Bitcoins Reduce Global Warming

Motherboard reported 24TWh/year in the article 'Bitcoin’s surge in price has sent its electricity consumption soaring' on November 1. If you read the article, they say this much electricity consumption would be profitable -- not that this was what was being used. Let's look at how much is being used -- albeit still an estimate:

Current network hashrate is 11,584,496TH/s.

Current mining hardware (using the Antminer S9) uses 24.64W/s (with a 93% efficient powersupply) and produces 13.5TH/s

That works out to 1.83W/TH/s, and for the current network hashrate, that's 21,145,204 watts per second. Yearly that works out to (rounding and converting to terawatts) 11.12TWh/year..

That's a lot of energy, but still less than half what Motherboard is reporting. You could argue that miners are not using the latest equipment, but energy is their biggest cost so it makes economic sense to have the latest and most efficient.

So, it would be profitable at 24TWh/year, not that they're using that much (just profitable), so the number would be somewhere between 11.12 and 24 TWh/year, my guess it's closer to the 11 figure.

A simple way to look at the problem is that dollars convert to watts. Each dollar spent is for energy, either now or in the future. This works because if you buy a product (paper), the cost of the paper is the energy to make the pulp and dry it into sheets. The cost of the pulp is the energy to turn the wood into pulp (using chemicals, then cost of energy to make the chemicals). Eventually every raw material in the process can be reduced to the energy it takes to process, grow or dig out of the ground.

The cost of paying workers still goes back to why you have to pay them -- they need to eat and buy clothes and houses, which all need raw materials which all took energy.

Profits at each step is still just future energy -- eventually profit is spent to buy something that needs raw materials that need energy. Profits are just energy that we haven't consumed yet, either corporate profits or individual workers retirement savings.

Bitcoin has many uses, and so we could pick from a lot of alternatives. Keeping things simple, let's pretend it's the equivalent of a credit card processor.

This gives us a few big companies -- let's just look at Visa and Mastercard. In 2016 according to their financials they had expenses of 5.3 and 4.5 billion respectively. I pulled out holdbacks for legal expenses, which you could leave in (lawyers still consume resources).

So just the two main payment processing companies are using 9.8 billion dollars a year, and we're not looking at the other payment processors or any of the banks.

If I make a simple assumption that electricity is worth $0.12/KW, then I can convert dollars to energy (could be more or less depending on location, but I'm hitting the low side, most countries are much more).

That works out (at $0.12 per KW) to 81TW per year.

The cost of mining is between 11-24TW per year, and that cost doesn't change if the block size increases to handle more transactions.

Add in the cost of mining hardware and personnel to run the data centers, assume you double or triple the total annual budget, it's still not even close to our current system.

The US Federal Reserve Bank alone has an expense budget for 2017 of 4.3 billion (35TW if we want to use energy as cost). Add in the other countries, and all the businesses that transfer payments and the suggestion that cryptocurrency is not efficient is too simplistic.