Bitcoin mining is about to get a way more sophisticated business

I’ve been thinking about the convergence of bitcoin mining, the energy markets in general and the electricity markets in particular and I believe it is almost as fascinating as Bitcoin itself, so I thought that in order to structure my ideas I could write this post hoping to explain my views and a few more things at the same time for those not familiar with the matter.

In terms of geopolitics and geostrategy, it doesn’t get more fun than combining Bitcoin with energy, so there it goes:

The electricity markets have a very peculiar characteristic whose implications few outside these markets understand, which is that electricity production always needs to match electricity demand plus the transmission losses. The reason is simple, as unlike with almost any other consumption good, electricity storage at large scale is hardly economically viable at this point.

Lots of different Companies are researching in the field of storage with many different technologies, such as Tesla with its batteries and very innovative ones based on gravity (they use the excess energy to lift large stones, and drop them down when they need to generate it). There are many approaches to tackle the problem.

With hydro electricity for example, pumped storage has been working for a really long time, that is, pumping large amounts of water when there is energy in excess and dropping them down to a turbine when there is a need for it. But the amount of potential storage is limited with hydro technology.

Up until the large increase in installed capacity of renewables, it was easy to size the grid, that is, the “pipelines” through which electricity reaches its destination because energy could be easily dispatched at will, which means, you simply had to turn on or off the switch to make it work or stop it. Most energy plants could be simply be working at near full capacity with just a tiny fraction reserved as back-up.

So basically if the energy capacity of a market would be around 100GW, the peak demand would be around 90GW and the grid should be able to transport around 100GW so that there would always be a margin. The part of the grid that was idle was very small for security purposes.

If demand would grow because that country was doing good economically, then you could simply invest a bit in the grid every year. Since production would be smaller than demand, the price would rise making new investment in new capacity more interesting for utilities. So demand, production and grid used to grow at a similar speed. Problem solved.

The fever for renewables

Fast forward a few years and the renewables fever kicks in. As it turns out, these generation technologies are producing energy only when the wind or the sun wants to. There is no button that says “generate” or “stop it”. It generates at the will of nature as a result.

But since it wouldn’t make economic sense to install a wind farm and only sell the energy depending on the demand, markets integrate ALL renewable generation, and the polluting ones are stopped or initiated as needed.

What this is creating is electricity markets that are completely dependent on the weather (the industry where I worked before I went nuts about Bitcoin).

When the wind blows enough or the sun is shining enough, you get wholesale prices at zero or close to zero, stopped gas and coal plants and nuclear working at near full capacity (unless there is a maintenance stop). Nuclear works at full capacity because it takes days to switch on and off.

The problem for gas, coal and others is that if they are not paid for “being available” as back-up (availability or capacity payments), it doesn’t make economic sense to have them available without producing when the wind blows and/or sun shines enough. Even with this kind of “capacity payments” many utilities decide to shut them down.

Therefore, the whole “electricity system” has to pay the owners of these plants only to be available. The electricity system at the end, is btw, paid by all of us, i.e. consumers.

So summarizing, the situation with renewables is the following:

There are times when the wind stops and the sun goes away (it’s called nights).
We need generation as back-up available when this happens. Financing this back-up increases the retail price for consumers.
The grid has to be sized to be able to absorb as much renewable generation as possible, even if when this generation is low, the grid infrastructure stays idle. That’s a massive fixed cost nobody wants to pay for.
Interestingly, our consumption is based on the number of kwh we consume, so we are paying a (mostly) variable price when most of the cost of the system is fixed!!!!! (grid infrastructure and back-up technologies)

So the situation is that many markets are already saturated and grid operators have to choose between increasing their capex expenditure in infrastructure that needs to stay idle a lot of the time which would require increasing the end price to consumers, OR, start decreasing the amount of energy from renewables that can be integrated into the grid. At the beginning they were choosing the former, yet prices in for example Germany or Denmark are so high, that is becoming unpopular to do so.

In some nodes of some grids, there is already a structural excess of renewable energy, that is, renewable generators are not able to sell all the energy they produce and are not compensated as a result.

But if people want to keep having renewables, there is no other way to move forward than getting more and more renewable energy stranded.

Bitcoin mining

Although wind farms tend to work at a capacity factor of around 30% of the time (depending on the location, as offshore ones have higher factors for example) you could buy a 5% of that capacity and expect to get it around 80% of the time at the right location.

If you add a PV plant to that, you could expect that percentage to increase to for example 90% or even 95% of the time as solar and wind are very complementary:

Low pressures typically mean high winds and lower global radiation
High pressures mean lower winds and higher global radiation
Wind tends to blow more at nights
In the summer, radiation tends to be higher yet the heat makes the air pressure smaller and wind generation decrease too. Winter is the opposite.

You could consume a small % of a wind or wind/solar farm generation for mining. You could buy the remaining 10%-20% from wasted natural gas (if available) or from the grid, from small scale storage or even better, if your operations are spread around the world you could do something really interesting, which is:

If you are operating at 80% to 90% of the time, the capex of your operation would make you uncompetitive during 20%-10% of the time, which in mining is similar to economic suicide, unless you are buying that 10%-20% from a cheap enough different source.
Yet interestingly, when there is no wind nor sun, the spot wholesale price will tend to be way more expensive (2 to 4 times higher depending on the market) except for periods of low demand, so it could be interesting to stop your miners and sell the electricity you purchased in advance for a way higher price (I haven’t ran the numbers, but such margin should be enough to buy BTC in an exchange at a cheaper cost than mining it + capex) (you can even automate this process with hour-ahead 10-minutal wind and power forecasts, both site-specific and market-wide), that is, trading future price for spot price.
Also, if you are large enough and stop your operations in an area number 1 where the kwh is punctually very expensive due to the weather, you could mine in geography number 2 with a lower network hashrate (since you stopped miners in location 1), which means, to mine at a cheaper price than usual.

Basically you could be selling expensive kwh in one location, and mining bitcoin at a lower cost in another therefore profiting from the different weather patterns across continents.

Besides, you could make some pretty decent profits with fiat, that could finance your fiat expenditure, and you would decrease substantially the need to sell BTCs at the wrong time, thus protecting your business from volatility. Given that bitcoin’s price is the biggest factor in mining’s profitability, this would be a very interesting competitive advantage.

Also, the fact that you can power on and off miners is extremely valuable for the network as it is equivalent to having the ability to inject and stop power into the grid at will at a pretty decent scale in a short time.

Just as back-up technologies are expected to inject power to the grid, large consumers are paid to stop consuming energy, so that more is available for other purposes. In Spain this is called “interruptability”. Large interruptability at large scale is equivalent to owning several nuclear plants that you can switch on and off in minutes. This is extremely disruptive and helpful for the integration of renewables.

The electricity markets are only a fraction of the entire energy market and it is going to be really interesting to see how all this plays out.

Published at Tue, 03 Dec 2019 13:35:30 +0000

{flickr|100|campaign}