Poverty lines tangled: An influential report claiming that one in eight Australians lives in poverty is wrong

Our paper, Poor Arguments: A Response to the Smith Family Report on Poverty in Australia – questioning the claims that one in eight Australians is living in poverty today, and that poverty has risen in the last decade – attracted considerable media interest on Tuesday.  Yesterday, the Smith Family vigorously defended its claims.

So where do things stand now that the dust has started to settle?

There is clearly some common ground between us.  Everybody accepts, for example, that ‘absolute poverty’ is very rare in Australia.  Everybody also accepts that even the poorest people in Australia are better off today than they were ten years ago – the Smith Family’s own figures show that the poorest individuals increased their incomes by an average of $38 per week over this period in real terms (i.e. after allowing for inflation).

But wait a moment!  If the ‘poor’ became better off over the last ten years, how could the Smith Family conclude in its report that poverty had worsened?  The answer is that it was talking about ‘relative poverty’, and that the way it measured this depended less on what was happening at the bottom of the distribution than on what was happening at the top!

We spent a lot of time in our paper explaining what is wrong with this.   Unlike ‘absolute poverty’, ‘relative poverty’ is said to exist when people’s living standards fall a long way short of what is ‘normal’ in their society.  According to the Smith Family, this point is reached when people’s incomes fall below half the average income – and this is where they drew their ‘poverty line’.  Measured in this way, one in eight Australians is ‘poor’, and poverty increased in the 1990s.

There is, however, a fundamental problem with measuring poverty in this way (and it is a problem which many researchers in this field readily acknowledge).  If a small number of high earners increase their incomes, this will pull up the average income, thereby dragging the ‘poverty line’ up too.  The result is that more people will now fall under this line, even though they may well have got better off. This is exactly what happened in Australia over the last ten years.

To avoid this problem, most researchers nowadays define the ‘poverty line’ as half the median (i.e. the middle-ranking) income, rather than half the average.  Defined in this way, the number of Australians in ‘relative poverty’ comes out at one in 12 – not one in eight as the Smith Family claimed.  Furthermore, this proportion remained virtually unchanged throughout the 1990s.

Responding to all this yesterday, The Smith Family claimed that it calculated no fewer than twelve different poverty estimates, and that eleven of them showed that poverty had increased over the last decade.  But this again really is misleading.  Bear with us while we explain why.

Their original report actually identified six different ways of calculating a poverty line, each of which was repeated both before and after housing costs were taken into account.  Because the Smith Family report chose to look at figures before housing costs, we followed suit.  This leaves us with six estimates, not twelve.

Of these six measures, two were based on the well-known but badly-flawed ‘Henderson’ poverty line.  Both of these produce absurdly-inflated estimates suggesting that getting on to a quarter of the population is in poverty!  Wisely, the Smith Family rejected these measures, and so did we.  That leaves four.

Of these four, one – based on half-average incomes – was the one they chose to use in their report, and it shows a rise in the number of people under the poverty line.  The other three were all based on different versions of our preferred measure, based on half the median (middle-ranking) income.  In all three of these cases, the number of people under the poverty line hardly changed over this ten year period (there was a slight increase according to two of these measures and a decrease in the third – essentially, the trend line is flat).

Claiming that 11 out of 12 measures all point in the same direction is, therefore, misleading.  Some of these measures are effectively duplicates; others are useless; and of the four that are left, three indicate that nothing much changed over the last ten years.

Anyone trying to make sense of this debate should bear in mind three further points.

First, there is little doubt that the income figures used by The Smith Family are unreliable, particularly at the bottom end of the distribution.  We know from many previous studies that those claiming to have the lowest incomes are often under-reporting how much they actually receive.  Many of them regularly spend more than they say they earn, and it may actually make more sense to gauge people’s living standards by their expenditure rather than their reported incomes.

Secondly, the Smith Family calculations completely ignore the value of government services like Medicare and public schooling, yet these are designed precisely to supplement people’s final living standards.  Add these into the calculations and the ‘poor’ get a lot better off, while the gap between the top and bottom narrows.

Thirdly, the report ignores the fact that many of the people who find themselves below the poverty line this year will be above it next year.  There is a lot of ‘churning’ in and out of poverty, and the number of people who are in long-term, sustained poverty is much smaller than the number who appear under the ‘poverty line’ at any one time.

Put these three points together, and it is clear that our estimates of the size of the poverty problem need to be revised downwards quite dramatically.  We suspect that there are no more than one in twenty Australians living in real, long-term poverty – and it may well be even fewer than that.

This debate may sound like two sets of researchers nit-picking, but its implications are enormously important.  If the Smith Family was right that one in eight people in this country are poor, we would be looking at a major problem possibly requiring massive and hugely expensive new policies to rectify it (ACOSS, for example, has already urged that welfare benefits be raised).

Fortunately, though, the Smith Family are not right.  Poverty today is a manageable, targetable problem, caused mainly by lack of paid work, and it can in large part be overcome by policies designed to get the long-term poor participating actively in the labour force.

About the Author:
Professor Peter Saunders is Director of Social Policy Programmes and Dr Kayoko Tsumori is a Policy Analyst at The Centre for Independent Studies.