Do foreign nationals really live longer than natives in Germany? Results based on 2011 census data

Felix zur Nieden, Federal Statistical Office, Germany
Bettina Sommer, Federal Statistical Office, Germany

Differences in life expectancies of foreign nationals and natives in Germany have been the subject of extensive research during the past decades. Based on intercensal population estimates, researchers found large differences in life expectancies in favor of foreign nationals. These findings have been called into question due to the assumption that especially the foreign population was gradually overestimated with greater time since the last census. 2011 census data now gives the opportunity to calculate complete life tables for both population subgroups while eliminating error in the intercensal population updates as a potential explanatory factor for differences in life expectancy. The present study produces these estimates by replicating the methodology of the official general life table 2010/12 for Germany. Results show that the differences in life expectancy at birth are reduced for males from 5.2 to 2.9 years and for females from 5.1 to 2.1 years, when estimates based on census results are used in contrast to updated intercensal population estimates. A closer look into the age-specific mortality differences of natives and foreign nationals in Germany reveals that children with non-German citizenship have higher probabilities of dying than German children. The balance is turning at about age 16. Before age 25 the differences are not significant (except for infant mortality) – above age 40 the differences are constant in relative terms and significant up until the highest ages. Despite the methodological constraints implied by using period and not cohort data, these findings support the famous selection hypothesis of a strong healthy-migrant-effect. In conclusion, the census based findings support earlier findings that foreign nationals do on average live longer than natives in Germany, although the difference between both groups is smaller than that found based on intercensal population updates.

  See extended abstract

Presented in Session 121: Ethnicity, migration, and mortality