This post was triggered by Tyler’s new post:
With apologies to Scott Sumner, I say Bitcoin is a bubble.
Here’s a Tyler Cowen post from April 26, 2011:
Jerry Brito reports:
“The arrow notes the date my column on the virtual currency was published in TIME.com. The day after that piece was published, the Bitcoin exchange rate reached an all time high at $1.19. Yesterday, just over a week later, it was pushing $2.”
“In late April, 2011, one bitcoin is approximately at parity with the US Dollar — a 2000% percent appreciation against the Dollar in less than a year.”
Of course, that’s the appreciation you might expect from a highly successful private asset! Or you can take that as a sign of the dependence of the Bitcoin upon expectations, and a sign of its bubbly nature. Just think how hard it is for a major country to establish credibility for its currency, and then ask how much of the Bitcoin value is expectations-dependent and multiple-equilibria dependent. What will be the rate of Bitcoin appreciation five years from now? I guess it will be falling.
If I read that correctly, Bitcoins started at a nickel and had reached $2 by April 2011. That sure looks like a bubble—even to me. But, what does the term ‘bubble’ actually mean? One definition has to do with prices not equalling a rational expectation of fundamental values. But there’s really no way of testing that definition. If you say you don’t think the price is right, you’ve basically assumed the answer. The question is; how do we test whether the bubble proponents are right?
In my view a bubble claim must be regarded as a sort of implied prediction about the future path of prices. More specifically, that prices will be substantially lower at some point in the not-so-distant future. And I think one really ought to give a date, as a random walk with no trend is exceedingly likely to fall below current levels if you wait long enough. So it’s good that Tyler gives a date—5 years.
As an aside, unlike Tyler many bubble proponents don’t realize this and they give no date. Then when prices rise after their prediction they brush it off, saying prices will later fall. If prices later fall they say it proves they were right, even if the later decline leaves prices higher that when the bubble prediction was made. This false gloating will probably happen soon in Australian housing. I once did a post accusing The Economist magazine of making exactly that error, in an ad touting their accurate prediction that housing was a bubble (which was actually wrong in 5 out of the 6 countries they referred to.)
Unfortunately the wording used by Tyler is a little vague, so I’m not quite sure he was calling a bubble back in 2011. He refers to a prediction that prices will be falling in 5 years, not lower than today. But let’s assume he was.
In any case, in today’s post Tyler links to a report of spectacular price appreciation, and then comments:
The Bitcoin price has just broken the all-time high of $31.9099 that it set on June 9, 2011 on MtGox. After a persistent, one-and-a-half month rally from $13 to $28, followed by nearly two weeks of bumping up against $30 and then hovering around the $28-$31.5 range, the bulls have finally won…
With apologies to Scott Sumner, I say Bitcoin is a bubble. Outside of war and rebellion, do “normal” new currencies behave this way?
I’m too cowardly to go out on a limb with price predictions for such a volatile asset. Notice that it soared to $31 dollars soon after Tyler’s April 2011 post, then collapsed, and has again soared to $31. For those keeping score, here’s what I’d suggest:
Assume the price drops to $3 in 2016 (5 years after 2011.) Should we give Tyler credit? I say we should say he’s batting .500, which is excellent in baseball but only so-so in terms of binary choice predictions. That’s because if it drops to $3 his current post will be prescient, but the earlier post will be inaccurate–bitcoins were not overvalued at $2. In fairness to Tyler he didn’t quite say that in 2011, I’m cheating a bit by referring to the inference most readers would have taken from the earlier post.
I am able to state with complete assurance that Tyler made one inaccurate prediction in his April 2011 post:
Last post on this topic!
He’s done 19 posts since (although in fairness most were merely links.) Given how much smarter he is than me, I find it reassuring that he’s not much better at predicting his future behavior than I am.
PS. I actually agree with Tyler that the price path in his graph looks fishy. What do my finance readers think? Do random walks have such long swings that look serially correlated? Or are my eyes fooling me? Doesn’t it look like a momentum trader could profit?
PPS. Tyler has put a crushing burden on my shoulders. By singling me out as the bubble denier (and why not Fama?), I face the unhappy prospect of being taunted by commenters every time an asset price moves in a way that to all the world looks like a bubble and crash. Imagine I had started blogging in 2006. The Great Housing crash would have (unfairly) ruined my reputation just as surely as the 1929 stock market crash (unfairly) ruined Fisher’s reputation. Fisher was never taken seriously again, even though he had the right prescription for ending the Depression–reflation via devaluation. In contrast I started blogging in early 2009 and have been mostly able to avoid big asset price crashes—so far. Thus my reputation is not ruined and people are willing to at least listen to my NGDP targeting ideas. I’m very lucky.
PPPS. Because low interest rates are the new normal, higher P/E ratios are the new normal. Thus the US stock prices “have reached what looks like a permanently high plateau.”