Pension adequacy and the pension review.

Saunders, Peter ; Wong, Melissa

1. Introduction

The intense political debate that preceded the government's establishment of the Harmer Pension Review illustrated that, in certain circumstances, payment adequacy is an issue that resonates with the public and taps into deep-seated views about deservingness and the ethos of the 'Fair Go'. Claims by then Opposition Leader Brendan Nelson that pensioners were forced to 'get by on baked beans and jam sandwiches' focused public attention on a feature of the social security system that had been largely neglected in broader debates about welfare reform and population ageing. This reflected the fact that the automatic indexation of (most) payments provided regular increases that took the spotlight off the base to which these increases were applied. Asked to respond to the Opposition's concerns, Families and Community Services Minister Jenny Macklin agreed that she could not live on the existing pension of $273 a week, paving the way for the establishment of the Harmer Review, which led eventually to a substantial increase in the single rate of pension (Harmer 2009). The need for such a review was reinforced by community concern over cost-of-living pressures which had emerged in the wake of the global financial crisis and other developments that had pushed upwards the price of many basic items, including food, petrol, medical care and rents (Harmer 2008: v).

At first glance, it is somewhat surprising that a government which, like its predecessor, was keen to ease the financial pressures resulting from the ageing of the population would even contemplate a change that would add directly to current and projected budgetary costs. It was clear from the outset that the review was pursuing a largely predetermined agenda and would recommend an increase in the pension, the only issue being how large it would be and who would benefit. In the event, the government managed to offset the longer-run budgetary effects by introducing a phased increase in the pension eligibility age when the pension rise itself was announced in the May 2009 Budget. Even so, the fact that pensions were raised at all seems at odds with the underlying demographic imperatives, although this can be explained by the increasing numbers of older people, public support for older Australians, and the growing political voice and influence of the 'grey vote'.

More generally, this whole episode illustrates how political forces can distort the underlying economic logic and disrupt what might otherwise be an orderly process of policy reform. At one level, the events leading up to the establishment of the Pension Review can be seen as an example of where policy coherence can give way to political expediency. However, it can also be seen as a vivid illustration of the powerful impact that concern about the adequacy of payments can have on policy when the underlying arguments (about pension adequacy), the evidence (about pensioners doing it tough) and political interests (of older people and other pensioners) coalesce and are mobilised in a coordinated way. This was achieved by focusing the debate on the adequacy of the Age Pension and on the financial difficulties of a group that is not only large enough to exert political influence, but is also widely seen as 'deserving'. There seems little doubt that had the focus become broader to include groups such as the unemployed and sole parents that are seen as far less deserving of additional (even existing) levels of support, the adequacy arguments would have been less compelling and the political support weaker and hence less influential.

However, the events leading up to the pension increase placed the issue of payment adequacy at the centre of the policy agenda (albeit temporarily), and it is reasonable to examine the wider implications of the evidence and arguments used to support the decisions that were taken. This is the main aim of this article. We adopt a particular perspective on the adequacy issue that is based on work on deprivation originally developed to study poverty by Townsend (1979) and refined by Mack and Lansley (1985), Nolan and Whelan (1996) and Pantazis et al. (2006). These methods have been applied in Australia by Saunders et al. (2007) and we have drawn on that work to examine some of the arguments developed in the Pension Review in an earlier article (Saunders and Wong 2011a; see also Saunders 2011).

This article adds to this literature by examining two specific issues: first, it provides a more thorough discussion of the concept of adequacy and explains why the deprivation framework is particularly suited to addressing adequacy questions; second, it extends the empirical analysis of payment adequacy reported in Saunders and Wong (2011a) by using new survey data to examine the impact of the pension changes that were introduced following the Pension Review.

The article is organised as follows. Section 2 provides a brief review of the meaning of adequacy and how it has been interpreted and applied in the Australian policy context. Section 3 then outlines the deprivation approach and explains why it is a useful tool for studying adequacy questions. Section 4 describes the data and methods we use to examine adequacy, while Section 5 presents results on the adequacy of different income sources and how they have changed since the review recommendations were implemented. Finally, Section 6 summarises our main conclusions.

2. Defining Adequacy

It is important to emphasise at the outset that the review's Terms of Reference restricted it to addressing only the Age Pension, Carer Payment and Disability Support Pension, although it acknowledged (Harmer 2009: 1) that many of the issues it considered 'are also relevant to the wider system of pensions and allowances'. (2) While this narrow focus can be explained by the political factors that had led to the establishment of the Harmer Pension Review, it is important to acknowledge that changes implemented in one area will have consequences for the system as a whole. Thus, if adequacy improvements are restricted to some payments only, this may prevent the system as a whole from achieving some of its other design objectives. It may, for example, make the system as a whole less equitable in its treatment of different groups, and create greater incentives for recipients to move between different payment categories. (3) These broader, system-wide issues are important although they are not discussed further below, where the focus is on the narrower approach adopted by the Pension Review.

It has long been accepted that the adequacy of payments is one of the criteria used to assess the degree to which the social security system is achieving its objectives. Although not referred to specifically by this term, achieving adequacy is the first of the five design principles identified in the Pension Review Background Paper (Harmer 2008: 4), which is its ability to support 'a basic acceptable standard of living, accounting for prevailing community standards' (emphasis in the original). This goal was identified as 'the central question for the Review' in the report itself, which noted that:

The Review's approach to this question was to test whether current rates of pension are providing a basic acceptable standard of living, accounting for prevailing community standards. The Review considered that the full rate of pension should provide a basic acceptable standard of living for those who are wholly reliant on it, often for extended periods of time, without any assumptions about access to private income or assets ... while the question of adequacy can be conceived of in both absolute and relative terms, ultimately it needs to be answered in the context of contemporary society, and the living standards of others. (Harmer 2009: xii-xiii)

The key features of this definition of adequacy are consistent with the approach taken in other reviews of the Australian social security system undertaken over the last three decades. (4) Thus, as input into the assessment of the family benefits system undertaken in the 1980s by the Social Security Review, Whiteford (1985: 24) argued that:

Adequacy [exists when] the level of income support is sufficient to allow individuals and families to achieve an adequate standard of living ... [and] may be judged by reference to a poverty line or relative to other standards of living in the community.

The same broad approach was adopted when the Department of Social Security (DSS) embarked upon a comprehensive review of the adequacy of its payments a decade later (see Nicolau 1998), where it was argued that:

 Statements regarding the adequacy of incomes are essentially based
 on community attitudes and values ... A core criterion for both
 provider and recipient is likely to be that DSS payments are
 adequate if they allow the recipient to live modestly but decently
 within his or her community, having regard to that community's
 norms and values. (Strategic Development Division 1995: 1)

Two key ideas are embedded in all three of these definitions. The first is the notion that, in the social security context, adequacy is reflected in the ability of a given level of income to support a standard of living that is acceptable; the second is the idea that the notion of acceptability itself can only be judged with reference to prevailing community standards, as reflected in the norms and values that exist in society at a particular point in time.

It is also clear from these definitions, however, that adequacy is an elusive concept that defies a precise definition. Terms such as 'basic', 'acceptable', 'modest' and 'decent' do not lend themselves to precise definition, while the notion of the standard of living itself presents formidable conceptual challenges (see Sen 1987). There is the added complexity that an income level that may be adequate for one person may not be so for another, because of differences in their needs, even though both may be eligible for the same social security payment. While some of these differences are recognised by the payment system through the existence of supplementary payments (such as for those who are renting), others are not, so that any specific payment may be adequate for some but inadequate for others. It is not possible to resolve all of these complications, so it is normal to think of adequacy as applying on average, rather than in every specific instance.

Reflecting these problems, the Henry Tax Review cautioned against drawing definitive conclusions about payment adequacy, arguing that:

 There is no single agreed definition of income poverty, nor is
 there an agreed way to measure the adequacy of support rates. What
 would have been seen as an adequate level of payment according to
 'community standards' in 1950 might not be seen as adequate today,
 and what might be seen as adequate today might not be seen as
 adequate in 2020. Aside from changes in living standards and the
 distribution of work and income, views about the best way to avoid
 or alleviate poverty can also change. (Treasury 2010: 495)

The first point is obviously correct and is captured by the idea that adequacy can only be judged against prevailing standards, as the earlier definitions acknowledge. The final point--that changes in how we respond to poverty will influence how we judge adequacy--seems to confuse the issue of identification with the question of how to respond to it and does not logically follow. The important point that is highlighted in this statement is the difficulty of deciding how changes in overall living standards should affect judgments about income adequacy--this is itself a judgment, although it is also implicit in the definitions cited earlier.

In light of these difficulties, some assessments of adequacy seek to provide only an ordinal ranking of different situations. The most common approach involves arguing that if payment levels have increased over time, they must have become more adequate. However, such comparisons are only valid if the methods used to produce them are consistent with the underlying notions of adequacy. Thus, if payment levels are adjusted for changes in the CPI but not for improvements in real community incomes, the implicit assumption is that adequacy is being evaluated in an absolute sense that takes no account of changes in general living standards (which would require a more explicitly relative approach). Such an approach thus does not avoid the problems involved in judging adequacy, but merely embodies a specific assumption that may or may not be relevant to 'prevailing community standards'.

A more demanding approach to adequacy requires that an assessment is made that is not merely ordinal ('payment A is more adequate now than it was in the past') but provides a cardinal ranking against an external benchmark ('payment A is adequate now but was inadequate 10 years ago' or 'payment A is adequate but payment B is not'). It is this latter approach that is relevant when judging the quality and performance of the social security system, but it requires an external benchmark against which the adequacy of the income provided by a particular payment can be assessed in terms of whether or not it can support an acceptable, modest or decent standard of living. The most common such benchmark is a poverty line, and comparisons of the incomes provided by the social security system with poverty lines defined for specific family types remain important in Australia because of the prominence given to poverty alleviation in the design and structure of the system (see ACOSS 2010; Melbourne Institute 2010).

Nevertheless, poverty lines have come under attack for being too narrowly focused on income, for embodying arbitrary assumptions about needs and how much is required to meet them, and for not being grounded in the living standards actually experienced (Saunders 2005: 2011). One specific problem with the use of poverty lines to assist in establishing the relative adequacy of the single and married rate of pension--one of the principal goals of the Pension Review--is that the results will depend on the equivalence scale embedded in the poverty line, which may itself be challenged. The two most common poverty lines used in Australia are the Henderson poverty line (see Johnson 1996) and the international poverty line set at 50 per cent of median income (see OECD 2008; Wilkins 2008). (5) Neither is useful by itself to judge adequacy, Henderson because it is outdated (and was originally based on relative needs estimated for New Yorkers) and the international line because it does not relate specifically to Australian needs and circumstances. The Pension Review went further, noting (Harmer 2009: 34) that 'it considers neither of these [poverty lines] to be a particularly robust measure of wellbeing, and it is difficult to dispute the general claim that existing poverty lines should thus be used with caution when judging the adequacy of Australian pensions.

Rather than commit itself to any single benchmark, the Pension Review used a range of benchmarks to assess the adequacy of the pension. It also avoided some of the more difficult conceptual problems by focusing on the relative (to each other) adequacy of the single and married rate of pension. Two cardinal/threshold indicators were used--a relative poverty line based on 50 per cent of median income and the updated budget standards originally developed by the Social Policy Research Centre (Saunders et al. 1998). In addition, three ordinal approaches were examined, the first based on changes in the real (CPI-adjusted) and relative (minimum wage and average earnings-adjusted) values of the pension, and two indicators of well-being, self-assessed prosperity and the incidence of adverse outcomes associated with financial stress (Harmer 2009, Table 3 and Charts 4-12). There was also a brief mention (ibid.: 43) of the results produced by the deprivation approach described further below, although no specific results were presented.

The review did not try to identify which method was superior--a formidable challenge, given the problems identified above. Instead, it followed the approach set out in the Background Paper, which argued (Harmer 2008: 14) that '[t]here is a need to look to a range of measures to judge whether current payment rates are appropriate' and drew on the evidence from the range of approaches examined. This is perhaps understandable in the circumstances, but the review's failure to reach some more definitive conclusions about the relative merits of the different approaches represents a lost opportunity that could, had it been pursued, have been of enduring value. (6)

3. The Deprivation Approach

The deprivation approach seeks to identify who is unable to afford items that are widely regarded as essential. In order to achieve this, it is necessary to conduct a survey in which people are asked which of a list of items they regard as essential, where that term is defined as covering 'things that no-one should have to go without in Australia today'--thus, they are asked which items are essential for people in general, not just for themselves. The items included in the list should not be arbitrarily selected but should reflect the experiences of those living in, or close to, poverty. In our case, this was achieved by obtaining the views of low-income Australians, obtained from focus group discussions in which they were asked to identify which items are needed to achieve a minimal but decent standard of living (Saunders and Sutherland 2006). The items that emerged from this process were supplemented by those used in overseas deprivation studies (conducted in New Zealand, Ireland and Britain) and some of the items used to identify hardship and financial stress in Australia (Bray 2001; Wilkins et al. 2011).

Only those items that receive majority support for being essential constitute the essentials of life--items that meet basic needs to an acceptable community standard in contemporary circumstances. Those surveyed were also asked if they had each item and, if they did not, whether the lack of the item was because they could not afford it. Deprivation was then identified as existing when someone did not have and could not afford items that were regarded as essential by a majority in the community (see Saunders 2011; Saunders et al. 2007). (7)

Clearly, being deprived provides a better indication of who is actually experiencing poverty than simply comparing someone's income with a poverty line: whereas the latter approach implicitly assumes that low income equates with poverty, the deprivation approach seeks to establish that a lack of economic resources results in an unacceptable standard of living. Importantly, the use of focus groups and a community survey to identify which items are essential grounds the deprivation approach in the experience of poverty in a way that reflects prevailing community norms. Most importantly, by identifying who cannot afford items regarded as essential by a majority, the approach provides direct evidence on what constitutes an unacceptable standard of living that is consistent with prevailing community standards. It thus conforms directly to the definitions of adequacy identified earlier and is therefore particularly suited to inform questions about the adequacy of people's incomes. (8)

The approach is not without its problematic elements, however, and this has led to some criticism that it is no less arbitrary in some respects than setting a poverty line. One example is the use of majority support to identify which items are essential, an approach that has been criticised for being arbitrary--why not set a benchmark level of support at 75 per cent or 90 per cent? In response, Gordon (2006) has argued that the use of majority rule has political validity and is easily understood by ordinary people. Some have argued that the distinction between essential and non-essential items (or between necessities and non-necessities, in economic jargon) is flawed on both conceptual and practical grounds--the former because of the diverse nature of individual needs and preferences, the latter because the evidence shows that many people who lack necessities do own many non-necessary items, an observation that seems at odds with the former items being necessary (and hence acquired before other items) (see McKay 2004; Van den Bosch 2001). In addition, many economists would also argue that asking people whether or not they can afford items that they do not have is incapable of distinguishing between choice and constraint in observed patterns of consumption and ownership, with the result that the responses will capture the effects of both and cannot thus be regarded as entirely objective (Berthoud and Bryan 2008; Brewer et al. 2008).

One final issue relates to aggregation. While it is unproblematic to identify deprivation rates for specific items, there is also general interest in whether or not people are deprived overall. This involves setting a threshold (the deprivation equivalent of a poverty line) that specifies how many items must be lacking before an individual or family is identified as deprived in overall terms. The choice of threshold is again arbitrary, although it is possible to vary the definition and check the sensitivity of the results. How many items constitute overall deprivation will depend on how many essential items there are to start with, although it is common to use factor analysis to try to identify the core items that explain most of the total variation in the complete list and to then define deprivation as existing when any one of these core items cannot be afforded (Nolan and Whelan 1996). An alternative approach (used below) involves calculating a deprivation score equal to the number of essential items that each respondent does not have and cannot afford, and comparing mean scores between subgroups in the population. This avoids classifying people as either deprived or not deprived and allows the living standards of different groups to be compared directly.

Against such criticism (most of which can be responded to by conducting sensitivity analysis: see Saunders and Wong 2011b), the great strength of the deprivation approach is that it avoids many of the major criticisms that have been levelled at poverty line studies. In particular, there is no need to set a poverty line or rely on the judgments of 'experts', since the deprivation approach relies instead on what the community regards as essential. There is also no need to impose an equivalence scale, making the approach particularly suited to examine the living standard relativities between different groups. Thus, for example, if the relativity between the single and married rate of pension is appropriate in the sense that it allows both groups to achieve the same standard of living, then one should find that there is no difference in the average level of deprivation experienced by the two groups.

4. Data and Methods

The results reported and analysed below are based on data collected from two surveys. The Community Understanding of Poverty and Social Exclusion (CUPSE) survey was distributed by mail to 6000 adult Australians randomly selected from the electoral rolls in April 2006. (9) It generated 2704 responses, equivalent to a response rate of 46.9 per cent--somewhat higher than that achieved by other similar social surveys conducted around that time. (10) The detailed comparisons reported by Saunders et al. (2007, Table A.3) indicate that the CUPSE sample is broadly representative of the general population, although the following groups are underrepresented: males; those who have never been married; those who live alone; Indigenous Australians; those with lower levels of education; those in private rental accommodation; and those with incomes between $1000 and $2000 a week. Some of these differences are interrelated, while others may reflect the difficulty involved in conducting a mail survey.11 The Poverty and Exclusion in Modern Australia (PEMA) survey was distributed to a new sample of 6000 adults in May 2010 and generated 2645 responses--equivalent to a response rate of 46.1 per cent. (12) It replicated the CUPSE deprivation questions, and also included additional questions relating to the impact of the global financial crisis and aspects of community participation and location.

Both surveys included the three key questions identified earlier relating to a series of potential essential items (61 in the case of CUPSE, 73 in the case of PEMA). The items themselves included a substantial meal at least once a day, a car, a television, being able to buy prescribed medications, regular social contact with other people, and having up to $500 in savings for an emergency. As indicated earlier, only those items that attracted majority support for being essential were identified as constituting 'the essentials of life. A total of 26 items satisfied this condition in 2006, although one of these (the television) was subsequently dropped after conducting reliability and validity tests (see Saunders and Naidoo 2009). All but one of the remaining 25 items (a separate bedroom for older children) also received majority support for being essential in 2010, and this item was also dropped, bringing the number of items regarded as essential in both years down to 24. (13) A further five items were excluded because they relate specifically to the needs of children and are thus not relevant when examining deprivation among older people in receipt of an Age Pension, leaving 19 items as the basis of the following analysis. (14) The items themselves are identified in Table 1, which also shows the (age-weighted) percentages that regarded each item as essential in each year. (15)

Two measures of deprivation were examined: the first was the mean deprivation score, derived by summing the number of essential items that people did not have and could not afford and averaging across relevant groups; the second was the percentage in each group that was deprived of at least three of the 19 adult essentials of life. The initial results (see below) indicated that both measures produced very similar rankings of the groups considered, and so the latter measure was subsequently dropped. The groups themselves were defined in order to provide an insight into the issue of payment adequacy being addressed by the Pension Review, but were broader in scope in order to provide a wider perspective on its findings. The approach involved identifying groups on the basis of their main source of income in the week preceding each survey, an approach that allowed five groups of social security payment recipients to be identified that differed according to the payment received. Two further groups were included in the analysis to provide a broader perspective on the findings: low-wage earners and self-funded retirees. Although identification of these latter two groups involved making some assumptions, their inclusion expands the scope of the analysis and generates results that are of interest in their own right.

The definitions of each group and the sample sizes in each year are presented in Table 2. It is clear that some of the samples are very small and this has obvious consequences for the robustness of the estimates, which needs to be borne in mind when reviewing the results.16 Those respondents in each group who were living with others in mixed household arrangements were removed from the samples shown in Table 2, further reducing sample sizes in some instances. Although the number of age pensioners is quite large in both years, this group will be disaggregated by payment type later and this will also generate some rather small samples.

5. Main Findings

Figure 1 compares the values of the two deprivation measures in 2006 across the income groups identified in Table 2. It is clear that the two measures produce very similar results, and so the following discussion focuses on the mean score differences only. On average, those reliant on the Age Pension were deprived of about one essential item, similar to the level of deprivation among service pensioners. In contrast, there was almost no deprivation among the self-funded retiree group, a finding which provides reassuring evidence that the pension income and assets test are doing their job, and that the mean deprivation scores do indeed track the living standards of each group.

However, the most striking aspect of the results in Figure 1 is the high levels of deprivation among the other groups included in the analysis. Thus, in round terms, low-wage workers were deprived of around two essential items, disability pensioners of around three items, Newstart allowees of around four items, and sole parents of close to five items. All of these latter levels of deprivation are significantly higher than that experienced on average by the age pensioner group. These results thus imply that however strong the case may have been for improving the adequacy of the Age Pension in 2006 (when the first survey was conducted), there was a far stronger case on adequacy grounds for improving the payments to the last three groups listed in Table 2, particularly sole parents.

[FIGURE 1 OMITTED]

It is also interesting to note from Figure 1 that deprivation is higher among disability support pensioners than among age pensioners, even though the basic rates of payment were the same for both groups and despite the fact that deprivation among those disability pensioners with children has been understated by the removal of the child-related essential items. This provides compelling evidence that the needs of the disability group are, on average, higher than those of the age pensioner group, and that the incomes they receive are thus less able to support the same standard of living. Put differently, even though the Disability Support Pension was set at the same level as the Age Pension in 2006, it was less adequate.

It has already been noted that a focus of the Pension Review was on the relative adequacy of the payments to different groups of age pensioners, specifically on the relativity between the single and married rate of pension. This issue is examined in Figure 2, which compares the two deprivation measures for different categories within the age pensioner group defined on the basis of gender, age, living arrangements and housing tenure. (17) This disaggregation reduces sample sizes (shown under the group names in Figure 2) and makes it harder to establish that the differences are statistically significant, but the general patterns are nonetheless interesting. There are clear differences in the mean deprivation scores among age pensioners classified on the basis of age and whether they are single or living as a couple, with the average difference being around one in each case (meaning that the more deprived group was deprived of one more essential item in each case). However, by far the most marked difference is that based on housing tenure, with the average level of deprivation among renters approaching five times that among homeowners or purchasers. (18)

[FIGURE 2 OMITTED]

Again, it is clear that there were inadequacies in the payment structure that existed for age pensioners in 2006. Recalling that the average deprivation score among age pensioners as a whole was around one (Figure 1), the results in Figure 2 indicate that payments were far less adequate for those pensioners living alone and those in rented accommodation. (19) Even though the numbers are small, these differences are statistically significant and the fact that the two deprivation measures produce similar results is further evidence that the underlying picture is not being distorted. It is difficult to conclude other than that the pensioner payment structure was in need of review in 2006, although the political storm that erupted in the wake of the 'baked beans and jam sandwiches' depiction should have alerted policy makers to the inadequacies of the support provided to renters as well as to those living alone.

The above approach is now used to examine the impact of the changes that were introduced to the Age Pension (and other pension) levels following the Pension Review. This involves replicating the results in Figure 1 and using the data for 2010 and comparing changes between 2006 and 2010. Before presenting the results, it is important to note that the numbers in some of the groups listed in Table 1 in 2010 were below those shown for 2006, and this was a particular problem for the Newstart allowee and sole parent pensioner groups. This had the effect of making it harder to establish that the observed changes were statistically significant, and in fact none of the differences shown below in Figures 4 and 5 are significant.

Before presenting the results, it is useful to present some background information about what happened to key economic variables in the four years between June 2006 and June 2010 (when both surveys were in the field). Figure 3 summarises movements in the CPI, several income aggregates and the payments received by specific groups. It is clear that real incomes increased sharply over the period--and this resulted in increases in the real value of average earnings and household incomes, as well as in both the Henderson and international poverty lines. And while those in receipt of allowances experienced no real income growth (because their benefits were indexed to prices), the real incomes of age pensioners and sole parents both increased because payments were indexed to movements in earnings. The increase between 2006 and 2010 for single age pensioners (who benefited from the indexation arrangements as well as the substantial rise that followed the Pension Review) exceeded 35 per cent in nominal terms, or around 24 per cent in real terms.

[FIGURE 3 OMITTED]

The changes in deprivation experienced by the groups identified in Table 1 are shown in Figure 4. Consistent with the picture of real income growth (and hence rising living standards) displayed in Figure 3, there was a fall in deprivation (measured using the mean deprivation score, although the pattern is similar if the other measure is used), although the decline was modest for all categories. The largest declines were experienced by sole parents and the unemployed on Newstart Allowance, and in both cases this is probably mainly due to the growth in employment among both groups, reinforced by the one-off payments many received as part of the fiscal stimulus measures introduced in response to the global financial crisis. Mean deprivation among age pensioners declined from around 1.0 in 2006 to 0.85 in 2010. Although this might appear to be a small change--a good deal less than the rise in the real value of the pension shown in Figure 3--it needs to be remembered that the pension increase applied only to single pensioners, whereas the deprivation score shown in Figure 4 is averaged across all pensioner categories, single as well as married. Despite this, the proportionate decline in pensioner deprivation was around 15 per cent--a substantial change over a relatively short period.

[FIGURE 4 OMITTED]

Figure 5 presents the changes in mean deprivation for the different pensioner categories shown in Figure 2. (20) These results indicate that although single pensioners received a much bigger pension increase than those receiving the married rate, the decline in deprivation among single and couple pensioners was of a similar magnitude (around 9 per cent). Differences in sample size may explain part of the difference between what is observed (in Figure 5) and what was expected (given the different pension increases received by different groups of pensioners), although it may also reflect the fact that the mean deprivation score is a sluggish indicator that moves slowly and possibly after a lag (which may not have been picked up in mid-2010, less than nine months after the pension increase was first paid). Even though the pension increase was not targeted specifically on pensioners who rent, Figure 5 indicates that, on average, this group benefited considerably from the increase that was paid, presumably because many renters live alone and hence benefited from the increased single rate of payment.

[FIGURE 5 OMITTED]

It is important to emphasise that the changes shown in Figures 4 and 5 cannot be attributed solely to the pension changes introduced in 2009. Many other changes occurred over the period covered, including the global financial crisis and the policy responses that followed. These provided substantial one-off payments to pensioners and other groups that would have had an impact on the level of deprivation experienced by those most in need. Although the economy stalled in 2008, the anticipated recession did not materialise and by 2010 employment was growing strongly and unemployment had almost returned to its level in 2006. Average incomes increased strongly between 2006 and 2010 (see Figure 3) and, despite widespread community concern (fuelled by political opportunism) about cost-of-living pressures, the increase in average incomes far outstripped the increase in the CPI between 2006 and 2010. Finally, the experience of the financial crisis may have affected people's attitudes in ways that impacted on their willingness to spend or incur debt, and this may in turn have changed the relationship between the rise in average incomes and the decline in deprivation. Finally, it is important to note that the timing of the two surveys was not ideal for assessing the impact of a change that occurred in late 2009, although it is rare for social science to come so close to generating the 'before' and 'after' data needed to conduct a set of comparisons that represent a 'natural experiment.

6. Summary and Conclusions

This article has examined the short-run impact of the substantial increase in the single rate of pension that was paid, following the Pension Review, in late 2009. It began with a discussion of the concept of adequacy--a key feature of the social security system (and an important factor in its interaction with the tax system)--but one that has rarely featured in a policy debate that has become increasingly focused on improving incentives, tightening targeting and containing costs. Unless its payments are adequate, no social security system can hope to maintain living standards and promote social equity. A review of reports produced over the last three decades from within the federal department responsible for the social security system indicates that adequacy can be judged by assessing the ability of payments to sustain a standard of living that is consistent with prevailing community standards of acceptability.

This conception of adequacy makes the deprivation approach particularly well-suited to examining payment adequacy, because it identifies who cannot afford those items that are essential or necessary for everyone to achieve a minimum standard of living. By drawing on community views to identify essential items, deprivation thus provides a benchmark against which the adequacy of social security payments (and other incomes, including the minimum wage) can be assessed. Importantly, the approach avoids the need to set a poverty line expressed in income terms and does not require assumptions to be made about relative needs (as encapsulated in an equivalence scale).

Having outlined its advantages, the deprivation approach was used to examine the relative adequacy of different incomes, including those derived primarily from different social security payments. The results reveal that, when judged using two deprivation measures, the payment structure that existed in 2006 did not provide equally adequate incomes for different groups of payment recipients. On average, age pensioners experienced lower levels of deprivation than those receiving income support because of disability, unemployment or sole parenthood. This suggests that the case for increasing the Age Pension was weaker than the case for implementing improvements elsewhere in the system--most notably, in relation to the levels of support provided to those affected by disability, sole parenthood or unemployment. The deprivation evidence also suggests that, within the age pensioner group, the strongest case for a payment increase applied to those living in rented accommodation--not to those living alone who were the beneficiary of the substantial pension increase awarded following the Pension Review.

When the deprivation approach was used to examine the impact of the increases that were paid in late 2009 on the living standards of different income groups and different categories of age pensioners, the evidence shows that deprivation declined among those groups who received the largest increase, but was not restricted to these groups. In fact, the mean level of deprivation declined among all groups included in this analysis, and even though these declines were not statistically significant, they can be attributed in part to the strong economic conditions that existed over the period examined, despite the global financial crisis. However, many other things changed over this period, making it difficult to identify any single factor as causing the observed declines in deprivation, although the changes in the pension that resulted from the Harmer Review clearly had an impact. The key point, however, is that the deprivation approach has been shown to shed important new light on payment adequacy and, it is hoped, will be used more frequently in future to examine this important social policy issue.

References

Australian Council of Social Service (ACOSS) (2010) Poverty Report: October 2010, ACOSS, Sydney.

Berthoud, R. and Bryan, M. (2008) 'Deprivation indicators' in R. Berthoud and F. Zantomio (eds) Measuring Poverty: Seven Key Issues, Institute for Economic and Social Research, University of Essex, Colchester, pp. 14-15.

Bray, J. R. (2001) Hardship in Australia: An Analysis of Financial Stress Indicators in the 1998-99 Australian Bureau of Statistics Household Expenditure Survey, Occasional Paper No. 4, Department of Family and Community Services, Canberra.

Brewer, M., Muriel, A., Phillips, D. and Sibieta, L. (2008) Poverty and Inequality in the UK: 2008, Institute for Fiscal Studies, London.

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, Policy Press, Bristol, pp. 29-69.

Guio, A-C. (2005) 'Material deprivation in the EU', Statistics in Focus 05/2005, Eurostat, Brussels.

Harmer, J. (2008) Pension Review: Background Paper, Department of Families, Housing, Community Services and Indigenous Affairs, Canberra.

Harmer, J. (2009) Pension Review Report, Department of Families, Housing, Community Services and Indigenous Affairs, Canberra.

Johnson, D. (1996) 'Poverty lines and the measurement of poverty, Australian Economic Review, 1st Quarter, pp. 110-126.

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

McKay, S. (2004) 'Poverty or preference: What do "consensual deprivation indicators" really measure?, Fiscal Studies, 25(2), pp. 201-223.

Melbourne Institute (2010) Poverty Lines, Australia: December Quarter 2010, Melbourne Institute of Applied Economic and Social Research, University of Melbourne, Melbourne.

Nicolau, L. (1998) 'Introduction: Recent poverty research and DSS--an overview of the Adequacy Project', Social Security Journal, 1, pp. 3-8.

Nolan, B. and Whelan, C. T. (1996) Resources, Deprivation and Poverty, Clarendon Press, Oxford.

Organisation for Economic Cooperation and Development (OECD) (2008) Growing Unequal? Income Distribution and Poverty in OECD Countries, OECD, Paris.

Pantazis, C., Gordon, D. and Townsend, P. (2006) 'The necessities of life' in C. Pantazis, D. Gordon and R. Levitas (eds) Poverty and Social Exclusion in Britain: The Millennium Survey, Policy Press, Bristol, pp. 89-122.

Pech, J. (2011) Relative living standards and needs of low-paid employees: Definition and measurement, Research Report 2/2011, Fair Work Australia, Melbourne.

Saunders, P. (2005) The Poverty Wars: Reconnecting Research with Reality, University of New South Wales Press, Sydney.

Saunders, P. (2011) Down and Out: Poverty and Exclusion in Australia, Policy Press, Bristol.

Saunders, P., Chalmers, J., McHugh, M., Murray, C., Bittman, M. and Bradbury, B. (1998) Development of Indicative Budget Standards for Australia, Policy Research Paper No. 74, Department of Social Security, Canberra.

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

Saunders, P., Naidoo, Y. and Griffiths, M. (2007) Towards New Indicators of Disadvantage: Deprivation and Social Exclusion in Australia, Social Policy Research Centre, University of New South Wales, Sydney.

Saunders, P. and Sutherland, K. (2006) Experiencing Poverty: The Voices of LowIncome Australians. Towards New Indicators of Disadvantage Project: Stage I: Focus Group Outcomes, Social Policy Research Centre, University of New South Wales, Sydney.

Saunders, P. and Wong, M. (2011a) 'Using deprivation indicators to assess the adequacy of Australian social security payments', Journal of Poverty and Social Justice, 19(2), pp. 91-101.

Saunders, P. and Wong, M. (2011b) Measurement and change in deprivation and exclusion in Australia: A report on research in progress, paper presented to the Second Townsend Memorial Conference, University of Bristol, 22-23 January.

Sen, A. K. (1987) The Standard of Living, Cambridge University Press, Cambridge.

Strategic Development Division (1995) Developing a framework for benchmarks of adequacy for social security payments, Policy Discussion Paper No. 6, Department of Social Security, Canberra.

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

Treasury (The Henry Review) (2010) Australia's Future Tax System: Final Report to the Treasurer, Part 2: Detailed Analysis, Vol. 1, Treasury, Canberra.

Van den Bosch, K. (2001) Identifying the Poor: Using Subjective and Consensual Measures, Ashgate, Aldershot.

Whelan, C. T., Nolan, B. and Maitre, B. (2008) Measuring material deprivation in the enlarged EU, Working Paper No. 249, Economic and Social Research Institute, Dublin.

Whiteford, P. (1985) Issues in assistance for families: Horizontal and vertical equity consideration, Social Security Review Research Paper No. 29, Department of Social Security, Canberra.

Wilkins, R. (2008) 'The changing socio-demographic composition of poverty in Australia: 1982 to 2004, Australian Journal of Social Issues, 42(4), pp. 481-501.

Wilkins, R., Warren, D., Hahn, M. and Houng, B. (2011) Families, Incomes and Jobs, Volume 6: A Statistical Report on Waves 1 to 8 of the Household, Income and Labour Dynamics in Australia Survey, Melbourne Institute of Applied Economic and Social Research, University of Melbourne, Melbourne.

Wilson, S., Meagher, G., Gibson, R., Denemark, D. and Western, M. (2005) Australian Social Attitudes: The First Report, University of New South Wales Press, Sydney.

Notes

(1.) This article is based on presentations to the Australian Social Policy Conference and the Colloquium of Superannuation Researchers in July 2011. The authors thank participants in both events (and two anonymous referees) for their comments and acknowledge the financial support provided by the Australian Research Council under project grants DP0452564, LP0560797 and LP100100562.

(2.) This broader perspective was being taken by the Henry Tax Review, and its work on the retirement income system was brought forward so that it could inform the work of the Pension Review.

(3.) One example of the latter is the increased incentive facing people receiving unemployment assistance (Newstart Allowance) to seek to become eligible for the higher Disability Support Pension.

(4.) Similar definitions can also be found in the reviews of the taxation and social security reviews undertaken in the 1970s by the Asprey Committee and the Commission of Inquiry into Poverty, respectively.

(5.) The equivalence scale embodied in the international poverty line assumes that the ratio of the needs of a single person to a couple is equal to 0.67 for all family types, while the Henderson poverty line ratio varies according to workforce status and housing tenure, falling between 0.61 and 0.74 (see Melbourne Institute 2010, Table 1).

(6.) A similar exercise has since been conducted by analysts at Fair Work Australia, who have reviewed the alternative approaches used to assess the living standards and needs of low-paid workers (Pech 2011). The same range of approaches were assessed and compared, with emphasis given to the need to explore new approaches based on recent Australian research on deprivation and social exclusion. There was also acknowledgement of the role that qualitative research can play in providing a better understanding of adequacy and living standards issues. Of the methods reviewed, the report concluded (ibid.: iii) that 'relative poverty lines would appear to be the most adaptable for use in research', but went on to note that 'they should always be used with caution as benchmarks of income adequacy'.

(7.) The deprivation approach is now widely applied in Europe (using data generated by the European Union Survey of Income and Living Conditions, EUSILC; see Whelan et al. 2008) and the estimates are used to monitor country progress towards reaching the goals identified in the European Union social policy agenda (see Guio 2005).

(8.) Strictly speaking, the survey question used to identify whether items lacking reflects a lack of affordability rather than a lack of income, although those most likely to experience deprivation will have few if any economic resources other than income so it is not unreasonable to draw inferences about income adequacy from the responses to deprivation surveys.

(9.) Because voting is compulsory in Australia, the electoral rolls provide a good sampling frame for the adult (aged 18 and over) population.

(10.) The 2003 Australian Survey of Social Attitudes (AuSSA) achieved a response rate of 44 per cent; see Wilson et al. 2005: 7.

(11.) One area where the difference between the sample and the adult population was most pronounced is in relation to age structure. Older people (aged 50 and over) are overrepresented relative to younger people (particularly those aged under 30) among the respondents. Because of this, population-based weights have been applied to the raw data when identifying whether or not items attracted majority support for being essential. Further details of how the age structure of the two samples compares with that of the general population is provided in Saunders et al. (2007, Figure 2) and Saunders and Wong (2011b, Figure 1).

(12.) The PEMA survey was accompanied by a follow-up survey of 1000 of those who responded to the 2006 survey, which attracted a response rate of 60.1 per cent.

(13.) The separate bedroom only just exceeded the 50 per cent support benchmark in 2006.

(14.) Some of the groups included in the analysis (e.g., sole parents) will have children and the exclusion of those essential items that relate to the needs of children will understate the extent of deprivation experienced by these groups. This will have an important bearing on some of the group comparisons presented later, and this should be kept in mind when reviewing those results.

(15.) It should be noted that a car is not included in the list of essential items shown in Table 1 (because it received only about 48 per cent support for being essential in 2006 and around 45 per cent in 2010). Despite this, comprehensive motor vehicle insurance is included in the list, which implies that a majority think that those who do own a car should have adequate insurance coverage.

(16.) The decline in the numbers receiving Newstart Allowance and Parenting Payment (Single) appears to reflect the growth in employment that occurred between 2006 and 2010.

(17.) The categories shown in Figure 2 are not mutually exclusive and will overlap in some instances.

(18.) The renter group in Figure 2 includes both public and private renters because of the small numbers in each separate group.

(19.) Because the categories shown in Figure 2 are not exclusive, some of those who live alone will also be renters and vice versa. This implies that the circumstances of single age pensioners living in rented accommodation are particularly dire.

(20.) The pensioner sample sizes used to derive the estimates in Figure 5 differ from those used in Figure 4 because not all of the age pensioner respondents included in Figure 4 provided the information needed to include them in Figure 5.

About the Authors

>> Peter Saunders is a Scientia Professor and Research Professor at the Social Policy Research Centre, University of New South Wales. He has published extensively in the areas of poverty and income distribution, household needs and living standards, social security reform, comparative social policy, and ageing and social protection in China. He can be contacted at p.saunders@unsw.edu.au.

>> Melissa Wong is a Research Associate at the Social Policy Research Centre, University of New South Wales. Her research interests include the measurement of poverty, deprivation and social exclusion, child well-being indicators and development economics. She can be contacted at melissa.wong@unsw.edu.au.

Peter Saunders * Melissa Wong *

* Social Policy Research Centre, The University of New South Wales

Table 1: Support for items being essential in 2006 and 2010
(weighted percentages)

Item 2006 2010

Warm clothes and bedding, if it's cold 99.8 99.9
Medical treatment if needed 99.9 99.9
Able to buy medicines prescribed by a doctor 99.3 99.5
A substantial meal at least once a day 99.6 99.4
Dental treatment if needed 98.5 98.4
A decent and secure home 97.3 97.1
A roof and gutters that do not leak 91.5 91.3
Secure locks on doors and windows 91.6 92.4
Regular social contact with other people 92.5 91.6
Furniture in reasonable condition 89.3 89.0
Heating in at least one room of the house 87.4 87.0
Up to $500 in savings for an emergency 81.1 81.4
A washing machine 79.4 77.7
Home contents insurance 75.1 72.4
Presents for family or friends at least once a year 71.6 71.4
Computer skills 68.7 72.6
Comprehensive motor vehicle insurance 60.2 59.9
A telephone 81.1 59.7
A week's holiday away from home each year 52.9 53.9

Source: Saunders and Wong (2011b, Table 1).

Table 2: Income group definitions and sample sizes

Sample size

Income group Definition 2006 2010

Low-wage worker Main source of income (MSI) is 188 82
 wages/salaries, aged 17-64,
 and at least one full-time
 earner in the household

Self-funded Respondent is above pension 98 116
retiree age and MSI is interest, rent,
 or dividends, etc.

Age pensioner MSI is Age Pension 306 264

Service pensioner MSI is Department of Veterans' 45 33
 Affairs Pension

Disability MSI is Disability Support 69 55
pensioner Pension

Parenting payment MSI is Parenting Payment and 38 19
recipient a sole parent

Newstart allowee MSI is Newstart Allowance 34 20

Source: CUPSE and PEMA surveys (see text).