I find this strange. If someone wasn't able to get work done in the office the solution wouldn't be "OK, we all need to work from home because Gary can't get stuff done otherwise".
That’s only one factor, though it’s not just “Gary” IME many people are in this category, the very effective remote people are less common. The junior engineer issue and the comms issue affects everyone. Companies are teams of people and you want to make the collaboration as low effort and fast as possible. In person is very good for that and I suspect better on net.
It’s why companies with the leverage to push for it (Apple and Amazon, Elon Musk companies) do so - they believe it to be an advantage.
How many people do you believe were "very effective" in office? The truth of the matter is most people are "effective enough".
I honestly don't know if it's actually better or it just appears better. Sure, there are times when fast communication is important, but as we see with Twitter lately it's obvious that rapid fire decisions sometimes aren't the best and while things seem very active and dynamic maybe it'd be better if communication was more considered and thoughtful. Isn't that why that whole "this meeting could have been an email" meme exists?
I agree the companies believe it's an advantage, but I don't know that it is. I don't know that they know it is. Leadership definitely thinks so, and there's a ton of institutional momentum that still hasn't shifted after the last few years, but I don't that it's actually empirically better.
I think we’ll get to see the experiment happen which is cool. If either truly has a decisive advantage in some contexts we should see that materialize in the market.
Like all startups, it’ll be in a new field. If it’s an advantage some sort of AGI/LLM remote company should be able to out compete non-remote variants.