When you have been labeled an enemy of Bitcoin, you don’t expect to find yourself giving a speech at a Bitcoin conference. Yet, on June 3, 2019, I found myself in the middle of the lion’s den at the Value of Bitcoin conference in Munich. And interestingly, that didn’t turn out the way one might expect.
During March, 2019, I received an invite from Daniel Wingen to be a speaker and panelist at a conference that would be hosted at BayernLB in Munich. Daniel was trying to gather a group of both proponents and critics of the digital currency to cover an array of topics. My topic would, unsurprisingly, be the sustainability of the Bitcoin network. Now, I’m not a stranger to conferences (e.g. I recently appeared on stage at Consensus in New York), but my audiences generally don’t consist of mostly Bitcoiners. Hence I saw this event as an unique opportunity to engage directly with the community of the digital currency that is the subject of most of my research.
On the day itself I arrived just in time for the kick-off. The first talk of the day is by Giacomo Zucco, who discusses some of the myths surrounding Bitcoin. Here I find the first reference to the sustainability of the Bitcoin network, as one of the listed myths is that “Bitcoin will boil the oceans”. This reference first appeared in an op-ed on Bloomberg at the end of 2017, when the media started paying more attention to the energy consumption of the Bitcoin network. The articles argues that the Bitcoin network, consuming ~8 TWh per year at the time, isn’t much of an environmental concern as U.S. Christmas light decorations consume ~7 TWh per year. Even though “X wastes energy too” isn’t a particularly strong argument to defend Bitcoin’s energy use, it remains a popular red herring to this day.
In fact, Dan Held, who is the other speaker with regard to Bitcoin’s sustainability this day (and strongly advocating that Bitcoin’s energy use isn’t much of a problem), actually cites it during his speech later on (right before mine). This despite the fact that the network now consumes almost 10 times the electricity consumed by U.S. Christmas lights, and only took one year since the original statement was made to get to that point. During the rest of his speech he repeats many of the arguments that can be found here. After 20 minutes, I was up next, and I started drilling down on the numbers that can be associated with Bitcoin’s environmental impact.
Inefficient by design
Before doing so, however, I first reminded the audience that blockchain technology is inefficient by design. It’s worth repeating this here, since companies still seem to consider blockchain as a tool for optimizing efficiency. That’s not the purpose of a blockchain. In fact, independent validations by different nodes in the network are part of the core of blockchain networks. You can only create a “trustless” system if none of the participants are required to trust each other. That means a lot of repeated effort is required for processing updates. It also adds a lower bound to the efficiency of blockchain technology that is naturally higher than when using a centralised server. It’s not a bug, it’s a feature.
Proof of Work impact
Bitcoin’s Proof of Work mechanism is ultimately a tool to align and secure the network, that is independent of the other steps. In terms of sustainability, this is where the biggest pain is. Millions of machines around the globe are being used to generate 53 quintillion potential proofs every second of the day, non-stop. Altogether, these machines consume as much electrical energy as a country like Switzerland, with a total carbon footprint comparable to that of Denmark. On top of that, one has to consider that mining is being done with specialised non-repurposable hardware. With the constant introduction of more efficient machines, the network also generates a steady stream of electronic waste in the form of obsolete devices. The total electronic waste output of the network could equal that of a small country like Luxembourg.
These numbers don’t mean Bitcoin will be the end of the world at this time, but they’re also far from being insignificant. In the meanwhile, Bitcoiners do realise that the size of the Bitcoin system is insignificant no matter what angle you look at it. For example, the network processed 81 million transactions in 2018, whereas the banking industry handled around 500 billion. At the conference I learned that these numbers concern Bitcoiners more than I had initially anticipated. Current thought leaders in the industry have very little to put against this apart from wishful thinking and red herrings. And one participant commented specifically that this too easily dismisses the concerns.
Of course I received some critical feedback as well. One thing that Bitcoiners generally dislike is expressing the footprint per transaction, as they argue Bitcoin is more than a payment system. And of course there’s truth in this statement, but at the same time there is a need to have a relative measure on top of an absolute measure (due to significant differences in scale). The best measure will ultimately depend on the user’s intend for using Bitcoin. If you intend to use Bitcoin as a payment system, it makes sense to consider the carbon footprint of the alternative ways of making a payment (either through a “trustless” network or through a financial intermediary). If you intend to use Bitcoin purely as digital gold, you might be more inclined to compare Bitcoin mining to real gold mining. There’s probably not one satisfying measure, yet, what’s clear to everyone is that a disproportionate amount of resources is being used.
One thing that helped me in this regard was to show participants what happens to the value of Bitcoin (and its respective environmental impact) under the condition of mainstream adoption. I specifically asked participants what they thought would be the value of Bitcoin should it replace the banking industry (also assuming scaling is solved). I gave four options, and asked the audience if they thought the price would either remain the same, go up 10 times, go up 100 times, or even go up 1,000 times. Nobody raised their hands at the first two options, even though a tenfold increase already raises Bitcoin’s energy consumption potential to around 3% of the current global electrical energy consumption. That would be more than three times the current global data centre energy use, and also more than the total solar PV energy output.
Below is an adjusted version of this slide, as part of the feedback I received was to let go of potential energy consumption numbers for the bigger increases. I already mentioned during my speech that this becomes increasingly hard to predict due to more and more production constraints (limited resources) as the number goes higher. What, however, doesn’t change is the relation between price and the budget available to miners to spend on resources. At the moment they are receiving $6 billion in mining revenues per year. Should the Bitcoin price increase a 1,000 times (as a good part of the audience appeared to expect), those revenues would also increase a 1,000 times. We’d be talking about a budget of $6 trillion per year, which is more than the entire US federal budget ($4.7 trillion per year), just for energy. Anyone will realize that this won’t bode well for life on this planet. There’s no better way to explain why a scaling solution alone won’t fix Bitcoin’s efficiency problem.
Of course, the suggestion to consider a proof of stake algorithm instead was met with laughter. No matter how many environmental concerns a Bitcoiner may have, this is a non-starter. Hence perhaps a more acceptable solution could initially be a change to a hybrid algorithm rather than a pure PoW algorithm. I don’t intend to advocate for a specific algorithm. PoS addresses both energy and electronic waste concerns, but with many smart minds in the Bitcoin community a better alternative may be found.
After my speech, it was time for the panel discussion between me, Dan and Michel Rauchs, moderated by Peter McCormack. The only thing that really stood out here was that the audience realized that I’m “actually quite positive towards Bitcoin” as I recognize that Bitcoin allows us to do something unique (send and store value without an intermediary) and that I personally just “don’t feel it’s worth the cost”. I dislike the Proof of Work mechanism, but to me that’s not the same as disliking Bitcoin. Proof of Work is just a part of Bitcoin’s software, that has a very specific purpose, and could potentially be replaced with something better without harming the idea behind Bitcoin. Bitcoiners might just be more open to this than generally assumed, and the community should be less fearful of having this discussion.
This story was originally published on Medium.