This is an essay that in its first part, covers how the financial sector strongly influences our political leaders; what financialization is, how it stagnates innovation and stops us from taking action on climate change. In the second part, it addresses why Bitcoin has value and is not a pump-and-dump scheme, and how it can disrupt the financial sector, giving us a chance to act on climate change and create a newly transformed world (one which exists with or without Bitcoin). The third part addresses environmental concerns regarding Bitcoin’s proof-of-work approach to securing the network, contextualizing the problem and providing some solutions through both external and internal pressures. Finally, the essay concludes with a call to end financialization if we want to maintain a habitable planet.
A contents section is below to make it easier for the reader to jump between sections. There’s a lot to cover, but this was the only way to make the case so please, take breaks when needed, but do read to the end.
Bitcoin’s environmental impact
I have alluded to some issues related to the digital currency, but the most important one for the climate change community is that of its environmental impact. There are issues with mining companies and underlying energy sources that drive the high-computation costs. The Cambridge Center for Alternative Finance estimates that today, Bitcoin uses roughly 70 terawatt hours per year or 0.31% of the total global electricity production (although estimates peaked at over 130 terawatt hours during the recent bull run). To put this into scale, Austria and Colombia, with populations in the tens of millions use similar amounts. In comparison, the Center reports that refrigerators in the United States use around 100 terawatt hours per year of electricity and global air conditioning use requires 2199 terawatt hours per year of electricity. Cement, a problematic source of CO2 emissions, requires 384 terawatt hours.
In some cases, Bitcoin mining companies claim to rely on excess electricity in the system to run their hardware. This suggests that mining could be a buyer of last resort, but it is difficult to know how many miners operate this way. It is true that not all Bitcoin mining is renewable energy derived, but Bitcoin is not an entrenched institution, it is malleable. Miners will move to wherever energy costs are cheapest and with the right incentives and external pressures, it is not hard to see mining being fully renewable in a very short period of time. As it stands, the Bitcoin Mining Council estimated in June 2021 that around 56% of mining is based on renewables and nuclear. A 2018 study puts it around 28% renewables. For perspective, roughly two-thirds of California’s electricity production is renewable and nuclear (~36% renewable)* and in Georgia it is 12% renewable. Further, there exists historical precedent from a different energy intensive technology sector from which Bitcoin miners can learn from: data centers.
In 2010, data centers faced criticism for their rightly concerning, fossil fuel-based energy consumption (when talking about TWh, we’re talking in general energy terms, but their energy is sourced as electricity). Did data centers ignore the pressure? No, the data center industry is now the largest user of renewable energy on Earth. Due to pressure from Greenpeace and other environmentalists, companies who relied on the industry, like Google, Facebook and Microsoft, compelled data centers to shift. To quote some statistics from Data Center Frontier, “Electricity usage by global data centers grew just 6 percent from 2010–18…while the number of physical servers rose 30% and compute instances rose by 550 percent. This marked a complete reversal from the 90 percent growth in data center energy from 2000–2005.” In other words, the more we understand the operations of Bitcoin, the better we can situate ourselves to put pressure on Bitcoin miners and electricity providers to switch to renewables.
In a 2017 study, data centers — the infrastructure backbone for our Internet-based data sharing needs — were found to use 416 terawatt hours per year or approximately 3% of the total global electricity production. The Cambridge Center reports 200 terawatt hours for data centers based on a different study, so the estimates do vary. Using the Cambridge numbers for data centers, data centers have a lower bound of around 1.5% of global electricity production. This means that data centers alone use roughly 40-percent more energy than the United Kingdom. This is 1.2-2.7% more than the Bitcoin mining network uses (~0.31%). This means that Bitcoin’s energy use is within and below the industry standard for computational purposes. If Bitcoin is truly the resilient and adaptive disruptive technology that its proponents claim it to be, then there is no reason to believe that Bitcoin will not follow the data centers’ path toward renewable energy reliance. Plus, with continued pressure from climate-concerned members of the community, the shift to renewable energy will be difficult to escape. In fact, since Elon Musk declared that Tesla would no longer accept Bitcoin until it adopted a renewable energy standard, there has been intense discussion around alternative approaches and solutions within the Bitcoin developer mailing list.
Finally, there are a number of additional possibilities for improving energy efficiency in Bitcoin without compromising the proof-of-work algorithm, although within a transition from a fossil fuels framework, these would be temporary. A few examples are mining stranded energy, or mining from flared methane. The wasted energy from flaring alone would power almost 10 times the amount of miners currently online.
There is an additional environmental concern, and that is the built-in obsolescence of Application Specific Integrated Chip miners (ASICs). Like a lot of technology devices that we use, devices are built to be thrown away within a few years of use. For example, consumers are incentivized to upgrade their smart phones every two years and ASICs has a similar lifespan. ASICs are designed to do only one task and to do it well. Once they burn out or become outdated due to faster chips on the market, they will be thrown away. Some miners are sold for scrap metal, but I do not doubt that most end up in a landfill. ASICs are not the only device that contributes to e-waste. The EPA estimates that in 2009, US consumers and businesses produced 2.37 million tons of e-waste. Only 25 percent was collected for recycling and the rest went to landfills. We, humans, are wasteful. The EPA reported that if Americans recycled 1 million cell phones, they could recover more than 35,00 pounds of copper, 33 pounds of palladium, 772 pounds of silver, and 75 pounds of gold. Given the design of ASICs, which are metal boxes filled with 3–4 large circuit boards and two or so plastic fans, there is so much potential for metal recovery and reuse. ASICs do not need to contribute to landfills and environmental pollution.
There have been some questionable claims thrown at Bitcoin over the e-waste that mining causes. Alex de Vries, found of Digiconomist published data based on back-of-the-envelope calculations that claimed that Bitcoin miners were major e-wasters. For now, we will assume his assumptions to be true. Based on his calculations, he concludes that Bitcoin’s e-waste generation is 6.37 kilotons per year, roughly half of the waste generated by Luxembourg. Based on the EPA’s 2009 numbers for the United States, this makes up 0.27 percent of American electronic waste (2.37 million tons). De Vries makes a second estimate, which is where he assumes erroneously. First, de Vries does not compare the VISA network’s e-waste to city-sized countries. Second, he concludes that for each individual Bitcoin transaction, approximately a little more than 80 grams of e-waste is generated (we assume this is true for now). He then says that this is in comparison to the approximately 45 grams generated per 10,000 VISA transactions.
However, this argument is incorrect because it ignores transactions generated on the Lightning network. As more users move to Lightning for day-to-day transactions, the number of transactions that happen with just one transaction on the Bitcoin layer become competitive with VISA. VISA can handle 60,000 transactions per second, whereas the Lightning network is capable of millions or billions of transactions per second. Therefore, de Vries’ argument breaks, because it shows that Bitcoin is no more wasteful than the existing sector, if anything, it has the potential to be less so. To be clear, this is not to excuse the e-waste problem, but only to put it within perspective. E-waste is a byproduct of a single-use society and not a Bitcoin-only problem. We must transform our society consumption habits if we want to see an end to e-waste.
Mining Alternative Has Network Security Risk
Proof-of-stake is a well-known alternative to proof-of-work. Its proponents claim it is less energy consuming and that it works as well as proof-of-work. However, a big concern with proof of stake is that it is based on who has the most money staked in the network. Therefore, if you are already extraordinarily wealthy, what stops you from becoming the biggest stakeholder in the network? With Bitcoin mining, you need some technical expertise and a willingness to chase down the cheapest energy sources.
What we know from financialization is that proof-of-work is not easy money so trying to take over a proof-of-work network like Bitcoin’s to attack it would be difficult given finite resources and very expensive. It is much easier for Jamie Dimon to stake money than mine. In fact, that is what the ultra-rich do already, they stake their money in shares and other financial assets. Moreover, the benefit of tying proof-of-work to a physical resource, is that it is finite, and electricity must obey the laws of thermodynamics. However, any government could simply inject more liquidity to purchase more of a proof-of-stake currency until it and major banks hold the greatest stake in the network. With majority stake, they can now perform an attack on the network and take it over. In Bitcoin, even if a government owned the majority of coins, it would still not have control of the network, because mining and transaction verification are external to the number of coins owned.
Bitcoin mining is without a doubt energy intensive, however, as discussed in this section, it is not an outlier per se when compared to other industries, especially if we believe that Bitcoin and the network provide a certain value to us. E-waste is an issue, but as discussed, mining is not a lone contributor to this problem and accounts for a small percentage of overall annual e-waste production. Moreover, solutions already exist, such as recycling, which would reduce mining metals and be overall a positive for the environment. In the end, so long as we believe Bitcoin provides value, as I have tried to argue, then focused external and internal pressures, through grassroots campaigns and community discussion and innovation are worth our effort in the goal of keeping global temperatures below 1.5 degrees Celsius.
*While it may not seem like much of a humble brag, in 2021, California was able to operate its grid at ~95% renewable energy for a few seconds. This is in light of criticisms that achieving that — at all — would be impossible. Thanks to Alex Epstein for catching my typo where I wrote “energy” instead of “electricity”. :) Remember: RENT-FREE.
