This essay talks about Bitcoin a lot, but it is not about Bitcoin per se, that is merely the technology that first illuminated the problem I’m addressing. Bitcoin could crash and burn tomorrow and it would not change the analysis here: it is a placeholder for any decentralized “virtual” currency. Similarly I will be using Gold as a placeholder for all non-fiat currencies that might be proposed: after all, many people propose switching to a gold standard, no one proposes switching to an iron or coca bean standard despite those having been mediums of exchange at one time or another.
A few years ago when discussion of Bitcoin was all the rage one of the common objections was that Bitcoin is just another fiat currency and doesn’t have any real value. The strong version of this objection is that Gold is Real™ and has Intrinsic Value™ in some bizarre philosophical sense. I am not going to waste pixels knocking down that particular concept when so many others have already done so, if you believe this idea is true then please go study Subjective Value Theory until you stop being silly (some General Semantics would also be advisable).
The weaker version of this objection is that Gold has value to people before its use as currency, giving it an effective “Intrinsic Value”, as opposed to fiat currencies which have no value outside the decrees of the Government Bank. This is mostly true, however due to historical accident the theory which explains how currency comes into existence has rested on assumptions that are only justified under primitive conditions.
The Standard Model of how a currency forms goes something like this: People barter the goods they produce for the ones they want; the butcher trades meat for apples, the carpenter chairs for shoes, etc. But not everyone wants every commodity, or wants it in the quantity that they would get by trading what they have. As a result barter makes trading complicated, with the difficulties increasing rapidly as the society built on it becomes larger.
Eventually someone, let’s say the orchard owner, realizes that he can trade his apples to the miner for gold, and then trade the gold to the butcher who doesn’t like apples. The butcher accepts the gold because he can trade it to the smith for a new knife. Once this process gets started people start thinking of, and valuing gold differently because it now has value as a medium of exchange, as well as the value it had before. Once a medium of exchange gains traction it is an enormous boon to trade: each member of a transaction need only account for two commodities; money and whatever they are buying/selling, rather than the dozens they would have to track in a barter system.
The people objecting that Bitcoin is fiat correctly recognize that “a Bitcoin” has no value to anyone on its own: you may as well write random data to the drive. But they misidentify what it is about fiat currency that makes it fiat. It is not about the actual value of the commodity, but the fact that some entity (almost always a government) has decreed that FooBucks Shall Be The Official Currency. This means that anyone who does business in Fooistan can rely on being able to pay their taxes with FooBucks, and that everyone else in Fooistan will also be willing to take FooBucks for the same reason.
What makes fiat bad is that the same entity which decreed FooBucks to be money, can also change how many FooBucks there are, manipulating the value of the currency at will. Done slowly and subtly this means a slow inflation, with the manipulating entity continuously siphoning off a portion of the economy for its own ends. Done quickly (or when they lose control of the slow method) it can destroy an economy, causing it to revert to a barter system.
Note that I haven’t even mentioned the “intrinsic value” of FooBucks yet; that is because it doesn’t matter. Consider this scenario: Fooistan is actually firmly on the gold standard, and FooBucks are just a specific mass of gold. Then the Fooistan Reserve Bank gets a magic replicator, starts pumping out millions of FooBucks worth of gold every day, and dumps it all on the market. The effects on the economy will be the same as if the bank were printing bales of paper money. But the use value of the gold may actually increase even as its exchange value collapses. This could happen if the abundant supply and low cost allows people to use gold any place they need good corrosion resistance, or electrical conductivity.
Bitcoin is the inverse of Gold FooBucks: it has no use value, but no one can manipulate its value by pulling more Bitcoins out of thin air. Historically the Gold FooBuck scenario never happened (Spanish New World idiocy notwithstanding) because of a severe shortage of replicators. Similarly all currencies not pegged to a stable commodity have relied on the beneficence of some Trusted Authority to not manipulate the currency, and to hunt down counterfeiters.
All of this means that the Standard Model of Currency Formation is wrong in the same way that Newton’s Laws are: they will accurately predict low energy phenomena like the path of an artillery shell, but will wildly mis-predict the trajectory of a ship traveling through the ergosphere of a rotating black hole. Correspondingly, the Standard Model of Currency correctly predicts that under conditions where coordination is difficult and transaction costs are high, a currency must evolve from a pre-existing commodity. But if transaction costs are low (acquire endless gigabytes of data of your choice with only a few keystrokes!), and coordination is easy (download this program to join the network!) the theory must be retrofitted to correctly predict what is.
The reason why the old model did not take this into account is because the pre-existence of a valued commodity chops out a lot of the coordination difficulties, and until recently high coordination and transaction costs were the norm. Because of this, not basing a currency off of a commodity was difficult or impossible without a government to force the issue. Even then, they usually started with a commodity base and later removed that base. But with the wholesale obliteration of transaction costs brought about by modern (as of the early 21st century) communications and computer technology, it is easier to get people to agree to accept a new fangled currency. Especially when they can automagically index their prices in the currency of their choice, and have the new currency exchanged for the old one in real time, which negates the possibility of them getting stuck with junk money over timescales greater than a few minutes.
In this new situation traditional commodity based currencies will continue to carry the value they have always had: if the currency collapses you can still extract a tiny fraction of its former value by consuming the commodity. Maybe, it depends on the commodity. But their ease of coordination advantage will suffer because there isn’t enough efficiency margin to gain over the next best alternatives. If this is true we may have already seen the last commodity based currencies, with the only exceptions I can see being on frontiers where new sources of raw materials are being extracted and normal economic infrastructure has yet to be setup. And of course the ever present low level bartering that exists in any economy.