Energy consumption of Banking (related to Bitcoin)

Energy consumption of the Bitcoin blockchain is huge and could therefor be a potential threat to the sustainability of the cryptocurrency. In this video, Andreas Anatopoulos makes an interesting statement the energy consumption of the Bitcoin blockchain is far from problematic for the following three reasons:

  • Is is a driver of sustainable energy: installing a mining farm at a new renewable power generation site can be used to consume all generated power. This is based on the assumption that not all power of the new site can be transported to consumers right from the start. Converting the excess power in Bitcoin provides and additional revenue that accelerates the ROI. In the case of Bitcoin, this is possible as there are no geographic restrictions of mining.
  • It is expected to level off: The powerconsumption of the blockchain is expected to stabilize at some point in time.
  • It is not worse than traditional banking: the energy consumption of traditional banking ICT infrastructures is also very high. Bitcoin does not make it worse.

While it is very interesting to dive into each of these statements I would like to address only one: the energy consumption of the traditional banking sector. There must be more research on this topic, which I did not dive into. An estimation of 100 TWh per year was made by Carlos Domino. In this case, the energy consumption was calculated by using the number of banks and estimation of energy consumption of offices, ATMs and datacenters to get an estimate.

Let's try another approach and see where that leads. Nowadays, banks are more transparent on their impact on the environment. As a result data on energy consumption is typically included in annual reports. As an example, the annual energy consumption of ING in 2017 was 315GWh (Annual report ING Group 2017) while serving 37.4 million customers (from the same report). This corresponds to an annual energy consumption of 8.4 kWh per customer. This corresponds to approximately the daily energy consumption of an average household in the Netherlands.

My assumption is that, on average, each individual on this planet is customer of a bank. Some have no access to banking at all, while others have multiple accounts at multiple banks. With 7.6 billion people on this planet (July 2018) results in 8.4 kWh * 7.6 billion = 64 TWh (Terawatt hour).

While my estimation of 64 TWh annual energy consumption is a typical case of n=1, I find it interesting that the result is of the same order of magnitude as the 100 TWh estimated by Carlos, which appears to me like a confirmation.

The current annual energy consumption of Bitcoin is estimated at 71 TWh (Digiconomist, July 2018), and is of the same order of magnitude.

My conclusion is that the energy consumption of Bitcoin is not such a problem at all. It is however interesting to see where it ends up, will it continue to increase or level of at some point? What is the impact of extra layers (like the Lightning network) on top of the Bitcoin blockchain? Have other banks the same energy consumption per customer?

Update 20 July 2018: The ABN AMRO bank consumes 137 GWh per year (Annual report) while serving approximately 6,8 million customers ( This corresponds to an annual energy consumption per customer of 20kWh, which is considerably higher than that of the ING bank. The average energy consumption is no 10kWh per customer per year or 78 TWh per year globally.

Bank Energy Customers Energy per customer
ING 315GWh 37.4M 8.4kWh
ABN 137GWh 6.8M 20kWh
Totals 452GWh 44.2M 10kWh

results matching ""

    No results matching ""