Ethnic networks in neighborhoods affect mental health: Evidence from a quasi-random assignment of applicants in the public social housing system.

This paper examines the impact of residence-based ethnic networks on mental health; such networks are defined as the concentration of residents from the same country of origin in a neighborhood. To estimate the effect, we utilize administrative registry data, together with data on quasi-random assignment of apartments to non-Western households with housing needs to various neighborhoods. After controlling for individual characteristics, time-invariant neighborhood characteristics, and general practitioners (GP) fixed effects, we find that a 1-percentage-point increase in the concentration of residence-based co-ethnics (RBCEs) increases the probability of being treated with psychiatric medications by 0.7-percentage point over a 5-year period after the assignment. With 19% of the population being treated with psychiatric medications the year before assignment, the result translates into an effect size of 3.7%. The results indicate that relatively high concentrations of co-ethnics treated with psychiatric medications increase the probability of being treated with psychiatric medications. The positive impact on treatment with psychiatric medication reflects an increase in the demand for these drugs when moving into a neighborhood with neighbors of the same ethnicity. If new residents are in good mental health condition when moving, these results suggest that moving into a neighborhood with a high co-ethic concentration worsens mental health status. However, as the population in this study is a vulnerable group an increase in treatment with psychiatric medications likely reflects that untreated mental health problems are treated, and the mental health status improved. The group of non-Western immigrants in this study differs significantly from the population in general, thus, results may not be generalized to all non-Western immigrants.