[Cryptography] An academic paper on Bitcoin's energy consumption

[Matt Corallo]

> On Jan 10, 2021, at 02:50, Ray Dillinger <bear at sonic.net> wrote:
> 
> Because as long as the *rate* at which it can be produced is limited,
> there will always be the opportunity cost of energy.  Whatever I do
> with it, I can't get the benefit of any of the other things I could
> have done with it.  If I could have sold it to someone for a couple of
> bucks, then using it for hashing bitcoin instead costs me a couple of
> bucks.  If I could have used it to keep the lights on in my house for a
> day, then using it for hashing bitcoin instead costs me the price of my
> house staying dark.

The price of bulk power occasionally goes *negative* in some markets, you’re way of the mark, here. The reality is people get really angry when the power goes out, so power companies generally make sure that they have enough power to cover unexpected peaks. More importantly, the time it takes to turn on or off a large power plant is usually measured in hours or days (the exception being natural gas, often used as “peakers”), so at night there’s occasionally an excess. This type of thing only becomes more common with green energy, which appears when it’s windy/sunny and you don’t get a lot of say in the matter. Driving a few MW of energy to ground has a cost, so occasionally instantaneous power costs can go negative at the price it takes to drive to ground. Sure, “grid-scale” battery stuff is the real answer here, especially around green energy, but short-term market inefficiencies will survive.