Saturday, April 26, 2014

Bubbles, bitcoins and the decision irrelevance of non-existent information

What do the following things have in common?
Shells (especially cowries), beads (whether made of stone, shell, glass or whatever), salt (including stamped salt cakes), cloth (from silk to wadmal, a coarse wool fabric), tool metals (iron, copper, tin, bronze) in various shapes, die cakes, gold dust, weighted gold, teeth, feather coils, strings of coconut discs, carved stone (also a tool material), animal skins, cattle, grain (notably barley and rice), pigs, coconuts, buffaloes, seeds, slaves, silver in lumps or shaped, tea (including in bricks), plaited palm-fibre rings, tobacco, beeswax, camphor, porcelain jars, human heads, cocoa beans, balls of rubber, coca leaves, logwood (mahogany) 
They are all things cited in anthropologist A. H. Quiggin's A Survey of Primitive Money: the Origins of Currency (1949) as being used as money. What this great variety of things with varying degrees of "moneyness" demonstrates is how money is a network good. At its simplest, money is something you pay for things with and hold so you can. It value as money depends on expectations that you will be able trade it with other people for goods and services that you want. In other words, that you can be part of a network of transactions, with money being transacted along transaction chains.

The most widely used (across time and space) money item.
Which means that goods can have consumption utility, production utility or transaction utility. The above goods may also have production or consumption utility, but their appearance in that list is because of their transaction utility. With particular goods having high levels of transaction utility in particular contexts--that is, in particular transaction chains. With cowrie shells being by far the most widespread money item in human history, being used for thousands years across several continents.

How do they have that transaction utility? Well, apart from various desirable physical characteristics (durable, divisible, not too heavy nor too bulky, etc) that they may have to greater and lesser extent, they have transaction utility from information about the expected behaviour of other economic agents in transaction chains. Hence the existence of such a wide variety of local monies.

Evolution of money
Hunter-gatherers have no use for money. They live in small groups where they know everyone and engage in repeated, indeed daily, interactions with them. There is constant pooling of resources in a life of connections, not transactions. Barter is something you do with outsiders.

Money grows up when societies get big enough, or connected enough, that people start interacting regularly with a wider range of people to build up expectations about what they value, but not so often as to create a personal connection such that resources are pooled. Instead, people start transacting with people, developing information about what they value. At this point, the well-known disadvantages of barter begin to bite and money evolves. With what becomes used as money being based on local circumstances.

Once any good begins to be used as money, its use as money drives up its value beyond its production or consumption value. If it does not, it ceases to be used as money and is consumed or used in production. Hence the reality of transaction utility. Which is a real utility, as money permits transactions that would otherwise not occur.

Information has to exist to matter
Note that expectations are crucial to the process. Note also that those expectations are entirely based on current information. As a principle of modern physics, there is no information from the future, and economics has no way around that. So, expectations can only be based on what we actually have information about.

At this point, it may be attractive to reason from a regress, particularly for money which has no production or consumption utility, only transaction utility. Such as bitcoins. We know that at some point its transaction utility will terminate. So, since its moneyness depends on expectations of future transaction utility, and it has no value beyond its transaction utility, the knowledge that such utility will certainly terminate should [so the argument goes] destroy any current transaction utility. As econblogger David Glasner puts it:
The problem I have is that bitcoins can’t be used for anything except as a means of payment for something else. Bitcoins provide no real service distinct from being a means of payment. Think about it; if a bitcoin can’t be used for anything except to be given to someone else in exchange, that means that someday, someone is going to be stuck holding a bitcoin with no one left to give it to in exchange. When that happens, that stinky bitcoin won’t be worth a plum (or plugged) nickel, or a red cent. It will be as worthless as a three-dollar bill. 
Now I grant you that that final moment of clarity might not happen for a long time – maybe not even for a very long time. But if anything is certain, it is certain that, sooner or later, such a moment must certainly come. But if it is certain that ultimately no one will accept a bitcoin in exchange, then it follows that no one forseeing that inevitable outcome would accept a bitcoin in exchange prior to that moment unless he or she is confident that there is some sucker out there who will accept in the interim. But since when does a theory of asset valuation premised on the existence of an unlimited supply of suckers count as an acceptable theory? Under the normal rationality assumptions that economists like to use, it is not possible to rationalize a positive price for a bitcoin at any point in its history.
Is the transaction utility of bitcoins weakly anchored, so likely to be highly variable on the basis of new information? Absolutely. So, in that sense, I agree with David Glasner, bitcoins are a "bubble" in the sense of being an asset likely to be highly unstable in value.

Where I disagree is that the regress from the certainty that at some (completely unspecified) time in the future they will cease to be traded has any implications for current decision-making as long as we have no specific information about when that event is likely to occur.

It is certain that, at some stage, homo sapiens will cease to exist. At that point, all assets will have no value. Does the certainty of our future non-existence as a species affect the price of any existing assets? No, because we have no information about when that certain event is likely to actually occur. So, the certainty of our future non-existence becomes part of the background uncertainty that does not affect current decision making because we have no information about it apart from its inevitability. Since we have no information about when it is likely to occur, it is not a risk differentiating one time period from another, so not part of current decision-making, so does not affect the price of any asset.

A point that applies to any particular asset class. Including items with transaction utility, but no consumption or production utility, which will inevitably cease to have transaction utility at some point, but whose timing we have no information about. Not even whether it will happen before all asset classes cease to have any future. Though it is clearly likely to be happen earlier, since much less has to happen for the event to occur. Hence I agree that the transaction utility of bitcoins in particular is weakly anchored, as remarkably little has to occur for them to lose transaction utility.

I also agree that reasoning on the basis of a regress can be very useful. It is why I don't think the concept of a "bubble" is very useful. If the turning points of "bubbles" could be reliably predicted, then bubbles would not occur, as no one would buy the asset at a price to be caught by the drop in price. Volatility in asset prices depends on the reality that turning points cannot be reliably predicted. But if you cannot predict the turning point, how can one usefully talk of a "bubble" beyond broad comments about asset price volatility?

But that regress depends on the implications of not having information. It is not a regress which turns out to require information before we actually have it.

(And if you think the above provides support for the Efficient-market Hypothesis [EMH], at least in its weak version, yep.)


[Cross-posted at Skepticlawyer.]

No comments:

Post a Comment