We seem to misunderstand each other, I am not talking about a miner making changes to their own mining code. There are, to my out-of-date knowledge, two fairly small groups with a lot of power within bitoin:
* The maintainers of the bitcoin core code
* The miners proving a large portion of the hashing power
These two groups have a large amount of centralised power. Last time I was involved with bitcoin the maintainers were essentially five people for example.
The network majorly runs the core client in various forms [1] by far the largest portion even runs the newest, or at least last two major releases. If those two groups decide to change the monetary policy via the source code of bitcoin core your options are basically to take your losses and fork the official chain or go with it.
But as I mentioned, I am out-of-date on my bitcoin knowledge, so these things might have changed.
> If somebody were to propose adjusting the monetary policy, they need overwhelming consensus of all bitcoin users.
I don't believe that is true, you just need consensus by hashing power and maybe harassing/ddosing power [2], otherwise you risk them sabotaging and damaging your new network too much to recover.
edit: I just read your other response to better understand your point instead of just trying to make mine clearer. It seems you are of the opinion that the consent of most economic participants, especially end users, is required to make that change. Which, obviously, is in contrast to my hypothesis that most people will follow whatever the current core version of bitcoin does as long as the primary hashing power of the system agrees.
I wonder if that's a dispute that could be settled with data/arguments at all, or if that's just something we have to disagree on.
[1] https://bitnodes.io/nodes/
[2] https://www.reddit.com/r/Bitcoin/comments/3jj2hf/bitcoin_xt_...