Gender and human capital formation in developing countries

Abstract

Gender inequality hinders economic development, not least through its adverse effect on human capital accumulation. This thesis is comprised of three essays that shed light on the role of different policies; social preferences and practices that contribute towards gender inequality in developing countries.
Chapter 1 is set in the context of Colombia and is the first study to analyse the causal effect of age at first cohabitation on intimate partner violence outside Southeast Asia, where legal marriage is universal. Using age at menarche (first period) as an instrument, I find that age at first cohabitation only has a significant effect on IPV if a woman is legally married, which is consistent with the hypothesis that legally married women are more vulnerable to IPV as they face higher costs of separation as compared to those in de facto union.
Chapter 2 augments the existing literature on the effect of trade liberalisation on investments in child human capital by disentangling the effect of fathers- and mothers’ potential earnings exposure to Indian trade liberalisation in 1991 on girls’ and boys’ school attendance separately. The results are striking as they show that girls’ school attendance suffers more in districts where female employment is more exposed to the reforms, while boys’ school attendance suffers more when male employment is more exposed to the tariff reforms. Further analysis suggests that the decrease in girls’ school attendance is neither due to an income effect nor substitution effect.
Chapter 3 exploits the national scale-up of safe, affordable abortion centres following an abortion regime change in 2002 in Nepal that unconditionally legalised abortion up to 12th week of gestation to estimate the causal effect of improved supply of abortion services on sex selection. I also test for heterogeneous treatment effects given that sex selection is more prevalent at higher parities and among higher socio-economic status. Results show that sex selection did not become widespread in the wake of improved access to abortion. However, they do suggest that sex selection increased among certain subgroups.

Date of Award	23 Jun 2020
Original language	English
Awarding Institution	University of Bristol
Supervisor	Christine M F Valente (Supervisor) & Zahra Siddique (Supervisor)