Three Things #25: July 10, 2022
Why bitcoin mining is a good thing
I said two weeks ago that I want to write more about Spacemesh. That’s still the plan, but to understand Spacemesh—its design, motivations, and plans—we have to start from the basics and explore Bitcoin a little bit. While the Spacemesh design is more influenced by Ethereum (since both are smart contract platforms), it inherits more of its values and culture from Bitcoin. So, this week, I want to write about Bitcoin.
Bitcoin FUD is never ending and know-nothing journalists are always excited to write the next Bitcoin obituary. While I generally don’t pay attention to it, one perennial topic that I actually find interesting is the question of the energy consumption of bitcoin mining (and of cryptocurrency mining more generally). The topic is interesting because it has surprising depth, complexity, and nuance. Folks like Nic Carter have covered it much more intelligently and thoroughly than I possibly could, but I thought it would be interesting to summarize here the best arguments for why Bitcoin’s energy consumption isn’t nearly as terrible as those dismal journalists would have you believe.
This topic is especially interesting and relevant in light of Ethereum’s upcoming transition to proof of stake, and the launch of Spacemesh and proof of space time. I’ve written and spoken about why I dislike proof of stake, and I think it’s interesting to consider why Bitcoin will never make such a transition. One reason is that, as described below, proof of work mining has some surprising properties.
Thing #1: It’s the Renewables, Dummy!
One of the most common forms of FUD we often hear about bitcoin mining is that it’s polluting the atmosphere, killing polar bears, and the like. There are concerns about mining facilities having an outsized contribution to greenhouse gas emissions, and about old gas and coal facilities being brought back online to mine bitcoin.
The people making these claims haven’t taken the time to look at the data, and they don’t really understand how energy markets work. According to the most recent survey by the Bitcoin Mining Council, which surveyed miners accounting for half of the entire Bitcoin network hashrate, around 58% of bitcoin is currently mined using electricity from sustainable sources. Other estimates put the total even higher, at nearly 75%.
To put things in perspective, only 4% of the energy consumed by the transportation industry and 9% of the energy consumed by factories comes from renewable sources. On average, 7% of energy consumed residentially and 3% of energy consumed by businesses comes from renewable sources. (source) The bitcoin mining industry is lightyears ahead of just about every other industry in terms of its renewable energy consumption mix. It’s also way ahead of the average US energy mix, which comes in at 21% sustainable.
What’s more, numbers such as these are incredibly hard to come by, so bitcoin miners are ahead of the game not just in the sustainability of their energy sourcing but also in their voluntary disclosure of this important information.
It’s also meaningless to talk about the total amount of energy consumed by bitcoin miners without comparing it to the energy consumed by other industries. Bitcoin currently consumes around 92 terawatt-hours (TWh) of electricity annually. This is around 0.14% of all of energy produced globally—which means that, if you were to hypothetically shut down all bitcoin mining globally, it would have effectively no impact whatsoever on emissions (especially given how much of it comes from sustainable sources). For comparison, gold mining consumes 131 TWh per year. The paper and pulp industry—which, among other things, produces physical fiat monetary notes—consumes 586 TWh, or more than 6x what Bitcoin consumes. Air conditioners consume 2200 TWh, or 24x what Bitcoin consumes. And all of these consume way more energy generated from fossil fuels than bitcoin mining.
For more: Read the above-linked Bitcoin Mining Council report to get a better sense of the energy mix used by miners today. This letter to the EPA is also a good read.
Thing #2: Opportunity Cost
To consider cost is only one side of the coin: you also have to consider what you’re buying with this cost. For the sake of comparison, air conditioning uses 24x the energy of bitcoin mining, but… would we really want to go back to a world without air conditioning? I don’t recall hearing anyone say we should ban air conditioners.
Critics like to decry bitcoin mining as wasteful. This is naive and ignores the big picture. What are we buying with that mining? We’re buying a global, decentralized, seizure and censorship resistant, self-sovereign money and ledger system that’s free from government interference. The fact that users are willing to pay fees, both in the form of explicit transaction fees and implicitly in inflation (block rewards), and the fact that miners are willing to pay for the equipment and power needed to mine it shows that lots of people consider the price worth paying. In other words, Bitcoin has clear value to many people, just as other things we choose to use power for are also valuable—like air conditioning.
Secondly, we also have to consider opportunity cost. Bitcoin doesn’t exist in a vacuum, it’s designed to replace existing systems and infrastructure. As I pointed out above, gold mining uses more electricity than Bitcoin. It’s possible we could actually reduce emissions by replacing gold with bitcoin entirely. And that’s just the tip of the iceberg. There’s the energy cost associated with printing, distributing, managing, and storing paper (fiat) money, checks, credit cards, and the like. There’s the cost of building and operating all of the centralized data centers that host the centralized ledger systems of Visa, Mastercard, Amex, and all the world’s banks. There’s an unbelievable amount of redundancy in such a system since each bank network is proprietary and operates independently. Imagine how much energy we could save and how much more productive we’d be if we transitioned all of that commerce onto one open, decentralized, global network, with “data centers” (i.e., miners) colocated next to the cheapest, cleanest energy in the world. Imagine the layers of bureaucracy that could be spared if the ledger could be maintained in a truly decentralized fashion, coordinated by a protocol rather than by dozens of redundant corporate structures, each backed by an army of lawyers and accountants, all doing more or less the same thing.
We also have to consider the fact that a lot of the power that’s used to mine bitcoin simply wouldn’t be used for any other purpose. The key to understanding this is to understand that not all energy is created equal: in other words, energy is far from perfectly fungible. Energy created in the wrong place or at the wrong time, far away from demand or during times of low demand, can easily become “stranded” and wasted. Enormous amounts of such energy are wasted every day (this is referred to as curtailment).
For one thing, energy demand isn’t stable. The grid needs to be able to produce far more than the average load because of peaks: e.g., in hot weather, midday, on weekdays, demand spikes. And energy is notoriously hard to store and hard to transmit (transmission is a much bigger cost than most people realize). So energy producers often have no choice but to curtail such energy.
Bitcoin miners figured this out ages ago. This is why lots of miners are colocated near energy producers to take advantage of energy that would otherwise have gone to waste. And they’re incentivized to do so because of four golden qualities of bitcoin mining: it’s modular, attenuable, location-independent, and interruptible. It’s almost totally unique in this way. Unlike the vast majority of other industries that consume power, bitcoin mining can be turned on and off, i.e., it can make use of energy that’s only intermittently available. Bitcoin miners have a “turn off price,” above which they’d rather shut down than continue to mine, since it would be uneconomical to do so. This “turn off price” is very low compared to other industries, which means that, in fact, Bitcoin is not competing with other use cases: it’s complementary to them, and makes way for them when needed.
And bitcoin mining can happen literally anywhere, even far from population centers and sources of demand for energy. Miners are happy to relocate to wherever they can find cheap power and reliable internet (which can happen pretty much anywhere via satellite). And it’s not at all a problem for the protocol if miners temporarily turn off during demand spikes when energy prices soar: other miners will continue to produce blocks, and the interrupted miners can turn back on at any point and continue to mine profitably. This is simply not possible for factories, which have very long, complex lead times that are required to turn on or off, to scale up or down production, or (of course) to relocate.
For more: Read the books Nic links in this tweet thread (before attempting to discuss the impact of bitcoin mining on the power grid)
Thing #3: You Can’t Stop It
Last month, the New York State legislature passed a bill to ban cryptocurrency mining. As some commentators pointed out, the bill is ridiculous and quite possibly illegal as it proposes to dictate not only how electricity fairly bought on the market can be used, but also what specific computer algorithms the purchaser is allowed to use it to run. (Of course, this wouldn’t be the first ridiculous cryptocurrency-related bill passed by the state.)
The government should absolutely have a say in the supply side, i.e., how power is generated. Climate change is real, and the market has largely failed to capture the negative externalities related to it, a classic Moloch trap. It makes sense for the government to step in and offer sticks and carrots: to incentivize certain types of energy production (those that are cleaner, safer, and more sustainable) and disincentivize others. Similarly, the government can and should help make sure that energy markets are robust and sustainable, that they have enough generation and transmission capacity to meet peak demand, and that critical consumers of electricity, like hospitals, always have enough power.
But how that power is used is none of the government’s business. Consumers and businesses should be free to use power that was legally and fairly bought and paid for however they choose, within the limits of the law. The idea of the government dictating which algorithms we can run on our computers is chilling, and anyway it’s unenforceable. New York State’s proposed law is dystopian and it sounds more like it belongs in Venezuela, famous for finding bitcoin miners by their energy signatures and confiscating them.
The global nature of bitcoin mining is another reason it’s in the interest of policymakers to understand and engage with the industry. When one state attempts to ban mining, it will migrate elsewhere, but the mining will continue. Miners will migrate to wherever they can find the cheapest, greenest sources of energy and the most favorable regulation. And as China’s recent ban shows, miners are happy to move elsewhere when those conditions change. Other things being equal—the emissions associated with bitcoin mining are going to happen anyway, somewhere—isn’t it better if it happens close to home, in a country like the USA where relatively green energy is abundant, and where rule of law and competent enforcement mechanisms allow us to keep a close eye on miners and make sure they’re holding up their end of the bargain? And wouldn’t we rather have the associated jobs close to home?
Finally, putting things in context, the story of human society over the past century is one of more and more efficient use of energy. Society thrives when energy is abundant, dense, and cheap. Even if you could theoretically stop bitcoin mining, you wouldn’t want to, because it has an important role to play in our energy transition. The world needs to build a lot more electrical capacity in the coming years. We’re in the midst of a major transition from fossil fuels to electrification of everything: cars, trains, and buses to HVAC to heavy industries and everything in between. The only way we stand a chance of meeting greenhouse gas reduction goals is to speed up this process of transition, and electrify basically everything. Bitcoin mining is a great bootstrapping mechanism to get there. Due to the unique nature of the mining process and the “four golden properties” described above, bitcoin mining can be used to bootstrap new renewable energy projects regardless of location, even far from population centers, before they’re connected to the power grid, and before local demand rises to meet supply. As discussed above, it can also be used to monetize excess supply in times when demand is low and that supply would otherwise be wasted.
For more: Set up a miner! You can find instructions easily enough. Miners are cheap right now! Note that they tend to be large, loud, hot, and power hungry, so if you don’t live in a home with a basement or a spare shed, in a cool climate, with cheap power, you might want to find a friend who does.