One point I don't see in the other responses: that conflation is way more coherent when you're looking at specific areas. Citylab is looking at relative wealth

One point I don't see in the other responses: that conflation is way more coherent when you're looking at specific areas. Citylab is looking at relative wealth in cities, and urban poverty really is hugely tied to race in the US for a bunch of reasons ranging from inherited wealth (having ancestors who owned land is a very big deal if they bought it in now-expensive cities) to overt discrimination in the recent past (white flight, blockbusting, etc.) Looking around the country as a whole, things get much less clear. There are heavily-white rural areas where lots of people are in extreme, <$2/day poverty. (Although nonwhite people in comparable places are generally even worse off.) And then other areas (e.g. northern US suburbs) are heavily white, but (in large part thanks to selection bias like housing prices) they don't necessarily have major racial wealth gaps among the people who do live there. The national numbers are a pretty confusing mixture of inequality in comparable situations and differences between situations, but looking at specific areas like Los Angeles the association is disturbingly strong.