This paper takes a ‘beyond the mean’ perspective on physical and mental health differences between natives and immigrants and among immigrants themselves. We test the ‘healthy immigrant effect’ (HIE) and assess its deterioration, focusing on the evolution of the entire health distributions over time. Indeed, mean differences can have very different consequences in terms of health care costs and health inequalities, according to the underlying differences at the top and at the bottom of the health distribution. Using unconditional quantile regressions on data from the Italian Health Condition Survey, we find a HIE for both physical and mental health, which is mainly due to large differences in the lowest quartiles. Detailed decompositions show that observed characteristics (such as age, gender, and occupation) are associated with better health for both natives and long-stay immigrants compared to short-stay immigrants. At the bottom of both physical and mental health distributions, these gains are more than offset by the negative impact of some unobserved characteristics. Our results point towards the need of improving the data collection on health determinants, especially among immigrants, in order to uncover what is behind the unobserved component.

JEL Codes: I14, C21, J15, O15.

Keywords: Immigration; health; unconditional quantile regression; decomposition analysis; Italy