Childhood Events and Long-Term Consequences

Abstract

Health and experiences in early childhood are strongly associated with adult outcomes. In this dissertation, I explore the association in detail with a focus on identifying the causal mechanisms that generate variation in early health and uncovering the parental behaviors that determine whether early health and living conditions evolve into long-term deficits or advantages.

In chapter 1, I explore whether pre-conception maternal desire for children of a particular sex has implications for the health of children in Indonesia. I show that a simple fertility stopping model predicts that when a child is born of the mother’s preferred sex, they will receive more resources, and I test this prediction empirically using a longitudinal data set. I find that children born of the mother’s preferred sex are heavier, have a higher body mass index, and experience fewer illnesses. I provide evidence that reductions in subsequent fertility are the primary mechanism for these effects.

The existing research measuring the long-term implications of early childhood conditions frequently fails to identify the mechanisms through which early deficits become life-long disadvantages. In chapter 2, I examine one instance where deficits may matter for long-term well-being. Using data from Indonesia, I find that when third trimester rainfall is fifty percent higher than expected, birth weight and relative size are approximately .23 standard deviations higher. Despite this early advantage, I find no persistent positive impact fifteen years later. However, parental investment appears to be negatively influenced by in utero exposure to rainfall, suggesting that parents compensate for early health conditions.

To date, research on the long-term effects of childhood participation in subsidized housing has been limited by the lack of suitable identification strategies and appropriate data. In chapter 3, I, along with my co-authors, create a new, national-level longitudinal data set on housing assistance and labor market earnings to explore how children’s housing affects their later employment and earnings. We find that while naïve estimates suggest there are substantial negative consequences to childhood participation in subsidized housing, household fixed-effects specifications attenuate these negative relationships for some demographic groups and uncover positive and significant effects for others.