Gender wage inequality in the urban Chinese labour market
This thesis presents an analysis of the gender earnings gap in urban China. The focus is on the impacts of human capital, family circumstances, social and political capital, occupation, ownership sector, industrial sector, and geographic locations on the gender wage gap. Fixed and random effects regression techniques are applied to analyse the November 2001 to January 2002 cross-section of the CULS dataset. Our findings indicate that the gender earnings gap in China was remarkable in 2001, when women made only 78.57% of men’s earnings. This gap persists after controlling for individual, industry, and city level characteristics: women now make about 16 to 20% less than men do. The size of earnings in urban China is highly dependent on individual and industry level characteristics. It is positively affected by work experience, educational level, health status, marriage, social capital, and CCP (Chinese Communist Party) membership. And it varies with and within occupations, ownership sectors, industrial sectors, and geographic locations. Except for two: marital status and administrator, gender wage differentials are not identified in other variables, indicating that rather than different income returns due to individual and industrial level characteristics, disproportionate allocation of men and women in those variables is the major force behind the earnings gap. Family circumstance is another critical part of the story of gender differences in income, where gaps are concentrated among married women, especially mothers. It is also the primary contributor of the vicious cycle against women, which has been proved to exist in the Chinese labour market. Although inter-regional income disparities are significant, they are not associated with a city-level gender effect on income. Net of controls, the gender gap in income remains constant among cities of different income and economic development level. At last, by controlling all of these variables, the gender gap in wages is still statistically significant, which demonstrates that discrimination also plays a role in this issue.