Working Paper: NBER ID: w18429
Authors: Anna Aizer; Flavio Cunha
Abstract: We study how endowments, investments and fertility interact to produce human capital in childhood. We begin by providing empirical support for two key features of existing models of human capital: that investments and existing human capital are complements in the production of later human capital (dynamic complementarity) and that parents invest more in children with higher endowments due to the complementarity between endowments and investments (static complementarity). For the former, we exploit an exogenous source of investment, the launch of Head Start in 1966, and estimate greater gains from preschool in the IQ of those with the highest stocks of early human capital, consistent with dynamic complementarity. For the latter, we are able to overcome the potential endogeneity and measurement error associated with traditional measures of endowment based on health at birth. When we do, we find that parents invest more in highly endowed children. Moreover, we find that the degree of reinforcement increases with family size. Thus, an increase in quantity leads not only to a decline in average quality (the quantity-quality tradeoff) but to an increase in the variation in quality, due to both greater variation in endowments (from more children) and greater reinforcing investments. These findings can be explained by extending the quantity-quality trade-off model to include heterogeneous child endowments and parental preferences that feature complementarity between quality and quantity and moderate aversion to inequality in child human capital within the household.
Keywords: human capital; early childhood; parental investments; fertility; Head Start
JEL Codes: I24; J13; J24
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| investments in preschool (I21) | child IQ (I12) |
| higher initial stocks of early human capital (J24) | greater gains from preschool (A21) |
| parents invest more in highly endowed children (D29) | static complementarity (D10) |
| one-standard-deviation increase in birth weight (J19) | 10% increase in average investment differences between siblings (G59) |
| family size (J12) | greater variation in child quality (J13) |
| investments in preschool (I21) | significant impacts on cognitive achievement (I24) |