Unique properties of Bitcoin mining
Marcus Dapp: Let’s dig a bit deeper: Bitcoin’s uniqueness is revolving around the mining process. Processing transactions, rewarding miners, and spending a lot of energy in the process is all to secure the network. What is special about the energy dynamics of Bitcoin mining in the context of climate change?
Harald Rauter: There are a lot of beautiful innovations in the Bitcoin protocol. Immutable economics is very powerful, that leads to very net-positive social outcomes. In an inelastic supply, cooperation is rewarded. Your win is my win. Inelasticity, risk rewards, and cooperation are all extremely powerful.
As to mining, I think the real innovation happened in a combination of how the Bitcoin protocol has been designed: proof-of-work plus the difficulty adjustment. Those two are the key innovation elements and they create all these positive externalities. What do I mean? We know every four years, whatever you do as a miner, your block reward will be cut in half. That puts a lot of pressure on your operations and your economics as well. But at the same time, we have the “difficulty adjustment.” We adjust every two weeks how much hash power is in the system. To really understand the dynamic of how this has been playing out, we need to go back to the days when difficulty adjustment was low, energy was abundant and energy price was not really an issue. You plugged in your GPU, you mined, and you had much wealth. Soon, many people had the same idea. What happened? The difficulty adjustment skyrocketed, and sud- denly a very interesting dynamic came to play because mining is designed to be expensive from the start through the machines that you need to deploy to be competitive in the race, but also – and this is the main determinant – the operational costs under which you can run your machines.
There’s a big argument that mining uses an enormous amount of energy, the same amount as Argentina or whatever the country comparison is at the moment. But these absolute numbers are useless and misleading. They are irrelevant because you need to ask where this energy is coming from. In the past, when the operational costs for the kilowatt hour were on par, miners might have consumed energy that was competing with your toaster. Then you can decide: do I use the toaster, or do I use my miner? As an incentive driven individual, you will say it depends on what I want: to save or toast. Now, let us fast forward: difficulty adjustment increases, block reward decreases. Suddenly the economics and your incentive structure are changing. Because now, you deploy this amount of input energy, but you get way too few Bitcoin out of it to be economically viable. So you go for the toaster, and you are right. Yet, you still want to mine, you still want to save. What is the next logical step? Asking yourself: what is the next most accessible, non-demand competing energy source that I can tap into that the toaster cannot use?
Marcus Dapp: Are you disagreeing with the common opinion that the energy Bitcoin uses is taken from other people who would need it?
Harald Rauter: Yes, because the more difficulty adjustment increases, the lower the block reward gets. And the miners’ capital expenditure (capex) is a bit of an inelastic measure, too, so it averages out the at a very reasonable price. The only variable that you can influence is the operational expenditure (opex) of your energy which determines your profitability. I am just looking for the cheapest. Once we understand that we can have a fantastic environmental conversation because now we shift gears: we focus on the design of energy systems. Our energy systems are typically designed in a very centralized way with a few big actors, because that reduces the complexity to better ma- nage grid stability. Grid stability translates into reliability, and all incentive structures in an energy system are designed towards reliability. Costs do not really matter, reliability does.
However, over the past 15 years, we have seen a new game in town which is called renewable energies, and we have seen the price deflation of photovoltaics (PV), and we have seen the efficiency gains in terms of wind. There is a lot of wind and PV that is now cheap and accessible. This allows you to generate the kilowatt hour of power for less marginal cost than in the legacy system. If you take oil, for example, you need to dig deeper, build more complex infrastructure; you need to transport it, refine it, transport it again, and you need to store it. All that adds to the marginal cost of the kilowatt hour. We observe this divergence between cost depreciation for renewables and cost appreciation for fossil fuel.
Marcus Dapp: Based on this insight, would you say it is correct to claim that Bitcoin mining fosters renewable energy sources because they are cheaper?
Harald Rauter: That is one part of the equation. Let me dissect the different arguments that could come up here. First, the marginal cost of production for a kilowatt hour from a fossil energy source gets more expensive, renewables get cheaper.
Second, an argument that is brought up very often: there is a big advantage to the expensive energy that fossil fuels and natural gas can provide. It is “baseload“ energy because it is continuously available. We need this because if we are only on variable energy sources, this generates problems to stabilize the energy grid. So, fossil fuels have a particularly important role to play to provide the baseload energy that is necessary for a system to work, which is one of the major arguments against renewables, because obviously, you need wind to produce wind energy and sun to produce sun energy.
But now, we have a dilemma. We have energy generating infrastructure that gets cheaper and cheaper and cheaper. People and institutions want this energy. Not because it is environmentally friendly, nobody cares. It is incentive driven. It is economically driven that on your property in Switzerland, in Germany, in the Netherlands, you can have sovereign power based on those renewable energy sources that gets cheaper and cheaper. Yet we have this problem that it is an intermittent energy source.
Marcus Dapp: You are saying there is a maximum percentage of renewable energy that can be feasibly integrated into the grid due to their intermittent nature despite their cost-effectiveness and environmental benefits compared to fossil fuels?
Harald Rauter: That is a particularly good question. The answer is yes and no. We also need to think how this dynamic plays out over time. Let’s assume you are a miner, and you now identify that there is a lot of intermittent energy that the grid cannot use because it cannot handle the variability.
For example, the UK currently has about 100 gigawatts of wind energy infrastructure which is not online on the grid because the grid infrastructure is not able to handle it. Now, someone has incentivized building out this infrastructure. A pension fund that you and I are insured with, could be holding those projects on its balance sheet. But it is producing zero economic value! That is a problem. Not only now, but also for follow-up projects. What hap- pened in the UK? They did not find any investors for the new wind infrastructures anymore because they were not willing to accept the minimal cost for the kilowatt hour, given the project risk and costs.
Marcus Dapp: Are you saying that the barrier in the transition to renewables is economics?
Harald Rauter: Yes, 100%. We have the same situation for photovoltaics. We have an enormous amount of photovoltaic infrastructure in Europe. The problem is that across Europe the sun shines at about the same time. That means we have a surplus of generation which coincides around lunchtime right when demand is the lowest because we’re at work: we’re not consuming, not watching television, not using our toaster and so on. So, high demand in the mornings and evenings and high supply around lunchtime. This leads to a demand-supply mismatch: the energy produced and generated at lunchtime needs to be consumed there and then! The first law of thermodynamics says energy cannot be destroyed, only converted.
If you cannot bend physics, what is the only variable you can shift? It is economics. If I hold a hot potato in my hand and ask you to hold it for me, but I want twenty cents for it, you will say: no, I am not taking the hot potato. Then I change tactics and say, take it and I give you twenty cents. You still decline. So, how about one hundred cents? Okay, you take it for one hundred. So, I need to have the costs of generation and distribution, which is self-cannibalization. That puts a natural limit on the scalability of the rollout and deployment of renewable infrastructures.
This is where Bitcoin mining comes in. For me as a biochemist, miners are like mushrooms: If you have a hostile environment, the first species that comes to pioneer this hostile environment are mushrooms and algae. They prepare the ground, prepare the soil, so higher species can come to live there. When the higher species die in the end, the fungi come in again and regenerate the carbon of the higher species to make it accessible again for new life. And so, fungi have this enormous role in biology to make carbon cycles circular and get the most out of the chemical properties of the carbon itself.
Bitcoin miners have the same role in electricity cycles. They are not competing. They are pioneering, for instance, the wind farms in the UK. But once these wind farms get access to the grid, higher utility purposes will come in. Demand that is willing to pay more than a miner is ever able to pay, so they will marginalize the miner out. Then, the miners can come in again and say, you do not have to pay me anything, I do not have to pay you anything. I just take your energy, and it is even more economically viable than any other form of economic action you could think of.