Unravelling Bitcoin and the Crypto Puzzle
Ethan Gray / April 11 / Finance
A cryptocurrency mine in Canada - Credit: James MacDonald Getty/Bloomberg
On February 8th, 2021, Tesla purchased $1.5 billion in Bitcoin and plans to accept it as payment for vehicles. Tesla's titanic purchase of Bitcoin may have been one of the most high-profile institutional investments, and it will not be the last. Cryptocurrencies and their associated blockchains have been buzzwords in the media and classrooms for nearly a decade. That will soon change. Accompanying the headlines of Bitcoin's surging price are stories highlighting the network's substantial energy use. The network was consuming 137 TWh of electricity at the end of March 2021. That is up 3,526% since early 2017. For perspective, that amount of electricity can power the country of Ukraine. Bitcoin's algorithm prevents it from scaling efficiently, a rather important factor for a global system. Bitcoin catalyzed a revolution in decentralized networks, but should it be adopted in the coming decades?
Bitcoin is handicapped by Proof of Work, its protocol for securing transactions. In short, a network of miners validates transactions within a block that contains information. They do this by taking the hash (unique identifier) of the previous block, adding a number used once (nonce), and running the combined input through a hash function, until they find an output lower than a target set by the network. As of February 2021, it had 2^189 leading zeros. Miners painstakingly adjust the nonce one character at a time until the output they get happens to be smaller. This is called brute force computing because it is purely trial and error. In early 2021 it took roughly 2^69 changes in the nonce before miners validated a block. Energy is consumed by processing these numbers and cooling the hardware. The incentive for validating blocks is Bitcoins.
In its infancy, the network did not use much power to validate a block. The Proof of Work model, coupled with poorly designed incentives, has led to a runaway problem driven by market principles. The first miner to find a number below the target gets the reward. Naturally, miners add more computing power to increase their odds of being first and getting their compensation. Common sense leads you to believe this makes the network operate faster. That is not the case. For various technical reasons, such as keeping the chain succinct across the network, Bitcoin's algorithm dynamically adjusts the target value mentioned earlier to lock the release of blocks on a 10-minute cadence. Adding more processing capacity to the network just makes creating the next block more difficult. Typically, when a task has more resources committed to it, more gets done.
Rising energy use is entirely tied to the rising price of Bitcoin itself, not pushing through more transactions. It is easy to see how the issue compounds through the incentive structure of the network. As a result, mining has become more centralized, cutting small miners out of potential rewards. Only those with immense computational capacity have reasonable odds of solving the puzzle first. Satoshi Nakamoto, Bitcoin's creator, said, "A purely peer-to-peer version of electronic cash would allow online payments to be sent directly from one party to another without going through a financial institution." This is still technically true because miners do not operate as a bank. But as they become more consolidated, they diverge from the philosophical goal of the Bitcoins creation, a peer-to-peer approved network for settling transactions with digital currency. Currently, two mining operations already control over 50% of total mining capacity, and that trend is unlikely to change.
Bitcoin is the household name of cryptocurrencies. It’s even being advertised on the London tube.
The true impact of climate change is difficult to assess. Claims that Bitcoin will soon consume as much energy as the entire United States are embellished. But there is still cause for concern. The electricity sources in mining are not public. This makes mapping sources of electricity used in mining nearly impossible to do with certainty. Some estimates point to a carbon footprint the size of London. What we do know for certain is that we can do better. Other blockchains can accomplish the same philosophical goal with greater speed and efficiency. As it stands, the number of daily transactions executed on Bitcoin's network has remained essentially unchanged for the last four years. Fixed block processing times, a cap on block size, and mining incentives pegged entirely to the price of a cryptocurrency lead to unending run-ups in computing power and do not make for a sustainable decentralized network. Thankfully, other creators have innovated.
The second-largest cryptocurrency by market cap is Ether, and it operates on the Ethereum blockchain. It is transitioning to a protocol called Proof-of-Stake (PoS). Proof of stake shifts the validation process away from using sheer computing power and instead allows nodes in the network to vote on whether a proposed block should be approved. Refer to Ethereum's website for more detail. In brief, the voting weight of each node is pegged to their stake in the network. If you attempt to validate an invalid block, your stake is destroyed, which forces nodes to operate as good-faith actors lest they risk losing their assets. The central "beacon chain" monitors 64 chain "shards" and randomly picks validators on each shard based on their stake. Voting takes place by nodes with at least 32 Ether, $62,258 as of April 2021, staked in the chain. Transactions are validated in six minutes, as opposed to an hour on Bitcoin. This verification method is more energy-efficient, processes transactions faster, maintains lower barriers to entry for validators, and has comparable security to Bitcoin.
The third and final protocol is called Federated Byzantine Agreement (FBA), a complicated name that builds trust through simple ideas. In short, FBA systems remove the mining concept from Bitcoin and the Staking concept from Ethereum. It relies strictly on a peer-to-peer voting system to validate transactions. For example, in the Stellar blockchain, individual nodes in the network conduct rounds of voting to determine which transactions to approve. They choose other nodes in the network they wish to trust and form a slice. These slices contain nodes that overlap with each other, thus linking the network. If a quorum of overlapping slices approves a transaction, it is added to the network and cannot be refuted by any other nodes. For further reading, refer to Stanford's David Maziers' work. This consensus-building method is much more efficient, with transaction times taking roughly four seconds on Ripple and Stellar, two pioneers of Byzantine voting consensus algorithms.
Proof of Work, Proof of Stake, and Federated Byzantine Agreements are only three of dozens of different protocols, but their variations are the most commonly used. Discussing blockchain processes is a bit monotonous, but it is essential to highlight the differences. Hidden technologies that underpin modern life are complex yet seamless. That makes it easy to gloss over how variations in how they operate can have tangible impacts. A few design choices at the early stages of Bitcoin's development have ballooned to a sizable problem. Proliferating access to secure transaction systems on a network that's security is assured by its users is a lofty goal. A lofty goal that has undoubtedly been improved upon since its inception with Bitcoin.
Bitcoin has considerable scaling issues in its current state. As prices rise, transaction throughout remains static, and miners compete more aggressively, further lowering the efficiency per transaction. Decentralized networks have sizable advantages over current systems, and they are here to stay. Perhaps Bitcoin will find a way to decouple rising prices with energy consumption. So far, that has not happened yet. This is why the growth in Bitcoin prices and bets by large institutions on a network that consumes entire countries' worth of electricity seems somewhat foolish when considering very robust alternatives exist. If investors truly believe in public blockchains reshaping how we transact with each other. Why is one of the least scalable solutions still leading the charge beyond name recognition and longer run proof of concept?
The decentralized nature of blockchains allows them to enable transactions across the globe between anyone connected to the internet. The potential use cases for blockchain networks touch nearly every element of modern commerce. The beauty in its innovation aligns with what most benign technologies do. Proliferate opportunity to those previously isolated. When institutions choose protocols to build their products upon, they should look carefully at their choice's potential impact. Bitcoin mining is not going to destroy the planet singlehandedly. But why double down on a network built on an algorithm that cannot efficiently solve the problem it set out to? The most significant restructuring of transaction technology since the credit card will not happen overnight. But it will have lasting implications in far-flung places, and the protocols selected to build these new networks will have consequences. Hopefully, they are chosen with some care.