Sadly, there are far too many “OH SH-T” moments for those paying even the slightest attention to climate chaos (and related environmental issues/crises). One came for even that “slightest attention” crowd as a peer-review article about Bitcoin’s gluttonous electricity habits made a splash leading to global headlines like “Bitcoin Predicted To Be The Nail In The Coffin Of Climate Change” with the potential that “Bitcoin alone could push global temperatures over the 2ºC catastrophic threshold by 2034.” ‘Oh s—, Bitcoin miners could well kill kill us all’ was a thought that passed through likely not just this blogger’s mind. Now, on second reflection, the thought was ‘this seems a bit click bait exaggeration potential’, then ‘need to check into it’, and, then, life caught up and the check-up never occurred. And, like probably so many of us, in the back of the mind was this idea that Bitcoin’s electricity demands posed a serious, menacing, and real threat to hopes for driving a clean energy future.
Those click-bait headlines and that back-of-the-mind, semi-forgotten thought were — like all too many fantastical claims — simply off by orders-of-magnitude. As Churchill said, a lie is halfway around the world before the truth got out of bed. Well, as to the fantasy claims of Bitcoin burning the world, the truth got out of bed wide-awake today as the Coin Center just issued a report it commissioned from energy efficiency expert Jonathan Koomey entitled Estimating Bitcoin Electricity Use: A Beginner’s Guide.
“Periodically and predictably someone makes wild claims about information technology electricity use. Because I’ve been working on this topic for a quarter century, those claims always make their way into my inbox. I wrote this report to address misconceptions about Bitcoin electricity use and give those not versed in its complexities some simple tools for assessing whether an estimate of Bitcoin electricity use is credible.” Jonathan Koomey
In brief, Koomey makes clear that Bitcoin is responsible for perhaps 0.2% of global electricity use with uncertain projections about its future growth. That Bitcoin mining is using 1/500th of the world’s electricity production, when hundreds of millions have no electrons, isn’t something to laugh at but it isn’t going to drive global warming.
As to that 2 degree prediction, Koomey makes clear the fundamental analytical assumption failures:
The authors assumed that the efficiency of Bitcoin mining as well as the emissions intensity of electricity would stay constant over the next century. [p. 12]
Sure, based on the past 50+ years, it seems ever so reasonable to assume that computing efficiency wouldn’t improve an iota. Sure, based on rapidly reducing emissions per kilowatt hour through the global electricity system over the past decade, let’s assume that the grid of 2109 will be no cleaner than that of 2019. Sure …
Reading that one line reminded me of how forecasting agencies are constantly getting it wrong as to (even near term) future clean energy developments: assuming that efficiencies won’t improve and costs won’t fall pretty much guarantees underestimating how well clean energy systems will perform in the market place.
Koomey’s opening underscores the importance of being awake to clickbait risks about information technology energy demands since debunking claims about growing IT electricity usage is Groundhog Day for him. For example, there were claims that the Internet used 8% of electricity in 2000 (actual figure, about 1%), all computers plus the Internet used 13% (actual about 3%), and that a wireless Palm VII used as much electricity as a refrigerator to support its networking (oops, this was an exaggeration of the networking electricity demands by some 2000%). Thus, Koomey’s opening section’s title: Know Your History (Beware the Hype). Or, as Jonathan summarized it last fall when he blogged on this Bitcoin will burn the planet hysteria.
“The amount of energy needed to refute BS is an order of magnitude bigger than to produce it.” (Brandolini’s law)
In any event, with yet another quality analytical product, Jonathan Koomey’s work should put to rest aside any lingering concern you might have had that Bitcoin miners were frying humanity’s future.
NOTEs:
2 responses so far ↓
1 John Egan // May 9, 2019 at 11:11 am
Mr. Siegel, Mr. Siegel –
You could not be more wrong.
If Bitcoin continues at this pace, it will suck up massive energy.
From “Bitcoin Market Journal” which one would suspect is rather Bitcoin-friendly, it is estimated that there are 32 million Bitcoin wallets, but fewer actually users since most users have multiple wallets. But let’s stick with 32 million.
https://www.bitcoinmarketjournal.com/how-many-people-use-bitcoin/
32 million represents .004 of the world’s population of circa 8 billion or 4/1000. Fair to say that 1/500th equals 2/1000?
Thus, each bitcoin wallet uses the equivalent of 50% of the per capita electricity use worldwide. Of course, that doesn’t include heating & cooling, transportation, and all other electricity uses. Not of mention that climate activists want to switch fossil-intensive transportation to electricity, also.
I am well aware that using a global per capita electricity use figure has its limitations since electricity access and usage is so skewed towards developed nations. But, at best, this is yet another example of neocolonialism.
“We get to gobble up electricity with Bitcoin and you get mini solar stoves that don’t work well when it’s raining.”
PS – I got 800s in math on both my SAT and GRE – – so I think I know my numbers pretty well.
2 A Siegel // May 10, 2019 at 4:34 pm
Are you simply purposefully misconstruing the post (and the paper it discusses)?
Written in the post:
Perhaps you couldn’t get the ‘hint’ that I am not endorsing Bitcoin, here, and that I see real problems with Bitcoin (users) eating up a disproportionate amount of electricity. This post, however, is not ‘about’ whether Bitcoin is a worthwhile thing or a good use of electricity or an equitable use of energy, but about how ‘click bait’ stories burst onto the scene based on faulty analysis.