Building a Better Bitcoin
How much is a bitcoin worth?
Well, it's worth whatever somebody will pay for a unit of the online currency, which as I write this is $209, up from $142 last Friday, $44 a month ago, and $4.93 a year ago. This huge run-up — the latest spike began with EU's botched rescue of Cyprus's banks — has led to much talk of a bitcoin bubble.
The word "bubble" has been greatly overused in recent years. My understanding is that you're in a bubble when the price of an asset becomes completely detached from its intrinsic value. It's a bubble when the price you pay for share of stock cannot conceivably be recouped from the earnings of the company (this was Cisco in 2000) or the price you pay for house cannot conceivably be recouped from rental earnings (this was Phoenix in 2005). The only way you can avoid losing money on your investment is for a greater fool to come along — in the case of real estate, a greater fool backed by an even-greater-fool lender — and take the asset off your hands.
Bitcoins have no intrinsic value. They lay claim to no stream of future earnings. A price of $198 per bitcoin is surely not justified by the fundamentals. But neither is a price of 10 cents. There are no fundamentals.
So as an asset, Bitcoin (I'm trying to follow Maria Bustillos' rule of capitalizing the system but lowercasing the coins) is clearly in a bubble, and always has been. But maybe asset pricing is the wrong lens to be looking through here. A dollar bill lays claim to no stream of future earnings, yet nobody says there's a "dollar bubble" because somebody's willing to give you a candy bar for one. This even though a dollar is almost certain to buy you less in a few years than it does now. According to the Bureau of Labor Statistics, a 2013 dollar has one-tenth the purchasing power of the 1950 version.
By contrast, bitcoins have been skyrocketing in value. This sounds like a good thing, but for a currency it's really not. An economy where bitcoins were the means of exchange would have experienced 98% deflation over the past year. No one would be able to repay any loans, or really do business at all. What we want out of a currency is not price appreciation but stability. Monetary economists differ on whether the optimal stability is inflation of 0% or in the low single digits. Nobody thinks 98% deflation is healthy, and all but a small minority seem to think any deflation at all is a bad thing.
So ... bitcoins are without intrinsic value as assets, yet they have risen too fast in value to be much use as a currency. Kind of makes your head hurt, doesn't it? But it also sounds a bit like a familiar commodity, gold, that's also been on a roll, with its dollar price quintupling over the past decade. Gold has, over time, not been the greatest of assets to invest in. It's not the greatest of currencies, either: Back when the gold standard was widely adhered to, nations struggled regularly with deflation. There's persuasive evidence that the primary cause of the Great Depression was a refusal to unlink currencies from gold until too late. Still, gold has held onto its purchasing power over time. It remains something that people turn to in times of financial uncertainty such as now. And while there are skeptics these days who talk of a gold "bubble," they don't really mean it. That is, they may expect the price of gold to decline from the current $1,575 an ounce, but they don't expect it to suddenly lose most of its value, as assets tend to do when real bubbles burst.
There are some key differences between gold and bitcoins: Gold is a shiny metal that can be made into jewelry, electronic components, and dental fillings — meaning it has some intrinsic value, albeit nowhere near $1,575 a troy ounce. Bitcoins are made of otherwise valueless digits. And while mankind has treated gold as a store of value for millennia, bitcoins were first unleashed upon the world in January 2009, by a mysterious and pseudonymous cryptographer (or cryptographers).
But there are important similarities. Both bitcoins and gold are pretty much impossible to counterfeit. (That is, whatever fakes you might be able to produce won't get past an expert.) Also, bitcoins are "mined" — by computers that have to solve a tough mathematical problem in order to free a block of 25 coins. This isn't exactly the same as gold mining, but in one crucial aspect it's the same. Unlike dollars, which can be created at will by the Federal Reserve, the supply of both bitcoins and gold is determined by forces outside the control of elected or appointed government officials. Given the long history of governments debasing their currencies to the point of worthlessness, the limited-supply, non-governmental nature of gold and of bitcoins has its attractions.
Bitcoins actually have an advantage over gold in this regard, because bitcoin mining generates a steady, predictable increase in supply, whereas the gold supply grows in fits and starts. Bitcoin creation is thus a bit like Milton Friedman's "k-percent rule," which proposed that the money supply be made to grow automatically at a steady rate that averts both inflation and deflation.
The advantage of a "quasi-commodity money" like Bitcoin, writes University of Georgia economist George Selgin, "is precisely that by resorting to it one can avoid leaving the management of money either to central bankers or to the blind forces of nature. Instead, supply is determined once and for all by artificially-arranged resource constraints."
The question, really, is whether you can pick the right "resource constraints" ahead of time. Friedman's rule of a steady percentage increase in the money supply proved problematic when the Federal Reserve actually tried it from 1979 to 1982 — it wasn't clear which measure of the money supply was the right one to target. Since then the focus has moved on to inflation or, recently, nominal-GDP targeting.
With Bitcoin it's the supply that's targeted, with steadily decreasing returns to mining and the number of bitcoins set to top out at 21 million around 2040. From the Friedman perspective, this is not a great k-percent rule. Instead, it's a recipe for severe deflation — fewer coins chasing around after a growing number of things available for sale. The flipside of deflation is rising prices for bitcoins in other currencies, which is what we're witnessing now. Prices have risen so far, so fast, that it seems inevitable that they'll collapse at some point (because remember, they have no intrinsic value). That doesn't have to be the end: Bitcoin already survived a price boom and bust in 2011. But far fewer people were paying attention then, and on its current trajectory the market seems destined to become dominated by gamblers and greater fools.
Then again, lots of important financial innovations start out like that. They become topics of public fascination, prices are bid up to crazy levels, then they crash. Sometimes that's the result of design flaws, but often it's just because we don't know how to use them yet. The infamous 1720 South Sea Bubble in Great Britain springs to mind. It wasn't that publicly traded corporations such as the South Sea Company were such horrible things, just that nobody really knew how to behave around them. Eventually, we figured it out. Sort of.
I tend to agree with Felix Salmon's verdict that Bitcoin would be better off if it had been "designed to be used primarily as a payments mechanism, rather than as a store of value and a unit of speculation." (The 21-million-coin limit seems like the most obvious flaw in this regard.) But I also know that nobody really knows what the right currency for this networked, globalized age will be. It'll take experimentation, trial and error, and the occasional financial bubble to get us there.
Update: Since there seems to be a widespread misunderstanding about this among less-sophisticated members of the Bitcoin fan club (who keep posting about it in the comments), the fact that bitcoins are divisible into itty-bitty bits of bitcoins is in no way relevant to the concerns I raise about the 21-million-bitcoin limit. The issue is simply that if the overall money supply can't grow, you will probably get deflation. Something that costs 10 bitcoins today will cost 5 next week. Which may sound great, but is no way to run an economy.
Then again, as the very-sophisticated Bitcoin fan Florian Bösch comments below, it is a really effective way to get people interested in a fledging cybercurrency.