Yahoo's experience says otherwise, and going into denial about that serves no one's interests. If we want to continue being taken seriously about telework not hurting productivity, we would be far better served by finding the explanation for why it did hurt productivity in Yahoo's case.
As a former Yahoo employee, I would say the answer is obvious: talent. Yahoo had an inconsistent talent bar: some teams had a very rigorous interview process and managed under-performers out, others did not do either.
Autonomy requires trusting your employees to do what's necessary even without direct supervision. That's not to say there weren't many strong engineers at Yahoo: indeed, I can think of one or two who have only remained at Yahoo due to ability to work from home -- so Mayer's policy will send them scuttling, while retaining those who are only at Yahoo because they can't find a comparable position (even sans the ability to work from home) elsewhere.
did it hurt productivity? Can that be quantified? Most likely not. This is 'management theatre' - a series of actions meant to reassure the BOD that the CEO is taking charge of things and making changes. Will it work? I guess we'll have to wait and see.
How is it just theater if it will change the way their teams work together? Team members that telecommuted, won't - and their productivity might change as a result/they might quit. I think it's pretty likely that it'll change things. She's making the bet that it'll improve Yahoo.
