ABSTRACT
This paper presents an empirical analysis of return migrants’ entrepreneurial decisions after returning home using individual survey data in Jiangxi Province. We apply an extended hurdle model to investigate how social-economic factors, in particular, social network, local government support and previous migration experience may play different roles in affecting return migrants’ entrepreneurial decisions in different stages. A standard probit model is incorporated in the first stage to separate potential return migrant entrepreneurs from those who are not willing to consider establishing enterprises. In the second stage, a lognormal hurdle model is used to identify factors affecting return migrants’ entrepreneurial establishment and investment amount decisions. Results from the extended hurdle model suggest that government policy support is the most important factor for successfully attracting return migrant entrepreneurs back in Jiangxi Province. In addition, previous migration experience and household attributes are crucial and play different roles in affecting both return migrants’ willingness to consider entrepreneurship decisions and entrepreneurial investment decisions. On the other hand, individual characteristics are shown to affect only return migrants’ entrepreneurial investment decisions but not willingness to consider entrepreneurship decisions.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 This period is purposely selected to cover as many return migrants as possible in this snapshot survey. Migrants are more likely to be home before and after the Chinese Lunar New Year. They tend to stay home until Lunar 15th (the Lantern Festival) before heading to a migration destination. Ideally, a two-wave survey with time lag should be implemented to investigate the sequential entrepreneurial decisions of return migrants. It is yet practically infeasible as return migrants not self-employed are likely to be out-migrated again afterwards and difficult to track for follow-ups. An underlying assumption of this snapshot survey method is that return migrants’ entrepreneurial decisions reach a stable and equilibrium status and our sample is perfectly random and representative. Therefore, even a snapshot survey still provide sufficient information to identify migrant’s decision to start new business.
2 As shown in , the average number of migrants per household in our sample is 1.5, indicating on average one household has one to two persons out-migrated.
3 In this sense, we do not distinguish one-time migrants and repeated migrants. Most return migrants do not have continuous contracts with previous employers upon return. It could be the case that a temporary return migrant who is willing to consider entrepreneurship stops out-migration if the attempt of starting a new business were successful. However, we do control for the status of temporary return in our regression analysis. In addition, as a robustness check, regression analysis was replicated using a subsample of ‘intended’ permanent return migrants only by excluding observations who reported ‘temporarily come back to visit family and relatives.’ Results are not significantly altered and available from authors upon request.