Essays on Childhood Environments and Human Capital Formation

Abstract

This thesis comprises three papers that examine childhood environment effects on individual cognitive and health outcomes in adulthood.

Chapter 2 examines the relationship between fertility and the gender gap in cognitive skills in a society where son preference is prevalent. Drawing on Becker's Quantity-Quality trade-off model, I empirically test if the trade-off between quantity and quality of children is larger for daughters than that for sons. To consider the endogenous nature of the demand for children, I exploit an exogenous variation in fertility due to China's family planning policy. We utilise the policy intensity information collected from hundreds of county gazetteers as an instrument for family size. The main results suggest that an additional sibling widens the gender gap in cognitive test scores by 33.2% of a standard deviation in the rural sample and 9.8% in the urban sample. The pattern is more pronounced in regions with a higher proportion of people who prefer a son over a daughter, and among households who face tighter budget constraints. I also provide suggestive evidence that our findings are strongly associated with belief in a son's role to carry on family lineage in Confucian tradition.

Chapter 3 explores childhood environment effects on adult height. Between 1950 to 1990, Chinese average adult height increased more than 1 cm a decade. This is an impressive achievement. This paper describes the trend of adult height growth during this period and explores the reasons behind such a trend. The result suggests that during the 40-year period, the growth of adult height increased, with the most substantial increase occurring in the last decade. One important reason behind the increase in adult height is the continued increase in government per capita spending on health and education throughout the forty-year period. However, the recent impressive growth was mainly due to the market-oriented economic reforms introduced in the 1980s. The observed height growth in the 1980s is a combination of two opposing factors, economic reforms and the introduction of the One Child Policy (OCP). We find that once we control for the 'number of siblings' after the introduction of the OCP, the revealed economic reform effects become much larger. The effect of OCP is, in general, negative on height and, in particular, for rural females owing to the widespread son-preference in the rural society.

Chapter 4 tests whether long-term exposure to single-sex schooling is beneficial for female cognitive performance. While recent studies have demonstrated that single-sex schooling improves girls' academic performance, little evidence is available as to how long it can be effective. By exploiting South Korea's random school assignment policy, we attempt to answer this question. The results suggest that attending single-sex schools improves female academic performance for the first three years of exposure, consistent with existing evidence. However, girls exposed to single-sex education for their entire secondary school life perform worse than girls with less exposure at the college entrance exam. The estimates are robust to the inclusion of controls of individual, parental, and school characteristics. These patterns are not clear among boys. We provide suggestive evidence that show a gender-salient environment in single-sex schools may drive the results. Show more