There's a difference between pools not accepting txs in their mempool, and pools refusing to build on other miners' blocks that contain them. Pools are beginning to do the former (which doesn't cost them), not the latter (which would cost them dearly). Only if a majority of hashpower does the latter will censoring be effective (and then no longer cost them again).

A hashpower majority is not needed. For the latter, if a single digit percentage of the hashpower decides to enforce the ban, other miners in the network have an economic incentive to join the ban as well.

The reasoning is explained here https://juraj.bednar.io/en/blog-en/2020/11/12/how-could-regu...

In short, if a miner chooses to include banned transactions in its block, it means that block cannot be built off of by fully regulatory compliant miners. Fully compliant miners will be still building off the previous block in the chain. If fully compliant miners win the hash race and mine the next 2 blocks, the original miner's block will not be part of the longest chain, and they will lose all their block reward.

Who gets to determine which addresses are "poisoned"? China? The US? I have to imagine the ideologues in the bitcoin community will fight back against that. Also, not being censorable is a big part of bitcoin's value prop so this type of regulation would likely crash the price as well.

If different countries had different ban lists then it would still censor the BTC that are included in enough lists

As BTC continues to grow, the 'bitcoin community' changes in character; it's already fallen far from the cypherpunk anti-censorship / privacy roots

Of course they can't start orphaning blocks yet, but they can once it is more popular. It wouldnt take much government pressure, obviously governments will want to stop people from processing illegal transactions, and miners won't care

The price of bitcoin will plunge to 0 if that happens, which is the reason why it won't.

The new breed of users doesnt care at all about such things. Usually the price goes up as it becomes less cypherpunk and more of a regulated surveillance instrument - speculators will argue that it increases legitimacy and encourages institutional investment. It will still be a scarce asset that hedges against inflation

You're missing that for this to work, miners, those who invest into processing bitcoin transactions, would be willingly lowering the value of what they do. Other comments bring issues they'd already face in cost trying to adopt banning more transactions. But if they accepted those costs, bitcoin's reputation of stability and trust would be broken, if 1 BTC = 0.75BTC because .25 was from tainted transactions, nobody wants to risk that happening to them so they'd abandon quickly and the price would plummet. But if the price would plummet then miners would be inclined to do the opposite to avoid that happening.

but it is already happening:

https://decrypt.co/48025/bitcoin-mining-pool-that-censors-tr...

Big pools need to buy cutting-edge equipment and use huge resources, they have to abide by regulators.

Most users would be unaffected because they just "hodl" and only care about the price going up + encouraging institutional adoption. Bitcoiners don't really care about making transactions.

The lack of privacy built into the protocol makes this inevitable. You're asking miners to willingly help move illegal funds while regulators watch

Only if you can keep up with the new addresses the coins are sent to, and if they decide to send some of the coins to random legitimate accounts en masse, now pools will be blocking countless legitimate accounts that have no connection to the censored address. It would require some kind of central agency that can track and immediately update a list of censored addresses.

Monero is mostly traceable still.

Source: https://www.sciendo.com/article/10.1515/popets-2018-0025

That article has been debunked for years, it is very old

https://www.getmonero.org/2018/03/29/response-to-an-empirica...

It's not, that analysis is from 2017 and the issues described have been mitigated. The Monero team updated mixin sampling algorithm to better reflect real world usage. They increased the ring size from a minimum of 4 in the paper to 11. The team is working on increasing the ring size from 11 to 64, and possibly 128 (provided there is enough block chain space).