Publication on internal labour migration

In their article Local attributes and migration balance - evidence for different age and skill groups from a machine learning approach, Moritz Meister, Annekatrin Niebuhr (both Institute for Employment Research), Jan Cornelius Peters (Thünen Institute) and Johannes Stiller (formerly Thünen Institute, now Expert Commission on Research and Innovation) show the extent to which regional migration balances resulting from internal labour migration are related to regional attributes. In doing so, they also analyse the heterogeneity of the correlations between age and qualification groups.

Their econometric model can be understood as an aggregate formulation of a two-region random utility model. To account for the fact that a variety of factors are potential determinants of internal migration and that these characteristics are often interrelated, the authors apply machine learning techniques (Lasso, complete-subset regression). The results indicate that, from the perspective of many workers, regions that are located outside but close to large urban centres are particularly attractive. However, the strength of this correlation varies greatly depending on the age and qualification group. Furthermore, labour market conditions and some amenities correlate significantly with regional net migration. The former and especially the availability of apprenticeships are, as expected, particularly important for young workers.

Meister M, Niebuhr A, Peters JC, Stiller J (2023) Local attributes and migration balance – evidence for different age and skill groups from a machine learning approach. Reg Sci Policy Pract. https://doi.org/10.1111/rsp3.12652