Tianning Zhu, London School of Economics
Being networked to migrants has been argued to affect (1) people's propensity to migrate and (2) migrants' destination choices. This project explores these issues using Chinese genealogical data, which provides evidence on the structure of extended families unavailable in other sources and in which migrants can be identified. The pilot project uses reconstructed family trees for 8 branches in one genealogy, covering 2,071 male members born between the 17th and 20th centuries. The clan recorded in this genealogy is based in Guangdong, one of the major migrant-sending provinces in China. Many of its members settled elsewhere during this period. Based on one's position on the family tree, I construct a connectedness score measuring how close he was to all the earlier male migrants in his paternal kinship network. I then use logistic regressions to see if this network measure matters in people's decisions to move, controlling for personal, family, and branch characteristics. The baseline result shows that this measure is significantly positively related to one's odds of migrating. I also break down the connectedness score by migrant's destination to explore how the networks in the destination affected the migrants' destination choices. I found that connectedness with Malaysian migrants improves migrants' odds of moving to Malaysia but reduces the odds of non-Malaysian migrants going to other destinations outside the province. This is consistent with the chain migration theory, which suggests that new migrants follow the previous migrants to a particular destination. On the other hand, non-Malaysian migrants with a migrant network in Malaysia were more attracted to short-distance destinations. This suggests that having a migrant network, regardless of its location, improves one's migration propensity. However, whether to go on a long-distance trip might depend on whether one has a destination-specific network.
No extended abstract or paper available
Presented in Session 178. Migrant Networks and Systems