Indigenous households are more likely to be multi-generational and have several families in residence than other Australian households. Equivalence scales attempt to control for family size and composition and the relative costs of maintaining various families. We use the 1995 National Health Survey to examine how variations in the assumptions underlying equivalence scales, such as household composition and economies of size, affect poverty measures for Indigenous and non-Indigenous Australians. Our results show that the choice of equivalence scale affects the level and composition of poverty. This is evident in that variations in the assumption about the costs of children can increase Indigenous poverty by a factor of two-and-a-half. We also examine how equivalence scale variations can induce large threshold effects, and demonstrate the influence of zero and negative incomes in our data set on the composition of poverty.
ISBN: 0 7315 4918 X
ISSN: 1442 3871