AUSTRALIAN ATTITUDES TO UNEMPLOYMENT AND UNEMPLOYED PEOPLE.

Introduction

Over the past decade social security policy in Australia towards people of working age has become increasingly conditional on their demonstrating thorough and continuing job search effort. Though work tests have long been part of entitlement to unemployment benefits, the current regime of activity testing has moved well beyond earlier requirements. Initial moves in this direction began under the Labor Government in the mid- 1980s, but have been taken much further by the Coalition Government.

In emphasising `mutual obligation' as a guiding principle in recasting social security for the unemployed, the Howard Government argues that it is going with the grain of public opinion. Is this claim justified? Does the policy shift over the last decade reflect a movement in public opinion ? Do Australians support the idea that benefits should only be available on condition that unemployed people demonstrate ever greater job search effort: and, if so, does this apply across the board or only to particular groups of the unemployed? More generally, what does the public see as the responsibilities of government towards the unemployed?

In fact we know little about how Australians view the balance between rights and obligations for unemployed people, although we do have more attitudinal evidence on broader questions relating to unemployment and public support for job seekers. This article is mainly based on that evidence.

We begin by briefly outlining the changes in social security policy towards the unemployed in the last decade. We then review evidence from polls and attitudinal surveys on changes over time in the significance of unemployment as a social problem, the responsibilities of government, and public attitudes towards activity testing.

The policy shift from entitlement to obligation

The first distinct elements of the new framework of support for unemployed people emerged from Labor's Social Security Review (Cass 1988). In line with the `active society' policy framework promoted by the OECD (Gass 1988; OECD 1990), the new emphasis on active job search was signalled by changes in the nomenclature of payments, from the `passive' Unemployment Benefit to the more `active' Newstart and Job Seekers' Allowance. Under the initial Newstart strategy, from 1991, intensive individualised job seeker interviews were introduced, and sanctions for noncompliance increased.

From 1994, Labor's Working Nation policy expanded labour market programs and introduced a job placement guarantee, along with `case management' for the long-term unemployed, and individual activity agreements. In return, stronger penalties were introduced for job seekers judged not to have met their obligations.

When the Coalition Government took office in 1996 there was a move away from labour market programs and job guarantees toward greater levels of obligation on the part of the job seeker, including a strengthening of the activity test. This included a requirement to provide more details of job search activity; an increase in the issuing of Employer Contact Certificates -- used to verify job seekers' approaches to prospective employers; and the introduction of the Jobseeker Diary, for recording efforts to find work. Penalties for non-compliance with activity testing were again increased.

The Coalition's second term has seen further emphasis placed on the principle of `mutual obligation', typified by the expansion of `Work for the Dole' from a program only for the young unemployed to one covering those aged up to 34. In January 1999 the Government also announced that young unemployed people with literacy and numeracy problems would be obliged to undertake remedial courses or face penalties.

Another dimension of change has been the progressive contracting out of employment assistance to private and non-government agencies. This culminated in 1998 with the creation of a quasi-market system of competitive tendering and payment by results (the Job Network). Now, not only are the details of a job seeker's activity test a matter of individual agreement with an employment service provider, but the responsibility for recommending a sanction for breach of an agreement often lies with an agency outside the public service. This in particular has disquieted some critics (e.g. Carney 1998) who argue that traditional welfare rights are being undermined.

Overall, these changes represent a major shift from generalised entitlement to benefits when out of work. Yet relatively little has changed in the external environment. Registered unemployment has declined somewhat over the decade, but still remains high, at around 7.5%, with surveys indicating that many more people who are not registered as unemployed would like to work. We need to consider, therefore, whether public attitudes towards the problem of unemployment have also shifted in some way.

The significance and causes of unemployment

Given the persistently high numbers of people out of work over the last 20 years, it is not surprising that polls tend to find unemployment one of the most frequently mentioned issues of concern. Yet the importance accorded to it has fluctuated. The Morgan Poll over the decade from 1982, for example, regularly asked, `What are the three most important things the Federal Government should be doing something about?' The response `reduce unemployment' showed a progressive decline from 79% in February 1983 to 15% in 1990 (Morgan Poll, various dates). The corresponding rate for September 1983, however, was 63%. It seems possible that the seven-year trend reflected the movement of the business cycle, as concern about joblessness declined in parallel with the fall in the rate of unemployment under the Hawke Labor Government. More recently, a similar question found unemployment among the issues most frequently mentioned -- by 56% of respondents in July 1994, falling (again with the rate of unemployment) to 39% in March 1995 and then rising again to 46% by November 1995 (Morgan Poll 1995).

As Table 1 shows, unemployment was a major issue in the 1996 federal election campaign, second only to health and Medicare as `extremely important' in voting choices (Jones, McAllister & Gow 1996). This seems to contradict the usual finding that `welfare' issues do not figure as a major element in people's accounts of their electoral deliberations (Zaller 1992). We might also note Smith and Wearing's observation (1987) that unemployment is usually seen as more of an economic than a social issue.

Table 1: Views on Electoral Issues of Importance, by Party Affiliation

Party closest to               Liberal-       No       Don't
own view on ...        Labor   National   difference   know

Taxation               21.0      42.1        20.6      16.4
Immigration            19.6      46.1        15.7      18.6
Education              28.8      31.1        23.8      16.4
The environment        29.5      29.7        25.8      15.0
Industrial relations   33.3      41.8        10.0      14.9
Health and Medicare    38.6      38.4        14.0       8.9
Links with Asia        34.5      25.6        20.4      19.4
Defence                15.8      24.5        27.2      32.5
Interest rates         20.2      31.0        24.7      24.1
Unemployment           23.6      47.4        16.9      12.1
Privatisation          24.3      33.6        23.0      19.2
Inflation              21.6      29.5        24.0      25.0
State/Territory        13.8      20.5        26.9      38.7
  issues

                       Extremely     Quite     Not very
How important is ...   important   important   important

Taxation                  44.0       36.1        19.9
Immigration               30.8       36.4        32.8
Education                 50.4       33.1        16.5
The environment           41.9       42.0        16.1
Industrial relations      45.9       35.8        18.3
Health and Medicare       67.7       24.6         7.7
Links with Asia           21.4       40.0        38.6
Defence                   25.3       36.9        37.8
Interest rates            49.5       33.5        17.0
Unemployment              61.4       28.4        10.2
Privatisation             30.0       42.0        28.0
Inflation                 47.5       36.7        15.7
State/Territory           15.5       36.4        48.2
  issues

Source: Australian Election Study 1996 (Jones, McAllister and Gow, 1996).

A number of polls have canvassed perceptions of the causes of unemployment. A summary of Morgan polls concluded that the public shifted its preferred explanation from `laziness' from 1975 to 1985, through `union wage claims' over the next few years, to `world economic pressures' in the 1980s, as the structural and enduring nature of the problem became apparent (Smith & Wearing 1987: 62-3). At no stage, however, did fewer than about a quarter of those polled see wilful idleness as a major factor.

Figure 1 extends this analysis to the early nineties. Throughout the period, the proportion ascribing unemployment to the action of employers remained low and fairly stable, between 10-15%, with a small decline during the last recession. There was also a downward trend from the early eighties in the tendency to blame trade unions and their wage demands. This perhaps reflects the decline in both union membership and employment in traditionally unionised industries. People may be less inclined to blame the unions the less effective they perceive them to be.

[Figure 1 ILLUSTRATION OMITTED]

The most significant pattern to emerge from these data concerns the belief that joblessness results from people's reluctance to work. This is plotted separately for 1975-93 in Figure 2, along with the level of unemployment. The scales of the left and right hand axes are different, but the pattern of movement over time suggests that the two series are mainly negatively correlated. Although there were time lags and periods of positive association, it appears that as unemployment rose people were less inclined to ascribe responsibility to the unemployed themselves. Then, as the jobless rate fell, there was a growth in the feeling that work must be available for those who wanted it. The 1980s saw a steady rise in willingness to personalise responsibility for unemployment, followed by a dramatic drop, to under one-fifth, in the first two years of the recession.

[Figure 2 ILLUSTRATION OMITTED]

After the end of 1991 the figure began to rise again, although unemployment was still increasing. It is not clear why this should be so, though we might speculate that the `active society' concept which was informing changes to social security provision during this period may have percolated through to the public consciousness. Thus unemployment again perhaps began to be seen as a matter of personal responsibility.

One interpretation of Figure 2 would be that the Australian public has come to appreciate the argument that in times of high unemployment unwillingness to work on the part of some unemployed people will have a negligible effect on the overall rate of unemployment, as other more motivated job seekers will fill any vacancies. Bradbury et al. (1988) also note that the peaks in the propensity to blame the unemployed tend to occur under Labor governments and the troughs under the Liberals. The more recent data, however, show both a rise and a dramatic fall in this viewpoint under periods of Labor administrations.

In Beed and McNair's The Changing Australian, 1983 data set (cited in Graetz 1987), explanations of unemployment were sought from samples of both the general work force and of government, business and union elites. The authors found a notable contrast between senior bureaucrats and union leaders, who emphasised macro-economic and structural factors, and business executives and ordinary employees, who saw things in more individualistic terms. Neither Graetz nor the original researchers speculated at any length on the reasons for this difference, but one explanation might be that people's understanding of causation is relative to what they believe they can influence.

One of the explanations frequently put forward for shifts in attitudes towards the unemployed is the role of the media (and politicians) in creating public furores about `dole bludgers'. Windschuttle (1980), for example, directly attributed the high percentage of people blaming the unemployed for their own predicament in the late 1970s to media campaigns against alleged dole cheats. He argued that the public is receptive to such stories because they represent conflicts between readers' (or viewers') own desires and the constraints of capitalist society. The interplay between fascination with the idea of breaking the rules of a society and the constraints felt by most people against actually doing so then emerges in the form of moral indignation at deviant activity.

Bradbury et al. (1988) suggested that if this moral indignation stems from constrained desire, those who are most constrained might be expected to be the most indignant. This is supported by the tendency of working class respondents to be more likely than others to blame the unemployed for their joblessness. A Morgan Gallup Poll in 1985, for example, found that blue-collar workers were substantially more likely to take this view than white-collar workers.

It is generally argued (for example, Deacon 1978 in the case of the UK) that this effect is more pronounced when blue-collar workers themselves face declining living standards. The hypothesis is also consistent with an increased incidence of blaming the unemployed during the 1980s, when real wage levels were falling. The shift towards blaming government and world economic pressures after the late 1980s could perhaps represent a response to Labor's policies of fiscal conservatism and deregulation in the face of industrial globalisation.

Graetz (1987) also noted that in the early 1980s the public was particularly likely to attach personal responsibility to young people for their unemployment. This was a sentiment apparently shared by young people themselves; the Morgan Gallup Poll (1985) found that nearly half (47%) of 14-17 year olds attributed unemployment to personal unwillingness to work. Twelve years later, the controversial `Current Affair' television program involving young members of the Paxton family can be read as an indication of public concern about individuals receiving income support when they are not genuinely seeking employment. Yet it is also, arguably, a sign that public doubts about the strength of the work ethic among youth can still easily be exploited by the media, in spite of widespread concern about the current shortage of opportunities for young people.

Given the ubiquity of the `dole bludger' in popular mythology, it is surprising how little formal research has been carried out on the subject. There have been surveys of unemployed people themselves and descriptive accounts of their lives and problems (Van Moorst 1983, 1984). Yet Australian research on people's personal or ethical evaluations of their unemployed fellow citizens is scarce.

One indication of how unemployed people are currently viewed in Australia in relation to various other `unpopular' social groups comes from a 1997 Morgan poll. Respondents were invited to rate a list of groups they would not like to have as neighbours. Unemployed people ranked low on this list, included by only 4%, compared, among the higher scoring groups, with `drug addicts' (65%), `people with a criminal record' (41%), `emotionally unstable people' (29%) and `members of new religious movements' (20%). It may be that the high level of unemployment and its chances of occurring across a wide range of social groups means that it touches too many people for the individuals concerned to become objects of personal disapproval. Yet, as we have seen, Australians are often ready to blame unemployed people themselves for their circumstances. It seems more likely that being unemployed does not in itself impart as much of a social threat as that offered by the other groups.

The role of government

Although the inclination to blame government for unemployment seems to have been growing since the mid-1980s, Australians still do not appear to rate assistance to the unemployed high among priorities for government action.

One of the difficulties with interpreting survey data on attitudes to public expenditure arises from respondents' tendency to favour more government spending on a wide range of items, while simultaneously supporting tax cuts (Welch 1985). Answers to questions about taxation are also sensitive to the wording, so that enthusiasm for tax cuts tends to be lower in response to questions which mention reduction in social spending as a consequence (Hedenius 1985).

Braithwaite (1988:31-2) attempted to deal with these problems in an Australian study using the first of the National Social Science Surveys. He subtracted the percentage of respondents favouring reductions in specified areas of public spending from the percentage favouring increases. The differences were used to rank the issues in order of priority. As might be expected, all (except foreign aid, at -24%) were positive, but varied from margins of 67% and 63% for education and `dealing with drug abuse', to 16% for `improving the conditions of Aborigines' and 23% for `providing assistance to the unemployed'. The unemployed thus seemed to be a more popular focus for assistance than foreigners and indigenous Australians, but still ranked low in overall priorities for greater intervention.

By seeing other priorities as more urgent than unemployment, Australians resemble citizens of most countries (Taylor-Gooby 1995). Even so, Australian attitudes seem ungenerous by international standards, as indicated in Table 2. This is based on analysis of data from the International Social Survey Programme (ISSP), a collaborative study by attitude survey research teams in a number of countries.

Table 2: Comparative Preferences for Spending on Unemployment Benefits:1985-90

                Much more   More   Same as now

Australia
  1985             3.7       8.9      35.5
  1990             1.9       8.4      31.6
West Germany
  1985             7.9      26.8      52.1
  1990             9.4      27.2      49.8
Great Britain
  1985            12.5      29.4      39.3
  1990             7.8      28.5      46.5
USA
  1985             7.6      18.0      49.4
  1990             7.2      20.2      51.4
Italy
  1985            14.2      41.4      27.0
  1990            15.4      38.0      31.2

                Less   Much less   No.

Australia
  1985          30.1     21.8      1475
  1990          39.4     18.7      2375
West Germany
  1985          10.5      2.7      1006
  1990          11.0      2.4      2719
Great Britain
  1985          14.7      4.1      1598
  1990          13.5      3.7      1114
USA
  1985          17.1      7.8       881
  1990          15.2      6.0      1125
Italy
  1985          10.7      6.8      2316
  1990           9.4      6.1       978

Source: International Social Survey Programme, Role of Government I and II, 1985 and 1990

In 1985 nearly 52% of Australian respondents were in favour of reducing spending on unemployment benefits, compared with only one-quarter of those from the US and 19% of Britons. By 1990 the Australian majority in favour of reductions had grown to 58%, while falling in all the other countries.

The question is whether cross-national differences reflect something specific to popular attitudes toward the unemployed, or whether they simply correspond to national levels of enthusiasm for government per se. It is also possible that attitudes are influenced by payment levels. If benefits are perceived as generous, or at least adequate, a preference not to increase expenditure might not necessarily indicate a lack of concern for the unemployed.

In order to address the first question, Figure 3 compares attitudes towards spending on the unemployed with other areas of expenditure. In each national population `mass' social services were more popular than assistance to the unemployed: most people, that is, supported spending on pensions, health, education, police and `the environment'. Even so, the apparent hostility to unemployment benefit in Australia stands out. Whereas in other countries the areas in which most people would have preferred less public spending were defence and the arts, in Australia unemployment payments were the primary target for reductions.

[Figure 3 ILLUSTRATION OMITTED]

Figure 4 further breaks down the national responses on unemployment payments by labour force status. In comparison with the other countries, Australians overall showed a marked preference for less spending on unemployment benefits, with nearly one-fifth of the unemployed themselves voting for less and under 40% calling for more.

[Figure 4 ILLUSTRATION OMITTED]

One possible explanation for Australians being less supportive than other nations of expenditure on unemployment benefits lies in the absence of insurance elements within Australian social security -- a key difference between Australia and all the other countries surveyed. Even the United States, where popular opinion has historically been hostile to `welfare', has insurance for (short-term) unemployment. Given that the pattern of opinion for the US, which is normally comparable with Australia, was much closer to that of the European countries, it seems possible that the survey picked up attitudes towards unemployment insurance rather than the residual, means-tested welfare payments.

Debate about the structure and financing of social provision in Australia throughout the 20th century has frequently returned to the introduction of contributory social insurance in areas such as age pensions, health and unemployment (Saunders 1995). In the 1980s, the Social Security Review speculated that `a non-contributory scheme may have less legitimacy because there is no explicit perception of the employees having "earned" their benefit through their years in the labour force' (Cass 1988: 37). Finally, however, the Review rejected this argument on the grounds that `many Australians are unfamiliar with insurance principles, and would feel that they have earned their benefit by paying their taxes' (Cass 1988: 37). Castles (1985), on the other hand, has argued that the labour movement's traditional opposition to contributory benefits has placed serious obstacles in the way of welfare state financing.

If it is true that comparative surveys are picking up differences in attitudes towards different types of provision, then perhaps the `wage earners' welfare state' (Castles 1985), with its reliance on flat-rate, means-tested support for the unemployed, may have bred a lack of generosity in public attitudes towards benefit expenditure. Australians may be negative about spending on assistance benefits, which are met from general taxation, whereas other national respondents may be offering more positive views of insurance benefits, which are at least partly funded by people's own contributions.

However, this interpretation runs counter to the possibility canvassed earlier that the more generous the payment the less likely people are to prefer increases in the level of expenditure. Most European insurance-based unemployment benefits are earnings-related and thus tend to be paid at a higher level than Australia's assistance payments. Unfortunately the data available from the ISSP surveys do not allow any more detailed exploration of how different funding mechanisms for benefits may affect public attitudes towards them.

The idea that `residual' welfare states like Australia ascribe a particularly strong stigma to unemployment benefit receipt is an argument dating back at least to the work of Titmuss (1974). Though warning of the dangers of generalising across countries, Titmuss contended that the residual model of social policy fostered images of pauperism among recipients, making them feel like `chisellers, cheaters, welfare bums, and abusers of the system' (1974: 45).

The categorisation of Australia as a residual welfare state like the United States has, however, been challenged in recent years (Mitchell, Harding & Gruen 1994; Eardley et al. 1996). There is also little direct empirical evidence to suggest that stigma is any more of a feature in Australia than in other countries. Furthermore, the limitation to drawing such a conclusion from the ISSP survey data is that like is not being compared with like. The actual levels and type of provision in unemployment differ considerably among the countries surveyed, which is likely to affect perceptions of how much more or less expenditure is desirable. This is a point demonstrated by Taylor-Gooby (1995), who shows that for most European Union countries the incidence of the belief that the jobless are well protected increases with the proportion of per capita GDP spent on each jobless person.

An alternative approach, given the difficulty of comparing the preferences expressed by populations differently served in terms of existing expenditure levels, is to ask what people see as the appropriate role of the state with respect to employment and unemployment, compared with other activities. Table 3 presents respondents' views on the responsibilities of governments in Australia and four other countries in or around 1990. The predominant views in different countries varied considerably according to the issue. Overwhelming majorities in each country saw providing health care for the sick or ensuring an adequate standard of living for older people as legitimate concerns of government. There was, by contrast, substantial variation in the proportions who believed that finding a job for anyone who wanted one, providing a decent standard of living for the jobless, and reducing income differences between the rich and the poor, were the kinds of things governments should be doing.

Table 3: Governmental Responsibilities, Comparative Attitudes: 1990

                  Provide Jobs   Control     Provide
Country             for All      Prices    Health Care

Australia
  Should be           40.4        81.9        93.1
  Should not be       56.9        16.9         6.1
  Don't know           2.7         1.3         0.8
West Germany
  Should be           71.3        67          93.2
  Should not be       24.8        29.2         4.4
  Don't know           3.9         3.8         2.3
UK
  Should be           64.3        86.9        96.9
  Should not be       30.5         9.7         0.9
  Don't know           5.1         3.5         2.2
USA
  Should be           40.3        71.1        84.3
  Should not be       51.6        22.2        10.3
  Don't know           8.1         6.7         5.4
Norway
  Should be           80.6        89.1        97
  Should not be       15.5         8           1.4
  Don't know           3.9         2.8         1.6

                  Look After     Aid      Look After
Country              Aged      Industry   Unemployed

Australia
  Should be          93.4        82.8        53.3
  Should not be       5.8        14.8        42.3
  Don't know          0.8         2.4         4.4
West Germany
  Should be          92.6        49          74.3
  Should not be       5.2        44.7        20.5
  Don't know          2.2         6.2         5.2
UK
  Should be          97          89.9        79.9
  Should not be       1.2         6          15.7
  Don't know          1.8         4.1         4.4
USA
  Should be          82.6        63.8        48.2
  Should not be      11.8        26.8        43
  Don't know          5.7         9.4         8.8
Norway
  Should be          97.6        62.8        87.1
  Should not be       1.1        30.4         9
  Don't know          1.3         6.8         3.8

                    Reduce                  Provide
                    Income      Aid Poor   Low-income
Country           Differences   Students    Housing

Australia
  Should be          49.7         86.5        75.3
  Should not be      47.4         11.6        20.7
  Don't know          2.9          1.9         4
West Germany
  Should be          59.4         82.3        75.9
  Should not be      33.9         13.5        19.2
  Don't know          6.7          4.2         4.9
UK
  Should be          72.6         90          89.5
  Should not be      22.3          7           6.9
  Don't know          5.2          3           3.6
USA
  Should be          40.3         80.9        67.7
  Should not be      50.4         11.6        24.4
  Don't know          9.4          7.5         7.9
Norway
  Should be          67.2         75.1        70.3
  Should not be      26.6         19          23.5
  Don't know          6.2          5.9         6.3

Note: Question: Should it be the responsibility of government to ...?

Source: ISSP, Role of Government II, 1990

There are indications that these differences might be linked to prevalent institutional arrangements. The United States and Australia, two societies whose institutions are commonly characterised as liberal and market-oriented, showed the lowest rates of support for all three of these propositions, while with respect to full employment and income redistribution, the US had slight majorities against any government responsibility. By contrast, in Norway, a full-employment welfare state with extensive tax-transfer and labour market programs, more than two-thirds of the sample favoured each of these three areas of responsibility, with 87% agreeing that the Government should provide a decent standard of living for the unemployed.

Australia trailed the field in support for unemployment relief programs. Endorsement for government providing a decent living standard for the jobless was also comparatively low, though similar to that in the US. The even lower level of support for `finding a job for everyone who wants one' would seem to preclude the idea that the public preferred an `active society' approach over a passive policy of income support. However, this might be traced to cross-national differences in interpretation of the question. Does the government actually have to give people jobs or simply ensure that economic and social conditions are conducive to employment? This indeed is the thrust of current Australian policy on unemployment.

Other questions in the 1990 ISSP Role of Government survey included a series of possible economic policy initiatives for the respondents' consideration and evaluation. Of all the countries included, Australia had the lowest level of support for `government financing of projects to create new jobs' and little enthusiasm for most of the alternatives. On the other hand, there was considerable support for assistance to industry.

More recent data from the ISSP are not vet available for analysis, but some comparable information on public attitudes can be found from other sources. A recent survey carried out for the Uniya Jesuit Social Justice Centre, for example, asked about Australians' views on taxation for certain kinds of social expenditure (Baldry & Vinson 1998). Table 4 summarises the main results. It indicates substantial support for increased spending in certain areas, including job training for unemployed people, even if this resulted in extra taxation. However, as in the ISSP, unemployment expenditure remains the least popular of the areas canvassed, though only by a few percentage points, and there seems to be only minority support for tax and spending cuts in this area. Again there are problems of interpretation: a lack of support for job training may indicate doubts about its effectiveness at reducing unemployment rather than a wish to reduce support for the unemployed per se.

Table 4: Australians' Preferences for Taxation and Social Expenditure, 1998

                          Option best for
                              country

                          Increase spending
                          by increasing
                          income tax by
                          one or two cents
Spending Area             in the dollar

Education                       49
Health                          48
Aged care                       48
Support for families in
  need                          42
Job training for
  unemployed people             39
The environment                 37

                                Option best for country

                            Keep income       Cut spending by
                           tax as is and      reducing income
                          spend about the    tax by one or two
Spending Area               same as now     cents in the dollar

                                      Percentages

Education                       32                  15
Health                          33                  16
Aged care                       33                  16
Support for families in
  need                          36                  18
Job training for
  unemployed people             36                  21
The environment                 40                  19

                           Option best
                           for country

                           Not
                          known/
                           not
Spending Area             stated   Total

Education                   4       100
Health                      3       100
Aged care                   3       100
Support for families in
  need                      4       100
Job training for
  unemployed people         4       100
The environment             4       100

Note: The survey included 969 persons aged over 18 across all States and Territories of Australia except the ACT, interviewed in February 1998. Respondents were asked to assume all tax evasion loopholes were closed, that taxes remained progressive, and changes in tax levels could result in direct changes in public spending.

Source: AC Nielsen Omnibus Survey for Uniya, reported in Baldry and Vinson, 1998

A further recent source of attitudinal information is the Middle Australia Project, carried out from the School of Sociology at the University of New South Wales (Tables 5 and 6). In 1996 the project surveyed a national sample of 400 adults identified as `middle Australians' on the basis of their residence in certain census enumeration districts. The respondents did not represent a fully random sample of the population, but the survey included some of the same questions as in the ISSP.

Table 5: Responses to Selected Statements and Questions from the Middle Australia Project: 1996 (row percentages)

                                      Agree
                                     Strongly   Agree

The welfare state makes people
nowadays less willing to look
after themselves                       14.9     41.2

People receiving social security
are made to feel like second class
citizens                                9.3     42.9

The government should spend
more on welfare benefits for the
poor, even if it leads to higher
taxes                                   7.6     26.0

Around here, most unemployed
people could find a job if they
really wanted to                        5.3     32.0

Most people on the dole are
fiddling in one way or another          2.3     13.2

It is everyone's responsibility to
find a job for themselves, and
nothing to do with the government       3.8     26.9

Young people today should keep
on challenging traditional
Australian values                      10.9     38.9

                                      Neither
                                     Agree nor
                                     Disagree    Disagree

The welfare state makes people
nowadays less willing to look
after themselves                       15.9        18.7

People receiving social security
are made to feel like second class
citizens                               25.8        17.4

The government should spend
more on welfare benefits for the
poor, even if it leads to higher
taxes                                  23.2        30.8

Around here, most unemployed
people could find a job if they
really wanted to                       18.5        28.4

Most people on the dole are
fiddling in one way or another         27.8        32.7

It is everyone's responsibility to
find a job for themselves, and
nothing to do with the government      23.6        33.0

Young people today should keep
on challenging traditional
Australian values                      27.5        17.3

                                     Disagree   Don't
                                     Strongly    Know

The welfare state makes people
nowadays less willing to look
after themselves                        8.1      1.3

People receiving social security
are made to feel like second class
citizens                                3.5      1.0

The government should spend
more on welfare benefits for the
poor, even if it leads to higher
taxes                                  10.4      2.0

Around here, most unemployed
people could find a job if they
really wanted to                       12.4      3.3

Most people on the dole are
fiddling in one way or another         18.7      5.3

It is everyone's responsibility to
find a job for themselves, and
nothing to do with the government      11.9      0.8

Young people today should keep
on challenging traditional
Australian values                       3.8      1.5

Note: n = 394-396

Source: Middle Australia Project, SPRC analysis

Table 6: Views on Government Expenditure, from Middle Australia Project: 1996 (a) (row percentages)

                           Spend            Spend
                           much    Spend    same as
                           more    more     now

The environment            17.9    41.9     33.0
Health                     26.3    50.3     18.6
The police and law
  enforcement              12.0    41.2     38.1
Education                  27.6    50.4     19.2
The military and defence    3.8     8.4     36.8
Old age pensions            7.4    33.5     50.9
Unemployment benefits       3.3    16.1     49.7
Culture and the arts        5.6    23.7     41.0

                           Spend   Spend       Can't
                           less    much less   choose

The environment              5.1      1.0       1.0
Health                       3.1      0.8       1.0
The police and law
  enforcement                6.4      1.3       1.0
Education                    1.0      1.3       0.5
The military and defence    30.4     18.9       1.5
Old age pensions             4.6      1.8       1.8
Unemployment benefits       22.7      6.6        .15
Culture and the arts        20.9      7.1       1.8

Notes: (a) The question was: `Please show whether you would like to see more or less government spending in the following areas. Remember that if you say `much more' it might require a tax increase to pay for it'.

(b) n = 391-393

Source: Middle Australia Project, SPRC analysis

Table 5 presents a somewhat conflicting picture. A substantial proportion of respondents -- between 17% and 33%, depending on the question -- either offered no firm view on the statements or did not know. A majority of those expressing a definite view supported the idea that the welfare state saps personal responsibility -- this proposition receiving the highest level of strong agreement -- but also agreed that social security recipients are treated as second class citizens. Just over one-third supported more spending on welfare benefits even at the risk of higher taxes, but more than two-fifths opposed this extra spending. Over one-third also felt that jobs in their area were available to people who looked hard enough, but a larger percentage disagreed.

Only a small minority (just under 16%) clearly supported the idea that dole cheating is widespread, despite the popular `dole bludger' rhetoric discussed above. Also, notwithstanding the view that the welfare state undermines personal responsibility, the idea that government has no role in helping people to find work received only limited support.

In Table 6, on attitudes to public spending, a similar picture emerges to that from the 1990 ISSP, and to a lesser extent from the Uniya survey, in that unemployment benefits are not among the main candidates for increased spending. Expenditure on defence is even less popular than on unemployment benefit, but otherwise the patterns emerging from the various surveys are fairly consistent.

Activity testing and `mutual obligation'

As stated earlier, there is little information on public attitudes towards the detail of activity testing. One of the few studies in this area is by Weatherley (1993), who surveyed the attitudes of social security claimants themselves towards compliance with eligibility rules and procedures. Even among this group he found considerable support (76%) for the then requirement to lodge the mandatory `work effort' form, even though the survey took place at the peak of the recession when there were likely to be large numbers of job seekers for each listed vacancy. Views expressed by claimants were fairly similar to those of many DSS staff, such as that the requirement acted as a safeguard against abuse. Weatherley suggested that for some respondents it seemed that meeting these requirements also helped assuage guilt about being unemployed and requiring help.

One further question from the Middle Australia Project is relevant here. Respondents were asked to record their views on a series of issues by marking a point on a scale between two opposing statements. One set concerned the obligation of job seekers to accept any work available. Figure 5 shows the responses on a scale of 1 to 10. Opinion clearly varied considerably, but with a majority (58%) leaning somewhat towards the obligation end of the spectrum.

[Figure 5 ILLUSTRATION OMITTED]

The one recent policy change which has been tested in published opinion polling is the `Work for the Dole' scheme. In February' 1997, the Morgan Poll asked respondents whether they approved or disapproved of the proposed scheme, first on a compulsory and secondly on a voluntary basis. Overall, 51% said they approved strongly of a compulsory scheme and a further 21% mildly. Mild disapproval came in at 9% and strong disapproval at 16%. Women were slightly more likely to approve than men (73% to 71%) and the highest approval rating by age (77%) came, perhaps surprisingly, from the 16-19 year age group. In terms of occupational groups, there was relatively little difference in overall approval levels except for semi-skilled and unskilled workers, who gave the scheme the lowest approval levels of 56%. The slimmest majority in favour also came from unemployed people seeking work (55%), as opposed to 73% for all those in work. Making the scheme voluntary brought slightly higher levels of approval overall.

Clearly such a poll has limitations as an indicator of public attitudes. Respondents would have had little idea of the actual operations of the scheme and were commenting on a concept rather than a set of concrete proposals -- one, moreover, that had been strongly promoted in the media by the Government. Nevertheless, it does appear that a substantial majority of the Australian public favours, at least in principle, a greater degree of obligatory participation by young unemployed people.

Discussion

At the beginning of this paper we posed a number of questions about public attitudes towards unemployment and unemployed people. In particular we asked whether recent governments have been moving in line with public opinion in imposing increasingly strict conditions for the receipt of unemployment payments and reinforcing the idea of `mutual obligation'.

Overall, the evidence seems ambiguous. The literature on Australian attitudes to unemployed people is not extensive and is particularly lacking when it comes to the details of activity testing. Australians do seem in the past to have been noticeably inclined to attribute at least part of the responsibility for joblessness to people's own unwillingness to work. However, the tendency to hold this view seems to have decreased sharply in the late 1980s in line with a drop in unemployment. This suggests that people's views shift according to how far they regard work as being available for those who want it. Nevertheless, after 1991 there was another apparent rise in willingness to personalise responsibility for unemployment, leading us to speculate that the `active society' discourse promoted by government may have encouraged people again to see unemployment as a matter of personal responsibility.

In spite of this, Australians do still have sympathy for unemployed people, resist the notion that they are all welfare cheats or should assume full responsibility for their predicament, and see a role for government both in helping them to find work and in providing them with an income when out of work. These views should not necessarily be seen as contradictory, but more an example of the complexity of the packages of attitudes and opinions -- what Offe (1993) has described as the `moral repertoires' -- upon which people draw in formulating their responses to issues of public welfare and governance.

Nevertheless, it is hard to escape the view that, by international standards, Australians appear to take a relatively hard line on the responsibilities of unemployed people -- especially the younger unemployed -- to actively seek and accept work. A majority also seem disinclined to support increases in public expenditure on unemployment benefits. It may be that Australia's system of means-tested income support for the unemployed and its lack of insurance-based payments have tended to foster hostility to `welfare'. On the other hand, it could represent a perception that levels of income support are already adequate, or alternatively that people would prefer the problem of unemployment to be solved -- since it consistently figures high on the list of public concerns -- rather than see extra spending on benefits per se.

The apparent support, in principle at least, for `Work for the Dole' suggests that the notion of mutual obligation arising from the provision of support for unemployed people is one which has broad community backing, particularly when applied to young people. Yet the limited data from opinion polls on this topic tell us little about why people support the scheme, what they think it actually involves or what it might be expected to achieve.

Although unemployment remains a matter of deep public concern, we have little detailed information on what people think government should be doing to solve the problem. In spite of evidence for support at a general level for obligations on recipients of unemployment allowances, we still know little about how far people believe activity testing should go; which categories of unemployed client should or should not be obliged to demonstrate their job search effort; in what ways job search should be demonstrated; and what support exists for the level of sanctions now operating for non-compliance.

Attitudinal surveys can, of course, go only so far in answering these questions. In-depth exploration is also required in order to understand how people exercise their `moral repertoires'. As Dean (1998: 148) has put it, the concern is not only about the expressed opinion of respondents, but also `about the way in which those opinions [are] constructed out of everyday experience and discourse'.

Acknowledgements

This paper is based on a wider study of attitudes to unemployment, including those of employers, commissioned by the former Department of Social Security (Eardley & Matheson 1998). The authors would like to thank Michael Pusey for access to the Middle Australia data set, and John Powlay and Anne Gregory from the (now) Department of Family and Community Services for their helpful comments. Responsibility for any errors of fact or interpretation remains with the authors.

References

Aalberg, T. (1998) Founding the Welfare State: Comparative Public Opinion on Taxes and Redistribution of Income, Luxembourg Income Study Working Paper No. 180, Syracuse, LIS.

Baldry, E. & Vinson, T. (1998) `Majority of Australians favour taxing for social expenditure', Uniya Brief Research Report No. 2, Sydney, Uniya.

Bradbury, B., Encel, D., James, J. & Vipond, J. (1988) Poverty and the Workforce, SWRC Reports and Proceedings No. 72, Social Welfare Research Centre, University of New South Wales.

Braithwaite, J. (1988) `Economic policy: what the electorate thinks.' In J. Kelley & C. Bean (eds) Australian Attitudes, Sydney, Allen & Unwin.

Burgoyne, C., Swift, A. & Marshall, G. (1993) `Inconsistency in beliefs about distributive justice: a cautionary note', Journal for the Theory of Social Behaviour, 23 (4), 327-42.

Carney, T. (1998) `Merits review of "contractual" social security payments', Journal of Social Security Law, 5 (1), 18-43.

Cass, B. (1988) Income Support for the Unemployed in Australia: Towards a More Active System, Canberra, AGPS.

Castles, F. G. (1985) The Working Class and Welfare: Reflections on the Political Development of the Welfare State in Australia and New Zealand 1890-1980, Sydney, Allen & Unwin.

Deacon, A. (1978) `The scrounging controversy: public attitudes towards the unemployed in contemporary Britain', Social and Economic Administration, 12 (2), Summary, 120-32.

Dean, H. (1998) `Popular paradigms and welfare values', Critical Social Policy, 18 (2), 131-56.

Eardley, T. & Matheson, G. (1998) Employer and Community Attitudes to Unemployment and Unemployed People, Report for the Department of Social Security, Social Policy Research Centre, University of New South Wales.

Eardley, T., Bradshaw, J. Ditch, J., Gough, I. & Whiteford, R. (1996) Social Assistance in OECD Countries, Research Reports Nos 46 and 47, Department of Social Security, London, HMSO.

Gass, J. (1988) `Towards the "active society"', OECD Observer, 152, 4-8.

Graetz, B. (1987) `Unemployment, social policy and public opinion', Australian Journal of Social Issues, 22 (4), 321-31.

Hedenius, A. (1985) `Citizens strike a balance: discontent with taxes, content with spending', Journal of Public Policy, 5 (3), 349-63.

International Social Survey Programme 1985-1992 (1994), [CD-ROM of unit record data and documentation], Cologne, Zentralarchiv fur Empirische Sozialforschung.

Jones, R., McAllister, I. & Gow, D. (1996) Australian Election Study, 1996 [computer file], Social Science Data Archives, Australian National University.

Morgan Gallup Poll (1985) No. 698, November, Roy Morgan Research Centre Pty. Ltd., Melbourne.

Morgan Poll (formerly Morgan Gallup Poll) (various dates) Roy Morgan Research Centre Pry Ltd, Melbourne.

Morgan Poll (1997) No 1409, Roy Morgan Research Centre Pty. Ltd, Melbourne.

Miller, D. (1992) `Distributive justice: what the people think', Ethics, 102, 555-93.

Mitchell, D., Harding, A. & Gruen, F. (1994) `Targeting welfare -- a survey', The Economic Record, 210, September, 315-40.

Organisation for Economic Co-operation and Development (1990) Labour Market Policies for the 1990s, Paris, OECD.

Papadakis, E. (1992) `Public opinion, public policy and the welfare state', Political Studies, 40, 21-37.

Saunders, P. (1995) Social Security Contributions: An Australian Perspective, Australian Tax Research Foundation Research Study No. 24, Sydney, ATRF.

Smith, R. and Wearing, M. (1987) `Do Australians want the welfare state?', Politics, 22 (2), 55-65.

Stilwell, F. (1994) `Working Nation: from Green to White Paper', Journal of Australian Political Economy, 33, 110-23.

Taylor-Gooby, P. (1995) `Who wants the welfare state? Support for state welfare in European countries?' In S. Svalfors (ed) In the Eye of the Beholder: Opinion, Welfare and Justice in Comparative Perspective, Umea, Scandbooks.

Titmuss, R. (1974) Social Policy: An Introduction, London, George Allen and Unwin.

Van Moorst, H. (1983) Problems of Being Unemployed, Special Report No. 3, Urban and Social Research Centre, Melbourne, Footscray Institute of Technology.

Van Moorst, H. (1984) From Dole Bludging to Dole Pathology -- The Transformation of an Ideology, Special Report No. 5, Urban and Social Research Centre, Footscray Institute of Technology.

Weatherley, R. (1993) `Doing the right thing: how social security claimants view compliance', Australian and New Zealand Journal of Sociology, 29 (1), 21-39.

Welch, S. (1985) `The "more for less" paradox: public attitudes on taxing and spending', Public Opinion Quarterly, 49 (3), 310-16.

Windschuttle, K. (1980) Unemployment: A Social and Political Analysis of the Economic Crisis in Australia, (revised edition), Ringwood, Penguin.

Zaller, J. (1992) The Nature and Origins of Mass Opinion, Cambridge, Cambridge University Press.

Tony Eardley is Senior Research Fellow and George Matheson Senior Research Officer at the Social Policy Research Centre, University of New South Wales.