Household Composition, Equivalence Scales and the Reliability of Income Distributions: Some Evidence for Indigenous and Other Australians* BOYD H. HUNTER  Centre for Aboriginal Economic Policy Research, Australian National University, Canberra, Australia STEVEN KENNEDY AND DANIEL SMITH Australian Bureau of Statistics, Canberra, Australia Indigenous families experience substantial and multiple forms of economic burden arising from the size and structure of their families and households. Indigenous households are more likely to have more than one family in residence than other Australian households and are more likely to be multigenerational with older Indigenous people living with younger people in extended family households. This paper seeks to characterise the economies of household size in Indigenous and other Australian households using equivalence scales that cover the range of feasible values and 1995 National Health Survey data. I Background Poverty and inequality studies almost always use an equivalence scale to adjust raw income to account for the cost of maintaining households and families. These costs are said to vary with household size and composition, and sometimes the number of employed in the household and other household characteristics. Unfortunately, variations in the assumptions about the relevant costs, and the relative complexity of the trans- formations involved, mean that measures of equivalent income are difficult to compare directly. For example, different groups may be classified as poor depending on which equivalent income is used. The purpose of this paper is to explore the intricacies of equivalent income cal- culations and to identify whether a particular group, Indigenous persons, are being dispropor- tionately re-ranked by several widely used meas- ures. The paper also examines to what extent Indigenous people move along the overall distri- bution of Australian income when different measures of equivalent income are applied. One of the most widely cited international studies of poverty claims that Ôequivalence scales haveingeneralnogreateffectontherankorderof measured inequality across countries as long as averagefamilysizeisnotextremelylargeÕ (Buhman et al. 1988).Astherationaleforchoosingaspecific scaleisrathervague,theimportanceoftestingthe sensitivity of income inequality estimates to the choiceofequivalencescalehaslongbeenacknow- ledged – especially where there are substantial differencesinfamilysizeandcomposition(Coulter etal. 1992;DeVos&Zaidi1997;Aaberge&Melby 1998; Lancaster & Ray 1998). The qualification *The authors would like to thank Jon Altman, Bruce Bradbury, Matthew Gray, Peter Saunders, and two anonymous referees for comments on an earlier version. The views expressed in this paper are those of the authorsanddonotnecessarilyrepresenttheviewsofthe Australian Bureau of Statistics. Where quoted or used, they should be clearly attributed to the authors.  Correspondence: Boyd Hunter, Centre for Aboriginal Economic Policy Research, The Australian National University, ACT, 0200, Australia. Tel.: +61 2 6125 8207; Fax: +61 2 6125 2789. Email: Boyd. Hunter@anu.edu.au 70 Ó 2003. The Economic Society of Australia. ISSN 0013–0249. THE ECONOMIC RECORD, VOL. 79, NO. 244, MARCH, 2003, 70–83