In the sci-fi Matrix reputation world, I can absolutely see somebody curating a reputation list called #bad-people:example.com which they prime by finding every user ID in every room they don't like and blanket assigning them -1000 reputation.
If then a moderator was dumb enough to subscribe to #bad-people:example.com and use it to impose a ban list on their rooms, then I'd hope that their community would arch an eyebrow at the crassness and treat them like a rogue moderator and either get them removed, or fork and go elsewhere... assuming that it's possible to visualise the filters which have been put in place.
There's a huge responsibility on the tool author to ensure that the users can see what filters are in place, and what they do, and encourage the user to challenge them - but again, hopefully, the market will vote with its feet and users will adopt the best tools available, and avoid being trapped under primitive moderation systems like the ones you refer to here.
tl;dr: we need better, morally relative reputation systems - rather than zero reputation system. liberal plurality >> anarchy ;)