Glad to see this is the top comment here. If anything, I think jobs where productivity can be easily measured (think customer service agents that handle a particular volume of calls, radiologists that already have a very well defined "Relative Value Unit" system for calculating productivity, etc.) are the ones that make the most sense working from home, and these are also the types of tasks most likely to be compared in these remote work studies.

But the thing that tech companies are the most worried about is a long term slide in innovation and competitiveness, and the effect of WFH on that has definitely not been proven one way or the other.