The prevalence of low income for children, especially for children in lone-parent families, varies considerably across countries. This paper considers five sets of hypotheses that may be relevant to the cross-national variability of child poverty. The tentative conclusion from this analysis in 20 countries is that reducing child poverty, and in lone-parent families in particular, requires several approaches. Provisions that would discourage teenage childbearing would have their importance, as would opportunities for lone mothers to work. More important is the generosity of social expenditure applying to individuals and especially to families. The present analysis also shows the advantages of encouraging joint custody, along with special provisions for lone parents, and child support through advance maintenance payments.Keywords: child poverty, lone parent, social assistance, custodyThe prevalence of low income among children, especially for children living in lone-parent families, varies considerably across countries. Many authors have pondered over the differences showing that the rates of low income for children in lone-parent families are under 10 percent in countries like Belgium, Finland, and Sweden, while the rates are over 40 percent in Australia, Austria, Canada, Germany, United Kingdom, and the United States. Several questions have been analysed in seeking to explain these differences: the extent to which mothers are considered 3 to be employable (Baker, 1996;Gornick et al., 1996;Millar, 1996;Lewis, 1993), the relative economic status of women (Kilkey and Bradshaw, 1999;Solera, 1998;Bianchi, 1996), the generosity of social assistance provisions (Phipps, 1999(Phipps, , 2001Kenworthy, 1998;Smeeding et al., 1997;Baker, 1996), or the extent of compliance to support obligations from non-resident parents (Kunz et al., 2001;Garfinkel et al., 1998). Many studies focus on lone motherhood rather than on children in lone-parent families (e.g. Kiernan et al., 1998).The purpose of this paper is to compare the relative importance of various possible factors in understanding the variation in these child poverty outcomes. While other studies make reference to the demographics (Nichols-Casebolt and Krysik, 1995;Rainwater and Smeeding, 1995;Gornick and Pavetti, 1990), we will systematically consider the demographics associated with the prevalence of lone parenthood, the teenage fertility rate or the percent male in loneparent families. After considering the average per capita income, the next second set of variables considers the generosity of social transfers, and the transfers toward families in particular. The next set of considerations relates to women's labour market status, particularly for lone parents.These explanatory factors are compared to three specific policy orientations: the extent of state involvement in ensuring child support, the extent of differential state support for lone parents, and the extent of joint custody.
Theoretical context
4The broad theoretical context is that of the relative prio...