Blockchain and energy consumption

Today, we would like to address the issue that has made cryptocurrencies and especially Bitcoin a target for criticism — the fact that the underlying mechanism makes cryptocurrencies highly energy-intensive. It was an especially hot topic a couple of years ago when an article published in Nature Climate Change was widely commented on and numerous news articles were written about how Bitcoin use would thwart efforts to meet the Paris Climate Agreement goal of keeping global warming under 2ºC.

The energy-hungry nature of Bitcoin remains relevant in 2020, especially due to its record-breaking performance this year. The more its value rises, the more difficult it is to mine; the more difficult it is to mine, the more powerful hardware and ancillary equipment for cooling/lighting is used. Thus, particular attention should be paid to the efficiency of computing power that keeps the underlying mechanism of blockchain going. Nevertheless, it is important to handle data cautiously and put things in the right context.

Hard to estimate

Bitcoin and other cryptocurrencies are not the only expressions of our modern, digital way of life that have come under criticism for being energy-intensive. In spring, when lockdowns were imposed across the globe, there was a myriad of headlines referencing to a report according to which the CO2 emissions of watching Netflix for 30 minutes is equivalent to driving for over 6 km. Was it then that the much-acclaimed reduction in mobility and, thus, the resulting cleaner environment was actually offset and invalidated by our binge-watching? No, most probably not.

It is actually very difficult to give precise estimates about how much power such worldwide-spread digital phenomena require. But depending on the methodology, certain estimates can be considered (more) valid/invalid. Reports about the scandalous carbon footprint of video streaming have been contested and debunked (laying bare that the impacts were exaggerated by 30- to 60-times).

With Bitcoin, it is a similar story. Estimates about its current and future energy use have been diverse. The most widely used indicator is the Digiconomist’s Bitcoin Energy Consumption Index (BECI) which has repeatedly found its way to headlines, news articles, reports, and investment newsletters. According to BECI, Bitcoin currently uses almost a whopping 80 TWh per year.

European Environment Agency, in their recent post about blockchain’s energy use, has emphasized that interpreting any estimates is very tricky. It must be done “with caution, due to methodological issues, limited data availability and highly variable conditions across the industry”.

The International Energy Agency (IEA) agrees, saying that apocalyptic predictions about Bitcoin’s energy use are merely sensational. The IEA cites a range of publications and estimates, to finally conclude that the cryptocurrency’s electricity use in 2018 was probably around 45 TWh (i.e. almost half of what Digiconomist has claimed).

Several experts, as cited by ThinkProgress.org, as well as research conducted by CoinShares have clearly refuted the notion that Bitcoin’s carbon footprint is as colossal as some seek to portray. CoinShares’ article argues that using a more appropriate methodology (than Digiconomist), one finds out that the contrary is true — Bitcoin is actually ‘cleaner’ than almost any given sector. Not only is the energy use smaller than alleged, but there are actually several advantages cryptocurrency mining offers from an energetic perspective. First of all, Bitcoin mining is mostly done in areas powered by renewable energy. Secondly, thanks to its mobility, Bitcoin mining “might actually be a boon for global stranded renewables”, i.e. cryptocurrencies very often use renewable energy in remote areas that would otherwise go to waste or incur in high costs and losses along the long-distance power transfer infrastructure.

Of course, it is clear that Bitcoin is not the only cryptocurrency. There are hundreds of altcoins out there. Moreover, blockchain has many more applications than just cryptocurrencies. So when estimating the carbon footprint of blockchain, looking only at Bitcoin is far from enough. At the same time, newer applications tend to use more efficient systems than the highly-criticised energy-hungry proof-of-work consensus mechanism that Bitcoin relies on.

Brighter outlook

We should avoid falling for erroneous extrapolations to the future saying that with the imminent growth in cryptocurrency use, energy consumption will skyrocket. As the IEA has brought out, mining equipment has evolved massively, changing from the initially used CPUs to the modern-day ASICs that are “around 50 million times faster (H/s) and a million times more energy efficient (H/J)”.

According to research into data centre electricity use, the same conclusion can be drawn: over the last decade, computing efficiency has been increasing quickly enough to offset most of the ever-growing demand. Looking ahead, we can feel optimistic as well because “current efficiency potentials are enough to keep electricity demand roughly constant for the next doubling of computing service demand”.

In the 1990s, there was an analogous situation where dire predictions were made about internet use (Forbes is one example). Back then, front-line experts such as Dr Jonathan Koomey had to fight those allegations. Looking back, we can indeed conclude that despite the huge surge in internet use, those gloomy forecasts never came true. With the past mistakes and the current allegations about Bitcoin’s energy use in mind, Dr Koomey stresses that we should be careful with that kind of extrapolations because they “can have real-world consequences”. Furthermore, he emphasizes that projecting the ICT sector electricity use in a reliable fashion more than a few years ahead is simply impossible.

Bigger picture

The proposed figures about Bitcoin’s energy use are clearly overestimated but when we put them in a wider context, with the global electricity demand growing by 900 TWh a year, then they look even more ridiculous. It does not mean that we should stop paying attention to cryptocurrencies’ energy efficiency, but it gives us a better perspective.

Secondly, all goods and services require power — some a lot — but, again, we should look at the wider context. Energy consumption is only as good as the benefits the consuming phenomenon offers. As discussed in our last post, blockchain and cryptocurrencies have the potential to make the world a better place. Even though improving efficiency has to remain relevant, the array of positive impacts blockchain has should allay any worries. Taking all the potential environmental effects and applications into consideration, the deriving climate-change-related implications of blockchain are rather positive.

Thirdly, there is a long list of other human activities that are (arguably) less useful and require even more power than Bitcoin. But if we were to engage in criticism of how much energy watching TV, playing video games, or heating hot tubs consumes, then we would miss the bigger picture. Instead of getting stuck in details, we should strive for a more holistic vision.

Conclusion

Ominous warnings about the impacts of new technologies can be a good way to get the public’s attention and keep the stakeholders’ fingers on the pulse of recent developments. Moreover, nobody argues that policies and measures to ensure the continuous improvement of different sectors’ energy efficiency are indispensable. However, criticism of blockchain’s and the related applications’ energy use is largely unjust. If we really want to reduce our carbon footprint, it is essential that we have real data and use rigorous methodologies in order to avoid wasting time pointing the finger in the wrong direction.