1 Peer Migration in China*

This paper uses a new instrumental variable (IV) strategy to identify the causal effects of peer migration. The existing literature has long recognized the importance of social interactions in migration, but it is difficult to separate causal peer effects from omitted variables. Since 1984, rural Chinese families are allowed to have a second child if the firstborn is a girl. This policy, plus the nature-determined sex of the firstborn, generates exogenous variations in family size and gender composition of children, which in turn affects rural adults ’ migration decision. Using the sex of peers ’ firstborns and whether peers ’ first and second births are multiples as instruments, we find whether one migrates or not is strongly dependent on how many peers migrate in the same village. The peer effects are most concentrated by same gender, similar age, similar education, same surname, same destination and same occupation. These cluster patterns suggest that peers in the same village play an important role in reducing migration cost and information uncertainty in the destination