This dissertation contains three chapters in the fields of development economics and labor economics. In the first chapter, I highlight the social impact of empathy on school bullying reduction. I collect unique data by conducting a randomized control trial---a parent-directed empathy education intervention in middle schoolers in China. Program evaluation shows it reduces bullying by indirectly changing the network structure, making bullies less popular in the classrooms. I estimate a unified framework incorporating an empathy production function, a network formation model, and a social interaction model of the final bullying outcomes. I find that the social channel of empathy accounts for almost half of its human capital effect. Policy counterfactuals suggest that targeting bullies’ friends is more effective than targeting bullies directly.

The second chapter (co-authored with Flavio Cunha, Yiming Xia, and Naibao Zhao) provides the context and a comprehensive introduction to the empathy intervention program. The program leads to more parental investment and higher empathy levels. Using a state-of-the-art generalized random forest method, we find more reductions in bullying for those with lower parental investment and academic stress. Cost analysis shows that reducing one bullying incident costs $16.30 for the intervention, suggesting a scalable and low-cost strategy to inform public policy on bullying prevention in other similar settings.

In the third chapter, I develop a new approach to identify intrahousehold resource allocation and the extent of joint consumption for extended families using a collective household model. It allows for endogenous living arrangement decisions with fewer data requirements. Application to nationally representative household survey data in China reveals that the elderly are allocated the least resource shares, and women tend to be more altruistic in consumption sharing than men. I also confirm that co-residence households enjoy economic efficiency gains compared to nuclear households, but there is still room for explanations on why multi-generations choose to live together.