In general, "Ponzi schemes" in the context of asset pricing refer to "rational bubbles", or a failure of the transversality condition you need when passing from a flow identity to a present value identity. They are not "fraudulent schemes".
For instance, there can be a rational bubble on a "fundamentally" worthless asset in a perfect foresight world if its price grows at the real interest rate available in the broader economy. (This is actually what you're talking about, in case it's not clear to you - each investor assumes they can offload the asset at a higher price to another investor, and you push the problem of the "greatest fool" to infinity, which makes the assumption a rational one ex post.)
Is there any evidence for the existence of such rational bubbles?
Cochrane (2002) points out that rational bubble theory does not account for the association of "bubbles" with high trading volume.
In a rational bubble, it is rational to hold the asset in every period: there's no reason to observe high turnover along with high prices. They are also silent on the association of high prices with high volatility - in a rational bubble context, volatility (induced by shifting from one equilibrium to another) is always present, and the choice of equilibrium (the fact that we're on an explosive price path) does not produce more volatility than usual.
Rational bubbles appear to be a poor explanation of events that people have historically described as "bubbles" (and of the price behavior of cryptocurrencies).
In contrast, the convenience yield theory outlined in Cochrane (2002) and also in this blog post seems to fit the data much better. If "overvaluation" is driven by a liquidity premium induced by shorting constraints and a short-term constraint on supply, then we would expect to see the standard prediction of monetary economics that high prices should be associated with high turnover.
In that case, overvaluation relative to "fundamentals" should also require shorting constraints - in the case of dollars vs treasuries, for instance, it's illegal to print banknotes or close substitutes to money to finance a purchase of bonds, which is a kind of shorting constraint.
Indeed, shorting constraints are what we see in the cryptocurrency market: BTC futures have only recently started trading on stock exchanges, and even then the "electronic paper" that you hold when you take the long side of a BTC futures contract is not the same thing as holding BTC directly, so what appears like a possibility to short BTC through these contracts is in fact not so.
(It satisfies "speculative demand", so it's possible to "bet on cryptocurrencies" without having to hold any, but it's not possible to use these contracts in the place of cryptocurrencies in transactions.)
If you want to ask whether cryptocurrencies are a bubble, then you have to define what "bubble" means. If it is "overvaluation relative to fundamentals", then is USD a bubble?
Inside The Cryptocurrency Revolution - VICE on HBO
After all, it's trading at a premium relative to T-bills. If you include any convenience yield the asset may have in your definition of "bubble" so that USD is not one, then it's likely that cryptocurrencies are not "bubbles" either.
It's very difficult to say anything about whether cryptocurrency valuations are "justified" or not by this explanation, since the convenience demand of cryptocurrencies, like of cash, has components which are not transparent. We're mostly in the dark, and I wouldn't try to outguess the market price of an asset in a circumstance where we know next to nothing about the factors which are driving the convenience demand for it.
answered Jan 30 '18 at 9:32
62222 silver badges55 bronze badges