Determinants of low work intensity in households with disabled family members. Empirical analysis for the EU-15

Despite the employment gains in many EU countries in the decade prior to the crisis of 2008, the integration of disabled individuals in the labor market has often proved a difficult task. As already shown in other studies, the consequences of this reach further than the individual level with the employment status of other family members often being negatively affected when someone leaves the labor market because of health-related issues. In addition to this, the social expenditure patterns of European countries in terms of facilitating the need for care (in the area of incapacity) with paid work show very pronounced differences. In this paper we therefore (1) establish the exact correlation between the individual employment of disabled individuals and the corresponding household work intensity and we (2) investigate whether and to what extent the observed household joblessness rate deviates from a random distribution of employment by means of a polarization index. We take into account distinct socio-demographic characteristics of both disabled and non-disabled family members when evaluating the observed polarization to finally (3) correctly evaluate whether welfare states that explicitly take the issue of care into consideration in their social expenditure pattern (in the areas of residential care and home-help services) have corresponding household work intensity levels. We find that the lower employment opportunities of disabled individuals translate into high proportions of jobless households. Positive polarization indices point to more being at play than solely the lower employment level of the disabled household members. When taking age and educational clustering into account, a substantial reduction in the observed polarization index is achieved. For six countries within the sample, a pattern is found between the residual polarization and social expenditures for the relevant domain. Bivariate association, however, fails to underscore the hypothesis that high-spending countries display low polarization counts.