Yes, as is the subsequent remark: the only way to know is to collect and analyze data. I don't think Musk has done so, I don't think most management who gets quoted in newspapers does so, and I think anyone who attempts to declare authoritatively that working remotely is good or bad needs to consider backing up their assertion with data rather than declarations on the philosophy of labor.
This is essentially the conversation some internal auditor/QA types are having with a lot of the 'business leader/middle manager' types in various places.
Leaving aside that The E-loon in all probability isn't doing this to increase productivity (in all likelihood this is a combination of his 'American UberBusinessTyrant' theater and wanting to avoid layoffs by making people quit), there have been several studies (some high quality
Owl Labs consolidated report) that show an amazing average increase in productivity for people doing work from home, almost 50%. This includes internal company studies. One response from the Strongman Businessman people has been to demand mechanisms, essentially expressing incredulity at the possibility.
Which is the wrong response of course. Now there are many plausible mechanisms from not needing to get dressed in stupid costumes every day with the accompanying grooming, removing commutes, being able to get maximum recovery from actual breaks, lack of interruptions from coworkers, less theater to make it look like you're doing the work you're actually doing to prove to the middle manager you're doing what your output obviously shows, and a whole host of things that add up to workers being happier, healthier, with more sleep, doing their work better as a result. However, one can leave all that aside to point out that the output is still there. 'How are they doing it' is an interesting question that is in no way evidence it isn't happening.
That in turn leads to them questioning how you could even measure such a thing. The response to this is the quite obvious 'aren't you the ones who approve the metrics of work done? Isn't that, you know, one of the
main functions of YOUR job?' Either you trust your metrics of work performed or you don't. There is no 'I trust it when it tells me what I want to say and I don't when it doesn't' that gets out of this argument being a handwave.
Of course there are times it doesn't work. That can be because people are doing it wrong or because the tasks aren't well suited to it. That doesn't mean one should just return everyone to offices 'just to be safe'. That cowardly demand for control doesn't actually fit their 'smart business choices are cold and lead towards advancements' image anyway. It's the desire to be a petty tyrant. Now there are other lessons to be learned about going into why work from home has worked so much better than predicted that can be applied to an office environment (stop with the stupid costumes, don't over manage, make people take
effective breaks, stuff that can all be summed up as 'stop getting in your worker's way'), but that's another battle to be had once this one is accepted.