> The chain's security is primarily paid by the block subsidy

> The block subsidy trends to zero.

These statements are misleading. As the block subsidy trends to zero, the chain's security will be primarily paid by transaction fees. This will likely happen within two decades, in which the block subsidy will diminish by a factor 2^5 = 32.

> Either the fees need to increase or the security property falls away

The security will diminish, but not fall away completely as long as bitcoin maintains a constant backlog of high-fee paying transactions.

Totally agree, the key insight is that Bitcoin in the limit e.g. 50 years from now will need regular high txn fees to stay secure, we've only seen high txn fees periods twice in 12 years.

In the limit Bitcoin either is a chain that you can use for under 60 bucks a txn (two of which are needed for each lighting network channel) and it has meh security or it costs greater than 60 bucks a txn and is like an extra expensive wire txn system and used as financial infra.

As long as the problem is somewhere in the future it is in the present because it's all about speculation & speculation depends on future value.

There are two other things to think of in this thought experiment: other cryptocurrency, they can set higher subsidy, use different L2 solutions

And dark miners. Even if bitcoin's security parameter is stable is it goes down then miners will turn off and try to sell their hardware at a discount. If that happens >50% then there's a persistent threat that a single actor bought it all and is waiting to turn it on.