It is well known that there is a strong correlation between the income of parents and that of their offspring. The same applies to cognitive skills and to measure of IQ. The big question is whether this is due to genes or the environment. One more addition to this literature is based on the German Socio-Economic Panel, a large and incredibly rich panel that has measures of IQ along with plenty of other socio-demographic characteristics.
Silke Anger and Guido Heineck show that genes are certainly not the only ones responsible for this correlation, but they are part of the story. Interestingly, this study can differentiate between two types of cognitive abilities: cognitive speed, which is supposed to be independent from the environment, and verbal fluency, which is acquired. It finds that mothers matter more than fathers (no surprise), that there is an own-gender effect in the transmission of language skills, and there is stronger intergenerational transmission of verbal fluency that cognitive speed.
What all this means is that without help from the government, the development of cognitive skills is likely to be inefficient. As a society, you would want the innately brightest children to benefit the most from school to enhance the total quantity of human capital. But this is very elitist and may not satisfy fairness or some egalitarian goals. If the goal were to give everyone the same IQ, state intervention would be necessary to counteract the naturally occurring positive correlation. The trouble is that societies have difficulties establishing what their actual goals are: efficiency or fairness.