The flawed foundations of Austrian economics

[Mengerian]

In my opinion, Bitcoin is completely compatible with the Regression Theorem. I don't see what all the fuss is about.

The Regression Theorem is just an explanation of how it is possible for people to assign value to something that no-one intends to actually use directly, everyone is valuing it for exchange. It explains how people can arrive at a subjective value for the unit being exchanged by considering the process in time. People can observe the value that others traded the unit for in the past, then use that as a basis for expectation of what others will trade it for in future, with allowance for changes they foresee, and their own time preference.

The connection to commodity use value is not an important part of the theory. It's just the realization that the regression back in time can't go back forever, so at some point in the past the exchange unit had to be valued based on something other than past exchange values. This initial value could be anything, even a trivial value as a trinket, it doesn't matter. It doesn't need to be an important use, or a high value item.

Once exchange is established, the people deciding how to value the unit don't need to consider end uses anymore, and the value they assign can evolve over time based solely on observed exchange values and expectations of future factors affecting future exchange values.

 

[cypherdoc]

i remember debating the Regression Theorum back a few years ago with Pierre Rochard on Twitter when Jeffrey Tucker and Robert Murphy were trying to sort thru this apparent dilemma. TBH, i don't even remember what i said back then but it appears to me that Bitcoin has discredited it. i'm sure @Melbustus could help jog my memory. reason being, Bitcoin was designed, implemented, & traded as money from the get go based on unique properties that had never been seen before in human history. all possible b/c it is digital and rides a supposed free and open internet. just read the WP as evidence. there never was an alternative function or usage upon which it could be valued. this is actually why i have a problem with it's current definition as a commodity as ordained by our various gvt agencies. i think it is a currency/money.

 

[Erdogan]

The best strategy for revamping the regression theorem, would be to consider bitcoin a succession of fiats. A new fiat can be created directly (without first being backed and redeemable), because people have experienced fiats and know that they can work. I am sure that thought was also the minds of the enthusiasts, while the value was still nothing. It went on for years, but they knew it could become valuable. It is not obvious that people thousands of years ago, before any fiat (and written debt obligations, that looks suspiciously like money) would think that. Maybe they needed the stepwise approach.

What is this talk about bitcoin not being money? Does it have intrinsic value, does it support life on its own, can you eat it? No. Does it require someone to pay you some other money in the future, depending on their ability and will? No. Does it guarantee something valuable for support of life later? No. Therefore it is money. Its value is market based, you get something for it depending on the one you trade with, his preferences based on his valuations. You have to speculate that you can get something useful in exchange for them in the future. It's money.

 

[Erdogan]

@Mengerian

Good stuff.

I think we also have learned, that a completely new money design needed a shakeup, a rattle, an ignition, to establish a value above nothing, like the well communicated pizza trade. At that point, the market kicked in.

 

[Zangelbert Bingledack]

Bitcoin is not money in the Austrian sense. The Austrian definition of money could use revising, as it only covers one thing that can be used to serve the purpose we expect money to serve.

In any case, Bitcoin doesn't violate the regression theorem, as all the posts I quoted agree.

 

[Erdogan]

I strongly disagree, it is money in the Austrian sense, as I explained above.

 

[Zangelbert Bingledack]

I think we're just interpreting the definition in different ways. You seem to be saying the Austrian definition as written ("the most saleable commodity"), smartly interpreted to include non-physical goods, applies to Bitcoin.

I agree, and likely Menger/Mises/Rothbard would have interpreted it that way if they had encountered things like Bitcoin, but I think the word choice makes it clear they didn't include that intent. (Commodity mainly means: "a raw material or agricultural product that can be bought and sold, such as copper or coffee.") This is no discredit to these great figures, but it does mean that many Austrian writings have probably fallen into speaking with the assumption of money as a physical good. This is thus an important distinction to draw.

Now if you prefer the "medium of indirect exchange" definition of money, we probably do disagree, as Bitcoin is definitely not a medium of indirect exchange in the usual sense in which "medium" is used. But then, neither is gold (mostly). The money-as-medium-of-indirect-exchange story is a broken aspect of Austrian economics that should be jettisoned (I can explain why this is the case if there is interest).

 

[Erdogan]

Well, we see that a lot of traditional austrians do not include bitcoin, so you have a point there. But I think bitcoin can be regarded as tangible, do you remember the bitcoin company Tangible Cryptography, now deceased? The secret code needs to be in tangible form at least while the key is used. (The copyright people define tangible in a way practical for that purpose, but we don't have to follow their definition). Else, the need for the money to be tangible could also be just the result of looking on the world and seeing that all forms of real money were tangible at the time.

These considerations is the reason, in my mind, why the traditional austrian finance people are so slow to fathom bitcoin. But it will be clear to them as time passes, because bitcoin will behave exactly how you expect good money to behave.

To comment also on your last point: Medium of exchange is in my mind the practical effect of using money and to exchange something of life-supporting value for money with only speculative value. It is practical, because it reduces the problem of valuation of tradable stuff in half. We don't really need that function to be a part of our definition of money, we need only the store of value and the liquidity.

If this makes me a non-austrian, so be it. I try to use good language to help spread information on bitcoin (and to clear my own mind), and if we adjust the austrian theory based on our knowledge of bitcoin, the sound money with nil direct use value, it is even better.

 

[Mengerian]

Early Austrian writers, like Menger, used the word "commodity" to refer broadly to any goods traded in the market, beyond what modern use of the term commodity would apply to.

And anyway, in the Austrian framework, what matters for economic classification is how people perceive and treat the good, rather than its physical properties. The physical properties only matter insofar as they affect the characteristics that people care about.

So if people treat Bitcoin as a physical token that they are trading, then it should be treated economically as a physical token. The physical reality of how Bitcoin works as a decentralized ledger with cryptographically verified transactions with inputs and outputs, nakamoto consensus through proof of work, etc. only matter (in the economic analysis) to the extent that they affect the properties that people care about, like transferability, durability, and so on.

Regarding the Regression Theorem, I think many people have over exaggerated the importance of the connection to commodity value because they were trying to use it as an argument for gold over Fiat money. If you go back and read Mises' Theory of Money and Credit, it seems obvious that the historical connection of the value of money to commodity value is basically a minor detail to tie up a loose end in the theory. All it says is that the first time that good is ever traded, there needs to some basis to decide it's value other than looking at past exchange value (obviously, since there are no past exchanges). So for something like Bitcoin, which has no obvious use other than exchange, it would be reasonable to expect that the value established for the first trades would be very low, essentially arbitrary, and treated like a trinket or novelty. And this seems to align with what actually happened.

 

[Zangelbert Bingledack]

That is accurate, and in fact I have raised this point with Austrian Bitcoin-skeptics before, that it's all about perception. I just don't think most Austrians generally "take seriously" that praxeological point; they tend to compartmentalize it. Here's Hayek taking it seriously, complete with what I above called a "smartly interpreted" definition of commodity:

That the objects of economic activity cannot be defined in objective terms but only with reference to a human purpose goes without saying. Neither a “commodity” or an “economic good,” nor “food” or “money,” can be defined in physical terms... . Economic theory has nothing to say about the little round disks of metal as which an objective or materialist view might try to define money....Nor could we distinguish in physical terms whether two men barter or exchange or whether they are playing some game or performing some ritual. Unless we can understand what the acting people mean by their actions any attempt to explain them, that is, to subsume them under rules...is bound to fail.

 

[Roger_Murdock]

In my opinion, Bitcoin is completely compatible with the Regression Theorem. I don't see what all the fuss is about.

The Regression Theorem is just an explanation of how it is possible for people to assign value to something that no-one intends to actually use directly, everyone is valuing it for exchange.

Exactly, it's one possible answer to this sort of chicken-and-egg problem of how something can acquire exchange usefulness. Bitcoin is tricky because it doesn't quite seem to fit this story. And so the question becomes: if Bitcoin didn't initially have any non-exchange usefulness, why would anyone be the first to value it for its exchange usefulness in light of the fact that exchange usefulness requires a network effect? Daniel Krawisz provides a nice "crystallization" analogy:

Currencies are unusual in that their usefulness as currencies is a result of their demand. The more deals that can potentially be made with one, the better it is. It is only indirectly because of a currency’s inherent properties that it can turn into a good currency. To an alien with no interest in human culture or technology, both dollars and bitcoins would be equally worthless, even though he would be able to see that bitcoins are a superior medium of exchange.

In economics, this is often called the network effect because it is the network of people using it that convinces everyone that the good has value. This can appear to lead to circular reasoning: everyone believes it has value because everyone else believes it has value because everyone else believes it has value… and so on. It is almost like a cartoon monkey that picks itself up by the end of its own tail!

However, a better physical metaphor might be the formation of crystals. A liquid can be supercooled into a state that can support crystals, but no crystals form because there are no initial crystals that the liquid particles can attach to. However, once a tiny impurity, or “seed” is introduced into the liquid, a crystal will form around it and grow quickly until the liquid has been absorbed.

By this metaphor, the supercooled liquid is the world as of a few years ago, ripe for a monetary revolution, and the seed is the initial bitcoin purchase.

In other words, the seed was speculation. Exchange usefulness is speculative / forward-looking. All it took to get the ball rolling was a few visionaries who recognized that Bitcoin's fundamental properties gave it the potential to, over time, become valued by others for its exchange usefulness, which led them to place immediate value on it. In other words, they recognized the potential for a virtuous cycle whereby greater adoption would give rise to greater utility, leading to greater adoption, leading to greater utility, etc., etc.

 

[Mengerian]

@Roger_Murdock
"All it took to get the ball rolling was a few visionaries who recognized that Bitcoin's fundamental properties gave it the potential to, over time, become valued by others for its exchange usefulness, which led them to place immediate value on it."

Agreed, This this is plausible.

But it is a very interesting question how the first Bitcoin exchangers determine what initial value to assign to the unit. All the Regression Theorem says is that this initial value had to be based on something other than just previously observed exchange values. The initial demand could be based on something as trivial as "having fun". For example, many people spend countless hours collecting in-game currencies like World of Warcraft Gold. So I see no reason why initial Bitcoin acquirers couldn't have been motivated by similarly trivial reasons. This means the initial value could have been based on some combination of the amount of leisure time people felt like spending on mining, and an incremental electricity cost that was small enough to seem trivial.

Krawisz's speculation explanation could be true, but it's also possible that initial value was based on more frivolous reasons.

 

[cypherdoc]

@Mengerian

how do you think Lazlo's initial 10000BTC/ two pizzas tx fits into the picture?

 

[Mengerian]

I think Lazlo's pizza transaction fits well with what we have been describing.

The fact that he went out of his way to set up the exchange shows that he was anticipating future use of Bitcoin as a medium of exchange. So that fits with the speculation/crystallization concept.

At the same time, it seems that there was an element of fun and frivolity to the transaction, so part of his motivation was also probably fun and amusement.

When people heard about the transaction, the psychological precedent likely added support to people's perceptions of Bitcoin's value as an item of exchange. It was one step along the regression ladder towards monetary value.

 

[cypherdoc]

Assuming you agree that the pizza tx was Bitcoin's first, why is it so taboo to think that the author of the regression theorem just couldn't conceive of a digital currency designed for an Internet?

Its not like anyone could disqualify all his other great contributions to humanity.

 

[lunar]

Exactly, it's one possible answer to this sort of chicken-and-egg problem of how something can acquire exchange usefulness. Bitcoin is tricky because it doesn't quite seem to fit this story. And so the question becomes: if Bitcoin didn't initially have any non-exchange usefulness, why would anyone be the first to value it for its exchange usefulness in light of the fact that exchange usefulness requires a network effect?

I'm not convinced travelling back in time to the big bang, or spontaneous first point of value is entirely meaningful. It strikes me as similar to the resolution point in philosophy Cogito ergo sum "I think, therefore I am" where Descartes asserted that the very act of doubting one's own existence served—at minimum—as proof of the reality of one's own mind. Similar to the first spoken words of communication. Was the first word spoken "hello world?". Obviously not. it's in continual flux or evolution from the previous form or state of communication. Prior to Descartes philosophy was in a vicious circle, this point of clarity even though it can't be proved, was necessary as a reference frame to advance the debate.

My point is to answer when the first point bitcoin has value you would need to consider why data, words or unique combinations of numbers have value. Even the simple act of saying 'I owe you one' has some value (it is a verbal token of value communication) dependant on the trust between the two parties. Digital tokens had value long before bitcoin came along, The 'Warglaives of Azzinoth' have value to some, pokemon to others and linden dollars to more. Bitcoin is simply better than these, as it was the closest to perfect money the world had yet created. (we can argue the form perfect money should take - another debate).

Bitcoin had some value the second the first block was mined but the free market had not yet assigned value accordingly. Bitcoins value has spontaneously emerged through consensus on the properties and present/future utility of the network.

my tuppence

 

[cypherdoc]

The Regression Theorem is just an explanation of how it is possible for people to assign value to something that no-one intends to actually use directly, everyone is valuing it for exchange.

But there was clearly intent by Satoshi from the get go. Just read the WP. Speculation from guys like me and @Roger_Murdock was undoubtedly the main driver of initial adoption as hardly anyone was actually using it yet the price kept going up as they accumulated. Even the term "speculation" is a bit muddy as " speculators" eventually become "users" once the value goes high enough to satisfy their preference.

 

[Erdogan]

I think the rationale behind the regression theorem, is that since money has no reference to the real world, it is impossible to value them without prior knowledge. Any time you go to the market, you have some idea of what cheap and expensive is, and that is based on prior prices. If you don't have that, if you wake up from a hundred year coma with a few deutsche mark (to use a unit that have to be unknown to you), there is no way to know the value of it. So if you need prior prices, that must always have been true, and you have the regression theorem.

But in light of bitcoin, the whole theorem seems not so important. The early adopters saw the good money qualities and therefore assigned some unknown random value to it in their minds. Then, as trade started, prices were revealed. The Lazlo moment was a high profile trade, and it communicated that price to everybody. We had a starting point, and the market developed.

 

[tar]

I think the rationale behind the regression theorem, is that since money has no reference to the real world, it is impossible to value them without prior knowledge.

That is a general misinterpretation of the foundation of money and when why and how it emerged.

The foundation of money (credit) is power. With the first claim of tribute the reference is there. Whatever is claimed the subjects have to provide it on the fixed date(!) or they will face penalty. This is the reference of power: force and violence - the foundation of money, markets and prices.

 

[Erdogan]

Nah.