Association for Public Policy Analysis & Management

Parental rural-to-urban migration has theoretically ambiguous effects on the "left-behind" children who stay in rural area because different mechanisms. The net impact is also often hard to tell because of the selection and therefore endogeneity issue of migration decision due to unobservable characteristics particularly. This paper uses a data set from rural China called Gansu Survey of Children and Family. We overcome the issue of endogeneity by applying Instrumental Variable approach, which provides causal interpretation. Moreover, the unique features of our data set allow us to try to understand the different mechanisms such as income effect, time input and parental believes. Better understanding of both the causal impact and the underlying mechanisms will shed light on our understanding of child development and intergenerational mobility in the development context.