Health-related disabilities and matching of spouses: Analysis of Swedish population data

Abstract

Recent research concerning the incidence of reported work-limiting disabilities in the married population indicates a degree of interdependence between spouses' disabilities. This pattern is consistent with several hypotheses. Spouses tend to share many lifestyle traits that might lead to common health outcomes. Alternatively, their joint reports might reflect a shared preference for income benefits or workplace accommodations available to disabled individuals. Another possibility is that disabled individuals tend to be matched in the process of marital formation. This paper investigates the latter hypothesis. Taking advantage of a unique data set from the Swedish population, we select a sample of recently married couples and trace them back in time to their single years. Our analysis indicates nonrandom matching on the basis of disability status. After controlling for observed traits such as age and education, we find a residual correlation between future spouses that is positive and strongly significant. The magnitude of the correlation is within the range of residual correlations obtained from other studies that address marital matching in the contexts of education and earnings.