I wish more "real world" managers believed in this principle.
Academics tend to want their students and post-docs to function independently. They don't have the money to hire "managers" (a.k.a. "people who boss other people around and don't contribute"), and therefore prize autonomy. It's not universal, but it's common. Even on large projects, there tend to be very few formal "managers" on a scientific team. The research organization tends to be flat, and the leaders are the people who produce.
When I first left grad school, I assumed that this sort of autonomy would be a prized commodity in the real world, but instead I've found that most businesses have a natural revulsion to independence. Politics, money and laziness conspire to produce a class of people who want mainly to exercise authority without doing much work. These folks are easily threatened, resort to micro-management to control situations that they don't understand, and generally just try to get in the way of any progress that occurs without their consent. It's a weird pathology of the commercial world, but I don't think I've ever encountered a professional manager who thought that his most effective role was "getting out of the way".
Also, Michael Lopp, aka rands of rands in repose seems to hold that philosophy.
Most new grad students drift for a while until they find their internal ass-kicker. The job of the "manager" is to help them find the thing that lets them excel, then get out of the way.
Most smart people carries an awful lot of weight in that sentence. And respect can be traded off against liking. Valve is an existence proof that this is a feasible mode of organisation but I would be surprised if it worked with a workforce that was not above average in intelligence, conscientiousness or both.