The Single Most Important Truth About Bitcoin Mining, Energy and the Environment

Last week, Greenpeace USA splashed laser-eyed animations of JPMorgan Chase CEO Jamie Dimon and BlackRock CEO Larry Fink on Manhattan skyscrapers, along with shocking statistics about bitcoin mining’s energy use and emissions.

Bitcoin, according to its bombastic ad copy, uses more energy than many countries, emits more carbon than millions of automobiles and its pollution “WILL ONLY GET WORSE WORSE WORSE WORSE!” Four building-width WORSES were stacked vertically, just in case you weren’t persuaded by the first three.

This story is part of CoinDesk's 2023 Mining Week, sponsored by Foundry. Troy Cross is a professor of philosophy and humanities at Reed College, a co-founder of The Nakamoto Project and a fellow at the Bitcoin Policy Institute. You can follow him on Twitter at @thetrocro.

As an environmentalist who has been studying bitcoin mining’s energy footprint since 2011, I have a radically divergent perspective. I think bitcoin holds tremendous promise as a tool for decarbonizing the grid, cleaning up waste methane and accelerating the electrification of heating. I see bitcoin mining’s emissions not as increasing, but steadily trending lower in the future. And I predict it will actually make energy cheaper, not more expensive.

How have Greenpeace USA and I come to such polar opposite views?

Bitcoin’s critics have yet to appreciate the most important fact about Bitcoin’s impact on energy systems: bitcoin mining trends towards using only the cheapest electricity in the world, whenever and wherever that is. At present, most miners already pay well below $0.08 per kilowatt hour (kWh) for their electricity, with some reporting rates as low as $0.02/kWh.

But those numbers will continue to drop in the coming months and years, until it is only profitable to mine bitcoin on free or nearly-free power.

In what follows, I will first support this claim, and then show that the effects of bitcoin buying up nearly-free energy are a net positive for energy systems and the environment.

Why bitcoin miners will only use the world’s cheapest electricity

Here are some undisputed yet underappreciated facts about bitcoin.

• Miners earn bitcoin from two sources: the block subsidy and transaction fees. Transaction fees currently are a negligible percentage of miner revenue, so we will set them aside. The block subsidy is how new bitcoin enters circulation: 900 fresh bitcoins are awarded to miners worldwide every day, roughly in proportion to their computing power.
• This issuance rate drops by half every four years, in an event known as the halving. In April 2024, that supply of new bitcoin will drop to 450 BTC per day and stay there until 2028 when it drops to 225 BTC per day.
• Bitcoin is a fungible good produced from the same input – electricity – at roughly the same rate by the same specialized mining machines (application-specific integrated circuit, or ASICs) wherever they operate around the world. Once mined, bitcoin is sold in a global marketplace and instantly “shipped” to its buyer simply by updating the shared ledger of bitcoin itself.
• Bitcoin miners are interruptible and attenuable; they can turn off and back on quickly, and can fine tune their consumption of electricity, with minimal loss of profit.
• The largest expense of mining – upwards of 70% – is energy.

How do these facts show that mining is headed for nearly-free energy use?

First, the uniformity and availability of the means of bitcoin production – ASICs – and the fungibility of the product means there’s an extremely low barrier to entering this market. The bitcoins you produce will be as good as anyone else’s and you’ll use the same machines to produce them. So if it is possible to mine one bitcoin for less than one bitcoin then someone, somewhere in the world will do it, because people like money.

Second, once the mining market is saturated, those miners with the highest expenses will sit right on the edge of profitability – just enough profit to make it worthwhile to continue – and no one else will be tempted to enter the market unless they can improve on the expenses of that marginal miner. If some new miner does make an entrance during saturation, and they’re profitable, then the existing marginally-profitable miner will go out of business.

Third, miners with access to cheaper energy enjoy a tremendous advantage. If energy is 70% of your cost and you can find a 50% savings on energy, you’ve just cut expenses by 35% making for a dramatic increase in margins. Other business advantages matter too – firmware, cooling, taxes, personnel, access to capital, uptime, etc. – but the 800-pound gorilla is your electricity bill. If you have a significant disadvantage on a cost that accounts for 70% of your expenses, then no matter your other advantages, your business is on the chopping block; cheaper energy reserves will be tapped somewhere in the world until you hoist the white flag and auction off your assets.

Fourth, the halving of bitcoin issuance, which will next happen in April 2024, violently shakes out less-efficient miners every four years. When the total amount of money bitcoin miners are expected to spend worldwide on energy decreasing by nearly 50% in nine months – which it will, absent a doubling in bitcoin’s price or a dramatic spike in the fee market – any miner with thin, or even moderate, margins will be forced to find cheaper energy or perish.

Impacts of an energy cheapskate

So, we can conclude that bitcoin will trend towards the cheapest energy in the world. What follows from that? One immediate implication is that bitcoin is not going to push up utility rates for anyone very much for very long. It simply can’t. If the entry of bitcoin miners into an electricity market raises most users’ prices significantly, those prices will subsequently be higher than other regions of the world and it will be more profitable to mine somewhere else. The miners in the region with newly-expensive power will then be on the chopping block.

Furthermore, miners, since they are interruptible, will never use power at times when power is most expensive – when it costs more than a bitcoin just for the electricity to mine a bitcoin – so will never contribute to the peaks of demand within a grid. Those demand peaks bear much of the responsibility for infrastructure needs and high power costs. While miners may raise price floors – not too significantly or they’ll be outcompeted! – they will most definitely not raise price ceilings.

Another conclusion on the environmental front is far more dramatic: the cheapest energy in the world is energy that no one, at present, wants. This means energy that is stranded in space or produced at the wrong time. Such nearly-free, excess electricity is mostly the result of generation that cannot readily adjust to changes in load (nuclear, hydro, wind or solar).

For a gas plant, if demand drops, one simply throttles back the gas. But when demand drops during some part of the day, or as heavy industry leaves a region, these non-fossil sources of power cannot drop their production as well, and consequently, prices fall. With ambitious plans for deploying renewable generation, such over-abundant pockets of electrical power will only expand in the coming decades.

See also: Bitcoin Mining Is Good for the Environment and Better for the Energy Grid | Opinion

Bitcoin buying up unwanted power produced by intermittent renewables at the wrong time, or by nuclear or hydro facilities in local power markets with insufficient demand, improves the economics of those forms of electricity generation.

An analogy may help here. Suppose a bakery customer agreed ahead of time to buy a certain amount of pastries each day, unless it was a busy day, in which case, they'd cancel their daily order. Suppose the customer also agreed to buy up any unsold pastries at the end of each day. That's bitcoin mining.

Just as the ideal bakery customer would be regular, but also deferential to other buyers, miners agree to buy a certain amount of power in advance, in a predictable way, for a low price, but they turn off their machines to make way for other buyers with higher bids during demand spikes. And just as the bakery buyer lies in wait for wasted baked goods at the days’ end, so too do bitcoin miners with older, less efficient machines wait for the only energy on which they are profitable: the free stuff.

In both cases, the seller – of pastries or power – is able to meet their customer’s demands, but also count on steady revenue as a result of this very special and flexible buyer.

Another consequence of bitcoin’s stingy energy shopping is that it is well-suited to clean up energy that is also waste. Flared methane is not earning any money at all, but can be used to generate electricity and mine bitcoin. It follows that bitcoin miners will seek out flared gas, whether on landfills, at wastewater treatment facilities, farms or oil fields, and take advantage, mitigating what the United Nations calls “the strongest lever we have to slow climate change over the next 25 years.”

Finally, all of the energy entering an ASIC as electricity leaves as heat. Miners who find a way to sell this heat will have a competitive advantage over those who do not. This is already happening in a variety of settings – spas, distilleries, district heating – where it makes sense. And we can infer it will continue if it is profitable.

In sum, bitcoin mining is a nearly perfect market with margins trending to zero, whose most significant input is energy. Given that mining is location-agnostic, scalable and flexible, it will inevitably happen where and when power is cheapest. That means non-rival consumption of wasted energy, which in our current energy landscape is the unavoidable by-product of non-dispatchable generation.

The consumption of what would otherwise be wasted energy promises to be good, not bad, for energy economics, and for the emissions of our energy systems as a whole, given that it improves the economics of non-dispatchable power generation, incentivizes the cleanup of waste methane and incentivizes the electrification of heating.

Real-world complications and the difficult questions that remain

The inferences I’ve drawn above all have caveats and exceptions. Debt can allow miners to operate at a loss, so it might be that in the short term miners thrive who have access to capital rather than cheap energy. Jurisdictional risk can outweigh electrical price advantage – who wants a mining operation that can be seized at any moment by a corrupt government?

In bull runs, ASICs, or other power infrastructure, can be in short supply, so that mining margins remain fat, allowing miners the luxury of consuming expensive energy until ASIC and other infrastructure availability catches up.

Further, the halving event can trigger a bull run of sufficient magnitude that it fails to shake out inefficient players. New electrical generation projects are difficult to finance against promises of bitcoin mining, because the volatility of bitcoin makes lenders uncertain. And cheap electricity from intermittent renewables comes with significant downtime, while ASICs depreciate at a steady rate, posing a challenge for mining purely on renewables’ excess.

Finally, states subsidize energy and regulate energy markets in ways that make my Econ 101 analysis locally fail – e.g., when coal-fired power plants are subsidized, they're suddenly economical ways to mine bitcoin. Any electrical market may be rigged such that bitcoin miners lock in low prices while other buyers are forced to pay more as a result of the increased demand.

All of these qualifications are true, and partially explain the events triggering a barrage of negative bitcoin mining press over the past two years. When China banned bitcoin mining during a historical bull run in price, electrical rates mattered little, and easy money flooded the bitcoin mining industry, leading to a mad scramble for electricity at any price, straining some local grids and keeping open some fossil fuel plants that were scheduled to close.

But the qualifiers above are exceptions that prove the rule. Each represents some kind of temporary aberration or failure of markets, which, in the long run, must give way to the relentless march towards nearly-free energy. Nothing can long keep the most portable, scalable, flexible and price-sensitive energy consumer in history from abundant, nearly-free, unwanted energy. It’s destiny.

See also: Bitcoin Halving Is Coming and Only the Most Efficient Miners Will Survive

Of course, Greenpeace will continue to argue that bitcoin is world-ending while demanding changes to its code. They have shown, in a decades-long campaign against nuclear energy, a willingness to attack a climate-saving, but new and scary technology on the most meager evidence.

But serious conversations among policy makers and industry leaders about bitcoin mining’s impacts on energy systems and the environment must do better. They must investigate this new technology from the ground up, branching out into disagreement about how bitcoin mining happens in the messy real world only after reaching a point of convergence on its essential nature.

That point of convergence, the foundation of rational discourse, is bitcoin mining’s relentless drive towards cheaper energy.

The author would like to thank Andrew Bailey and Colin Brown for feedback on an earlier draft.