This is so matter-of-fact I consider that the more important question is - why don't managers recognize this and how can we make them recognize it? If I was a steel worker, they'd go insane over a 1% increase in my productivity, but as a developer they chop up my time into tiny little increments, and then wonder why the heck I can't get anything done.
I'm not sure, but probably because productivity of programmers is hard to measure? You, as a programmer, solve different problem each time. Productivity can be defined by
Within these parameters, managers can only see time-to-implement clearly, amount-of-the-task vaguely, individual-productivity-coefficient relatively (to other coworkers), and difficulty-of-the-task hardly. Suppose you spend a week to solve X in busy environment, then spend a day to solve Y when the noisy manager is on vacation and you know X and Y are more or less the same amount and difficulty, so it's 5x boost! But the only clue the manager has is that you finish Y in 1/5 of the time of X, and each counts as single "feature" so the amount of task should be the same, and they conclude that the difficulty of Y is 1/5 of X instead of thinking that your individual productivity coefficient raised 5x.
Sure, it's hard to measure (the root of the problem) - but there's plenty of people out there (Joel Spolsky for starters, Peopleware is another) that suggest some of the things that adversely affect it. Why are those factors so hard to communicate to managers?
The existence of those factors is not hard to communicate. Your manager will probably understand and sympathize, but without a measurable cost or benefit, how is she going to argue for them in a committee with other managers? Sure, we get that your private offices are going to make your team more productive, but is that $100k more productive or $1M more productive? If you can even begin to answer that question without the rest of the crowd falling asleep, you're amazing. There is always pressure to make simple, easily justifiable decisions, rather than nuanced, complex ones; unless you can find simple terms to argue in, you just brought a paintbrush to a gun fight.
To quote my boss (it's almost like a mantra) - "whatever's easier." I'm sort of a student of management (and in academic terms actually more than "sort of") so this stuff fascinates me. It's easy to point to companies that don't do whatever's easier, but actually do whatever's correct (Google would be a good example) but day to day, the message doesn't get executed.
The reason that managers do what they do to developers is because a manager does not have to think. Because of that, he has no understanding of what kind of environment supports or destroys one's ability to think. He sees his job is to make snap decisions based upon no information and, as a consequence, thinks that your job too.
