How does income inequality compare to more traditional measures of disadvantage, like poverty?
In a recent article for Inside Story, John Quiggin wrote that the verdict is still out on whether equality and social well-being are correlated. Some research shows that they are – Richard Wilkinson and Kate Pickett’s book, for example – while other research has shown that the correlation is weak once other factors are taken into account. Because of the uncertainty surrounding these results, John Quiggin’s advice is to promote equality and social solidarity in as many dimensions as possible.
Given this debate on inequality, and the links to social well-being, people may be asking three related questions. What is income inequality? How does it compare to more traditional measures of disadvantage, like poverty? And how does Australia compare over time and with other countries in measures of income inequality?
When a researcher in Australia talks about poverty, they are usually talking about the number, or proportion, of people who earn below a set poverty line. This poverty line can either be relative (set in relation to everyone else’s incomes) or absolute (a fixed dollar amount, possibly calculated by looking at what someone needs to survive). A relative poverty line will increase as everyone’s income increases; a fixed poverty line is usually increased in line with inflation.
A relative poverty line could be set, for example, at half the median Australian income. In 2005–06 (the latest ABS data available) this was $281.50. An international example of a fixed poverty line is the $1 per day used by the World Bank until 2008, when it was increased to $1.25 per day. This is the amount that was required for people to meet their basic needs for adequate food, water, shelter, clothing, sanitation, health care and education. While this amount is appropriate for third world countries, in Australia, the amount required to meet these basic needs will naturally be much higher because prices of these basic necessities are higher. An absolute poverty line was calculated in Australia for the Henderson poverty inquiry in 1973. It was $62.70, which was the disposable income required to support the basic needs of a family of two adults and two dependant children at the time. This poverty line has been updated regularly by the Melbourne Institute according to increases in average incomes; for a single employed person it was $391.85 per week (including housing costs) in March 2009. If we inflated the poverty line from 2005–06 using the same inflation factors as applied to the Henderson poverty line, we get a figure of $320.77.
Inequality is quite different from poverty. It is a measure of how rich the rich are compared to how poor the poor are. The easiest way to calculate inequality is to look at how much the very rich earn, and divide this by how much the very poor earn. This can be measured using what is called a P90/P10 ratio, in which the income that the top 10 per cent of the population earns is divided by the income that the bottom 10 per cent of income earners receives. In Australia, this ratio was 3.92 in 2005–06, which means those at the bottom of the top income earner group earn around four times more than the top of the bottom income earner group. It has been at this level since 1999–2000, whereas from 1994–95 to 1999–2000 it was lower, suggesting lower income inequality (or greater income equality) in that period.
An alternative measure to the P90/P10 ratio for inequality is the Gini coefficient. This is a measure of inequality that ranges between 0 (when all incomes are equal) and 1 (when one household receives all the income and every other household receives none). The closer the gini coefficient is to 0, the more even the distribution of income.
In Australia in 2005–06, the gini coefficient was 0.307, a figure that has not changed much since 1997–98, but was lower (P90/P10 ratio) in the mid 90s, suggesting greater income inequality since 1997–98.
Internationally, the 2009 OECD Factbook ranks Australia sixteen out of thirty OECD countries, or about half way, using data from 2003–04. The country with the least income inequality was Denmark (0.232), and the country with the greatest income inequality was Mexico (0.474).
So how is income inequality distributed geographically across Australia? In recent work at the National Centre for Social and Economic Modelling (NATSEM), Riyana Miranti, Rebecca Cassells, Yogi Vidyattama and Justine McNamara calculated measures of income inequality using the gini coefficient for small areas across New South Wales and Victoria. These measures of income inequality give some idea as to which small areas in Australia suffer from greater income inequality.
The report finds that remote areas in New South Wales like Central Darling, Bourke and Brewarrina suffered greater income inequality. In Sydney, the western suburbs and some inner-city suburbs like Randwick and Waverley suffered greater income inequality when compared to the rest of the state. In Victoria, remote areas near the South Australian border, like Wimmera and Hindmarsh, had higher income inequality. In Melbourne, many inner city areas like Melbourne City and Stonnington had higher inequality.
It is in these areas that we would expect lower wellbeing, according to the analysis in Wilkinson and Pickett’s book. And in many cases, that appears to be true. When we look at some variables representing disadvantage in an area (proportion of indigenous people, unemployment rates, occupation, education, poverty rates and housing tenure), the areas with high income inequality also seem to be areas of high disadvantage. In other words, the areas with the highest inequality tended to have a higher proportion of Indigenous people, a higher proportion of public housing and a higher proportion of people in poverty. But we can also find areas where high income inequality is associated with low disadvantage, so a higher proportion of managers and professionals and a higher proportion of people with a university degree.
So the link between income inequality and disadvantage is not clear. This is probably not surprising – when we go back to the definition of income inequality, it needs both rich and poor in an area for an area to be unequal. So, by their very nature, areas with high income inequality can also have high income, more advantaged households, and this would offset the low income, disadvantaged households.
Which is why income inequality is such a difficult concept to grasp, compared to other measures of disadvantage like poverty rates and indexes of disadvantage. And, possibly, why there are different results coming out of different studies mentioned by John Quiggin; because the measure is complex and requires both high income households and low income households in an area to show high inequality.
Robert Tanton is Principal Research Fellow at the National Centre for Social and Economic Modelling (NATSEM)
Photo: Andrew Jeffrey