In many low‐fertility countries, immigration is increasingly seen a solution to the twin problem of rural depopulation and skill shortages. In Australia, this takes the form of regional visa schemes that require both skilled and humanitarian migrants to reside initially in nonmetropolitan regions for a minimum of 2 years. In the absence of nationally representative longitudinal data, the efficacy of this policy is yet to be assessed. Applying survival analysis to novel administrative data from the Person Level Integrated Data Asset (PLIDA), this paper establishes the level and determinants of rural retention among immigrants who arrived between January and August 2011 on eight different visas and compares these to the Australian population to the end of 2019. Our results suggest that regional visa schemes are effective in attracting permanent skilled migrants but not in retaining them, even when controlling for socio‐demographic characteristics. Migrants on regional skilled visa and temporary skilled workers display a 40% 9‐year retention rate compared with over 50% for Australian and New Zealand citizens, permanent family, skilled and humanitarian migrants and 30% for students. In contrast, the low retention of temporary skilled migrants is largely the product of their younger and more educated profile. We identify a negative selection process by which immigrants with less‐education, lower incomes, or less English proficiency—including humanitarian migrants—are more likely to stay in nonmetropolitan regions. This outcome signifies a process of socio‐spatial polarisation and a segmented labour market. At a regional‐level, we find that regions with a diverse occupational mix and co‐ethnic networks are more likely to retain immigrants whereas those with high housing costs are significantly less likely. These results provide policy levers to boost rural retention.
