A fixed-effects multilevel analysis of how community family structure affects individual mortality in Norway

Using register data for the entire Norwegian population aged 50-89 in 1980-1999, in which there are ¾ million deaths, it is estimated how the proportions who are divorced or never-married in the municipality affect all-cause mortality, net of individual marital status. The data include individual histories of changes in marital status and places of residence, and provide a rare opportunity to enter municipality fixed-effects, capturing the time-invariant unobserved factors at that level, into the models. The positive health externality of marriage that has been suggested in the literature is supported by some of the estimates for women, while other estimates, and especially those for men, point in the opposite direction. These findings may indicate that a high level of social cohesion is not as beneficial as often claimed, at least not for both sexes, that marriage perhaps undermines rather than strengthens social cohesion, or that other mechanisms are involved, for example related to people’s perceptions of their health relative to that of others. Estimates from models without such municipality fixed-effects are markedly different, but these also shed doubt on the notion that a high proportion not married generally increases individual mortality.