Great post, the only valid counter argument in the entire thread.
Maybe the pay will be enough to put aside any conflicts of interest? Or if you were at the meeting where the management decides not to fix lifechanging or "so easy but still not approved for reasons" will it make you personally feel better? Then what you say to the community? "I know but management sucks, we're actually aspiring to be activision internally?".
Showing up to work every day counts for a lot, but you make a really good point.
They don't necessarily have to be actively hostile towards management to get a point across and to appear to be on the side of the community, simply giving factual statements through gritted teeth could be used to achieve the exact same effect without the drama. Don't say "I know, but management sucks", just say "we are aware of the issue, but management has decreed that this is issue is commercially non-viable to resolve until further notice". Simple, factual and without any unnecessary vitriol or accusations of being a mouthpiece - just a quick highlighting of the situation and an explanation as to why nothing is being done.
Similar things can be used to give the illusion that a community manager is on our side for different questions. No news on the Panther Clipper? Rather than saying "no news at the moment", they could give a much more human response such as "no mention of this in the dev offices recently, so no news there" or "would love to discuss that, but that whole issue is NDA'd up to the eyeballs at the moment".
And that's the whole point, they should give the illusion that they are with us. That they are just one step above the moderators here, that they are shepherds and leaders for the community, as a good manager should be. They don't need to actually be on our side, just enough to pretend to be when we are in earshot. It's not good for the flock if the shepherd goes around plainly in wolf's clothing, they should at least give the illusion that the aren't just preparing to send us off to the abattoir.
Also, within marketing it helps a lot to know a lot about the product you are allegedly marketing. If I'm speaking to a salesman about a prospective product, I'd like for them to know beyond just the very basics and up to what it is reasonable for a user to encounter during normal use; if half of my questions of "what does that feature do" and "how do I use this functionality to good effect" are met with referrals to other people or stock phrases like "I'll get back to you on that" then I'm quite likely to simply ask to speak to someone who actually understands the product rather than a clueless marketer. Granted, that's more of the role of their upcoming influencer manager though. That's probably also why I tend to ignore regular marketing folk and instead focus on reading reviews, specifications and datasheets to make my own mind up.
Similarly, part of marketing is to research for market data and get feedback to allow for further improvements and developments to the product - how can you expect to keep on delivering better products if the business doesn't actually know what is good? CMs, alongside whatever analytics division FD have (assuming they even have an analytics division who go datamining through everything to identify issues), are their primary source of feedback on the reception and effectiveness of new features as they are FD's main eyes and ears within the community; it's very hard to collate and make sense of community feedback if all you are seeing are unfamiliar sci-fi buzzwords mashed together (in particular if the community when the community uses acronyms or slang when discussing issues, someone not well acquainted with the community might now understand terms like the "Federal Bricks" or "Thargoid Thursdays"), but it's much easier if you actually know what all the terms mean within the context of the game.
More generally on the topic, part of the problem that I see here is how locked down the whole thing is and how distant we actually are from the real work being done. From what I have seen in other game communities, nothing makes a particular individual developer in a dev studio more popular with the users than simply communicating with them. People get angry with what they do not understand, they are given a black box and if it doesn't work the way they want it to and they don't know why it was made to work a particular way then it just frustrates them. Granted, some people will never bother trying to understand things, but its a lot easier to remain calm and discuss things if have sufficient information to actually reason though design decisions. Even developers who have made controversial design decisions often get a Rockstar's welcome on the forums once they have established themselves as bastions of logic and reason in the pursuit of a perfect game.
For example, many people would say that the Mamba is bad design. It doesn't add any particular niche that the FDL doesn't, it's weaker overall than the FDL despite a similar price point, it doesn't even seem to add any lore-based reason such as being a competitor to the FDL as it is by the same manufacturer and it's similar enough in its characteristics that it barely even adds variety from the FDL. Personally, I can't see a single good reason for the Mamba to have been made. However, that's not to say that there is a solid design process behind why the Mamba was implemented the way it was, just that it isn't clear in-game and hasn't been communicated to us. Now, if the Mambas introductory livestream had a guest star from one of the developers who worked on the Mamba who could explain where it fits in the manufacturer's product stack and how it enriches the game by virtue of its very existence, then I would probably have a quite different opinion on the Mamba. The livestream showed what the ship is and does, but not why it even exists in the first place.