Despite the well-known limitations of statistical measurement, there is still a vital need to document the extent of overall poverty and its incidence among different groups in Australia. Without such evidence, it is impossible to establish how successful social and economic policies have been in alleviating poverty, or where additional resources need to be directed. The absence of such evidence is all the more concerning given the reliance placed in Australia on the efficient targeting of resources through means-testing and other policies (e.g. mutual obligation) designed to ensure that the assistance provided is cost-effective. At the very least, one would expect trends in poverty to be monitored regularly using a range of indicators.
Community concern over poverty remains at a high level. Evidence from a large social survey conducted in the middle of 2006 by the Social Policy Research Centre (SPRC) indicates that when asked how big a problem poverty is in Australia today, almost one-third (32.1 per cent) thought it was 'a big problem' and almost everyone else thought that it was 'a problem for some'. Less then five per cent thought that it was ‘a small problem’ and around one per cent thought it was ‘not a problem at all’.
In response to this situation, the was commissioned to undertake a detailed analysis of the extent of poverty in Australia in 2003-04 (the latest year for which this is possible with existing data), describe who is most affected by poverty, assess the robustness of the findings, to examine trends in poverty levels over the preceding decade, and update a limited range of poverty estimates to 2005-06 using recently released data.
Although research on income poverty using a poverty line approach can reveal only a single dimension of what is a complex and multi-faceted issue, it remains important – particularly in an era where 'money matters' more than ever before. Poverty research thus provides the statistical foundation on which other methods of measuring deprivation and social exclusion can build. Those other approaches may eventually replace the conventional approach with its emphasis on comparing incomes with a poverty line, but even those countries that are moving in this direction (e.g. the UK and Ireland) still include poverty rates measured using a range of poverty lines as part of the battery of indicators used to identify who is poor. The approach taken in this report examines the sensitivity of poverty estimates to variations in the methods used to produce them. The basic idea is to check the robustness of estimated poverty rates by varying some of the underlying assumptions in order to establish their quantitative importance. Without such analysis, it is difficult to be confident about the findings from any single study. With them it is possible to check the robustness of the estimates and accumulate a body of knowledge that adds to our understanding in ways that can generate new insights and implications for policy.
