I agree that we're quick to look at other people's small wastes with condescension, but bitcoin is not a small waste.
If an evil overlord wanted to exacerbate our climate and energy problems, pushing Bitcoin to mainstream use is probably the ideal way to do it.
Incorrect, growth curve is driven by mining power and forces innovation in lower cost of electricity. The worldwide cost of Bitcoin usage in KW has been compared to the headquarters of Bank of America's usage.
You seem to be missing a couple units there...
Here's my reasoning. Miners spend electricity. They're getting block rewards and fee rewards, they'll exchange earned bitcoins to dollars and pay their bills. So ultimately energy spent per block = (block reward + fee reward) x bitcoin price.
Right now block reward is 6.25 BTC. Last block fee reward was 2.13 BTC. Block reward will go down as time goes on.
I don't really understand economics of fee rewards and whether they'll grow over time linearly or exponentially. I checked historical data and it does not seem to grow at all. So I'll suppose that reward fee will not grow.
So in the end you'll have block reward of 10-20 BTC. That's all you have to pay your electricity bills after you mined block. So basically energy spent per block is directly correlated to bitcoin price, and that's about it.
Obviously bitcoin price won't grow to the moon and its price is limited by some economical factors.
So while bitcoin network might draw non-negligible amount of electricity, it still is limited by design.
Now if miners will decide to change bitcoin design, for example to increase block size, that's another matter. But it's not clear if that would happen.
Since supply will become more and more constrained over time, while demand has no realizable upper limit (that we know of), this assumption is wrong:
> Obviously bitcoin price won't grow to the moon and its price is limited by some economical factors.
If bitcoin were to be fully adopted by the mainstream, BTCs would be easily in the M$ range.
So that's a rough estimate upper bound if Bitcoin would become our civilization currency of choice: 1.2% of total energy spent.
Not the end of the world, if you ask me.
Mining bitcoin is essentially printing money, so miners with profitable setups have an incentive to expand these setups. As mining capacity grows, a parameter called "difficulty" - which represents how much power you need to waste to mine a block - is adjusted so that the mining rate remains constant. Since mining is necessary to maintain network security, difficulty will grow to compensate for expanding mining capacity, but not high enough to kill it off.
If you look at the history of the adjustments to difficulty, you'll find a pretty much clean exponential.
(In other words, Bitcoin is as close as we've ever come to expressing greed in units of kilowatt hours.)
I see this take quite regularly and it seems a normal response to clickbaity headlines like "BTC has same power usage as Switzerland", but I think it misses some important nuance to the debate.
* Most studies locate miners and then assume their CO2 output based on energy use assuming a general energy mix of the country, this is not useful because energy usage != CO2 output, it depends very much what energy source is used * Miners are strongly incentivized to find the cheapest possible electricity, often means finding pockets of excess wasted energy, which is also often renewable or hydro and using it rather than letting it go to waste * BTC can be thought of as converting frequently excess wasted units of energy into a commodity which can be sold * The above might change and if for some reason coal became the most competitively priced energy source then this would be very bad for the environment indeed.
Whether or not BTC is useful, it undeniably holds value, in the literal sense that if I gave you 1 now you could sell it at a market rate
All the above notwithstanding, a lot of research is going into finding an alternative to Proof-of-Work that is less energy intensive and I would hope that in the future a secure alternative will be found.