i think it’s pretty clear GP is not saying it is worthwhile and is actually implicitly criticizing the practice.

> high school performance is a statistical proxy for class (wealth). In the US, it is a statistical proxy for ethnicity as well.

the degree to which this claim about wealth is true is impacted by confounders. it is generally less true than commonly stated. outside of the public sector, that a measure is correlated with race/ethnicity/class does not make it a priori illegal to hire based on.