While it’s true that mining for cryptocurrency consumes significant amounts of power, in many cases the media makes highly inaccurate estimates regarding power consumption. In addition, journalists often lack an understanding of how blockchains or even just electricity works.
“Ethereum could be using more than a country’s worth of electricity” reads a title from futurism.com. Below is a quote from the article:
“the entire Ethereum network could be consuming as much as 4.2 Terawatt-hours (tWh) — which is only a little bit more than what’s consumed by the Middle Eastern island of Cyprus.”
There are a number of problems with this sentence. First of all, the 4.2 Terawatt-hour figure is meaningless without any time frame as to how often that much energy is consumed. A watt is a unit of power, whereas watt hours are units of energy.
I decided to take a look at their source that they linked for that statistic, only to find that the link was to an article on Bitcoin’s power consumption. I think he meant to link this article instead. It appears they meant 4.2 Terawatt-hours per year. Another quote from the Futurism article:
“All those household computers turned into Ether miners each have blockchain transactions consuming at least, if not more than, 45 kWh of electricity.”
The statistic the author was trying to describe was the average energy consumption per transaction on the Ethereum blockchain. Basically, it takes the estimated daily energy consumption from Ethereum miners, and divides it by the number of transactions that occur per day.
This is a completely misleading way of describing the efficiency of a proof-of-work based blockchain. When people say that miners validate transactions, that’s actually a huge oversimplification and misrepresentation of what actually goes on.
When miners perform calculations to try and find blocks to earn coins, those calculations aren’t actually validating transactions. The calculations that are made by miners, which are known as hashing operations, are much like guess and check math equations, in which there is a very tiny chance of succeeding, and being the first one to find a solution entitles you to create the next block and claim the prize. Miners compete, performing millions of operations per second, in order to try and find the next block.
When the miner creates a block, they can include transactions in the block, and can claim the fees. This is where mining and transactions are connected. The sole purpose of these calculations, known as “proof of work”, is to decentralize the network, by providing a way of establishing consensus on who has the right to publish the next block. As long as no entity has greater than 50% of the hashing power, there is no need to trust a single entity such as a bank.
Without a blockchain, there would be no consensus or proof on the network that coins are owned and not sent, making the currency worthless.
The reason this is important is because the power consumption is unrelated to how many transactions there are. If Ethereum’s transactions doubled overnight, but the price remained the same, the power consumption would remain the same, and the energy consumption per transaction would decrease, as there’s more transactions on the same amount of power.
The reason power consumption is so high is because block rewards create an incentive to mine. The good news is that Bitcoin’s mining reward gets cut in half 4 years, meaning the incentive to burn electricity to mine will decrease over time.
The amount of power consumed will likely follow market forces. The incentives that encourage supply of miners is the newly minted coins via block rewards, plus any fees from transactions. With too little of an incentive to mine, a proof-of-work blockchain is vulnerable to 51% attacks due to low costs to bruteforce it. With too high of an incentive to mine, a tremendous amount of power is used to compete for new blocks.
So for example, approximately 1800 Bitcoins are created per day from new block creation as of today. At a price of $2,636, that means $4,744,800 worth of Bitcoin is created daily, meaning total investment in mining will likely remain near that amount(this includes hardware costs, not just power). But in 2021 when block rewards are cut to 6.25 BTC, if prices are the same, the incentive to mine is then cut in half, meaning older miners will be taken offline, and mining hardware production/purchases will likely decrease.
The key takeaway is that as usage of cryptocurrency usage grows, and block rewards shrink, energy consumption per transaction will decrease. Right now, cryptocurrency is in it’s experimental stage, and most usage is for speculative means and development. That does not mean the power consumption is currently useless. With each block created, Bitcoin and Ethereum’s blockchains continue to grow larger, making them stronger and more difficult to attack.
There is a compelling argument that proof of work wastes too much power, and luckily, there are alternatives, such as proof of stake and proof of storage, which consume very little. However, it is incredibly misleading to claim that it takes 45 kilowatt hours to verify a single transaction. If cryptocurrencies solve scaling problems that result from lack of high-speed internet infrastructure, and lack of affordable storage for storing large chains, the transaction capacity of Bitcoin and other digital currencies could increase almost infinitely while having very little impact on total power consumption.