I've never really had an office job, and I am having a very hard time believing this. Can I get a corroborating account or two?
I think that 20% time has a lot to do with this. I dunno about other folks, but I view 20% time as "All those other things I could be doing with access to Google's data that I try to squeeze in when I don't have an official project to do." Given the near infinity of nifty things to do with the web and a few zillion computers at your fingertips, that's quite a lot.
Another big factor seems to be whether companies assign tasks to individuals or to products. The number of tasks you're assigned is limited by your manager's bandwidth: if you're a fairly talented worker at the bottom of the totem poll, it's very likely that you'll be able to complete tasks faster than your manager can dream them up. But if tasks are just assigned to the product - "Well, we could be better in X, Y, and Z ways. Oh, and anything else you can think up" - there are a potentially infinite number of ways that the product could be better. If you just pick them off one by one as soon as you finish your previous task, without worrying about who owns them, you end up perpetually busy.
The disgusting part is this is accepted as normal behavior. There's rarely punishment because these people earn their employer serious money. Contractors tend to be paid 1/2 - 1/4 of the money they actually bring in. For gov't contractors, there go your tax dollars.
People complain about government workers being lazy. No, they are just as lazy as most regular Americans[1].
[1] I have no idea if this applies to other countries, but human nature tells me it probably is.
Most people think they do much more.
At several large companies I worked at, it was absolutely impossible to do anything more that. Their structure made crushed any attempt at it with meetings. There were weeks where I had over 30 hours of meetings scheduled. Meetings where everyone felt productive but nothing really was accomplished.
Yet they were the social bugs, and in this part of the organization, that apparently was what succeeded. If where you are doesn't value your kind of input, then however good it is, it is others who will "succeed".
And they did "finish" things. Just in such a state, that others had to clean up the shortfall. But that wasn't really measured. I find most corporate "metrics" I've encountered akin to all the jokes you hear about statistics (lies, damned lies, and...).
Well, I've said before that I need to curb my "bitter old man" comments. But you asked for collaboration, so: Yes, this really does exist.
To remedy this, I try never to give them tasks, but instead to give them expectations, targets, and responsibilities. Then my job becomes one of checking in on them and assisting from time to time if they appear to be struggling. They know they'll be held accountable for how things turn out, so they tend to work hard.
That said, getting away from task-driven management is a constant struggle.
I used to work 3-month contracts, broken up by 9 month roadtrips, so coming back to the office I'd be stoked to get back to writing code, with months of pent-up energy ready to burst out in the form of productivity.
On two occasions, I took over projects from another developer who had been working on them the entire time I'd been away, and rebuilt them from scratch in less than a week. While it's certainly possible that I'm 100x-500x as productive as those guys, more likely is that they'd simply been coasting along that whole time.
It took a lot of diplomacy not to rock that particular boat. "Uh... yeah... it must be this new technology that's just so darn efficient. Certainly not any fault of the team or anything... Say, did you catch the game last night?"
IMHO, if you get the job done in 4 hours, someone else does it in 3 hours, someone else in 10 hours, and someone else in 13 hours, there is nothing wrong with being that 4 hour guy.