In March, The New York Times declared to its readers that a “single bitcoin transaction now requires more than 2,000 kilowatt-hours of electricity, or enough energy to power the average American household for 73 days”.
The claim will have seemed credible to the vast majority of people who eye bitcoin with deep suspicion. It coincided with the launch of a lobbying campaign by Greenpeace, which directly linked bitcoin’s proof-of-work mining system to climate change. Around the same time, lawmakers in the European Parliament voted on a proposal to ban bitcoin’s energy-intensive form of mining (but decided against doing so).
For anyone not intimately familiar with the design and purpose of the bitcoin network, it’s becoming increasingly difficult to support the cryptocurrency without muzzling your environmental conscience. Yet bitcoin is not the gas-guzzler depicted by The New York Times and others.
What the newspaper should have told its readers is the following: the energy required to keep bitcoin running is so vast that – when measured against the currently low levels of user engagement – individual transactions appear to reflect a high capital cost. That statement is factually accurate; it acknowledges that bitcoin’s environmental impact would be harmful if the network were not scaled up efficiently; but it also places the status quo in its proper context. Bitcoin transactions categorically do not use 2,000 kilowatt-hours of electricity; they occur in the blink of an eye with the entry of a short string of text into the network’s transaction ledger; they cost no more than sending an email. Of course it’s true that, behind the scenes, the towering digital infrastructure that supports bitcoin – like the sprawling web of data centers and exchange points that hosts the internet – is capital-intensive and costly to maintain. But, like the internet, bitcoin is worth it.
Imagine an airline with one plane and one route: London to New York. On Christmas Eve, the airline carries a full payload of 200 passengers to New York. Its next scheduled departure from London, on Boxing Day, is also fully booked. But only 20 seats have been filled on the intervening return leg from New York, as most people avoid traveling on Christmas Day. Environmental critics might decry this latter flight as socially irresponsible, arguing that each passenger’s carbon footprint is 10 times higher than it needs to be. But if the airline didn’t run the return leg, its plane would be stranded in New York and it couldn’t carry any of the (highly efficient) Boxing Day passengers. The unpopular Christmas Day service is a scheduling anomaly – necessary for operational reasons, but in no way illustrative of the airline’s carbon footprint.
This essay will make the argument that bitcoin’s current stage of development should be seen through the same prism of long-term averages and projected utility. An airline cannot be judged on the performance of one individual flight. A manufacturing plant cannot be judged after its first product rolls off its first production line. And bitcoin – which is a peer-to-peer monetary network designed to protect the interests of billions of global citizens – cannot be judged after a few million prescient individuals cotton onto its benefits.
It’s entirely normal for society to be confounded and skeptical at the advent of a technological revolution. People struggle to understand how and why their old ways of doing things are becoming obsolete. They worry about the changes that will bear down on their future selves. Any attempt to write objectively about bitcoin must, therefore, be led by the facts: accurate, unambiguous ones. And it must be accompanied with a firm caveat that we do not and cannot truly know the extent to which this technology will be good, bad or neutral for humanity. Predicting social, economic and environmental change is a purely conjectural endeavor. The best approach to take – but the one that most people seem unwilling to do – is to invest your time in better understanding novel concepts, and in humbly contemplating how they may, hypothetically, impact on your world of tomorrow.
With that in mind, let’s begin with a refresher on how bitcoin functions as a monetary network and why it’s deliberately been designed in this way. Only once we understand these basics can we assess bitcoin’s carbon footprint from an informed perspective: what purpose it serves; whether it’s too large; and, ultimately, whether it’s detrimental or beneficial to society.
Why does bitcoin use so much electricity?
Bitcoin is the first decentralized monetary blockchain. That means it relies not on a central authority like a bank to move money around, but on multiple identical copies of one ledger – distributed all over the world – which are updated about every 10 minutes with a new “block” of data. Each block confirms details of the latest transactions across the network. So if you want to send me some bitcoin, your proposed transaction enters a queue and sooner or later (depending on the fee you paid) it’ll be confirmed in a newly mined block. That’s the point at which the coins are moved to my wallet, where they remain until I make a new transaction with my private key. All the copies of the ledger circulating around the world verify that this transaction took place, so everyone participating in the network agrees that I now own the coins. If someone wanted to steal them by rewriting the ledger, they would need to mine a new block that contains different historical data to all the other copies of the ledger. That kind of reverse engineering is impossible thanks to the mathematical process through which new blocks are mined (cryptographic hashing). The fraudulent block would be automatically rejected by the network.
Now, don’t worry if you don’t understand the precise way in which cryptographic hashing achieves immutable data. I’m a journalist – not a mathematician – and I don’t understand it either. But I also don’t understand the aerodynamic processes that allow planes to fly, and that doesn’t stop me boarding them in total confidence. The relevant point is this: for several decades, the world’s greatest mathematical minds have dissected public-private key cryptography and cryptographic hashing to the nth degree. No flaw has ever been found with blockchain technology’s use of them. Mere mortals like you and I can safely assume that these algorithmic functions are mathematically pure, and that they form a sound basis for distributed ledgers. If you’re not comfortable with that assumption, I suggest you neither buy bitcoin nor board any more planes.
None of which, you’ll notice, answers the question in my subheading: “Why does bitcoin use so much electricity?”. It does lay the groundwork for that explanation, though, if you’ll just stay with me.
Recall that any attempt to fraudulently alter a blockchain’s history will be automatically rejected by the network. That’s fine. Very good. But who is the network? At the time of writing, the network is 15,259 people sharing copies of the ledger (nodes) and an unknown number of people attempting to mine the next block (miners). The latter group deploy power-hungry computer hardware in a race to guess the unique cryptographic hash they’ll need to mine that block. Why bother? Because, when they’re successful, they earn newly minted bitcoin for their trouble. And why is the network designed around this convoluted proof-of-work game? Because energy is security. If it were easy to mine and validate fraudulent blocks, then it would be easy for hackers to launch a “51% attack” – a scenario in which more than half of the network is controlled by nefarious actors. Put another way, if fraudsters or hostile governments wanted to steal your bitcoin, they’d be able to do so. Bitcoin would be useless.
Think once again of aviation. It’s not enough for an aircraft manufacturer to understand aerodynamics and design principles on paper. The likes of Boeing and Airbus must take that scientific knowledge and apply it in the real world through a robust engineering process that uses the best raw materials, components and practices. Only then can they build a safe, reliable plane that people are willing to fly on. The same holds true for bitcoin. Without an energy-intensive mining infrastructure, bitcoin would lack security and would have no appeal as a monetary network. It would exist solely as an idea. That’s why it matters who the network is, and that’s why only a costly and arduous process like proof-of-work mining can ensure that the wide interests of society trump the narrow interests of thieves and aggressors.
What’s the cost of this magic trick?
Estimates of the network’s total energy consumption vary – due in large part to the anonymous nature of mining – but the most rigorous analysis by the University of Cambridge currently pegs it at about 138 terawatt-hours per year. Tech website Digiconomist suggests it’s more like 205 terawatt-hours per year. Critics fret that these figures are comparable to the total primary energy consumption of a small country like Ireland.
However, such analogies are misleading. Society cannot judge the cost of something until it first identifies and agrees on the “thing” being costed. In Ireland’s case, the energy consumed by the country fulfills – on aggregate – a relatively clear and uncontroversial purpose; namely, economic life. That includes electrifying homes and businesses; powering industries and public services; shuttling around people and goods, and so on. Of course, much energy is wasted at a microcosmic level. There’s much room for criticism of Ireland’s carbon footprint, as with any country. But, on a macro scale, even the most radical environmentalist wouldn’t dare to contend that Ireland should cease all energy consumption. The “thing” being costed is the very ability of the State and its people to function, and that’s got to be worth a few emissions.
When it comes to bitcoin, things are more nebulous. Thirteen years after the cryptocurrency was created, humanity is still only scratching the surface of what decentralized sovereign digital money means for our species. We’ve only begun to play with ideas: some farcical, like meme-coins; some exploitative, like pump-and-dump schemes; some so potent they scare superpowers into banning them outright, like bitcoin and decentralized finance. The world’s richest man thinks that a joke cryptocurrency with an odd-looking dog on its imaginary coinage may one day become viable money, despite its glaring security flaws. He’s wrong. So are most people – supporters and critics alike – who speak with confidence on the subject. And yet bitcoin’s potential, if fulfilled, could not be more worthy of attention: the minting of a new form of money that society – not government – imbues with value and fungibility; one that’s more secure than any physical asset that’s ever existed; that’s easier to transact between people and across borders than any other medium of exchange; and that has such a puritanically classical approach to monetary policy that it banishes the threat of inflation by fixing supply for all eternity.
It’s a lot to take in, especially considering that none of it’s actually happened yet. But, if we return to the analogy with Ireland, this abstraction becomes quite enlightening. Everyone agrees what Ireland’s economy is and why it should be allowed to exist. Very few people agree about much to do with bitcoin. How, then, can any of us claim to authoritatively define it for costing purposes?
If you’re in the camp that sees bitcoin as a frivolous game with no real-world application then, of course, its carbon footprint seems abhorrent and it should probably be banned. With each passing year, though, the number of people dismissing bitcoin in this way shrinks. Younger generations who have only ever experienced rising inequality, falling opportunity, and negative real interest rates have a strong incentive to consider new economic models. Their digital intuition also softens the learning curve for bitcoin. This is no “frivolous game” to them. Nor to Africans who have spent a lifetime excluded from the financial platforms and instruments that Westerners take for granted, and that in many cases reap their profits from the developing world. Nor to Iranians nor Venezuelans whose life savings could at any moment evaporate in a puff of hyperinflation. These are complex, nuanced issues that each deserve their own debate. But the point is universal: if bitcoin is a force for good in the world – if it moves humanity away from an epoch in which governments control and manipulate the money supply; towards an epoch in which money is a personal, incorruptible asset – then the justifiable cost of running the network is immeasurably higher than the environmentalists suggest. Even a seemingly exorbitant cost will still have an immense return on investment.
Attempts have been made to quantify bitcoin’s utility – for example, by comparing its market cap ($900 billion) with Ireland’s GDP ($418 billion) – but none is able to capture its abstract benefits or future potential. Predicting bitcoin’s eventual scale is, as already mentioned, impossible. So the best metric to fall back on is comparative efficiency: is bitcoin more environmentally efficient than the existing global financial infrastructure?
A better question: How efficient is it?
We have to be careful here. Most of what’s been written about bitcoin’s environmental credentials is deeply biased, with authors picking and choosing the unscientific statistics that validate their stance. Those who defend bitcoin mining solely on the basis of renewable-energy consumption gloss over the fact that little is known about the make-up of fuel sources powering the network. Once again, therefore, let’s stick with the facts and be mindful not to over-interpret them.
The starting point has to be an acknowledgment that the existing global financial infrastructure is far from environmentally friendly. Statistics are difficult to come by and those that exist are riddled with misleading information. Greenpeace’s analysis of the UK financial sector, for example, blamed it for 805 million tonnes of CO2 emissions in 2019 (nearly double the UK’s entire carbon footprint). But it reached that figure by including “financed emissions”, or indirect emissions arising from the loans and investments made by financial institutions. Clearly, this is an unhelpful methodology designed to inflate the numbers. The researchers said they broke down their data into more detailed channels that would give a more accurate picture – the use of company offices and vehicles; the use of leased assets; commuting and business traveling by employees, and so on – but no specifics appeared in their report. All we can say with confidence is that the global financial sector has a significant environmental cost. Its skyscrapers, computer systems and jet-setting bankers are not helping climate change. We can also safely assume that central banks and their money-printers are no greener.
Against this backdrop, bitcoin’s use of energy begins to feel less like a dirty secret and more like a clean break. When you remove centralization, you remove the need for a vast number of financial middlemen. The advent of algorithmic “smart contracts” means that almost nothing that traditional banks do is beyond the scope of bitcoin and decentralized finance. Combine these efficiency gains with the societal advantages of financial sovereignty, and you can see why some people argue that the environmental cost of bitcoin mining is justified. However, there’s a more compelling argument to be made.
The biggest obstacle to the global expansion of renewable-energy harvesting is the “stranded power” problem that handicaps developing nations. It’s one thing for a poor country to build a wind, solar or hydropower plant; foreign investors are queuing up to finance such projects. It’s another thing altogether to connect that plant – which is likely located in a remote region – to the developing nation’s power grid. You don’t just need a functioning plant to generate energy; you need functioning infrastructure across the length and breadth of the country to suck up that energy and distribute it to people who can use and pay for it. Without powerlines and transformers, and without stable demand from local populations, Africa’s renewable energy supplies are simply not commercially viable. It’s a bitter blow. If only there were a way to build those renewable plants in those developing countries; to serve communities to the maximum extent that infrastructure and local demand permits; and to convert all the surplus energy into a guaranteed financial return, 24 hours a day, 365 days a year. If you’re a bitcoiner, it’s hard not to smirk at that sentence. The solution – bitcoin mining – is apparent to all who understand this technology. Advocacy groups like Ethiopia’s Project Mano are working hard to raise awareness among governments. All it takes is time and effort.
There are other examples of bitcoin solving an environmental problem that traditional finance cannot. In the developed world, gas flaring at oil fields is tolerated as a necessary evil due to the logistical and commercial impracticality of transporting the fossil fuel for processing. Excess gas from wells is vented and burned off, creating nothing but a visual spectacle and ozone-damaging methane. Thanks to bitcoin mining, oilfields across the US are now feeding that gas into on-site generators and creating value from otherwise wasted combustion.
This unique ability to harness stranded and wasted energy adds weight to the argument that bitcoin may indeed – for all its emissions, and as absurd as it sounds – be a green technology. But there’s one final piece of the efficiency puzzle.
Bitcoin is often criticized for the relatively small number of transactions that its blockchain is able to process by design: the network’s primary layer currently handles about 4.6 transactions per second, versus 1,700 by payment processor Visa. Speeding things up to match – let alone surpass – Visa’s capability means either increasing the amount of data in each block, or reducing the amount of time it takes to generate blocks. The former would lead to centralization, as fewer nodes would have the bandwidth needed to keep pace. The latter would compromise security in a different way, as blocks could not be propagated quickly enough to allow nodes to fulfill their oversight function.
Critics claim that this inherent limitation is the Achilles’ heel of bitcoin – guaranteeing that the network will never be scalable; transactions will never be truly efficient; and the environmental costs will always be disproportionate. They are wrong. Bitcoin is no more bound by the limitations of its primary-layer blockchain than the internet is bound by the limitations of the TCP/IP protocol on which it was launched in 1983. Second-layer solutions such as the Lightning network are already in widespread use, bundling a theoretically infinite number of off-chain payments into a single on-chain transaction.
Bitcoin is more secure, more egalitarian and more efficient than any other form of money; it incentivizes renewable-energy harvesting and optimizes non-renewable plants; and it’s already the most scalable payment system ever developed. We cannot know the abstract ways in which bitcoin will shape our world. But noone who truly understands bitcoin fails to see it as a technological and financial leap forward for humanity.