On the Impact of Separate Fertility and Mortality Assumptions for Native and Migrant Subpopulations in Population Projections

Christina Bohk, University of Rostock

Several demographic studies have shown that fertility and mortality often differ between natives and migrants. Hence, to accurately project a population on a macro level, a population should be divided into such subpopulations that do not differ substantially in their demographic behavior. However, common population projections often neglect this aspect due to the lack of data. Yet, if data are available, a single projection of natives, immigrants, emigrants, and their descendant generations with specific fertility and mortality assumptions reduces the effect of demographic heterogeneity and increases a population projection's accuracy. The impact of separate assumptions for native and migrant subpopulations is evaluated by comparing four projections, based on real-world data for Germany, using the probabilistic population projection model (PPPM). It turns out that assuming equivalent mortality and fertility for natives and migrants induces a considerable error in long-term projections, for both gross and net migration.