Game Theory of Bitcoin
Source 1: https://blockgeeks.com/guides/cryptocurrency-game-theory/
Source 2: https://dl.acm.org/doi/pdf/10.1145/3212998
With Bitcoin being mentioned on the latest problem set, I thought it would be interesting to delve further into it for this blog post. It will be about Bitcoin and the game theory topics it encompasses, specifically with the mining process. One of the main breakthroughs that Bitcoin brought forth in 2008 was the creation of what is known as the Nakamoto Consensus, a solution to the Byzantine Generals Problem. This problem revolves around the question of how to reach provable consensus between nodes when they cannot trust each other. Before Bitcoin, the best way for reaching consensus of this type would require every node to give their input in order to reach consensus, which would clearly not scale to the size that something like Bitcoin would require it to, among other problems.
For Bitcoin, the main driver of it reaching consensus is its Proof of Work (POW) protocol, in which miners consume electricity and computing power to compete to mine blocks every 10 minutes for the block reward (currently 6.25 BTC). At first glance, it may seem like miners hold too much power over the network, since after all, they are the ones responsible for adding new blocks to the blockchain. However, the miners are incentivized to not cheat or otherwise harm the network. If trust were to be lost in the Bitcoin network, all the Bitcoin held by miners would likely become worthless, and they would also lose their investments in mining equipment and infrastructure (Bitcoin is mined primarily using ASICs – these ASICs are specifically made so that they are optimized for Bitcoin and will not work for anything else). Another incentive is the process of how the Bitcoin network is structured, which creates a self-enforcing Nash Equilibrium. It is very technologically consuming for miners to create blocks, while on the other hand, it is easy for any node to verify the correctness of any block. This means there is a disincentive for miners to waste resources to complete a malicious action such as “double spending,” since it can easily be caught. Overall, this self-enforcing Nash Equilibrium should work out well, and it has in the 11 years Bitcoin’s existed, however these incentives are possibly skewed if miners do not act independently.
Although the Bitcoin network has never experienced colluding miners on any large scale, it is theoretically possible for miners to “selfish-mine” and pool together until they reach 51% of the hash power and thus have control of the network (known as a 51% attack). Given that Bitcoin is the largest decentralized network in the world, backed by a very large amount of computing power, a 51% attack is unlikely. It is not out of the realm of possibility, though, as there have been 51% attacks already on less secure blockchains such as Ethereum Classic, Bitcoin Cash, et cetera.