The United States, in living memory, was a better/more affordable place for the working class in part because of government retirement benefits, union-friendly laws, federally subsidized home loans, and federally subsidized education. One of the reasons that there was so much of this “free stuff” available to white people back in the day is that it wasn’t available to nonwhite people. Discrimination meant there were bigger slices of the pie to go around, so to speak, for every white person.

Especially when compared with other more strikingly abhorrent harms like lynching and water-fountain segregation, this kind of (relatively!) subtle economic discrimination is not part of the story of civil rights in the 20th century that white Americans learn in school and elsewhere. But it’s a pretty good reason why black voters and other nonwhite voters would not want to make America more like it used to be.

Ben Mathis-Lilley slices the pie.

I think this is interesting trivia but I’m not sure the maths entirely works out. For one thing, it seems to assume that the “economic pie” is fixed, and that greater access to the inputs of it (i.e. allowing people of colour comparatively greater access to jobs, particularly in the middle- and upper-working class) doesn’t impact the outputs (i.e. the tax base collected and, thus, available for allocation to social services). If this were true, the US’s GDP would track only against the growth of working class white males which, yeah. It doesn’t (q.v. the impact of women’s participation in the labour market, for one example).

So while I think the premise of the argument is true (i.e. people of colour have historically been denied access to government benefits in favour of whites), I’m not sure that you can say the reason government benefits now are less is because the pie has outgrown the diners.