The Impact of the Parenting Style on the Formation of Adolescent Human Capital

In the literature on early formation of human capital and the intergenerational correlation of human capital, there is more and more attention to human capital investment within the household by parents’ behavior and interaction with children, while the parenting style has received less attention in empirical economic studies. This paper attempts to fill in this blank by examining the impact of the parenting style on adolescent cognitive and non-cognitive abilities.　　Based on the data of China Family Panel Studies （CFPS, 2010 to 2014）, and drawing on the mainstream literature of psychology, we select relevant questions and construct two continuous variables that measure the variance on two different dimensions: demandingness” and responsiveness”. Further, we divide the parenting style into four types according to the existing practices in the literature: authoritative, authoritarian, permissive and neglecting. As a dependent variable, we use academic performance to measure children’s cognitive abilities, and use the standard CES-D scale as a measure of children’s mental health, which is a proxy variable for non-cognitive abilities.　　In the regression analysis, we use the leave-out averages of the two dimensions of the parenting style in the community as the instrumental variables. For the first time, we make use of the instrumental variable method to control the possible endogeneity problem. We find that the degree of parental demandingness and the degree of responsiveness both have a positive impact on children’s academic performance, while only the responsiveness has a significantly positive effect on children’s mental health. High demandingness may not be conducive to children’s mental health development. As for the comparison of the four types of the parenting style, we find that for children’s academic performance and mental health, the authoritative type is the best. On the basis of the benchmark regression, we also try a different measurement of academic achievement to verify the robustness of the conclusion.　　In addition, we also conduct subsample regressions based on gender and urban/rural status, and find that the impacts of the parenting style are different in different sub-samples: from the gender perspective, the impact degree of the parenting style on boys is generally stronger than that of girls; the demandingness only has a significantly positive impact on boys’ academic performance; and girls’ mental health is almost not affected by the parenting style. According to urban and rural subsamples, we find that urban children’s academic performance is only affected by the degree of responsiveness; while in rural areas, demandingness and responsiveness both have a positive effect on children’s academic performance. As far as mental health is concerned, the effect of the parenting style for the mental health of rural children is much weaker than that for urban children.