> For instance, do people who declare their occupation as landlord have some demonstrable income distribution x std deviations above some control group?

Not really. Being a landlord takes a bunch of money to sink into the real estate assets. If you don't put it in there you'd just leave it in a mutual fund or whatever and be making passive income of about the same order. Real Estate is a solid investment choice but it's not *that* much better.

Basically this argument is for the form "being a landlord looks like a working class job but pays better and that's not fair".

But the actual truth on the ground is "being a landlord is just a different way of being wealthy, and you have to fix plumbing on the side instead of working a day job in an office".