According to a new Brigham Young University study, human capital endowments passed from father to son — perhaps in the form of genes, advice, work ethic, or some other intangible — could be more important to a son’s success than the size of dad’s paycheck. The findings are published in the Journal of Political Economy.
“We know there’s a correlation between fathers’ income and sons’,” said David Sims, an economics professor at BYU and one of the study’s authors. “What’s gotten less attention is the mechanism. We wanted to see if the intergenerational income correlation is due to money — what we can buy for our kids — or if human capital plays a role as well.”
The problem is that separating the two inputs is tricky. On average, fathers with higher human capital endowments also tend to have higher incomes, so it’s hard to tell which factor is doing what. Sims and his colleagues used a statistical model and a rich dataset to try to disentangle the two.
Sims and his colleagues used remarkably detailed government administrative data on a large sample of Swedish fathers with sons born between 1950 and 1965. The dataset included salary information for fathers and sons as well as clues about fathers’ human capital — education level and the nature of their occupations. Men in jobs that require specialized skills or education are considered to have higher human capital endowments.
First, the researchers used the data to look for a raw correlation between fathers’ income and sons’, which, as expected, was quite high. Then they employed a statistical methodology to examine how this correlation changes when they isolated differences in fathers’ income due to something other than human capital. For example, a smart, well educated father may lose his job because the company he works for closes, or he may have a lower income because he lives in a town with a depressed labor market.
Under these types of circumstances, the researchers found, a father’s income was not a good predictor of his son’s income.
These findings allow the researchers to conclude that, for the men in their dataset, a father’s income alone isn’t the only thing that matters. Differing human capital endowments, passed from father to son, actually account for about two-thirds of the overall intergenerational income relationship.
“Obviously, our measure of human capital isn’t perfect. For example, father’s education is just one dimension of human capital,” Sims said. “If we had perfect measures of human capital, this question wouldn’t be so hard to answer. But what this gives us is an upper bound for the extent to which money is responsible for the intergenerational transfer of income. We show that the monetary side is responsible for about a third of the correlation.”
“We don’t offer a final answer here,” Sims added. “But we do offer some boundary conditions and present a methodology that could help unravel the question.”