“The Nurture of Nature and the Nature of Nurture: How Genes and Investments Interact in the Formation of Skills”, 2024-02 ():
This paper studies the interplay between genetics and family investments in the process of skill formation. We model and estimate the joint evolution of skills and parental investments throughout early childhood.
We document 3 genetic mechanisms: the direct effect of child genes on skills, the indirect effect of child genes via parental investments (nurture of nature [ie. “the environment is genetic” or evocative/reactive gene-environment correlation]), and family genetic influences captured by parental genes (family genetic associations).
We show that genetic effects are dynamic, increase over time, and operate via environmental channels. Our paper highlights the value of integrating biological and social perspectives into a single unified framework.
…Our approach allows us to gain additional insights into the process of skill formation. We find that genetic influences accumulate over time and gradually increase over the early childhood period. Genetic influences on initial skills are small, but by ages 6–7, a one standard deviation increase in the child’s genetic factor leads to almost a 0.2 standard deviation increase in skills. This pattern is consistent with earlier findings on the increasing importance of genes over the life-span (2013; Tucker-Drob et al 201311ya; Tucker-2014; et al 2016). Unlike previous work, our approach allows us to rule out several potential sources of bias, including confounding from the environment and differences in measurement error over time, and, at the same time, to gain additional insight into why this pattern appears. We find that the increase is due to two main mechanisms. First, conditional on their current stock of skills and parental investments, genetics make some children better able to retain and acquire new skills, the direct effect of genes. Second, parents reinforce initial genetic differences by investing more in children with higher genetic factors and higher stock of skills, the nurture of nature effect.1 We show that the second mechanism is more important at early ages, and the first is more important at later ages.
We also document a strong association between parents’ genes and children’s skills. This association captures the effect of family genetics and unobserved environmental factors correlated with parental genes on the environment experienced by the child. These influences explain 40–82% of the association between the child’s genetic factor and her skills. We show that these genetic influences are completely mediated by parental educational attainment. This suggests that controlling for parental education may be enough to capture the family background in analyses of child development that have access to a child’s genetic data but not their parents’.
Another key contribution of our paper is documenting what child development models that ignore genetic influences miss. First, we show that neglecting genes leads to an overestimation of the returns to parental investment, although adding parental controls eliminates most of this bias. Second, we identify substantial heterogeneity in the returns to investments across the child’s genetics that is not captured by models that ignore genes. This heterogeneity is only partially captured by observable family characteristics. Third, the genetic heterogeneity that is captured in models without genes is misattributed to observable nongenetic factors, such as family income, that do not reflect the underlying causal mechanism. This exercise highlights the importance of incorporating genes into models of skill formation.
…The paper is organized as follows. In Section I, we outline the theoretical framework describing the various channels through which genes influence skill accumulation. In Section II, we introduce the ALSPAC dataset, discuss our measures of the genetic factor, and conduct a preliminary descriptive analysis. In Section III, we describe our empirical model, including the measurement system used to identify the genetic factor, latent skills, and investments, along with the estimation procedure. Section IV presents our main results. In Section V, we study mechanisms and highlight the implications of our findings for our understanding of the skill formation process. Section VI offers brief concluding remarks, discusses the limitations of our work, and makes suggestions for future research.