Interdependent hazards, local interactions and the return decision of recent migrants

We propose a new statistical model of spatially interdependent hazards in order to examine whether interactions at the level of the neighbourhood are present. To address the endogeneity of location choices, we also implement a weighting scheme that combines insights from inverse propensity score weighting and Bartik-shift-share instrumentation. The good performance of our approach is demonstrated in a Monte Carlo study. In our empirical application a unique large administrative dataset of recent Turkish labour immigrants to The Netherlands is used. We find a positive spatial interaction parameter and show that the local spatial multipliers are substantial.