Bitcoin & The Environment, Pt. 1: Why China (Still) Matters
Laying the groundwork for a provincial-level analysis of Chinese electricity generation and Bitcoin mining
Bitcoin: Green or Otherwise?
“Bitcoin mining’s environmental reputation is less than stellar.” Bitcoiners Troy Cross and Andrew Bailey have it right. Both proponents and detractors of Bitcoin can agree that its environmental impact has garnered no shortage of negative press.
Past that, they are unlikely to agree on much.The topic of crypto’s environmental footprint is polarized and vitriolic. Numerous commentators claim that Bitcoin is an environmental disaster, enriching a favored few and consuming profligate amounts of electricity while providing limited social benefit.
Bitcoin’s defenders strongly contest these claims. Some note that Bitcoin compares favorably with industries such as gold mining and banking. Others argue that Bitcoin utilizes stranded assets that would otherwise go to waste, creating positive value where previously none existed.
Adjudicating between the two requires a clear-eyed assessment of the sources of electricity that power Bitcoin mining. Without knowing how clean or dirty a miner’s power source is, one cannot reliably assess its carbon & environmental footprint.
This can often be an arduous task. Christopher Bendiksen, Head of Research at the digital asset management firm CoinShares, put it this way:
“In order to assess Bitcoin’s energy use, you can’t just look at the energy mix in a country as a whole; you have to zoom in on specific regions, down to provinces, cities, and sometimes individual mining facilities.”
In a Harvard Business Review piece, Nic Carter concurs:
“Furthermore, many high profile analyses generalize energy mix at the country level, leading to an inaccurate portrait of countries such as China, which has an extremely diverse energy landscape.”
Seems simple enough. And yet, discussions of Bitcoin’s electricity consumption and its provenance remain heavy on furor and light on facts.
Why is this? More importantly, what can we do about it?
What’s The Controversy?
Until recently, the overwhelming majority of Bitcoin mining occurred in China. Although many Chinese-language sources exist that could shed light on how much electricity mining has consumed, few analysts in the wider Bitcoin community have given these primary sources a deeper examination. To date, not much has been published in English, cryptocurrency’s lingua-franca, regarding China’s provincial-level electricity mix and its connection to crypto.
Of the sources that have broken down Bitcoin’s mining footprint, by far the best is the Cambridge University’s Center for Alternative Finance and their China Bitcoin Electricity Consumption Index (CBECI). Virtually every source that talks about the issue references the CBECI as a key pillar of their argument, if not the foundation.
The CBECI is excellent, but it is not flawless. We’ll point out some of its limitations in future pieces. More importantly: no matter how good the CBECI may be, placing undue emphasis on a single point of failure is as unpalatable in research as it is in crypto. To paraphrase Do Kwon, co-founder of the popular Terra blockchain, “decentralized currency merits decentralized analysis.” Including Chinese sources lessens our reliance on a single analytics provider.
Data quality is a third issue precluding adequate analysis of Bitcoin’s ecological footprint. CBECI notwithstanding, the sources invoked to support or criticize Bitcoin’s environmental impact are often biased, several years out of date, or misquoted. A full literature review is beyond the scope of both this article and the pieces that will follow it, but in later editions, we will focus on pieces centered around China’s data.
The debate around Bitcoin’s electricity footprint deserves research that is up-to-date, accurate, and objective — or as objective as possible for a contentious topic. This research, in brief, is what we aim to provide.
Let’s get to it.
Objective
In brief, this article seeks to outline a cogent case for including Chinese-language sources in the discussion of Bitcoin’s mining footprint. Several difficulties confound simple analysis of Bitcoin mining. We seek to highlight these difficulties and to propose a course of analysis that addresses them.
In doing so, we hope to lay a foundation for future analysis, including:
- Bitcoin’s electricity footprint in 2022, as well as scenarios that may unfold beyond 2022
- The changing role of China as a player in the space
- The electricity consumption of Ethereum and other major cryptocurrencies
- The opportunities and constraints of using renewable power for cryptocurrency creation
- The extent to which low-cost stranded assets can power the Bitcoin network
This piece is the first in a series of articles that will examine Bitcoin mining’s electricity demand in China in 2020 (the most recent year for which twelve months of data is available). Specifically, we’ll review mining in three locations: Xinjiang, Sichuan, and Yunnan. As recently as mid-2021, these regions collectively accounted for nearly half of global Bitcoin mining. Studying them can illuminate how the Bitcoin mining industry has sourced its electricity, and how it may continue to do so.
At the time of writing in early 2022, the state of mining is in flux. Geopolitical turmoil in Kazakhstan and Russia has clouded attempts to definitively trace Bitcoin’s current mining operations. Put another way: it will be some time until we can accurately assess the current state of global mining. Yet, while we cannot reliably assess the present state of mining, we can gain a better understanding of its past, in order to illuminate assumptions about its potential future.
We are aware that China is changing its own energy mix. We acknowledge that miners have been leaving China en masse due to regulatory crackdown, and that novel applications of existing resources — such as using flared natural gas to fuel Bitcoin mining — may increase in the coming years. As well, we note that some members of the crypto community believe crypto mining could incentivize the production of more renewable energy.
These future possibilities will be a topic for future pieces. For now, our main purpose is to bring more Chinese sources into the debate over crypto-mining. We believe that doing so will empower reasonable discourse around Bitcoin’s past and present environmental impact. Even (especially?) if you support Bitcoin, a nuanced understanding of its past and present electricity consumption can be valuable to you.
Well-known proponents of crypto, e.g. Jack Dorsey and Cathie Wood, have pointed out that hesitation around environmental consequences has made people and institutions stop buying. Calling these concerns “FUD” is unlikely to change that hesitation. If mass acceptance for cryptocurrency is ever to occur, detailed answers to the “Bitcoin is an environmental disaster” line will be necessary.
Below, we address two major objections to conducting further study of Bitcoin’s electricity consumption:
- First, that the data surrounding Bitcoin mining is clear-cut and easily interpreted to prove straightforward conclusions about its environmental consequences
- Second, that China is not worth bothering with now that mining has largely fled the country
If you’re already sold that these propositions are lacking, feel free to skip to the end. Conversely, if you’re dead-set on believing that historical analysis is a waste of time or that Bitcoin mining is clearly a) boiling the oceans or b) overwhelmingly powered by clean sources, close your browser and accept our best wishes on your crusade.
And now to business.
Reviewing the Discourse
A common refrain in media pieces about crypto is that Bitcoin is not just a negative force to the environment — it’s a major negative force.
Consider the following statement from The Economist:
“As the price of bitcoin has soared, so has its thirst for energy. According to the Cambridge Bitcoin Electricity Consumption Index, a tool created by researchers at the British university, in the week ending May 13th bitcoin mining used electricity at a rate equivalent to 150 terawatt-hours per year (see chart), more than the entire annual energy consumption of the Netherlands.”
Seems pretty cut-and-dried, doesn’t it?
Let’s fact-check the claim.
According to the Netherlands’ Central Bureau of Statistics, annual energy consumption was 3,080 Petajoules (PJ) in 2019. Converting this figure to terawatt-hours produces a figure of 855.5 TWh. In other words: the Bitcoin network doesn’t consume more energy annually than the Netherlands — the Netherlands consumes more than five and a half times the energy that Bitcoin does. Hm.
At best, the original statement is lazily written. Taken literally, it’s flat-out-wrong.
But let’s give The Economist the benefit of the doubt. We’ll edit the statement: Bitcoin mining consumes more electricity annually than the Netherlands.
Okay, you might say, but you’re just quibbling over terminology. Besides, isn’t an entire country’s worth of electricity still an astronomical amount?
To assess that claim, let’s bring in another source. According to a September 2021 New York Times piece, the Bitcoin network consumes about as much electricity as Washington State. They even made an awesome graphic to display it:
Set the magic of the graphic aside. Let’s compare the two statements:
- Bitcoin mining consumes as much power as the Netherlands
- Bitcoin mining consumes as much power as Washington State
Both are true. But “Bitcoin consumes as much electricity as Washington State” doesn’t exactly command the same rhetorical gravitas as “Bitcoin consumes as much electricity as an entire country.”
Let’s go a step further. Here’s another summary of Bitcoin’s electricity consumption, from a recent report by NYDIG:
“Bitcoin’s absolute electricity consumption and carbon emissions represent…0.2 percent of global electricity generation, and 0.1 percent of global carbon emissions.”
“Two-tenths of a percent” implies inconsequential impact. A similar flavor of argument suggests that Bitcoin uses dramatically less energy than traditional banking or store-of-value assets:
And yet another take, from a Q3 2021 Bitcoin Mining Council study, suggests that the Bitcoin network consumes slightly less electricity than holiday lights.
The point of all this is that reasonable observers can interpret existing data to reach wildly different outcomes about Bitcoin’s environmental footprint — and all of them can be correct. Depending on your priors — the beliefs and assumptions that you, the reader, bring to the table — you can easily make a data-driven case for two strongly opposing viewpoints.
Moreover, even the process of interpreting existing data can lead to a range of conclusions. Here’s a figure from the Global Cryptoasset Benchmarking Study that the Cambridge team ran:
In the very same report, the team shared an “alternative methodological approach”, suggesting a 29% figure:
“In July 2019, CCAF launched a real-time estimate of Bitcoin’s electricity consumption, the Cambridge Bitcoin Electricity Consumption Index (CBECI), which was followed by the release in May 2020 of an interactive map of the geographic distribution of Bitcoin hashpower from September 2019 to April 2020… The findings from the map combined with survey data for SHA-256 hashers offer an alternative methodological approach to estimate the energy mix of Bitcoin mining. The results of this top-down calculation are displayed in Table 3 and indicates (sic) that about 29% of Bitcoin mining is powered by renewables.” (emphasis ours)
Here’s a handy table that displays the 29% estimate for the miner-reported share of renewables:
At the risk of stating the obvious, the difference between 29% or 39% is significant. If even leading research institutions acknowledge significant uncertainty around the true figure, then the rest of us should be cognizant of the challenges facing this type of analysis. Close examination of primary data sources may help resolve some of these challenges.
Why Bother With China?
Even if you’re convinced that the data merits a closer look, you might wonder: why does one need to examine China, if Chinese mining has practically disappeared?
Here’s the CBECI’s display of Bitcoin mining by country:
The display shows the overwhelming prominence of China on the global mining scene — until summer 2021, when it dropped to nearly zero.
Many might assume that with China’s new laws against mining, and the movement of the industry to places like the USA and Kazakhstan, a deeper look at China’s contribution to crypto’s energy mix is not important anymore. It’s undeniable that the global landscape has changed dramatically since 2020, when China accounted for two-thirds of Bitcoin’s total hashrate.
So why are we — a former renewable energy analyst and a professional Chinese translator — proposing a thorough investigation of Chinese-language sources on Bitcoin mining?
To be clear: the CBECI is awesome. To their credit, the team behind it transparently lays out the model’s methodology and limitations.
But conclusions shouldn’t be based on one source alone, no matter how good that source is. Here’s one reason why, based on an excerpt that describes how the CBECI’s works (emphasis ours):
We have partnered with several Bitcoin mining pools to collect geolocational mining facility data in a non-obtrusive and privacy-preserving manner. This geolocational data is based on IP addresses of mining facility operators (‘hashers’) that connect to the servers of mining pools.
Assumption 1: IP addresses of mining facility operators are an accurate indicator of hashrate location…
The quoted section goes on to note that to preserve the privacy of cooperating miners, the Cambridge University team doesn’t have access to the underlying IP addresses.
The limitation here should be obvious: the survey relies on mining pools to provide accurate IP addresses. No mining rig operator is going to provide this data during a government-mandated mining crackdown. Of course we’d expect the CBECI’s method to produce a 0% answer.
Other sources contest the view that mining has left China entirely. If these articles are true, they conflict with the chart above that shows how China’s share faded to nothing in mid-2021. Moreover, this wouldn’t be the first time in China that rules have been put in place, enterprising individuals find ways around them, and rules subsequently loosen. Just look at VPNs, video game usage, video game consoles, dual-citizenship, or how families are maneuvering around the new regulations on after-school education. Moreover, the political climate in China in 2022 may not be representative of the climate in, say, 2025 (more on that in upcoming pieces). If illegal mining activities in China continue, it is crucial that researchers remain on top of the electricity situation in the country.
Conversely, if Chinese mining does fade altogether, a study of 2020 may help us better grasp the current facts on the ground. Recent articles suggest that mining may have become dirtier since China’s crackdown last year. To reliably assess whether Bitcoin’s electricity sources have gotten cleaner or dirtier recently, we need to know how clean or dirty they were before. Historical analysis can tell us that — and historical analysis means taking a close look at China.
With many governments, including that of the United States, having discussions about this topic — and some even considering more regulations themselves — the future is not yet set in stone. If miners eventually move to a country with a similar political landscape as China, a study like ours may act as a blueprint for later investigations.
What’s Next?
China-focused analysis remains valuable for those who seek to understand Bitcoin’s environmental footprint, both today and in the future. We aren’t embarking on providing more insight into China’s energy mix just because we feel previous discussions haven’t zoomed in enough. We also want to shine a light on the fact that despite all that has happened in the past year, China’s role in the crypto world shouldn’t be regarded as dead and done. If this topic interests you, keep an eye out for the next installment in this series.
—
About the Authors
Christian Dean Manley 司马天
Christian is a British American with an undergraduate degree in International Relations and Medieval History from the University of St Andrews and a Masters on Chinese affairs from the London School of Economics. His adventures with China have taken him anywhere from writing articles on current affairs to selling cider to bars in multiple cities. Currently, he lives in Chicago and works as a freelance consultant for Chinese gaming companies wishing to expand their business internationally. In his downtime, you can find him traveling, biking, or hiding behind a rock in Destiny 2 as GJ protects him from any (most) players better than him.
GJ Flannery 范哲
GJ is the community lead at MetricsDAO, special projects contributor at Angel Protocol, and Senior Manager of Special Projects at Flipside Crypto. Prior to joining Flipside, GJ spent five years as an analyst & consultant for several Fortune 500 energy companies, e.g. First Solar and Centerpoint Energy, including several months on-site in locations ranging from Manila to Pennsylvania to West Texas. He has a master’s degree in international economics & Mandarin Chinese, specializing in China’s role in the global energy trade, as well as a passionate belief in the power of both renewable energy & decentralized finance to empower individuals and communities.
