Chairman, Australian Catholic Council for Employment Relations
It has been said that you cannot solve poverty until you measure it. Perhaps we should add the words ‘as best you can’, in recognition of the challenge of setting a precise monetary value. We have to do the best we can to measure poverty if we are to have a real chance of eliminating it and its attendant deprivation. A measure of poverty is just as important as a compass in a jungle.
A surprising feature of debate in Australia about poverty, and the assessment of a range of government policies that are meant to alleviate it, has been the absence of a measure of poverty, or any reasonable approximation of it, that can provide the focus for advocacy and assessment. It wasn’t always so.
In the 1960s and 1970s, the work of the Melbourne Institute of Applied Economic and Social Research, in particular that of Professor Ronald Henderson, developed and disseminated a set of evidence-based assessments that did influence public policy. So influential was this work and the national Poverty Commission headed by Professor Henderson that we had political and community debate about ‘poverty gaps’ that had dollar values. The Melbourne Institute still publishes quarterly updates of what are commonly called the Henderson Poverty Lines to reflect increases in Household Disposable Income per head, but they don’t cause a ripple amongst activists and the commentariat.
Since the mid-1990s the Social Policy Research Centre at the University of New South Wales has done some very valuable work on the needs and costs of various kinds of households, but this budget standards research does not set out to identify poverty as such.
Relative poverty lines
Another way of identifying poverty is by the use of relative poverty lines. These are estimates of poverty by reference to a measure of community income, the most common of which is the median equivalised disposable household income. Relative poverty can be set at different percentages; for example, where the amount of disposable income available to a household is 60% of the median figure.
There is debate about where a poverty line should be set, eg at 50% or 60% of the median, or somewhere in between. The appropriate level needs to be set by reference to empirical research, of which there is very little. This potential for debate is illustrated in, for example, the case of a single pensioner. At December 2012 the single pensioner’s disposable income was 54.8% of the median. Is that a poverty standard of living?
Differences in household size and composition are taken into account by the use of various equivalence scales that identify the differences in income that are required to yield the same standard of living across the households.
These scales mean that we can estimate each household’s percentage of the median income and compare different kinds of households. We can compare the standards of living, say, of a pensioner couple and a family of two adults and two children.
The Organisation for Economic Cooperation and Development (OECD) publishes a wide range of data on relative poverty lines for individual member countries and estimates of the number of people who are under these poverty lines. This is data well known to the Australian Government, but (for reasons noted below) it does not make use of them in public discourse. The OECD’s equivalence scales are used by the Australian Bureau of Statistics (ABS).
The use of poverty lines in minimum wage cases
Relative poverty lines can provide important information about the setting, operation, and assessments of national safety nets. The review of pensions in 2009 (see the Harmer Report) made use of aspects of the OECD methodology.
The use of these poverty lines in minimum wage-setting has been limited, but it should be of interest to those engaged in a broader range of policy issues.
For years, the Australian Catholic Council for Employment Relations (ACCER) has been highlighting the plight of low-paid workers, especially those with family responsibilities. It has argued that the National Minimum Wage (NMW) and other low-paid wage classifications deliver poverty wages for workers with family responsibilities, even after taking into account the substantial family payments.
From 2006 to 2009 the Australian Fair Pay Commission (AFPC) made a substantial effort to use the Henderson Poverty Lines as well as the 60% relative poverty lines to illustrate the impact of minimum wage levels on various kinds of households.
Afterwards, Fair Work Australia, now renamed the Fair Work Commission (FWC), continued to publish the relative poverty lines, but said that it regarded the Henderson Poverty Lines as outdated, and has not used them.
With the effective demise of the Henderson Poverty Lines, the attention is now on relative poverty lines. The term ‘poverty line’ can now be used without qualification to refer to the relative poverty line. We should make it clear what percentage we are using, but in the absence of work that will provide another percentage, the 60% level can be regarded as the default position.
While poverty lines are important to the setting of wages, few would argue that the minimum wage would be set at the poverty line. We would expect that the disposable income outcome, in terms of its proportion to the median, would be higher in the case of an employed person than it would be for pensioners, even if only because of the costs of work. Given that single pensioners are on 54.8% of the median, 60% of the median for workers does not appear to be an unreasonable relativity.
The impact of minimum wage rates on different households, after taking into account taxes and transfer payments, will depend on the composition of those households. The more children there are, the lower the standard of living. So a major issue for wage-setting is the number of people who may depend on a worker’s pay packet.
A simplistic approach would be to set wages on the basis that the worker is single and without family responsibilities, leaving it to government to provide for the needs of the other family members. Two points need to be made. First, the FWC has to take into account, across a wide range of matters, the needs of workers with family responsibilities. The adoption of a single person test would breach that obligation and discriminate against workers with family responsibilities. Second, despite the very large increases in family payments over the past decade, we are a long way short of the government covering the costs of dependants. Government transfers do not cover the costs of dependants; and they are not intended to do so. All the indicators show that this spending has come to an end.
The setting of a fair safety net wage, as the FWC is obliged to do, raises fundamental issues about the nature and purpose of a safety net and the need to take into account the fact that some workers have family responsibilities. How far must the safety net go to cover the sometimes unusual circumstances in which workers with family responsibilities find themselves?
ACCER has argued that a safety net wage is a wage which, together with taxes and transfers, is sufficient to support workers with family responsibilities in the ordinary and expected circumstances. In contemporary Australia this means that we have to focus on workers, whether partnered or not, with two children. A safety net has to be sufficient for these couple parent and sole parent families.
We have concentrated on three kinds of households; couples with two children, sole parents with two children and single workers. The single worker’s position is a little academic in circumstances where family payments do not cover the needs of dependants.
Poverty Lines in the Annual Wage Review 2013
As a result of the work of the Australian Fair Pay Commission and the later Fair Wages Commission, we have a number of annual poverty lines. Over the last few years ACCER has presented aspects of this data, going back to 2006, in tabular form.
In 2013 we revised these figures in the light of further data and calculated poverty lines back to 2000. Our calculations showed that from December 2000 to December 2012 median disposable incomes rose from $438.18 to $813.49 per week, an increase of 85.7%. We produced very detailed calculations of the disposable incomes of single workers and workers with family responsibilities (two children) over the period December 2000 to December 2012.
We presented a graph to show how the family of four (with children aged 8 and 12) has fared under three different safety net wage rates; see Supplementary Reply, April 2013, at http://www.fwc.gov.au/sites/wagereview2013/inreply/ACCER_suprepsub_awr1213.pdf
Disposable Incomes of Safety Net-dependent Families Relative to Poverty Line (Couple and two children).
December 2000 to December 2012
After being only $1.03 per week under the poverty line in December 2000, the NMW-dependent family of four had a poverty gap of $109.46 per week in December 2012. The value of the breadwinner’s work had been reduced by $108.43 per week over the past 12 years, at the rate of about $9.00 per week per year.
We showed that many more fell below the poverty line. The family of four dependent on the C12 classification in the Manufacturing and Associated Industries and Occupations Award, which now pays only $648.00 per week, fell from 4.8% above the poverty line to 7.4% below the poverty line. Trade-qualified workers in the same award (on a wage of $710.00 per week), whom we would have assumed can support a family of four on their C10 wage rate, have seen their family’s position fall from 11.4% above the poverty line to 2.8% below it.
The C10-dependent single workers have fallen 24.3 percentage points relative to the poverty line, compared to 14.2 percentage points for C10 families. The discrepancy between the two is the result of increased family payments over this period. These kinds of changes have occurred across the whole range of awards.
These figures are based on full-time work. Many sole parents are unable to work full time because of its unavailability or because of the high costs of childcare.
About one in six workers is only paid the safety net rate. Most workers are covered by collective agreements or individual arrangements that have delivered much larger wage increases since 2000. In the 12 years to December 2012 the C10-dependent worker received a wage increase of 43.3%. By comparison, over the same period Average Weekly Ordinary Time Earnings (AWOTE) rose by 74.4%.
The contrasting outcomes for safety net-dependent workers and the rest of the workforce was the major cause of increasing wage inequality in Australia over this period and the primary reason for the increasing numbers of workers and their families falling below the poverty line, or getting much closer to it.
It should be noted that the extension of family payments to middle income groups exacerbated this process; for example, the family of four on AWOTE had an increase of 88.8% in their disposable income. In contrast to safety net-dependent families noted above, over the 12 years the AWOTE family increased its margin above the poverty line, from 24.7% to 26.9%.
Making more use of Poverty Lines
While much of ACCER’s work is original, it is based on the work of others; in particular, ABS data, quarterly assessments of changes in Household Disposable Income by the Melbourne Institute and the methodology of the AFPC and the FWC.
In 2012 research commissioned by the Australian Council for Social Services showed that 7.1% of full-time workers are below poverty line, including many with family responsibilities. The report estimates that 20.5% of those living below the poverty line (estimated to total 760,000) are in, or rely on, full-time employment.
The OECD material is the source of some concerning material. Australia has the third highest poverty rate among 23 OECD countries (calculated by reference to 50% of median income) and child poverty across 27 countries shows that Australia has the 11th highest child poverty rate (also by reference to 50% of median income).
These figures are startling for a wealthy country that has prided itself on its egalitarianism.
The statistics are important, but aggregates need to be underpinned by data that people can understand. If these worrying figures about overall and sectoral poverty rates are to cut through we need poverty lines that are understood.
Why don’t we have a widely-used and credible poverty line as a point of reference for public debate in Australia?
The basic data is easy enough. As ACCER has demonstrated, we can calculate median equivalised disposable household incomes back to 2000. We have at hand a very important benchmark for the evaluation of wage and pension increases over that period.
You can see why the Commonwealth Government, for example, is reluctant to make much public use of these kinds of figures. They do not fit the narrative of the early 2000s, as written by successive governments. But it is about more than a narrative. The development of a credible and accepted poverty measure will expose government programs to greater scrutiny. The best example of this is in regard to the Newstart allowance for the unemployed.
At the Senate inquiry into the level of this allowance and other matters the Government’s defence of the status quo was partly based on the claim that measures of adequacy have little or no utility. (See ACCER’s discussion of these matters at paragraphs 129 to 139 of its March 2013 submission to the Annual Wage Review, at http://www.fwc.gov.au/sites/wagereview2013/submissions/ACCER_sub_awr1213.pdf
There are, however, two challenges that need to be addressed.
First, the 60% relative poverty level needs to be subjected to empirical testing. That testing must take into account the society in which the person lives and the social value that income has. The social value of money is important because it is the means by which people participate in their societies. There has been a tendency to concentrate on the maintenance of the real value of wages to the detriment of the recognition that the social value of wages is the principal determinant of social inclusion.
Second, inadequacies of the current equivalence scales need further work; for example, a comparison between the costs of sole parent and couple parent families takes no account of childcare costs for the sole parent and, a comparison between the living standards of a single pensioner and a single worker takes no account of the costs of work. Childcare costs have the capacity to drive sole parent families into poverty or force them into inadequate childcare arrangements.
I referred earlier to the budget standards research of the Social Policy Research Centre at the University of NSW. That research was initially commissioned by the Keating Government in the mid-1990s and is currently being revised and updated pursuant to a grant from the Australian Research Council. This research should help address equivalence scales and the setting of poverty lines at some point in the 50% to 60% range. While there is debate about the appropriate level of a poverty line, there can be no debate about its trajectory over the past decade or more. This is sufficient for us to ask some very substantial questions about inequality and social inclusion.
Among other things, the FWC has to take into account relative living standards and the needs of the low paid when setting safety net wages. Poverty measures are relevant to that obligation, as are the changes in poverty levels since 2000. Those who have followed labour market changes since 2000 are well aware of the fundamental cause of the increases in poverty: growing wage inequality. It is evidenced by what is called the ‘minimum wage bite’, which relates the NMW, or some other safety net rate, to average earnings.
In the 2012 Annual Wage Review decision, increasing inequality was treated with equanimity and it made no reference to poverty issues, even though there was substantial material on both aspects. This year the FWC has more material on poverty, including the material set out above. At the time of writing this essay the decision is reserved. How it responds in 2013 may shape the future direction of wage setting. However, our experience of the review shows that these poverty line lines are powerful tools and that they deserve greater levels of research support.