I want to start with a story about a new, exciting form of money. It is a financial technology which promised to change the world, making it easier for people to transact with each other. Unlike other forms of money, which could be easily debased, driving up inflation and potentially impoverishing millions, this medium of exchange promised to retain its value forever. There was one problem, however. The way this new technological wonder was produced was via mining, and mining this currency involved a lot of pollution.
The innovation, the form of money I’m discussing is, of course, gold.
OK yes, this is one of those trick intros we journalists are so fond of, the point being that all of the above could apply to Bitcoin as well. But before we get to that let’s think about gold for a moment, since, aside from providing us with a useful framework for thinking about cryptocurrencies, it’s also a fascinating strand of economic history which often gets ignored.
For a long time, gold was the foundation material for the global monetary system. Under the 19th and 20th century gold standard, and then the Bretton Woods system which lasted a few decades after the war, much of the world’s money – even the paper stuff and coins which were decidedly not made out of gold, were effectively backed by the precious metal. That meant that in theory you could take a ten pound note to the Bank of England and ask for the equivalent weight of gold in return.
Now in practice the system wasn’t quite like that for much of its existence. Even so, the principle was that money was yoked to a physical material, in this case gold. From the industrial revolution to the 1970s, gold was the ultimate yardstick of value in the global economy.
This was of course long before anyone was especially worried about climate change, but if you haven’t visited a gold mine, let me tell you it is an extremely energy intensive exercise. The ore is scraped or exploded out of the ground, crushed in enormous mills, ground down to a dust, processed in solvent and heated to incredibly high temperatures.
And among the oddities of the gold standard era was that fixing countries’ currencies to a notional weight of gold (under Bretton Woods, for instance, an ounce of gold was deemed to be worth $35) meant odd things would happen when, say, a mine shut down or a new seam of gold was discovered.
The deeper issue, though, was that countries would often have to implement harsh domestic economic policies, limiting the growth in money (the obvious modern analogy is raising interest rates) purely to keep to the rules of the gold standard. When William Jennings Bryant said “you shall not crucify mankind on a cross of gold” he spoke for many American workers and farmers. When Winston Churchill decided, as Chancellor in the wake of World War One, to return Britain to the pre-war gold standard, it triggered a deep recession, not to mention fury among workers, and the General Strike.
It isn’t just that the gold standard – or to put it more abstract terms, the apparatus that determined the price of currencies – was a big deal. It was the biggest deal. It was a topic of widespread public debate. Discussion about these things wasn’t just buried in the financial pages: it was front page news.
We forget this too easily these days. When you read stories about international economics today they are invariably framed in terms of trade rather than in terms of currencies. Except that this is, in the historic spectrum, historically a bit odd. When politicians on the Allied side began to consider what to do with the global economic system after WW2 they didn’t start by talking about trade deals, but by talking about how to replace the gold standard.
The indifference so many people have these days towards the current state of the international monetary system is something I’ve found curious and frustrating. So important do I think this strand of history is that a few years ago I wrote a whole bloody book about it. The one thing I regret, in hindsight, is not including Bitcoin. For while there are many problems with the cryptocurrency, some of which may be existential, to my mind Bitcoin is the most interesting thing to happen to monetary economics in a long, long time.
You see, monetary economics didn’t go away when Bretton Woods and the gold standard came to an end in the 1970s. At that point, the link between currencies and gold was broken and central bankers lurched from one pseudo-scientific set of rules to another in an effort to reinforce trust in their currencies (everything from targeting monetary aggregates to targeting inflation to “forward guidance”). The rules that underpinned the gold standard and Bretton Woods dissolved away, with all sorts of implications for the financial system and for all of our lives (rather than boring on about it here can I might just subtly point you towards the final chapters of this).
Anyway, my point is that this stuff matters, and that we don’t talk about it enough. The end of the link between currencies and something physical might have been the right thing to do, but it has had all sorts of consequences, among them this:
This is the British pound being devalued. Or, to put it another way, look at how the price of gold, as denominated in British pounds, shot through the roof at precisely the same moment:
Now there’s nothing new about people fretting about the diminishing value of “money”. The point of that chart is that since the 1970s inflation has shot through the roof, which is another way of saying that the purchasing power of your average five pound note has been eroded in a way it hasn’t ever really been. Is this a problem? That’s an interesting debate. Is there a better alternative for the way we run our monetary system? That’s also an interesting debate.
And, while this might disappoint those in Silicon Valley, none of these debates – about the value and purpose of money, about its legitimacy and use – are especially new. The problem was just that the mainstream stopped talking about it, and these debates instead became ones that mostly appealed to heterodox monetary economists and gold investors.
But Bitcoin has certainly supercharged them. Whatever happens to the cryptocurrency in the coming years, it has got a lot more people thinking more deeply about the nature of money, currency and the role of government in it.
Bitcoin is of course not gold. But it has plenty in common with the precious metal. It is, by design, a scarce resource. It must be “mined” – though the mining happens electronically, with computers attempting to crack a fiendishly complicated code in exchange for a “block” of coins. Bitcoin is not dependent on governments: “it” is countless computers around the world run by anyone, which form a network run on the basis of a set of rules laid down by its creator, who called him/herself Satoshi Nakamoto.
But there are differences. For the algorithms behind Bitcoin were designed not only to allow people to mine coins but to exchange them as well. If you’re selling a gold bar – or for that matter a house – you will invariably end up relying on an intermediary to help connect you to a buyer, to help with the financing, to help move money from their account to yours. There is an enormous infrastructure we have to do this stuff, called the financial sector. Bitcoin promised a technology – the blockchain -which would allow people not only to mine coins but to securely transact without having to rely on the banking system. This is genuinely fascinating and potentially revolutionary.
Similarly intriguing is the way it works: rather than relying on trust and credit, which is the basis of most financial systems, Bitcoin relies instead on something Satoshi called “proof-of-work”. “What is needed,” he wrote, “is an electronic payment system based on cryptographic proof instead of trust, allowing any two willing parties to transact directly with each other without the need for a trusted third party.”
That cryptographic proof effectively involves getting each computer (or “node”) in the network to spend a lot of energy doing these fiendish cryptographic challenges. This is not a bug; it’s the very thing keeping the network secure. As Satoshi put it:
Proof-of-work is essentially one-CPU-one-vote. The majority decision is represented by the longest chain, which has the greatest proof-of-work effort invested in it. If a majority of CPU power is controlled by honest nodes, the honest chain will grow the fastest and outpace any competing chains.
Note the “CPU power” bit. The whole point here is to put computers to work here. “Proof of work” depends on there being lots of computers around the world chugging away to keep the system secure.
Before we get onto the environmental consequences of this – which I think are both an enormous deal but also, confusing as this might sound, slightly less of a big deal than you might have thought – it’s worth noting there’s another way in which Bitcoin differs from gold. While there are often new discoveries of gold ore around the world (or for instance gold resources that suddenly become economic because of new technologies) the amount of Bitcoin in existence is, by design, finite and unmoving. The system was designed so that there would only ever be 21 million coins, and that it would take 100 years to mine every single one.
The upshot of this enforced scarcity is that Bitcoin prices are skewed upwards in a way few other assets are. In a recent paper on crypto, Bank of America worked out that
a net inflow into Bitcoin of just $93mn would result in price appreciation of 1%, while the similar figure for gold would be closer to $2bn or 20 times higher. In contrast, the same analysis for the 20y+ Treasuries shows that multi-billion money flows do not have a significant impact on price, pointing to the much larger and stable nature of the US Treasuries markets.
And the higher prices go, the greater the incentive to keep on mining, so there has been increasing competition among miners to get the next new block of coins. It’s a kind of feedback loop which has pushed up Bitcoin prices into the stratosphere. There’s a genuine demand phenomenon – people piling in because they see an exciting new thing, people piling in because they see other people making money – but there’s also something else: the structure of the market makes it especially sensitive to inflows of capital.
Anyway, as more Bitcoins are mined, the more “CPU power”, as Satoshi put it, goes into the network, making more secure but also a lot, lot more energy intensive. That is the unassailable logic of “proof of work” – in the absence of trust, you need cryptographic might. This is mirrored in the nature of what constitutes a bitcoin miner. Back in the day you could mine bitcoin with a personal computer provided you left it on long enough. These days Bitcoin mines tend to be vast collections of dedicated computers all chugging away trying to earn coins. If you’ve ever wondered what this looks like in reality, check out this from my Sky colleague tom Cheshire who recently filmed in one secret Bitcoin farm in China.
Estimates of how much energy is being used to mine coins can vary, but according to Bank of America it’s running at nearly the level of the Netherlands, or just below the entire US Federal Government. Now you might think this is worth it, but there’s no denying it’s, well, a lot.
This extraordinary energy consumption is, as I wrote in my Times column this week, becoming somewhat awkward for many Bitcoin proponents. Many of them have invested because they earnestly believe this is an exciting new frontier for money. And many are now deeply concerned about the ecological consequences. There are many people thinking deep thoughts about how the system could be reformed (which we’ll get to in a moment). But there are also many people who deny that there’s a fundamental problem.
There are, broadly speaking, two routes these rebuttals take. The first is to claim that this energy consumption is both justified but also green – and that most bitcoin mining servers use hydroelectricity and other forms of power. Quite what evidence these claims have is unclear, and if anyone has it I’d really like to see it. Because the vast weight of evidence suggests that instead, Bitcoin is primarily reliant on fossil fuels for its energy.
How do we know this? Well, we don’t for sure. But we do a pretty clear steer on where most of the bitcoin mining is happening: China. And we do know that the majority of Chinese electricity is generated via coal.
Ah, you might respond, but there are lots of hydroelectric projects in China. What about the Three Gorges Dam? What about all the hydro in Sichuan or the solar farms in Qinghai Province, which are generating more energy than anyone knows what to do with? The short answer is while there is certainly a decent chunk of Bitcoin mining happening in Sichuan, the majority is happening in Xinjiang province. And the vast majority of Xinjiang electricity comes from… fossil fuels.
The upshot of all of this is that, as far as we can tell (and as I say, this is reliant on some estimates – though I’ve yet to see compelling work that disproves this) the carbon footprint of bitcoin mining is incredibly high. Higher than almost any other activity. Put it this way: According to Bank of America, a single Bitcoin purchase worth $50,000 has a carbon footprint of 270 tons. Or, to translate, buying a Tesla Model 3 with Bitcoin (you can do that these days!) is equivalent to driving a petrol car round the equator 34 times. And that’s before we get onto the carbon footprint of the Tesla itself.
(Small but important note: I’m not sure whether BoA are simply talking here about the cost of mining $50k or using $50k of already-mined-Bitcoin for a purchase. I assume the former, which isn’t quite the same as a typical purchase. But I’m not sure. I’ve reached out but am yet to hear anything back)
And, frankly, most intelligent Bitcoin investors and advocates admit that the emissions are significant and somewhat concerning. But that brings us to the second justification, which has been knocking around in cryptocurrency circles for a while but which I most recently encountered in this letter from Kjell Inge Røkke, one of Norway’s richest men.
Here’s the key bit:
Bitcoin is, in our eyes, a load-balancing economic battery, and batteries are essential to the energy transition required to reach the targets of the Paris Agreement. Our ambition is to be a valuable partner in new renewable projects.
It’s quite something. Bitcoin is not a problem for the climate. It’s the solution!
But what does this mean about bitcoin being a battery? It’s a somewhat odd idea but once you get your head around it it makes some sense. A normal recyclable battery takes in electrical energy (from your mains socket) and converts it into chemical energy, which is then expended when you discharge it. Rokke’s point is that Bitcoin could be much the same, absorbing electrical energy and converting it into financial energy which can then be used elsewhere.
But here’s the thing: this hinges on a few big (and I fear incredibly over-optimistic) assumptions. The overarching assumption is that there are lots of places in the world where there is excess power. In China there are some dams which produce more electricity than the local grid can handle. Same thing for some solar farms and wind farms.
You can see the story in the map above. In some provinces there is more supply from these renewable resources than there is demand. So the provinces are forced into curtailment – cutting the output of that green power so as not to overload the grid. (sidenote: a lot of people in the bitcoin world call this “stranded” energy, but in the energy world, the word “stranded” has a subtly different meaning, referring to old fossil fuel era assets which will soon be obsolete. So in the interests of clarity I’ll use the word “curtailed”).
So there is indeed a legitimate issue here where, in the coming years we’re likely to have some temporary oversupply of green power in certain areas. The idea behind Bitcoin-as-battery is that miners could locate in these places, only using the excess energy – flipping on when there’s a lot of wind or sun and switching off when there isn’t.
It’s a nice idea. And, like most nice ideas it’s not altogether original. Actually this model – of high energy products swarming to places where there’s lots of green energy – is precisely the basis of the modern aluminium industry. That’s why you get smelters near hydroelectric plants or the geothermal plants in Iceland.
But here’s where it starts to break down. First, this model relies on the notion that rather than mining in places where energy is cheapest, which is to say China, Russia, Iran and other developing economies with a lot of cheap fossil fuel power, all bitcoin mining suddenly shifts to places where there are lots of renewable resources. Given Bitcoin mining is, by design, unregulated, how do you stop small black market operators (or for that matter state sanctioned operations in China) flouting your Bitcoin-as-battery model and just working away off fossil fuel power? I’m not sure there’s a compelling answer to this that doesn’t involve the kind of big government intervention Bitcoin enthusiasts are not keen on.
Second, given this model is already used by aluminium smelters, isn’t there a risk that Bitcoin ends up crowding out the aluminium industry in these areas? And frankly it isn’t just aluminium which uses this model. We are in the early days of a green revolution where many different industries are about to switch from reliance on fossil fuels to reliance on electricity. Consider steelmaking: green steel production involves far more electricity demand than regular steel. Consider green hydrogen production: it likewise is highly energy hungry.
My point is that there is no shortage of other sectors attempting to seek out areas where there’s excess green energy.
Now on the one hand there’s nothing wrong with creating a little extra demand. Might Bitcoin perhaps serve to increase investment in green energy in these places? Perhaps. Except that we already have so much investment going into these new green fuels and industrial techniques that the real constraint on getting more wind farms and solar farms built is not so much money or price but the availability or raw materials. And those raw materials include, well, steel and aluminium and hydrogen. There’s a circularity here…
Anyway, the final questionable assumption behind the bitcoin-as-battery model is the notion that these areas with curtailed energy will always have curtailed energy. Since that map above was made, the proportion of curtailed solar and wind energy in China has fallen considerably. This is a good thing: China’s electricity grid is getting better so less renewable energy is getting lost. But it also somewhat undermines the argument that this is a sustainable long-term model for such an energy intensive activity.
Now, there is a solution – in theory if not in practice – for all of the above. Bitcoin could change the model whereby coins are minted. That’s what many of the other cryptocurrencies which currently use proof-of-work are considering. Ethereum is contemplating shifting to “proof of stake”, where computers get awarded coins occasionally, provided they’ve put in enough money. I’m not sure of the constraints that would or wouldn’t prevent Bitcoin itself from doing likewise, but would be interested to hear what they are, if any readers have any ideas.
Bitcoin advocates get a bit irritated that when people refer to Bitcoin they don’t reference the existing climate impact of other activities. Isn’t gold mining terribly environmentally damaging? What about paper currency? What about the fumes from private jets chartered by big banking chiefs? These are legitimate points.
However every single one of those emissions sources happens to be going down these days. Big time. Bitcoin is heading in the opposite direction. The more money goes into it, the more carbon it emits. It is a deeply troubling issue, especially for the people who hope that it could cause genuine reform to global finance.
And overshadowing all of this – everything from gold to currencies to Bitcoin – is another, deeper question: value. Do we value proof-of-work enough to think the potential damage to the climate is worthwhile? Does it matter more than, say, the green aluminium or steel or hydrogen industry it threatens to crowd out? More broadly, does the promise cryptocurrencies hold to change the nature of money prefigure a new era where fiat money meets its maker? Or will central bankers and governments do what they’ve invariably done when the private sector devises something which threatens their ability to tax and regulate the financial system: clamp down on it or try to kill it?
I have hunches about all of the above. So do you, no doubt. It’ll be interesting to see what happens next.