Intergenerational transmission of inequality: A view within the family

How important is the family in shaping children's capabilities and opportunities? Extensive research across many disciplines has been devoted to studying whether and how social and economic status is transmitted from generation to generation, including infant health right at birth. In this work, I provide evidence for another social return to education by studying the effect of maternal schooling through a view within the family. Even though considerable work has been devoted to understanding whether maternal schooling affects the levels of children's outcomes, to the best of my knowledge, this is the first study to focus on how maternal education affects the distribution of outcomes within the family.

My empirical analysis is based on the matched mother-child data file from the National Longitudinal Survey of Youth 1979. I show evidence that siblings from low educated mothers have lower levels of and more unequal birth outcomes relative to siblings from highly educated mothers. For birth weight, which has been associated with substantial long-run effects, I find that the gap within family is 550g for mothers who finished middle school while this gap is 400g for mothers with a college degree, which represents a 0.2 effect in standard deviation units. Similar results are obtained for measures of prenatal investments. These results have been shown to be robust to the addition of mother-level as well as child-level control variables.

My explanation to these results emphasizes the ability to smooth investments over time -- higher educational attainment not only provides higher income, but also makes women less vulnerable to the aggregate economic environment. The novelty of this framework lies in showing that differential prenatal investments and consequent human capital gaps between siblings happen due to uninsurable shocks. Currently, I'm working on evidence for less educated women being more sensitive to the business cycle using employment data at the county-level from the Bureau of Economic Analysis.

Then, I consider several alternative explanations that do not rely on a role for income uncertainty and market imperfections and argue that these are hard-pressed to explain the within-family dispersion results. My findings have several implications for the interdisciplinary study of inequality. Given that these initial disparities have been shown to persist, policies that aim at reducing inequality by targeting future mothers will have another effect not yet documented before.