This article is missing important info:

(1) This is much less interesting if the effect doesn't propogate, i.e. if income levels only lasts one generation. This is a crucial part of eugenics -- which the articles compares the effect to -- and yet nothing in the article suggests that income transfers generations. (I'm not saying it does or doesn't; just that the article doesn't address this at all.)

(2) The same is probably true for good looks, culture, athletic ability, hobbies, education, geography, race, IQ, and wealth. I suspect there are a number of confounding variables. Omitting any mention of them doesn't invalidate the reported results, but does make misinterpreting the results of the article much easier.