Are Cryptocurrencies Destroying the Environment?

It is no surprise that it takes a significant amount of computing power to keep track of the blockchains which hold the data about the various digital currencies which exist. By some estimates, the largest cryptocurrency network, Bitcoin, consumes as much electricity as the entire country of Denmark!

Why are cryptocurrencies so resource intensive? That is because most of them rely on a “proof-of-work” algorithm in order to reach consensus on the state of the blockchain. Remember that blockchains are just ledgers which trace the state of a digital economy from some genesis point to the present day. The current state of digital economy is uncertain since many transactions have occurred since the last time everyone in the network agreed on the state of the economy. In order to eliminate that uncertainty, each “miner” on the network seeks to add a block of recent transactions to the blockchain. Theoretically, this is a relatively easy thing to do- you just look at the list of unconfirmed transactions that have been broadcast to the network and organize them into a block. The computational complexity of this procedure is increased because blockchains have very specific rules which govern when a block is allowed to be appended to the end of the blockchain.

In a proof of work cryptocurrency, the rules which govern when a new block may be added to the end of the blockchain depend on the output of what is known as a hashing function. A hashing function is like a fingerprint reader for data- you know that if you put the same piece of data into the hashing function, you will always get the same output: the hash. However, if you change the data in any way, the hash will be different. A popular hashing algorithm is SHA256, which can take any arbitrary piece of data as an input and return a 256 bit hash for that piece of data as an output.

Hashing functions used in cryptocurrencies have several important features. First, they are collision-free. This means that it is infeasible for someone to find two inputs to the hashing function which produce the same hash output. We say infeasible instead of impossible because there are always going to be collisions in a fixed length hashing function, since the function must be able to take in data strings longer than the hash output length. This means that there exist data pairs which produce the same fixed length hash. However, as long as there is no way for someone to easily find such pairs, hashing functions are said to be collision-free.

Second, they must be one-way. This means that given a particular hash output, it should be very difficult to determine the input to the hash function which produced that output. Third, it should be deterministic, meaning that the same input should always produce the same hash output. Fourth, small changes in the input should result in large changes in the hash output.

Fifth, and most important for proof of work cryptocurrency applications, hashing functions should be easily “puzzlefied”. This means that it should be easy to set a condition for the hash output which allows you to vary the length of time it should take a computer to find a hash output which meets your criteria. This property is what the blockchain uses in order to decide which “miner” gets to add a block of transactions to the end of the blockchain.

Rather than just accepting the first block of transactions it receives, a proof of work cryptocurrency waits until it receives a block of transactions which meet a particular hashing difficulty level. This is because each block of transactions in a proof of work cryptocurrency blockchain must include a “nonce”- a Number Only used oNCE. This number meets the special requirement that if the block of transactions plus the nonce are run through a hashing function, the resulting hash output satisfies the rules of the “puzzle”. In Bitcoin, the difficulty level of mining a block is governed by how many leading 0s it must have. When demand for Bitcoin is low, then the difficulty level decreases, and miners are more quickly able to claim rewards from the blockchain. When demand for Bitcoin is high, the difficulty level of the puzzle increases, and it takes miners longer to claim block rewards. In this way, the Bitcoin ecosystem is self-regulating- the rate at which new transactions are added to the end of the chain fluctuates based on the volume of transactions.

This is why people who started mining Bitcoin early were able to earn higher rewards than people are able to today. Since the volume of transactions was low, it was very easy to add transactions to the end of the blockchain because the difficulty level of the cryptographic puzzle was low. As Bitcoin became more popular, more and more transactions were going through the network and the difficulty of the puzzle went up. Now people have to spend much more computing power and have a lower chance of getting to claim mining rewards since there are so many more miners on the network.

All of these cryptographic calculations take energy in order to happen, and they also produce waste heat which needs to be managed. Large cryptocurrency mining data centers spend huge amounts of money on electricity and cooling in order to the machines up and running. And since most electricity is produced through the burning of fossil fuels rather than through the use of renewable energy sources like solar or wind, all of this cryptocurrency activity also leads to more carbon emissions.

Clearly, if cryptocurrencies are going to be adopted worldwide, then the problem of the energy consumption of the network will need to be solved. Whether or not the environment gets destroyed before that happens is anyone’s guess.