THESIS
2021
1 online resource (xiv, 126 pages) : color illustrations
Abstract
This thesis consists of three essays on income inequality in China.
The first essay provides new Gini index estimates of income inequality in China for the years 2012, 2014, and 2016, which are obtained by applying a new enhanced semi-parametric Lorenz curve method to data from the China Household Finance Survey (CHFS). Our estimates, which appropriate accounts for the influential observations of top income earners, yield estimates of the Gini index for China of 0.609 in 2012, 0.612 in 2014, and 0.613 in 2016. These results are similar to those obtained when we apply our method to the data from the China Family Panel Studies (CFPS), and are consistent with other income inequality estimates in China which use top income adjustments in addition to the survey data. However, the results a...[
Read more ]
This thesis consists of three essays on income inequality in China.
The first essay provides new Gini index estimates of income inequality in China for the years 2012, 2014, and 2016, which are obtained by applying a new enhanced semi-parametric Lorenz curve method to data from the China Household Finance Survey (CHFS). Our estimates, which appropriate accounts for the influential observations of top income earners, yield estimates of the Gini index for China of 0.609 in 2012, 0.612 in 2014, and 0.613 in 2016. These results are similar to those obtained when we apply our method to the data from the China Family Panel Studies (CFPS), and are consistent with other income inequality estimates in China which use top income adjustments in addition to the survey data. However, the results are significantly higher than estimates based solely on household survey data.
In the second essay, we conduct the first systematic empirical analysis of income inequality in China at the provincial level. More specifically, we use the data from China Household Finance Survey (CHFS) and a combined Lorenz curve semi-parametric approach to estimate Gini indices in Chinese provinces for the years 2012, 2014, and 2016. We find that an important factors explaining differences in inequality across provinces are the “prices” and “quantities” of human capital across provinces. Provinces with higher educational inequality, lower average years of schooling, and higher returns to schooling have higher levels of income inequality. At the same time, inter-provincial differences in the returns to schooling could be mainly attributed to the remaining barriers to the labor mobility. Our findings suggest that poor provinces are currently severely disadvantaged as compared to rich provinces: they face higher income and educational inequality as well as higher returns to schooling, while at the same time lower average level of education. We conclude that the reduction of existing inter-provincial human capital gaps and the acceleration of labor markets’ integration through appropriate government policies could substantially contribute to the reduction of income inequality and disparities in inequality across regions in China.
In the third essay, we extend the Bergstrom et al. (1986) model of voluntary provision of public goods so that individuals also care about income mobility when deciding on their contribution towards public goods. To address this feature we incorporate an additional term in the individual utility function that accounts for the expected change in the distance to mean income between time periods. We claim that a higher degree of mobility reduces the distance to mean income, which scales up the individual identification with the society and raises the willingness to contribute to public goods. Using the data from the China Household Finance Survey (CHFS), we estimate income mobility indices for 29 Chinese provinces in 2014 and 2016, and test our theoretical predictions. We find that a narrower expected gap between individual and mean incomes in the future, reflecting a higher level of provincial mobility, strengthens the willingness to pay for environmental protection, and that these results are robust to controlling for the effects of income expectations, income inequality, and linguistic heterogeneity.
Post a Comment