Closing the gap: the growing divide between poverty research and policy in Australia.

Saunders, Peter

Introduction

The fiftieth anniversary of The Australian Journal of Social Issues coincides closely with another milestone in Australian social policy. Next year will mark the fiftieth anniversary of the landmark Melbourne Poverty study conducted by Ronald Henderson and colleagues from the Melbourne Institute of Applied Economic and Social Research, the main fieldwork for which took place between April and June 1966 (Henderson, Harcourt & Harper 1970; see also Williams 2012: 14-21). That study had a profound impact on subsequent generations of Australian poverty researchers, establishing a framework for collecting data and measuring poverty, defined as a lack of income relative to need. The framework allows the dimensions of poverty to be estimated, its proximate determinants identified, and the impact--and inadequacies--of social security and other government provisions to be assessed. Importantly, Henderson also emphasised the need--frequently forgotten in subsequent developments--for complementary studies examining in detail how families on low incomes manage to survive, and the role that services, local communities and other factors play in this process (see McCaughey, Shaver & Ferber 1977). Together, the quantitative and qualitative studies pioneered by Henderson increased public awareness of the issues, produced a better understanding of the underlying causes of impoverishment, and resulted in the establishment of the Poverty Inquiry under Henderson's leadership, which was a landmark in Australian poverty research.

These were formidable achievements. Nonetheless, a broader perspective on the impact of this work provides a fuller appreciation of its contribution to public understanding of the issue and the momentum for change that resulted--in large part because this helps explain why poverty rarely features in contemporary policy debate in Australia. Henderson and colleagues displayed consummate skill not only in conducting the research, but also in disseminating their findings in digestible chunks--using language that spoke to the public--to policy makers and to government decision-makers. It was thus not just the research itself that mattered, but also its role in initiating a dialogue between researchers and policy makers, which built the momentum for change. Although the extent of change was less than Henderson had hoped for, the shortfall had as much to do with the turbulent economic conditions in the 1970s than with any lack of political will to improve social policy (Manning 1998). In retrospect, the episode serves as a classic example of 'evidence-based policy', with cutting-edge research conducted by respected experts identifying an issue that needed addressing, tapping into public awareness of the need for change, and convincing policy makers to respond with the commitment--and resources--required to address the problem. (1)

How things have changed. Today, the word poverty has virtually disappeared from policy discourse--in large part because the research lacks credibility among decision-makers, but also because policy makers have become suspicious of evidence that can be used to hold them to account on a sensitive issue with strong 'moral imperative' overtones (Whiteford 2001). This is an example of what the OECD Secretary-General has recently described as the 'ostrich' approach to policy, which involves 'ignoring difficult issues, in the hope that no-one will talk about them and that they will eventually go away' (OECD 2008). There is a ray of hope in a recent Productivity Commission report on disadvantage in Australia--'disadvantage' constituting bureaucratic code for poverty--in which it is argued that: 'A lack of understanding about disadvantage can ... be an impediment to good public policy ... [which] ... should be built on an evidence-based understanding of the nature, depth and persistence of disadvantage' (MeLachlan et al. 2013: 3, 4; italics added). But it comes at a time when one of the most important social security payments for people of working age--Newstart Allowance--remains well below the poverty line, and where the growing chorus of national (Whiteford 2010) and international (OECD 2010) calls for an increase to restore its adequacy has fallen on deaf ears--even by the previous ALP Government, which had placed social inclusion at the centre of its social policy agenda (Saunders 2013a). (2)

Australia lags behind many other countries in the priority given to addressing poverty. While recent initiatives introduced by the Australian Bureau of Statistics (ABS) have improved the quality of its income data, it has increased the difficulty in tracking movements over time in poverty rates, defined as those with income below the poverty line. (3) The Bureau has not recently published estimated poverty rates, preferring instead to report on the circumstances of 'low income households' that fall within the second and third deciles of the income distribution (see for example, ABS 2012), or on 'households with low consumption possibilities' (ABS 2009b) or 'low economic resource households' (ABS 2013b), which combine low income with other indicators of economic adversity. (4) The resulting vacuum in regular official reporting on trends in the numbers of low-income--or poor--households is reflected in the absence of an informed national debate about poverty, even in relation to child poverty, where the evidence is clearest about the negative longer-term consequences and the benefits--economic as well as social--associated with early intervention policies. (5) Despite this, increased attention is being paid to the economic, social and political consequences of growing inequality and to the risks associated with financial and other forms of economic instability, and when substantial progress has been made internationally in identifying and measuring poverty. With an increasing number of other countries drawing on an expanding evidence base to design and implement improved anti-poverty policies, Australia lags behind--not only in terms of its commitment, but also in terms of its performance. (6)

Against this background, the remainder of this paper reviews recent developments in international poverty research, identifies some existing gaps in Australia, and explores what needs to be done to improve the situation. It does not attempt to be comprehensive in its coverage, but focuses instead on those elements of the measurement framework established by Henderson where new approaches are most urgently needed. One issue that is not addressed--primarily for reasons of space--is the role and impact of social exclusion in providing a new framework for poverty analysis and examining social disadvantage more broadly. This omission does not reflect the view that the exclusion literature is incapable of contributing to poverty research and anti-poverty policy, but rather because the two are best kept conceptually distinct and thus warrant separate, but integrated, examination. (7) In what follows, the next two sections present recent evidence on income poverty using the current version of the Henderson framework, and compare this with evidence produced using a deprivation approach. This is followed by discussion of the three crucial ingredients of a revitalised anti-poverty strategy: better measures, better data, and better dialogue. The main threads of the argument are drawn together in the concluding section.

Another issue that is not addressed explicitly is the role that this journal has played in publishing articles on poverty and promoting academic debate on the topic. A review by the current author published a decade ago to commemorate its fortieth anniversary began by noting that many of Australia's leading poverty researchers had not published their work here, and those who had often addressed topics other than poverty (Saunders 2005: 13). It went on to observe that despite this, many articles had addressed aspects of poverty, and this continues to be the case. The last decade, for example, has seen the publication of important contributions on the profile (Wilkins 2007) and dynamics (Headey, Marks & Wooden 2005) of poverty, and many others on specific aspects, including the role of domestic violence (Costello et al. 2005), the costs of disability (Saunders 2007), and energy pricing (Cheshire & Morris 2011). These papers highlight extensions to the Henderson framework and, by focusing on specific manifestations of poverty, are consistent with the material deprivation approach discussed later. However, it is also important to acknowledge that a number of significant developments in the Australian literature on poverty measurement have appeared in economics journals such as The Economic Record, The Australian Economic Review, and The Australian Journal of Labour Economics. While this reflects the greater availability of unit record file data, it is also indicative of the increased technical complexity of measurement and its interest to economists rather than to the wider audience to which this journal is directed.

Income Poverty

Poverty line studies provide a valuable starting point for any assessment of the extent of poverty because income plays a pivotal role in determining living standards in a country like Australia. While much has been written about the problems inherent in this approach, these weaknesses do not obviate the need to estimate poverty rates. (8) Australia is a world leader in targeting social benefits to those in greatest need; however, it is remiss in failing to construct, monitor and debate income-based measures of payment adequacy and the risk of poverty, which income poverty rates capture. (9) Income is the form in which most social security assistance is provided and it is also the basis of a major component of the tax system, which contributes to the funding of social security benefits. Poverty rates capture the risk of poverty because some of those with incomes below the line may be able to call on other resources to escape poverty, if only in the short term, for example, by running down savings, borrowing from relatives or friends, or drawing on emergency assistance provided by a welfare agency. Nonetheless, it is necessary to know how many and what kinds of people are at risk and how poverty is changing. Without that information, it is difficult to establish how well current policies are working and what more needs to be done, or done differently.

Recent international developments in income measurement (reflecting Australian leadership and input) have seen the definition of income expanded to better capture those elements that are difficult to measure in household surveys and/or have grown in importance. A Task Force chaired by the ABS has been established by the UN 'to incorporate new developments in the area of household income measurement and to expand the guidelines to take into account these new developments' (United Nations Economic Commission for Europe 2011). Its report notes (p. 88) that people living in low income households 'are more likely to have insufficient economic resources to support an acceptable standard of living ... [although] ... a full understanding of poverty requires information in many non-economic dimensions ... [including] ... cumulative deprivation in relation to income, housing, education and health care'. This assessment is consistent with the now widely accepted approach to measuring poverty, which combines income poverty measures with evidence on material deprivation (see below).

Ensuring that our income measures keep pace with the changing forms in which people receive their income is important if the statistics are to capture reality accurately. With this goal in mind, the ABS has been improving the quality and expanding the scope of its income statistics for more than a decade (see ABS 2002a; 2003; 2009a; 2013a). These changes--documented for the 1990s by Siminski et al. (2003) and for the 2000s by Wilkins (2014a)--have without doubt raised the quality of the data but have also compromised their inter-temporal comparability, making it difficult to establish the extent of change in poverty and income inequality. It is difficult to disagree with Wilkins (2014a: 88), who argues for 'data providers to be cognisant of the importance of consistency in data over time' and expresses the hope that 'future revisions to ABS concepts and survey methods for its household income survey collections will be kept to a minimum'. Well said! (10)

Reluctance on the part of the ABS to publish estimates of income poverty reflects a deeper concern about the concept and its measurement. In an important report on Measuring Wellbeing, released in 2001, it argued that:

People with a low standard of living may be considered to be in poverty if the range and quality of resources available to them and their ability to participate in society falls below community standards. There is no general agreement on what the 'community standards' are--indeed many standards exist.... Thus poverty, as with living standards, is a relative and quite subjective concept (ABS 2001: 183-184).

It went on to note that measures that rely solely on income 'can be misleading' because some people--the self-employed were mentioned as an example--may have low incomes but still be able to finance a higher level of consumption, arguing that 'alternative approaches to measuring poverty such as those based on levels of consumption would also be useful' (ABS 2001: 206). It is also possible that a perceived lack of agreement among Australian poverty researchers at the time about where to set the poverty line may have contributed to the ABS's reluctance to publish estimates of income--or consumption--poverty. (11) This gap has been filled by The Australian Institute of Health and Welfare (AIHW), which includes among its suite of indicators of Australia's welfare the percentage of households with incomes below 40 per cent, 50 per cent, and 60 per cent of the median (see AIHW 2013: Figure 11.23, Table All.20). (12) Estimates of relative poverty using a poverty line linked to median income are also produced regularly as part of the Household, Income and Labour Dynamics in Australia (HILDA) project's regular reporting and monitoring (see for example, Wilkins 2013). The Social Policy Research Centre (SPRC) and the National Centre for Social and Economic Modelling (NATSEM) also publish poverty estimates (Saunders, Wong & Bradbury 2014; Phillips et al. 2013) that draw on the ABS income survey data. All of these studies set the poverty line as a percentage--normally 50 per cent--of median income, with some varying the percentage in order to assess the sensitivity of the estimates. There is thus no shortage of poverty estimates for Australia, although we lack an official detailed series with all the authority that this conveys. Such information would contribute to a revitalised poverty research agenda focused on better informing future policy, not just berating the government about the failings of past and current policy.

It is now widely accepted that income-based poverty line estimates provide a valuable starting point from which to document the extent of poverty and begin to grapple with its underlying causes. Although these estimates depend upon an arbitrary poverty line, setting the line at 50 per cent of median--equivalised income is now widely practiced and accepted internationally. (13) With this in mind, Table 1 presents estimates of poverty derived from the ABS 2011-12 Survey of Income and Housing (SIH) on a before housing costs (BHC) and after housing costs (AHC) basis. (14) The estimates apply the latest (introduced in 2007-08) and previous (applying in 2005-06) ABS income measures in order to assess the impact of the measurement changes. The poverty line has been set at 50 per cent of the relevant median income, the modified OECD equivalence scale has been used to adjust for need differences, and the estimates refer to the numbers of persons in households with income below the poverty line. (15)

The estimates in Table 1 highlight the vulnerability of single person and sole parent households and the role of housing costs in protecting older Australians from poverty. They also show that high housing costs expose other groups to poverty, particularly where there is only one adult present, although the overall between-group disparity in AHC poverty rates is lower than for BHC poverty. Poverty rates for families with children, particularly sole parents, have declined, reflecting the impact of strong employment growth and increased family payments. It is clear that how poverty is measured affects who is identified as most at risk, and also that the measurement changes introduced in 2007-08 had a considerable impact on poverty, whichever measure is used. The AHC poverty rate increased by just over one percentage point as a result of the change, and the number of people in poverty rose by almost 200,000. These observations illustrate the value of head-count poverty rates which, despite the shortcomings, provide a broad picture that promotes improved understanding and encourages debate about underlying causes.

Material Deprivation

Despite all of the measurement and statistical uncertainties, the main limitation of poverty estimates like those in Table 1 is conceptual, not empirical. It relates to the fact that the existence of poverty has not been demonstrated. (16) If poverty is defined, following the World Bank, to include 'those who do not have enough income or consumption to put them above some adequate minimum threshold' (Haughton & Khandker 2009: 1) or, as proposed by the Irish Combat Poverty Agency (2004: 1), as those whose incomes 'are so inadequate as to preclude them from having a standard of living which is regarded as acceptable', then Table 1 has little to say about the actual experience of poverty--as opposed to the risk of exposure to it. One way of addressing this limitation involves examining more directly the living standards actually achieved in order to establish whether or not they meet prevailing standards of adequacy or acceptability. (17)

A framework that does this is provided by the deprivation approach developed originally by Townsend (1979), which is now widely used to identify those who cannot afford items regarded as necessary or essential by a majority in the community. This lack of affordability of 'socially perceived necessities' (Mack &c Lansley 1985: 38) is taken as indicating that those so identified are unable to attain a standard of living that is consistent with prevailing community standards of acceptability (Gordon 2006; Nolan & Whelan 2007: Nolan & Marx 2009; Saunders 2013b; Saunders, Naidoo & Griffiths 2008). The deprivation approach is designed to supplement, not replace the poverty line studies discussed earlier. The two approaches produce distinct but overlapping measures and can be combined to identify those who are both income poor and unable to buy a minimum number of necessities, as is illustrated in Table 2. Poverty is defined here as those with equivalised income below 50 per cent of the median, as before, although the income data used to produce the estimates in Table 2 are derived from a survey conducted by the author and colleagues and is of much poorer quality than the ABS income data used to produce Table 1. Deprivation is defined to exist when a household does not have and cannot afford at least 4 of the 25 items identified as being essential 'for all Australians'. (18) These definitions of poverty and deprivation allow the population to be split into four groups, defined on the basis of the different combinations of poverty and deprivation; the size of these four groups is shown in the first row of Table 2, followed by the average value of a series of objective and subjective outcome and well-being indicators for each group.

If poverty results--as it surely should--in such concrete examples of unmet need as an inability to buy basic items or pay existing bills and, for some, results in a need to seek external support or assistance, then Table 2 indicates that identifying poverty on the basis of income alone provides a far inferior measure to one based on deprivation. According to all eight objective indicators, those with incomes below the poverty line appear between three and five times less adversely affected than those who are deprived. If the definition of poverty is restricted to those below the poverty line who also experience deprivation, the indicators all rise sharply--implying a lower standard of living--while the increase associated with having an income below the poverty line if one is already deprived is positive, but far more modest. A similar picture is revealed by the subjective indicators, although the differences here are smaller--a reflection of the bounded nature of the indicator scales and the tendency for people to adapt to their circumstances, causing the indicators to remain relatively stable. Even so, the fact that less than one in ten of those with incomes below the poverty line describe themselves as poor, compared with two-fifths of those who are deprived and two-thirds of those who are both poor and deprived is compelling evidence in support of the inclusion of deprivation in the poverty measure.

Table 2 provides a case for including deprivation in extended measures of poverty--as has been done in countries such as the UK and Ireland, which have adopted official poverty reduction targets. This approach is now widely used by international agencies including the OECD and UNICEF to supplement income-based poverty rates (see OECD 2007; de Neubourg et al. 2014). This increased reliance on deprivation measures does not mean that the identification and measurement of deprivation does not also raise issues that need to be thoroughly examined and assessed. One such issue is how the 'essentials of life' are to be identified. The standard approach involves including only those items that receive majority support for being essential--as 'things that no-one in Australia should have to go without today'. An alternative approach would identify these items not on the basis of what a majority in the community thinks, but according to the items that individuals themselves identify as essential. This would bring the deprivation approach more into line with the capability approach developed by Sen (1985) by defining deprivation using items that individuals themselves value as essential or necessary. This approach has been examined by Saunders and Wong (2012a: Table 5.2), who show that it makes a large difference to the estimated deprivation rates of some items. (19) Issues such as this warrant further examination to ensure that the deprivation approach is capturing what it is intended to capture--an inability to achieve a socially acceptable minimum standard of living.

If a deprivation approach is to form the basis of future Australian poverty research --as it should--then this would represent a substantial shift beyond the Henderson framework described earlier, though not one that would have been of concern to Henderson himself. At the outset of the Poverty Inquiry report it was noted that:

Poverty is not just a personal attribute; it arises out of the organisation of society. Australian society has failed to adapt ... with the consequence that many are ... cut off in many ways from participation in community groups and urban society (Commission of Inquiry into Poverty 1975: 1).

This statement is entirely consistent with the ABS living standards approach discussed earlier, and with the deprivation approach, which seeks to identify the adverse outcomes--or lack of basic necessities--that are associated with poverty. In a similar vein, the 2004 Senate Report on poverty identified poverty as a multidimensional concept that goes beyond just material deprivation ... when it refers to poverty in this report, it is referring to a concept of deprivation, of lack of opportunity to participate fully in society, of social isolation and exclusion (Senate Community Affairs References Committee 2004: 3).

The need for a paradigm shift in Australian poverty research has thus been apparent for at least a decade, although the failure to promote and use deprivation measures as part of an anti-poverty strategy under the ALP Government's social inclusion agenda was a major lost opportunity.

Better Data, More Research

The only source of reliable national income data other than that produced by the ABS is the HILDA survey, which has been used to track movements in poverty over the period since the survey began in 2001 (see for example, Wilkins 2013). Estimates have also been derived of changes in 'absolute' poverty, using a poverty line anchored at a fixed point in time and adjusted only for changes in consumer prices. This makes a substantial difference to the poverty trend over a period when real incomes are rising. Thus as Wilkins (2014b: Figure 4.3) shows, although the relative--half-median--poverty rate remained stable at between 12 and 13 per cent over the decade to 2011, the absolute poverty rate declined markedly from just below 13 per cent in 2001 to below 6 per cent in 2011. (20) However, the most significant development associated with the HILDA survey is that for the first time it allows the dynamics of Australian poverty to be examined. This is a second important development that has taken Australian poverty research beyond the Henderson framework. Wilkins (2014b) examines this issue using an estimate of 'permanent income', defined as income averaged over five years, and shows that the poverty rate--assessed using a poverty line set at 50 per cent of median permanent income--fluctuated over the decade to 2011 between 8.5 and 10.0 per cent--considerably below the static (one-year) estimates cited earlier (Wilkins 2014b: Table 4.7).

Wilkins asserts (2014b: 31) that the concept of permanent income poverty 'is of considerably greater policy importance than one-year poverty' although no justification is provided to support this claim. While it is true that poverty becomes a more serious problem for those affected the longer it is experienced, it does not follow that the long-term poverty rate is more important for policy purposes than the short-term (annual) poverty rate: the former is more intense but affects fewer people, while the latter is more prevalent and provides clues as to how the former arises and evolves. Some of the language used in recent studies of poverty dynamics may have inadvertently contributed to the perception that short-term poverty is of less concern for policy. For example, Rodgers and Rodgers (2009) used the first six waves of HILDA to estimate 'chronic' and 'transitory' poverty, where the former is defined as those whose permanent income--defined over the whole six years--falls below fixed percentages of median permanent income, and the latter refers to those who are not chronically poor overall but whose income falls below the poverty line in any single year. Use of the word 'transitory' in this context might be construed as implying that the problem is short-lived and self-rectifying and thus of less concern for policy makers. More generally, there is need for further examination of whether the poverty line--and equivalence scale--used in these studies of long-term poverty should be the same as those used in conventional studies based on weekly or annual income.

The only sources of Australian data that can currently be used to estimate deprivation on a national scale are the surveys of poverty and social exclusion conducted by the author and colleagues at the SPRC (Saunders, Naidoo & Griffiths 2008; Saunders & Wong 2012a; 2012b). (21) However, limited funding has restricted the size of the surveys and necessitated a mail-out approach, which reduces the response rate and is not ideal for asking the questions used to identify which items are essential. Despite this, results obtained from the two surveys conducted to date confirm that the methodology produces plausible and robust results that shed important new light on the nature and extent of poverty. (22) Larger samples and better survey methods are needed to reduce response bias and allow more detailed examination of the deprivation experienced by those groups --for example, sole parent families, the long-term unemployed, and Indigenous Australians--that are known to be particularly prone to poverty. National statistical agencies in countries such as the UK and Ireland have included a suite of deprivation questions in household surveys--like the UK Family Resources Survey--that collect the data needed to produce robust estimates of deprivation and poverty. (23) The items and precise formulation of the questions followed an exchange between policy makers, poverty researchers, NGOs and the general public about what measures are needed, and experts have been engaged to refine the wording of questions and provide general advice on survey methodology (see Department for Work and Pensions 2003; McKay 2011).

In Australia, the ABS has been collecting information on indicators of hardship and financial stress for the last decade, but no attempt has been made to identify which items are relevant to deprivation--the 'necessities or essentials of life' that are widely seen as things that everyone should have. (24) However, beginning in 2014, the HILDA survey will include a suite of deprivation questions based on the items identified as essential in the SPRC deprivation studies referred to earlier. (25) If successful, the questions will be repeated on a four-year cycle, thus allowing the items included--and possibly the wording of the questions--to be modified to capture and reflect changing community norms and attitudes. The first wave of this important new data will be available for analysis at the end of 2015; this will provide exciting new opportunities to bring Australian poverty research into the twenty-first century and in line with international developments.

The discussion so far has focused entirely on quantitative studies of poverty that examine its overall level and extent within a national context. There have been relatively few studies of poverty among specific vulnerable groups (even among Indigenous Australians, aside from work undertaken by Hunter and others at the Centre for Aboriginal Economic Policy Research: see Altman & Hunter 1998; Hunter 2012). With the exception of recent work by NATSEM on the locational profile of poverty and social exclusion using census data (see Phillips et al. 2013), there has been relatively little research into the role of location as a poverty risk factor, although lack of appropriate survey data--which often suppresses detailed information about location to protect respondent confidentiality--has been a contributing factor. Studies of child poverty that incorporate children's views on how poverty is identified and builds on their experiences of dealing with financial adversity are beginning to emerge in the international literature (see for example, Main & Bradshaw 2012) and are generating similar research in Australia, although to date it has been mainly qualitative (e.g. Skattebol et al. 2012), and the resulting insights have not yet been incorporated into quantitative studies of child poverty. These studies raise important questions about whether the approach should be tailored to reflect the characteristics of specific groups, or if a uniform approach should be applied to all: if the way child poverty is identified and measured incorporates the views of children, should not a similar approach be used for people in other groups--defined on the basis of such factors as age or ethnicity--and if so, how far should this process be allowed to extend?

There have been far too few qualitative studies that seek to improve understanding of how poverty is experienced, how people cope and adapt, and its consequences. Despite their expressed interest in these issues, policy makers have been notoriously reluctant to take such studies seriously, often in the mistaken impression that small sample size and lack of generalisability mean that such studies are not capable of providing any insights or lessons for policy formulation and direction. This view reflects a misunderstanding of the role of such studies, which is to enhance understanding of the processes that create poverty and the motivations and behaviour of those affected. There is much to be gained by paying greater attention to the important insights that can only be provided by well-designed and conducted qualitative studies. These studies are grounded in the lived experience of poverty and, like deprivation studies, can provide new evidence on the nature and impact of poverty that is an important complement to the statistical findings of poverty line studies.

Better Measures, More Dialogue

The lack of communication between poverty researchers, data collection agencies and policy makers that exists in Australia is neither typical of other national poverty responses, nor is it restricted to those holding any single political viewpoint. In the UK, extensive community consultation preceded the selection of the child poverty reduction targets that have been enshrined in legislation under the Child Poverty Act, 2010. (26) In Ireland, which launched its National Anti-Poverty Strategy in 1997 with considerable input from researchers at the National Institute for Economic and Social Research (Nolan 2000), the recently revised National Action Plan for Social Inclusion 2007-2016 includes a national social target for poverty reduction that will reduce consistent poverty--the combination of income below 60 per cent of the median and deprivation of at least 2 out of 11 basic need items--to 4 per cent by 2016 and to 2 per cent by 2020, from a baseline level of 6.2 per cent in 2010 (Department of Social Protection 2012). The fight against poverty is a central feature of the EU's revised social agenda (the Europe 2020 Strategy), which is targeting the reduction in poverty and exclusion using three indicators: the at-risk poverty rate; an index of material deprivation; and the percentage of people living in very low work-intensity households.

Even the United States--hardly a world leader in anti-poverty policy--has recognised the weaknesses in its current poverty measure. In 2010, the US Census Bureau introduced a supplemental poverty measure 'to provide an alternative view of poverty in the United States that better reflects life in the 21st century, including contemporary social and economic realities and government policy' (Institute for Research on Poverty 2014). (27) There are no official poverty reduction targets in Canada, although UNICEF Canada has recently argued that:

Canada needs to set an official definition of child poverty and adopt more informative ways to measure it, as a fundamental means of monitoring the well-being of Canada's children (UNICEF Canada 2012: 5).

In New Zealand, an Expert Group appointed by the Children's Commissioner recommended after extensive community consultation that legislation be enacted that requires the measurement and monitoring of child poverty as part of a national strategy that includes the setting of short-term and long-term poverty reduction targets (Expert Advisory Group on Solutions to Child Poverty 2012).

These developments show that most of the countries that Australia traditionally compares itself with are using expert research and community input to inform how to measure and address poverty, particularly child poverty. This does not mean that there is universal agreement on all of the issues, as the recent debate over the UK government's child poverty reduction measures and targets illustrates. In late 2012, the Coalition government launched a public consultation on measuring child poverty that identified, in addition to income and material deprivation, seven other dimensions of poverty: worklessness, unmanageable debt, poor housing, parental skill level, access to quality education, family stability, and parental health (HM Government 2012). There was little mention of the income and deprivation measures that are captured in the existing Act, although these were acknowledged. The consultation exercise was widely criticised by academics and service providers--and by the UK Treasury--and to date the proposed changes have not been implemented. Among its leading critics, Bradshaw (2013a: 1) argued that the consultation document was 'of very poor quality', going on to describe (2013b: 9) the proposed new multidimensional indicator as 'a ragbag of risks, correlates and consequences of poverty rather than poverty itself'.

The UK experience is an example of the debate that has been generated in countries that have encouraged community engagement with the setting of official poverty reduction targets, in the process increasing their acceptance and legitimacy. Overall, the picture is one of a broad consensus about key issues, including the need for several targets that include both income and deprivation measures, covering both the relative and absolute dimensions of poverty. Governments have also agreed to report on progress and on the strategies being developed to reach the targets, thus providing clear articulation of how poverty is defined and measured that holds them to account for their actions, and for how they conceive of--and hence address--the problem.

The fact that most published poverty reduction targets have so far failed to be achieved in practice does not mean that the approach itself is flawed. As Stewart (2012) has noted in relation to the UK's failure to achieve the interim child poverty reduction targets set by Tony Blair in 1999,

to spin this as another broken promise is disingenuous, ignoring the real and sustained improvements in living conditions and opportunities enjoyed by hundreds of thousands of children as a result of policies implemented over the last decade (Stewart 2012:5).

Compare this with the flood of criticism directed at the Australian Government's failure to deliver on Bob Hawke's pledge to 'end child poverty by 1990'. That experience marked a turning point that ended Australia's--in retrospect, visionary--flirtation with poverty targeting, and its legacy has had a profound negative impact on subsequent debate on the topic. It left policy makers feeling that targeting in this area was a 'lose/lose' situation in which the government would be condemned for not achieving its target, but not praised for its attempts to do so.

Concluding Reflections

What can be done to escape from the cul-de-sac in which anti-poverty policy in Australia currently finds itself? There is a critical need to build on international developments and encourage dialogue between the key stakeholders, including policy makers--politicians and bureaucrats--researchers, NGO service providers, and advocates and those for whom the policy is intended: Australians living in poverty. This will only occur if the government is prepared to take steps to monitor and address poverty--including through the introduction of poverty reduction targets. However, this cannot happen unless there is agreement on definitional matters and that, in turn, must involve input from experts and stakeholders, as is the case in other areas of policy. But which must come first: the policy commitment or the development of better measures (the plural signifying that no single measure will suffice)?

We are currently a long way from any form of poverty reduction policy commitment in Australia, but substantial progress has and is being made to improve our data and poverty measures. The reluctance of governments to set formal poverty reduction targets reflects government concern over measurement problems and an unwillingness to expose itself to the harsh moral judgements associated with failure. However, improvements in data quality and availability and better poverty measurement, combined with advances in policy design and implementation capacity, have made other governments willing to place poverty reduction at the centre of their policy platforms. Poverty can be significantly reduced--though probably not eliminated entirely--and this will bring major economic and social benefits that no government can afford to ignore. Poverty reduction targets can be agreed and achieved, but this requires leadership from government, and input from others.

Another major obstacle to any shift towards poverty targeting is the concern of politicians and policy makers over the likely fiscal implications. Such concern is well-founded. Past attempts to achieve poverty reduction targets have shown that this is a far harder--and more expensive--task than was initially thought. (28) This is compounded by legitimate concerns that whatever progress is made will not be fully acknowledged, with public attention focused on what remains to be done, not on what has already been achieved. In short, it is a situation that combines the virtual certainty of high economic cost with the strong possibility of low political return. This conflict cannot be completely resolved, although it can be eased by setting realistic goals--in terms of both targets and timeframes, ensuring that the measures used to monitor progress capture what has been achieved, and focusing on areas of achievement, not just on failure. Examples of how this can be done include using poverty rate measures based on both 'absolute' (price-anchored) and relative (to median income) poverty lines, tracking movements in the average real incomes (and poverty gaps) of those below both poverty lines, and supplementing these income-based measures with deprivation measures that capture improvements in absolute and relative living standards. (29) Regular reporting of trends by either the ABS or AIHW should be required, along the lines of that currently produced as part of the HILDA project, and the whole process could be overseen by Treasury or COAG. Researchers can fill in the gaps and examine specific issues.

These are some examples of how to make those with the power to bring about change more willing to engage in dialogue about how to measure and alleviate poverty. Those who conduct and use Australian poverty research must recognise that progress towards the design and implementation of a national anti-poverty strategy that includes poverty reduction targets will require compromise and the search for common ground. Against this, policy makers need to be reminded of the importance of addressing poverty--not as an overriding objective, but as an important goal that helps to define the kind of nation we are. Australia has the resources and capabilities--economic, technical and practical--to ensure that poverty no longer denies people the chance to realise their potential, and the challenge is to pull those resources together and get on with it.

Acknowledgements

The author acknowledges the detailed and helpful comments provided by the two anonymous referees, and the statistical assistance provided by Melissa Wong.

References

ABS (Australian Bureau of Statistics) (2001) Measuring Wellbeing: Frameworks for Australian Social Statistics, Catalogue No. 4160.0, Canberra, ABS.

--(2002a) 'Upgrading household income distribution statistics', Australian Economic Indicators, April, Catalogue No. 1350.0, Canberra, ABS.

--(2002b) Measuring Australia's Progress 2002, Catalogue No. 1370.0, Canberra, ABS.

--(2003) 'Revised household income distribution statistics', Australian Economic Indicators, June, Catalogue No. 1350.0, Canberra, ABS.

--(2004) Measures of Australia's Progress 2004, Catalogue No. 1370.0, Canberra, ABS.

--(2009a) Household Income and Income Distribution, Australia, 2007-08, Catalogue No. 6523.0, Canberra, ABS.

--(2009b) 'Report on low consumption possibilities research project', ABS Methodological News, June, Canberra, ABS.

--(2012) 'Life on "struggle street": Australians in low economic resource households', Australian Social Trends, March Quarter 2012, Catalogue No. 4102.0, Canberra, ABS.

--(2013a) Household Income and Income Distribution, Australia, 2011-12, Catalogue No. 6523.0, Canberra, ABS.

--(2103b) 'Low economic resource households', Household Income and Income Distribution, Australia, 2011-12, Catalogue No. 6523.0, Canberra, ABS.

ACOSS (Australian Council of Social Service) (2014) Poverty in Australia in 2012. Poverty and Inequality Report No. 2, Sydney, ACOSS.

AIHW (Australian Institute of Health and Welfare) (2013) Australia's Welfare, 2013, Canberra, AIHW.

Altman, J. & Hunter, B. (1998) 'Indigenous poverty', in R. Fincher & J. Niewenhuysen (eds), Australian Poverty: Then and Now, Melbourne, Melbourne University Press, 238-57.

Australian Government (2012) Social Inclusion in Australia: How Australia is Faring, 2nd Edition, Canberra, Australian Social Inclusion Board, Department of Prime Minister and Cabinet.

Baum, S. (2004) 'Measuring socio-economic outcomes in Sydney: an analysis of census data using a general deprivation index', Australian Journal of Regional Studies, 10 (1), 105-128.

Blank, R. (2008) 'Why the United States needs an improved measure of poverty. Testimony to the subcommittee on income security and family support, House Ways and Means Committee', Washington DC, The Brookings Institute.

Bradshaw, J. (2013a) 'Consultation on child poverty measures', PSE Policy Response Working Paper No. 8, Poverty and Social Inclusion in the UK, Bristol, University of Bristol.

--(2013b) 'Measuring child poverty: can we do better?' Poverty, 144, 6-9.

Brewer, M. (2012) 'Child poverty and financial support for children', in L. Judge (ed.), Ending Child Poverty by 2020: Progress Made and Lessons Learned, London, Child Poverty Action Group, 28-31.

Cheshire, L. & Morris, A. (2011) 'A new form of energy poverty is the hallmark of liberalised electricity sectors', Australian Journal of Social Issues, 46 (4), 435-59.

Costello, M., Chung, D. & Carson, E. (2005) 'Exploring alternative pathways out of poverty: making connections between domestic violence and employment practices', Australian Journal of Social Issues, 40 (2), 253-267.

Combat Poverty Agency (2004) What is Poverty?, Dublin, Combat Poverty Agency.

Commission of Inquiry into Poverty in Australia (1975) First Main Report. Poverty in Australia, Canberra, AGPS.

De Neubourg, C., de Milliano, M. & Plavgo, I. (2014) 'Lost (in) dimensions: consolidating progress in multidimensional poverty research', Working Paper WP-2014-04, Florence, UNICEF Office of Research.

Department for Work and Pensions (2003) Measuring Child Poverty, London, Department for Work and Pensions.

Department of Social Protection (2012) National Social Target for Poverty Reduction, Dublin, Department for Social Protection.

Department of Social Security (1995) Developing a Framework for Benchmarks of Adequacy for Social Security Payments, Policy Discussion Paper No. 6, Canberra, Department of Social Security.

Expert Advisory Group on Solutions to Child Poverty (2012) Understanding Poverty and Wellbeing. A Note with Implications for Research and Policy, Wellington, Children's Commissioner.

Gordon, D. (2006) 'The concept and measurement of poverty', in C. Pantazis, D. Gordon and R. Levitas (eds), Poverty and Social Exclusion in Britain. The Millennium Survey, Bristol, Policy Press, 29-69.

Harding, A., Lloyd, R. & Greenwell, H. (2001) Financial Disadvantage in Australia 1990 to 2000. The Persistence of Poverty in a Decade of Growth, Sydney, The Smith Family.

Haughton, J. & Khandker, S. R. (2009) Handbook on Poverty and Inequality, Washington, The World Bank.

Headey, B., Marks, G. & Wooden, M. (2005) 'The dynamics of income poverty in Australia: evidence from the first three waves of the HILDA survey', Australian Journal of Social Issues, 40 (4), 541-552.

Henderson, R.F., Harcourt, A. & Harper, R.J.A. (1970) People in Poverty. A Melbourne Survey, Cheshire, for the Institute of Applied Economic and Social Research, Melbourne.

HM Government (2012) Measuring Child Poverty: A Consultation on Better Measures of Child Poverty, London, Department for Work and Pensions.

--(2014) Consultation on Setting the 2020 persistent Child Poverty Target, London, Department for Work and Pensions.

Hunter, B. (2012) 'Is Indigenous poverty different from other poverty?', in B. Hunter & N. Biddle (eds), Survey Analysis for Indigenous Policy in Australia, Canberra, ANU ePress, 193-221.

Institute for Research on Poverty (2014) How Is Poverty Measured in the United States? Wisconsin, Institute for Research on Poverty, University of Wisconsin.

Kennedy, S. (2014) 'Child Poverty Act 2010: a short guide' London: House of Commons Library.

Lloyd, R., Harding, A. & Payne, A. (2004) Australians in poverty in the 21st Century, NATSEM Conference Paper Series, Canberra, NATSEM.

Mack, J. & Lansley, S. (1985) Poor Britain, London, George Allen & Unwin.

Main, G. & Bradshaw, J. (2012) 'A child material deprivation index', Child Indicators Research, 5(3), 503-21.

Manning, I. (1998) 'Policies: past and present' in R. Fincher and J. Niewenhuysen (eds), Australian Poverty: Then and Now, Melbourne, Melbourne University Press, 10-32.

McCaughey, J., Shaver, S. & Ferber, H. (1977) Who Cares? Family Problems, Community Links and Helping Services, Melbourne, Sun Books.

McDonald, C. (2013) 'Poverty in Australia and the social work response', Asia Pacific Journal of Social Work and Development, 23 (1), 3-11.

McKay, S. (2011) Review of the Child Material Deprivation Items in the Family Resources Survey, Research Report No. 746, London, Department for Work and Pensions.

McLachlan, R., Gilfillan, G. & Gordon, J. (2013) 'Deep and persistent disadvantage in Australia', Productivity Commission Staff Working Paper, Canberra, Productivity Commission.

Nolan, B. (2000) 'Targeting poverty--the Irish example', Australian Social Policy, 1, 25-41.

Nolan, B. & Marx, I. (2009) 'Economic inequality, poverty and social exclusion', in W. Salverda, B. Nolan and T. M. Smeeding (eds), The Oxford Handbook of Economic Inequality, Oxford, Oxford University Press, 315-41.

Nolan, B. & Whelan, C. T. (2007) 'On the multidimensionality of poverty and social exclusion', in S. P. Jenkins and J. Micklewright (eds), Inequality and Poverty Re-examined, Oxford, Oxford University Press, 146-65.

OECD (2007) Society at a Glance: OECD Social Indicators--2006 Edition, Paris, OECD.

--(2008) Growing Unequal? Income Distribution and Poverty in OECD Countries, Paris, OECD.

--(2010) Economic Surveys, Australia 2010, Paris, OECD.

--(2014a) Income inequality Update. Rising Inequality: Youth and Poor Fall Further Behind, Paris, OECD.

--(2014b) Society at a Glance. OECD Social Indicators--Highlights Australia, Paris, OECD.

Phillips, B., Miranti, R., Vidyattama, Y. & Cassells, R. (2013) Poverty, Social Exclusion and Disadvantage in Australia, National Centre for Social and Economic Modelling, University of Canberra.

Rodgers, J.R. & Rodgers, J.L. (2009) 'Contributions of longitudinal data to poverty measurement in Australia', Economic Record, 85, Special Issue, S35-S47.

Rodgers, J.R., Siminski, P. & Bishop, J. (2009) 'Changes in poverty rates during the Howard era', Australian Economic Review, 42 (3), 300-320.

Saunders, P. (1998) 'Setting the poverty agenda: the origins and impact of the Henderson report', in R. Fincher and J. Niewenhuysen (eds), Australian Poverty: Then and Now, Melbourne, Melbourne University Press, 52-70.

--(2005) 'A valuable contribution to research and policy: reviewing four decades of Australian poverty research', Australian Journal of Social Issues, 40 (1), 13-32.

--(2007) 'The costs of disability and the incidence of poverty', Australian Journal of Social Issues, 42 (4), 461-80.

--(2013) 'Researching poverty: methods, results and impacts', Economic and Labour Relations Review, 24 (2), 205-218.

Saunders, R & Naidoo, Y. (2009) 'Poverty, deprivation and consistent poverty', Economic Record, 85 (271), 417-432.

Saunders, P., Naidoo, Y. & Griffiths, M. (2008) 'Towards new indicators of disadvantage: deprivation and social exclusion in Australia', Australian Journal of Social Issues, 43 (2), 175-194.

Saunders, P. & Wong, M. (2012a) Promoting Inclusion and Combating Deprivation: Recent Changes in Social Disadvantage in Australia, Social Policy Research Centre, Sydney, University of New South Wales.

--(2012b) 'Estimating the impact of the global financial crisis on poverty and deprivation', Australian Journal of Social Issues, 47 (4), 485-503.

Saunders, P., Wong, M. & Bradbury, B. (2014) Poverty in Australia: New Estimates and Recent Trends--Research Methodology for 2014 Report, Social Policy Research Centre, University of New South Wales.

Sen, A. K. (1985) Commodities and Capabilities, Amsterdam, North-Holland.

Senate Community Affairs References Committee (2004) A Hand up Not a Hand Out: Renewing the Fight Against Poverty. Report on Poverty and Financial Hardship, Canberra, The Senate.

Siminski, P., Saunders, P. & Bradbury, B. (2003) 'Reviewing the inter-temporal consistency of ABS household income data through comparisons with external aggregates', Australian Economic Review, 36 (3), 333-349.

Skattebol, J., Saunders, P., Redmond, G., Bedford, M. & Cass, B. (2012) Making a Difference: Building on Young People's Experiences of Economic Adversity, Social Policy Research Centre, University of New South Wales, Sydney.

Stanton, D. (1980) 'The Henderson poverty line--a critique', Social Security Journal, December, 14-24.

Stewart, K. (2012) 'Child poverty: what have we really achieved?', in L. Judge (ed.), Ending Child Poverty by 2020. Progress Made and Lessons Learned, London, Child Poverty Action Group.

Townsend, P. (1979) Poverty in the United Kingdom, Harmondsworth, Penguin Books.

United Nations Economic Commission for Europe (2011) Canberra Group Handbook on Household Income Statistics, Second Edition, New York & Geneva, United Nations.

UNICEF (2013) Child Well-Being in Rich Countries. A Comparative Overview, Report Card No. 11, Florence, UNICEF Office of Research.

--(2014) Children of the Recession: The Impact of the Economic Crisis on Child Well-Being in Rich Countries, Innocenti Report Card No. 12, Florence, UNICEF Office for research.

UNICEF Canada (2012) Poverty ... The One Line We Want Our Children to Cross. UNICEF Report Card No. 10: Measuring Child Poverty--Canadian Companion, Toronto, UNICEF Canada.

Whiteford, P. (2001) 'Understanding poverty and social exclusion: situating Australia internationally', in R. Fincher and P. Saunders (eds), Creating Unequal Futures? Rethinking Poverty, Inequality and Disadvantage, Sydney, Allen & Unwin, 38-69.

--(2010) 'Why unemployment benefits need to be increased', Inside Story: National Affairs, 7 December, http://insidestory.org.au (accessed 20 February 2015).

Williams, R. (2012) The Policy Providers. A History of the Melbourne Institute of Applied Economic and Social Research 1962-2012, Melbourne, Melbourne University Press,

Wilkins, R. (2007) 'The Changing Socio-Demographic Composition of Poverty in Australia: 1982 to 2004', Australian Journal of Social Issues, 42 (4), 481-501.

Wilkins, R. (2013) 'Relative income poverty' in R. Wilkins (ed.), Families, Incomes and Jobs, Volume 8. A Statistical Report on Waves 1 to 10 of the Household, Income and Labour Dynamics in Australia Survey, Melbourne Institute of Applied Economic and Social Research, University of Melbourne, 23-29.

--(2014a) 'Evaluating the evidence on income inequality in Australia in the 2000s', Economic Record, 90 (288), 63-89.

--(2014b) 'The distribution and dynamics of household income' in R. Wilkins (ed.), Families, Incomes and Jobs, Volume 9. A Statistical Report on Waves 1 to 10 of the Household, Income and Labour Dynamics in Australia Survey, Melbourne, Melbourne Institute of Applied Economic and Social Research, University of Melbourne, 23-32.

Endnotes

(1) As Williams (2012: 19) recently observed, when Prime Minister MacMahon was discussing the changes to social security payments introduced in the 1972 federal budget, he described the research as 'a study in depth by people who know what they are talking about'.

(2) The 2011-12 ABS Survey of Income and Housing indicates that median (equivalised) income was $790 a week, which implies that the poverty line (50 per cent of the median) was $395 a week (ABS 2013a: Table 1). At the mid-point of that year (that is, at the end of December 2011) Newstart Allowance for a single adult who received the maximum rate of rent assistance was $303 a week, or 23.3 per cent below the poverty line.

(3) The most recent changes to the ABS' income measure are claimed to have mainly affected those in the top two quintiles of the income distribution (ABS 2013a: Appendix 4), although studies by Wilkins (2014a) and Saunders, Wong and Bradbury (2014) indicate that they have caused the poverty rate to vary by around one percentage point in some years. It is also important to acknowledge that the ABS has made substantial progress over the last two decades in making its data more accessible to users, including provision of data to the research community in the form of confidentialised unit record files (CURFs).

(4) The ABS did publish estimates of the percentage of households with income below 50 per cent of the median in the report Measuring Australia's Progress (ABS 2002b). However, the headline progress dimension into which that indicator fell ('Economic Disadvantage and Inequality') was replaced in 2004 by the dimension 'Financial Hardship' that 'covers material better suited to discussions of progress in this area' (ABS 2004:v) and the two income-based poverty indicators were dropped.

(5) A recent UNICEF report notes that the 'failure to protect and promote the well-being of children is associated with increased risk across a wide range of later-life outcomes ... [including] ... impaired cognitive development to lower levels of school achievement, from reduced skills and expectations to lower productivity and earnings, from higher rates of unemployment to increased dependence on welfare, from the prevalence of antisocial behaviour to involvement in crime, from the greater likelihood of drug and alcohol abuse to higher levels of teenage births, and from increased health care costs to a higher incidence of mental illness' (UNICEF 2013: 4).

(6) In 2012, public spending on cash benefits in Australia accounted for 8.3 per cent of GDP, compared with an OECD average of 12.8 per cent. Australia's spending ratio ranked 28th out of 32 countries, with only Chile, Iceland, Korea and Mexico spending less. See OECD Social Expenditure database, www.oecd.org/social/expenditure.htm (accessed 10 September 2014). In 2011, Australia also ranked low--25th out of 34 countries--in terms of its poverty rate of 14.6 per cent, which was over 3 percentage points above the OECD average (See OECD 2014a: Figure 5; OECD 2014b).

(7) AJSI has published several important papers on social exclusion, including one Special Issue devoted to the topic (Volume 45, No. 2) and another due for release in 2015.

(8) In a forerunner to later developments, Stanton (1980: 22) argued that the Henderson poverty line was flawed in that it relied on an equivalence scale based on New York spending patterns in 1954. More importantly, Stanton argued there was need for 'a major improvement in the area of descriptive statistics' that should include 'an empirical examination of the actual living conditions of major groups in society ... to assess how in fact low income groups cope and what problems they actually face'.

(9) McDonald (2013:3) has recently argued that the capacity of the Australian safety net 'to protect people from poverty is dissipating and the more prevalent indicator of poverty is receipt of payment ... As a result of this system, Australia's poor languish; their incomes are manifestly inadequate'.

(10) Wilkins' analysis of changes in income inequality lead him to conclude that the measurement changes introduced by the ABS exaggerated the recent increase in inequality, particularly between 2002-03 and 2005-06 by capturing more of the income of those at the top of the distribution.

(11) This lack of agreement has been exaggerated and exploited by those who are critical of the concept of relative poverty and of all research based on it. There has in fact been wide agreement among Australian poverty researchers since the early 2000s about where to set the poverty line (at 50 per cent of median income) and how to adjust for differences in need (by using the modified OECD scale). The only major Australian report that has not followed this approach (published by NATSEM in conjunction with The Smith Family: see Harding, Lloyd & Greenwell 2001) was heavily criticised at the time, but subsequent NATSEM poverty research studies reverted to applying the standard approach described above (for example, Lloyd, Harding & Payne 2004; Phillips et al. 2013).

(12) In contrast, the social inclusion indicator framework developed by the previous government's Australian Social Inclusion Board (ASIB) does not include the poverty rate, but does include measure that combines low income with other indicators such as low wealth and/or reported financial stress (Australian Government 2012).

(13) The EU poverty alleviation target uses a poverty line at 60 per cent of median income and it is common for international agencies such as the OECD (2008) to use poverty lines set at 40 per cent, 50 per cent and 60 per cent (see OECD 2008: Figure 5.1).

(14) These estimates have been produced as a byproduct of research commissioned by ACOSS on behalf of a consortium of community sector agencies (see ACOSS 2014) and reflects the input of my colleagues Melissa Wong and Bruce Bradbury. Neither of these individuals, nor ACOSS, nor any of the organisations who financially supported the research are responsible for any errors or for the views expressed here.

(15) Table 1 has been derived from the confidentialised unit record file (CURF) based on the 2011-12 SIH available in August 2014. An error in the recording of income from the Disability Support Pension on the CURF has been corrected. Households containing individuals who are self-employed or who report zero or negative income have been excluded. Further details of the methods used to produce the estimates are provided in Saunders, Wong and Bradbury (2014).

(16) Another aspect of the approach that has received far too little attention relates to the fact that the equivalence scale only takes account of how needs differ between adults and children and according to household size. This makes the approach ill-suited to examine poverty among groups with specific needs such as Indigenous Australians and people affected by disability or care responsibilities.

(17) This would go some way towards capturing the living standards approach to poverty measurement that is captured in the ABS comment quoted earlier.

(18) Further information about the data and methods used to derive the estimates in Table 2 is provided in Saunders and Wong (2012a).

(19) For example, the deprivation rate for the item 'regular social contact with other people' is 4.9 per cent using the conventional (majority support) approach but 10.1 per cent when identified on the basis of individual preferences. In contrast, the deprivation rate for the item ' a week's holiday away from home each year' declines from 19.8 per cent (majority support) to 16.8 per cent (individual preferences) (see Saunders & Wong 2012a: Table 5.2).

(20) For analysis of relative and absolute poverty rate changes during the Howard Government years, see Rodgers, Siminski and Bishop (2009). The use of an anchored 'real' poverty line to measure movements in 'absolute' poverty has become an accepted approach among international organisations like the OECD (2008: Figure 5.4) and UNICEF (2014)--although only as a supplement to relative measures.

(21) Baum (2004) uses the term 'deprivation' to describe low socio-economic status areas, but the classification is not based on the approach associated with the work of Townsend.

(22) In this context, the word 'plausible' is interpreted to mean that the items identified as essential conform to commonsense understanding of how society functions and is changing. In relation to robustness, all but one of the 25 items identified by a majority as being essential 'for all Australians' in 2006 after validity testing (see Saunders & Naidoo 2009) were also seen as essential by a majority in 2010, and no new items exceeded the cut-off. The one item that was dropped in 2010 was 'a separate bedroom for each child aged over 10' which also received just less than 50 per cent support in 2006, but was included after examining the age breakdown of responses.

(23) The Survey of Income and Living Conditions (SILC) provides the corresponding data for EU countries that is then used to monitor progress against the EU Social agenda poverty reduction targets.

(24) The ABS financial stress questions were included following the recommendations of the Adequacy Project conducted by the (then) Department of Social Security (DSS) in the 1990s. They were seen at the time as providing a new framework for poverty measurement in Australia (DSS 1995). The approach did not survive the change of government in 1996.

(25) In the first instance of all the 26 items identified as necessary by a majority of those responding to the SPRC surveys will be included in the HILDA household questionnaire. Their inclusion in the personal questionnaire was not possible because of time and space constraints. The self-completion questionnaire was also considered as an option, but worked less effectively in the piloting phase. Minor changes have been made to the wording of some questions following feedback provided during piloting.

(26) Four targets have been set for achievement by 2020: reduce the percentage of children living in households with incomes less than 60 per cent of the median (BHC) to less than 10 per cent; reduce the percentage of households experiencing material deprivation and with an income less than 70 per cent of the median (BHC) to less than 5 per cent; reduce persistent poverty among children to less than 7 per cent; and reduce the anchored poverty rate to less than 5 per cent (see Kennedy 2014: 5). The persistent poverty target has been proposed by the government and been the subject of community consultation, but has not yet been finalised. Persistent poverty is defined as being in relative poverty for at least three years over a four-year period (HM Government 2014).

(27) As Blank (2008) has noted, any change to the historical US poverty definition would have to come from the Executive Office of the President, which oversees the Office of Management and Budget, responsible for producing the US poverty statistics. Given the political difficulties and risks associated with this, she argues that the only realistic way forward is to supplement the existing measure with new ones.

(28) Brewer (2012: 28) notes that even though the British government 'massively increased' the level of spending on financial support for families with children, it did not achieve its interim target of halving the child poverty rate by 2010. Even so, he observes that a 900,000 reduction of the numbers of children in poverty was a 'remarkable achievement' (p. 30).

(29) If the focus on the acceptability of living standards is to be maintained, these increases will eventually affect which items to include when measuring deprivation.

Table 1. Estimated income poverty rates in 2011-12 (percentages)

                          Based on the 2007-08   Based on the 2005-06
                             income measure:       income measure:

                           Before     After      Before     After
                          housing    housing    housing    housing
Family type                costs      costs      costs      costs

Single non-aged (15-64)     24.0       29.4       21.8       27.6
Single aged (65+)           45.3       25.2       34.1       20.1
Non-aged couple (15-64)     6.9        8.4        5.4        7.4
Aged couple (65+)           25.3       12.7       15.0       10.3
Couple with children        6.8        11.6       6.6        12.1
Lone parent                 21.9       33.7       16.5       28.6
Mixed family households     6.1        8.4        5.0        7.3
All households              11.9       13.9       9.6        12.8
Number of poor            2,192.4    2,548.5    1,765.1    2,350.5
  individuals ('000)
Number of poor             362.9      602.6      323.7      577.3
  children ('000)

Notes: The poverty line is set at 50 per cent of the
relevant median in each case. Family type definitions are
based on the characteristics of the Household Reference
Person (HRP).

Source: ABS, Survey of Income and Housing 2011-12;
confidentalised unit record file.

Table 2. Income poverty, deprivation and living standards indicators,
2010

                               Income                         Neither
                               poor    Deprived   Poor and   poor nor
Living Standards Indicator     only      only     deprived   deprived

Estimated percentage           7.7       11.9       5.4        75.0
Objective:
Does not have enough           7.5       21.9       37.4       2.1
  to get by on
Had to go without food         4.1       14.9       34.4       1.8
  when hungry
Got behind with rent           5.8       26.7       31.4       4.6
  or mortgage
Couldn't keep up with          13.6      46.7       50.7       7.9
  utility bills
Had to pawn, sell              12.8      29.5       30.1       3.2
  or borrow money

Had to ask a welfare agency    1.9       9.8        22.3       0.9
  for support
Wore bad-fitting or worn-      12.7      44.2       44.2       5.6
  out clothes
Unable to go out with friends  31.7      67.1       72.2       13.4
  and pay one's way
Subjective:

Describes self/family as poor  9.0       40.3       67.7       4.5

Self-rated standard of         3.09      2.63       2.25       3.40
  living (5-point scale)
Life satisfaction              3.69      2.73       2.53       3.90
  (5-point scale)

Happiness (4-point scale)      3.00      2.64       2.46       3.14

Autonomy and control           6.64      5.72       5.27       7.38
  (10-point scale)
Satisfied with financial       5.38      3.80       3.36       6.43
  situation (10-point scale)

Source: Author estimates derived from the Poverty and Exclusion in
Modern Australia (PEMA) survey (see Saunders and Wong 2012a).