This paper relaxes the single factor model of intergenerational educational mobility standard in the literature, and develops a research design to study the effects of parents' education and occupation on children's schooling. We use survey data from rural China that cover three generations and are not subject to coresidency bias. The evidence from recently developed matching and propensity score weighted estimators shows that the mean effects of parents education from the standard model miss substantial heterogeneity. Within the low education subsample, a son (girl) attains about 0.80 (0.60) years of additional schooling when born into a non-farm household compared to a farm household, and among the farming households, a child gains a one year of schooling when at least one parent has more than primary schooling. Having nonfarm parents, however, does not confer any advantages over the farmer parents if the farmers are relatively more educated, even though nonfarm households have significantly higher income. This suggests that income plays a secondary role to parental education. Estimates of cross-partial effects without imposing functional form show little evidence of complementarity between parental education and non-farm occupation. The role of family background remains stable across generations for girls, but for boys, family background has become more important after the market reform.Key Words: Educational Mobility, Inequality, Rural China, Nonfarm, Education and Occupation, Family Background, Heterogeneity, Complementarity, Market Reform, Gender Gap JEL Codes: O12, J621 We would like to thank Li Shi, Zhan Peng, and Song Jin for help with the CHIP survey data, and Matthew Lindquist and Forhad Shilpi for helpful comments and/or discussions.
(1) IntroductionIncreasing inequality despite sustained economic growth and poverty reduction in last few decades has become a central policy issue in both developed and developing countries (Stiglitz (2012), Rajan (2010), Picketty (2014)). Inequality and its negative effects have also been high on the agenda of the Chinese policymakers in recent years. 2 There is a broad consensus among the policy makers and academic economists that education is one of the most important policy instruments to make the playing field level for children from the poor socio-economic background, and moderate the inequality generated by market reform. It is thus important to understand potential disadvantages faced by children from a poor socio-economic background in attaining the education and skills required in an increasingly globalized and skill-based labor market.The economics literature on intergenerational educational mobility has focused on parental education as the relevant indicator of family background for understanding intergenerational linkages (see Bjorklund and Salvanes (2011) for a recent survey, and on China, see, among others, Knight et al. (2013), Sato and Li (2007), Emran and Sun (2011)). This emphasis on parental education is eminently appropriate when the goal is...