Fate or opportunity?Impact of family background on household welfare in rural China Open Access
Downloadable ContentDownload PDF
This dissertation consists of two essays on household welfare in rural China as it has evolved during the post-reform era. The first essay presents evidence on intergenerational educational and occupational mobility in Rural China from 1988 to 2002 using data from two rounds of the Chinese Household Income Project. To understand whether the estimated intergenerational persistence is driven solely by unobserved heterogeneity, I implement biprobit sensitivity analysis (Altonji et al. (2005)) and heteroskedasticity based identification approach of Klein and Vella (2009). The empirical results show that there have been dramatic improvements in occupational mobility from agriculture to non-farm occupations. While there was significant persistence in occupational choices across generations in 1988, this occupational persistence had virtually disappeared by 2002. In contrast, intergenerational mobility in educational attainment remained largely unchanged for daughters and deteriorated significantly for sons. There is strong evidence of a causal effect of parental education on a son's schooling in 2002. I provide some possible explanations for the dramatic divergence between occupational and educational mobility in rural China from 1988 to 2002. China has maintained "one-party rule" in its political structure during rapid economic expansion. In the second essay, I examine the magnitude of economic returns to political connections of rural households, where households are defined as having a political connection if a member is a village cadre or is a Communist Party member. I examine whether returns to political connection changed over from 1988 to 2002 and whether the sources of returns derived from increases in agricultural or from non-agricultural income. The empirical analysis uses a recently developed econometric approach by Klein and Vella (2009) that corrects for unobserved variables affecting both household income and political status. I find that measurement error causes a considerable downward bias in the OLS estimates. When using IV estimation with the Klein and Vella (2009) approach, the resulting estimates for the returns to political connection are all considerably higher than those from the OLS. The empirical analysis finds that political connections in rural China yielded increased returns as a market economy expanded, which contradicts the "Market Transition" theory. Although the returns of political connection were positive and statistically significant in both years, the source of political power differed in 1988 and 2002. In 1988, the return to political connection came from its benefit to agricultural production; in 2002 political connection led to more profitable off-farm opportunities.