This thesis contains four essays discussing how institutions, economic policies, and culture could have significant impacts on gender inequality, demographic changes, and decisions regarding investment in human capital development. Reducing inequality, including gender inequality, and improving human capital are the key priorities to achieve sustainable growth for developing countries. 

Chapter 2 investigates the role of local political institutions in influencing decisions regarding public education in Sweden during the early twentieth century. We leverage a unique setting where local municipalities were mandated to switch from direct to representative democracy, determined by a population threshold. At the same period, the national government was promoting the harmonization and extension of instructional time (regarding term length and years of education) at primary schools by financing most of the required costs while allowing local governments to decide the timing of these changes. Our analysis reveals that municipalities with representative systems implemented these educational extensions earlier, thus contributing to the limited literature on the role of representative democracy in promoting development. While existing studies mainly focus on expenditure indicators, our findings further show that representative institutions were also more pro-education in a non-pecuniary dimension. On examining the underlying mechanism, we provide some suggestive evidence of the persistent influence of local political elites. Exploiting information about the distribution of votes at the municipalities back to the late nineteenth century, which was based on the amount of tax contributions, we find that municipalities with a high concentration of political power were more likely to remain in direct democracy decades later if they were not subject to the mandatory political reform. In addition, the effects of representative democracy on education were only statistically significant in municipalities without political elites (either landed or capitalist elites). Our study highlights the potential problems of decentralization in which elites can dominate the policy agenda at the local levels. These results provide relevant insights for countries seeking to promote inclusive and sustainable growth, as well as those in the process of decentralizing decision-making and encouraging civil participation in local policies. 

Chapter 3 addresses the impacts of cultural norms on gender inequality in human capital formation. There is extensive literature documenting the problems associated with the preference for male offspring, especially in societies with patrilineal or patrilocal practices. Firstly, this preference is driving the manifestation of the sex ratio by girl neglect or sex selection. Secondly, it creates several disadvantages for girls, as they are more likely to grow up in larger or lower socioeconomic status families and face intrinsic discrimination where parents allocate more resources to boys even before birth. Our study on the context of Vietnam offers a unique perspective to the literature by demonstrating that son preference is not always associated with disadvantages for girls in many dimensions. We conduct a rigorous analysis using various datasets covering the period from 1989 to 2019 and employ different approaches to address the issues of child gender non-randomness and endogenous parity progression. We show that the increase in the child sex ratio in Vietnam is entirely driven by sex selection without evidence of girl neglect, despite the country's long-standing family planning policies. In addition, we find no evidence of preferential treatment towards boys, and girls outperform boys in several health and education outcomes. Our study, therefore, contributes to the comprehensive understanding of son preference and gender dynamics in Vietnam and the broader discourse. 

Chapter 4 explores how trade policies affect women's marital and fertility choices in Vietnam. Limited literature in this field investigates the impacts of rising import competition and emphasizes the mechanism of changing labor market outcomes of women \textit{relative} to men. However, their findings are diverse and do not always align with theoretical predictions that depict women's specialization in domestic work and men's in market work. Our study contributes to the literature by assessing the impacts of export growth induced by the U.S.-Vietnam Bilateral Trade Agreement in a developing country with an exceptionally high level of female labor force participation and a persistent preference for boys. Our analysis finds an adverse effect of the trade shock on the marriage and fertility of young women (aged 18-28) but no impact on older women, suggesting a delay or postponement. In addition, we do not observe any significant effect on the changes in the sex selection behavior of women or the aggregate child sex ratio at the provincial level. These results might be driven by the structural transformation with more manufacturing jobs and women staying longer at school. In addition, this trade shock does not affect the working status of women in absolute and relative terms compared to men, nor does it affect gender segregation in the labor market. This also helps explain the null effect of the shock on the male-to-female sex ratio among young children in Vietnam. Our study thus contributes to understanding demographic changes, which are currently of core concern in both developed and developing countries. 

Chapter 5 assesses the impacts of women's economic empowerment on domestic violence. Most of the existing literature focuses on women's employment in non-paid jobs and outside the households. This paper contributes to the literature by examining the context where most women engage in non-paid jobs and work inside the household's territory. By exploiting a unique longitudinal survey of rural households in Bangladesh, we find evidence that the women who work or actively contribute to the household's income suffer significantly more verbal abuse by other household's adult members. Our study highlights the need to collect more information on verbal abuse, which is absent in established data, but this form of domestic violence could have negative and lasting impacts on women and children. 

By examining the contexts of historical Sweden, Vietnam, and Bangladesh, the thesis not only broadens our understanding of these specific cases but also contributes to the general literature on gender, demographic, and human capital development. The findings of these papers underscore the importance of considering local political structure, cultural practices, and socio-economic conditions when designing policies aiming at fostering inclusive and sustainable growth in developing countries.