This paper estimates a sibling-difference model of the impact of sibling height-for-age differences on parental investment in each sibling’s cognitive development. The aim of the paper is to uncover parental preferences for equality or efficiency in human capital investments by determining whether parents make compensating or reinforcing investments. Understanding parental behavioural responses to endowment differences is an important intermediary step in understanding how these decisions in childhood impact the developmental outcomes of interest such as health and earnings in adulthood.

I estimate a sibling-difference model using OLS and instrumental variables in order to control for the endogeneity of child’s height-for-age (stunting). To instrument sibling height-for-age difference, I use child’s quarter of birth, household idiosyncratic shocks during the critical stage of development, age 0-3 years, and rainfall shocks as instrumental variables. Using the Young Lives Round 2 and Round 3 data for the older cohort of Ethiopian children, I compare parental investments in siblings in order to determine the difference in school hours between siblings who are the same age (the index child is 11-12 years old in Round 2 while the sibling is a similar age in Round 3) due to different height-for-age scores. This approach controls for household and community fixed effects in order to identify whether parents make compensating or reinforcing investments. How poor parents respond to endowment differences in their children is not obvious ex-ante and is thus an empirical question.

This paper finds evidence of reinforcing investments such that healthier siblings spend, on average, 0.19 hours more per day at school relative to their siblings. This result is robust to measures of parental expenditure on siblings and finds that parents spend relatively more on uniforms and stationary for healthier siblings constituting another type of reinforcing investment.