We have no idea how much time or effort it would take to do such a thing. I'm sure they want to aim all of their time, resources and personal towards polishing the game and fixing the various bugs. They delayed it for a reason. I highly doubt they'd delay it, only to divert precious man hours and resources to a demo.
Frankly, I don't need a demo of any kind. Let the team spend their resources in a way that they deem necessary to produce the best game possible.
How do you know it's not?
The decision to make a demo is one of the most disruptive events that can happen in the timeline of the development of a game. They will derail development for the majority of the team due to the need to treat it like a mini-release for a full game. Content and Code each have to be locked down, forcing developers who are otherwise madly working overtime to come to a sudden and screeching halt. Ideally codebases need to be branched (and if they're not, that's even more disruptive). They need to be given a rigorous QA pass (and then sent to first parties for their ceritification pass). All the while, development on the full game continues, and due to the timeline of releasing a demo before the main game (I assume this is what you are asking for, rather than the release of a demo after the game release), all demo issues take priority over issues with the main game, further disrupting main game development.
And contrary to what an earlier poster implies, it takes significantly more than 5 people.