If I understand correctly, when the author say "hierarchical rank" he means that the rank (let's call it R) is larger for people at the top of deeper hierarchies, and that seems to be confirmed by the graphic in Figure 5.
Anyway, if we consider that deeper hierarchies are generally also larger overall and therefor produce more in economic terms, and that people in a hierarchy are considered to be responsible for the productivity of the levels below them, then R value is really a proxy for productivity, so the primary thesis of this whole essay, that R better correlates with income than productivity, doesn't make any sense because they are not independent. In the sense that matters to neoclassical economics the CEO of a big corporation is more "productive" than the CEO of a smaller company even if in practice the two do the same amount and quality of work, have the same education, etc.
Whether this should be so is another question, but if neoclassical economics defines productivity in this way (as I think it does), then income correlating to R is in no way contradictory.
Agreed, there are too many correlations to make hierarchal rank a useful metric for directly comparing to income.
I think the author would get more interesting results by looking at how much hierarchal rank is passed down through generations and comparing that against expected income.
Anyway, if we consider that deeper hierarchies are generally also larger overall and therefor produce more in economic terms, and that people in a hierarchy are considered to be responsible for the productivity of the levels below them, then R value is really a proxy for productivity, so the primary thesis of this whole essay, that R better correlates with income than productivity, doesn't make any sense because they are not independent. In the sense that matters to neoclassical economics the CEO of a big corporation is more "productive" than the CEO of a smaller company even if in practice the two do the same amount and quality of work, have the same education, etc.
Whether this should be so is another question, but if neoclassical economics defines productivity in this way (as I think it does), then income correlating to R is in no way contradictory.