Correlation doesn't always imply causation, right? For example, there's a correlation between IQ and income. There are several causative possibilities:

* IQ causes income

* Income causes IQ

* Some unknown thing causes both income and IQ

* The correlation is spurious; IQ and income are unrelated

Evidence strongly suggests that the third bullet point is true; socioeconomic class causes both income and IQ. Richer people living in nicer neighborhoods both have better opportunities for income, and also better opportunities for education; education causes IQ. This is why redlining is brought up so often as a root cause of so many of the disparities in quality of life; redlining deepened socioeconomic divides.

Haven't there been studies untangling causation here? Using random events that hit socioeconomic class as a natural experiment?

Adoption studies might also be informative.