The social impact of the global financial crisis in Australia.
The Australian economy has weathered the storm that followed the
global financial crisis (GFC) better than most other OECD countries. The
reasons for this are complex, although the fiscal stimulus measures
introduced by the federal government in 2008 and 2009 boosted domestic
consumption and investment and helped to sustain economic growth.
However, even with these measures, concerns have been raised over the
social impact of the financial crisis, with a number of studies
suggesting that those with lowest incomes and/or reliant on welfare
services for support were most adversely affected. This paper presents
new estimates of the social impact of the GFC using data from two
national surveys, conducted in 2006 and 2010--before and after the
crisis hit Australia. The impact is assessed using a range of different
approaches, including people's own perceptions of the impact,
changes in their subjective wellbeing, reported changes in financial
stress and changes in deprivation and economic exclusion. The results
suggest that the social impact of the crisis has been small, although
some evidence suggests that those already facing the most severe levels
of social disadvantage were most adversely affected. In this sense, the
GFC may have led to greater inequality in living standards, at least in
Key words: deprivation, economic exclusion, global financial crisis, perceived impact, subjective wellbeing
Economic conditions (Social aspects)
Financial management (Social aspects)
|Publication:||Name: Australian Journal of Social Issues Publisher: Australian Council of Social Service Audience: Academic Format: Magazine/Journal Subject: Sociology and social work Copyright: COPYRIGHT 2011 Australian Council of Social Service ISSN: 0157-6321|
|Issue:||Date: Spring, 2011 Source Volume: 46 Source Issue: 3|
|Topic:||Event Code: 290 Public affairs|
|Product:||Product Code: 9915100 Financial Management; 9915500 Financial Systems & Controls|
The economic impact of the global financial crisis (GFC) that emerged during 2008 has been lower in Australia than in most other OECD countries. Between 2008 and 2010, the average unemployment rate in OECD countries increased from 6.1 per cent to 8.6 per cent, while it increased only marginally in Australia, from 4.2 per cent to 5.2 per cent after peaking at just over 6 per cent in early 2009 (OECD 2011: Annex Table 14; Saunders & Deeming 2011). As a consequence, the post-crisis unemployment rate in Australia remained below the pre-crisis level experienced elsewhere.
Many factors contributed to the resilience of the Australian economy over this period. One was the decisive fiscal stimulus measures introduced by the federal government in 2008 and 2009 that have attracted wide praise from international agencies like the OECD and the IMF, as well as from expert commentators. (1) Another was the rapid return to growth in the Chinese economy which stimulated the demand for Australian mineral exports, while the strong financial regulatory framework introduced in Australia in the early 1990s and reforms to the labour market that increased flexibility have cushioned it from external shocks like the GFC. A large budget surplus and relative low public debt also provided room for the government to expand its fiscal stance without compromising its longer-term fiscal sustainability.
However, while debate over the relative importance of these and other factors continues, less attention has been paid to the social impact of the GFC. Economic statistics like those cited above have important flow-on effects on social outcomes, where the critical issue is who bears the burden and for what duration. A sharp rise in unemployment, even if reversed fairly quickly, can have lasting effects on those who lose their jobs, particularly if they have to wait some time before regaining employment or have to settle for lower wages or poorer quality jobs. Both long-term unemployment (those unemployed for 12 months or more) and youth unemployment (among those aged between 15 and 19) remained high following the GFC, highlighting the uneven incidence of increased unemployment and the potentially broad social effects that it caused (ABS 2011a; ABS 2012; CBA/NATSEM 2010).
This paper estimates the social impact of the GFC in several dimensions using a range of indicators derived from data generated by two large-scale national surveys conducted in 2006 and 2010. The goal is to establish whether or not the GFC exerted an adverse social impact in Australia and, if so, to identify those impacts. A subsidiary question is whether the GFC contributed to growing inequality between those who were most disadvantaged before it emerged and the rest of the population.
It is important to acknowledge at the outset that the GFC has not been the only factor giving rise to concern about potentially undesirable social effects of recent economic developments. Much attention has focused on the cost of living pressures associated with rising fuel and energy bills, reinforced by the rising food prices that have been a consequence of natural disasters in food-producing areas of the country. Although these have occurred against a background of rising national income, and hence increasing living standards, it has been argued that the increased cost of living has been a particular burden on those on lowest incomes. Because these pressures have emerged since the GFC struck, it is difficult to untangle the effects of each, and this qualification applies to the results reported below.
Recent studies of the impact of the GFC
The emergence of the financial crisis led to considerable consternation among government and non-government agencies responsible for addressing the anticipated increase in social distress. A number of studies produced by non-government agencies examined its impact so that remedial measures could be planned and implemented. In 2008, Access Economics was commissioned by a consortium of leading community sector NGOs to prepare a report on the impact of the GFC on social services in Australia. The report, released in November 2008 when concern about the impact of the crisis was at its peak, began with the bold claim that:
The current global financial crisis and its likely impact on the Australian economy will have an acute impact on the most disadvantaged members of society, as well as pushing increased numbers of low and middle income earners to seek the services of welfare agencies... Economic growth will inevitably slow, the extent to which is uncertain [but the] impact will vary across different segments of society, with the unemployed and other vulnerable groups particularly hard hit (Access Economics 2008: 2; emphasis added).
The Access Economics report was discussed at an NGO summit that generated a series of recommendations for government action which began:
We are deeply concerned that the deepening impact of the global financial crisis will impact most harshly on those Australians that are least able to weather the gathering storm on their own. Many Australians who did not enjoy the prosperity of the recent boom will bear the greatest impact of the bust (Anglicare Australia et al. 2009: 3; emphasis added).
The recommendations themselves called for prompt action to avert the impact of the crisis through stimulus measures that should 'target low income citizens, low income households and chronically disadvantaged locations' (Anglicare Australia et al. 2009: 6).
In the same year, the Department of Families, Housing, Community Services and Indigenous Affairs (FAHCSIA) commissioned a survey to examine the impact of the crisis 'to help understand the impact of the worldwide economic downturn on Australian families' (FAHCSIA 2009: 1). The main fieldwork was conducted in May and June 2009 and involved telephone interviews with 1,650 families with at least one child under 18. The study found that almost 34 per cent of those surveyed considered their financial situation to have worsened over the previous six months (since around the beginning of 2009), while only 11 per cent considered it to be better. (2) In general, there was a clear tendency for those with lower incomes to be more likely to report a worsening of their financial situation: for example, 57 per cent of households with income below $20,000 a year reported a worsening compared with 14 per cent of households with incomes over $150,000.
There was also a clear gradient linking perceived changes in financial situation over the last six months and the respondents' assessment of their current financial situation. Thus, while around 14 per cent of those who rated their situation as 'prosperous' when surveyed thought that it had worsened over the last six months, this was the case for over 70 per cent of those who rated their circumstances as 'poor' and 100 per cent for those who rated themselves as 'very poor' (FAHCSIA 2009: 8). Even though the numbers in these latter two groups are small (around 100 households in total) there was a clear relationship between a perceived negative impact and current circumstances which suggests that those least able to cope had been hit hardest.
A similar pattern emerged from a study conducted in May 2010 by The Salvation Army (2010), which surveyed almost 700 clients who had requested emergency assistance and compared the results with those produced by an online survey of 'the general public' that generated 375 respondents. Over half (55 per cent) of the client survey believed they were worse off as a result of the financial crisis, with a similar proportion reporting that they had 'cut down on basic necessities' or 'felt depressed about their situation'. The first of these figures was well above the corresponding figure of 37 per cent of respondents to the public survey who reported being worse off as a result of the financial crisis. The findings lead the report's authors to conclude that: 'people most disadvantaged in the economy, the people The Salvation Army has regular contact with, are facing the worst of the current economic situation and the negative psychological impact is greater on them' (The Salvation Army 2010: 3).
Another study, conducted by the Wesley Mission (2010), drew on an online and telephone survey with over 620 adults in New South Wales. It found that more than one-third of those surveyed reported being financially stressed, as indicated by having to struggle to pay utility bills, going without meals or being forced to pawn items. It noted that 'where six in 10 would have been able to cover a $2,000 emergency expense in 2006, only four in 10 can do so now' (Wesley Mission 2010: 6). These and other findings lead the CEO of Wesley Mission to claim in his Foreword to the report that:
While Australia has escaped the full impact of the Global Financial Crisis (GFC) at the macro level, many households have nevertheless been affected and, according to this report, are likely to keep feeling the pain for some time, particularly those surviving on low incomes (Wesley Mission 2010: 5; emphasis added).
Although these latter two reports would have captured the effects of cost of living pressures as well as the impact of the GFC, the overwhelming impression they present is that the GFC itself resulted in a considerable adverse impact on those least able to absorb it. However, all of the surveys reviewed above were small in scale and were targeted on recipients of cash support or service assistance. This is an understandable focus for agencies with limited resources and whose primary goal is to protect and promote the interests of their clients, but it does not allow conclusions to be drawn about the wider social impact of the GFC. Only when a broader approach is taken is it possible to establish whether or not those at the bottom have experienced not only adverse, but the most adverse effects.
A somewhat different picture emerges when a broader perspective on the impact of the GFC is adopted. A Melbourne Institute study based on waves 7 and 8 (covering 2007 and 2008) of the Household, Income and Labour Dynamics in Australia (HILDA) survey found that mean job security satisfaction dropped sharply between 2007 and 2008, but remained above its level in 2006 and earlier years. Disaggregated results also show that mean job satisfaction increased between 2007 and 2008 for fixed-term and casual employees but declined for those in permanent positions (full-time and part-time) (Melbourne Institute 2010: Figure 12 and Table 12). The latter finding suggests that those in the most advantaged labour market positions may have fared worse over this period.
Analysis by the current authors of financial stress data in waves 7 and 9 of the HILDA survey reveals a slight decline in the incidence of 2 of the financial stress or hardship indicators between 2007 and 2009 but an increase in the incidence of 6 indicators. (3) However, the increases were all small in magnitude (less than one percentage point in all but one case), although it is interesting to note that the two indicators with the largest increase were asking for help from a welfare/ community organisation (up 0.7 percentage points) and could not raise $2,000 in an emergency (up 1.2 percentage points).
Overall, the evidence from these national studies suggests that it is important to look more broadly at how the GFC impacted on the general population. This enables us to compare how different socioeconomic groups were affected and to establish whether those experiencing social disadvantage did worse than others, leading to an increase in between-group inequality.
Data and methodology
The Community Understanding of Poverty and Social Exclusion (CUPSE) survey was the second in a sequence of surveys designed to develop better indicators of social disadvantage in Australia. The survey was distributed by mail to 6,000 adult Australians randomly selected from the electoral rolls in April 2006. (4) By the end of July, it had generated 2,704 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. (5) The detailed comparisons reported by Saunders, Naidoo and Griffiths (2007: Table A.3) indicate that the CUPSE sample is broadly representative of the general population, although the following groups are under-represented: 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 gross incomes over $1,000 a week. Some of these differences are inter-related, while others may reflect the difficulty involved in conducting a mail survey. (6)
The Poverty and Exclusion in Modern Australia (PEMA) survey was distributed to a new sample of 6,000 adults in May 2010 and generated 2,645 responses over the next three months--equivalent to a response rate of 46.1 per cent. It was accompanied by a follow-up survey of 1,000 of those who responded to the CUPSE survey, which attracted 533 responses (the PEMA follow-up sample), equivalent to a response rate of 60.1 per cent. Both surveys replicated the CUPSE questions, aside from the removal of those relating to attitudes to poverty and inequality and the addition of questions relating to the impact of the GFC and aspects of community participation and location. A small number of minor modifications were made to the questions used to identify deprivation. For example, both surveys include in the list of items a telephone and a mobile phone. However, the increased usage of mobile phones resulted in the first item being identified in 2010 as 'a telephone (landline)' to avoid confusion.
Any bias in the composition of each of the two samples can have an important impact on key aspects of the survey results (for example, when identifying whether an item attracts majority support for being essential) and differences in the composition of the two samples can distort comparisons over time. Because the largest bias was related to age in both surveys (see Saunders et al. 2007: Figure 2; Saunders & Wong 2011a: Figure 1), population-based weights have been applied to the raw data so that the estimates better reflect community-wide effects, rather properties of the two samples.
The weighted results adjust for any differences in the age composition of the two samples, but it is possible that other differences still exist that could have a bearing on the comparisons between them. A detailed comparison of the socioeconomic structure of the two samples reported in Saunders and Wong (2012: Table 3.1) indicates that the main areas of difference between them relate to educational status (more respondents with undergraduate degrees and fewer with only high school completions in 2010), family status (fewer couples with children and sole parent families in 2010) and (gross) income (far more with incomes over $1,000 a week in 2010). These differences suggest that the PEMA sample may be somewhat more economically advantaged overall than the CUPSE sample, and this could mitigate the estimated negative impacts of the GFC and needs to be borne in mind.
The over-representation of older people in the CUPSE (and PEMA) samples was magnified in the PEMA follow-up sample because the sampling frame itself contained an over-representation of older people. Although this can also be adjusted for by applying weights to the raw data, this is a more complex exercise when the data are longitudinal and so only unweighted estimates are presented for those comparisons that include the PEMA follow-up sample.
All three surveys included questions relating to a series of 'basic need' items (61 in the case of CUPSE, 73 in the case of PEMA), some of which had been identified as necessary to achieve a decent standard of living by participants in a series of focus groups with low-income clients of selected welfare services (see Saunders and Sutherland 2006). Other items were drawn from overseas deprivation studies conducted in New Zealand, Ireland and Britain and included some of the indicators used to identify hardship or financial stress in HILDA and other Australian surveys (Bray 2001; McColl et al. 2001; Breunig & Cobb-Clark 2006; Hahn & Wilkins 2008).
Survey respondents were asked three questions about each item: Is it essential? Do you have it? And, if not (and where the item is purchasable by individuals), Is this because you cannot afford it? The word 'essential' was defined in the surveys as referring to 'things that no-one in Australia should have to go without today': the aim is thus to identify items that are essential for people in general, rather than for individual respondents.
Following international practice (see Gordon 2006; Pantazis et al. 2006), 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 & Naidoo 2009). All but one of the remaining 25 items also received majority support for being essential in 2010. The exception was a separate bedroom for older children, but support for this item varied greatly with age and the overall level of support only just exceeded the majority support threshold in 2006. This item was therefore also dropped, bringing the number of essential items down to 24.
Deprivation was defined to exist when people did not have and could not afford those items that received majority support for being essential for all Australians. (7) Economic exclusion (the only dimension of exclusion considered in this analysis) was identified using the indicators developed by Saunders, Naidoo and Griffiths (2008) and details are provided later.
Before presenting results, it is important to note that the timing of the CUPSE and PEMA surveys was not ideal for examining the social impact of the GFC by comparing "before" and "after" outcomes. This is illustrated in Figure 1, which highlights the relatively short impact of the crisis in Australia and the speed with which it was reversed. Clearly, the observed differences between the two surveys will reflect the impact of all the changes that took place between mid-2006 and mid-2010, of which the GFC was only one. Other factors include the policies introduced to combat the impending crisis (including the one-off stimulus payments, which were targeted at many low-income groups), and (potentially) the differences in sample composition mentioned earlier. It is also important to note that real incomes (including average weekly earnings, household disposable income per capita and the real value of most social security payments) increased considerably between 2006 and 2010 (Saunders & Wong 2011b: Figure 3) and this could have offset the short-run negative impact of the GFC.
[FIGURE 1 OMITTED]
Despite the existence of these confounding factors, it is rare for social scientists to have access to data that provides the basis for making direct 'before' and 'after' comparisons using data designed specifically for the purpose (including longitudinal data). Importantly, the CUPSE and PEMA surveys allow the impact of the GFC to be examined using data that, while not ideal, were specifically designed for the purpose.
In light of the above discussion, and in order to avoid the obvious interpretational problems involved in relying on a single measure of impact, the issue has been explored using three approaches. The first involves examining what respondents themselves said the impact had been on them and their families. The second draws inferences from comparing how a range of indicators of subjective wellbeing (SWB) changed between 2006 and 2010. Finally, changes in the reported incidence of financial stress and in estimated rates of deprivation and economic exclusion are compared. (To the extent that the timing problem discussed earlier applies, it will only be relevant to results derived from the second and third approaches.) If all three approaches reveal a similar story, then one can be more confident that the findings are robust, whereas if different results emerge from the different approaches, a more cautionary conclusion will be appropriate.
The PEMA survey included a question in which a number of possible effects of the GFC were identified and respondents were asked to indicate which of them had been relevant in their case (so that multiple responses were allowed). The effects included and the responses produced are summarised in Table 1. The aggregate estimates in the first column of Table 1 indicate that almost one-quarter of the population experienced some form of employment impact following the GFC and close to one-half had their incomes affected in some way. The survey question does not specifically ask respondents whether their changed circumstances were a direct consequence of the GFC, although the wording of the question (see Table 1) clearly implies this. Against this, 40 per cent reported no impact on themselves or their family--a finding that confirms that the impact was much less than was originally predicted.
The effects are, however, different for different age groups, with those under 30 bearing the biggest employment impact (substantially larger in all cases than those experienced by people aged 30-64), with many young people finding it harder to obtain work or get promoted. Income effects were generally more pervasive than employment effects, and although the impacts also vary significantly across age groups, these differences are not as large as those that relate to employment. Many older people experienced the impact of lower interest rates, whereas younger age groups were more likely to reduce their spending or pay off debts, but both responses suggest a decline in living standards. It is important to acknowledge that, even when 'normal' economic conditions prevail, some people will lose their jobs, others will experience reduced hours of work and those affected will experience lower incomes that will in turn have further flow-on effects. Some of these effects will vary with age and hence may in part explain the patterns observed in Table 1. It is thus not possible to attribute the survey responses unambiguously to the GFC (even though the survey question refers specifically to it) because respondents will not know with certainty if the change they experienced was a direct consequence of the crisis.
The income impacts also vary with income--not surprisingly, since both income and employment participation vary systematically with age--although the differences are not as large as those for age and are less frequently statistically significant. In terms of absolute size, one of the employment effects and all three of the income effects are bigger for those in the middle income category than for those in the lowest income category. This finding is at odds with the findings produced by the studies reviewed earlier that the biggest effects were felt by those on lowest incomes, and suggests that a somewhat different assessment of the distribution of the impacts is produced when a broader perspective is taken. Note, however, that no adjustment has been made to income to reflect differences in household size and composition (for example, by using an equivalence scale) and this may also partly explain the difference between the results presented here and those reported in the studies reviewed earlier.
The second approach used to examine the impact of the GFC compares the reported levels of subjective wellbeing (SWB) before (in 2006) and after (in 2010) the financial crisis struck. Six indicators were used for this purpose: assessed standard of living (ASL); satisfaction with standard of living (SSL); happiness (HAP); ability to manage on current income (IM); choice and control over 'one's life and the things that happen to you (C&C); and satisfaction with one's financial situation (SFS).
The wording of the relevant survey questions from which the SWB indicators were derived follows standard practice, with respondents asked to choose between a small number of options or (in the case of SFS and C&C) to rate their situation on a 10-point scale from very dissatisfied/no control to very satisfied/a great deal of control. For ease of presentation, each response has been assigned a score that varies between one (for the lowest response category) to the highest response category provided in each case and the mean value of each indicator in each year have been derived. Results are presented for the cross-sectional comparisons for the two surveys, and for the linked panel of respondents to both surveys.
The results in Table 2 provide little evidence that there was a decline in SWB as a result of the GFC. Although the absolute values of both indicators reveal important information about the high level of wellbeing among the population in both years, interest here focuses on how things changed over the period. Most of the indicators remained fairly constant, but those that did move (the last two indicators, both of which are defined using a ten-point scale) increased significantly in the cross-section comparisons. (8)
There is, of course, a large literature on the determinants of subjective wellbeing, much of it showing that SWB indicators tend to be stable even over long periods. Drawing on his earlier research, Cummins (2003: 225) reports 'remarkable stability in the means scores of population estimates of life satisfaction', and goes on to argue that this is a result of strong internal homeostatic tendencies that restrict life satisfaction to a narrow range of values, and explain the weak relationship between life satisfaction and objective measures of the external conditions of life (Cummins 2003: 253).
This line of reasoning suggests that the finding that there was little change between 2006 and 2010 in the SWB indicators examined here provides rather weak evidence that the GFC had no impact, since stability in the indicators examined is to be expected. However, the logic of this view depends on how quickly the homeostatic mechanisms operate to maintain life satisfaction in the face of major external shocks (like the GFC). The idea of homeostasis is consistent with short-run volatility in life satisfaction in the period before the homeostatic mechanisms exert their influence. If this takes more than four years, then one should observe some degree of short-run volatility in life satisfaction (and other SWB indicators) and the tests reported above should be able to detect it--as some of them do.
The third approach begins by examining changes in the reported incidence of a range of direct and indirect indicators of financial stress or hardship between 2006 and 2010. These comparisons are again based on the two cross-section samples (weighted by age, as before) and on the (unweighted) linked panel sample. The indicators of financial stress are defined in Table 3, which also presents the incidence rates of each indicator in each year. (9)
The results in Table 3 provide no strong evidence that financial stress was any worse in 2010 than in 2006. The individual estimates move in both directions and although many of them declined over the period across both sets of comparisons, the only differences that are statistically significant are the cross-sectional decline in the proportion unable to raise $2,000 in an emergency, and in the proportion of the panel sample that went without food when they were hungry. Doubts have been raised, for example, by Bray (2001), about the ability of the kinds of indicators shown in Table 3 to capture who is most disadvantaged, and a degree of caution thus applies to the interpretation of these results. In contrast, the indicators now considered have been developed specifically to capture the extremes of disadvantage and should thus provide a more accurate picture of the changes that have taken place.
As explained earlier, the analysis of changes in deprivation are based on the 24 items that received at least 50 per cent (age-weighted) support for being essential in both years. In addition to comparing the overall incidence of deprivation across these items in each sample, changes in deprivation for those identified as facing severe deprivation in each year are also examined. For this purpose, severe deprivation was defined to exist when the household was deprived of 4 or more of the essential items. Two summary measures of overall deprivation are used, the mean deprivation incidence rate across all 24 items, and the mean deprivation score, equal to the average number of deprivations experienced across the whole sample. The (unweighted) mean deprivation score is commonly used in deprivation studies and there is statistical support for its use as a summary measure of multiple hardship or deprivation (see Butterworth & Crosier 2005; Cappellari & Jenkins 2007). The results are shown in Table 4.
The first point to note about these estimates is that both the mean incidence of deprivation and the mean deprivation score declined for the full sample between the two years. So too did the size of the severely deprived sub-sample relative to the total sample--from 13.6 per cent in 2006 to 11.7 per cent in 2010. However, within the severely deprived group there was no change in the mean incidence of deprivation, while the mean deprivation score increased slightly, but not significantly. Thus, while overall deprivation fell somewhat, its severity did not change or rose slightly, suggesting that those who were most deprived in 2006 suffered most over the period.
The overall deprivation incidence rate declined for 19 of the 24 items and increased in only five instances--generally by a very small margin. Some of the declines are substantial, with five items experiencing declines of more than 1.5 percentage points: dental treatment if needed; an annual dental check-up for children; up to $500 in emergency saving; computer skills; and a week's holiday away each year. (10) However, few of the individual item (and none of the aggregate) year-on-year changes shown in Table 4 are statistically significant. The weight of the evidence thus suggests that the hypothesis that deprivation became worse over the period can be rejected.
Table 4 also indicates that deprivation declined less among the most severely deprived sub-group than among the community as a whole. For this group, deprivation fell for only 13 of the 24 items and increased for the other 11--in several cases by more than three percentage points. Both indicators show that dental deprivation also declined among this group, but remained high enough to suggest that there is still a need for further policy improvement in this area. With those already facing the most severe deprivation faring relatively worse than the community as a whole, the results in Table 4 suggest--albeit tentatively--that there was an increase in inequality in deprivation (and hence in living standards) over the period. This aspect of the findings is also consistent with the picture of the greatest burden falling on those least able to afford it that was given prominence in some of the community sector reports described earlier.
The final piece of evidence concerns changes in economic exclusion--one of the three broad domains of exclusion identified in earlier research on social disadvantage (Saunders et al. 2008; Saunders 2011). Eight indicators have been used to capture economic exclusion and they are identified in Table 5, which shows changes in the mean incidence of each indicator and in the overall incidence of economic exclusion among the full sample and among those identified as in deep exclusion (defined as experiencing at least 8 of 27 indicators across all three dimensions of social exclusion). The mean exclusion score is defined in the same way as the mean deprivation score. This summary measure has been used in work on social exclusion conducted by the Melbourne Institute (see Scutella et al. 2009). The definition of deep exclusion has been based on all three domains of exclusion (rather than the multiple incidence of just economic exclusion) for consistency with earlier work (see Saunders et al. 2008) and with the analysis of severe deprivation presented earlier.
The evidence for the full sample suggests that economic exclusion declined overall and in most dimensions, with the overall decline and several of the individual falls (particularly those relating to labour market exclusion) being statistically significant. A similar pattern is evident for changes in deep exclusion, although fewer of the individual changes are statistically significant (although those that are all indicate a decline). Although both the average incidence of economic exclusion and the mean exclusion score declined overall and for those in deep exclusion, the decline was proportionately less (and not statistically significant) for the latter group, which suggests a rise in exclusion inequality over the period.
The exclusion indicators based on unemployment and joblessness declined over the period even though overall unemployment increased slightly (Figure 1) and these declines contributed substantially to the overall decline in economic exclusion. Perhaps of greater significance is the sharp drop in the proportions who did not have $500 in emergency savings, could not raise $2,000 in a week if they needed to and had not spent money on a special treat, which is consistent with the fact, noted earlier, that the average real incomes of most groups increased over the period. In contrast, the (small) rise in the percentage experiencing difficulty getting by on their income may reflect the rising costs of living that were taking place over the period.
Summary and conclusions
This paper began by noting that the Australian economy has weathered the storm caused by the global financial crisis better than most other OECD countries. Despite this, concerns have been raised about the extent and distribution of the social burden that accompanied the economic downturn. A series of reports released by those working at the coalface of poverty relief have suggested that declines in actual incomes and perceived wellbeing were experienced by many of those who were least able to afford it. The results presented here have examined the validity of these propositions using data from two national surveys that were conducted around 18 months before the crisis emerged in Australia and about 18 months after its effects had begun to dissipate.
The approach explored whether the overall picture is one of decline or improvement across a range of indicators that capture the effects of the GFC, including people's own perceptions of its impact, changes in a series of indicators of subjective wellbeing and reported financial stress, and changes in two objective indicators of social disadvantage, relating to deprivation and economic exclusion. A substantial proportion of those surveyed in 2010 reported that their employment circumstances and incomes had been adversely affected by the GFC, with younger people reporting the worst employment outcomes and middle-aged and older people the worse income effects. In contrast, although the evidence suggests that some dimensions of subjective wellbeing improved slightly between 2006 and 2010, many of these indicators remained static over the period.
The picture revealed by movements in the more objective indicators is one of greater stability, with very few of the indicators of financial stress and none of the 24 indicators of deprivation showing a decline (or an increase) that is statistically significant. The evidence that economic exclusion declined over the period is stronger, particularly in relation to labour market exclusion and some of the indicators that track changes in material living standards.
There is also evidence that those who were initially most disadvantaged in terms of their exposure to deprivation and economic exclusion tended to fare somewhat worse than others in the community. In this sense, the results are consistent with the reports of increased demand for welfare assistance produced by community sector studies conducted immediately after the onset of the GFC, and suggest that inequalities in living standards (like income inequality--see ABS 2011b: Table 1) increased between 2006 and 2010.
However, despite many qualifications, the over-riding impression conveyed by the results is one of a general (if modest) improvement in living standards, broadly conceived--a finding that is consistent with the increased real incomes that many Australians experienced over the period. This suggests that the resilience demonstrated by the Australian economy in response to the GFC, reinforced by the direct (income-enhancing) and indirect (macroeconomic) impacts of the fiscal policies introduced to counter its effects, have meant that the adverse social impacts that many had feared when the crisis first broke in late 2008 have been avoided. The fact that most people were better off in 2010 than in 2006 suggests that any negative social effects of the GFC were modest and temporary.
Preliminary versions of this paper were presented to a seminar at the Social Policy Research Centre in May 2011 and to the 2011 Australian Social Policy Conference in July. The authors would like to thank the participants in both events who provided helpful suggestions for improvement and acknowledge the incisive comments provided by three anonymous referees. Financial support was provided by Australian Research Council grants DP0452562 and LP100100562.
Access Economics (2008) The impact of the Global Financial Crisis on social services in Australia, Canberra, Access Economics.
Anglicare Australia, Catholic Social Services, Australia, UnitingCare Australia and the Salvation Army (2009) Averting the social impact of the Global Financial Crisis, Canberra, Anglicare Australia.
ABS (Australian Bureau of Statistics) (2011a) Australian Social Trends. September 2011. Long-term Unemployment Cat. No. 4102.0.
--(2011b) Household Income and Income Distribution 2009-10, Cat. No. 6523.0.
--(2012) Australian Economic Indicators February 2012. Long Time Series Spreadsheets Cat. No. 13500DO016_210202.
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, Canberra, Department of Family and Community Services.
Breunig R. & Cobb-Clark, D. (2006) 'Understanding the factors associated with financial stress in Australian households', Australian Social Policy 2005, 13-64.
Butterworth, P. & Crosier, T. (2005) 'Deriving a measure of financial hardship from the HILDA survey', Australian Social Policy, 4, 1-12.
Cappellari, L. & Jenkins, S.P. (2007) 'Summarising multiple deprivation indicators'. In S.P. Jenkins & J. Micklewright (eds), Inequality and Poverty Re-examined, Oxford, Oxford University Press.
CBA / NATSEM (Commonwealth Bank of Australia & National Centre for Social and Economic Modelling) (2010) The economic vitality report: the impact of the Global Financial Crisis on Australia, Issue One, CBA/ NATSEM.
FAHCSIA (Department of Families, Housing, Community Services and Indigenous Affairs) (2009) Preliminary Report: Survey on the Impact on Families of the Economic Downturn, Canberra, FAHCSIA.
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.
Hahn, M. & Wilkins, R. (2008) A multidimensional approach to investigation of living standards of the low-paid: income, wealth, financial stress and consumption, Research Report No. 5/09, Melbourne: Australian Fair Pay Commission.
McAllister, A. & Clark, J. (2007) Trends in Australian political opinion: results from the Australian Election Study, 1987-2004. Canberra, Australian Social Science Data Archive.
McColl, R., Pietsch, L. & Gatenby, J. (2001) 'Household income, living standards and financial stress', Australian Economic Indicators. June 2001, Cat. No. 1350.0, 13-32.
Melbourne Institute (2010) Household, Income and Labour Dynamics in Australia (HILDA) Survey. Annual Report 2009, Faculty of Business and Economics, University of Melbourne.
OECD (2011) Economic Outlook 2010, Paris, Organisation for Economic Co-operation and Development.
Pantazis, C., Gordon, D. & Townsend, P. (2006) 'The necessities of life'. In C. Pantazis, D. Gordon & R. Levitas (eds), Poverty and Social Exclusion in Britain. The Millennium Survey, Bristol, Policy Press.
Saunders, P. (2011) Down and Out: Poverty and Exclusion in Australia, Bristol, Policy Press.
Saunders, P. & Deeming, C. (2011) 'The impact of the crisis on Australian social security policy in historical perspective', Social Policy & Administration, 45 (4), 371-88.
Saunders, P. & Naidoo, Y. (2009) 'Poverty, deprivation and consistent poverty', Economic Record, 85 (271), 417-32.
Saunders, P., Naidoo, Y. & Griffiths, M. (2007) Towards new indicators of disadvantage: deprivation and social exclusion in australia, Social Policy Research Centre, University of New South Wales.
--(2008) 'Towards new indicators of disadvantage: deprivation and social exclusion in Australia', Australian Journal of Social Issues, 43 (2), 175-94.
Saunders P. & Sutherland, K. (2006) Experiencing poverty: the voices of low-income australians. Towards new indicators of disadvantage project stage I: focus group outcomes, Social Policy Research Centre, University of New South Wales.
Saunders, P. & Wong, M. (2011a) 'Measurement and change in deprivation and exclusion in Australia: a report on research in progress', presented to the Second Townsend Memorial Conference, University of Bristol, 22-23 January.
--(2011b) 'Pension adequacy and the pension review', The Economic and Labour Relations Review, 22 (3), 7-26.
--(2012, forthcoming), Promoting inclusion and combating deprivation: recent changes in social disadvantage in australia, Social Policy Research Centre, University of New South Wales.
Scutella, R., Wilkins, R. & Kostenko, W. (2009) 'Estimates of poverty and social exclusion in Australia: a multidimensional approach' Working Paper no. 26/09, Melbourne Institute of Applied Economic and Social Research.
The Salvation Army (2010) Perceptions of poverty: an insight into the nature and impact of poverty in australia, Sydney, The Salvation Army.
Wesley Mission (2010), Making ends meet: financial stress is not just about money, Sydney, Wesley Mission.
Wilkins, R., Warren, D., Hahn, M. & 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.
Wilson, S., Meagher, G., Gibson, R., Denemark, D. & Western, M. (eds) (2005) Australian Social Attitudes: The First Report, Sydney, UNSW Press.
(1) Nobel Laureate Joseph Stiglitz was reported as noting during a visit to Australia in 2010 that by introducing the stimulus measures the ALP Government 'did a fantastic job of saving Australia from the global economic crisis' (Saunders & Deeming 2011).
(2) Results from the 2007 Australian Electoral Study (AES) reported by McAllister and Clark (2007: Section 4) indicate that 27 per cent of those surveyed indicated that the financial situation of their household had improved over the past year. This is well above the corresponding figure of 11 per cent among FAHCSIA respondents.
(3) Declines were experienced for the percentage not able to pay electricity, gas or telephone bills on time, and not able to pay the mortgage or rent on time. Increases took place for the percentages who: pawned or sold something; went without meals; was unable to heat home; asked for financial help from friend or family; asked for help from a welfare/community organisation; and could not raise $2,000 in an emergency (HILDA unit record file data).
(4) Because voting is compulsory in Australia, the electoral rolls provide a good sampling frame of the adult (aged 18 and over) population. Household members aged under 18 are defined as children.
(5) Response rates have been calculated after removing returned surveys indicating that the address was incorrect. The CUPSE response rate was above that of 44 per cent achieved in the Australian Survey of Social Attitudes 2003 (AuSSA)--see Wilson, Meagher, Gibson, Denemark and Western (2005: 7).
(6) 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 over-represented relative to younger people (particularly those aged under 30) among the respondents. The AuSSA sample also contains an under-representation of those aged under 35 and an over-representation of those aged over 65 (Wilson et al. 2005: Table 1.1).
(7) It is worth observing at this stage that the identification of essential items is driven by community views as reflected in the survey responses rather than imposed by the researchers, and also that the method avoids the need to set a poverty line or make any assumptions about relative needs (as captured in an equivalence scale).
(8) Because of the differences identified in Table 1 across age and income groups, the analysis reported in Table 2 was repeated on a disaggregated basis for the same sub-groups. There are some interesting between-group differences in the estimated impacts in both years, particularly those based on income. However, while all of the year-on-year differences broken down by income are statistically significant in the cross-sectional comparisons, few of those disaggregated by age are, and almost none of the disaggregated linked panel comparisons are statistically significant. These disaggregated results are available on request from the authors.
(9) Trends in the incidence of financial stress using the HILDA data for the period 2001-08 are presented in Wilkins and colleagues (2011: Table 9.1). Although the general patterns are similar, the 2006 estimates reported in Table 3 are somewhat higher than those derived from HILDA for either 2005 or 2007. This may reflect differences in the definition of the indicators and in the detailed wording of the questions from which they are derived.
(10) The first two of these can in part be attributed to the improvements in the Commonwealth Dental Health Program that were introduced in March 2008.
Table 1: Reported impacts of the financial crisis (weighted percentages) Question: The global financial crisis (GFC) began to affect Australia towards the end of 2008. Since that time, have you been affected by any of the following? Age: All < 30 30-64 65+ I lost my job 6.2 8.1- 7.0 0.5 My partner or another 10.3 15.8** 9.7 5.0 close family member lost their job I found it harder to 12.3 26.2** 10.2 0.3 obtain work or get a promotion At least one of the above 23.9 41.0** 22.3 5.5 three 'employment effects' I was forced to reduce my 7.1 12.5** 7.0 0.3 hours of work or take a pay cut My income fell because of 13.9 6.2** 10.4 37.3 lower interest rates I reduced my 31.3 32.4** 34.4 18.8 spending/paid off some debts in case I was affected At least one of the above 46.6 44.3* 46.4 50.5 three 'income effects' The GFC did not affect me 40.7 33.7** 41.4 48.3 or my family Gross weekly family income: <$500 $500- $1,500+ $1,499 I lost my job 8.2** 7.2 3.7 My partner or another 8.5 12.7 8.8 close family member lost their job I found it harder to 17.1 ** 13.3 8.8 obtain work or get a promotion At least one of the above 27.2** 26.6 18.8 three 'employment effects' I was forced to reduce my 6.8 7.9 6.0 hours of work or take a pay cut My income fell because of 14.9** 17.6 7.7 lower interest rates I reduced my 25.6 35.1 29.4 spending/paid off some debts in case I was affected At least one of the above 42.2* 53.1 39.3 three 'income effects' The GFC did not affect me 41.2** 34.3 49.6 or my family Note: The asterisks (**/*) indicate that the three disaggregated (by age and income) estimates are significantly different from each other (p=0.01/0.05). Source: PEMA survey. Table 2: Changes in subjective wellbeing between 2006 and 2010 (percentages or mean scores) Indicator definition Cross-sectional Linked panel comparisons comparisons (weighted by age) (unweighted) 2006 2010 2006 2010 Assessment of standard of living (ASL)--% very high 31.3 32.8 34.2 32.7 or high Satisfaction with standard of living 66.4 68.2 70.8 73.6 (SILS)--% very or fairly satisfied Happiness (HAP)--% very 89.4 88.6 92.4 90.2 happy or happy Income managing (IM)--% have enough or more than 57.9 58.8 64.1 61.3 needed Degree of choice and control (C&C)--mean score 6.88 7.05** 6.97 7.14 on a 10-point scale Satisfaction with financial situation 5.82 5.97* 6.24 6.33 (SFS)--mean score on a 10-point scale Note: The asterisks (**/*) indicate that the year on year differences are statistically significant (p= 0.01/0.05). Sources: CUPSE, PEMA and PEMA follow-up surveys. Table 3: Changes in financial stress between 2006 and 2010 (percentages) Question: Have there been times over the last 12 months when you have experienced any of the following because of a SHORTAGE OF MONEY? Cross-sectional Linked panel Financial stress indicator comparisons comparisons (weighted by age) (unweighted) 2006 2010 2006 2010 Had to go without food 4.5 4.7 2.5 0.6 * when I was hungry Got behind with the rent 9.0 7.6 4.0 5.4 or mortgage Moved house because the 2.8 2.8 1.5 1.2 rent/mortgage was too high Couldn't keep up with 13.4 13.3 10.7 10.2 payments for water, electricity, gas or telephone Had to pawn or sell 7.2 7.4 4.2 4.8 something or borrow money from a money lender Had to ask a welfare 3.1 2.7 1.7 0.8 agency for food, clothes accommodation or money Wore bad-fitting or 11.6 10.9 8.0 8.3 worn-out clothes Couldn't go out with 24.1 21.9 15.8 17.3 friends because I was unable to pay my way Unable to attend a 3.1 4.0 2.5 4.4 wedding or funeral Couldn't get to an 5.7 5.2 4.6 3.3 important event because of lack of transport None of the above 64.5 65.3 75.1 72.3 I/we have not got enough 6.1 6.2 5.2 3.8 to get by on Unable to raise $2,000 in 14.6 11.2 ** 9.4 9.7 a week in an emergency Note: The asterisks (**/*) indicate that the year on year differences are statistically significant (p= 0.01/0.05). Sources: CUPSE and PEMA surveys. Table 4: Changes in deprivation and severe deprivation, 2006 to 2010 (weighted percentages) Incidence of deprivation Incidence of severe - full sample deprivation - sub sample (D [greater than or equal to] 4 2006 2010 2006 2010 Item (n=2,589) (n=2,574) (n=353) (n=300) Warm clothes and 0.3 0.4 1.8 2.7 bedding, if it's cold Medical treatment if 2.1 1.7 12.9 11.8 needed Able to buy medicines 4.5 3.5 26.9 23.4 prescribed by a doctor A substantial meal at 1.2 0.9 7.6 5.6 least once a day Dental treatment if 14.5 13.1 69.3 64.3 needed A decent and secure 7.1 6.7 33.3 34.6 home Children can 3.7 3.0 21.0 20.5 participate in school activities & outings A yearly dental 9.8 8.0 * 50.2 44.0 check-up for children A hobby or leisure 5.7 5.2 30.9 34.5 activity for children Up to date schoolbooks 4.0 3.4 21.6 21.5 and new school clothes A roof and gutters 4.8 5.1 21.7 27.3 that do not leak Secure locks on doors 5.0 4.4 25.1 25.4 and windows Regular social contact 4.7 4.9 26.2 30.6 with other people Furniture in 2.8 2.2 14.5 14.4 reasonable condition Heating in at least 2.1 2.5 12.2 14.5 one room of the house Up to $500 in savings 19.6 17.8 72.2 71.0 for an emergency A separate bed for 1.7 2.1 7.8 12.2 each child A washing machine 1.1 1.0 4.7 5.7 Home contents 11.1 9.5 53.9 50.2 insurance Presents for family or 6.8 5.5 38.2 33.4 friends at least once a year Computer skills 4.7 2.9 ** 23.3 15.9 * Comprehensive motor 9.8 9.1 40.7 44.5 vehicle insurance A telephone 1.9 3.8 ** 10.8 19.9 ** A week's holiday away 23.6 19.8 ** 84.1 83.3 from home each year Mean incidence of 6.4 5.7 29.6 29.6 deprivation Mean deprivation score 1.43 1.30 6.80 6.89 Note: The asterisks (**/*) indicate that the year on year differences are statistically significant (p= 0.01/0.05). Sources: CUPSE and PEMA surveys. Table 5: Changes in economic exclusion among full and deep-excluded samples, 2006 to 2010 (weighted percentages) Incidence of Incidence of deep exclusion - full exclusion - sub sample sample 2006 2010 2006 2010 (n=2,626) (n=2,605) (n=349) (n=292) Does not have $500 in 26.1 23.2 * 80.0 78.4 emergency savings Had to pawn or sell 7.2 7.4 32.9 36.7 something or borrow money Could not raise $2,000 14.6 11.2 ** 62.5 54.9 * in a week Does not have $50,000 27.7 25.5 69.6 67.6 worth of assets Has not spent $100 on 8.6 6.2 ** 27.7 24.4 a special treat Does not have enough 6.1 6.2 30.2 32.9 to get by on Currently unemployed 4.2 2.7 ** 14.4 11.2 or looking for work Lives in a jobless 19.9 14.4 ** 38.5 27.1 ** household Mean Incidence of 14.3 12.1 * 44.5 41.6 Exclusion Mean exclusion score 1.08 0.92 ** 3.49 3.26 Notes: Deep exclusion is defined as those experiencing at least 8 of 27 indicators of social exclusion across the three broad domains identified by Saunders, Naidoo and Griffiths (2008): disengagement; service exclusion; and economic exclusion. The asterisks (**/*) indicate that the year on year differences are statistically significant (p= 0.01/0.05). Sources: CUPSE and PEMA surveys.
|Gale Copyright:||Copyright 2011 Gale, Cengage Learning. All rights reserved.|