Regarding the blockchain technology, observers are endlessly debating many points which are, for the most part, totally false or at least only a very small side issue. And if there is one subject that has been debated for many months, it is for sure the energy-consuming aspect of blockchain. Thus, many experts argue about the misdeeds of blockchain on the environment, assuming that it is a real energy calamity for our planet. But then, who is really right in this battle between pro and anti blockchain on its energy consumption? Is there only one truth?
First and foremost, some facts
Speaking about “blockchain” in general is a misnomer that everyone uses, but there are dozens of different technologies, each with its own characteristics and features. All blockchains do not work the same way. The hard point, which experts are tackling, concerns one of the many features of a blockchain: its consensus algorithm. The algorithm that most raises an issue of energy consumption is called “proof-of-work”. The two best-known blockchains that use this algorithm are Bitcoin and Ethereum, respectively world number one and number two in terms of capitalization.
It is necessary to reach a consensus on a blockchain from the participants, because data and transactions are validated by all the stakeholders of the network, called miners. Their role is essential for the functioning of a technology that is based on this network: they are the guardians of the temple. They guarantee that all transactions made on the blockchain are valid, in the sense that there is no attempt at fraud. This validation operation is performed in a completely distributed way, all the miners realize it at the same time. However only one of the miners will be drawn to give his analysis of the transactions he had validated. When he gains this right to express himself for the whole network, he is paid back by the creation of a certain number of cryptocurrency units of the blockchain.
The consensus mechanism is therefore completely random to ensure that it is never the same miner that provides the result of its analysis, and thus to prevent the easy introduction of fraudulent transactions. To gain this right, the miner must solve a complex cryptographic equation that requires a lot of computing power to be solved, which will allow him to prove the work done to get there. But this computing power has a hardware cost, only highly specialized machines can carry out this calculation, and they require each miner to consume electricity. So, the more miners on the network and the more energy consumption there is, but this has nothing to do with the number of transactions made.
If we take Bitcoin, which is the most criticized blockchain for its energy consumption, it uses:
- the equivalent of the annual electricity consumption of Ireland
- the equivalent of the electrical production of 4 nuclear power plants
- more electricity than 159 countries in the world
- 30.25 TWh of electricity (figures in 2017)
Let’s compare what is comparable
First of all, numbers significantly differ from one study to another. In addition, the few recent studies are widely discussed and questioned on the assumptions used to achieve them. For the moment, no serious scientific study has been done to accurately assess the power consumption required to secure Bitcoin.
Many people give an opinion on the level of consumption of Bitcoin, because data are available and at least computable, so it is possible to imagine what its energy cost is.
In this case, it makes a lot of sense to ask what the energy consumptions of other activities are. Did you know, for example, that a single search query on Google search engine is equivalent to the energy consumption of a light bulb for an hour, so the data centers of this web giant alone are spinning 14 power plants?
Even better (or worse), according to a 2013 study, the web as we know it consumed at the time the equivalent of the electrical production of 30 nuclear power plants and data centers consumed nearly 3% of the world’s energy! These figures have obviously progressed since 2013.
Now let’s take the comparison with an activity like gold production! We often compare Bitcoin to a digital gold, we can agree that this analogy is debatable. However, as part of the exercise, we will make the comparison with the energy consumption required for global gold mining. In 2012, 7,400 kilograms of gold were produced per day, making a total production of 2,700 tons of gold over the year. A 2014 study estimates that producing one kilo of gold requires 310,000 MJ / kg. If we convert this need into energy in KWh, for a kilogram of gold, we need about 86,112 KWh. We now have a good basis for calculating the energy cost of world gold production over a year, so it is just under 233 TWh, almost 8 times more than the annual consumption of Bitcoin!
The last comparison we can consider is the total annual energy consumption of the banking system. Here again the comparison is delicate since the banks do not only manage the monetary transactions. An evaluation of this consumption was carried out at the end of 2017, landing on a result of 100 TWh per year with, in this calculation, 60% of the energy consumed by agencies. Agencies that have been deserted by customers for many years and that the banks have decided to massively close in the years to come. In France, between 2012 and 2016, around 3% of branches closed and French banks are currently planning to close 12.6% of outlets by 2020.
Problems that push for innovative solutions
Even with the best will in the world, it is difficult to fully defend the level of energy needed to run Bitcoin, regardless of the studies or opinions expressed on the subject. At the time of preparing COP24, it is urgent to find a way to reduce the ecological footprint of blockchains on the planet, not just Bitcoin.
The advantage in this kind of situation is that the actors of the blockchain ecosystem are the first concerned by this issue and therefore, the first to find innovative solutions to reduce its electricity consumption. Take, for example, the machines needed to carry out the mining; the latest generation are 2.5 times more efficient in energy consumption than the previous generation. Miners themselves seek to optimize this energy consumption, particularly by moving to countries where the ambient temperature makes it possible to avoid having to cool the machines with installations that are themselves expensive in energy. Promising solutions based on liquid cooling are also being studied by industry professionals.
A new feature that could reduce the need for energy to secure Bitcoin has also been available for a few weeks now: the Lightning Network. Even if it is still marginal in the volume of transactions carried out, this feature allows two Bitcoin users to perform transactions in cryptocurrency without having to go through the main blockchain. Lightning Network makes it possible to conduct peer-to-peer transactions in a secure and transparent manner while notifying the main blockchain only with the final result of the transactions.
Moreover, let’s not forget that Bitcoin is not the only blockchain alternative, there are dozens of other blockchain technologies using other algorithms to achieve consensus or on whose development teams are actively seeking to develop other forms of less energy consuming consensus. Thus, Ethereum which is currently based on proof-of-work like Bitcoin, is finalizing the development of a major update to introduce the proof-of-stake in the coming months. This new consensus algorithm will no longer need to involve computing power but rather cryptocurrency directly. The draw, in this case, will be based on the amount of cryptocurrency in the miner’s possession.
Let’s not talk about blockchain in general but always specify the one we are referring to and when it comes to energy consumption, it’s mainly the Bitcoin blockchain or the Ethereum one. Without excusing the fact that Bitcoin consensus algorithm is clearly energy-intensive, it turns out that other equally comparable industries also have significant power consumption, some of which far exceed that of Bitcoin. Many tracks are currently being activated to solve this problem of energy consumption, from the Lightning Network to the proof-of-stake or even the optimization of the energy performance of miners. Finally, it is quite easy to harm the development of a technology like blockchain, and more precisely of Bitcoin itself, by highlighting the weak points while obscuring the real benefits it brings in different business sectors. Blockchain carries with it the genes of disruption, the extent of the possibilities offered by this technology are unimaginable and it would be a pity to reduce it only to the quantity of energy necessary to make it work.