Data sources

Both the poverty and inequality estimates used in this website have been sourced from the basic Confidentialised Unit Record File (CURF) data from the Survey of Income and Housing  by the Australian Bureau of Statistics (ABS).

This survey is run by the ABS every two years, and every six years is integrated with the Household Expenditure Survey.  The purpose of the survey is to collect information on where people get their income from, what people’s households are worth, their housing status, household characteristics and more.

The survey asks a sample of the population across all areas of the country questions from the same questionnaire. In the most recent survey, around 14,000 dwellings were contacted..

Summary results from these surveys are published by the ABS in their Household Income and Wealth  reports every two years.

All the estimates reported on this website are adjusted to 2017-18 prices using the Consumer Price Index (CPI).

International comparisons are made using data from the Organisation for Economic Cooperation and Development (OECD)’s statistical database.

Changes in income definitions

During the period of the Poverty and Inequality in Australia series, the ABS has introduced a series of measurement changes that improve the quality of the income data that they collect. The most important changes were introduced in 2007-08, and include changes to the way in which income is recorded to include non-cash benefits, bonuses, termination payments and irregular overtime payments. These changes resulted in an $85 increase in the average weekly gross household income, and affected 43% of all households.

The changes also resulted in an increase in inequality measured by the Gini Coefficient, and increased the poverty line (as it is based on median income) and hence the poverty rate.

The 2018 Poverty in Australia report uses the 2007-08 income measure, as this is now the most comprehensive measure of income. It was not possible to use this measure when examining the trend in poverty back to 2003-04, because the data that uses the new measure was only collected from 2007-08 onwards. However, there is a consistent measure that gives us the capacity to measure poverty from 2005-06 to 2015-16, and this forms the basis of the trend analysis. The poverty graphs reflect the impact of using the pre- and post-2007 income measure on poverty rates. As noted above, poverty rates are higher using the post-2007 measure, because median income is higher.


Within the poverty data, two population groups are excluded. These are:

  1. All households that have reported either no income or less than no income (about 1% of the population); and
  2. All self-employed households (about 15% of the population).

Self-employed households have been excluded because it is often difficult to distinguish between personal and business incomes, and especially to work out which households are able tto improve their living standards by drawing down business assets.

However, when estimating the total number of people in poverty in Australia, an assumption has been made that the self-employed, and those with negative incomes have the same poverty rate as the non-self-employed population. This adjustment is not carried out when measuring the different population groups where the estimates of the number of people in poverty include the non-self-employed population only.

In the inequality results, only households with zero or negative incomes are excluded. Households reliant on self-employed income are included in the data.

Poverty calculations

Two methods are used to calculate poverty on this website. The first is based on disposable income (after-tax income), adjusted for family size using the modified OECD equivalence scale. The poverty line is defined as either 50 or 60% of the median income, before housing costs.

The second method is similar, except instead of being based on disposable income, it is based on disposable income minus housing costs. There is a wide variety of housing costs throughout Australia. People who own their own homes and have paid off their mortgage, for instance, have much lower housing costs than those still paying mortgages or rent. This is the after housing poverty line, used for most of the results shown on the website.

Household types

Household types are defined on the basis of age, labour force status or family status. Where necessary, they are based on the status of the person nominated as the reference person in each household.

By splitting the analysis up by household types, we can provide a more in-depth account of the profile of poverty and inequality; and help identify the factors contributing to overall poverty and inequality by showing variations across income and wealth groups.

Read our detailed methodology papers

Full poverty research methodology paper

Full inequality research methodology paper