For a few days / weeks now some numbers have been circulating all over the Internet calling out Bitcoin on its huge consumption of power.
If you were scared by the number let me tell you something: it's a lot more than what they quoted (they were only talking about Bitcoin), and it's gonna be bigger, a lot bigger, for a while.
But the whole idea of saying "one Bitcoin transaction costs x amount of energy" is a fallacy and it's not the only one...
First of all energy consumption isn't necessarily bad: there is plenty of wasted energy out there, think of all those photons hitting the ground and not being used. Think of all the wind we're just letting pass by, all the water flowing without its energy being harnessed, ...
We're as far as a type I civilisation as we can be with our technological knowledge...
We suck at energy anything really: production, conservation, transport, ... I'll come back to that.
Secondly think of all the power consumed by other industries: comparing the power consumption of a worldwide monetary system to the one of a household is stupid. Do you have any idea how much energy the current monetary system consumes? From ATMs to offices to servers to printing and pressing currency,...
Short answer: only the first hand of this equation, the ATMs, consume in 2 days about as much as Bitcoin does a year (3.5m ATMs each consuming 5~6KWh / d), and it's growing as well. I'm not seeing a lot of "ATMs consume more power than whatever amount of countries" anywhere. Why? Because that's idiotic.
Now where the interesting part of all this starts: crypto-currency mining is a market play. No one is gonna use (and pay for) energy if they don't get a return on it. So companies (people mining in their garage are anecdotal in this industry) have a huge incentive to go where the conditions are favorable to it: cheap, abundant and reliable electricity, cold weather and good connectivity.
The cost of entry is very huge as well (which is a problem for crypto-currencies but a good thing for what we're talking about right now). Which means that investing in solar panels or hydro generators or what not is basically mandatory unless you can find the Icelandic deal (more on that later). Anyone who is gonna spend $10m on equipment is gonna consider how much energy they can produce with a fraction of that cost invested in any form of renewable ("free") energy and if it is interesting on the long term (short answer: if it's not, the problem isn't Bitcoin mining).
Now some mining operation, mainly in Asia, are still relying on fucking coal... They are going to be priced out slowly and my bet is that the next BitMain mega-farm will be solar powered. I could be wrong on that, but if I am it is another problem than Bitcoin mining: it is the fact that we still use fucking coal for anything. This should be banned or at least heavily taxed (which would result in mining operations switching to a more favorable solution faster than any other industry would, or die).
Some companies were forward thinking enough to find what I call the Iceland deal: a country which has way more (clean) power production than it uses and can't really export it (in a cost effective way); it also has a cold weather... Those companies have built huuuuge facilities (pictures attached of one I visited and the geothermal source it consumes power from) which draw power like crazy and cost them $1m a month to run, in electricity alone (figure they gave me, which sounds about right). Where does that money go? To companies which are producing clean energy, who in turn can use this to expand, refine their process, cut costs for other clients...
The equipment itself is improving faster than any other computer component: think graphics cards or CPUs are improving? Look at the evolution of the efficiency of ASIC miners. With the same power input, the amount of computation you get out is skyrocketing. Less and less energy is wasted in heat and what not with each generation of miner. Graphic cards are probably also gonna get a good kick thanks to the afflux of money from altcoin mining rigs (sorry for the shortage, enjoy your 4k 120fps gaming in a few months instead of a few years).
The simple fact we have a number for the overall power consumption is a good thing! We know how much money costs to run, which is a crazy improvement. I defy anyone to find me a good number of how much the USD costs, or the EUR, or whatever... We have this figure for crypto-currencies (albeit an estimate, but it's easy to get) and can keep an eye on it. Were it to become out of hand (which it isn't, no matter how many times you read it's a lot of energy, it's still way more efficient than any other monetary system and way way cleaner) we can see it.
And this isn't even my final form! If today a lot of the (biggest) crypto-currencies are based on mining there is no indication that it's going to be the system that survives, or the pre-dominent one. A lot of research is currently financed by crypto-currencies to find alternative solutions to decentralized concensus. PoS (Proof Of Stake) for instance would consume very little energy, with a few (tens of) thousands of consumer grade computers maintaining the whole currency that relies upon it on roughly idle level power consumption... Imagine a monetary system that costs less per year to run than a family household... (Yup I did that and it's also stupid in reverse).
Thanks for reading, have a crypto-cookie!