Decomposing the Effects of Household Characteristics on Child Poverty in Ten Countries
Tsui-o Tai, University of California, Irvine
This study examines the effects of household characteristics on poverty risk for children and their household members across ten countries in the Luxembourg Income Survey. Robust-cluster logistic analysis relates the risk of poverty to state welfare regime type and household characteristics. Persons in households with children are significantly less likely to be poor in social democratic countries but more likely to be poor in liberal countries compared with East Asia, an exemplar of limited social welfare programs. Controlling for household composition reduces the difference in poverty risk between liberal and East Asian countries. Fairlie decomposition analysis shows the contribution of the less prevalence of single motherhood to the relatively low poverty risk in East Asian countries.
Presented in Poster Session 3