Which feature of Bitcoin makes it censorship resistant?
Is it asymmetric cryptography? Is it block chain? Is it difficulty adjustment? Is it proof of work? Or is it the incentive to mine transactions?
Miners have two incentives:
- block subsidy, which is currently 3.125/block
- transaction fees, which is the difference between the sum of inputs and outputs in each transaction. Currently 0.3/block
Miners who want to censor a certain transaction will have to forfeit the corresponding fee while the non censoring miners would collect that.
If the censoring miner has deep pockets, it can continue making that loss. Imagine a state funded censor. Such well funded censors might outcompete the non-censors if the transaction fees are low. It would be up to the Bitcoin users to pay higher fees to keep the non-censors in business. Alternatively, the non-censors must accept the loss and keep mining.
Transaction fees won't matter if the censors hold the majority hash power. For example, if Foundry and Antpool join hands and start censoring transactions, other miners can't compete. It would be up to the mining groups to change pools in time. But how will they know if their pool is censoring transactions? I am not aware of any alerting system to detect active censorship. This centralization of mining pools is a common concern and is discussed in another chapter.
So, if you want to keep Bitcoin alive, be ready to mine at a loss. Be ready to mine in hiding. Be ready with your own source of energy. Learn to mine. Learn to keep the miners running. If you can't mine, be ready to support miners with higher fees. If governments start subsidizing the censors, they will also increase taxation to balance the subsidy. At that point, Bitcoiners would have to choose between paying taxes and transaction fees.
In conclusion, Bitcoin is censorship resistant because of transaction fees. It may not remain so if market doesn't pay high fees, if non-censors can't mine at a loss or censors hold majority hashpower.
PS: This is an attempt to rewrite a simplified version of Eric Voskuil's Cryptoeconomics one chapter at a time. Read the last chapter: #528424