That's the hard part about things like this. It's almost completely subjective, so doing a study would be difficult. We're left with dueling anecdotes from credible people.
Different fields vary a lot also, so it's probably not that useful to try to talk about "academia" as a whole. Not only do we have dueling anecdotes, but they aren't even anecdotes on the same subject. =] Imo, the pros and cons of being a physics versus a philosophy versus a CS academic just aren't that similar.
I can believe that the 'hard' sciences are roughly like he says, at least at many places. It's common for there to be a sort of "lab" mentality, with a lot of grad-student cogs in a famous-professor-lab machine, and credit tends to go to the head of the lab (especially if the paper has 50 authors or something, as is common in some areas). Partly that's because it takes a lot of money to set up a physics/chem/bio lab, and there is a lot of grunt-work to be done.
That's less common in CS, I think. Not inexistent, but you can find a research group that isn't like that. It's even less common in the humanities, but then you have a whole different set of problems (less money, fewer jobs).
