[dTal]

It helps make them go away, over time, in the same way that refusing to use group statistics to make judgements about individuals (at a cost to your own predictive power) helps redress the inequalities that led to differences in groups in the first place.

[Spivak]

Ohh don't get me wrong, I strongly agree with this. But I think it's okay to take the blinders off to answer "why do these seemingly unrelated groups of people not buy hybrid cars?"

I really think the distinction is whether you look at these labels in good faith, care, attention to their history, and always through the lens of intersectionalism.