> There could also be a simpler causal link: books and lucrative jobs could both caused by higher family socio-economic status.
ISTR in the last few years seeing similar research that controlled for most typical measures of SES that are associated with positive outcomes (parents income, parents educational attainment, etc.) which found that growing up with more books in the home had a positive impact on a number of outcome measures (leveling off once at about 100 books.)
I suspect that there is some degree in which it is serving as a proxy for as-yet-unidentified socio-economic, genetic, and parenting style and parental personality traits, but I also suspect even once all those were identified and taken into account, simply having books handy has some positive effects.