Human Capital Investment in the Presence of Child Labor
Policies that improve early life human capital are a promising tool to alter disadvantaged children's lifelong trajectories. Yet, in many low-income countries, children and their parents face tradeoffs between schooling and productive work. If there are positive returns to human capital in child labor, then children who receive greater early life investments may attend less school. Exploiting early life rainfall shocks in India as a source of exogenous variation in early life investment, we show that increased early life investment reduces schooling in districts with high child labor, especially for girls and lower castes. These effects persist and are intergenerational, affecting fertility, per capita household consumption, and other measures of household poverty, and lead to a divergence in the next generations' educational outcomes. Our results are robust to the inclusion of rich controls for district-level characteristics and an IV strategy. We provide evidence that reductions in educational investment in response to positive early life shocks are inefficient.
We are grateful to seminar and conference participants at Duke, IFPRI, Kent, NBER Children's meetings, Northwestern, NYU, UCSD, and PacDev for insightful comments and suggestions. Shah and Steinberg gratefully acknowledge funding from NSF grant#1658852. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.