Trends in Economic Inequality over 27 Years of Economic Growth
Commissioner Jonathan Coppel delivered a speech to the Economic and Social Outlook Conference in Melbourne on 12 October 2018.
Download the speech
Read the speech
Thank you for that introduction Gary.
This session poses a fundamental question: “Are we becoming more unequal as a society?”
It is a question that draws diverse and competing views.
As Gary highlighted in his introductory remarks, some in the community would say, yes it has risen based on this or that indicator. Others might say no, it has actually fallen, based on a different metric.
And yet others will express a view solely on the basis of their perception of what is happening.
This is not a good place to begin a discussion on inequality and its social impacts.
For a serious discussion, it is first necessary to have a thorough and authoritative grip on where we stand today and how we got there.
This is precisely the contribution the PC’s recent report seeks to make and hence its title “Rising inequality? A stocktake of the evidence”.
What we have done is to bring together and take stock of the latest and most complete evidence measuring the level of and trends in inequality, poverty and disadvantage in Australia.
The report contains many insights – some of them quite different from popular perceptions.
So to set the scene for this afternoon’s discussion I would like to take you through a selection of these findings.
Let me begin with some of the context and motivation for our study.
As you know, Australia has clocked up an unprecedented 27‑year period of uninterrupted economic growth.
This has prompted many to ask how the benefits from growth have been shared.
This slide shows how sustained growth has delivered for the average Australian household, in every income decile, significantly improved living standards.
This is what distinguishes Australia from most other developed countries.
It is certainly in stark contrast with the United States and the United Kingdom, which experienced much weaker income growth.
But as we also know, income growth is not the same as inequality.
And with such patterns of income growth it is not clear from looking at this slide what the impact is on overall measures of inequality.
Not so with this slide, which informs our bottom line conclusion that inequality has risen slightly over recent decades, as measured by the Gini coefficient – a standard measure of inequality.
This is most evident with respect to the distribution of wealth and consumption.
The trend in income inequality is more contested territory.
The Gini coefficient derived from ABS data reveal a small upward tick. However, HILDA data shows no trend change in inequality since the early 2000s.
One could get easily distracted dissecting the reasons for these differences. But this would blur the bigger picture.
The key point is that just looking at the distribution of income in isolation gives you a lopsided view of inequality.
A few examples help to illustrate why:
- Many retirees live on low incomes, but have high wealth.
- There are also many young adults with higher consumption than income and little wealth.
- And income patterns alone do not capture the importance of in-kind transfers from government, such as health, education and government housing.
This is why our approach deliberately eschews the specific and often self-serving use of any one measure of inequality.
Another reason why commentary on income inequality is contested territory is because the data reveals no uniform trend. It varies over time – sometimes in unexpected ways.
We were surprised, for example, to discover that ABS data since the GFC shows a decline in income inequality.
In fact, most of the uptick in income inequality over the past 27 years occurred during the mining boom period when wages growth was strong.
It reminds us that economic growth is no guarantee against a widening disparity between rich and poor.
International comparisons provide yet another perspective on the level of inequality in Australia, and the scale of any increase.
This slide shows income inequality in Australia is close to the OECD average.
We also know that Australia’s wealth distribution is relatively equal compared to other countries.
Among 28 OECD countries, Australia ranks eighth most equal, as measured by the Gini coefficient of wealth.
The fact that inequality levels are so different among developed countries strongly hints at the scope for policies, institutions and political environments to shape inequality.
Indeed, a further clear message from the data is that Australia’s tax and transfer systems substantially reduce income inequality.
Relative to other OECD countries, Australia redistributes less income. But we do a much better job of targeting this redistribution to low income earners.
Overall, our progressive tax and highly targeted transfer system lower the summary measure of income inequality by about 30 per cent.
Government funded services, such as health, education and public housing have an additional equalising effect.
When these in-kind transfers are included in a more expansive measure of consumption, inequality is about 30 per cent lower again than that for disposable household income.
This is because people with low incomes receive the largest amount of in-kind transfers.
While there has been no material change in these equalising effects in the past 30 years, there is nothing inevitable about them either.
The inequality measures considered thus far give a snapshot of the distribution at a point in time.
While they show a widening gap between “rich” and “poor” that does not mean the rich and the poor households at the beginning and the end of the period are the same households.
The distinction is important, and is why our report also looks at economic mobility – the gauge of whether the rich always remain rich and the poor always poor.
It matters, because a society with a low level of mobility can limit the sources of economic growth and give rise to feelings of social exclusion.
Mobility prospects are also an important determinant of life satisfaction and wellbeing.
So how does economic mobility stack up in Australia? Well, there is good, “in between” and bad news.
The good news is life course mobility is high in Australia.
This slide shows that those who were in the top income decile at the beginning of the century are now distributed across all of the deciles.
And three quarters of the people in the top income decile today were not there fifteen years ago.
We also looked at the movement from the bottom and middle deciles and a similar picture emerges.
In fact, almost everyone moves across the income distribution through the course of their lives.
To give a few examples:
- Over a 16-year period, the average Australian was classified in five different income deciles.
- Only one per cent of people remained in the same income decile over the whole period.
- And 9 per cent of the Australian population spent time in both the top and the bottom income decile.
Life events, such as transitioning from education into work, career advancement, household formation, having children, divorce and retirement, underpin some of these observed trends in economic mobility. Policy design also plays a part.
The “in between” news refers to another type of mobility, known as intergenerational mobility.
It concerns the relationship between a person’s economic position and that of their parents, and is a proxy measure for equality of opportunity.
It is affected by the importance of factors such as family wealth, connections, education and inherited characteristics such as intelligence and personality traits.
Our report does not examine intergenerational mobility in any detail. Instead we cite estimates from the literature.
These estimates suggest intergenerational mobility in Australia is neither very high nor very low compared to other developed countries.
The literature also finds a positive correlation between income inequality and inequality of opportunities.
In other words, and contrary to the expectations of many, countries with a higher degree of income inequality are also characterised by less equality of opportunity.
This is yet another reason why the inequality story is nuanced.
And the bad news is we found life course mobility at the ends of the distribution to be ‘stickier’ than the middle.
By stickier I mean households in the top and bottom income deciles at the beginning of the period were the most likely to be in the same decile 15 years later.
This is of particular concern because low mobility at the bottom typically extends beyond earnings to other dimensions of wellbeing, such as educational attainment and health outcomes.
Moreover, it may be symptomatic of low equality of opportunity.
To be sure, some ‘stickiness’ at the ends of the distribution is to be expected for technical reasons, which I won’t get into.
But how much stickiness is too much?
Well, there is no consensus on how much matters, but international benchmarking provides a basis for comparisons.
Relative to the United States, for example, economic mobility in Australia is higher overall and less sticky at the tails. A similar pattern of income mobility holds across a broader range of OECD countries.
Our way of summing it up thus far is that many Australians experience economic disadvantage at some stage in their lives, but for most of us, it is temporary.
For this group, sustained economic growth and reliable access to employment will continue to be the source of new opportunities and better living standards.
But another group of Australians experience entrenched economic disadvantage; they are stuck at the bottom end of the mobility ladder and less able to embrace new opportunities.
Disadvantage is a multidimensional concept.
It can take the form of low economic resources (poverty), an inability to afford the basic essentials for an acceptable standard of living (material deprivation), or being unable to participate in the normal economic and social activities of their community (social exclusion).
We consider disadvantage through these three lenses.
This slide shows our estimates based on the relative poverty metric.
You can see there are major differences in the level between the various estimates.
But in terms of trends, and with the exception of private consumption poverty, they have basically fluctuated around their mean.
So despite 27 years of uninterrupted economic growth, unemployment stabilising at a lower level and a significant redistribution of income, about 9 per cent of Australians live on very low incomes.
This translates to about 2.2 million people.
Of course, this number is a bit misleading.
It includes many people for whom low incomes have an explanation that does not warrant a policy response, like students in part time jobs or start up business operators building a capital asset.
It also captures all spells of poverty, regardless of their duration.
This slide shows that short‑term spells in income poverty were by far the most common, with around 60 per cent of people exiting income poverty after one year, and another 18 per cent exiting after two years.
However, we also observed significant poverty recurrence.
Of all those who moved out of income poverty over the period, a little more than half later fell back into poverty.
Once you account for all these factors, the number of households that are experiencing protracted poverty shrinks, but still accounts for a significant proportion of the population.
We estimate about three per cent of Australians – roughly 700,000 people – have been in income poverty for at least the last four years.
People living in single‑parent families, the long term unemployed, people with disabilities or other long term health conditions are particularly likely to experience protracted income poverty.
For people in these circumstances, there is an elevated risk of economic disadvantage becoming entrenched, limiting their potential to seize economic opportunities or develop the skills with which to overcome these conditions.
These risks are particularly elevated for children living in jobless households, which is a group that has stood out among the multiple measures of inequality and disadvantage.
These are just a handful of the stories on inequality we drew from the data and the multitude of indicators.
Hopefully they dispel the popular perception that the benefits of growth are not being broadly shared.
But inevitably they beg the question: Can we do better and how?
Our report does not directly enter into debates about what types of policy responses might be appropriate.
It does, however, provide pointers on where we could do better.
One such area I have highlighted today relates to the small, yet significant number of Australians in entrenched disadvantage.
The issues with entrenched disadvantage are multiple and complex, for which there are few precedents.
We know a bit about the drivers in general terms. For example: chronic disease, mental illness, addictions, long term unemployment and poor educational outcomes. Or more likely some toxic cocktail of these conditions.
We also know from evaluations of past efforts and analyses that policy designs need to be much more adaptable and targeted to individual disadvantage. This is what the previous Chair of the PC referred to as “handmade” policy.
But there is still a lot we don’t know, even though we increasingly have the data to be able to improve on this work.
In this regard one relevant dimension relates to the Government’s announcement of a Productivity Commission inquiry into mental health.
I am sure that through our analysis and processes of consultation we will build on the current knowledge base to identify where we can do better by our fellow Australians and recommendations to governments on how to get there.
Rising inequality? A stocktake of the evidence
The purpose of this research paper is to contribute to an informed discussion in Australia by bringing together and taking stock of the latest and most complete evidence measuring the prevalence of, and trends in, inequality, economic mobility and disadvantage