Ruyi Shi

"Internal migration and formal employment: evidence from Indonesia"

This paper examines the causal impact of internal migration on formal employment in Indonesia using longitudinal household survey data from the Indonesia Family Life Survey (IFLS). We use rainfall as the instrumental variable for migration to address the endogeneity bias led by reverse causality. Our empirical results show that, in general, internal migration is expected to increase the chance of having a formal job by approximately 16 percentage points. This impact is heterogeneous across gender and levels of education. The magnitude of this impact rises to around 25 percentage points for males but drops to only 7 percentage points for females. We also find migration only has a significant impact on individuals with senior high school education or higher, with the estimated impact being roughly 20 percentage points. 

Following are some interesting graphs/satellite data used in this paper:

      Household locations                           Rainfall data                                Nighttime lights data 

"Adult children’s education and parents’ mental health in China: a quantile approach"

 This study examines the heterogeneous effect of adult children's education on elderly parents' mental health in China. We use children's exposure to the compulsory schooling law as an instrument to address the endogeneity bias of education. Our results from the instrumental variable quantile estimation reveal that the effect of children's education is heterogeneous across the distribution of parental mental health. Specifically, children's education has no effect on parents with mild to moderate mental health issues (below the 0.7 quantile) but has a positive and significant impact on those with severe issues (0.7 quantile and above). Our split-sample analysis indicates that a son's education has a greater positive impact on parental mental health than a daughter's. We also find the positive effect is greater for parents living in rural regions compared to those in urban regions. In the comparative analysis, we show that when parents are not depressed, the schooling of the least educated children improves parents' mental well-being more. In contrast, when parents have moderate to severe mental health issues, the schooling of the most educated children has a greater impact. Additionally, the educational impact of first-born children is more substantial compared to that of second-born children.

"Birth order effects, gender composition and old-age support: new evidence from China"

This paper provides rich new evidence on China's evolving birth order effects on education. A positive birth order effect on education is observed among older generations (born before 1967), while the birth order effect reverses to negative among younger generations (born after 1978). For the positive effect, school dropout among older siblings is the primary driver, whereas child labour laws and resource dilution are the key factors for the negative effect. Sons have consistently attained higher levels of education than daughters, but this gap has narrowed over time. In addition, differences in old-age support levels are negligible both among siblings and between sons and daughters. This implies that children who are expected to provide more support to their parents in return for the favours they received during childhood do not, in fact, offer more support. The altruistic motive emerges as a key reason why siblings, as well as sons and daughters, provide equal levels of old-age support to their parents.