Deprived housing conditions have long been recognized as a source of poor health. Nevertheless, there is scant empirical evidence of a causal relationship between housing and health. The literature identifies two different pathways by which housing deprivation affects health, namely, neighborhood effects and the effects of the individual dwelling unit. However, a joint examination of both pathways is absent from the literature. Moreover, endogeneity is a substantial concern in analyses of these two problems. Thus far, studies addressing endogeneity concerns have done so through experimental design or instrumental variables. While the first approach suffers from problems of external validity, we demonstrate the substantial diffculty in identifying robust and reliable instruments for the latter. Consequently, we adopt an alternative strategy to identify the causal effects of housing on health in 21 European countries by estimating fixed-effect models and considering both sources of endogeneity, neighborhoods and dwellings. Furthermore, using the panel dimension of our data, we reveal the accumulation dynamics of poor housing conditions. Our results indicate that living in poor housing is the chief socioeconomic determinant of health over the four-year observation period and that bad housing is a decisive, causal transmission pathway by which socioeconomic status affects health.
Housing Health Europe EU-SILC data Fixed-effects model