Start What You Finish! Ex Ante Risk and Schooling Investments in the Presence of Dynamic Complementarities

We study the relationship between risk and schooling investment in a low-income setting. Our theoretical model reveals that parents respond to variance in future child schooling by reducing investments ex ante if the human capital production function exhibits dynamic complementarity and parental preferences for human capital are not too concave. We further explore the implications of the theory using multiple rounds of panel data from rural India that contain, in each round, three seasons of time allocation for each sampled child. In particular, we first use seasonally-measured study times to estimate the key parameters of the structural model. These estimates suggest that elimination of variance would result in a 15 percent increase in acquired human capital, 50% of which is attributable to an ex ante response. We then use cross-village variation in risk over time to estimate these effects, using the idea that the consequences of rainfall variation for income in a rural economy are importantly affected by the presence of irrigation. Using this variation, we estimate an ex ante elasticity of study time with respect to variance that is somewhat smaller, but of the same order of magnitude as those predicted by the structural model. Finally, we simulate the effects of an implicit social insurance program, modeled after the National Rural Employment Guarantee Scheme (NREGS). Our results suggest that the risk-reducing effect of the NREGS may offset adverse effects on child education that were evident during the NREGS phase-in due to rising wages.