still does not resolve any of the contradictions in his arguments. Dave needs to face the contradictions and resolve them. My recommendation would be to start with the basic gaps:

  1. Explaining why a trade between two non-consumption goods is not indirect exchange, and provide a method for classifying such trades (as according to Mises, the only two options are direct and indirect exchange)
  2. Providing a clear distinction between a non-money medium of exchange and money, and providing examples of such non-money media of exchange (preferably ones that he considers sustainable)
  3. Explaining how Bitcoin is to be classified (as according to Mises, the only three options are consumer good, producer good and medium of exchange)
  4. And last but not least, as a bonus, he can try to describe the process by which Bitcoin will collapse
My expectations of high level of coherence from a random guy on the internet may be pedantic and unrealistic. But that's just how I am. For me, it is irrelevant who makes an argument, rather it is the logical consistency of the argument that I strive for. I also do not view my personal opinion about the conclusion as relevant. Indeed, if Dave's conclusion was correct and Bitcoin really was to collapse, we should know about it, and we need to understand the reasons for it so that we may adapt our future action accordingly. But at the present level, Smiling Dave's arguments make it impossible to draw any conclusion. Maybe someday, they will.

Friday 1 February 2013

Economics of Bitcoin discussed at UT Mises Circle

On January 28th 2013, UT Mises Circle discussed my master's thesis "Economics of Bitcoin" and I was invited as a guest. The meeting was conducted over Google+ Hangout, and the recording has now been made available on youtube. If you're interested in finding out about Bitcoin and hear some comments about my thesis, check it out below.