Isn't Bitcoin useless anyway?
tl;dr: No. Hundreds of millions of people worldwide are finding it useful for a lot of different use cases. It's a pretty arrogant stance to assume you know better than all of them.
Bitcoin can be seen as a circuit breaker or relief valve for financial systems everywhere. It gives people a way to circumvent financial censorship, capital controls or inflation, and it enables permissionless access to an alternative financial system. It cannot discriminate against people for whatever reason. Let's look at some examples:
- Here's a video by Alex Gladstein of the Human Rights Foundation enumerating humanitarian causes that Bitcoin has been useful for. Here's a thread, if you prefer text.
- As a prominent example, Wikileaks probably wouldn't exist anymore if it wasn't for Bitcoin enabling donations for them when the US government pressured other payment providers into blocking them.
- Here is a fascinating analysis of Bitcoin serving as a vehicle currency in Latin America.
If you find Bitcoin useless for yourself, you should call yourself lucky.
I have read somewhere that every Bitcoin transaction is "equivalent to the power consumption of an average U.S. household over 22.86 days.". Is that true? Isn't that horrible?
tl;dr: Most people don't value Bitcoin as a payment mechanism, and even if they were, most payments are not recorded on-chain. So calculating the energy cost of a transaction is as impossible and nonsensical as it would be for gold.
It's true in a similar way that the lack of pirates is causing global warming. Yes, this is a valid estimation of Bitcoin mining's energy use divided by a valid estimation of on-chain transactions, compared to a probably valid estimation of household electricity use.
However, connecting these three figures in this way doesn't yield any insight beyond shock and awe. That's because Bitcoin's primary value is not as a payment network, so the number of on-chain transactions is not an interesting metric for it. This can be seen easily: On one hand the number of onchain transactions has always been limited — and that limit has been pretty much hit constantly since at least 2019. On the other hand, the value of Bitcoin, as well as the estimated number of its users, has been rising over many orders of magnitude since 2009. How can this be possible?
- In recent years, long term store of value has become Bitcoin's main use case. That is, people use it more for storing their savings than for paying for goods and services.
- Its current usage is thus more comparable to gold, which is also rarely used in transactions. I have never seen anybody calculate the energy cost of gold mining plus handling divided by the number of physical gold shipments. Probably because that's as misleading a number as the equivalent in Bitcoin.
- Just as with gold, there are ways to make even small transactions quickly and cheaply with Bitcoin. Most real-world transactions use one of these off-chain methods. Probably the most popular today are custodial services like exchanges. Sending bitcoin from one customer to another on a service like this just requires changing two numbers in their database and is the equivalent of gold in a bank changing owners. It's hard to say how many such transactions there are, since there are many "bitcoin banks" and their numbers are proprietary, but it's safe to say they dwarf on-chain transactions by orders of magnitude.
- Unlike gold, there are methods of transacting Bitcoin off-chain that require less trust in intermediaries, such as the Lightning Network or Blockstream's Liquid federation. With a growing demand in transactions, such alternatives are refined and gain popularity, without this being reflected in the above numbers. With the Lightning Network, it's even difficult to estimate how many transactions are being performed, since there is no central ledger recording them all.
But I have read that Visa transactions are much more economical!
That's because these are very different things. Visa transactions are just a few companies changing a few numbers in their databases. It's basically just a fancy way of signalling an intent to spend something. It's not final. There might not be a real transfer of value happening at all (all the "money" is still debt in the same bank, just on another account).
On the other hand, a Bitcoin onchain transaction is a real transfer of a bearer instrument, comparable to a gold shipment from one central bank to another. Those are also quite expensive, I imagine. It's total overkill if you want to buy a coffee. For that, it might be enough to just update some numbers in a (Bitcoin) bank. Or use the Lightning Network.
If you have a little time, Nic Carter has explained it much better than me.
Will Bitcoin boil the oceans?
tl;dr: No. It does use a lot of energy and this also leads to a lot of CO2 emissions. See the current best estimates on how much here. It's hard to predict how this will develop in the future, but it will not grow exponentially forever.
It is difficult to make predictions, especially about the future. This is why even within the Bitcoin community, you will find contradicting (but dead-sure) prognoses for future hash rate, mining cost, energy use, and CO2 emissions. But in the first place, it is important to clearly distinguish those four values.
Bitcoin's hash rate has grown in leaps and bounds from some megahashes in the beginning to triple digit exahashes (that is, xxx**10^18 hashes per second, or xxx million million megahashes) today. That means it has increased several trillion fold in 12 years. If electricity demand had scaled up comparably, mining would use more energy than is produced in the world by now. But luckily, hash rate is not what is important to Bitcoin's economics at all.
The metric that ties everything together is the cost of mining. If miners behave economically (which they mostly appear to do) it should converge toward the block reward, which in turn is composed of the block subsidy ("new bitcoins being mined") and the mining fees ("what people pay to get their transactions into a block"). The block subsidy is halving every 210.000 blocks (about 4 years), which has happened thrice already, so it has gone down from 50 BTC to 6.25 bitcoin today and is further exponentially declining — when calculated in BTC. The mining fees have varied considerably over the last few years. They are now about 1% of the block reward most of the time, but can reach more than 30% in hype phases. It is important for Bitcoin's future security (when the subsidy has declined further) that these remain substantial. How substantial? Nobody knows. The hope is that the market will find out.
Of course, when denominated in, say, USD, the cost of mining has not fallen but risen exponentially in the past 12 years, because the price of bitcoin has risen even more. But it is important to observe that
- mining cost is rising exponentially slower than hash rate (because of technology improvements)
- mining cost is rising exponentially slower than bitcoin price (because of halvenings)
- mining cost in BTC is relatively easy to predict in the near future (as long as it's dominated by subsidies)
- mining cost in USD is mostly dependent on the price of bitcoin, has been growing exponentially so far, but might top out when the bitcoin price stops growing by more than 100% every four years
Now for the third metric. Energy use depends on many factors. Mining cost is probably the most obvious one. Availability, longevity and cost of mining equipment have to be taken into account as well, and they are also changing in interesting ways. However, the most interesting aspect in this is probably the price of the electricity used. As Bitcoin mining has become more mature, miners have become both industrial size energy users and mobile enough to seek out the cheapest energy on earth. This has fascinating consequences:
- energy use might have grown more than mining cost, due to miners gaining access to cheaper energy
- the energy that miners use today is mostly very cheap, because it is mostly surplus
- energy use is inversely correlated with energy price. So if regulators succeed in raising the price of dirty energy, miners will use less of it.
Which brings us to the final and most decisive metric, CO2 emissions. It is the most difficult to measure and predict, because it depends critically not only on the amount of energy used but on the kind of energy used. Unfortunately, many mining operations are not very public. This leads to the existing research (see our sources) being imprecise in deducing CO2 emissions from energy use. Although the estimates of the percentage of renewable energy used in mining vary, it is relatively safe to say it exceeds that of most countries by far. Since the studies extrapolate from power usage to CO2 emissions by estimating the amount of mining in a region and then multiplying by the average carbon footprint of energy production in that region, they are probably overestimating. In fact, there are two key questions here, and the second one has not been addressed at all in academic research, as far as we know:
- how much CO2 is emitted by the energy production?
- what would have happened otherwise, if the Bitcoin mining didn't happen?
There is substantial evidence that for both of these questions, the trend is positive. States will tolerate miners abusing fossil energy subsidies only so long. As pointed out in another question, energy providers are using Bitcoin mining in order to overprovision renewable energy. Mining is used in place of flaring or even just venting surplus natural gas. However, there are also reports of mining making stranded oil or gas fields profitable again. It remains important to look closely at all of these examples and see if they are really net positive, but one could be forgiven for being optimistic.
Isn't proof-of-work useless? What is it for?
tl;dr: No. It's a subtle topic.
Adam Gibson wrote a slightly rambling but still quite elegant exposition on this topic. I recommend reading it. That said, here's me trying to explain it a little more concisely:
Proof-of-work is short data that allows for cheaply verifying the fact that it was (with overwhelming probability) expensive to obtain. This short data (a 256-bit hash in Bitcoin) can also be referencing other — often larger — data sets, and thus testifies those must have existed when the proof-of-work was generated. It is serving multiple important functions in Bitcoin. I think these are the two most irreplacable ones:
Firstly, by looking at the proof-of-work attached to multiple competing versions of transaction history ("blockchains"), it is easy to see which history has needed more work to write down. The version of history with most work attached is deemed the currently valid one by all bitcoin clients. No context or other prior knowledge is needed to verify this. As long as a client that has been newly connected to the Bitcoin network is able to get one copy of the correct history, no amount of falsified histories is going to mislead him, for it is easy to spot the most expensive one. This is comparable to trusting a bank because is has skyscrapers and lots of people working in them over any amount of other "banks" that have just opened a stall in your street. The real bank wouldn't go to all these lengths just to trick you into giving them your money. Just like miners wouldn't spend giant amounts of work just to trick you by giving you made-up version of history in which you received money that in reality was spent to someone else.
Secondly, because it is very expensive to make, proof-of-work is the fairest distributed way of distributing a new asset that we know of. Miners spend almost as many dollars on hardware, energy and work as the bitcoin are worth that they receive in return. But because there is a lot of uncertainty in the process — hard-to-forecast things like the number of competitors, the price of electricity, hardware improvements and the bitcoin price can change mining profitability drastically — there is often a premium to be earned over the pure cost, which keeps miners interested.
Couldn't we just use something else like proof-of-stake or proof-of-anything-else?
tl;dr: As far as we know, there is no free lunch here.
Proof-of-stake is similar in concept to perpetual motion: It tries securing an asset against forging of multiple new histories by pricing punishment in terms of the asset itself while rewarding honest behavior with the asset itself. There is no indication that this is possible, in spite of many market participants making claims to the contrary, just like with perpetual motion.
One popular misleading approach is to make the system so complex that even experts cannot determine what security properties are even being pursued. It's possible to just define all practical problems away in theory. When these systems are finally built, and inevitably attacks are surfacing, the system is simply made even more complex, which might rule out one concrete attack, but opens up several others. Notice that this sleight of hand wouldn't even be possible in a truly decentralized system like Bitcoin, because the system can't simply be changed by a few insiders.
Notice too, that even if securing the ledger with proof-of-stake were feasible, that wouldn't solve the distribution problem at all.
As for more exotic proof-of-something, there have been many claims made of "the better Bitcoin" here as well, but alas, something has always prevented those from making real progress. Of course, just because so many extraordinary claims have been made and serious researchers have failed to come up with a better system for 12 years now, doesn't mean it's impossible. Just not something we can rely on for now.
Can't we just change Bitcoin?
No. Bitcoin's most cherished value proposition is its resistance to changing central economic properties. Proof-of-work is an important economic principle, even if there were a technical alternative. But most importantly, there aren't any known alternatives with similar properties.
Can't we just ban Bitcoin?
No. It was designed to withstand opposition from state level adversaries. Banning it would require a global totalitarian regime. Do you really want that?
How could Bitcoin be a net positive force for the environment?
tl;dr: It could help finance renewable energy developments, remove subsidies for fossil energy, replace legacy environmental hazards such as gold mining, incentivize people to save instead of spend, and maybe even be the foundation of a more sustainable economic system.
Many Bitcoiners have written passionate pieces full of conviction about this topic. Here's a twitter thread full of them. What I don't like about most of these pieces is the utter conviction that feels really inappropriate.
There are indeed all kinds of arguments why Bitcoin might be a net positive force in the fight against climate change. They reach from speculative to very speculative IMHO, but that doesn't mean they can't come true. Here's a selection:
- Bitcoin miners are seeking out the cheapest forms of electric energy on the planet, pretty much independent of location, and are guaranteeing a minimum price for it. This can have many interesting consequences: It builds more market demand for research and deployment of new energy sources. But without regulatory oversight, this might also make fossil energy sources even more profitable. So far, we have seen tentative examples of both: Miners have been used to finance overprovisioning of renewable energy as well as reactivate formerly unprofitable oil fields.
- Since Bitcoin miners are buying cheap energy wherever it occurs, they may hasten the demise of energy subsidies by simply making them economically unbearable. Or they might just be outlawed. Either way, no more fossil energy based mining.
- Bitcoin might replace parts of the legacy financial system including gold mining. Gold mining is an environmentally hazardous process and might become less profitable if Bitcoin takes over part of the monetary premium that gold is commanding now.
- Bitcoin is implementing the idea of sound money. That is, the Bitcoin supply is scarce (even more predictably than gold). This implies that it should always become more valuable over time. Therefore, saving Bitcoin is more appealing than saving fiat money, which tends to become less valuable over time. In order not to depreciate, fiat money always has to be invested. But every investment carries both risk and environmental costs. Bitcoin might become an alternative that is both less risky (in the long run) and less environmentally problematic than the average investment. Thus, people might be incentivized to hold Bitcoin over longer time frames, and this holding time might be, say, less carbon intensive, than having the same amount of buying power being spent potentially multiple times in the economy.
- From its inception, Bitcoin has been a counter movement to the legacy banking system built on credit expansion. Many people see this system of money production as a key driver for inequality, unlimited economic growth and environmental hazards. Bitcoin might be the foundation of a different system that could address some of these problems differently.
If Bitcoin could be eventually helpful against climate change, why do you want to do anything at all?
Notice the COULD. It means these visions are both uncertain and certainly not realized today. Climate change is a problem today and needs to be addressed ASAP. Even if Bitcoiners' visions of a better world through better money come true, it certainly won't be overnight, and it might be too late. In the meantime, there is no question that Bitcoin mining does contribute to climate change today. That's why we want to act now.
How could we help fight against climate change?
Bitcoin is money. Bitcoiners — hodlers, miners, exchanges, businesses — are making money. Money can help solve problems. So we want to encourage you to give some of that money to causes that you believe can make a difference in the fight against climate change. And tell us about it, so we can count it, and one day tell the world that we are sure that Bitcoin is a net positive money.
Where can I donate?
We do not want to take your money. In the spirit of Bitcoin, this is a voluntary and decentralized effort. So ideally, you select the cause and the amount you want to give, estimate the amount of CO2 offset, and report that to us, as anonymously or publicly as you want. In order to make that easier for you, we are curating a list of organisations that we believe are helpful and that work with us to automatically report the numbers, so we can measure our impact as bitcoiners. And we have made the calculator page to give you an idea of how much CO2 you might want to offset.
In general, we'd recommend indexing risk by donating (as a community) to a wide array of initiatives with different approaches. You can just select those that you find most appealing. If they aren't on our list, just contact us to let us know that you want to give to them. If they meet our standards for inclusion, we'll gladly add your contribution to the total and maybe even add them to our recommendation list.
How much money are we talking about?
Looking at the calculator page, the current estimate of total CO2 emitted by bitcoin mining in its lifetime is close to 100 Mt. Classical offsetting measures or carbon credits would price this at about $20 per ton of CO2 or $2 billion. Which is less than 0.4% of the current value of all bitcoins in existence, or about $120 per Bitcoin.
However, as you can see on our list of charities, there are initiatives that promise a much higher impact per dollar invested, to the tune of far less than $1 per ton of CO2. These tend to be a little more speculative in nature, in that there is more variance in the expected amount offset, which should be OK if we don't put all our eggs in one basket.
So it does seem reasonable that we might reach our goal of Bitcoin being a certified net positive money with less than $100 million in total donations. Which is less than 0.02% of the value of all bitcoins, or about $6 per Bitcoin.
Annual figures tend to be at about a third of the total lifetime amount at the current growth rate.
Is that a lot?
Not really, considering that fiat money is losing value at a rate of at least 2% per year. Or that Bitcoiners are already giving $300 million per year to nonprofits.