Bitcoin’s Energy Usage Isn’t a Problem. Here’s Why.

Bitcoin’s energy usage has been in the news for years.

It’s often criti­cized for using too much energy, not making efficient use of its energy, or in extreme cases, being an outright climate/energy disaster.

For example, back in December 2017, Newsweek ran a piece called,

Well, here in 2021, I can obviously say that didn’t happen. The Univer­sity of Cambridge is the most-cited source for estimating Bitcoin’s energy consump­tion, and they placed its estimated peak annual­ized rate at under 140 TWh so far. 

Since the world uses over 170,000 TWh of energy per year, the entire Bitcoin network, at its peak estimated consump­tion level, uses less than 0.1% of the world’s energy consump­tion. That’s for a network with 100+ million estimated users. 

Bitcoin’s energy usage is a rounding error as far as global energy usage is concerned. And I mean that liter­ally when scien­tists estimate that the world uses a certain amount of energy in a given year, they can easily be off by a couple percentage points in either direc­tion, let alone a couple tenths of a percent. Bitcoin is estimated to use less than one-tenth of one percent.

Plus, a large percentage of the energy that Bitcoin uses is from other­wise-wasted energy sources.

In the very long run, if Bitcoin is wildly successful and becomes a system­i­cally impor­tant asset and payment system used by over a billion people at 10 — 20x its current market capital­iza­tion, it should max out at just several tenths of one percent of global energy usage.

On the other hand, if it is unsuc­cessful and doesn’t grow much from current levels, its energy usage will stagnate and shrink as the block subsi­dies continue to diminish. I’ll dive into those numbers later in this piece.

The point is that the Bitcoin network is and forever will be a rounding error as far as global energy consump­tion is concerned, whether it’s successful or not. Its energy usage won’t exceed its long-run utility (however high or low that utility ends up being).

In fact, as someone with a background in electrical engineering, I was drawn to Bitcoin in the first place by seeing how efficiently its network uses energy. I could have picked any blockchain to invest in or could have avoided the digital asset space altogether, or could switch my invest­ment to another blockchain at any time. And yet, when I crunch the numbers, I think Bitcoin makes partic­u­larly elegant usage of energy and is getting more energy efficient over time because it was designed properly.

If that’s the case and Bitcoin’s energy usage is practi­cally irrel­e­vant on the global scale, how can journal­ists make such large, sensa­tion­al­ized errors? The answer is that it often comes down to them not under­standing the scaling process that the network is going through.

And it’s easier to sensa­tion­alize things for pageviews or polit­ical gain. For example, it’s commonly said that the Bitcoin network uses more energy than some countries. That’s true, but so does Google, Youtube, Netflix, Facebook, Amazon, the cruise industry, Christmas lights, house­hold drying machines, private jets, the zinc industry, and basically any other sizable platform or industry. From that list, Bitcoin’s energy usage is the closest to that of the cruise indus­try energy usage, but Bitcoins are used by more people, and the network scales far better.

It’s impor­tant to under­stand whether or not Bitcoin is an environ­mental problem and whether or not it is energy efficient because these obviously affect its proba­bility of being a good invest­ment and a good payment network.

If Bitcoin did indeed have serious energy scaling problems, it would eventu­ally fail against competi­tors in the private market­place by not offering enough utility for its energy consumption.

From an engineering perspec­tive, Bitcoin’s energy usage isn’t a problem when you actually run the numbers. Still, it takes an under­standing of how it works to calcu­late it properly and what the trade-offs are if you use a different approach than what Bitcoin uses.

Let’s take a look at how that works.

Article Chapters: 

• Under­standing Bitcoin’s Purpose

• Why Bitcoin Uses Electricity

• Bitcoin’s Efficient Scaling Pattern

• How Bitcoin Uses Other­wise-Wasted Energy

• Bitcoin’s Proof-of-Work vs Alter­na­tive Methods

• Final Thoughts and Summary

Understanding Bitcoin’s Purpose

Before we dive into its energy usage, it helps to summa­rize what bitcoins are used for, and what problems the network was designed to optimize for. From there, we can then look at the energy usage and decide whether or not it’s achieving its goal.

Bitcoin is not trying to be the cheapest payment network, although in many contexts it ironi­cally can be, when you consider the Light­ning network (one of Bitcoin’s second-layer scaling systems).

Instead, Bitcoin is trying to be a decen­tral­ized bearer asset and payment network, and for 13 years, has been succeeding.

In terms of the asset, bitcoins are something that users can self-custody with encryp­tion if they want to, and that has a fixed supply. This makes bitcoins infla­tion-resis­tant and diffi­cult to confis­cate. In terms of the payment network, Bitcoin can be used to send payments that don’t rely on the permis­sion or verifi­ca­tion from any central­ized entity. This makes the network censor­ship-resis­tant and inter­op­er­able with many different payment systems internationally.

Almost anything that claims to be more efficient than the Bitcoin network at making payments or storing assets for you is also more central­ized. Central­ized things inher­ently tend to be efficient. Sending payments between parties and tracking peoples’ balances can be as simple as updating an internal database, which is nearly free.

The problem, of course, is that central­ized things don’t tend to be resilient, so there is a trade-off between efficiency and resiliency. When you put all your eggs in one basket, give that basket to your friend, and she drops the basket or refuses to give it back, you’re out of luck. Relying on a central­ized entity cedes control to others, which rests on the premise that those others are moral and compe­tent. Plus, as far as payment networks go, they’re tied to fiat currency, which is useful as a medium of exchange but, over the long run, doesn’t hold its value.

Bitcoin is a publicly distrib­uted ledger with a series of private and public keys. As long as they have access to the internet (including satel­lite internet if need be), users can send bitcoins (including fractional bitcoins) to others by using their private keys. They can hold their private keys offline and receive Bitcoin when offline; they just need an internet connec­tion to send them, confirm their balance, and those sorts of actions.

This ranks Bitcoins among the most portable and hard-to-confis­cate assets in the world. They can be sent inter­na­tion­ally between different parties, can be memorized and brought anywhere in the world, and can be stored offline with no counter­party. Consid­ering that they also have a supply cap, it’s not surprising that folks in many emerging markets have turned to them rather than relying entirely on a local unsound currency that may be inflating at 10%, 20%, 50%, or 100%+ per year in some cases.

Even in devel­oped markets, which have lower rates of infla­tion, the popula­tion has to deal with negative infla­tion-adjusted interest rates. For example, the following chart shows the interest rate of US Treasury bills minus the prevailing official infla­tion rate. When it’s below zero, it means that T‑bills (and usually bank accounts as well) are not paying enough interest to keep up with the official rate of inflation:

While some countries can partially ban Bitcoin by disal­lowing banks to send money to crypto exchanges or by banning large-scale Bitcoin mining, it’s nearly impos­sible to stop peer-to-peer trading, especially in countries with reason­able property rights and freedom of expres­sion. Banning individual inter­ac­tion with Bitcoin is akin to banning infor­ma­tion or speech since it’s just an open-source public ledger. People can memorize numbers to access it and use satel­lite connec­tions to bypass local internet service providers if neces­sary. It would require a very author­i­tarian approach with draconian enforce­ment to truly stamp it out of use. 

On the other hand, some countries have embraced it, such as El Salvador, which made it legal tender. This is in signif­i­cant part because El Salvador is highly reliant on remit­tance payments. It’s nearly free to send remit­tance payments via the Light­ning network that runs on top of the Bitcoin network since it is open source and keeps getting more devel­oped over time. Several other countries that have previ­ously banned or were about to ban it have stepped back their bans. 

My article on bearer assets included a large section on the practical use cases of Bitcoins, both in devel­oped and devel­oping countries. Aside from being an asset with a fixed supply cap, here are some example use cases of the censor­ship-resis­tant payment and porta­bility aspect: 

1. Reuters reported in February 2021 how Russian opposi­tion leader and anti-corrup­tion lawyer uses Bitcoin as a censor­ship-resis­tant payment, as Putin’s author­i­ties block all of the tradi­tional permis­sioned payment rails: 

Russian author­i­ties period­i­cally block the bank accounts of Navalny’s Anti-Corrup­tion Founda­tion, a separate organ­i­sa­tion he founded that conducts inves­ti­ga­tions into official corruption. 

“They are always trying to close down our bank accounts — but we always find some kind of workaround, ” said Volkov. 

“We use Bitcoin because it’s a good legal means of payment. The fact that we have Bitcoin payments as an alter­na­tive helps to defend us from the Russian author­i­ties. They see if they close down other more tradi­tional channels, we will still have Bitcoin. It’s like insurance.” 

2. The Guardian reported in July 2021 how Nigerian merchants and protest groups used Bitcoin’s censor­ship-resis­tant payment attrib­utes to go around FX currency blocks to carry out their business and receive funds even when their bank accounts were suspended.

“The clamp­down was finan­cial too. Civil society organ­i­sa­tions, protest groups, and individ­uals in favour of the demon­stra­tions who were raising funds to free protesters or supply demon­stra­tors with first aid and food had their bank accounts suddenly suspended. 

Feminist Coali­tion, a collec­tive of 13 young women founded during the demon­stra­tions, came to national atten­tion as they raised funds for protest groups and supported demon­stra­tion efforts. When the women’s accounts were also suspended, the group began taking bitcoin donations, eventu­ally raising $150,000 for its fighting fund through cryptocurrency.” 

3. Alex Gladstein, with the Human Rights Founda­tion, discusses the human rights angle of the Bitcoin network in April 2021: 

“So for Venezue­lans to go through this has been nothing short of totally heart­breaking. However, there are a lot of folks who got involved early, like earlier, a lot of young people, a lot of young folks figured it out in 2015, ’16, they were mining at the time. One of these guys I inter­viewed, he actually helped start Ledn, which is one of the larger industry services now based out of Canada. 

But he and his brothers were mining bitcoin for a couple of years there and they ended up having to escape. The govern­ment came with like an armed squad and seized all their mining equip­ment. Thank­fully no one got hurt. The govern­ment was very perplexed. They saw the mining equip­ment and they thought that they meant they got the bitcoin. But that’s not how it works. So they were able to use the bitcoin to start a new life in Canada. I thought that was really amazing. 

I inter­viewed another guy who escaped to Argentina. He got involved in some sort of dispute with the govern­ment where they claimed he was a criminal, even though he wasn’t, but now he’s able to send money back to his mother who is in Venezuela in bitcoin. She uses it to support herself. There’s just so many stories like this. 

And I think for me, one of the most powerful things as someone whose family went through the Holocaust, was this idea of like, you could flee your country and back then you only brought what was on your back, like the clothing on your back. But today you can bring your wealth with you, which is truly remarkable. 

And I’ve given advice to people who are, for example, leaving Iran these days, help them out on this. People are selling their homes and they’re converting to bitcoin and they’re getting on a plane, getting the heck out, and they’re bringing it with them in this digital format. People are sending money in and out of Syria to people who are stuck there. 

People have escaped countries like Sudan. I inter­viewed a guy from Sudan, which has a horrible infla­tion problem that their infla­tion is in the hundreds. They have an infla­tion rate of something like 150 or 200%. And he’s living in Europe and he’s sending the hardest money around back to his family in Khartoum. And they’re able to get by through that. We’re early here. Again, I think the estimate based on Coinbase numbers is that maybe 10% of Ameri­cans have inter­acted with Bitcoin or cryptocurrency. 

The global number is lower than that, especially in these emerging markets. If the global is two percent and America’s 10%, it’s probably way less than that in a lot of these emerging markets, but hey, in some of them, look at Turkey, man, look at Argentina, look at Nigeria. These are huge countries, 200 million, 100 million, 45 million people. They’ve got the highest per capita usage in those countries. So it’s a changing world.” 

Perhaps more broadly and concisely, one of Alex’s best quotes about the topic is this: 

“We’re at the outset of great digital finan­cial trans­for­ma­tion, where the money we use on a daily basis is evolving from a bearer asset — one that doesn’t reveal anything about us — into a mecha­nism of surveil­lance and control. 

This is more urgent for some people in this world, and maybe less urgent for others, depending on the polit­ical regime they live under. When I’m looking at this new form of money that’s not controlled by govern­ments or corpo­ra­tions, I’m thinking about the big picture of today’s world, where we have 4.2 billion people living under author­i­tar­i­anism and 1.2 billion people living under double- or triple-digit infla­tion. When we talk about the fact that money is broken, this isn’t theoret­ical, and it isn’t just about one country. 

It’s much bigger than that. This is a world where hundreds of millions of people deal with 15 percent, 20 percent, 25 percent infla­tion, where their time and energy, and the currency that they earn their wages in, is liter­ally disappearing. 

At the same time, you have billions of people whose bank accounts can poten­tially be frozen based on their opinions or ideas.” 

4. The Rise Third Layer Applications

The recent creation of several new third layer appli­ca­tions, such as Sphinx Chat and Imper­vious Technolo­gies, that use Bitcoin’s Light­ning network as a way to transmit data, in addition to just trans­mit­ting value. Sphinx Chat is a peer-to-peer messaging platform that runs on Light­ning and allows users to attach micro­pay­ments with their messages, while Imper­vious Technolo­gies is releasing a VPN powered by Lightning. 

“As an example of what can be devel­oped utilizing the Imper­vious API, Imper­vious is releasing a dynamic VPN that lever­ages the Light­ning Network to provide on-demand, high-bandwidth VPN services. The Imper­vious API can operate from behind hostile networks and denied access areas using the Light­ning Network as a creden­tial exchange and trans­port layer to estab­lish peer-to-peer, secured channels for censor­ship resis­tant data transmissions. 

Appli­ca­tions being built on the Impervious.ai API include peer-to-peer streaming video, podcasts, live events, news reporting, distrib­uted storage, and VOIP-based commu­ni­ca­tions.”

• May 2021 Imper­vious Technolo­gies seed financing press release 

To summa­rize, as digital property with a set of censor­ship-resis­tant payment rails, Bitcoin is the appli­ca­tion of software towards finance. And not just towards the surface layer like fintech, but towards the root layer of bearer assets and settle­ment networks, which is part of why it’s so controversial. 

In addition, BTC/LN is emerging as a decen­tral­ized protocol for various layer-three appli­ca­tions focused on trans­mit­ting data in a secure and censor­ship-resis­tant way as well, similar to how it stores and trans­mits value. 

The firm Chainal­ysis, which is used by law enforce­ment to track public blockchains, has found across multiple years that only 0.50% to 2% of crypto trans­ac­tions are for illegal activity, such as scams, ransomware, or drug purchases (which happens to be lower than most estimates of the percentage of fiat currency trans­ac­tions that are for illegal purposes). The vast majority of the usage is estimated to be for invest­ment purposes or legit­i­mate payments. 

As described above, there are many author­i­tarian regimes where protesting and certain types of speech are consid­ered illegal and where basic economic inter­ac­tions can be blocked, and so we could say that bitcoin facil­i­tates those types of “moral yet illegal” behav­iors.

Oppressed people having access to open source software to aid in basic human freedoms and economic inter­ac­tions, making it more complex for author­i­tar­ians to deal with them, is not what I would consider uneth­ical, and indeed quite the contrary. 

Why Bitcoin Uses Electricity

The Bitcoin network is programmed to create a new block on average every ten minutes and add that block to the blockchain, which consists of hundreds of thousands of blocks since incep­tion in 2009.

A new block is produced by a Bitcoin miner (a special­ized computer) solving a crypto­graphic puzzle that the previous block created, and the miner can package thousands of Bitcoin trans­ac­tions currently in the queue into that block. That’s how trans­ac­tions get settled. The network is programmed to target average block times of ten minutes, meaning, on average, every ten minutes, a block of thousands of trans­ac­tions is added to the blockchain.

If miners drop off the network and new blocks, on average, start taking longer than ten minutes to produce, the network is automat­i­cally programmed to make the puzzle easier by a quanti­fied amount so that blocks go back to an every-ten-minute average schedule.

Likewise, if a lot of miners join the network and blocks get added to the blockchain faster than every ten minutes on average, the network will make the puzzle harder. This is known as the “diffi­culty adjust­ment” and is one of the key program­ming challenges that Satoshi Nakamoto solved to make the network work properly.

So, at any given time, millions of Bitcoin mining machines worldwide are looking to solve the puzzle and create the next block. There’s a natural feedback mecha­nism to ensure that blocks are created on average every ten minutes, regard­less of how many miners are on the network.

We recently saw in the first half of 2021 that China banned crypto mining, and approx­i­mately half the network went offline and started moving elsewhere. Bitcoin’s payment network briefly slowed down a bit but other­wise kept working with 100% uptime. The diffi­culty adjust­ment then kicked in and brought the network back up to its target speed.

Imagine if Amazon was told with one week’s notice that it had to move half of its server capacity inter­na­tion­ally; it would likely experi­ence uptime issues for its services for the rest of the year as it moved and rebuilt half of the system.

If a miner creates an invalid block, meaning one that doesn’t conform to the consensus rules of the node network, the network discards it. If two miners produce a valid block at around the same time, the winner will be decided by which one gets found by the rest of the network first and has another block produced and added onto it, becoming the longer (and thus official) blockchain.

This process is known as “Proof-of-Work.”

Millions of machines are using electricity to apply processing power to solve crypto­graphic puzzles left by the most recent block. This may seem like a waste of energy, but it keeps the system decen­tral­ized. Work is the arbiter of truth in this case. There is no central authority that decides what consti­tutes a valid block or a valid set of trans­ac­tions; the longest blockchain is verifi­able at any given time and is recog­nized as truth by the rest of the network based on code. The longest blockchain is the one with the most work put into it, and that also meets the consensus criteria that the node network checks.

The more energy Bitcoin’s network uses, the more secure it is against most attacks. Many of the tiny non-Bitcoin blockchains have been victims of 51% attacks, where a single entity temporarily or perma­nently gains control of over 51% of the processing power on the network and uses that majority of processing power to re-organize blocks and performs double-spend trans­ac­tions (which is essen­tially theft).

This chart, for example, shows Bitcoin’s network processing power compared to the processing power of some of its hard fork copycats:

Both of those other blockchains only have 1 — 2% or less of the Bitcoin network’s total processing power and have been hit by malicious block re-orgs. In fact, if just 2% of Bitcoin miners decide to do a 51% attack on either of those two hard forks, they can. The same is not true for the other direc­tion since the Bitcoin network has a far larger network of miners and energy usage than them, by two orders of magnitude. 

That shows the impor­tance of network effects in the blockchain industry and why Bitcoin’s energy usage has kept it uniquely secure. 

When someone asks, “Can’t you just copy Bitcoin?” 

“That’s why the answer is ‘no.’ 

You can repli­cate the open source code, but you can’t repli­cate the fact that millions of ASIC miners are securing the Bitcoin network and not your copycat network, and you can’t repli­cate the fact that thousands of devel­opers are working on making the Bitcoin network better every day rather than working on your copycat network. And Light­ning’s number of open channels and liquidity can’t be easily repli­cated either; it took years to build. 

Trying to copy Bitcoin would be like if I copied the content from Wikipedia and hosted it on my website. Techni­cally it could be done, but it wouldn’t do much. It wouldn’t gain the real Wikipedia’s traffic because it wouldn’t have the hundreds of millions of links pointing to it from other websites. And it wouldn’t be updated like the real Wikipedia because there’s no way I could convince the majority of those volun­teer editors to come to work on my version instead.

Unless I could somehow succeed in the Herculean task of convincing the majority of the network to move over to my version, it would always just be a shadow of the real one with a tiny fraction of the value. 

The same is true if I made a poor mimic of Twitter or something. I could make it look like Twitter, but it wouldn’t really be Twitter, full of users and developers. 

Bitcoin’s Efficient Scaling Pattern

When Bitcoin was created, it was designed so that every ten minutes, when a miner produces a new block of trans­ac­tions, the miner that produced it earns 50 Bitcoins. After four years, it was pre-programmed to drop to 25 new Bitcoins per block. Four years later, it was 12.5 Bitcoins per block. Four years after that, it’s down to 6.25 Bitcoins per block in the current era.

This pattern will continue every four years until new bitcoin gener­a­tion asymp­tot­i­cally approaches zero, and the hard cap of 21 million bitcoins is reached sometime after the year 2100. Miners will earn a vanish­ingly small number of fractional Bitcoins for producing new blocks within a few decades. Out of the 21 million, 18.7 million Bitcoins have already been created.

However, miners also earn trans­ac­tion fees. Senders pay trans­ac­tion fees, denom­i­nated in fractional bitcoin, to ensure their trans­ac­tion gets into the blockchain in a timely manner.

In the early days, blocks were often not full, so trans­ac­tion fees were minimal. However, as Bitcoin became more widespread, blocks reliably became full, and trans­ac­tion fees became a small but more meaningful part of miner fees.

So Bitcoin was highly infla­tionary in the begin­ning, but it has an increas­ingly disin­fla­tionary monetary policy until it approaches outright zero infla­tion, and its security budget scales similarly.

Here is a table of the Bitcoin network’s average market capital­iza­tion, annual security spending (total miner revenue, including block subsi­dies and trans­ac­tion fees), and the percentage of the market capital­iza­tion spent on security each year:

Each year so far, the Bitcoin network usually spent more on security than the previous year but always spent a smaller percentage of its market capital­iza­tion on security than the previous year. This isn’t a decision by any central­ized party; it’s a combi­na­tion of the algorithm, the value of the network, and individual miner decisions whether to mine or not. 

That’s what journal­ists and other people who don’t under­stand the algorithm often miss: the declining block subsidy. This results in Bitcoin’s infla­tion rate going down, along with miner revenue as a percentage of Bitcoin’s total market capitalization. 

Any serious analyst that under­stands Bitcoin’s algorithm would actually be more concerned about the possi­bility of Bitcoin one day not using enough energy for its security sometime in the future when it relies mostly on trans­ac­tion fees rather than using too much energy. 

Since the security spending repre­sents miners' revenue, and miners spend most of their costs on electricity, that security spending repre­sents the high-end for how much energy the Bitcoin network is using in dollar terms. In reality, it is less than that, due to miners usually making a profit. 

Looking at my own Bitcoin miners for example (I have some hosted via Compass Mining), my electricity costs are currently about 20% of my miner revenue. That will vary over time. 

Next, we can look at just the portion of the miner revenue that comes from trans­ac­tion fees, which is a subset of the previous chart: 

We can see that trans­ac­tion fees are a tiny portion of Bitcoin’s market capital­iza­tion each year. The highest year in percent terms was 2017, during the bubble peak. Efficiency improve­ments have been made since 2017, so even in the heart of the early 2021 bull run, the network didn’t reach those levels again.

The Bitcoin network is now down to less than 2% of its market cap being spent on miner revenue each year, including a fraction of one percent on fees. In 2024 there will be another block subsidy halving, which will probably bring miner revenue down closer to 1% of market capitalization.

In 2028 there will be yet another block subsidy halving, and another around in 2032. After that point, the block subsidy will be so tiny that a large portion of miner revenue will be made up of trans­ac­tion fees, and miner revenue will likely be less than 1% of market capital­iza­tion, approaching a steady state of maybe 0.25% to 0.50% on average, depending on fee levels.

Since we can’t know for sure what the steady state will be due to variable market-driven trans­ac­tion fees, here’s a table of poten­tial long-term Bitcoin market capital­iza­tions (vertical axis) and annual security spending (horizontal axis) in the future:

If Bitcoin fails to grow for one reason or another and becomes a failed project or perma­nently remains around its current market capital­iza­tion of less than $1 trillion, its miner revenue will signif­i­cantly decrease from current levels as block subsi­dies diminish. By the 2030s, Bitcoin miner revenue will probably be around 0.50% of market capital­iza­tion or less. So the network will be stuck at 2018 — 2020 energy spending levels or less.

If Bitcoin becomes system­at­i­cally impor­tant, let’s say $5-$10 trillion (repre­senting a per-coin price of $250k to $500k) with hundreds of millions of users, then at a 0.50% annual security cost, that would be $25-$50 billion. That would be 2x or 3x as much energy usage as Bitcoin was using at an annual­ized rate in the first half of 2021. This would repre­sent approx­i­mately 0.3% of global energy usage.

If we say it reaches an outra­geously high price of one million dollars per coin, for a criti­cally impor­tant market capital­iza­tion of $20 trillion, with billions of users, then at 0.50% annual security cost, that would be $100 billion, or about 6x as much energy usage as bitcoin was using at an annual­ized rate in the first half of 2021. This would repre­sent maybe 0.6% of global energy usage, which seems appro­priate for a network used by billions of people for multiple purposes, as it would need to be at that point to reach such a high value.

By that point, it would be big enough that it’s likely replacing energy used by parts of the global banking system. There are tens of millions of people working in banks and fintech compa­nies worldwide. The appli­ca­tion of software to money at the root layer, just like other indus­tries, brings efficien­cies and reduces the need for employ­ment and real estate in certain parts of legacy infra­struc­ture, freeing up those human resources and corre­sponding energy usage for other produc­tive purposes.

At the current time, the Bitcoin network is estimated to emit less CO2 than random things we don’t think about, like tumble driers or zinc production:

If Bitcoin becomes wildly successful with trillions of dollars of utility for users, we could poten­tially see it emit an amount of CO2 that is compa­rable to or higher than zinc produc­tion and likely still below that of aluminum produc­tion. In other words, despite reaching a massive scale and serving numerous purposes, it would still be compa­rable to various other random industries. 

Scaling By Layers and the “Cost Per Transaction” Fallacy 

The Bitcoin network can do a maximum of a few hundred thousand base layer trans­ac­tions daily. That’s about five trans­ac­tions per second. This number has increased slightly over time due to occasional upgrades that improve trans­ac­tion density. 

This trans­ac­tion limit is often unfavor­ably compared to a network like Visa, which can process tens of thousands of trans­ac­tions per second. 

Due to that, critics often point out that Bitcoin’s energy usage per trans­ac­tion is very high, and thus the network is ineffi­cient and should be avoided for ESG reasons. There are two problems with that reasoning, however. 

The first problem with that reasoning is that Bitcoin uses energy whether or not trans­ac­tions are occur­ring. The way to think about it is that a large portion of that energy is used simply for securing the network as a store of value. One block might have 1,200 trans­ac­tions. The next block might have 2,500 trans­ac­tions. The block after that might have 1,800 trans­ac­tions. Meanwhile, the same number of miners are hooked up to the network between those subse­quent blocks, verifying the blocks and paying for electricity. Whether blocks are full or not, they use roughly the same amount of energy. 

Whether you make a trans­ac­tion or not does not materi­ally change how much energy the Bitcoin network is using at that time. Bitcoin’s energy usage comes from miners earning the block subsidy and average trans­ac­tion fees and is denom­i­nated in bitcoin and thus based on the value per bitcoin, which mainly comes from people holding bitcoin as a store of value, not spending it. Trans­ac­tion volumes only affect the trans­ac­tion fee portion, and only the longer-run trans­ac­tion fee average matters. 

Think of it like running your dishwasher each night. Whether it’s 50% or 90% full when you run it, it still uses about the same amount of resources per run. The marginal extra dish or utensil doesn’t materi­ally affect the dishwash­er’s energy usage. 

Another analogy would be keeping your computer on all day and either sending 20 emails or 100 emails. The marginal amount of energy “per email” that you send isn’t relevant because regard­less of how many emails you send that partic­ular day, your computer is on and using its baseline resource level. 

The second problem with that reasoning is the idea that this limit of about 5 trans­ac­tion per second is the true limit, which it is not. In reality, the Bitcoin network has multiple layers, just like the current finan­cial system. 

Visa is merely a layer on top of a deeper payment network. In the United States, for example, we have the Fedwire system. That’s the gross settle­ment layer that banks use to perform large trans­ac­tions with each other. This system only does about 10 trans­ac­tions per second on average and has scaled up slowly as needed, but those trans­ac­tion amounts are very large, repre­senting millions of dollars each. On top of that layer are things like Visa, PayPal, people writing physical checks to each other, and so forth. 

If you send me a credit card payment, for example, that seems pretty instan­ta­neous to us, but in reality, it’s not. When the trans­ac­tion seems finished to us, in reality, our two banks just conversed and made an IOU between themselves. Sometime later, they will batch it with many other consumer trans­ac­tions and settle their books with a big Fedwire transaction.

Effec­tively, tens of thousands of consumer trans­ac­tions per second get condensed down to 10 much larger inter­bank Fedwire trans­ac­tions at a later time. There’s no limit to how many surface-layer trans­ac­tions can occur because there is no limit to the size of those massive settle­ment transactions. 

Similarly, the Bitcoin network has additional layers: Light­ning, Liquid, Exchanges, and more. However, unlike the banking system that depends on long settle­ment times and IOUs, many of Bitcoin’s layers are designed to minimize trust via software. 

Method 1: Lightning (Second Layer) 

The Light­ning network is a trust­less smart contract layer that runs on top of the Bitcoin network and is suitable for smaller transactions. 

Like the Bitcoin network itself, nobody “owns” the Light­ning network; it’s a set of open source standards that multiple compa­nies make open source imple­men­ta­tions of. Those Light­ning trans­ac­tions are instan­ta­neous and nearly free and have no upper limit in how many can occur per second as the network continues to grow. It went active in early 2018, and as of early 2021, it finally reached enough channel liquidity, with major apps and exchanges starting to use it, that it became truly functional and reason­ably mature: 

With Light­ning, two people can open a channel with each other using a base layer trans­ac­tion and then send any number of instant trans­ac­tions between each other. Days, weeks, months, or years later (whenever they want to), they can close that channel with a second base layer trans­ac­tion. That means dozens, hundreds, or thousands of mini-trans­ac­tions can be combined into two base layer trans­ac­tions. It’s like keeping a bar tab open and settling at the end of the night, or the end of the month, except it doesn’t rely on trust but instead relies on programmed smart contracts that ensure the bar tab is settled. 

In addition, you can send a Light­ning trans­ac­tion through the network to someone else. If Alice has a channel open with Bob, and Bob has a channel open with Cody, then Alice can send a payment to Cody through Bob as the inter­me­diary, even though she doesn’t have a channel with Cody. And it’s all trust­less; all based on programmed smart contracts using the Bitcoin base layer as the settle­ment assurance. 

If someone has a couple Light­ning channels open with well-connected counter­par­ties, she could send and receive many trans­ac­tions to/from any number of people through the network (including all the vast numbers of people on the network that she doesn’t have any channels open with), and settle them in a small number of base layer trans­ac­tions when opening or closing those channels. 

To give an idea of how efficient Light­ning is, there were some gaming demos at the 2021 Bitcoin Confer­ence in Miami involving “sat streaming, ” and THNDR Games processed 13,571 Light­ning trans­ac­tions during the confer­ence for an average fee of 1.4 sats per trans­ac­tion, or roughly $0.0005 USD per transaction. 

And as previ­ously mentioned, layer three devel­op­ments allow the Light­ning network to be used for secure, decen­tral­ized, and censor­ship-resis­tant data trans­mis­sion as well. 

Method 2: Liquid (Side Chain) 

Liquid is an open-source side chain of Bitcoin, meant for large entities like exchanges to settle bitcoins with each other faster and less expen­sive than base-layer trans­ac­tions. Beyond that, it has many other uses as well. 

With Liquid, bitcoins get locked up into L‑BTC tokens with a “peg-in” trans­ac­tion. Those L‑BTC tokens can run on this side chain with faster settle­ment times until an entity decides to unlock the bitcoins from that chain and bring them back to the base layer with a “peg-out” transaction. 

A feder­a­tion of nodes runs the Liquid network. The Liquid network can’t fully repli­cate the rock-solid security and decen­tral­iza­tion level of the bitcoin base layer, but it has a much faster trans­ac­tion throughput and has a reason­able degree of security and decentralization. 

In other words, an entity can make a known trade-off to lock up some Bitcoin and get L‑BTC tokens in return. Those tokens use Bitcoin as their founda­tion of value but provide additional speed and features. Between each peg-in and peg-out trans­ac­tion, L‑BTC will typically be used for many transactions. 

Similarly, any smart contract platform can lock up Bitcoin in a similar way that Liquid does. The largest one, Wrapped Bitcoin or “WBTC, ” consists of Bitcoins that are wrapped in an Ethereum token and can, there­fore, trade-in Ethereum-based DeFi ecosys­tems, with any advan­tages or security downsides that may come with that. There are also DeFi projects that run on Bitcoin, using wrapped bitcoins in their ecosys­tems as well. 

Method 3: Exchanges (Custodial Relationships) 

Custo­dial crypto exchanges and Bitcoin platforms are also scaling methods. When millions of people trade on Coinbase, for example, those are not millions of base layer trans­ac­tions. Those are trans­ac­tions within Coinbase’s central­ized database. Only when people deposit or withdraw Bitcoins to/from Coinbase, or Coinbase sends Bitcoins to or from cold storage or to another exchange in a big batch, would there be an on-chain transaction? 

Anyone looking to efficiently accumu­late Bitcoin rather than “trade cryptos” should, of course, use Swan Bitcoin instead since the fees are lower, and the customer service is better. It’s optimized for automatic recur­ring purchases, one-time buys, and self-custody if desired. 

As another example, millions of people use Cash App to buy and hold Bitcoins. They can send Bitcoins to another Cash App account for free. Those are not base layer trans­ac­tions; those are merely trans­ac­tions within Cash App’s database. It only becomes a base layer trans­ac­tion if they deposit or withdraw their bitcoins to/from Cash App with an external source or if Cash App moves its users’ Bitcoin around inter­nally, to or from cold storage in large batches.

(As a note, Cash App has tight limits on withdrawals). 

Unlike Light­ning and Liquid, those types of custo­dial accounts like Coinbase and Cash App require that you trust the central­ized platform, similar to how you trust your bank.

Many exchanges and custo­dial services already use Liquid between themselves for rapid settle­ment. Some of them are also starting to use Light­ning for their retail clients, so even when people withdraw, it might not need to be a base layer trans­ac­tion anymore. 

If Bitcoin continues to be successful and grows in users and market capital­iza­tion, base layer trans­ac­tions will become increas­ingly used for settle­ment trans­ac­tions, not day-to-day trans­ac­tions. The fee market is what modulates its usage. 

If trans­ac­tion fees are averaging $1 for the base layer because block space hasn’t been heavily used, for example, then $100 amounts are fine to do base layer trans­ac­tions with. However, if trans­ac­tion fees are averaging $50 for the base layer, those small trans­ac­tions would be better with Light­ning. There are many trans­ac­tions worth hundreds of thousands or millions of dollars on the base layer, and they don’t mind paying a $50 fee. 

Ultimately, Bitcoin is compa­rable to Fedwire as a massive settle­ment network that can do about 5 trans­ac­tions per second, with no limit to trans­ac­tion size and thus no limit to how much value it can settle per day. On top of that, multiple layers ranging from trust­less Light­ning to semi-trusted Liquid to fully-trusted exchanges can account for hundreds of thousands of trans­ac­tions per second, with no upper limit. As a result, the cost and energy per trans­ac­tion can become negli­gible over time. 

The Tech Stock Comparison 

When an investor is picking through early-stage unprof­itable tech compa­nies to invest in, they need to model out the future to ensure that, if successful, its revenue will scale faster than expenses. 

Consider a new social network, for example. 

It will typically be unprof­itable or minimally profitable in the begin­ning because it requires high base expenses to pay a team to run the service. Even if it has zero users in the begin­ning, it still has those base expenses to get it started and running. 

However, if it’s successful and well-managed, it becomes cheaper to add each user. Expenses still grow over time (more employees, more facil­i­ties, more servers, etc.). Still, if the company is well-designed, those expenses should grow more slowly than users and revenue, which leads the company towards profitability and robust profit margins. 

In other words: 

Bitcoin is not a company, but it is an efficient network. By design, its expenses scale more slowly than its utility due to its declining block subsidy that eventu­ally results in a security model based only on trans­ac­tion fees.

We can’t know for sure how much energy Bitcoin will ultimately use since we don’t know how many people will use it or what the fee market will look like a decade or more in the future. If it’s successful, it will consume more energy than if it’s not successful, but due to its declining block subsidy, its utility will greatly outpace its energy usage in either scenario.

That’s the key takeaway.

How Bitcoin Uses Otherwise-Wasted Energy 

Beyond a simple calcu­la­tion of how much energy Bitcoin uses, we should also consider the details of how it uses energy and the types of energy it uses. 

People often imagine bitcoin miners competing with other indus­tries for electricity, as though bitcoin mining must push out some other electricity use. However, because bitcoin miners inher­ently require extremely cheap electricity sources, they can’t usually compete with normal electricity users. As a result, bitcoin miners seek out ineffi­cien­cies around the world where electricity is being under­uti­lized and wasted. 

The vast majority of energy consumers can’t go to where the energy is; the energy has to be brought to them. Humans organize themselves based on geography, mainly around shipping channels. We live in coastal or river­side cities, in the suburbs of those areas, and around rural areas of fertile land, not around energy. 

We don’t move to where the oil and gas and uranium deposits are; we send folks out to go get the oil and gas and uranium and bring it back to us for consump­tion in our homes and at gas stations and nearby nuclear stations. 

Bitcoin miners are unusual energy consumers in that they can go wherever the energy source is, as long as they can get some sort of basic internet connec­tion, including a cellular or satel­lite connec­tion if needed. That means they use energy in quite efficient and unusual ways. 

Fideli­ty’s first digital asset analyst and later founding partner of Castle Island Ventures, Nic Carter, described Bitcoin’s energy usage in an insightful way back in 2018: 

“An inter­esting exter­nality of PoW coins — they are always-willing energy buyers at 3 — 5 cents/kWH. And some of the best energy assets are off the grid. This global energy net liber­ates stranded assets and make new ones viable. 

Imagine a 3D topographic map of the world with cheap energy hotspots being lower and expen­sive energy being higher. I imagine Bitcoin mining being akin to a glass of water poured over the surface, settling in the nooks and crannies, and smoothing it out.” 

 — Nic Carter 

Although some miners use cheap tradi­tional energy, here is a sampling of some novel ways that Bitcoin miners use other­wise stranded or unwanted energy to benefit themselves and their counterparties. 

1. Stranded Hydroelectric Power in China 

For a long time, China has been the largest Bitcoin mining juris­dic­tion. At one point, Chinese miners were estimated to account for over 70% of the network. However, by the spring of 2021, it was estimated to have gradu­ally dipped to under 50% as more compe­ti­tion arose elsewhere. Then, a 2021 ban on Chinese bitcoin mining, likely to enforce their capital controls, has sharply reduced Chinese bitcoin mining exposure, and those miners have gone elsewhere. 

For many years, China was an inter­esting example of Bitcoin mobility. The province of Sichuan has a ton of overbuilt hydro­elec­tric capacity. During the wet season, they produce more clean electricity than they can possibly use. So it would other­wise be curtailed, wasted. 

Since Bitcoin miners can go to where the energy source is, they used to flock to Sichuan during the wet season to use that other­wise wasted energy. Not because they are altru­istic environ­men­tal­ists, but simply because it is cheap and nobody else is making use of it. The electricity that would other­wise be wasted and generate no revenue for the operator can be sold for extremely cheap levels to someone who can find a use for it. 

Here is an estima­tion by Cambridge of how much each Chinese province contributed to China’s overall Bitcoin hash rate throughout the seasons. Sichuan is in yellow at the top: 

With the Chinese Bitcoin mining ban that came shortly after the end date of this chart, that hash rate and billions of dollars worth of annual revenue is moving to North America and other countries. But this was a great example of Bitcoin miners mopping up stranded and wasted energy for many years.

2. Flared Gas Bitcoin Mining 

Many types of petro­leum deposits come with associ­ated natural gas. 

If there is a suffi­cient quantity of this gas, it can be collected and sent via pipeline or other trans­port networks to be used as a primary energy source since, of course, natural gas is extremely useful for electricity and heating. 

However, if it’s a small amount, then it’s not economical enough to build a pipeline or other­wise collect that gas. 

So what happens? 

It gets vented or flared into the atmos­phere, and there­fore wasted. Venting means letting it out into the atmos­phere, mainly as methane (a stronger green­house gas than carbon dioxide). Flaring means it is burned and thus converted into carbon dioxide and emitted into the atmos­phere. A complete waste, either way and yet still contributing to global green­house gases. 

In terms of scale, the US Energy Infor­ma­tion Admin­is­tra­tion estimated in its 2020 natural gas annual report that 1.48 billion cubic feet of natural gas were vented or flared per day on average in the United States throughout 2019. That’s about 150 TWh of energy, which is higher than the estimated total peak level of Bitcoin’s annual­ized energy usage in 2021, according to the Univer­sity of Cambridge. 

In other words, virtu­ally, the entire Bitcoin network in its peak 2021 form could hypothet­i­cally be run off of stranded natural gas in the US, let alone the rest of the world. Considering energy lost in the conver­sion process to electricity, that’s probably not completely the case, but the point is, this stranded energy source can power a huge chunk of it. 

Several private Bitcoin mining compa­nies specialize in hooking up trailers of bitcoin miners to oil producers with stranded gas to use that other­wise-wasted energy. 

It’s a win/win scenario for producers and the Bitcoin miners. The producers get to sell their gas rather than waste it while earning higher ESG scores and meeting state flaring limits. Bitcoin miners get a super cheap source of energy.

Alter­na­tively, some of these Bitcoin miners are also willing to sell mining systems to oil and gas producers and set the hardware up for them so that the producer can directly capture any poten­tial upside in bitcoin’s price. 

North Dakota, in partic­ular, is a major area for stranded gas mining. According to the EIA, nearly 20% of produced natural gas is flared in North Dakota rather than collected. That wasted gas alone is equiv­a­lent to tens of TWh of electricity gener­a­tion per year. Back in 2014, North Dakota’s wasted amount was more like 35% and the state imple­mented rules to try to get that number down. Bitcoin miners can soak up much of this other­wise wasted energy, and Bitcoin mining is indeed growing in that state. 

I reached out to Marty Bent, the Director of Business Devel­op­ment for Great American Mining, to under­stand their opera­tions. Great American Mining is a private company that works with oil and gas producers to deploy mobile Bitcoin mining rigs onto the oil and gas produc­tion site to use their wasted gas to mine Bitcoin. 

Marty gave a useful clari­fi­ca­tion on why pretty much only bitcoin miners can make use of this stranded gas, rather than similar indus­tries like data center server farms: 

“Since the Bitcoin network is a distrib­uted peer-to-peer network that doesn’t depend on any one miner to facil­i­tate trans­ac­tions, bitcoin miners are better positioned to take advan­tage of the flare gas oppor­tu­nity compared to other energy inten­sive compute processes like server farms because they can stomach disrup­tions in the field without affecting the uptime of the network materi­ally. Whereas a server farm would not be able to because uptime disrup­tion could seriously affect critical business opera­tions. Beyond this, miners send the amount of data to mining pools is very small and doesn’t require much bandwidth, so they can operate in very remote areas using cellular data much more trivially than other energy inten­sive data processes.” 

 — Marty Bent, July 2021, via email 

I asked him which types of producers are better suited than others. His answers are the ones in chillier climates (current gener­a­tion Bitcoin miners need good air cooling, which uses energy) and/or in juris­dic­tions that have more strin­gent flaring require­ments. And due to various limita­tions, on-shore locations tend to be better suited than offshore locations: 

“The oppor­tu­nity is partic­u­larly large in juris­dic­tions with strict flaring regula­tions because producers are highly incen­tivized to reduce their flaring. If they flare too much they are forced to shut in their wells for a period of time. 

[…] 

Yes, shale producers certainly have an advan­tage over offshore drillers as the electrical require­ments and permit­ting neces­sary to operate offshore are much more restric­tive than onshore drilling. On top of this, the amount of surface area to drop containers and gener­a­tors on offshore opera­tions is extremely small compared to onshore well pads, so scaling could be an issue. Beyond this, producers in states with relatively cool climates have an advan­tage, at least in the short to medium term, until immer­sion setups mature. Also, producers who own the natural gas minerals and the produc­tion are better positioned because they wouldn’t have to deal with the headache of paying out royal­ties to mineral rights owners.” 

 — Marty Bent, July 2021, via email 

In my view, Bitcoin miners using stranded gas (and stranded hydropower) are about the cleanest miners out there. It’s better than solar panels, wind energy, or similar sources because the energy they are using is liter­ally wasted and sent into the atmos­phere other­wise, by compa­nies that are extracting petro­leum for other purposes. 

3. Bitcoin Mining as a Grid Battery 

Electrical grids have to compen­sate for two things: changing supply levels and changing demand levels. 

Some electrical sources are very consis­tent, like baseload nuclear power, which can run 24⁄7. Other sources, like wind and solar or hydro, are more variable based on what Mother Nature feels like providing in terms of wind, sun, and rain during a given timeframe. Due to this partial variability, electrical supply needs to be overbuilt, so that even on a partic­u­larly “low” day of supply gener­a­tion, it’s still suffi­cient to provide power to the community. 

Moving to the demand side, certain days require more electricity than others. Looking at my gas and electric bill, for example, I use a lot more gas in winter than in summer, because in the summer it’s only used for cooking while in the winter it’s used for cooking and heating.

Meanwhile, I use way more electricity in the summer since I’m using it for air condi­tioning in that season and for lights and electronics consis­tently throughout the year. Plus, there are peak days, such as the most danger­ously hot day of a given year, where just about every single house­hold has the air condi­tioning system on full blast. Days like that need to be accounted for. 

So, due to variability on both the supply side and the demand side, electrical grids need to be overbuilt and have a lot more power gener­a­tion capacity than is used on an average day. Some of that capacity could be variable, like natural gas peaking plants that can be rapidly turned on or off as needed. Other types might be ones they can’t control, like solar panels and wind turbines. If you overbuild solar capacity and wind capacity and aren’t using the excess or selling it to another grid, you just waste it. 

One of the problems with solar and wind power is that the cost of storage is very high. Despite all of our human ingenuity, we still can’t make very cost-effec­tive batteries at a utility-scale. It’s an extra­or­di­nary hard physics problem. We can make storage batteries for certain ideal condi­tions, but it’s not cost effec­tive to use them very broadly. 

This chart shows the energy return on invest­ment from various energy sources. In other words, it measures the multiple of how much energy you get out of what you put in. The lower yellow bars include the energy cost of storage for variable energy sources. Nuclear is on the far right with a 75x energy multiple (and no buffer/storage costs), while wind and solar are low, especially when the storage costs are considered: 

Bitcoin mining makes it profitable to overbuild renew­able sources of energy produc­tion, since it allows that surplus supply to be monetized. Every commu­nity that wants reliable power needs overbuilt electric capacity anyway, and for wind and solar and hdyro that’s even more impor­tant because they are variable. However, overbuilding is usually not very cost effec­tive, unless you can use it for something profitable and useful when it’s not other­wise needed. 

Bitcoin miners are a unique solution to that problem, can make overbuilding profitable, and thus play the indirect role of an energy storage solution. 

During the vast majority of the time when there is more supply than demand, bitcoin miners as one of the electricity consumers in the commu­nity can power their machines, earn revenue, and pay their electricity costs. If there is a surge in electricity demand or a reduc­tion in supply that would other­wise cause brown-outs in the region, those bitcoin miners can temporarily shut off. 

A well-structed commer­cial rates contract can make this work smoothly. The utility could offer the miner the lowest possible rate in the area, in exchange for them having a higher toler­ance for variability and other points of contract flexibility. 

Harry Sudock, VP of Strategy at a bitcoin mining company called Griid, explained this to Peter McCor­mack on his podcast in June 2021: 

“Curtailing is not the position you ever want to be in as the energy gener­ator. So, let’s use a wind turbine as a really easy example. The turbine goes around once, gener­ates electrons. The market price in some regions is negative, so what they’ll choose to do is just not to send the energy anywhere. It dissipates. 

So, if they’re able to strike a deal with another bidder on that energy who can tolerate some inter­mit­tent consump­tion, can use it some of the time, not the other part of the time, that’s a really valuable customer to be able to bring to a market that isn’t neces­sarily able to support the energy gener­a­tion on a broader basis. 

So, I think bitcoin miners are special and are a huge techno­log­ical upgrade from the tradi­tional consumers of electricity. We have two, I think of as “energy super­powers”: the first one is that energy is 80% or 90% of our monthly costs; the second is that we can consume on an inter­mit­tent basis without harming our business model partic­u­larly. So, if someone tells me I need you to shut off your miners 100 hours a year, or 500 hours a year, we don’t say no, we just say, “We need to reflect that in the energy price we pay.”

So, when I’m looking to negotiate a power contract, the way that I frame this is, “I need you to get me the lowest possible cost that you know how to offer. I’m willing to negotiate on every other part of the profile of the load. How big are we going to build the mine; how often do you need that power back; do you need us to serve any other creative purpose within your energy mix or system; do you need us to split our facility into two and to go locate at two different points within place? Great. Do we need to be able to contribute to the security budget of these other pieces of the operation?” 

Our job is to drive that energy price as low and compet­i­tive as possible and work with producers on every other variable.” 

For clarity, I would add that their third super­power is their ability to co-locate with the source of electricity gener­a­tion and thus cut down on trans­mis­sion losses to help keep their electricity cheap. Bitcoin miners are unique in that:

1. Almost their entire operating expense is electricity.

2. They can tolerate inter­mit­tent consump­tion.

3. They are flexible with their location.

As a result, they can sacri­fice variables that most other compa­nies cannot in exchange for rock-bottom electricity prices when electricity is abundant. 

This is why Jack Dorsey, CEO of Twitter and Square, has made the contro­ver­sial state­ment that the Bitcoin network incen­tivizes renew­able power. Right now, Bitcoin is too niche for grid engineers to incor­po­rate it into their plans. Still, if Bitcoin miners become more regular and visible, they can be incor­po­rated into grid designs and rate markets more thoroughly. 

In that podcast, for example, Sudock described this situation: 

“This is an anecdote that we’re in the midst of working through right now. A commu­nity is zoned to have a new hospital built in their area. There are 17,000 energy customers, house to house, in that utili­ty’s juris­dic­tion. They are going to bring a hospital that will double the amount of energy that this region pulls down. We can all agree that a hospital is a very worthy use of electricity; there is no argument there. 

They rebuild the trans­mis­sion lines, they build a new substa­tion that’s bigger, that can handle the additional load, and after they do all that, the hospital project falls apart. 

So they’re left having invested millions of dollars in this area to attract this large customer. They now have to pass that cost back to those 17,000 house­holds unless they can find another use for that energy. So what did they do? 

They called our VP of Energy Manage­ment and said, “We’ve got an overbuilt supply here. If we don’t bring in a large-scale energy customer, these costs are going to get passed to these house­holds that don’t have the budget to support rising energy prices.” 

So, we have this beautiful oppor­tu­nity to come in, backstop this utility, provide a customer to come in on the back of this deal falling apart, and provide the backbone to this commu­nity and to stabi­lize their energy prices for a decade to come. And so these are the stories of Bitcoin mining that don’t get to rise to the surface. It also happens that this energy source is over 60% carbon-free.” 

Due to their ability to go to the source of power, ZBitcoin miners can also fill in unexpected holes in demand or other special situa­tions.

4. Advancing New Energy Technologies 

The flexi­bility of Bitcoin mining allows it to fill in gaps for new clean energy technolo­gies, to help them scale and provide a proof of concept. Here’s a real-world example, and it also fits into the previous topic of acting like a battery to profitably soak up excess capacity. 

Oklo Inc is a startup that plans to make innov­a­tive new nuclear power facil­i­ties. Most existing nuclear facil­i­ties are massive multi-billion-dollar facil­i­ties. Still, Oklo plans to make micro-reactors with much smaller facil­i­ties, lower costs, lower energy output, and quicker construc­tion times. 

In addition, Oklo’s reactors are a type that can use the waste of conven­tional nuclear facil­i­ties as their fuel source. It actually reduces the radioac­tivity of existing nuclear waste. 

As ZME Science reports: 

“The startup Oklo plans to give us a reliable and cost-effec­tive source of power while also solving the issue of radioac­tive waste, which needs to be stored and managed in partic­ular condi­tions for hundreds of thousands of years. Their solution is to reuse the waste in autonomous reactors that don’t try to slow down the nuclear decay of the material. Effec­tively, such a reactor would be able to extract more power from fuel that has already been spent, giving us a use for the processes that happen naturally in a radioac­tive fuel dump, instead of letting them waste away as radioac­tive pollution. 

“What we’ve done is take waste that you have to think about managing for 100,000 or a million years … and now changed it into a form where you think about it for a few hundred, maybe thousands of years, ” Oklo’s co-founder Jacob DeWitte told CNBC.” 

However, with any new technology, there’s always a catch. You have to ask yourself, “Why hasn’t this already caught on?” especially consid­ering that Oklo is using a proven type of fission technology that has already existed for decades. 

Besides the process of going through regula­tion in an environ­ment that has not been very favor­able towards nuclear energy, the catch seems to be that Oklo’s small facil­i­ties naturally mean that profit margins are very tight. The company has proposed to operate them via automa­tion, unsuper­vised by any on-site humans. This has led regula­tors to raise their eyebrows; unattended facil­i­ties with nuclear materials are somewhat of a security hazard, to say the least. 

Enter Bitcoin miners because, appar­ently, Bitcoin fixes this. 

Oklo recently announced a 20-year partner­ship with Compass Mining. Oklo will provide Compass with at least 150 MW of power starting in the early 2020s. 

“Cryptocur­rency mining offers promising pathways to accel­erate the deploy­ment of clean energy technolo­gies, and Oklo is positioned to respond to commer­cial demands by offering end-users the conve­nience of buying clean, reliable, and cost-effec­tive power that they can depend on, ” added DeWitte. Oklo’s path to deploy­ment strives to optimize its power plant designs to be cost-compet­i­tive with the cheapest forms of energy. 

Oklo’s advanced fission power­houses can produce reliable power for up to 20 years without the need to refuel and have the capabil­i­ties to turn nuclear waste into clean energy. This commer­cial project is scalable, and Oklo can add additional capacity to accel­erate Compass’ sustain­able mining efforts further while driving the economics of Bitcoin mining activ­i­ties powered by advanced fission.” 

I reached out to Compass’ CEO, Whit Gibbs, to get some details. 

He explained to me that, for example, there might be a town that Oklo can do business with, to provide them clean power at a low cost by deploying one of their small facil­i­ties on the outskirts of the town. A given Oklo reactor might have 15 MW of output, in this partic­ular hypothet­ical example, but the town only needs 10 MW. So, Compass comes in and installs 5 MW of bitcoin miners, co-located with the Oklo facility, to monetize the remainder of the power output and make the project cost-effec­tive as a whole. As previ­ously described, bitcoin miners are uniquely suited for this, since they can go to where the power is and operate in remote areas if need be, to fill any gaps.

In addition, all of that bitcoin mining equip­ment naturally needs some personnel and security onsite and would be co-located with the Oklo facility. Due to that combi­na­tion, Oklo’s facility now has enhanced security and onsite monitoring as a side-effect of being teamed up with the bitcoin miners that are paying for the extra portion of Oklo’s power. That appar­ently solves their margin and security problem.

If the town needs more power years later, the Bitcoin miners could reduce their load and move their machines elsewhere. If the town ends up needing less power years later, Compass could bring in some additional miners to fill that gap. This reduces the risk for Oklo.

I don’t know if the Oklo facil­i­ties in partic­ular, will catch on or not, but this sort of proof of concept and scaling approach for new clean types of energy will be inter­esting to monitor, thanks to the flexi­bility of Bitcoin mining.

As another example, the startup PRTI converts wasted tires into hydro­carbon commodi­ties. The world produces over a billion wasted tires per year, made out of hydro­car­bons, and the majority of them are just burned or buried.

PRTI devel­oped a unique sealed boiler process to take those wasted tires and boil them down into their various hydro­carbon commodi­ties, and sell those commodi­ties. However, they also produce some natural gas in this process, which they can use to generate electricity. Since their locations tend to be wherever the tires are rather than in dense popula­tion centers, their local electrical grid doesn’t gener­ally have very much use for that electricity. And it’s not enough spare gas to build a pipeline or other­wise use it for many purposes.

So PRTI takes that extra natural gas that they generate and mines bitcoins (and ether, at least until Ethereum switches to proof-of-stake) onsite with it.

As DCD reports:

“Product Recovery Technology Inter­na­tional (PRTI) is processing discarded tires at a site in Franklinton, north of Raleigh, to create oil, syngas, carbon char, and steel. It is then using the gas to generate electricity which it uses in Bitcoin miners in its onsite office. The process is scalable, the energy could be used for other purposes, and the company is planning to roll it out in other sites, including Europe.”

The funny thing is that even that article gets it wrong. They went and stuck a random anti-bitcoin section later in the piece:

“Of course, making Bitcoin isn’t an environ­mental benefit in itself, as cryptocur­ren­cies simply burn energy to produce abstract value. Crypto mining uses more energy than a country the size of Argentina and has a very large carbon footprint because not all the energy they use comes from renew­able sources. Even when it is done with renew­able energy, it diverts that energy from other uses, thereby increasing the carbon footprint of the human race.”

If that author had researched the subject deeper, he would know why PRTI mines bitcoin and ether with their excess energy rather than send it back to the grid. As PRTI co-founder and former CEO Jason Williams explained on the Investor’s Podcast Network, the local grid providers offer an extremely low price per kWh to buy energy from PRTI since they don’t need that energy because these plants are located near waste sites, not in popula­tion centers.

Market prices dictate a lack of demand like that. This stranded energy that PRTI creates from recycling tires that would other­wise be burned or buried is not “diverting” energy from other uses.

Instead, PRTI merely keeps the relatively small amount of energy they create, which would other­wise be wasted or sent non-econom­i­cally into the grid, and mines bitcoin and ether. This helps keep their business profitable so that they can continue to grow to do the good work of recycling tires and cutting down on that massive global source of pollu­tants and litter. Crypto mining just happens to be the most economical use of the bit of energy they produce from their innov­a­tive process.

5. Old Power Plant Refurbishment

Since Bitcoin miners are mobile and can make use of energy that other sources cannot, sometimes previ­ously shut-off power plants can be refur­bished and turned back on to mine bitcoin and make money, create jobs in the commu­nity, and send electricity back to the local grid as well.

A partic­u­larly popular region for this is Quebec and certain other regions of Canada. They have plenty of unused hydro­elec­tric capacity just like China, so Bitcoin miners have grown in that region. 

As Bitcoin Magazine reported:

“For example, British Columbia and Quebec, two of Canada’s largest economies, were built on resource extrac­tion, mainly forestry and tradi­tional mining. But many of the region’s lumber and pulp and paper mills powered by hydro­elec­tric dams have moved on or closed, leaving power sources and infra­struc­ture behind.

Areas of Washington State and Upper New York State have similar advan­tages — lots of cheap stranded energy from a time when manufac­turing dominated the region and facto­ries needed to be near power sources to be economical.”

I can confirm this with Compass Mining. 

Looking through their facility list, many of them are in Canada, using some of this under­used hydro­elec­tric power.

However, not every old power plant is a stranded hydro asset. Some of them are old coal plants. A May 2021 article from the WSJ gave a contro­ver­sial example of this:

“One of the most ambitious — and contro­ver­sial — projects comes from private-equity firm Atlas Holdings. Based in Green­wich, Conn., the firm special­izes in turnarounds of troubled compa­nies. It bought the Greenidge coal-fired power station in 2014 after the plant in Dresden, N.Y., had been shut a few years earlier because it was econom­i­cally unattrac­tive to operate.

Atlas first converted the plant to natural gas from coal. Then, last year, it launched a data center for mining Bitcoin using power the plant gener­ated. The company said it currently has 19 megawatts of mining capacity and plans to raise it to 85 megawatts by the end of 2022.”

On one hand, that Greenidge power plant was unused, and thanks to Bitcoin, it was put back to produc­tive work as a natural gas plant (much cleaner than coal) and able to hire employees in the region. Plus, according to the article, it also sends electricity to the grid.

On the other hand, some folks in the commu­nity have observed that the facility heats a portion of the lake nearby, causing a negative local environ­mental impact. Of course, that facility would have the same negative environ­mental impact if it was being used for something else, and it was origi­nally built for a purpose other than Bitcoin. Bitcoin miners merely resur­rected it.

For any industry of this size, there will naturally be anecdotal examples of environ­mental issues. That’s for the local juris­dic­tion to decide whether it happens to be related to Bitcoin, aluminum refining, or anything else.

While some juris­dic­tions push back on that activity for environ­mental reasons, other juris­dic­tions specif­i­cally want to attract bitcoin miners. Kentucky, for example, signed a bill into law in 2021 to give tax breaks for Bitcoin miners to come to the state. These activ­i­ties can use stranded power assets, add power capacity to the grid, hire workers, and give the state a new taxable stream of income. Indeed, Kentucky is one of the states receiving a sizable influx of miners this year.

6. Bootstrapping Developing Country Electricity

Many low-income countries have a lot of energy resources, including hydro­elec­tric capacity. However, they often have a chicken-and-egg problem. It’s often too expen­sive for them to build the electrical trans­mis­sion and distri­b­u­tion infra­struc­ture to move that power from where it is gener­ated to where it would be consumed. And they can’t get a lot of capital because there is not a lot of produc­tive capacity sitting there ready to use electricity.

Bitcoin presents an inter­esting oppor­tu­nity for some of these devel­oping areas to build out their electrical capacity and generate revenue. If an energy source is devel­oped, bitcoin miners can come in and give that site immediate profits until that electricity is put to better use. 

As Alex Gladstein of the Human Rights Founda­tion described:

“Billions of people in devel­oping nations face the stranded power problem. In order for their economies to grow, they have to expand their electrical infra­struc­ture, a capital-inten­sive and complex under­taking. But when they, with the help of foreign aid or invest­ment, build power plants to try and capture renew­able energy in remote places, that power often has nowhere to go. 

In many countries across Africa, for example, there are vast solar, wind and hydro resources. These forces could drive economic activity, but local commu­ni­ties and govern­ments usually lack the resources to invest in the infra­struc­ture to kickstart the process.

Foreign donors and investors are not keen to support projects that do not have a pathway to sustain­ability or profits. Without strong trans­mis­sion lines to deliver energy from harvest points to popula­tion centers, power plant builders could wait years before they can run without foreign subsidy.

Here is where Bitcoin could be an incen­tives game-changer. New power plants, no matter how remote, can generate immediate revenue, even with no trans­mis­sion lines, by directing their energy to the Bitcoin network and turning sunlight, water or wind into money.

As local author­i­ties or customers gradu­ally link up to the power plant, and are willing to pay more for the energy than what miners can afford, the Bitcoin load is lowered, and commu­ni­ties can grow. In this way, economic activity and renew­able grids can be bootstrapped by Bitcoin mining. And inter­na­tional aid could provide the spark.”

 — Alex Gladstein, May 2021

Overall, I think this is an inter­esting and under-explored area. The concept got atten­tion in June 2021 when El Salvador made Bitcoin legal tender. Gladstein asked Presi­dent Bukele during an online event if El Salvador would consider using some of the country’s under-utilized geothermal resources for Bitcoin mining. Bukele responded by saying that it would likely be a good idea and then days later announced that he directed his country’s state-owned geothermal operator to develop a plan for geothermal Bitcoin mining.

It remains to be seen if it will pan out, but if it does, that would be a poten­tial revenue source for the impov­er­ished country, using clean energy. Once those sources are devel­oped for profitable Bitcoin mining, they could later hook up that power source for other purposes in the future if the popula­tion centers require more power at some point.

Bitcoin’s Proof of Work vs Alternative Methods

So far, we estab­lished that bitcoin uses only a sub‑0.1% fraction of global energy, and a signif­i­cant portion of the energy it does use is other­wise stranded or wasted, or renew­able. It can also help bring new types of energy to market thanks to its flexibility.

But can we make a blockchain that is even more energy efficient?

That’s what proof-of-stake purports to do, compared to Bitcoin’s proof-of-work model. Many newer cryptocur­ren­cies use proof-of-stake as their consensus and security model.

As previ­ously-described, proof-of-work is a system where miners compete with electricity and processing power to build the longest blockchain, which becomes the accepted blockchain.

In contrast to this, proof-of-stake is a system where holders of the cryptocur­rency lock up or “stake” their coins, use them to vote on the valid blockchain, and get rewarded with more coins for success­fully creating new blocks. Instead of commit­ting electricity and processing power to create new blocks on the blockchain, they’re commit­ting their stake of coins to do so.

Proof-of-work is simple because there is no need to punish bad miners that try to validate the wrong chain or make invalid blocks that don’t fit the rules of the node network. Their punish­ment is simply that they spent electricity on blocks that weren’t valid or weren’t included in the longest eventual chain, and thus lost money. They self-inflict their own wound, and thus it rarely happens on purpose. There is a tangible connec­tion between the blockchain and real-world resources.

Proof-of-stake is more complex because there is no connec­tion to real-world resources, and the system needs a way to punish stakers that improp­erly vote on the “wrong” chain.

In addition, they need a way to ensure stakers aren’t voting on all possible chains (which can’t be done with proof-of-work because it takes real-world resources). So, proof-of-stake is a much more complex system that will take away stakers’ coins if they vote improp­erly and has ways of checking to see if they are voting on multiple chains.

Ben Edgington, a devel­oper for Ethereum and someone who is in favor of Ethereum’s upcoming shift towards proof-of-stake, went on the Compass Mining Podcast and explained the long-term challenges that Ethereum has faced as it under­goes its multi-year (and long-delayed) shift from proof-of-work towards proof-of-stake:

“The reason it has taken a while, you know we’ve relied on proof-of-work in Ethereum for five plus years, is that proof-of-stake is compli­cated. Proof-of-work is funda­men­tally very simple, is easy to analyze, is easy to imple­ment and deploy, and proof-of-stake has a lot of moving parts. You can code up a proof-of-work algorithm in a hundred lines [of code] or so. Our current clients are a hundred thousand lines or so for proof-of-stake.

And I think the theoret­ical founda­tions for proof-of-stake have taken time to mature. It’s not obvious how to make it robust; there are attacks like long-range attacks and things that just don’t exist in proof-of-work that we’ve had to think through and come up with solutions to, so that’s just taken time. So we’ve relied on the tried and tested proof-of-work algorithm, which served Ethereum well.”

The host in that podcast discussed how early propo­nents of the Bitcoin network were initially inter­ested in proof-of-stake but deter­mined it had too many attack vectors. He then asked Ben how Ethereum and the proof-of-stake model defends against those attack vectors. Ben thinks it is robust and is in favor of it, and described proof-of-stake’s workarounds as follows:

“The initial diffi­cult one to solve was what we call ‘equiv­o­ca­tion’ which means that it is basically costless to produce blocks, so if I am a proposer of a block, I can propose two competing blocks, or three, or a hundred, and broad­cast them to the network that has no real way to distin­guish between these blocks. That can be extremely disrup­tive and attack the chain, certainly split the chain, and so, we deal with this through a mecha­nism we call “slashing”. And so for a proposer to propose conflicting blocks, that is a slash­able offence. The network can detect that. Another proposer can come along and say here are two blocks that were proposed by the same validator at the same time, their signa­ture is on it, so that can’t be faked, so here is a proof that they acted incor­rectly. And then part of their stake is taken away from them and they are rejected from the network at that point. So you only get once chance.

In proof-of-work, if your 51% attack fails, you can just crank it up and do it again and again. In proof-of-stake, you get one chance, you’re slashed, you’re out of the network, and your ETH is locked up for a while, and so it’s kind of self-healing, in that respect. So that was a major theoret­ical break­through that kind of made people think, “actually we can kind of do this, there are fixes to common attacks.”

“Another one is called a long-range attack and this is kind of subtle, but the idea is that once you’ve exited the network as a validator, you can then go back in time, effec­tively. So I exit the network and I can go back a month in time, and produce (if I have enough validator keys) as many histor­ical blocks as I wish, I can write a different history for the chain, effec­tively, which conflicts with its current history, and I’ve exited so I can’t be slashed anymore. So that’s a longrange attack. We have an analysis of this, and an under­standing of this, which Bitcoiners will hate but we call it ‘weak subjec­tivity’.

It’s the idea that anybody who is contin­u­ally online is always safe, because they are monitoring the chain and they always know what the correct chain is. If you sync from scratch, you know you sync from genesis, there is a danger that you follow an attacker chain, so you need a check­point, which guaran­tees that you are on the right chain, which you need to get from someone who has been online for the entire period or somebody who is guaran­teed to be on the right chain. Now that is called “weak subjec­tivity.”

There are rules about how frequently these check­points need to be produced, how we can rely on them, and we are building “somewhat trust­less” mecha­nisms for getting hold of these check­points. It is, I under­stand, a deep clash with Bitcoin ideology in that sense that anybody in vacuo should be able to sync up from genesis and know they are on the right chain without trusting anybody in any way, shape, or form. We’re not doing that. That seems to be very diffi­cult with proof-ofstake, that’s a compro­mise we made, but we believe in practice this is completely workable and will not lead to any practical attacks of any sort.”

Besides a larger amount of complexity, trust, and attack surfaces, the main issue with proof-of-stake is that it can be prone to central­iza­tion. The more coins you have, the more voting power you have, and those with the coins are also the ones earning the new coins from staking. Since they don’t need to expend resources to stake, they can simply increase their overall staking amount as they earn ongoing coins from staking rewards and exponen­tially grow their influ­ence on the network over time.

It would be like a polit­ical system where you get a vote for every hundred dollars you have and then also get paid a dollar by the govern­ment for casting a vote.

Mary, the high school science teacher with $20,000 in net worth, gets 200 votes and earns $200 from the govern­ment for voting. Jeff Bezos, with $200 billion in net worth, gets 2 billion votes and earns $2 billion from the govern­ment for voting. He’s a more valuable citizen than Mary by a factor of a million. Also, he gets paid more by the govern­ment for already being wealthy.

That’s not a system many folks would like to live in. Eventu­ally it would likely consol­i­date into an oligopoly, with a handful of multi-billion­aires control­ling most of the votes. If it gets too central­ized, that kind of defeats the purpose of a decen­tral­ized blockchain.

Instead, that proof-of-stake system mainly works well for stakes in private property, like corpo­ra­tions. In a corpo­ra­tion, each share is worth a vote for proposals and board seats since the owners decide what the company will do in propor­tion to their ownership.

So, I don’t consider the proof-of-stake model bad for other cryptocur­ren­cies to use if they are more like a corpo­ra­tion. In fact, proof-of-stake can increase the cost of attacking the protocol since an attack or group of attackers would need to acquire more than half of the coins (unless they find and exploit a bug due to the greater attack surface). There are certain Defi projects or platforms, for example, that can operate like a company and use proof-of-stake to be efficient and costly to attack if all goes well.

Instead, I just consider proof-of-stake to be unsuit­able for a democ­ra­tized, decen­tral­ized, censor­ship-resis­tant global money, especially as this article shows how negli­gible the negative impact of Bitcoin’s energy usage is for the world, along with bringing several positives.

Or as Adam Back, co-founder and CEO of Block­stream described:

“You see that with other commodity money, like physical gold. It’s a system that works because money has a cost. I think money that doesn’t have a cost ultimately ends up being polit­ical in nature. So people closer to the money, the so-called Cantillon Effect, will be advantaged.”

In a proof-of-work system like the Bitcoin network, power is more distrib­uted between miners, devel­opers, and individual nodes. Your ability to be a miner is based on your ability to put forth capital and find low-cost electricity. Rather than the entrenched miners having an advan­tage and increasing their advan­tage over time, newer miners actually have the advan­tage over existing miners because they buy the newer machines with more processing power per watt, thanks to Moore’s law. Mining businesses, old and new, are all constantly refreshing themselves, making use of new cheap or stranded energy resources.

Plus, the Bitcoin network’s designers went to great lengths to make it easy and cheap to run a full node (unlike almost any other cryptocur­rency). 

In the Bitcoin network, the real power rests with the nodes rather than the miners. If miners try to collude and invalid mine blocks, the node network won’t accept them.

Many other blockchains that have existed since the Bitcoin network make multiple trade-offs, including making the nodes require immense processing power, bandwidth, and storage so that only indus­trial-scale entities can run them, which central­izes the network into a handful of major providers.

Bitcoin’s proof-of-work and small block design keep a lot of power with the individual users. Anyone running a full node can audit the entire blockchain, verify their individual trans­ac­tions, and partic­i­pate in the network effect that ensures consensus.

How Bitcoin Survived By Not Being Proof-of-Stake

I recom­mend that folks inter­ested in Bitcoin and the broader crypto space read The Block­size War, which is a 2021 book that chron­i­cles the history of the Bitcoin network as different factions strug­gled with each other to shape the design of the protocol and to see who had the power (devel­opers, corpo­rate miners/exchanges, or individual users/nodes). It was a real-world test of Bitcoin’s level of decentralization.

Ever since the network’s early history, there was a growing divide between people who wanted to increase the block size and people who wanted to keep it small. Increasing the block size allows the network to process more trans­ac­tions per unit of time (not considering layer two solutions and side-chain solutions Light­ning and Liquid, which didn’t exist yet). However, increasing the block size also increases the bandwidth and storage required to run a full node. It thus puts it out of the reach of the everyday user on a laptop or Raspberry Pi.

If users can neither mine nor operate a full node themselves, they have to trust large-scale network providers, and Bitcoin would cease to be a trust­less, decen­tral­ized system.

After the seeds of this disagree­ment were laid from the proto­col’s incep­tion, it was from 2015 through 2017 that the block­size war went into full conflict.

At one point in 2017, over 80% of miner processing power, the biggest maker of bitcoin mining equip­ment, prior lead devel­opers of Bitcoin, and several major custo­dians and exchanges, including Coinbase and Grayscale, were in favor of increasing the block size with an upgrade called SegWit2x (not to be confused with the normal SegWit update). That’s an overwhelming amount of support among the corpo­rate-level players in the industry. And yet, they failed. The existing core devel­opers and the individual node opera­tors were not on board with the plan. So along with multiple other reasons, it was aborted.

“SegWit2x, (abbre­vi­ated B2X or S2X, and origi­nally called SegWit2Mb), was a failed contentious hardfork attempt outlined in the New York Agree­ment that intended to double the block size limit. The hardfork has been denounced as an attempt made by CEOs and owners of large Bitcoin businesses to intro­duce changes to the curren­cy’s protocol and devel­op­ment cycle with ulterior motives.

Though over 80% of miners signaled inten­tion for SegWit2x and the New York Agree­ment, it failed to gain any consensus among the commu­nity and Core developers.”

 — Bitcoin Wiki

If Bitcoin had been proof-of-stake, and without the real power in the Bitcoin network being among the individual node opera­tors (with those nodes specif­i­cally being designed so that anyone can run them), those big corpo­rate players might have been successful at reshaping the Bitcoin network, and that would have put running a full node out of reach of normal users and further central­ized the protocol.

If Bitcoin were built on a proof-of-stake model, where the more coins you have, the more votes you have on how the network functions, the large exchanges, and custo­dians could have used the millions of coins they held on behalf of clients to vote in their own favor.

Some folks on the big block side also forked their own blockchains out of Bitcoin throughout this war, creating large-block versions of Bitcoin, including Bitcoin XT, Bitcoin Classic, Bitcoin Unlim­ited, Bitcoin Cash, and Bitcoin Satoshi Vision. All of those have fallen signif­i­cantly vs. Bitcoin in terms of market capital­iza­tion and hash rate, as they have been rejected by the market.

Proof-of-work and small full nodes together are the main ways to keep a blockchain suffi­ciently decen­tral­ized and with the highest level of security, including the most hardened attack surface. If Bitcoin ever upgrades to a different system, it would only be with overwhelming consensus among users.

Every existing proposal to make the Bitcoin network less energy-inten­sive, and every proposed new superior blockchain to the Bitcoin network, comes with trade-offs. Those trade-offs might make sense for other blockchains but don’t make sense for Bitcoin.

Final Thoughts and Summary

There is a split in the Bitcoin commu­nity about how they should respond to environ­mental concerns.

Some of the larger compa­nies in the industry have formed the Bitcoin Mining Council to collect volun­teer data from miners on the sources of energy used. Their initial findings from Q2 2021 suggest that Bitcoin mining uses a higher ratio of sustain­able energy than just about any country’s typical electrical grid:

The general view of the Bitcoin Mining Council members and their supporters is that infor­ma­tion is useful to have, so collecting volun­teer data and presenting it to capital alloca­tors and policy­makers is inher­ently helpful. And as they found out from their surveys, bitcoin mining has a high ratio of sustain­able energy, which makes sense given the fact that bitcoin inher­ently seeks out cheap and stranded energy. This gives capital alloca­tors and policy­makers a useful set of infor­ma­tion to work with, rather than the hysteria that is often presented in media.

Others, such as some smaller miners and some individual users, have pushed back against the framing of the discus­sion and the idea of an industry council. In their view, the network shouldn’t have to bend over backward to meet environ­mental concerns to a greater degree than other indus­tries. They’re concerned that regula­tors might try to push for certain types of energy to be prior­i­tized in the network without under­standing the details of how the network consumes energy. 

For example, using stranded natural gas that would other­wise be burned away without use is about as clean as you can get, despite the fact that it’s a fossil fuel.

Both sides make valid points in my view. As an analyst, I find it helpful to have the facts about the industry. Still, I also see why many in the industry would be cautious about the framing of the debate and any attempts made by policy­makers (often based on miscon­cep­tions) to change or centralize their activities.

Research Summary

Overall, the main takeaways from this long report are as follows:

• Bitcoin provides a service that people can use to store and transfer value. So far, the market of millions of partic­i­pants has decided that this network has value. Like anything of value, it consumes energy.

• Bitcoin mining uses less than 0.1% of global energy and, by design, cannot use more energy than the utility it provides to users.

• A sizable chunk of the energy used by Bitcoin is other­wise stranded and wasted energy. This is because Bitcoin miners have the unique capability to go to remote locations and deal with incon­sis­tent power that other consumers can’t use, as long as it’s cheap.

• The network continues to be more energy efficient each year due to pre-programmed declining block subsi­dies (struc­tural disin­fla­tion). Plus, additional layers like the Light­ning Network dramat­i­cally expand its per-trans­ac­tion energy efficiency even further as they are built out and become increas­ingly operable. Like any functional finan­cial system, Bitcoin uses a layered scaling approach.

• Blockchains that use other consensus models with lower energy require­ments, like proof-of-stake, make trade-offs to do so. There is no free lunch, and these other forms of consensus are not strictly “better” than Bitcoin’s proof-of-work model since they have more attack surfaces and greater risks of centralization.

For those reasons, whether Bitcoin continues to be successful or fails in broader adoption, there’s no risk of the network using too much energy in the grand scheme of things. By any metric, it’s a rounding error as far as global consump­tion energy is concerned, with a sizable chunk of its energy usage consisting of sustain­able or other­wise wasted energy.

People focusing heavily on the environ­mental “E” side of ESG as it relates to Bitcoin often overlook the “SG” compo­nent- social and gover­nance. At the end of the day, I consider Bitcoin to be one of the most ESG assets around, just not in the corpo­rate sanitized concep­tion that the term ESG is often used in.