Essays on Health and Development Economics - Focusing on Early Stages in Life

This thesis presents three essays on health development when exposed to adverse shocks and first-year vaccinations and uses data from 35 African countries and the United States. Chapter 1 examines the impact of drought exposure in utero on sex ratio and child health at birth using data from 33 African countries between 1959 and 2019 to shed new light on the consequences of in-utero shocks on selection at birth. We find that drought exposure during the first month in utero increases the probability of female birth, in other words, this indicates that boys are more likely to be ‘lost’ when a heavy drought occurs during pregnancy. To strengthen this finding, we test for a causal effect on the sex ratio based on the cluster-month level dataset. The estimates confirm that drought exposure during the first month in utero results in a skewed sex ratio towards females. Additionally, we also find that drought exposure in utero decreases birth height, especially for girls. Chapter 2 explores the impact of wildfire exposure in utero on the birth–sex ratio and birth weight in a strongly institutionalised setting. We find that exposure to wildfires during the second trimester of pregnancy reduces birth weight by 0.3% and that boys are more sensitive to such adverse environments than girls. Although the average effect of wildfires exposure on the birth–sex ratio is insignificant, our heterogeneity analysis shows that wildfire exposure during the second trimester of pregnancy tends to increase the female–male ratio among less educated mothers, which indicates that the mother’s level of education is an essential mediating factor in determining the birth–sex ratio. This latter result aligns with the Trivers–Willard hypothesis, suggesting that exposure to negative shocks while in utero results in a female-biased birth–sex ratio.
Chapter 3 studies vaccination hesitancy and vaccination rates among Muslims in African countries using data for children born between 1999 and 2019. We first find that newborn Muslims are less likely to be vaccinated compared to other religious groups. Second, we study the mechanism behind this lower vaccination incidence by examining the hate crimes against Muslims recorded in the US. Our empirical results imply that one SD increase in hate crimes against Muslims leads to a decrease in the fraction of first-year vaccinations among the Muslim population by up to 3%, and this pattern is common across different types of first-year vaccines. A series of heterogeneity analysis tests identify that trust in institutions and mobile phone penetration are important mediating factors that contribute to vaccination hesitancy, which in turn decreases child immunisation rates.