There has been a long discussion in the literature on whether intelligence is inherited or acquired, the nature versus nurture debate. The traditional empirical strategy has been to look at twins separated at birth and raised in different families. While this seems to be a perfect data set for this kind of study, it suffers from at least two drawbacks: samples are very small, and there could be a selection bias, as twins given for adoption may not be representative of twins, and furthermore, twins may not be representative of the general population.
Anders Björklund, karin Hederos Eriksson and Markus Jäntti focus on the correlation of IQ across siblings as it compares with father-to-son IQ correlation. For Sweden they find that they are 0.473 respectively 0.347 from a data set of military conscripts (all male). As siblings share genes and environment, whereas father and son only share genes, they argue that this is evidence that the environment is important. An environment that may include the mother and her genes, by the way.
These results look interesting, but I do not quite know where to go from there. Does this mean that we have less to worry about the intergenerational persistence of skills? Or that the high correlation of earnings within a family is OK and not a sign of mis-allocated opportunities in society?