Review of the Bitcoin Mining Council’s Latest Report — Part 1
Q4 2021: Comparative Power Consumption data
In June 2021 a group of heavy hitters in the Bitcoin industry, including Michael Saylor and CEOs of large mining companies, formed The Bitcoin Mining Council, or BMC for short.
It was created — at least in part — thanks to Elon Musk’s famous “we’re now accepting bitcoin for Telsas” “Oh wait, we’re not anymore because it’s bad for the environment” flip-flop that had occurred a few weeks prior to that.
Musk’s reason for doing this were almost certainly less to do with actual environmental concerns and more to do with pressure from the board and issues over the enormous subsidies Tesla receives as a company.
After all, Musk had been well aware of how Bitcoin worked prior to either of these announcements and already acquired $1.5bn of it as part of its corporates reserves, something it still holds today.
But whatever the reason actually was, it once again publicly highlighted the ongoing — and much misunderstood — PR problem that Bitcoin has to carry with it everywhere it goes.
So, the idea behind the BMC was to create an official mouthpiece for the industry and dispel some of the myths that surround it. It would do this by sharing best practice, promoting transparency and educating the public on the benefits of Bitcoin and Bitcoin mining.
Since then, it has been producing quarterly reports on the state of the industry and last week the latest of these covering Q4 of 2021 was released.
As usual, it makes fascinating reading and continues to dispel the myths that surround it, but there are also some problems both with the content and the report itself.
This article looks to examine the data and conclusions made by the BMC in one key area — comparative power consumption.
A word about Personal bias
It is no secret, of course, that I am a Bitcoin miner and have been for over five years now.
These days my operation is a mish mash of M20s, M21s, M31s, T17s, S17s, T3+Pro 67’s and a mixed bunch of Avalon machines all made between 2019 and 2021 and all based in Irkutsk, Siberia. Recently I visited the operation and took this video.
It packs a pretty punch and the question most people ask is about power consumption and the implications thereof. It’s a fair question, and, although the answer is simple in my own case, the overall picture is a complex one.
This operation runs on excess hydro electric power. In fact, the entire area is fed by a single Soviet era dam which I also visited while I was there. It was built for aluminum smelting during the cold war, something that requires incredible amounts of power to operate.
Since then, however, the industry has all but died off and what remains is far more efficient than the older tech that preceded it. The result was an over capacity of an enormous amount of energy and, since Russia is unfathomably big and electricity cannot be transported very far without loss, this power was effectively stranded.
This remained the case until Bitcoin miners started moving into the area a few years ago. Miners have the luxury of being able to operate anywhere, so a cold area with cheap excess renewable energy carries strong appeal. The influx created jobs, bolstered the local economy and provided stability of demand for the power network.
It is one of the areas, in fact, that Putin was alleged to have referred to this week when challenged by his central bank to ban mining. The unconfirmed report suggested that the Russian leader wants instead to concentrate mining activities into areas like Irkutsk — which he named directly — where there is excess power and the activity can be managed.
In this case the advantages to the planet, economy, Bitcoin network and the human race as a whole are pretty obvious and, as a self proclaimed “Hippy Miner” it ticked the essential boxes. I am also extremely optimistic about the direction the mining industry is taking overall.
To me, not only does it seem that we have an unprecedented chance to combine global tech adoption with global environmental targets for the first time in history, we have a direct incentive to deliver on it. It’s a perfect once-in-a-generation opportunity to achieve something spectacular.
All this means is that I was prepared for my natural bias to lean towards seeing the best in the report and support it publicly, something I would need to keep in check.
As it turns out, however, there was more room for criticism on the report than I previously anticipated.
So let’s examine some of the details.
Comparative Global Power Consumption
Despite what the media says, Bitcoin is not an “ocean boiling” demon that is gradually consuming all the earth’s resources as a simple look at the top line numbers reveals.
Currently, Bitcoin mining (220 TWh) uses only 0.14% of the world’s entire energy production (154,750 TWh) and is miniscule amount when compared to the more pressing matter of energy lost due to inefficiencies at around 32% of the entire amount produced.
The top two figures don’t come from the mining council itself, but from respectable research carried out by BP and the International Energy Agency. Having read these reports in some detail, I am satisfied that they do appear to be reasonable estimates of the current picture. I know from my own experience getting this data is not a straightforward process.
These are, of course, energy consumption figures in Exajoules rather than pure electrical consumption number measured in TWh, so to compare like with like a multiplier of 277.778 needs to be applied.
I would argue, however, that “energy consumption” is subtly different from “electricity consumption” and should be borne in mind for the rest of this article.
It should also be noted that 2020 global consumption figures are being used against Bitcoin’s 2021 figures, but since this actually disadvantages Bitcoin, it can be assumed that critics would be accepting of that comparison.
This data is used as the basis for the chart below:
This clearly demonstrates that news reports using the line “Bitcoin uses more energy than Argentina” or “Bitcoin uses more energy than Switzerland” (both of these are from the BBC) might not necessarily be true. These articles tend to base their data on the Cambridge Bitcoin Electricity Index (CBEI) which is an admirable attempt to calculate “real time” consumption of power by the Bitcoin network. But how accurate is it?
At the time of writing, the CBEI states that Norway's’ consumption is 124.3 TWh a year, but BP states it is 536, which is an enormous discrepancy. The differences don’t end there. BP’s version of the other numbers shown above are as follows: Ukraine 919.44, Egypt 1013.89 and Poland 1113.89.
These differences apply across all countries, neatly highlighting the different research — and presumably methodologies — it is based on. We need to understand this.
The CBEI bases its calculations on this report from 2019. While data two years old is not so bad for a simple comparison exercise, drilling down to the underlying sources and analysis reveals that much of it was collated in the 2015–2017 timeframe, meaning at least some of the report is now up to seven years old. Much has changed that period — can it really be comparable?
According to BP’s report from 2021 (based on 2020 data), there are only two countries in the world where the Bitcoin network can safely claim it consumes more power by comparison — Ecuador and Azerbaijan. It certainly does not appear to include Switzerland or Argentina. So who is right?
Given the questions raised over the age of the data in CBEI report, I would tend to lean towards BP’s more modern and comprehensive numbers, remembering, again, that we’re dealing with a conversion of Exajoules to TWh based on energy consumption data.
But even if we take that as read, there is also the small matter of the amount of power than Bitcoin uses.
The BMC, who, arguably has a political motivation to keep this number as low as possible, actually reports a number far higher than the CBEI with the two figures being 220 and, as of today, 131.47 respectively — another substantial difference.
However, the CBEI also provides a lower and upper range estimate which puts the number somewhere between 49.26 and 330.11 TWh per annum, which, although enormous, does work with the BMC self-reported estimate. The CBEI also uses a method of calculation that I consider to be reasonable and logical. The BMC’s number is entirely based on extrapolated self reported data.
So what does all this mean?
If anything, it highlights just how incredibly difficult it is to get accurate data on global power consumption, something I have referred to previously. But it also means, in my opinion, that neither side can be categorically correct about any power comparisons.
It’s just a question of which methodology and data source you subscribe to and it will remain this way until the day we have solid, properly auditable data.
We live in hope of that.
However, since the consumption figure for Bitcoin itself is in the ball park of both methodologies discussed, I believe we can confidently say that Bitcoin’s energy use is likely to be less than people think (thanks to consistently misleading media reports) and that the overall percentage of global power used by Bitcoin mining is, genuinely, incredibly small.
Finally, it should also be mentioned that this is not really “additional” power. That 220 TWh is already being produced somewhere and the distribution of it is skewed by the countries that produce most hash power.
The question is more about how this power is produced, whether it was being produced before and whether it is really needed elsewhere. The report covers some, but not all, of that.
Anecdotally, of course, we already know there is at least one operation — the one discussed above — that is a perfect demonstration of this.
It stands to reason there will be others and, since I personally know of other operations based around similar situations, I can corroborate that statement.
Being, as I am, a single data point, however, it isn’t really enough to support a wider statement of fact.
Mining Power vs Other Activities
Comparing by country is interesting enough, but making similar comparisons against industry, activity and sector also produces a different viewpoint of note.
Again, numbers differ significantly by source, but all of them broadly agree that Bitcoin is a tiny element of consumption when compared against essential, non-essential or even completely superfluous activities, such as “holiday lights”
But the extent of the range is bewildering.
Taking “Air Conditioning” as an example. The BMC report claims it is 6283 TWh per annum in the U.S. alone, yet a 2020 IEA report puts it at 1885 TWh for the entire planet and the CBEI puts it at 2199 TWh globally, citing another IEA report.
“Appliances” are another enigma. The BMC report claims it is 3000 TWh (just for the US) but yet another IEA report puts it at 3636 TWh for the entire planet. This one may come down to simple definition and the BMC report does not make it clear what it is referring to here.
Further, the CBEI states that TVs and fridges in the U.S. alone accounts for 164 TWh, while this article claimed in 2019 (notably without source) that power wasted by units on standby account for 200 TWh.
“Gold mining” is another hot topic, with figures varying wildly from all sources. The CBEI’s figure is only 131 TWh against the BMC’s 571, but closer inspection reveals that the former is based on this 2007 report with the data extrapolated, which raises some questions about accuracy.
This also ignores the enormous amounts of direct environmental damage linked with gold production, the health hazards and the countless deaths associated with the activity. Gold mining is an incredibly expensive activity in more ways than just power consumption.
Once again, we cannot claim absolutes in any category, although the message- through any source — is broadly supported; Bitcoin uses minimal power not just when compared to regional consumption, but also when compared to other global activities, even those that can’t be classed as “essential” by any stretch of the imagination.
Conclusions
For this part of the report, my view is that the picture painted by the graphs in terms of trends and broad conclusions is more than likely accurate, although there is room for debate on the absolute numbers.
As a someone who has been mining Bitcoin for over 5 years, I am supportive on the industry and genuinely excited about the direction it is taking.
I am also openly supportive of the efforts made by the BMC to combat false narratives and to consistently drive for improvement, but, as an analyst, I have to also acknowledge that this report has some way to go, lacking as it is in clarity of reporting, clear definitions and specific data sources to support each number.
As my teacher always used to say, “You need to show your workings.” It needs to be bulletproof.
We also need to acknowledge that the BMC currently estimates that it only has data on 46% of the network, so some extrapolation is required. Can we but sure that the 54% is of similar ilk or is there a reason they are not yet reporting? Can we even be sure that the self-reported data is accurate?
Over time, I’m sure this percentage will increase and reporting accuracy will increase as a function of the industry maturing, but we’re not there yet.
And, to be clear, I don’t want to be critical of the BMC and its objectives. It’s a tankless task and, as we’ve seen, extraordinarily difficult to make sense of complex and conflicting global data that is subject to different collection techniques and subjectivity.
However, it could also be used as an opportunity to set the standard of what follows and, in my view, this is a worthy goal to strive for.
What better than to have the BMC’s data cited as the de facto source on these matters? This, in my view, should be a goal that both solves the problem of making absolute power comparisons and gives solid credibility to the project as whole.
But, in the meantime, perhaps we can all at least agree not to leave our devices on standby. It’s entirely possible this will save enough energy to power the entire Bitcoin network on an ongoing basis anyway.
And that’s quite a thought.
Want free access to articles, analysis, podcasts and training webinars? Why not subscribe to the ‘Bitcoin and Global Finance’ newsletter? Subscribers over 18, resident in Europe (see list on subscription page) & new to Bitcoin can claim £10’s worth of Bitcoin on joining! Unsubscribe at any time.
Disclosure: The author of this opinion piece has been heavily involved with bitcoin for several years and holds a substantial cryptocurrency portfolio, including bitcoin. He also has a mining operation running the SHA-256 algorithm based in Siberia and is a published author on the subject of promoting the understanding of cryptocurrency. Jason is an analyst at Quantum Economics and consultant to Luno.
Disclaimer: This content is for educational purposes only. It does not constitute trading advice. Past performance does not indicate future results. Do not invest more than you can afford to lose. The author of this article may hold assets mentioned in the piece.
If you found this content engaging, and have an interest in commissioning content of your own, check out Quantum Economics’ Analysis on Demand service.
It would be interesting if every industry has to report total energy usage as the measure of their worth. How about banks and atm machines, or maybe printing fiat and creating coins?
Spelling error in the Conclusion section in the paragraph starting ‘And, to be clear..’ - ‘tankless task’.