I guess I’ll play devil’s advocate here.

The NBER study you linked to was conducted on a group of Chinese call center workers. I don’t know if you’ve read the part where they describe the onsite environment they are normally working in, but it seems closer to a Taylorist factory than to the type of workplace that one could set for skilled knowledge industry people working collaboratively and autonomously to thrive. Isolation by cubicles and headsets, monotonous work, automated task assignment they have no say in, middle management constantly “patrolling” around the place to check on their performance.

You bet your home is a better context than this.

Of the 13% productivity increase generated by remote work in that particular study, 9% is attributed to working more minutes per shift (including by taking fewer breaks) and 4% to a quieter work environment.

The conclusion *I* would draw from it is that we (or they) should just have better workplaces because people aren’t cattle. Not that remote work is intrinsically better. I mean, it worked, but should we stop at one figure out of a bunch of statistics and not ask ourselves why it worked?

And more importantly, does it also work for the knowledge industry, for IT, development teams where collaboration and creativity is key? I’d like to see more studies about that.

One paradox that every WFH proponent I’ve talked to so far failed to solve for me (other than by objecting scorn and “you don’t get it” responses) is that the Agile movement and its derivatives all tend towards closer collaboration, more frequent feedback and broadening of the channels of communication. Physical and hierarchical structures are decompartmentalized. People pair program, people mob program. We brainstorm, we mind map, whole teams ask whole other teams to come diagram at the white board, we have entire walls covered with post-it notes. We meet our customers and end users more frequently, we do live demos, we invite stakeholders to come over, we do Gemba walks. Team building activities have never been so big, companies have “departments of fun”.

People in the dev community laugh remembering the now long gone cliché about nerdy antisocial programmers hiding behind their screens. But remote work is partly back to that. Remote work works the opposite way to the agile movement. You seclude again, at least on the physical level. How can you do all I mentioned when your channel of communication is reduced to voice and, if you’re lucky, your interlocutor’s face tightly enclosed in a video frame? When you don’t have your fingers to point, your hands to gesture, your eyes to exchange looks? When all the creative space you get is a little window on a screen? When there’s digital overhead to about every human interaction you need to have?

One reaction I often get is “But we hate meetings. No co-location, no more endless meetings.” So, better no information sharing at all than mediocre information sharing? Couldn’t we work on improving our human interactions and workplace habits first before calling it a settled case?

Besides, is working remotely a good solution 5 days a week? For everyone? Everywhere in the world? Every industry? Although this article goes further than most in terms of analyzing some positive aspects of the phenomenon, I have yet to see a single article or study on the subject explore these questions in detail. I am not against WFH, I do it sometimes, but I have to say the current one-sided, simplistic hype around it doesn’t help making it very credible as a holistic approach.