Now, I'd be the first to admit that my knowledge of economics is less than sound. We're told by the Great and the Good that the invisible hand of the market ensures that goods are bought and sold in such a manner that their price stabilizes at some magical equilibrium value.
Well, at least that's the idea. A loaf of bread has a certain visceral value - as does a pair of shoes and so on. Matter, in a sense, matters. But it's no longer the only thing of value and we merrily consume services at a cost that would confound earlier generations.
So, with the recent furore over cryptocurrencies, I started pondering the wider implications of having shiny hardware acting as glorified room heaters. I've heard it said more than once that the act of mining a coin (scrypt or SHA256, etc.) is 'wasteful', and at first glance the power currently driving the Bitcoin network is rather respectable - Forbes last year was bandying around a figure of 15M$ for the daily energy use, leading to something like 6GW of dissipation.
Probably accurate to a factor of a few, but let's not forget that then the market capitalization was around 1010$.
So, is it worth thinking of an efficiency figure? If so, to keep the present blockchain activity going 'costs' less than a cent for every dollar of value.
But how much energy is used by a traditional bank to keep itself ticking over? Think of the lighting, air conditioning, servers, etc. For fun let's plug in some figures.
HSBC employs 0.3M people and has assets of 3x1012$. Let's say that each person uses around 1kW continuously to do their job. In which case HSBC dissipates at the very least something like 3MW continuously. So it might seem that a traditional bank is about a million times more efficient than the Bitcoin network in terms of its cost per dollar - and that should surely count for something. Of course, the choice of whether or not to trust a bank has very little to do with its energy efficiency. The set of all Google returns for '<insert name of bank> bank + laundering' is not an empty set.
One point to think about is that the nodes of a cryptocurrency network don't need to be close to people or cities. Indeed, one might argue that communities with significant hydroelectric power reserves would do well to replace electrical heating schemes with 'plumbed-in' mining machines. After all, why waste the heat? With 'instant-on' PCs one could construct an electrical water heater that, at present rates, would actually pay to be used.
Now there's a thought.