Abstract: While many new technologies carry the promise of creating a better world, they often have unintended consequences. The promise of blockchains is trust-minimizing and immutable peer-to-peer interactions, but early blockchain platforms have faced a great deal of skepticism regarding their environmental sustainability, and continue to cast their shadow over the technology’s potential today.
Citation: Anderson et al. (2021). Encyclopedia of Sustainability, 2nd ed. Great Barrington, MA: Berkshire Publishing.
Any figures or illustrations or illustrations included here are not finalized for publication. Advance publication date as per post date. Copyright Berkshire Publishing Group.
Back to list of Advance Articles
The Impact of Bitcoin and Other Blockchains on the Environment
In 2008, an author or group of authors under the pseudonym Satoshi Nakamoto released a whitepaper entitled “Bitcoin: A Peer-to-Peer Electronic Cash System.” The whitepaper explained that Bitcoin was a new form of digital money that would allow for payments, with the help of what is now known as blockchain-technology, without having to go through traditional financial institutions. It is widely believed that the release of this whitepaper was strongly motivated by the financial crisis of 2007 to 2009, not just because it talked about bypassing traditional financing at a time when the trust in this sector had sunk to an all-time low, but also because the very first block in Bitcoin’s blockchain refers to the instability of the financial system at the time.
How Bitcoin Works
To understand Finney’s thoughts, one needs to have some understanding of Bitcoin’s design. Bitcoin’s underlying blockchain is literally a chain of blocks filled with bitcoin transactions, as each block links back to the one before it. As Bitcoin is open-source software, neither owned or controlled, anyone can run it and direct their computer hardware to participate in the process of creating these blocks. Nakamoto also added an incentive for people to join and support the network. It was defined that there would never be more than 21 million bitcoins, and that they would slowly be released over time as a reward to those participating in the block-creation process. The creation of these blocks was subsequently made computationally expensive. Only a block with a valid “proof-of-work” can be added to the chain. Obtaining this proof is done via a process that somewhat resembles a lottery. Participating hardware generates a series of random numbers hoping to find one that matches predetermined criteria.
As of March 2020, the whole Bitcoin network combined was generating more than 120 quintillion of such attempts every second. The Bitcoin protocol self-adjusts the difficulty of meeting the criteria, so that on average only a single attempt every ten minutes leads to a valid proof and the creation of a new block. Since network participants have to expend time and electricity for running their hardware in this system, Nakamoto described the block-creation process as “analogous to gold miners expending resources to add gold to circulation.” The latter is why media and academics commonly use the term “mining” to refer to this process.
By adding this proof-of-work-based system, Nakamoto achieved several objectives. First and foremost, it provided an incentive for network participants to behave in an honest way. In an open network, any number of malicious entities may enter, but the requirement to expend resources makes it costly to mount a successful attack. Also, an attacker would have to obtain control of more than half of the network’s total computational power to have a chance at succeeding, which would undermine the system in such a way that the rewards would become worthless. Furthermore, an attacker would be required to hold this control for a prolonged period of time to make changes to the history of transactions. Because blocks are chained, changing the history at a certain point in time requires redoing the accumulated work from there. The older the recorded events, the more computationally impractical it becomes to make changes. In this way, Nakamoto ensured that the Bitcoin blockchain would become practically immutable.
Still, almost all computations made by the network are inherently worthless. Every random number that isn’t a valid proof is immediately discarded. On average, the Bitcoin blockchain will only add around 52,500 new blocks per year, regardless of how many computations can be made. The rest serves no other purpose than making it expensive for an attacker to generate a majority of otherwise useless computations. But for the millions of devices that are active in the network, this means that the majority will not even manage to produce a single valid proof during their lifetime.
Another way of looking at this would be to consider the so-called “ghost flights” by airlines during the COVID-19 pandemic of 2020. In order to keep their slots in the skies, airlines continued to operate empty or near-empty planes (until the rules that encouraged this practice were ultimately changed). So while these flights did have a purpose, they didn’t contribute to anything useful (like transporting passengers) and they continued to emit environmentally harmful greenhouse gasses. In essence, this is not very different from how the Bitcoin design requires devices to run without producing anything useful.
Economic Incentives and Environmental Impact
As the value of the network increases, so does the value of the rewards that Bitcoin miners can obtain. This provides a strong incentive for participants to keep adding devices to the network while it remains profitable to do so. Moreover, because all participants compete for the same rewards, they try to gain a competitive advantage by using the latest state-of-the-art devices. The devices that can generate the largest amount of computations per unit of energy are the most profitable ones. Hence, market participants are also part of a race to develop and use increasingly powerful hardware. As a result, network participants quickly cycle through generations of highly specialized mining equipment, as the average Bitcoin miner becomes obsolete in less than 1.5 years.
Figure 1: Estimated annualized Bitcoin electrical energy consumption over time (www.bitcoinenergyconsumption.com).
To date, the precise environmental effects of this behavior remain subject to debate. While it’s possible to estimate the total computational power of all devices in the Bitcoin network, it’s not easy to translate this into specific numbers. This is because the network doesn’t provide a granular breakdown of what contributes to this computational power. Nevertheless, we’ve seen many attempts to quantify the environmental impact of the Bitcoin network. These estimates can be obtained from popular sources like the Bitcoin Energy Consumption Index, the Cambridge Bitcoin Electricity Consumption Index, and the Bitcoin Electronic Waste Monitor, which provide real-time updates. These sources estimated that in March 2020, all devices in the Bitcoin network consumed as much electrical energy as a country like Belgium, while generating as much electronic waste as a country like Luxembourg. Moreover, it was estimated that the energy consumption alone reflected a carbon footprint the size of a country like Denmark. Other studies reached similar conclusions, though often the content is already outdated the moment it is published given rapid increases during the prior years. From early 2017 to early 2020, the estimated environmental impact of the network increased more than tenfold, in line with the price of a single bitcoin, which increased from less than $1,000 to around $10,000 over the same period.
Though not insignificant in absolute terms, the most shocking part is in understanding what this means relative to the actual size of the system. Whereas the traditional financial system processes over 500 billion non-cash transactions per year, the Bitcoin system processes only around 120 million transactions per year. That’s just below four transactions per second on average. On a network-wide total of 120 quintillion computations per second, that’s a ratio of one transaction to every 30 quintillion computations. Knowing this, it should no longer come as a surprise that the energy footprint of just a single transaction exceeds the electricity consumption of a typical Chinese household during a full year.
Figure 2: Bitcoin’s energy requirement compared to the electrical energy requirement of a selection of countries as of 5 April 2020 (bitcoinenergyconsumption.com).
The Bitcoin community tends to argue that the associated carbon footprint is not that extreme, because many miners flock to the province of Sichuan in China, where they can get access to cheap excesses of hydroelectric power during the rainy season in the summer. As it is generally assumed that more than half of all Bitcoin miners are located in China, a substantial part of the network would be taking advantage of that. Unfortunately, the excesses available in Sichuan are temporary at best, and shortages do occur during the dry season. During these periods, the grid has to rely on back-up sources like coal- and gas-fired power plants. Since Bitcoin miners require a non-stop supply of energy, they will either add more pressure on the already strained grid during these periods, or migrate to other parts of the country like the coal-rich plains of Inner Mongolia. In either case, this substantially reduces the positive impact of renewable energy on the carbon footprint of Bitcoin. The carbon footprint of a single transaction still equates to well over 50,000 hours of watching YouTube. As if that wasn’t bad enough, the network also generates around 100 grams of electronic waste relative to each transaction. This means that, on average, a complete Bitcoin mining device is discarded roughly every fifty transactions. No amount of “green” energy could possibly make up for this.
Other Blockchain Solutions
After more than a decade since Hal Finney’s early thoughts on the sustainability of Bitcoin, we have to admit that these concerns couldn’t have been more valid. But while the Bitcoin community never inched any closer to a real solution, others have been more proactive. Nowadays, many alternatives to Bitcoin exist that use the same blockchain-technology, but have replaced the proof-of-work system with “green” solutions. The most well-known alternative to proof-of-work is called proof-of-stake, wherein one’s wealth rather than computational power is used for the algorithm . The latter eliminates the incentive to use highly specialized and energy-hungry equipment. Such a solution could ultimately be implemented in the Bitcoin software as well, though the community surrounding Bitcoin typically opposes any and all drastic changes like this. This is likely to remain the case, as other aspects of alternatives like the security of proof-of-stake-based systems remains contested until this day.
In any case, the environmental impact of blockchain technology isn’t necessarily high. Companies will normally avoid using the Bitcoin blockchain anyway, because on top of environmental concerns there are also concerns regarding privacy, costs, and scalability. Popular business solutions like Hyperledger, an open-source modular blockchain framework, don’t require a constant stream of computations by the network participants. These solutions offer a private and permissioned alternative to Bitcoin’s open nature. Because participants have to identify themselves to enter such a network, permissioned blockchains don’t face the same risks as public networks. For example, it won’t be possible for any random malicious entity to take part in the process of creating blocks. This more easily allows for an environmentally friendly design, and one could even argue that embedding proof-of-work in such a system would be a foolish decision since it would add unnecessary risks.
Towards a Positive Contribution to the Environment
Altogether, blockchain technology continues to hold great promise as a powerful tool for disintermediating processes. The Bitcoin blockchain, however, continues to cast its dark shadow over the rest. In order to harness the full power of blockchain, one should start by making a distinction between Bitcoin (an application that uses a very specific blockchain) and blockchain technology in general. Only then we can explore how the technology can actually contribute in the global fight against climate change. Think tanks like the Overseas Development Institute (ODI) already wrote in 2018 that blockchain “can be used to reinforce entitlements to use a natural resource, substantiate claims of reduced environmental impact and incentivize environmentally sustainable actions.” The ODI added, however, that blockchain also “depends on other technologies to reach its full potential.” The blockchain itself cannot observe or test physical reality, Blockchains remain dependent on external input (e.g. a human) to confirm that real world assets actually exist. The blockchain safeguards that this information cannot be changed afterward.
This may be solved by pairing blockchain with other developments in the broader digital ecosystem, like artificial intelligence and the Internet of Things (devises around the world that are connected and can transfer data). If set up properly, blockchain technology may therefore still be the backbone of a more sustainable society.
Alex DE VRIES, Founder, Digiconomist.net