This is a reasonable-sounding hypothesis that has been researched and disproved. Diverse teams perform better.

This would actually settle the whole thing. If these researches are true, companies which can afford it should diversify their teams (old, young, male, female, different nationalities, etc etc.) in order to increase efficiency and make more money. Which researches did you base this comment on?

And this is exactly what large software companies are doing. Microsoft's direct college hire program has been a good source of great black and women developers and architects. The teams I was on were definitely improved by their inclusion even though they were "inexperienced". Microsoft had the most diversity of any company I've worked at and it's not even close.

There's no shortage :-) Start here: https://hbr.org/2016/11/why-diverse-teams-are-smarter

That's precisely the sort of reasoning people use. It might even be true. The problem is that when I see a company like that I don't think "Wow, that company is a well optimised for team efficiency." and instead I think "I wonder why they only hire white marbles? Probably racist."