2+2=5
'Nuff Said.

On the other hand:

- There are often simple, almost foolproof tasks that can be done by new people; to keep with the raiding team analogy, you could ask the new members to farm gold/mats for the flasks the raid team is going to use, while he learns raiding on the guild's alt run, or even by running LFR with a guildie for a coach. There's a limited number of those tasks, though, so the number of new team members that can be integrated at a time without compromising efficiency or quality is often small.

- On the long term, the new team members should become productive. Of course the amount of people that can be thrown at a project is really dependent on how good the management is and how modular the whole project can be made, plus the overhead costs usually grow when teams grow, so there is a limit when it ceases to be worthwhile; I don't think Blizzard is even near that limit.

Of course, Blizzard knows this; they are often hiring. I don't think it's easy finding good enough local talent in sufficient numbers, though; the number of openings they have is fairly large:
http://us.blizzard.com/en-us/company/careers/directory.html

Surely the number of mats farmers (which in this analogy, I guess would be interns getting coffee and making copies) isn't the limiting factor in releasing content though.

Nine people can't make a baby in one month