AUS T R AL IAN S OCIAL AT T I T UDE S
Australian Social Attitudes: The First Report offers a new level of insight into the dynamics of Australian social life. Its 14 chapters develop an in-depth and accessible understanding of how Australia is responding to the new realities of work, globalisation, family and community life. The book draws on the latest social research and analysis from many of Australia’s leading social scientists to challenge conventional wisdom about Australia, and assesses the impact of the Howard government on the attitudes and behaviour of Australians of all ages and social backgrounds. Australian Social Attitudes: • is the fi rst major report on what Australians think about topics as diverse as the use of human genetic information, what makes a family, the costs and benefits of immigration, and crime and punishment • combines analysis of public opinion and social attitudes with reported behaviour and policy developments • compares Australian attitudes to those in other countries. Australian Social Attitudes will prove an excellent resource for students, teachers, researchers and policy makers, and anyone interested in understanding the social dynamics of contemporary Australia.
AUS TR ALIAN SOCIAL AT TIT UDES THE FIRST R E P O R T
edited by WILSON
UNSW PRESS ISBN 0-86840-671-6
MEAGHER GIBSON DENEMARK WESTERN
� 9 780868 406718 �
ASAcover2.indd
1
Shaun WILSON Gabrielle MEAGHER Rachel GIBSON David DENEMARK Mark WESTERN EDITED BY
UNSW PRESS
22/7/05
11:56:42 AM
AusSocAtts03
26/7/05
8:55 AM
Page i
AUSTRALIAN S OC I A L AT T I T U DE S
AusSocAtts03
26/7/05
8:55 AM
Page iii
AUSTRALIAN SOCIAL ATTITUDES The First Report
Edited by Shaun Wilson, Gabrielle Meagher, Rachel Gibson, David Denemark and Mark Western
UNSW PRESS
AusSocAtts03
26/7/05
8:55 AM
Page iv
A UNSW Press book Published by University of New South Wales Press Ltd University of New South Wales Sydney NSW 2052 AUSTRALIA www.unswpress.com.au © The Australian National University 2005 First published 2005 This book is copyright. Apart from any fair dealing for the purpose of private study, research, criticism or review, as permitted under the Copyright Act, no part may be reproduced by any process without written permission. While copyright of the work as a whole is vested in The Australian National University, copyright of individual chapters is retained by the chapter authors. Inquiries should be addressed to the publisher. National Library of Australia Cataloguing-in-Publication entry Australian social attitudes: the first report. Includes index. ISBN 0 86840 671 6. 1. Public opinion - Australia. 2. Australia - Social conditions - Public opinion. 3. Australia - Economic conditions - Public opinion. 4. Australia - Politics and government - Public opinion. I. Wilson, Shaun, 1971– . 303.380994 Layout Stan Lamond Cover art Di Quick Print Ligare
Australian Social Attitudes: The First Report has been produced with the support of the Academy of the Social Sciences in Australia.
AusSocAtts03
26/7/05
8:55 AM
Page v
CONTENTS FOREWORD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
vii
ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ix
CONTRIBUTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
x
CHAPTER 1
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Rachel Gibson, Shaun Wilson, Gabrielle Meagher, David Denemark and Mark Western
CHAPTER 2
WHAT MAKES AN AUSTRALIAN FAMILY? . . . . . . . . . . . . . . . . . . . . 12 Ann Evans and Edith Gray
CHAPTER 3
MAKING FAMILIES WORK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Deborah Mitchell
CHAPTER 4
HOW DO AUSTRALIANS FEEL ABOUT THEIR WORK? . . . . . . . . . . . 42 Bill Martin and Jocelyn Pixley
CHAPTER 5
VOLUNTARY ASSOCIATIONS AND POLITICAL PARTICIPATION . . . . 62 Andrew Passey and Mark Lyons
CHAPTER 6
ARE POSTMATERIALISTS ENGAGED CITIZENS? . . . . . . . . . . . . . . . 82 Mark Western and Bruce Tranter
CHAPTER 7
WHERE TO FOR THE WELFARE STATE? . . . . . . . . . . . . . . . . . . . . . 101 Shaun Wilson, Gabrielle Meagher and Trevor Breusch
AusSocAtts03
26/7/05
VI
8:55 AM
Page vi
• CONTENTS
CHAPTER 8
IS THERE A CRISIS OF TRUST IN AUSTRALIA? . . . . . . . . . . . . . . . 122 Clive Bean
CHAPTER 9
PERCEPTIONS OF CRIME AND JUSTICE . . . . . . . . . . . . . . . . . . . . 141 David Indermaur and Lynne Roberts
CHAPTER 10
HAVE AUSTRALIANS EMBRACED ECONOMIC REFORM? . . . . . . . . 161 Michael Pusey and Nick Turnbull
CHAPTER 11
IMMIGRATION, MULTICULTURALISM AND NATIONAL IDENTITY . .182 Murray Goot and Ian Watson
CHAPTER 12
KNOWING YOUR GENES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 Kristine Barlow-Stewart, Sandra Taylor and Margaret Otlowski
CHAPTER 13
MASS MEDIA AND MEDIA POWER IN AUSTRALIA . . . . . . . . . . . . 220 David Denemark
CHAPTER 14
ARE AUSTRALIANS OPEN TO GLOBALISATION? . . . . . . . . . . . . . . 240 Ian Marsh, Gabrielle Meagher and Shaun Wilson
APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
258
INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
272
AusSocAtts03
26/7/05
8:55 AM
Page vii
FOREWORD More than at any time in the past, the design of effective public policy requires the assistance of accurate, informed social surveys. What aspects of social policy concern people most? How do people view the reform of the health services? What level of immigration do most people regard as appropriate for Australian society? These and a host of other questions can only be properly addressed if we know what people feel about these issues and how their views are formed and have changed over time. Even when public policy has consistently been at odds with popular opinion – as in the case of capital punishment, for example – knowing what people feel about an issue provides an important backdrop to how political elites explain their policy decisions to the public. Yet Australia, despite a long history of innovative policy-making, has lagged behind other advanced societies in developing these critical empirical tools. In the United States, the General Social Survey is in its fourth decade of operation and provides an important adjunct to national policy, on everything from drug abuse to education policy. In Britain, a Social Attitudes Survey has been conducted since the 1980s, and the annual publication based on it is widely used by government to monitor public opinion. The Australian Survey of Social Attitudes (AuSSA), inaugurated in 2003, fills this important gap in our knowledge of what ordinary Australians feel about the major social, economic and political issues of the day. As policy problems and solutions have become more transnational through the inexorable spread of globalisation, we have seen a proliferation in comparative surveys. In Europe, the Eurobarometer surveys and, more recently, the European Social Survey, have monitored public opinion in the European Union states, as well as among aspiring members in Eastern Europe, while in Africa there is the Afrobarometer and in Asia, the East Asian Social Survey. In politics, the unprecedented wave of democratisation is tracked by the 50-nation Comparative Study of Electoral Systems project. In this comparative dimension, AuSSA again makes an important contribution, by placing Australia within this international context, notably as the Australian member of the International Social Survey Program and the World Values Survey. AuSSA is the brainchild of a new generation of enterprising social scientists, drawn mainly from political science and sociology. They come
AusSocAtts03
26/7/05
VIII
•
8:55 AM
Page viii
FOREWORD
from across Australia, with co-ordination from the Research School of Social Sciences at the Australian National University. Their dedication and foresight has enabled this ambitious project to reach fruition. In addition to providing important insights into what ordinary Australians feel about major issues, they have rejected the view that Australia is sui generis and instead placed it within a rich tradition of comparative social research. AuSSA’s success will depend on the extent to which it comes to influence the policy discourse. Based on the quality of the contributions to this first volume, the omens look decidedly good. Professor Ian McAllister Political Science Research School of Social Sciences The Australian National University
AusSocAtts03
26/7/05
8:55 AM
Page ix
ACKNOWLEDGMENTS Australian Social Attitudes: The First Report is the product of extensive collaboration. The editors and contributors would like to acknowledge the role of the ACSPRI Centre for Social Research (ACSR) in the Research School of Social Sciences (RSSS) at the Australian National University in establishing the Australian Survey of Social Attitudes. AuSSA is a significant new resource for Australian social sciences, and the primary resource for this report. In particular, we would like to thank Deborah Mitchell, Director of the ACSR, and Professor Ian McAllister, Director of the RSSS 1997–2004, for their commitment to AuSSA. We would also like to thank the AuSSA Advisory Team, whose experience and inspiration enabled us to develop this new questionnaire. In addition to the editors, the national team included Clive Bean, David Charnock, David de Vaus, Murray Goot, David Gow, Bill Martin, Ian McAllister, Deborah Mitchell, Jocelyn Pixley, Michael Pusey, Heather Strang and Bruce Tranter. International advisers included Roger Jowell, Hermann Schmitt, Stefan Svallfors, Yang-Chih Fu and Pippa Norris. The Australian Social Science Data Archive fielded the survey and delivered the AuSSA 2003 data with promptness and professionalism; for this Sophie Holloway, Rachelle Graham, Andrew Grubb and Stephen Gray deserve special thanks. We are grateful to the Academy of the Social Sciences in Australia, which generously sponsored a workshop for contributors at the Australian National University in June 2004. ACSR Senior Research Fellow Trevor Breusch provided expert advice on the more recondite points of statistical method. Peter McCarthy helped the editors polish the report’s prose. We also thank Brigid van Wanrooy for data assistance. Christel Cools and Louise Sims ensured that AuSSA meetings and workshops ran smoothly. Finally, we thank Phillipa McGuinness and John Elliot of UNSW Press for their confidence in this project.
AusSocAtts03
26/7/05
8:55 AM
Page x
CONTRIBUTORS Dr Kristine Barlow-Stewart is Director of the Centre for Genetics Education at Royal North Shore Hospital, Sydney. Dr Clive Bean is Professor in the School of Humanities and Human Services at the Queensland University of Technology. Dr Trevor Breusch is Senior Research Fellow in the ACSPRI Centre for Social Research in the Research School of Social Sciences at the Australian National University. Dr David Denemark is Associate Professor in Political Science and International Relations in the School of Social and Cultural Studies at the University of Western Australia. Dr Ann Evans is Research Fellow in the ACSPRI Centre for Social Research in the Research School of Social Sciences at the Australian National University. Dr Rachel Gibson is Deputy Director of the ACSPRI Centre for Social Research in the Research School of Social Sciences at the Australian National University. Murray Goot is Professor in the Department of Politics and International Relations at Macquarie University. Dr Edith Gray is Research Fellow in the Demography and Sociology Program in the Research School of Social Sciences at the Australian National University. Dr David Indermaur is Senior Research Fellow in the Crime Research Centre at the University of Western Australia. Mark Lyons is Adjunct Professor in the Australian Centre for Co-operative Research and Development at the University of Technology, Sydney. Dr Ian Marsh is ANZSOG Professor of Government at the University of Sydney. Dr Bill Martin is Associate Professor in the Department of Sociology at Flinders University.
AusSocAtts03
26/7/05
8:55 AM
Page xi
CONTRIBUTORS
•
XI
Dr Gabrielle Meagher is Senior Lecturer in Political Economy in the School of Economics and Political Science at the University of Sydney. Dr Deborah Mitchell is Director of the ACSPRI Centre for Social Research in the Research School of Social Sciences at the Australian National University. Professor Margaret Otlowski is in the School of Law at the University of Tasmania. Dr Andrew Passey is Senior Research Fellow in the Australian Centre for Co-operative Research and Development at the University of Technology, Sydney. Dr Jocelyn Pixley is Senior Lecturer in the School of Sociology and Anthropology at the University of New South Wales. Professor Michael Pusey is Visiting Fellow in the School of Sociology and Anthropology at the University of New South Wales. Dr Lynne Roberts is Research Fellow in the Crime Research Centre at the University of Western Australia. Dr Sandra Taylor is Senior Lecturer in the School of Social Work and Applied Human Sciences at the University of Queensland. Dr Bruce Tranter is Senior Lecturer in the School of Political Science and International Studies at the University of Queensland. Dr Nick Turnbull is Lecturer in Government and International Relations in the School of Economics and Political Science at the University of Sydney. Dr Ian Watson is Senior Researcher in the Australian Centre for Industrial Relations Research and Training at the University of Sydney. Dr Mark Western is Associate Professor in the School of Social Science at the University of Queensland and Director of the University of Queensland Social Research Centre. Dr Shaun Wilson is Research Fellow in the ACSPRI Centre for Social Research in the Research School of Social Sciences at the Australian National University.
AusSocAtts03
26/7/05
8:55 AM
Page xii
AusSocAtts03
26/7/05
8:55 AM
Page 1
CHAPTER 1
INTRODUCTION R a c h e l G i b s o n , S h a u n Wi l s o n , G a b r i e l l e M e a g h e r, D a v i d D e n e m a r k a n d M a r k We s t e r n
BRINGING TOGETHER SOCIAL RESEARCH AND PUBLIC POLICY Australians and their policy-makers demand sophisticated insights into social attitudes and behaviour and how these shape work, families, politics and Australia’s links to a larger world. Living in a dynamic and educated society, Australians increasingly reflect on the state of their institutions and their different ways of living. But this kind of self-reflection depends not only on values, but also on facts, which are often not available when they are needed most. Evidence – including feedback from the public, systematic social analysis, legal opinion, expert judgment and overseas experiences – is increasingly part of policy-making, as Meredith Edwards (2004) recognises in her exploration of the future policy role of Australian social scientists. And as German sociologist Claus Offe points out, many policy problems can no longer be ‘solved’ from above without a diverse public’s input and co-operation (Offe 1996, pp. 111–19). As governments are obliged to deal with active and informed citizens, considered social research will increasingly be important to make policy work. Australian Social Attitudes: The First Report takes up the challenge of providing high-quality social research to the Australian community and to policy-makers, research that informs both public debate and sound decision-making. The empirical social research found in this book also contributes to the development of the social sciences themselves. As well-known sociologist Gøsta Esping-Andersen (2000, pp. 72–5) makes clear, the future for the social sciences is to engage deeply with our changing community, and to answer the many hundreds of questions about
AusSocAtts03
26/7/05
2 •
8:55 AM
Page 2
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
contemporary society that interest the public at large. Reporting his findings to the Gulbenkian Commission on the Restructuring of the Social Sciences, another prominent social scientist, Immanuel Wallerstein, also affirmed the need for social sciences to engage in empirical research that combines researchers from different backgrounds to develop new methods and to construct research problems in innovative ways (Gulbenkian Commission 1996). In addressing the twin challenges to contribute to public debate and to the social sciences, Australian Social Attitudes is positioned between public opinion research, social commentary and ‘pure’ academic research. Public opinion polling is able to read the public pulse, but rarely tries to understand how social attitudes and social behaviour change over time, or to tell us why different constituencies find themselves in conflict. Social commentary reflects back to us images of Australian society that can prick our collective conscience and stir our emotions. But, without the support of detailed evidence, commentary risks repeating conventional wisdom or perpetuating myths. At the same time, ‘pure’ academic research can tell us about a subject in great detail, but often fails to speak to interested citizens beyond the walls of the university. This report brings together a widely shared interest in addressing topics at the top of the social agenda with a commitment to the social sciences. Our main tool for undertaking this report is the Australian Survey of Social Attitudes (AuSSA), a statistically representative national survey of Australians, first conducted by the Centre for Social Research at the Australian National University in late 2003 (Gibson et al. 2004). Drawing on a systematic analysis of the survey data, we track the attitudes and activities of a large and diverse sample of Australians – more than most of us could hope to meet and know well in a lifetime. We do this aware of the limitations of survey research. Conducting surveys is not the same as engaging in face-to-face communication, which gives us a more complete, nuanced picture of how people think and act. This report does not intend to have the ‘final say’, but seeks instead to add to the resources of the national conversation.
AUSTRALIA IN THE HOWARD YEARS As well as a reflection of Australia now, Australian Social Attitudes: The First Report is also a wide-ranging assessment of how social attitudes have changed during the Howard government, now in its tenth year. Many commentators feel that the country has become more conservative. They argue that this new conservatism is evident, for example, in the rejection
AusSocAtts03
26/7/05
8:55 AM
Page 3
INTRODUCTION
• 3
of a republic in 1997, in hostility to immigration and welfare, and in narrower ideals of family and the priority of affluence (see, for example, Manne 2004). Other commentators fear a decline of trust in others and in institutions, and falling participation in voluntary life as an affluent Australia ‘cocoons’ itself at home (see, for example, Cox 2002). The findings we present in this report do not entirely confirm this picture. Using data from AuSSA 2003 and from Australian Election Studies and other major surveys, this report reveals instead the nuances in public opinion – and some surprising patterns in Australian social attitudes. Below we give readers a sense of this complexity. Deborah Mitchell confirms in chapter 3 that family does remain central to Australian identity – three-quarters of all respondents to AuSSA 2003 list their family as one of the three groups most important to who they are. However, Ann Evans and Edith Gray show in chapter 2 that our understanding of what relationships make a family is quite broadly defined, especially among women and younger generations. Among these Australians, a majority believe that gay and lesbian couples with children, and single-parent households, are indeed families. And despite Health Minister Tony Abbott’s attempts to reopen the abortion debate after Howard’s election victory in October 2004, Evans and Gray find that Australians overwhelmingly support women’s right to choose. On welfare, Shaun Wilson, Gabrielle Meagher and Trevor Breusch also make surprising findings in chapter 7. Australians are less preoccupied with lower taxes than at any time in the last quarter century, with more now preferring higher social spending. A large majority of Australians are even prepared for taxes to rise to support more spending on health, Medicare and education. Still, the public appears to support the Howard government’s emphasis on increasing obligations on welfare recipients to find work. But this does not add up to support for cutting welfare, which is rejected by the majority. We find similarly complex patterns in attitudes to immigration, multiculturalism and national identity. In chapter 11, Murray Goot and Ian Watson report that a declining number of Australians support lower immigration: two-thirds wanted this in 1996, compared to less than two-fifths now. Attitudes to the economic impacts of migrants have also become more positive, as the economy has boomed through the 1990s and early 2000s. While attitudes to multiculturalism remain complex, there is little sign that there has been any great shift in public opinion on this question. Goot and Watson do find, however, that, compared to the
AusSocAtts03
26/7/05
4 •
8:55 AM
Page 4
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
mid-1990s, fewer Australians now are ashamed of things Australia has done. Though Australia’s long period of economic growth has sustained a new prosperity and affluence, some fear that the costs of this growth are higher debt, longer working hours and a degraded community and environment (see, for example, Hamilton 2003). Are Australians feeling prosperous? Certainly, a majority of respondents indicate that, with the ways things are at the moment, they feel that they have a good chance of improving their standard of living. But most telling is Goot and Watson’s finding that Australians are much more confident in the Australian economy than they were in the mid-1990s, with some 80 per cent now proud of Australia’s economic achievements (up 32 per cent since 1995). How this economic prosperity has assisted the Howard government is a story that will be revisited by many after the government’s easy win in 2004. One benefit of economic prosperity is relatively low job insecurity. As Bill Martin and Jocelyn Pixley show in chapter 4, only 13 per cent of AuSSA 2003 respondents feel insecure about their employment. But the costs of prosperity for work are also registered. Most working Australians see long working hours as a social problem, and many recognise the effects of long hours on personal and family life. A picture emerges, then, of low job insecurity, a desire for more spending on key public services and more confidence about Australia’s economic performance. But how are economic and social achievements anchored in social and community life? We have already mentioned that younger Australians and women have a broader idea of family and family life. But are we less trusting and more isolated in our communities? In chapter 8, Clive Bean shows that, on the figures, Australians are not losing trust in each other, but that their trust in some institutions – including major Australian companies and churches – has declined since the mid-1990s. The armed forces, active in campaigns in the region from East Timor in 1999 to the Asian tsunami operations of 2005, remain the institution in which Australians have most confidence, a finding mirrored in many other countries. When it comes to participation in civil society, Andrew Passey and Mark Lyons, in chapter 5, find that Australia is a nation of joiners: 86 per cent of AuSSA 2003 respondents belong to at least one voluntary association, and more than a quarter of members are officeholders or active members. Participation in voluntary associations is strongly linked to the political participation that is the lifeblood of democracy. Members
AusSocAtts03
26/7/05
8:55 AM
Page 5
INTRODUCTION
• 5
of these associations are nearly three times more likely than nonmembers to undertake political activity, such as contacting a politician or taking part in a protest. Participation also seems to be closely linked to our values. Although material happiness has featured a great deal in Australian life in recent years, many Australians continue to support environmental protection and other ‘quality of life’ or ‘postmaterial’ values. As Mark Western and Bruce Tranter show in chapter 6, Australians who hold postmaterial values are eager participants in environmental and social causes. No doubt, how we view the world is shaped by mass media, and, as David Denemark finds in chapter 13, television is our primary source of news, information and entertainment at home. But although we are absorbed by the electronic media, we are not uncritical of it. Most Australians think that television violence contributes to social violence and that media ownership is far too concentrated among a few rich families. The media may be playing some role in the contradictory reality that David Indermaur and Lynne Roberts find in chapter 9 when it comes to crime. They show that crime rates in most categories are actually falling in Australia, but perceptions of crime are yet to register this. As their international evidence suggests, the media (and politicians) have played a part in creating a climate of fear about crime in which public misperceptions thrive. And what of the future? Do we want more engagement in Asia, as former Prime Minister Paul Keating urged? Do we want further economic reform? And how are Australians responding to the potential benefits and threats of new biotechnologies? Australian engagement in the Asian region is bound to become increasingly important in the coming years. Yet, as Ian Marsh, Gabrielle Meagher and Shaun Wilson show in chapter 14, Australia is not particularly exposed to international trade compared to other small advanced democracies – and this is reflected in our comparatively protectionist attitudes to trade, local film content and the role of international organisations in national affairs. Australians want closer economic relations with Asia, but are less enthusiastic about closer political and cultural ties; few report feeling close to Asia. Many Australians also believe that globalisation is increasing insecurity at home. Marsh, Meagher and Wilson suggest that this might explain why attitudes to Australia’s future global engagement have been slow to change. The Howard government stunned political pundits when it managed to secure the first Senate majority in a generation after the 2004
AusSocAtts03
26/7/05
6 •
8:55 AM
Page 6
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
elections, which means it can pass legislation without the support of other parliamentary parties. Do Australians support further privatisation and freeing up of the labour market as the government intends? According to Michael Pusey and Nick Turnbull in chapter 10, Australians still want government ownership of key services like Telstra and strongly support award wages as the foundation of industrial relations. Pusey and Turnbull find that Australians are not ‘neo-liberals’, but rather economic pragmatists, continuing to envisage an active role for the government in utilities ownership, the economy and service delivery. Finally, Australians are enthusiastic about the potential freedoms to be gained from applying human genetic tests. But when pressed on the use of tests in insurance and employment, they are strongly opposed. Low trust appears to be a factor here. As Kristine Barlow-Stewart, Sandra Taylor and Margaret Otlowski argue in chapter 12, Australians will need to be convinced of the merits of extending these technologies to new areas of social and economic life.
ABOUT AuSSA 2003 The research in this report is mainly drawn from the first Australian Survey of Social Attitudes, completed in late 2003. AuSSA is a product of the Centre for Social Research at the Australian National University in collaboration with researchers from other Australian and overseas universities. AuSSA is also the official Australian contribution to the world’s two leading social surveys: the International Social Survey Program (ISSP), covering 39 countries, and the World Values Survey, covering around 90 countries. AuSSA 2003 included the ISSP’s 2003 National Identity module. AuSSA 2005 will field the next World Values Survey and the ISSP’s Citizenship and Work Orientations module. AuSSA takes the form of a mail questionnaire sent to more than 10 000 Australian citizens every two years. The survey itself was prepared by five Principal Investigators – Shaun Wilson, Rachel Gibson, Gabrielle Meagher, David Denemark and Mark Western – in co-operation with the AuSSA Advisory Panel, which met to draft the survey in early 2003. Survey questions were derived from national and international sources and the inspiration of AuSSA advisers.1 In developing the survey, the aim was to ensure not only that AuSSA questions provide comparability over time and with other surveys cross-nationally, but also to take up new topics and to contribute to ongoing innovation in social survey design. AuSSA 2003 was distributed to a stratified systematic random sample of 11 380 Australians aged 18 or over, who were selected from a
AusSocAtts03
26/7/05
8:55 AM
Page 7
INTRODUCTION
• 7
2002 updated version of the Australian Electoral Commission’s Electoral Roll.2 The sample reflects the state-by-state distribution of the Australian population. The overall net response rate for AuSSA 2003 was 44 per cent, with 4270 valid responses collected. Two versions of the survey were distributed between August and December 2003, providing samples with approximately the same number of respondents for each. Surveys A and B had a common core of about 130 items, and then included separate modules on bushrangers, genetic awareness and testing, share ownership, national identity (for the ISSP) and group membership in Australia. To view the AuSSA 2003 questionnaires, see the AuSSA website at:
The Australian Social Science Data Archive (ASSDA) conducted the fieldwork. Survey results are available publicly through ASSDA’s online data analysis system, NESSTAR at: or by contacting ASSDA at [email protected], which can also supply the dataset in machine-readable formats as well as provide the AuSSA 2003 User’s Guide (Study no. 1070).
HOW TO USE THIS BOOK This book combines two main types of evidence. The first type includes statistics about the social world (for example, government social expenditures, immigration statistics, trade data) and policy statements (for example, laws on genetic discrimination and insurance, laws on marriage, trade agreements with other countries, income-support regulations). The second type of evidence is data collected from individuals in surveys like AuSSA and the Australian Election Study (AES), which ask questions about social attitudes (for example, about family, work, politics), behaviour (for example, membership of organisations, media consumption) and demographics (for example, respondents’ education, place of residence, family type). Most of this data is presented in tabular form, and is easy to follow. In detailing survey questions and presenting data, we have aimed for simplicity and accessibility. We have been pragmatic in what we have presented, aiming to communicate the story to the reader without unnecessary detail. Chapters generally mention the full wording of survey questions except where they are obvious from the surrounding text. And, in tables and in data reporting, response categories like ‘strongly agree’ and ‘agree’ are regularly collapsed to ‘agree’. Tables do not report the numbers of
AusSocAtts03
26/7/05
8 •
8:55 AM
Page 8
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
responses in every cell; instead, tables include the frequency of responses where it matters for understanding the sample properties, and for making inferences. Some data is presented in more complex forms, mainly as regression models, which are included in the appendices at the back of the book. Readers will find that most chapters analyse the relationship between patterns in social attitudes and real-world developments, such as political behaviour (for example, voting patterns, protests), public policy change (for example, family payments, labour market deregulation, trade liberalisation) and economic and social trends (for example, economic growth, abortion rates, crime rates). Contributors do not try to ‘prove’ that social attitudes are decisive in producing public policy, but rather point out connections, tensions and feedback between the two.3 People interested in surveys often ask about the reliability of their findings. This is an important topic for both interested users of social science data and survey researchers, which we cannot elaborate fully here. However, we will briefly touch on three problems: survey method, sample bias and non-responses. All survey methods have benefits and limits. In the past, mail surveys have been criticised for not getting a good spread of responses across the community, and, as David de Vaus mentions, for obtaining lower response rates than face-to-face or telephone interviews (de Vaus 2002, p. 127). But mail surveys are making something of a comeback. The costs of interviewing large samples face-to-face in a country the size of Australia are increasingly prohibitive. At the same time, telephone surveys are having increasing problems with obtaining representative samples, as fragmentation in the telecommunications market makes compiling a comprehensive sampling frame difficult. Increasingly, too, people screen out social researchers along with market researchers. Using mail surveys, ASSDA consistently achieves net response rates in the 40–60 per cent range with the AES and now AuSSA, which compare well internationally. No matter what method is used, patterns of bias in social surveys are surprisingly consistent. The major known biases in the AuSSA sample are: age (the median age is older than the median age of the Census population), education (an over-representation of persons with post-secondary school qualifications), and gender (a slight overrepresentation of women). We provide some basic information about these biases in AuSSA 2003 by comparing its sample demographics with equivalent statistics from the Australian Bureau of Statistics (ABS)
AusSocAtts03
26/7/05
8:55 AM
Page 9
INTRODUCTION
• 9
(see table 1.1). Because the sampling frame of AuSSA 2003 includes registered voters, who are overwhelmingly Australian citizens, we have compiled the comparative ABS data for Australian citizens only. We hope that in the future AuSSA will also survey the 8 per cent or so of adult Australians who are not registered voters; that is, Australians who are not currently citizens. One way we can compensate for sample bias is by using weights that better align the characteristics of the sample with a population about which we have accurate information. This method is commonly used, for instance, in political opinion polling and in election forecasts. Although a sample weight is included in the AuSSA 2003 datafile (available from ASSDA), contributions to this book rely on unweighted data for a couple of reasons. Sample biases do not affect most of the headline frequencies used in this book enough to justify using weighted data. Researchers often use subsamples as one avenue for dealing with biases in an overall sample. More often than not, social researchers are interested in subgroups of the population (for example, blue-collar workers or Liberal voters) and their attitudes and behaviour, and perhaps how they compare to other subgroups. The contrast between the attitudes of subgroups can give accurate and useful clues about the likely social causes of differences. Weighted data also poses additional problems for statistical modelling, such as some kinds of regression analysis.
Table 1.1 Comparing AuSSA 2003 sample with Census 2001* AuSSA 2003
Census 2001
Sex Males
47.0
48.8
Females
53.0
51.2
20–34
17.0
29.9
35–49
30.5
31.6
50–64
30.6
22.7
65 and over
21.9
15.8
Year 12 and below
43.7
59.2
Trade qualification or certificate/diploma
34.0
25.7
22.3
15.1
Age
Education
Bachelor’s degree or higher n
(4270)
* Comparison is made for Australian citizens aged 20 years and over. SOURCES The Australian Survey of Social Attitudes 2003; Australian Bureau of Statistics Census 2001
AusSocAtts03
26/7/05
10 •
8:55 AM
Page 10
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
For some questions, survey researchers are confronted with a high level of non-response, which can add to our difficulties in understanding the true state of public opinion. This can be registered in different ways: respondents do not answer a given question, choose the ‘Can’t choose’ option, or choose ‘Neither agree nor disagree’. Some questions attract a higher level of non-response than others, and there are essentially two reasons for this. The first reason is the difficulty of the question (see Berinsky 2004). For instance, a question about the likely effects of a free trade agreement between Thailand and Australia might require a lot of knowledge of current affairs and economics. The second reason involves what Adam Berinsky (2004) calls ‘social complexity’, which obliges the respondent to give a potentially socially undesirable answer (on subjects such as racism and sexuality). Sometimes respondents will avoid revealing their true beliefs. Our strategies for dealing with non-responses are understandably limited. AuSSA tries to ask respondents questions in the simplest way possible, to maximise respondent participation. Additionally, the anonymous context of filling out a mail questionnaire no doubt encourages respondents to answer controversial questions. These are some of the problems confronted by social survey researchers and the ways that they can be accommodated. Acknowledging a few of these may help readers to use this book (and this series) to its fullest potential. We trust our readers will find the insights gathered in the first Australian Social Attitudes as fascinating as we have. NOTES
1
2
3
In particular, we drew on items included in British Social Attitudes (commenced 1983), the General Social Survey for the United States (commenced 1972), modules for ISSP (commenced 1984), the European Social Survey (commenced 2003), Australian Election Studies (commenced 1987) and the National Social Science Survey (commenced 1984). The Electoral Roll excludes permanent and temporary residents of Australia who are not citizens and any other adults in Australia who are non-citizens, except for British subjects on a Commonwealth electoral roll as at 25 January 1984. These exclusions amount to approximately 8 per cent of the adult population. The relationship between public opinion and public policy is comprehensively discussed in Manza, Cook and Page (2002).
REFERENCES
Berinsky, AJ (2004) Silent Voices: Public Opinion and Political Participation in America, Princeton University Press, Princeton, New Jersey.
AusSocAtts03
26/7/05
8:55 AM
Page 11
INTRODUCTION
• 11
Cox, E (2002) ‘Australia: Making the Lucky Country’, in R Putnam (ed.) Democracies in Flux: The Evolution of Social Capital in Contemporary Society, Oxford University Press, Oxford, pp. 333–58. de Vaus, D (2002) Surveys in Social Research, Allen & Unwin, Sydney. Edwards, M (2004) ‘Social science research and public policy: Narrowing the divide’, Occasional Policy Paper no. 2, Academy of the Social Sciences in Australia, Canberra. Esping-Andersen, G (2000) ‘Two societies, one sociology, and no theory’, British Journal of Sociology, 51(1): 59–77. Gibson, R, Wilson, S, Meagher, G, Denemark, D & Western, M (2004) Australian Survey of Social Attitudes 2003, Data file, Australian Social Science Data Archive, The Australian National University, Canberra. Gulbenkian Commission (1996) Open the Social Sciences: Report of the Gulbenkian Commission on the Restructuring of the Social Sciences, Stanford University Press, Stanford. Hamilton, C (2003) Growth Fetish, Allen & Unwin, Sydney. Manne, R (ed.) (2004) The Howard Years, Black Inc. Agenda, Melbourne. Manza, J, Cook, FL & Page, BI (eds) (2002) Navigating Public Opinion: Polls, Policy, and the Future of American Democracy, Oxford University Press, New York. Offe, C (1996) Modernity and the State: East, West, Polity Press, Cambridge.
AusSocAtts03
26/7/05
8:55 AM
Page 12
CHAPTER 2
W H AT M A K E S A N A U S T R A L I A N FA M I LY ? Ann Evans and Edith Gray
INTRODUCTION Like other modern industrialised countries, Australia has seen enormous change in family relationships and childbearing over the course of the 20th century. Median age at first marriage has increased to around 28 years, couples cohabit before more than 70 per cent of marriages, 31 per cent of births are exnuptial, and the median age of mothers is 30 (ABS 2003a; ABS 2003b). Additionally, many argue that cohabitation has become an alternative rather than a pathway to marriage (Hantrais & Letablier 1996), with many exnuptial births now occurring within cohabiting relationships. The speed of change has been rapid. Only 20 years ago, the median age at first marriage was about 22 years for women (ABS 2003a), only 14 per cent of all births were exnuptial (ABS 2003b), and only a fifth of marriages were preceded by cohabitation (ABS 1999). Over the same time, divorce has risen dramatically. In 2001, there were 55 300 divorces granted in Australia, the highest number in 20 years and a 34 per cent increase on the number granted in 1981 (ABS 2002). In 2001, 17 per cent of couples had divorced within the first five years of marriage, while a further 26 per cent had divorced in the following five-year period (ABS 2002). In this chapter, we answer two general questions: How much do Australian social attitudes reflect these rapid changes in family formation? How do social and demographic characteristics, such as gender, age, level of education, marital status and religion, shape attitudes about family life? We start by looking at what kinds of relationships Australians think constitute a family today. Given how much family life has changed,
AusSocAtts03
26/7/05
8:55 AM
Page 13
W H AT M A K E S A N A U S T R A L I A N F A M I L Y ?
• 13
do Australians consider some relationships to be more ‘like a family’ than others? We investigate the extent to which Australians view married and cohabiting couples as families, how the presence of children influences their judgment of the relationships that make a family, and attitudes about same-sex relationships. After exploring which relationships are considered to define family in contemporary Australia, we examine attitudes about children, reproductive rights and fatherhood.
DOES MARRIAGE MAKE A FAMILY? The Australian Survey of Social Attitudes 2003 (Gibson et al. 2004) asks respondents about which types of relationships they include in their definition of a family. The leading question is: ‘Generally speaking, which of the following living arrangements would you consider to be a family today?’ Respondents are asked about a married couple without children, a married couple with children, an unmarried couple without children, and an unmarried couple with children. The response options allow us to establish the relative importance of marriage and the presence of children in respondents’ views about what living arrangements constitute a family. Table 2.1 shows that marriage is an important defining criterion of the family in Australia: a majority (63 per cent) of respondents agree that a married couple without children is a family, while only 32 per cent agree that an unmarried couple without children is a family. Women (36 per cent) are more likely than men (29
Table 2.1 Attitudes to marriage and cohabitation by sex and age, 2003, per cent ‘Yes, a family’
Married couple Unmarried couple Married couple Unmarried couple without children without children with children with children
Sex Male
61
29
99
77
Female
64
36
99
82
18–34
66
34
100
92
35–49
61
37
99
89
50–64
65
34
99
80
65 and over
59
23
99
56
63
32
99
79
(1955)
(1951)
(2026)
(1955)
Age
Total n
SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
14 •
8:55 AM
Page 14
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
per cent) to agree that an unmarried couple without children is a family. Agreement also decreases with age: younger people are also more likely to agree that an unmarried couple without children is a family (34 per cent among 18–34 year olds), while respondents aged 65 and over are least likely to agree (23 per cent). Marriage is important, but Australians are even more likely to agree that couples are families if they have children. There is almost universal agreement that a married couple with children is a family, with little variation for sex and age of respondent. However, a majority of respondents also agree that unmarried couples with children are families (79 per cent), although men (77 per cent) and older respondents (56 per cent) are less likely to agree. Significantly, the proportion of respondents agreeing that unmarried couples with children are families is actually higher (79 per cent) than the proportion who agree that a married couple without children is a family (63 per cent). These findings suggest that the presence of children is more important than marital status alone to the way respondents define the family. It seems that only older respondents remain strongly attached to marriage as defining the meaning of family. Using the British Social Attitudes survey, Barlow et al. (2001) found that attitudes to marriage differed by marital status and religious beliefs. But we also know that marital status and religious beliefs both vary with age, so multivariate analysis helps separate out these effects. Such analysis shows that attitudes differ significantly and independently according to the age, gender, marital status, education and religious practice of respondents. Focusing on attitudes to unmarried couples without children, we estimate that 31 per cent of people agree that this living arrangement makes a family. But responses vary widely (see Appendix, table A2.1). Women are more likely to say an unmarried couple without children is a family than are men (35 per cent, compared to 27 per cent). Age also matters, with those aged 65 and over being least likely to say that this type of couple makes a family (25 per cent), while 34 per cent of respondents aged 35–49 years agree. Marital status makes a big difference to attitudes: respondents who are cohabiting themselves are most likely to agree that an unmarried couple makes a family (47 per cent), falling to 30 per cent among married respondents. Education and religion produce significant variations too: respondents with a Bachelor’s degree are more likely to agree (40 per cent), while respondents who attend religious services at least twice or three times a month are much less likely to agree (23 per cent).
AusSocAtts03
26/7/05
8:55 AM
Page 15
W H AT M A K E S A N A U S T R A L I A N F A M I L Y ?
• 15
The same factors – gender, age, education, marital status and religion – shape attitudes to unmarried couples with children, although respondents are much more likely to see them as families (we estimate 84 per cent). But we still find sharp divides: a striking 92 per cent of people aged 18–34 agree that this arrangement makes a family, while only 63 per cent of those 65 and over agree. Similarly, 95 per cent in a cohabiting relationship and 91 per cent with a Bachelor’s degree agree, while only 68 per cent of those regularly attending religious services agree. Clearly, marriage makes a difference to our perceptions about whether living arrangements make a family. Younger, more educated and less religious Australians are less likely to define families using marital status than are older and more religious people. For them, the presence of children becomes relatively more important.
CHILDREN, MARRIAGE AND DIVORCE We have seen that the presence of children is important to defining family. Given this, we examine responses to two statements in AuSSA 2003 that probe norms about marriage and children. The first statement is: ‘People who want children ought to get married’. The second statement is: ‘Children should be the main concern when couples divorce’. There is no strong support for, or opposition to, marriage if people want to have children – just over 50 per cent of AuSSA 2003 respondents agree that people who want children ought to get married, while 20 per cent neither agree nor disagree and one-quarter disagree (see table 2.2). But, again, we find the same variations in support among key groups of respondents that we found in the previous section. Men (56 per cent) are more likely to agree than women (49 per cent) that people who want children should get married. Older respondents are more likely to agree than are the young. Among those 65 and over, 78 per cent agree, while only 34 per cent of those aged 18–34 agree. Religious attendance is also important. Respondents who often attend religious services are more likely to agree that people who want children should marry (78 per cent), compared to 41 per cent of those who never attend religious services. Overall, the large number of respondents who did not agree with this view probably reflects the supportive views of unmarried couples with children that we found earlier. Although only half the respondents to AuSSA 2003 believe that having children demands marriage, almost 85 per cent agree that children should be the main concern in a divorce. And once more, older people and those who attend religious services regularly are more likely
AusSocAtts03
26/7/05
16 •
8:55 AM
Page 16
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 2.2 Attitudes about children and marriage by demographic group, 2003, per cent People who want children ought to get married
Children should be the main concern when couples divorce
Agree
Neither agree nor disagree
Disagree
Agree
Neither agree nor disagree
Disagree
Male
56
21
22
83
9
8
Female
49
22
29
83
8
9
18–34
34
19
46
76
11
12
35–49
37
27
34
80
9
10
Sex**
Age**
50–64
58
23
19
83
9
8
65 and over
78
13
8
91
4
3
Single, never married
37
22
40
76
11
12
Living with partner
20
24
55
73
9
18
Married
57
22
21
85
8
6
Separated/divorced
48
25
26
82
9
8
Widowed
77
13
9
91
4
3
Less than Bachelor’s degree
53
22
25
83
8
8
Bachelor’s degree or higher
45
21
33
82
10
7
Never
41
22
35
79
10
11
Sometimes
50
24
25
84
8
7
Marital status**
Education**
Religious attendance**
Often Total n
78
13
9
88
6
4
52
21
25
83
8
8
(2186)
(902)
(1082)
(3499)
(305)
(330)
NOTE ‘Can’t choose’ is the remainder for each distribution.
* p < 0.05; ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
to agree. Among 18–34 year old respondents, 76 per cent agree that children should be the main concern when people divorce, while 91 per cent of those 65 and over agree. That young people have more liberal attitudes towards what makes a family than older people is not surprising, and is confirmed here. What makes this finding particularly interesting is that young people are more likely than previous generations to have experienced the divorce of their parents, following increasing divorce rates since the 1970s.
AusSocAtts03
26/7/05
8:55 AM
Page 17
W H AT M A K E S A N A U S T R A L I A N F A M I L Y ?
• 17
I S A S I N G L E - PA R E N T H O U S E H O L D A FA M I LY ? We have seen that AuSSA 2003 respondents are much more likely to define couples with children as a family than they are couples without children. So do they consider a single-parent household to be a family? A large majority (74 per cent) do (see table 2.3). This proportion is greater than the percentage who agree that married or cohabiting couples without children make a family, but lower than the percentage who agree that married or cohabiting couples with children are a family (see table 2.1). Again, we see that men are more conservative than women on this question. Thus, men are less likely to recognise a single-parent household as a family (69 per cent, compared to 78 per cent of women), a pattern found in earlier analysis, though the difference is larger. This is not surprising, given that the majority of single parents are women. The fact that two-thirds of divorces in Australia are initiated by women (Wolcott & Hughes 1999) probably explains some of this gender difference too. And again, agreement drops with age: among 18–34 year olds, 92 per cent consider a single-parent household to be a family, compared to 70 per cent of the 50–64 year age group and just 48 per cent of those aged 65 and over. AuSSA 2003 also asks respondents to express their agreement with the statement ‘A single parent can bring up children as well as a couple’.
Table 2.3 Attitudes to single parents by age and sex, 2003, per cent Agree that a single-parent household is a family
Agree that single parents can bring up children as well as couple parents
Male
69
34
Female
78
49
18–34
92
55
35–49
85
45
50–64
70
39
65 and over
48
33
74
42
(1987)
(4205)
Sex**
Age**
Total n
* p < 0.05; ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
18 •
8:55 AM
Page 18
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Overall, 42 per cent of AuSSA 2003 respondents agree with this statement. Table 2.3 shows that men are less likely to agree (34 per cent) than women (49 per cent). Age matters, too. Among 18–34 year olds, 55 per cent agree, falling to 33 per cent of those aged 65 and over. Young women are the group most likely to agree that single parenthood is as good as two-parent arrangements. To predict the percentage of respondents who agree that single parents with children make a family, we modelled the effect of religious attendance, marital status, education, age and sex on respondent attitudes to single-parent families (see Appendix, table A2.1). Consistent with the patterns found so far in this chapter, women, younger respondents and more educated respondents are significantly more likely to see single-parent households as families. Those who agree that a single parent can bring up children as well as a couple are also significantly more likely to see single parents with children as a family (88 per cent), regardless of their age or sex. Interestingly, and perhaps surprisingly, religious attendance by itself does not shift attitudes about single-parent families (81 per cent of those who never attend religious services see single-parent households as families, compared to 79 per cent of those who attend often).
IS A SAME-SEX COUPLE A FAMILY? We have seen clear evidence of generational difference in attitudes to what makes a family: younger people are more likely than older people to see unmarried couples (with or without children) and single-parent households as families. Nevertheless, there remains a strong perception that parents are best to make a family and to raise children. So where do same-sex couples fit? Do Australians see them as families too? Evidence from AuSSA 2003 suggests that some do, but also that opinion is much more divided on same-sex relationships than on other household arrangements. Table 2.4 shows that only 20 per cent of AuSSA 2003 respondents agree that a same-sex couple is a family, but, in line with findings on other relationships, the presence of children makes a big difference: about 42 per cent of respondents agree that same-sex couples with children make a family. Again, women are more likely than men to agree that same-sex couples with children are a family and so are the young (65 per cent of those aged 18–34 agree, compared to just 14 per cent of those aged 65 and over). Multivariate analysis confirms that, in addition to significant differences in the attitudes of men and women, and of different age
AusSocAtts03
26/7/05
8:55 AM
Page 19
W H AT M A K E S A N A U S T R A L I A N F A M I L Y ?
• 19
groups, differences in education and religious attendance also affect attitudes to same-sex couples (see Appendix, table A2.1). Religious differences are particularly stark. The model predicts that only 15 per cent of those who regularly attend religious services agree that a same-sex couple with children is a family (compared to 50 per cent of those who never attend), falling to 8 per cent agreement that a same-sex couple without children is a family (compared to 21 per cent of those who never attend). The question about recognising gay couples as families is part of a larger debate about recognising same-sex relationships. In November 2003, the Supreme Judicial Court of Massachusetts ruled that barring same-sex couples from civil marriage was unconstitutional, sparking renewed debate in the United States and Australia about the virtues of gay marriage. In Australia, the debate has had some high-profile participants. On one side, the Catholic Archbishop of Sydney, George Pell, argues that promotion of gay marriage would ‘weaken significantly the place of traditional marriage and bring with it instability and further confusion’ (Pell 2004). On the other, former Australian Medical Association President, Kerryn Phelps, likens the ban on gay marriage to a form of apartheid (Symons 2004). In April 2004, Prime Minister John Howard proposed an amendment to the 1961 Marriage Act to include what he called a ‘commonly accepted definition of a marriage’ as ‘the union of a man and a woman to the exclusion of all others, voluntarily entered into for life’ (Howard 2004a). The amendment passed the Senate with bipartisan support in August 2004. At the same time, Mr Table 2.4 Attitudes to same-sex couples by age and sex, 2003, per cent Agree that same-sex couple without children is a family
Agree that same-sex couple with children is a family
Agree that the law should recognise same-sex couples
Male
14
34
28
Female
24
50
40
18–34
25
65
49
35–49
26
56
41
50–64
18
36
30
8
14
19
Sex**
Age**
65 and over Total n
20
42
34
(1986)
(1989)
(4207)
* p < 0.05; ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
20 •
8:55 AM
Page 20
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Howard announced changes to superannuation laws so that people who are ‘financially interdependent’, in effect including gay couples, would have the same access to superannuation as married or cohabiting couples (Howard 2004a). So how much support is there for gay marriage? AuSSA 2003 asks respondents a more general question about whether the law should recognise same-sex relationships (see table 2.4). Just over one-third of respondents (34 per cent) agree, while 41 per cent disagree. Again, women and young people are more likely to agree. However, it is interesting to note that the percentage of people aged 65 and over who agree that the law should recognise same-sex relationships is larger than the percentage of the same age group who agree that a same-sex couple with children are a family. A recent Newspoll found similar results when asking specifically about gay marriage. Overall, 38 per cent were in favour of gay marriage, with women and younger respondents more likely to favour them (Newspoll 2004). We develop a model for predicting support for legally recognising same-sex couples (see Appendix, table A2.2). Being a woman, younger and holding a degree remain significant factors in predicting support for legal recognition for same-sex couples. But we also find that respondent sexual identification matters. Heterosexuals are less likely to agree that the law should recognise same-sex relationships than non-heterosexuals (30 per cent, compared to 51 per cent). Frequent attendees of religious services are least likely to agree (15 per cent, compared to 31 per cent who attend sometimes and 42 per cent who never attend). Marital status is also significant: single respondents who have never married and cohabiters are more likely to agree than married, divorced or widowed respondents.
LIFE WITHOUT CHILDREN So far we have suggested that the presence of children is relatively more important than marital status in defining a family. Does this mean Australians devalue life without children? AuSSA 2003 respondents are asked: ‘Thinking about relationships and children today, please tell us how much you agree or disagree with the statement: A life without children is not fully complete’. Respondents do not support this statement as strongly as we might have expected, given the importance of children to defining families: around a quarter of people say that they neither agree nor disagree, 40 per cent agree and 35 per cent disagree. Interestingly, men are more likely to agree (45 per cent) than women
AusSocAtts03
26/7/05
8:55 AM
Page 21
W H AT M A K E S A N A U S T R A L I A N F A M I L Y ?
• 21
(34 per cent) and younger generations are less likely to agree than older ones. We use multivariate analysis to estimate the independent effects of age, gender, marital status, education and religious attendance on agreement with the idea that a life without children is not fully complete (see Appendix, table A2.2). The estimates show that men are still more likely to agree than women. Agreement increases with age, but after controlling for other factors, there is only a difference between the responses of the oldest age group (49 per cent of those aged over 65 agree) and other age groups (all around 35 per cent). The split by marital status shows that those who have been married are much more likely to agree that a life without children is not fully complete (42 per cent) than are those who have never been married (27 per cent). Education has a different impact than we find in previous models. Both certificate/diploma and university degree holders are less likely to agree to this statement than are those with less education. Finally, those who either sometimes or often attend religious services are more likely to say a life without children is not fully complete.
W H AT D O A U S T R A L I A N S T H I N K A B O U T REPRODUCTIVE CHOICE? Reproductive choice raises more highly charged questions about the role of children and marriage in Australian family life. In this section, we consider attitudes to abortion and access to fertility treatment in the form of in-vitro fertilisation (IVF). The public debate about abortion has a long history in Australia. Even after abortion laws were relaxed in some states in the early 1970s, some public opposition to abortion has continued. In July 2001, a man shot and killed a security guard at a Melbourne abortion clinic, committing Australia’s first abortion clinic murder. Yet abortion is not a rare procedure in Australia. According to data collected by the Health Insurance Commission, there were between 73 000 and 78 000 abortions performed per year in the late 1990s under the Medicare system (Health Insurance Commission, unpublished data). This figure does not include abortions performed by publicly funded doctors in public or private hospitals. Meyer, Evans and Rahman (2000) estimate a further 16 000 terminations are performed in Australia each year by publicly funded doctors in hospitals. Controversy over fertility treatment for single and lesbian women launched a public debate in 2000 when a Victorian provider, Dr John
AusSocAtts03
26/7/05
22 •
8:55 AM
Page 22
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
McBain, challenged the legality of Victorian legislation limiting access to IVF treatment to heterosexual married or de facto couples. Dr McBain claimed that this law contravened the federal Sex Discrimination Act (Rabsch 2002). The Federal Court agreed, and Justice Sundberg ruled in favour of McBain. The Australian Catholics Bishops Conference challenged the decision of the Federal Court and the matter was then heard in the High Court. The High Court upheld the decision by Justice Sundberg in April 2002. However, in June 2002, the federal Attorney-General introduced an amendment to the Sex Discrimination Act to allow states and territories to legislate to limit access to assisted reproductive technology services to married couples and couples in de facto relationships. So how do these protests and controversies reflect attitudes to women’s reproductive rights? A B O RT I O N
The majority of AuSSA 2003 respondents (83 per cent) agree that a ‘woman should have the right to choose whether or not she has an abortion’. Agreement differs by gender and by age (see table 2.5). Both men and women express a high level of agreement with a woman’s right to choose abortion (81 per cent and 84 per cent respectively), although women are more likely than men to express strong agreement (49 per cent versus 36 per cent). Men’s agreement is more consistent across the age spectrum: 84 per cent of men aged 18–34 agree with the right to choose, compared to 80 per cent of men aged over 65. By contrast, 90 per cent of young women agree, compared to 76 per cent of women aged over 65. Despite declining support for the right to choose with increasing age, a large majority of older respondents remain supportive. To explore attitudes to abortion further, we model the association between age, relationship status, education and religion with agreement to a woman’s right to choose abortion (see Appendix, table A2.3). We analyse the responses of men and women separately, because of the differences in their responses by age we discussed earlier. The model shows that attitudes to abortion differ significantly with religious beliefs and practices, marital status (for men), and age and education (for women). We use two measures of religion in these models because of the strong religious dimension in debates about abortion. First, we consider answers to the question ‘Do you have a religion?’ Those who indicate a religion are significantly less likely to agree with a woman’s right to choose abortion than are those with no religion. The model for men
AusSocAtts03
26/7/05
8:55 AM
Page 23
W H AT M A K E S A N A U S T R A L I A N F A M I L Y ?
• 23
Table 2.5 Right to choose an abortion by age and sex, 2003, per cent 18–34
35–49
50–64
65 and over
Total
Strongly agree
40
40
34
34
36
Agree
44
42
47
46
45
Neither agree nor disagree
7
10
8
10
9
Disagree
6
4
6
7
6
Strongly disagree
5
4
6
4
5
Men
n
(306)
(556)
(604)
(448)
(1914)
Strongly agree
57
55
46
39
49
Agree
33
33
38
37
35
Neither agree nor disagree
4
5
7
10
6
Disagree
3
4
5
8
5
Strongly disagree
4
3
5
7
4
Women**
n
(405)
(691)
(638)
(441)
(2175)
* p < 0.05; ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
predicts that 81 per cent of men who ascribe to a religion agree with a woman’s right to choose abortion, compared to 89 per cent of men with no religion. This pattern is the same for women. Second, we consider attendance at religious services as a measure of religious belief and practice. We find that the more frequently a person attends religious services, the less likely they are to agree with a woman’s right to choose abortion. Among men who attend religious services often, 60 per cent agree with a woman’s right to choose. Among men who never attend religious services, there is nearly universal agreement (92 per cent) with the right to choose. Again, the pattern for women is the same, with slightly higher levels of agreement (64 per cent agreement by those who attend often; 95 per cent among those who never attend). Among men, marital status is the only other factor that significantly affects their agreement with a woman’s right to choose abortion. Controlling for all other factors, men who are widowed, separated or divorced are less likely to agree with a woman’s right to choose abortion than are those who are married or cohabiting. There is no such difference among women. However, there are some significant differences among women other than religion. The model confirms that those aged 65 and over are less
AusSocAtts03
26/7/05
24 •
8:55 AM
Page 24
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
likely than are women under 35 years of age to agree with a woman’s right to choose abortion. Education also makes a small but significant difference, independently of religion and age. Women with a university degree (93 per cent) are more likely than those without a degree (90 per cent) to agree with the right to choose. Agreement with a woman’s right to choose abortion is high among AuSSA 2003 respondents. However, agreement is still not universal, and religion and attendance at religious services are crucial predictors of attitudes to abortion for both men and women. This finding reflects a recent study that found that young women are more likely to terminate a pregnancy if they agree with a woman’s right to choose abortion, which in turn, is affected by whether or not she reports having a religion (Evans 2001). IVF FOR UNMARRIED WOMEN
Access to IVF for unmarried women is also an issue of reproductive choice that has generated much public debate in Australia. Two polls conducted at the time of the Federal Court decision that we discussed earlier suggested opposition in the order of 80–90 per cent to single people having access to IVF treatment.1 AuSSA 2003 finds a higher level of agreement with the right of unmarried women to access IVF treatment: 38 per cent of respondents agree that ‘unmarried women should have as much right to fertility treatment (IVF) as married women’. Indeed, table 2.6 shows the now familiar pattern that, overall, women are more likely to agree than men, and that agreement decreases with age. The only variation on this theme is that, at older ages (50 and over), men are more likely to agree (57 per cent) that unmarried women should have the same access to IVF as married women than are women (54 per cent). We use the same model to predict agreement with access to fertility treatment for unmarried women as we did for abortion (see Appendix, table A2.3). However, the results for IVF are quite different. For abortion, religion and religious attendance are the key predictors of agreement. For IVF, other factors – including age, marital status and education – also matter. Controlling for all the other factors in the model, age is still an important predictor. For men, agreement ranges from 34 per cent of those aged 65 and over to 53 per cent of men aged less than 35. The pattern is the same for women, with agreement ranging from 25 per cent to 56 per cent. Older men are more likely to agree than older
AusSocAtts03
26/7/05
8:55 AM
Page 25
W H AT M A K E S A N A U S T R A L I A N F A M I L Y ?
• 25
Table 2.6 Unmarried women have as much right to IVF as married women, by sex and age, 2003, per cent 18–34
35–49
50–64
65 and over
Total
Strongly agree
20
12
11
7
Agree
32
23
21
18
23
Neither agree nor disagree
17
19
15
14
16
Men** 12
Disagree
18
27
30
38
29
Strongly disagree
12
18
24
23
20
(310)
(543)
(601)
(441)
(1895)
Strongly agree
34
25
12
9
Agree
29
26
19
14
22
Neither agree nor disagree
14
14
18
15
15
Disagree
15
22
33
39
28
8
14
19
24
16
(688)
(639)
(437)
(2159)
n Women**
Strongly disagree n
(395)
19
* p < 0.05; ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
women. Single men and women are the most likely to agree with IVF for unmarried women (47 per cent of men and 50 per cent of women). Over one-third of men who are married or cohabiting agree (38 per cent), with 40 per cent of men who are divorced, separated or widowed also agreeing. Married and cohabiting women are also less likely to agree with IVF for unmarried women than are women who are divorced, separated or widowed (38 per cent, compared to 44 per cent). Holding a university degree increases acceptance of IVF for unmarried women among both men and women. For men, 47 per cent of those with a Bachelor’s degree or higher agree with IVF for unmarried women, compared to 38 per cent of those with no degree. One half (50 per cent) of all women with a Bachelor’s degree or higher agree that unmarried women should have the same rights to access IVF as married women, compared to 38 per cent of women without a university degree. Religion matters to attitudes to IVF as it did for abortion. For both women and men, those who state that they have a religion are less likely than are those with no religion to agree that unmarried women have the same rights to IVF as married women. And, again, support drops as religious attendance increases. In general, AuSSA 2003 respondents are more supportive of women’s right to choose abortion than they are of unmarried women’s
AusSocAtts03
26/7/05
26 •
8:55 AM
Page 26
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
access to IVF. Notably, for abortion we see very little difference in the predicted percentages by age, whereas for IVF there is a sharp decline in the percentage agreeing for each successive age group.
W H AT R O L E F O R FAT H E R S ? We have seen that among AuSSA 2003 respondents two parents with children is the most commonly accepted definition of family. So how do respondents view the responsibility of fathers within families? In this section, we examine responses to the question ‘A father should be as heavily involved in the care of his children as the mother’. Over the past ten years, research on fatherhood has increased. Much of this research challenges the way men have been excluded or marginalised in parenthood research because ideas about parenting have too often been developed on the template of mothering (Marsiglio et al. 1998). Research investigating the division of caring between parents finds that shared caregiving (as defined by resident parents sharing the care of children) is not common, but families in which fathers are more involved in caring for children share some characteristics. These characteristics include having fewer and older children, or being older parents (Russell 1983; Ishii-Kuntz & Coltrane 1992), being economically active (Halle & Le Menestrel 2000) and having wives or partners who work (Radin 1994). Although fathers are likely to be more involved in families with these characteristics, mothers perform most child care. Australian estimates based on time-use data find that fathers spend around two hours and 20 minutes per day on child care, compared to mothers’ six hours and 50 minutes (Craig 2002). Further, fathers are more likely to be involved in passive and play-based care (Gray 2001). This much we know about time spent by fathers in intact family units. However, debate rages in Australia about divorced fathers’ involvement in their children’s lives. In this debate, shared care is defined as joint custody after divorce. A recent inquiry into child custody conducted by the House of Representatives Standing Committee on Family and Community Affairs attracted over 1700 submissions (HRSCFCA 2003). The result is a two-year trial of a Families Tribunal that will recommend joint custody for separating parents. The Australian Government has also recently released a proposal to establish Family Relationship Centres to assist separating parents and support joint parenting arrangements (Howard 2004b). On the face of it, AuSSA 2003 findings suggest that there is broad agreement with the idea of shared parenting, despite evidence to the
AusSocAtts03
26/7/05
8:55 AM
Page 27
W H AT M A K E S A N A U S T R A L I A N F A M I L Y ?
• 27
Table 2.7 A father should be as heavily involved in the care of his children as the mother by age and sex, 2003, per cent Strongly agree
Agree
Neither agree nor disagree
Disagree
Strongly disagree
Sex* Male
40
50
8
2
0
Female
45
46
7
2
0
18–34
52
39
8
1
0
35–49
39
49
9
3
0
50–64
39
52
7
2
0
65 and over
45
48
6
1
0
42
48
7
2
0
(1763)
(1995)
(84)
(11)
Age**
Total n
(306)
NOTE ‘Can’t choose’ is the remainder for each distribution.
* p < 0.05; ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
contrary when it comes to actual behaviour. When respondents are asked whether ‘A father should be as heavily involved in the care of his children as the mother’, 90 per cent agree (table 2.7). Differences in attitudes between men and women, or between different age groups tend to be in the strength of their agreement. Women are more likely than men to strongly agree (45 per cent versus 40 per cent) and those aged 18–34 are most likely to strongly agree (52 per cent), followed by those aged over 65. Respondents aged 35–64 are more likely to agree than strongly agree. An optimistic reading of these findings might signal good prospects for fathering behaviour. Alternatively, they may suggest that while people agree that a father should be as heavily involved as a mother in caring for his children, there are still significant barriers to his taking on this role.
CONCLUSION We can make four generalisations from the evidence that we have considered. First, young respondents accept a wider definition of the family. They are also the generation that has been most exposed to diverse living arrangements and their acceptance of this diversity perhaps reflects these experiences. Young people also take more liberal attitudes to reproductive rights and legal recognition of same-sex relationships. It will be interesting to follow these differences over time to
AusSocAtts03
26/7/05
28 •
8:55 AM
Page 28
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
see if younger generations become more conservative as they age. Second, with some minor exceptions (for example, older men’s stronger support for IVF for unmarried women), women are more accepting of a wider range of family types, more supportive of reproductive rights, and, intriguingly, less likely to think that a life without children is incomplete. More liberal attitudes among women may reflect their closer involvement in raising children. Third, religious affiliation is associated with less accepting views about family forms other than a married couple with children, and about reproductive rights. Fourth, and perhaps most important, is that the presence of children matters most to our definition of the family. Marriage is not as important: a sizeable majority recognise single parents with children as families, and a sizeable minority now recognise same-sex couples with children in the same way. The family, like other social institutions, is dynamic. Our perceptions of what makes a family are not constant, and family researchers, social commentators and policy-makers should be aware of current attitudes. Questions of family form, legal recognition for different family types, and access to reproductive technologies continue to be debated in Australian society, and policy will continue to respond to changing attitudes and behaviour. NOTE
1
Almost 80 per cent for A Current Affair and over 90 per cent for Melbourne Herald Sun. Cited from .
REFERENCES
Australian Bureau of Statistics (ABS) (1999) Marriages and Divorces, Australia, 1998, cat. no. 3310.0, Australian Bureau of Statistics, Canberra. —— (2002) Marriages and Divorces, Australia, 2001, cat. no. 3310.0, Australian Bureau of Statistics, Canberra. —— (2003a) Marriages and Divorces, Australia, 2002, cat. no. 3310.0, Australian Bureau of Statistics, Canberra. —— (2003b) Births, Australia, 2002, cat. no. 3301.0, Australian Bureau of Statistics, Canberra. Barlow, A, Duncan, S, James, G & Park, A (2001) ‘Just a piece of paper? Marriage and cohabitation’, in A Park, J Curtice, K Thomson, L Jarvis & C Bromley (eds) British Social Attitudes: The 18th Report, Sage, London, pp. 29–57. Craig, L (2002) ‘Caring differently: A time use analysis of the type and social context of child care performed by fathers and by mothers’, Social Policy Research Centre Discussion Paper no. 116, Social Policy Research Centre, University of New South Wales, Sydney. Evans, A (2001) Abortion or motherhood? Pregnancy resolution decisions of Australian teenagers, PhD thesis, Australian National University, Canberra.
AusSocAtts03
26/7/05
8:55 AM
Page 29
W H AT M A K E S A N A U S T R A L I A N F A M I L Y ?
• 29
Gibson, R, Wilson, S, Meagher, G, Denemark, D & Western, M (2004) Australian Survey of Social Attitudes 2003, Data file, Australian Social Science Data Archive, The Australian National University, Canberra. Gray, E (2001) Shared parenting in Australia?, PhD thesis, Australian National University, Canberra. Halle, T & Le Menestrel, S (2000) ‘How do social, economic, and cultural factors influence fathers’ involvement with their children?’, Child Trends Research Brief, Child Trends, Washington, DC. Hantrais, L & Letablier, M-T (1996) Families and Family Policies in Europe, Longman, London. House of Representatives Standing Committee on Family and Community Affairs (HRSCFCA) (2003) Every picture tells a story: Report on the inquiry into child custody arrangements in the event of family separation, The Parliament of the Commonwealth of Australia, Canberra. Howard, J (2004a) Press Conference, Parliament House Canberra, 27th April 2004. Available at: (accessed April 2004). —— (2004b) Reforms to the Family Law System, 29th July 2004. Available at: (accessed August 2004). Ishii-Kuntz, M & Coltrane, S (1992) ‘Predicting the sharing of household labor: Are parenting and housework distinct?’, Sociological Perspectives, 35: 629–47. Marsiglio, W & Day, R (primary authors), Evans, V, Lamb, M, Peters, E, & Braver, S (contributing authors) (1998) ‘Social fatherhood and paternal involvement: Conceptual, data, and policymaking issues. Report of the Working Group on Conceptualizing Male Parenting’, in Nurturing Fatherhood: Improving Data and Research on Male Fertility, Family Formation and Fatherhood, Federal Interagency Forum on Child and Family Statistics, Washington, DC, pp. 99–174. Meyer, P, Evans, A & Rahman, N (2000) Estimating the level of induced abortion in Australia, Paper presented at the Public Health Association of Australia Conference, Canberra, November 2000. Newspoll (2004) Newspoll 4–6 June 2004. Available at: (accessed August 2004). Pell, G (2004) ‘The case against gay marriage’, The Australian, 4 May 2004. Rabsch, M (2002) ‘The fallout from McBain: IVF, standing, international law, and human rights’, Law Society Journal, 40(6): 54–57. Radin, N (1994) ‘Primary-caregiving fathers in intact families’, in A Gottfried & A Gottfried (eds.) Redefining Families: Implications for Children’s Development, Plenum Press, New York, pp. 11–54. Russell, G (1983) The Changing Role of Fathers?, University of Queensland Press, Brisbane. Symons, E (2004) ‘Gay marriage ban just like “apartheid”’, The Australian, 27 April 2004. Wolcott, I & Hughes, J (1999) ‘Towards understanding the reasons for divorce’, Australian Institute of Family Studies Working Paper no. 20, Australian Institute of Family Studies, Melbourne.
AusSocAtts03
26/7/05
8:55 AM
Page 30
CHAPTER 3
M A K I N G FA M I L I E S W O R K Deborah Mitchell
INTRODUCTION Some social theorists claim that family is less important to how we define ourselves than it used to be. Instead, they argue, we increasingly think of ourselves as individuals and as part of friendship and social networks formed in the workplace and informal associations (Giddens 1991; Beck 1992). At the same time, how we balance family life with other pressures, especially work, continues to be hotly debated. These pressures have received much attention in the media, have come under scrutiny by governments, unions and industry, and have been analysed in publications with titles that have passed into popular usage – ‘juggling time’ (Bittman 1991), the ‘time bind’ (Hochschild 1997) and the ‘work/life collision’ (Pocock 2003). In the realm of policy and practice, ‘family friendly’ policies are much discussed by governments and employers, but work structures and family policies in Australia remain generally anchored in the male ‘breadwinner’ model; that is, a model that presumes men will work full-time while their partners take on family responsibilities (see Apps 2004). Against this background, changing gender roles and income needs have led couples with children to adopt a diverse set of strategies to cope with work and family life – even when these are not strongly supported by public policy (see Moen & Sweet 2003; OECD 2002, pp. 40–41). In this chapter, I use the Australian Survey of Social Attitudes 2003 (Gibson et al. 2004) to discover how different kinds of families are dealing with the problems of the work/life balance in three important policy-related areas – child care, working time, and living standards. My aim is to shed light on the interactions between ideas about identity and family, policies, and families’ strategies for managing work and life in contemporary Australia.
AusSocAtts03
26/7/05
8:55 AM
Page 31
MAKING FAMILIES WORK
• 31
Throughout the chapter, I focus on families with children aged 16 or under. I divide respondents into members of four family types: singleparent households; households with two full-time earners; ‘neotraditional’ households (with one full-time and one part-time earner); and ‘breadwinner’ households (with one earner and one stay-at-home partner).1 This analysis reveals how members of different types of families see themselves, how they adapt to child-rearing and income pressures, and how they view childcare provision, long working hours and their current and prospective living standards.
FAMILIES, IDENTITY AND THE CARE OF CHILDREN AuSSA 2003 asks respondents to rank the three groups that are most important in describing who they are. Respondents could choose from a list of ten items: their occupation, race or ethnic background, gender, age group, religion, preferred political party, nationality, family or marital status, social class, or the part of Australia that they live in.2 Their responses indicate that family and marital status matter greatly to identity – 52 per cent choose this first. As table 3.1 reports, in total, 74 per cent of respondents nominate their family or marital status as one of their three choices. The second most important choice is occupation – 16 per cent choose this option first, and, in total, 55 per cent of respondents choose work as one of their three choices. Third comes the part of Australia where the respondent lives: only 3 per cent of respondents identify first with the part of Australia they live in, but a high 14 per cent and 26 per cent choose this second or third respectively. Overall, then, despite recent theory to the contrary, these findings suggest that family remains central to Australians’ personal identity. Table 3.2 reports the percentages of those who choose family or marital status and occupation as their first choice by respondent age, sex Table 3.1 Identifying groups, 2003, per cent in rows Groups selected
First choice
Second choice
Third choice
Total
Your family or marital status
52
15
7
74
Your occupation
16
22
17
55
3
14
26
43
The part of Australia that you live in (n ≥ 2082) SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
32 •
8:55 AM
Page 32
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
and family type. We can see that women are more likely than men to choose their family or marital status as their primary identification (55 per cent, compared to 49 per cent). Age, clearly, is also an important factor distinguishing attitudes to family and identity, in part because our attachment to family follows the lifecourse. We can see that identifying with family rises for the 35–49 year old group and then stays much the same for older cohorts. Some of the 50–64 year old group are ‘emptynesters’ or have adult children still at home. The youngest age group – those 18–34 – show a surprising level of identification with family/marital status. Around 44 per cent identify with family first, and a total of 60 per cent of this group nominate it as one of their three choices. We also see in table 3.2 that identifying with work is highest among respondents aged 35–49 – the prime years in the labour market for many. When it comes to family type, we find strongest attachment to family among respondents from neo-traditional families, where it is usually the female partner who works part-time. Respondents from single-parent families are 10 per cent less likely to cite family as their primary identification than are others, and are a little more likely to choose occupation. Interestingly, we find the strongest attachment to occupation among respondents from breadwinner families (25 per cent). These respondents are overwhelmingly male earners, who presumably see themselves first and foremost as working to provide for their families. Table 3.2 Family and work identification by selected groups, 2003, per cent Family/marital status as first choice
Occupation as first choice
Male
49
15
Female
55
16
18–34
44
15
35–49
53
18
50–64
56
15
65 and over
52
16
Single parent (n = 101)
50
19
Two full-time (n = 122)
62
13
Neo-traditional (n = 197)
68
11
Breadwinner (n = 112)
56
25
Sex
Age
Family type
SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
8:55 AM
Page 33
MAKING FAMILIES WORK
• 33
Although family remains a mainstay of personal identity, evidence in chapter 2 suggests that Australians are beginning to adopt a more inclusive definition of family and are moving away from traditional family norms. Two examples of this are more accepting attitudes to child-rearing outside marriage and single-parent families. Taken with findings so far, the overall picture suggests that Australians are re-evaluating ideas about, and roles within, the family while still seeing family as central to identity. Tables 3.1 and 3.2 clearly suggest that family and work remain the anchors of personal identity for many Australians. Understandably, then, work and family are often in conflict. The care of children and the conflicting demands placed on working mothers sit at the heart of the family–work balancing act. On the one hand, studies reveal the guilt and anxiety that some working mothers feel when leaving preschool children in the care of others (Himmelweit & Sigala 2003, pp. 8–10). On the other hand, the financial, social and psychological rewards from working increasingly ‘pull’ women back into the labour market. As Pocock points out, virtually all mothers of young children have spent a considerable period in the labour market before childbirth and ‘now come to motherhood with an identity partly rooted in paid jobs’ (Pocock 2003, p. 74). Is this changing experience revealed in views about whether ‘a preschool child is likely to suffer if the mother works’? Policy, norms and experience all condition attitudes on this question. Table 3.3 again shows a marked divergence between the views of men and women. Men are far more likely to agree that preschool children suffer if the mother works as are older respondents.3 Because finding and managing child care are tasks predominantly performed by women (Bittman 1991; Baxter 2002; Pocock 2003), men may be less aware of the quality and impact of these arrangements on children. And younger women have fewer worries for their children because they are more likely to benefit from day care (especially regulated, centre-based care) than were older women who did not have as much access to these facilities when their children were very young. Do respondents in our four family types hold different opinions about child care? The answer is yes, and they hold views that match their actual labour market behaviour. Our breadwinners (with mothers at home) are most likely to agree that a child is likely to suffer if the mother works (46 per cent), compared to 29 per cent agreement among respondents from families in which both partners work full-time, who are presumably most likely to need, and be able to afford, child care. Here, neo-traditional families are much closer to families with two full-time
AusSocAtts03
26/7/05
34 •
8:55 AM
Page 34
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 3.3 Preschool child likely to suffer if mother works, 2003, per cent Agree
Neither
Disagree
Can’t choose
Male
44
24
30
2
Female
31
23
45
2
18–34
25
28
46
2
35–49
34
22
44
1
50–64
40
22
37
1
65 and over
49
23
25
3
Single parent (n = 187)
35
23
41
2
Two full-time (n = 237)
29
21
49
2
Sex
Age
Family type
Neo-traditional (n = 358)
30
18
51
1
Breadwinner (n = 225)
46
23
31
1
SOURCE The Australian Survey of Social Attitudes 2003
workers, which suggests that ‘juggling’ work and paid child care makes families less likely to think that a small child suffers if their mother works. The question of who is best suited to provide childcare services again provokes interesting patterns of responses among AuSSA 2003 respondents. This policy question matters not only for the care of children; it also matters greatly for the work and income prospects of mothers, especially for low-skilled women (Himmelweit & Sigala 2003, p. 22; for Australia’s current policies see OECD 2002, pp. 86, 100). In table 3.4, we see that nearly two-thirds of respondents think that governments and the private sector are better suited for delivering child care than families and relatives or charities (64 per cent, compared to 30 per cent). There is no difference in responses between men and women. The most striking finding is variation by age: older respondents overwhelmingly endorse government provision, well ahead of private provision or families. Those in the main child-rearing group (35–49 years), however, endorse the private sector just ahead of government provision, with families again being a third choice. Only the youngest cohort of the sample aged 18–34 – those most likely to have experienced non-family based child care – put families well ahead of both government and the private sector. Young respondents – who are either not raising children or have just started families – may lack the wider experience of child-rearing that presumably increases support for government and market-based care provision.
AusSocAtts03
26/7/05
8:55 AM
Page 35
MAKING FAMILIES WORK
• 35
Table 3.4 Best to deliver childcare services by sex and age, 2003, per cent Governments Charities
Private Families and businesses relatives
Can’t choose
Sex Men
34
4
30
26
6
Women
34
2
29
28
8
18–34
28
3
28
36
5
35–49
30
2
34
26
7
50–64
35
2
32
26
6
65 and over
43
5
23
23
7
Total
34
3
30
27
7
Age
(n ≥1996) SOURCE The Australian Survey of Social Attitudes 2003
Table 3.5 Delivery of childcare services by attitudes to child care, 2003, per cent Best suited to deliver childcare services … Fine for children under three years to be in full-time care
Governments
Charities
Private Families businesses and relatives
Can’t choose
Agree
36
2
34
21
6
Neither/Can’t choose
34
3
29
24
10
Disagree
32
3
26
33
5
(n = 2014) SOURCE The Australian Survey of Social Attitudes 2003
Table 3.5 suggests that attitudes to child-rearing matter quite a bit to our preferences for childcare provision. For the minority who agree that it is fine for children who are under three years to be in full-time care, governments and private businesses are clearly preferred providers. This is true for those who disagree as well, but support for family-based care is 12 per cent higher among this group than among those who agree. These responses suggest that, although there is a broad consensus that governments and private centres are best suited to deliver child care to the community as a whole, the preference for family care is strongly influenced by attitudes to child-rearing. However, again, experience is likely to condition attitudes. In their analysis of the British Household Panel Survey between 1991 and 1999, Himmelweit and Sigala show that British mothers with preschool children who entered employment
AusSocAtts03
26/7/05
36 •
8:55 AM
Page 36
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
in the previous two years and who thought that preschool children suffered were ‘more likely to change their attitude than their behaviour’ (Himmelweit & Sigala 2003, p. 13).
WORKING HOURS AND FAMILY LIFE When work and life collide for families, especially for families with working mothers, they do so for two main reasons. The first is the increasing length of the working day for many full-time workers. The mid-1980s saw a sustained reversal in the long-term trend of a decline in standard working hours (OECD 2004, p. 42). As the ABS has observed, Australia’s experience has been ‘relatively uncommon’ among OECD nations (ABS 2003, p. 120). Over the past decade, the kinds of jobs demanding long work hours have expanded out from management and professions to include a broad range of occupations and industries. Second is the lack of family-friendly policies in the workplace, especially for casual workers (OECD 2002, pp. 190–91), and the lack of affordable child care, even after recent government reforms (Marriner 2004). Together, longer hours for both men and women and costly childcare options contribute to a larger problem of ‘time-juggling’, which Bittman (1991) identifies as a major stress on families, especially where both parents work. Chapter 4 reports that around two-thirds of AuSSA 2003 respondents now believe that there are too many people working long hours in Australia. Only around one in six respondents disagree. Even so, around half of our respondents (51 per cent) believe that people make this choice freely; while 31 per cent believe that those who work long hours are not in that position by choice. Given talk of the ‘double shift’ for women and juggling work and family time, do working hours have different impacts on different kinds of families, and so shape respondent views about working time? Table 3.6 compares responses from the main earner in our four family types to four statements about work. Single-parent families are least likely to experience time pressure (because only a minority work full-time, a point taken up later). About a fifth say that their work hours are too long or that their employer expects them to work long hours, while around a third say that their hours interfere with family and personal life or that they would like to work fewer hours. Both breadwinner families and families with two full-time working parents experience more time pressure, with more than half respondents from both groups preferring fewer hours and finding that their working hours
AusSocAtts03
26/7/05
8:55 AM
Page 37
MAKING FAMILIES WORK
• 37
Table 3.6 Attitudes to working hours by family type, 2003, per cent agree Single parent
Two full-time
Neotraditional*
Breadwinner
My hours of work each week are too long
22
29
42
32
Employer expects me to work long hours
22
31
42
39
My hours of work interfere with family and personal life
33
51
61
50
Prefer to work fewer hours each week
30
58
67
51
(50)
(93)
(66)
(78)
n
* For neo-traditional families, respondent is the full-time earner. SOURCE The Australian Survey of Social Attitudes 2003
interfere with their family and personal life. But we find that neotraditional earners really stand out: they are more likely than respondents from any other family type to think that their working hours are too long (42 per cent), to feel pressured to work longer hours by their employer (42 per cent), to feel that work time interferes with family and personal life (61 per cent) and to prefer to work fewer hours (67 per cent). These findings are in broad agreement with qualitative research that shows that the lengthening work day of full-time men particularly affects fathers and interferes with the amount of time they can spend with their children (Pocock 2003, pp. 145–46). The reasons for pressure on neotraditional families seem obvious: they lack the ‘time resources’ of a stayat-home parent to take charge of family responsibilities, and, in the main, lack the ‘income resources’ of families with two full-time working parents to pay for support. Further research into adaptive labour market behaviour by families would provide further insight into the complex choices and binds that underlie the work/life balance.
FA M I LY L I V I N G S TA N D A R D S A N D C O P I N G Stresses on family life from the demands of paid work have led researchers, not to mention time-stretched parents, to wonder ‘why do they do it?’ As I noted earlier, apart from the social and psychological benefits that work and the workplace provide for women, and the pressure that employers place on men, financial benefits play an important role in choices about work participation and working hours. For many families, women’s entry to the labour market offers additional income resources to cope with tough labour markets and additional expenses associated with the lifecourse (mortgages, education and child care), even if income is gained at the cost of time and household work. How does
AusSocAtts03
26/7/05
38 •
8:55 AM
Page 38
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
women’s entry into the labour market affect how families judge their economic position and prospects? AuSSA 2003 includes several subjective measures of living standards and of expectations about economic mobility. We can use this data to assess how optimistic or pessimistic different kinds of families are, though obviously, other factors, such as income differences within family types, may explain some variation. Table 3.7 shows responses by family type to some subjective measures of living standards and expectations, as well as an objective measure in the form of gross annual household income.4 Optimism is the prevailing mood. Overall, we can see that a majority of respondents believe that their standard of living is better than their parents’. For every family type, more respondents agree that they have a good chance of improving their standard of living than disagree. On balance, respondents think that their children will have a better standard of living than they have. Finally, a majority of respondents from all family types are coping or comfortable on their current income. However, there is a clear divide between one-earner and two-earner families. Families in which both parents work full-time are the most optimistic, and coping best. Respondents from these families are more likely than those from other family types to say that their living standard is better compared to their parents (78 per cent), to expect that their children will have a higher standard of living than they have (45 per cent), to believe themselves to have a good chance of improving their own standard of living (56 per cent), and to be managing comfortably on their current income (33 per cent). These families are also most likely to report a household income in excess of $78 000 per year (57 per cent of this group). By contrast, single-parent families are least optimistic and coping worst (see also OECD 2002, p. 70). Respondents from these families are less likely than respondents from the other family types to say that their living standard is better compared to their parents (59 per cent), to believe themselves to have a good chance of improving their standard of living (38 per cent), and to be managing comfortably on their current income (12 per cent). Single-parent families are also least likely to report a household income in excess of $78 000 per year (9 per cent). Almost two-thirds of single-parent families reported an annual income below $31 199 in 2003 (compared to just 8 per cent and 7 per cent of two full-time earner and neo-traditional families respectively). Single parents are also most likely to report below-average incomes in their own estimation (54 per cent) and a high number indicate financial stress –
AusSocAtts03
26/7/05
8:55 AM
Page 39
MAKING FAMILIES WORK
• 39
fully 43 per cent of this group find it difficult or very difficult to get by, compared to 17 per cent in families with two earners. The absence of a partner’s income, coupled with often poor labour market experiences, put these families at the bottom of the household income distribution – and in need of government income support to maintain living standards (see McDonald 2001, and chapter 7 on the welfare state). Indeed, 45 per cent of single-parent respondents were not in paid work, similar to findings from the ABS (2004) and the OECD (2002, pp. 65–66). Breadwinners are in the middle, with 46 per cent saying they have a good chance of improving their living standards, presumably lower Table 3.7 Perceptions of current and future living standards by family type, 2003, per cent Single parent
Two full-time
Neotraditional
Breadwinners
Better
59
78
76
69
Same
22
13
16
19
Worse
18
8
7
12
Better
42
45
38
43
Same
22
28
29
22
Worse
29
21
26
25
Agree
38
56
50
46
Neither/Can’t choose
31
26
28
30
Disagree
31
18
22
24
Standard of living compared to parents
Children’s standard of living compared to respondents
Have a good chance of improving standard of living
Family income compared to average Below
54
14
13
34
Average
31
45
46
38
Above
11
41
41
27
How managing on current income Difficult
43
17
16
30
Coping
45
50
52
51
Comfortable
12
33
31
18
Gross household income
n
$0 – 31 199
65
8
7
31
$31 200 – 77 999
26
35
49
47
$78 000 or more
9
57
44
22
(232)
(344)
(214)
(175)
NOTE ‘Can’t choose’ not included for all variables. SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
40 •
8:55 AM
Page 40
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
than in dual-earner families because of more limited labour market participation. Interestingly, we find breadwinner families more evenly spread across the income distribution. More than a fifth of these families can probably afford a stay-at-home partner, while the bottom third, on both subjective and objective measures, might benefit from more labour market participation. Factors that influence family decisionmaking about female labour force participation include norms about child-rearing and gender roles (Himmelweit & Sigala 2003, p. 8), policy incentives for work or staying at home (Apps 2004), and labour market opportunities.
CONCLUSION This chapter reveals what families share in common as well as their major differences. We find that family life remains central to the identity of Australians, although breadwinning men hold a special attachment to work, perhaps in their role as hard-working providers. The same breadwinners are also more attached to the family for child care, but along with the others, they still prefer governments and private providers to deliver it. Working time pressures are hitting neo-traditional families hardest. These families, with a second part-time worker, have a bit more time than full-timers, and a bit more income than breadwinners – but perhaps not enough of either. Lastly, single parents are feeling the pinch. Fewer than half in the AuSSA 2003 sample have even one full-time income, underlining their reliance on government support. In recent times, public policy has recognised that protecting family incomes is central to living standards and a strong society (see also chapter 7). But policy-makers still face the challenge of understanding and assisting diverse families who, as we have seen in this chapter, face different income and time pressures (see also OECD 2002). A lot of this difference depends on whether women take up paid work and then how much they do. Three decades after Australian women started entering the labour market in large numbers, we find frictions between work and family life. By looking at attitudes and behaviour in different kinds of families, we have revealed the challenges that all families face, some with more resources than others. NOTES
1
Respondent’s work status is based on their hours usually worked. As usual, hours are not known for partners; their work status is based on the main activity they did last week.
AusSocAtts03
26/7/05
8:55 AM
Page 41
MAKING FAMILIES WORK
2 3 4
• 41
These items are part of the International Social Survey Program’s 2003 National Identity module fielded in more than 30 countries. See also ISSP findings for 1994–95, cited in OECD (2002, pp. 56–57). Only individual case analysis would allow us to link subjective and objective indicators of living standards to see how closely they match. But our purpose here is to look at how family types, rather than individual families, judge their living standards and compare for household incomes.
REFERENCES
Apps, P (2004) ‘The high taxation of working families’, Australian Review of Public Affairs, 5(1): 1–24. Australian Bureau of Statistics (ABS) (2003) Australian Social Trends 2003, cat. no. 4102.0, Australian Bureau of Statistics, Canberra. —— (2004) ‘The measures: Family, community, and social cohesion’, Measures of Australia’s Progress, cat. no. 1370.0, Australian Bureau of Statistics, Canberra. Baxter, J (2002) ‘Patterns of change and stability in the gender division of labour in Australia, 1986–1997’, Journal of Sociology, 38(4): 399–424. Beck, U (1992) Risk Society: Towards a New Modernity, Polity Press, Cambridge. Bittman, M (1991) Juggling Time: How Australian Families Use Time, Australian Government Publishing Service, Canberra. Gibson, R, Wilson, S, Meagher, G, Denemark, D & Western, M (2004) Australian Survey of Social Attitudes 2003, Data file, Australian Social Science Data Archive, The Australian National University, Canberra. Giddens, A (1991) Modernity and Self-Identity: Self and Society in the Late Modern Age, Blackwell, Cambridge. Himmelweit, S & Sigala, M (2003) ‘Choice and the relationship between attitudes and behaviour for mothers with preschool children: Some implications for policy’, Hawke Research Institute Working Paper Series no. 23, Hawke Research Institute, University of South Australia. Hochschild, A (1997) The Time Bind: When Work Becomes Home and Home Becomes Work, Metropolitan Books, New York. McDonald, P (2001) ‘Family support policy in Australia: The need for a paradigm shift’, People and Place, 9(2): 14–20. Marriner, C (2004) ‘Don’t make me scream’, Sydney Morning Herald, 22 November. Moen, P & Sweet, S (2003) ‘Time clocks: Work-hour strategies’, in P Moen (ed.) It’s About Time: Couples and Careers, Cornell University Press, Ithaca, New York. Organisation for Economic Cooperation and Development (OECD) (2002) Babies and Bosses: Reconciling Work and Family Life: Australia, Denmark and the Netherlands, vol. 1, Organisation for Economic Cooperation and Development, Paris. —— (2004) OECD Employment Outlook, Organisation for Economic Cooperation and Development, Paris. Pocock, B (2003) The Work/Life Collision, The Federation Press, Sydney.
AusSocAtts03
26/7/05
8:55 AM
Page 42
CHAPTER 4
HOW DO AUSTRALIANS FEEL ABOUT THEIR WORK? Bill Martin and Jocelyn Pixley
INTRODUCTION Paid work is an integral part of most Australian lives,1 and the world of work has been dramatically transformed in the past 20 years. Part-time, casual and contract employment have grown rapidly, women’s participation in paid employment continues to rise and most partners in couple families with children are now in paid jobs. The industrial relations landscape is also very different, with reduced rights and protection for many workers and reduced involvement of unions. Many employers have restructured work and sought to increase their flexibility to hire and fire workers (see Hancock 2002; Watson et al. 2003; Wooden 2002). Some commentators argue that recent workplace changes have made the experience of work less positive for many people (Watson et al. 2003; Tiffen & Gittins 2004, pp. 66–86), while others suggest the opposite (Mathews 1989; Wooden 2002). These debates mainly draw on objective indicators such as pay or conditions. In this chapter, we explore people’s attitudes and evaluations to understand how Australians feel about their working lives. Who is satisfied with their job? Who feels most insecure? How are longer working hours affecting us? And what about the work/family balance, or ‘work/life collision’ as Barbara Pocock (2003) calls it? By examining what the Australian Survey of Social Attitudes 2003 (Gibson et al. 2004) tells us, we can build an illuminating picture of how Australians are dealing with the challenges of working life.
AusSocAtts03
26/7/05
8:55 AM
Page 43
HOW DO AUSTRALIANS FEEL ABOUT THEIR WORK?
• 43
HOW DO AUSTRALIANS EXPERIENCE THEIR WORK? Among the questions about working life in AuSSA 2003 is the simple ‘All in all, how satisfied are you with your job?’. The vast majority of respondents express satisfaction. Table 4.1 shows that only 11 per cent express any level of dissatisfaction and, on an 11-point scale, more than 20 per cent place their satisfaction in the top two categories. While these findings certainly indicate that Australians are generally satisfied with their jobs, they are not unusual. A study of 21 countries in 1999 found similar results; only 11 per cent of Britons, 10 per cent of French, 7 per cent of New Zealanders and 4 per cent of Dutch respondents expressed dissatisfaction (Sousa-Poza & Sousa-Poza 2000). Two-thirds of AuSSA 2003 respondents also feel that they are able to use their abilities or qualifications in their jobs, though a fifth say that they do not use their skills at work. This is important, partly because maximising skill use matters for Australia’s economic performance, but also because people are generally happier at work if their skills are used. Indeed, 28 per cent of those who say they do not use their skills also express some dissatisfaction with their jobs, compared to only 6 per cent of those who say they do use their skills. Relations between management and employees are important in people’s everyday experience of work, and most AuSSA 2003 respondents (62 per cent) agree that these relations are ‘good … in my workplace’. However, few feel this emphatically (only 15 per cent strongly agree), and just over a fifth of respondents do not see workplace relations as good. Again, this is important because workplace harmony improves job satisfaction (see table 4.2). According to many commentators, one of the most important effects of workplace and labour market change in recent decades has been to increase people’s insecurity about their ability to keep their jobs and to get a new job if they are retrenched (Watson et al. 2003; Sennett 1998). AuSSA 2003 results partly confirm this view. Around one in eight respondents say that they are ‘very likely’ or ‘fairly likely’ to lose their jobs in the next year, although a little over half think that this is ‘not at all likely’. This level of insecurity appears to be out of proportion to the real risks of job loss, since ABS data shows that about 2 per cent of Australian workers actually experience retrenchment in any 12-month period (ABS 2002). Most Australians also doubt their marketability – that is, their ability to get another job that is at least as good as the one they currently have. Only 15 per cent of respondents say it would be
AusSocAtts03
26/7/05
44 •
8:55 AM
Page 44
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 4.1 How Australians experience work, 2003, per cent All in all, how satisfied are you with your job? (0 = extremely dissatisfied, 10 = extremely satisfied) 0–4
11
5
10
6–7
33
8
25
9–10
22
I don’t get a chance to use my abilities or qualifications at work Agree
20
Disagree
67
Management and employees have good relations in my workplace Agree
62
Disagree
21
Thinking about the next 12 months, how likely is it that you will lose your job or be retrenched? Very likely
5
Fairly likely
8
Not too likely
34
Not at all likely
54
How easy would it be for you to find a job with another employer that has approximately the same income and conditions? Very easy
15
Somewhat easy
35
Not easy at all
50
I feel under pressure to work harder in my job Agree
43
Disagree
34
(n≥ 2429) SOURCE The Australian Survey of Social Attitudes 2003
‘very easy’ to do so, while exactly half say it would be ‘not easy at all’. Fear of job loss and lack of confidence about one’s marketability may make Australians more compliant when employers demand more of them. Certainly, a large minority of respondents (43 per cent) feel under pressure to work harder, although a sizeable minority (34 per cent) feel no such pressure. Respondents who feel under pressure to work harder are twice as likely to think that they are fairly or very likely to lose their job in the next year as those who do not (25 per cent versus 12 per cent). Together, these initial results give us a picture of overall trends: most respondents like their jobs, think that their skills are used at work, report fairly harmonious management–employee relations and are not afraid of
AusSocAtts03
26/7/05
8:55 AM
Page 45
HOW DO AUSTRALIANS FEEL ABOUT THEIR WORK?
• 45
losing their jobs. However, responses do vary in some important dimensions of working life, notably perceptions of marketability and feeling under pressure to work harder. We return to these later.
W H AT A F F E C T S J O B S AT I S FA C T I O N ? There is reason to think that some kinds of workers, or workers in some kinds of jobs, are more likely than others to be less satisfied with their jobs – with their pay and conditions, with their job security, and with how compatible their job is with family life. We could speculate that older workers are less satisfied because they are more likely to be retrenched and to find it harder than younger workers to find another job. Perhaps women, too, are less satisfied because they face disadvantage in the labour market, because they are directly discriminated against, because they are concentrated in ‘feminised’ occupations with low pay and poor conditions, or because family responsibilities restrict their ability to demonstrate the outward commitment that many employers require for advancement. Loss of once exemplary pay and working conditions may undermine the job satisfaction of public sector workers as governments cut spending and privatise. People with fewer qualifications and those in casual and part-time jobs, particularly jobs demanding little formal training, might be less satisfied because their income and benefits (such as sick pay, leave entitlements, and job security) have been limited or reduced under new industrial relations rules. Other workers, meanwhile – notably those with higher educational qualifications and full-time managerial or professional jobs – may be more satisfied with their jobs because their rewards have increased. Perhaps surprisingly, many of these factors do not explain job satisfaction: men and women are equally satisfied, people’s hours of work are not related to their satisfaction, nor is their education.2 Even income is very weakly related to job satisfaction. Table 4.2 shows that age, occupation and workplace experience make a difference to job satisfaction. Workers aged 55 and over are more satisfied than those under 55, with an average score of 7.4 on our 11-point job satisfaction scale, compared to 6.9 for the younger age group. Perhaps these over 55s feel privileged just to have a job, and therefore have higher job satisfaction than younger people who feel that they have a right to a job. Table 4.2 also shows that people in some occupations are somewhat more satisfied than others with their jobs. On the 11-point scale, managers average about 7.5, while lower skilled workers (including routine clerical and administrative occupations) average about 6.6. These differences are significant,
AusSocAtts03
26/7/05
46 •
8:55 AM
Page 46
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
but they are quite small – around 1 point on our 11-point scale. Moreover, average scores obscure some wide variation in job satisfaction within occupational and age groups. For example, although the average job satisfaction level for lower skilled workers is 6.6, more than a third (38 per cent) gave their job satisfaction as 8 or above. Workplace experience makes much more difference to job satisfaction than age and occupation. Table 4.2 shows that being able to use one’s skills, having good workplace relations, and feeling secure in one’s job each make a difference of around two points on our 11-point scale, twice the variation by occupation and nearly four times that by age. Moreover, these factors are much more consistently associated with job satisfaction than occupation or age. For example, we can predict a person’s job satisfaction with much more confidence by knowing whether their workplace relations are good than by knowing their occupation or age. Table 4.3 confirms that workplace experiences are not strongly related to income or occupation. It is true that respondents in Table 4.2 Job satisfaction by occupation, age and workplace experiences, 2003 Mean job satisfaction*
Occupation Managers
7.5
Professionals
7.2
Associate professionals
7.4
Tradespersons
6.9
Advanced clerical and service
7.2
Lower skilled
6.6
Age Under 55
6.9
55 and over
7.4
Workplace experience Good relations between managers and employees
7.4
Not good relations between managers and employees
5.5
Use skills
7.4
Don’t use skills
5.5
Not under pressure to work harder
7.5
Under pressure to work harder
6.4
Not at all likely to lose job
7.5
Not too likely to lose job
6.7
Very or fairly likely to lose job
5.8
(n≥ 2422) * 0 = extremely dissatisfied; 10 = extremely satisfied SOURCE Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
8:55 AM
Page 47
HOW DO AUSTRALIANS FEEL ABOUT THEIR WORK?
• 47
Table 4.3 Skills and workplace relations by hourly income and occupation, 2003, per cent I don’t get a chance to use my abilities or qualifications at work
(per cent agree)
Income (n = 2353) Earn under $22 per hour
25
Earn over $22 per hour
14
Occupation (n = 2410) Low skill occupation*
30
Other occupation
14
Management and employees have good relations in my workplace
(per cent disagree)
Occupation (n = 2404) Managers
15
Professionals
21
Associate professionals
16
Tradespersons
23
Advanced clerical
20
Other (low skill) occupations
25
* Low skill occupations include the following four major groups from the Australian Standard Classification of Occupations (Australian Bureau of Statistics 1997): labourers; elementary clerical, sales and service workers; intermediate production and transport workers; and intermediate clerical, sales and services workers. SOURCE The Australian Survey of Social Attitudes 2003
lower paid jobs and lower skill occupations are less likely to feel that their skills are used. But most people still feel that their skills are used. Among respondents reporting incomes in the bottom 60 per cent of hourly earnings (that is, those earning less than $22 per hour), 25 per cent feel that their skills are not used, compared to 14 per cent in the top 40 per cent of the earnings distribution. And among respondents in lower skill occupations, 30 per cent of people feel that their skills are not used, compared to 14 per cent in other occupations. Those in lower skilled jobs are also more likely than others to report poor workplace relations, but the differences are small: 25 per cent of people in lower skilled jobs indicate that manager–employee relations are not good in their workplace, compared to 15 per cent of managers, 21 per cent of professionals and 20 per cent of advanced clerical workers. Overall, then, these aspects of workplace experience do vary significantly, but we cannot find strong connections between people’s experience and their occupations. Instead, skill use, job security and the quality of relationships seem to be matters internal to workplaces and therefore under the influence of management. Job satisfaction is widely recognised as a relative phenomenon – it largely reflects people’s assessment of various aspects of their job
AusSocAtts03
26/7/05
48 •
8:55 AM
Page 48
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
against what they think they might reasonably hope for. That status and income have relatively little impact on people’s job satisfaction means that most feel that they could not reasonably expect much more on these dimensions. On the other hand, most people view poor workplace experiences as unacceptable and changeable. Therefore, these experiences do translate quite consistently into lower job satisfaction. Because job satisfaction is quite closely related to people’s overall happiness, employers can significantly promote many employees’ job satisfaction and overall happiness by developing good relations between workers and management, by enhancing job security, by providing employees with opportunities to use their skills, and by ensuring that they do not feel under pressure to work harder. For most Australians, a positive experience of work depends more on resolving these everyday workplace issues than on achieving significantly higher incomes or ‘better’ jobs.
WHO FEELS MOST INSECURE ABOUT THEIR JOBS? Beliefs about job security can give us a sense of people’s perceptions of the wider world of work. Undoubtedly, some managers use the threat of job loss as a way of controlling workers and, as we noted above, Australia offers little legislative job protection (OECD 2004, p. 72; Tiffen & Gittins 2004, pp. 80–81). But factors beyond the control of individual firms, such as changing employment practices in general and growing competition between businesses, also increase the risk of retrenchment for employees. In this section, we find out which workers believe that they are most at risk of job loss, and which expect to have most difficulty finding another job. Surprisingly, some factors we might think are important seem to bear little on job security or people’s confidence about their marketability. AuSSA 2003 respondents who have higher levels of education, those whose jobs are well paid and those who work long hours are no more secure or confident of their marketability than others, once we take account of their occupations. The factors that do matter are gender, age and, as we see in the next section, sector of employment. We find that men are generally a little less confident about their job security and about their marketability than women. Although the differences are not large, this is a robust finding and indicates that, on this score at least, women are not worse off than men. Further, age affects people’s sense of both their job security and their marketability, but in
AusSocAtts03
26/7/05
8:55 AM
Page 49
HOW DO AUSTRALIANS FEEL ABOUT THEIR WORK?
• 49
different ways. People aged 45–54 are a little less confident about their job security than those older and younger. This probably reflects the tendency for employers to retrench or give early retirement to workers in this age group. But when it comes to confidence about finding another job, there is a strong and unequivocal age gradient. Only 25 per cent of those under 25 believe that it would be ‘not easy at all’ to find an equivalent job to their current one, compared to nearly half (46 per cent) of those aged 35–44 and fully two-thirds (67 per cent) of those aged over 54.
IS THERE A TRADE-OFF BETWEEN JOB S E C U R I T Y A N D F U T U R E M A R K E TA B I L I T Y ? Do people in better jobs feel more secure and have greater confidence in their marketability? Taking security first, respondents in higher skilled occupations are less concerned about losing their jobs. About 63 per cent of managers and professionals respond that it is ‘not at all likely’ that they will be retrenched in the next 12 months, compared to 44 per cent of those in lower skilled jobs. But, as figure 4.1 shows, public sector workers generally feel more secure than those in the private sector. The gap among tradespersons is particularly striking, with public sector tradespersons feeling as secure as all other groups except managers. The exception to this pattern is the large group of public sector workers in the lower skill occupations who feel no more secure than their private sector counterparts. Turning to marketability, it is clear that job security does not necessarily translate into confidence about marketability. Managers and private sector tradespersons provide the most striking examples. While we saw managers are among the most confident about holding their jobs, they are less confident than many others about their marketability; indeed they are the least confident of private sector workers. On the other hand, private sector tradespeople, though among the most uncertain about their job security, are among those most confident of their marketability. The apparent ‘trade-off ’ between security and marketability evident here applies if we compare the public and private sectors. While public sector workers are more likely than those in the private sector to feel secure, they are also more likely to doubt their ability to find a job equivalent to their current one if they were to lose it. The exceptions to this pattern appear to be that public sector managers and professionals feel just as marketable as their private sector equivalents, though they are more secure, and low-skill public sector workers feel no
26/7/05
50 •
8:55 AM
Page 50
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Figure 4.1 Perceptions of job security and marketability by occupation, public and private sectors, 2003, per cent
90 80 70 60 50 40 30 20
Private Public/Non-profit Low skill occupations
Advanced clericals
Tradespersons
Associate professionals
Professionals
10 0
Managers
Not at all likely to lose job (%)
Job security by occupation, public and private sectors
Marketability by occupation, public and private sectors 70 60 50 40 30
Private
20
Public/Non-profit
10
Low skill occupations
Advanced clericals
Tradespersons
Associate professionals
Professionals
0
Managers
Very or somewhat easy to find equivalent job (%)
AusSocAtts03
SOURCE The Australian Survey of Social Attitudes 2003
more secure than their private sector compatriots, though they feel less marketable. These results suggest that ‘market’ reforms to public sector employment have benefited more skilled and qualified workers, so that they understand that their skills are valuable in both the public and private sectors, while harming lower skilled workers by undermining their job security without engendering a corresponding increase in their sense of marketability.
AusSocAtts03
26/7/05
8:55 AM
Page 51
HOW DO AUSTRALIANS FEEL ABOUT THEIR WORK?
• 51
Despite these complications, we interpret figure 4.1 as conveying one very clear message about the contemporary Australian labour market. Professionals are the clear ‘winners’ in job security and marketability – they have better than average job security and great confidence about their marketability – while people with few formal skills are ‘losers’, having low job security and limited confidence about finding another job with equivalent pay and conditions to their current one. Given that the average pay and conditions of such workers have seen the least improvement during the boom of the past ten years (Watson et al. 2003, pp. 123–31), these findings show that the new flexibility does not raise all occupational boats, only yachts.
ARE WE WORKING TOO MUCH? Research suggests that workplace restructuring and deregulation of the labour market have led to a ‘lean’ workforce that is expected to work longer hours (Brenner 1998; Moody 1997). There is good evidence that the proportion of Australians working long and very long hours has been steadily increasing over the last 20 years, particularly among men. In 2002, 29 per cent of men worked very long hours (50 or more per week), compared to 22 per cent in 1982 (Watson et al. 2003, p. 86; see also OECD 2004, p. 42). Indeed, on average, Australians spend longer in their paid jobs than people in many other countries – on one recent analysis, longer than in any comparable country (Tiffen & Gittins 2004, p. 83). In this section, we examine how respondents to AuSSA 2003 feel about their working hours. Do they think that their working hours are too long? Do they want to work fewer hours? Do their working hours interfere with their personal and family life? And do people feel under pressure to work harder? Table 4.4 shows that nearly 30 per cent of respondents agree that their own hours of work are too long (37 per cent of full-time workers), though half disagree. But half want to work fewer hours, indicating that people do not have to feel that they are working too hard to wish to work less! Respondents are fairly evenly split on whether their hours of work interfere with their family and personal life. About half of all respondents agree that most people who work long hours choose to do so, though a sizeable minority (31 per cent) did not agree. Nearly a third say that their employers expect long hours from them (31 per cent of all workers and 36 per cent of fulltime workers). There can be little doubt, though, that long hours are widely perceived as something of a social problem: more than two-thirds agree that too many people work long hours in Australia today.
AusSocAtts03
26/7/05
52 •
8:55 AM
Page 52
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 4.4 Perceptions of working hours, 2003, per cent My hours of work interfere with my family and personal life Agree
41
Disagree
45
My hours of work each week are too long Agree
29
Disagree
49
I would prefer to work fewer hours each week Agree
49
Disagree
30
My employer expects me to work long hours Agree
31
Disagree
52
There are too many people working long hours in Australia today Agree
68
Disagree
15
Most people who work long hours choose to do so Agree
52
Disagree
31
(n≥ 1201) SOURCE The Australian Survey of Social Attitudes 2003
Unsurprisingly, as table 4.5 shows, Australians who do work long hours are more likely to express negative attitudes to their hours of work. However, only when people work long (more than 44 hours per week) or very long (more than 55) hours do these effects bite sharply. About half of those who work long hours and nearly three-quarters of those working very long hours agree that their hours interfere with their family and personal life and that their hours of work are too long, compared to a third or less of employees working shorter hours. But this is not to say that those working shorter hours feel no work pressure. Even among those who work 20 hours per week or less, nearly a third say that they feel under pressure to work harder in their jobs, while over half of those on very long hours report this. Overall, well over half of Australian full-time workers prefer shorter working hours, with fairly small differences between those working ‘normal’ full-time hours (35–44 hours) and those working long or very long hours. However, even a quarter to a half of those working part-time also want to work fewer hours. Table 4.6 shows strong correlation between feeling pressured to work harder, agreeing that work hours are too long and interfere with family and personal life, and preferring to work fewer hours. These
AusSocAtts03
26/7/05
8:55 AM
Page 53
HOW DO AUSTRALIANS FEEL ABOUT THEIR WORK?
• 53
Table 4.5 Attitudes to working hours by hours worked, 2003, per cent agree Work hours per week
20 or less
21–34
35–44
45–54
55+
29
39
41
51
55
9
8
20
48
70
Interfere with family/ personal life (n = 1181)
22
31
34
55
72
Prefer fewer hours
23
33
53
64
68
Feel under pressure to work harder (n = 2441) Hours are too long
(n = 1187)
(n = 1179) SOURCE The Australian Survey of Social Attitudes 2003
strong relationships indicate something of a syndrome of overwork. But who feels this syndrome? Can we be more specific than just saying it is those who work long hours? Table 4.7 suggests that we can dismiss some explanations. Age has little or no impact on how respondents experience work pressure and working too much. Work pressure varies slightly with age – people in the mid-career age group of 35–49 generally feel a little more pressure to work harder than those older or younger. People aged 50 and over are less likely to feel that work interferes with their family and personal life than those aged under 50. But there are no differences by age in agreement that one’s hours are too long or that one would prefer to work fewer hours. It is striking here that people in the prime childbearing age group do not report suffering the negative effects of long hours any more than other age groups. However, respondents in different occupations experience work pressure and the negative effects of working time quite differently. This is primarily because those in professional and managerial jobs work longer hours than others. Although more secure at work, respondents with these ‘better’ jobs are clearly stressed. Well over half of managers (57 per cent) feel pressure to work harder, as do nearly half of professionals (47 per cent), whereas only just over a third of tradespersons, advanced clerical workers and those in lower skill jobs feel this way. Respondents in higher skill occupations are also more likely to agree that their hours interfere with family and personal life. Managers and associate professionals feel this particularly frequently, perhaps since these groups generally have less control over their hours of work than professionals, but also because their hours of work are simply longer. Indeed, the effects of hours actually worked are substantial enough to account for virtually all of the association with occupation. These patterns confirm
AusSocAtts03
26/7/05
54 •
8:55 AM
Page 54
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
that higher skill jobs, particularly managerial ones, are demanding and stressful to the point that they interfere with private lives. People in the jobs work very long hours to try to complete all their work, but still feel that they need to work more. Those in higher skilled jobs are also more likely to say that their hours are too long, though most of this effect is because they actually work longer hours. However, managers are especially likely to worry about their hours, even given that they work longer hours than others, and tradespeople seem to be somewhat less likely to worry, considering that they also tend to work long hours. These differences are intriguing, Table 4.6 Attitudes to work pressures, 2003, bivariate correlations Feel under pressure to work harder
Feel under pressure to work harder
Hours too long
0.392**
Hours too long
0.392**
Work interferes
0.393**
0.640**
Prefer to work fewer hours
0.303**
0.544**
Work interferes
Prefer to work fewer hours
0.393**
0.303**
0.640**
0.544** 0.528**
0.528**
(n≥ 1194) ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
Table 4.7 Attitudes to work pressures by age and occupation, 2003, per cent Feel under pressure to work harder
Hours too long
Work interferes
Prefer to work fewer hours
Under 35
41
28
45
48
35–49
47
30
46
51
50 and over
39
28
33
48
(2489)
(1198)
(1191)
(1189)
Age
n Occupation
n
Managers
57
50
52
66
Professionals
47
31
44
53
Associate professionals
44
37
49
51
Tradespersons
37
21
35
51
Advanced clericals
38
24
26
54
Lower skill occupations
38
20
36
41
(2421)
(1175)
(1168)
(1166)
SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
8:55 AM
Page 55
HOW DO AUSTRALIANS FEEL ABOUT THEIR WORK?
• 55
possibly indicating that people assess the appropriateness of their hours within varying ‘occupational cultures’. The ‘long hours’ culture in management is well known and recent publicity and shifts in workplace messages, including moves towards more ‘family friendly’ policies, may have raised awareness among managers of long hours. At the same time, the culture within skilled trades may be much slower to change, especially given its masculine nature (nearly 90 per cent of tradespeople among AuSSA 2003 respondents are male). Another possibility is that work stress heightens concern about long hours. Perhaps managers suffer particularly high levels of workplace stress: they are certainly the ‘meat in the sandwich’ in much public sector budget cutting and private sector restructuring. Tradespeople may suffer less stress of this kind, perhaps because many feel that they can control their hours of work. AuSSA 2003 cannot answer these questions, but they are interesting lines for future research.
A ‘WORK/LIFE COLLISION’? Barbara Pocock (2003) has argued that long working hours are harming families and communities. For Pocock, families are at the centre of a ‘work/life collision’ as long hours and changes in the organisation of work limit the time available to care for children and other dependants, and for sustaining intimate relationships. Arlie Hochshild (1997) offers a different image. She finds Americans facing complex dilemmas that produce a ‘time bind’: as women have increased their participation in the labour force and as workplaces increasingly offer social support and stimulating experiences, home life comes to feel more like ‘work’, like an obligation with uncertain rewards and unrelenting demands. Those in better jobs with more positive experiences are particularly likely to feel the resulting dilemma of time allocation. So, are Australians experiencing a ‘work/life collision’ as Pocock argues, or are we confronting Hochschild’s ‘time bind’? The ‘work/life collision’ story implies that respondents in families, particularly those with younger children, are most likely to agree that their work hours interfere with their family and personal life. As table 4.8 shows, respondents in households with at least one child under 13 are more likely to agree (48 per cent). But the difference between them and other households is not overwhelming, and 39 per cent of those in households without younger children also agree. The idea that families feel work/life stress most also suggests that respondents in families in which both partners work full-time will be more stressed than those in
AusSocAtts03
26/7/05
56 •
8:55 AM
Page 56
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 4.8 Work interferes with family and personal life, by family type, 2003, per cent My hours of work interfere with my family/personal life
Agree
Children aged 12 or under in household Yes
48
No
39
(n = 1176) Partner Male full-time, partner works full-time
46
Male full-time, partner works part-time
54
Male full-time, partner not work for pay
52
Male full-time, no partner
45
Female full-time, partner works full-time
47
Female full-time, no partner
39
(n = 736) SOURCE The Australian Survey of Social Attitudes 2003
traditional arrangements in which a husband works full-time and his partner does not work for pay at all. However, table 4.8 shows that, if anything, men whose partners do not work or who work part-time are more likely to agree that their hours of work interfere, though the difference here is too small to be reliable.3 In short, while family circumstances have some effect on perceptions of time stress, the hours respondents work affect their perceptions of stress much more. A good test of people’s perceptions about their working hours is whether they would actually prefer to work fewer hours. Analysing these preferences helps us understand the trade-offs that people are willing to make. First, we have seen that family factors are unrelated to preferences for shorter hours: neither the presence or absence of children nor whether or not both partners in a household are in paid work makes a difference. Second, and unsurprisingly, table 4.5 shows that those working longer hours are more likely than those working shorter hours to prefer fewer hours. But, again, the differences are not large, so long hours themselves do not appear to be closely related to preferring to work fewer hours. Similarly, occupation makes less difference (table 4.7) on this question than other factors that we have considered. However, table 4.9 shows that whether people want shorter hours is very strongly related to their job satisfaction. Between 60 per cent and 70 per cent of those who express some dissatisfaction with their jobs want shorter hours, compared to
AusSocAtts03
26/7/05
8:55 AM
Page 57
HOW DO AUSTRALIANS FEEL ABOUT THEIR WORK?
• 57
Table 4.9 Preference for working less by job satisfaction, 2003, per cent Job satisfaction*
Prefer to work fewer hours
0–4
68
5
62
6–7
52
8
46
9–10
36
n
(1198)
* 0 = extremely dissatisfied; 10 = extremely satisfied SOURCE The Australian Survey of Social Attitudes 2003
around a third of those expressing the highest levels of job satisfaction. The reluctance of many with high job satisfaction to work fewer hours suggests that Hochschild’s ‘time bind’ may well be closer to the mark for those Australians than a simple ‘work/life collision’.
FAMILY VALUES AND WORK/LIFE BALANCE Many will be surprised at our conclusion that the day-to-day pressures of managing family life when children are young and both parents work make only a small difference to people’s perceptions of a ‘work/life collision’, and no difference to their desire to work fewer hours. But do differences in attitudes to parenting and domestic responsibilities affect how people perceive a ‘work/life collision’? When assessing their work/life balance, men and women may see the advantages and disadvantages of paid work for parents with children differently now that women are more likely to work. AuSSA 2003 gives us the opportunity to examine how people’s experiences of stresses of work and home might relate to general attitudes about the best ways to share family responsibilities. It may be, for example, that if people are able to establish family/work arrangements that are consistent with their general views about how family responsibilities should be shared, then they are less likely to perceive a ‘work/life collision’ or to prefer shorter hours, even if the demands of paid work and care responsibilities are great. AuSSA 2003 also asks whether a father should be as heavily involved in the care of his children as the mother. Fully 90 per cent of respondents agree, many very strongly. Only 2 per cent disagree. This suggests that Australians have completely egalitarian views about responsibilities for children, a further development from an already high level found in
AusSocAtts03
26/7/05
58 •
8:55 AM
Page 58
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
the 1980s (Bittman & Pixley 1997, pp. 118–20). However this unanimity is not maintained either in behaviour (see chapter 2) or in attitudes to the care of preschool children: responses to the statement ‘A preschool child is likely to suffer if mother works’ are evenly divided (see table 4.10). Note that the question asks about a mother who undertakes any work other than child care, and not about excessive hours of paid work. Thus we can infer a conservative view on the part of the 37 per cent who agree and a more egalitarian view for those 38 per cent who disagree. A quarter of the respondents are undecided. But how does this division relate to people’s views of whether there is a ‘work/life collision’ about their hours of paid work? Table 4.10 shows that women tend to hold views consistent with their behaviour. Only a quarter of women in paid employment agree that a preschool child is likely to suffer if their mother works, compared to 44 per cent of those not working. And among working women who have most choice about their work situation – those in managerial jobs – only 19 per cent take this conservative view, compared to 33 per cent of women in other occupations. Men are generally more conservative than women on this question: 45 per cent of men agree that preschool children of working mothers suffer, compared to 31 per cent of women. Interestingly, men’s work situation makes little difference to their attitudes – the greatest difference is between men in lower skilled occupations (45 per cent agree) and men whose partners work full-time (37 per cent agree). Women in paid work, then, may be relatively invulnerable to the argument that they should reduce their hours to conform to traditional ideas about gender roles. The perception that work hours interfere with family life clearly depends on comparing experience with a counterfactual ‘ideal’. Women who believe that mothers of preschool children should not work will see any paid work as interfering with family life. But we have found that men are more inclined to this view. Most women do not share this belief, so that their expectations about ‘proper’ family life allow for children to be looked after by others while they work. For these women, work only interferes with family and personal life when it takes up too much time, just as it does for women without children and for men. However, table 4.10 suggests that many men with working wives do not share their partners’ liberal views about paid employment for mothers with young children. We suspect that if AuSSA 2003 had asked men with working wives whether their wives should work fewer hours, the presence of small children would show up a large difference
AusSocAtts03
26/7/05
8:55 AM
Page 59
HOW DO AUSTRALIANS FEEL ABOUT THEIR WORK?
• 59
Table 4.10 ‘A preschool child is likely to suffer if mother works’ by gender, labour force status and occupation, 2003, per cent Agree
Neither agree nor disagree
Disagree
Can’t choose
Women Working full-time
21
21
57
0
Working part-time
26
20
54
0
Household duties
44
25
29
3
In household with children
29
20
49
2
Managers and administrators
19
20
60
1
Lower skilled
33
23
42
2
(n = 2208)
31
23
45
1
40
24
35
1
Men Working full-time Working part-time
44
28
27
2
Managers and administrators
43
26
30
1
Lower skilled
45
25
27
3
With a partner who works full-time
37
24
38
1
(n = 1950)
45
24
30
1
(n = 4158)
37
23
38
2
SOURCE The Australian Survey of Social Attitudes 2003
in attitudes. The point is that most men believe that it is women who should reduce their hours of work when work and family strains become excessive.
CONCLUSION Despite a decade or two of rapid workplace and labour market change, most Australians still get satisfaction from their jobs. Even so, people often feel under pressure to work harder and many are worried about their job security and lack confidence about their chances in the new labour market. Still, there are some important differences: people holding jobs requiring limited training tend to feel less secure in their jobs and more concerned about their ability to find another, older people are less confident about their marketability, and so on. But those in better jobs sometimes express the most negative responses, as when managers show the greatest propensity to feel under pressure to work harder. All in all, workers who express a high level of job satisfaction are not identified by their age or occupation, but by the quality of their workplace experiences.
AusSocAtts03
26/7/05
60 •
8:55 AM
Page 60
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
We do not read these results as challenging the reality of the cleavages identified by researchers who focus on other, more ‘objective’ variables like pay, conditions of employment and risk of unemployment. Rather, our findings remind us that the way people experience work is also important. In particular, we again emphasise that the day-to-day experience of work is affected by key relationships in the workplace, such as those between managers and employees, and by features of work organisation, such as whether workers can use their skills or whether they are under pressure to work harder. These matters vary from workplace to workplace, and employers can clearly choose to change them. American research using a very different approach has come to a similar conclusion (Hodson 2001). Our consideration of attitudes to working time and its effect on private life shows clearly that long hours can have negative effects. As we have emphasised, the ‘time bind’ best describes the experiences of Australians who work long hours, regardless of their family status. The implications of our findings for achieving a better work/life balance, however, turn out to be complex and equivocal. Australians clearly make trade-offs when deciding their work preferences, and these may involve many factors and are intimately tied to their values and beliefs about family and work. While better policies and family-friendly workplaces will make for better trade-offs, as Pocock (2003) suggests, they are unlikely to remove some real dilemmas. As Hochschild (1997) has emphasised, these arise from the fact with which we began – that many people gain great satisfaction from their jobs. NOTES
1
2
3
AuSSA 2003 reflects the overall Australian pattern in that about 68 per cent of respondents aged 18–65 are in paid employment of some kind. The vast majority of the remainder live in households where at least one person is in paid employment, or looking for it. The results we report do not control for any variables not mentioned. However, in multivariate analyses, all relationships we discuss are robust, statistically significant, and of similar magnitude to those shown in the tables, unless we explicitly indicate otherwise. Full multivariate analyses are available from the authors on request. There is no indication that confining attention to men who have younger children makes any difference to this pattern, though the numbers in AuSSA 2003 are too small to be certain.
REFERENCES
Australian Bureau of Statistics (ABS) (1997) Australian Standard Classification of Occupations, 2nd edn, Australian Bureau of Statistics, Canberra. —— (2002) Retrenchment and Redundancy, Australia, cat. no. 6266.0, Australian Bureau of Statistics, Canberra.
AusSocAtts03
26/7/05
8:55 AM
Page 61
HOW DO AUSTRALIANS FEEL ABOUT THEIR WORK?
• 61
Bittman, M & Pixley, J (1997) The Double Life of the Family, Allen & Unwin, Sydney. Brenner, R (1998) ‘The economics of global turbulence: A special report on the world economy, 1950–1998’, New Left Review, 229, May–June: 1–265. Gibson, R, Wilson, S, Meagher, G, Denemark, D & Western, M (2004) Australian Survey of Social Attitudes 2003, Data file, Australian Social Science Data Archive, The Australian National University, Canberra. Hancock, K (2002) ‘Work in an Ungolden Age,’ in Ron Callus & Russell D Lansbury, Working Future: The Changing Nature of Work and Employment in Australia, Federation Press, Leichardt. Hochschild, AR (1997) The Time Bind: When Work Becomes Home and Home Becomes Work, Metropolitan Books, New York. Hodson, R (2001) Dignity at Work, Cambridge University Press, New York. Mathews, J (1989) Tools of Change: New Technology and the Democratisation of Work, Pluto Press, Sydney. Moody, K (1997) Workers in a Lean World: Unions in the International Economy, Verso, New York. Organisation for Economic Cooperation and Development (OECD) (2004) OECD Employment Outlook, Organisation for Economic Cooperation and Development, Paris. Pocock, B (2003) The Work/Life Collision: What Work Is Doing to Australians and What to Do About It, Federation Press, Leichhardt. Sennett, R (1998) The Corrosion of Character: The Personal Consequences of Work in the New Capitalism, WW Norton, New York. Sousa-Poza, A & Sousa-Poza A (2000) ‘Well-being at work: A cross-national analysis of the levels and determinants of job satisfaction’, Journal of SocioEconomics, 29(6): 517–38. Tiffen, R & Gittins, R (2004) How Australia Compares, Cambridge University Press, Cambridge. Watson, I, Buchanan, J, Campbell, I & Briggs, C (2003) Fragmented Futures: New Challenges in Working Life, Federation Press, Leichhardt. Wooden, Mark (2002) ‘The changing labour market and its impact on work and employment relations’, in R Callus & RD Lansbury (eds) Working Futures: The Changing Nature of Work and Employment in Australia, Federation Press, Leichhardt.
AusSocAtts03
26/7/05
8:55 AM
Page 62
CHAPTER 5
V O L U N TA R Y A S S O C I AT I O N S A N D P O L I T I C A L PA R T I C I PAT I O N A n d r e w P a s s e y a n d M a r k Ly o n s
INTRODUCTION Do many Australians belong to voluntary associations? Does their membership have any impact on their participation in politics? Would it affect the way Australia is governed if associational membership dramatically declined? These are interesting questions that, surprisingly, have seldom been considered in this country. This chapter draws on the Australian Survey of Social Attitudes 2003 (Gibson et al. 2004) to consider the links between membership of the voluntary or non-profit associations that are so common in Australian social life and different kinds of political activity. Our focus, then, is on behaviour rather than attitudes. We explore two sets of behaviour: membership of voluntary associations and participation in a variety of forms of political activity (other than voting in national, state or local elections). Voluntary association is one of many terms used to describe organisations that are formed and joined by people acting voluntarily and collectively to serve their own interests or those of others, without seeking personal financial benefit. Other terms include non-profit or not-forprofit organisation or (to encompass a slightly wider grouping) third sector, civil society or social economy organisations. Voluntary associations are central to Australian life. It has been estimated that the nation has as many as 700 000 voluntary associations, more than 300 000 of which have obtained a separate legal identity via incorporation. Most are sustained entirely by the voluntary commitment of time by at least some of their members. Only about 35 000 associations employ staff, and the
AusSocAtts03
26/7/05
8:55 AM
Page 63
V O L U N TA R Y A S S O C I AT I O N S A N D P O L I T I C A L PA R T I C I PAT I O N
• 63
large majority of these are small (Lyons & Hocking 2000; Lyons 2001). Nonetheless, voluntary associations are of considerable economic importance. In 2000, they had over 600 000 employees and contributed 3.3 per cent to the nation’s gross domestic product (GDP). This is a similar economic contribution to that of the agricultural industry. When the value of voluntary work for all non-profit associations is added in, the contribution to GDP becomes 4.8 per cent, equal to that of the mining industry (ABS 2002). Voluntary associations provide the majority of community and sport and recreation services in Australia. They are major service providers in education, health care and the arts. They are ubiquitous in the representation of interests and religious worship. In 2000, almost 28 per cent of the population performed voluntary work for associations (ABS 2002). In 1997, just under two-thirds of the population made a financial donation to a voluntary association. More importantly, the same percentage of the population belonged to at least one voluntary association (Lyons & Hocking 2000). The most basic form of political activity in Australia is voting in national, state or territory and local government elections. But there are many other ways that individuals or groups engage in political activity. These include writing letters to, or speaking with, bureaucrats, MPs or ministers, or joining a demonstration. They also include lobbying government to act against particular firms or nations, or boycotting certain products to directly encourage firms to change. While voting is compulsory, these other political behaviours and who engages in them interests us here. This is important because participation in these activities is an additional measure of the strength of our democracy. In this context, any relationship between voluntary association and political activity becomes significant for democracy. Does being a member of an association make a difference to people’s levels of political activity? And does the kind of association matter?
V O L U N TA R Y A S S O C I AT I O N S A N D P O L I T I C S Voluntary associations contribute to the political system of democratic polities in two ways. Some associations are formed solely or primarily to lobby governments for a particular set of interests. These may be the interests of their members or of others. Other groups are formed with interest representation or advocacy as one of several goals. We call both these types of organisations ‘political’ associations. Political associations include political parties, only a few of which can realistically hope to
AusSocAtts03
26/7/05
64 •
8:55 AM
Page 64
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
form a government, as well as lobby groups and those campaigning on particular issues. However, the great majority of associations are formed to pursue non-political interests: to educate children, to support the chronically ill, to organise competition in a particular sport, to enable gardeners to swap exotic plants and so on. From time to time, a few of these associations may engage in political advocacy, but in general they do not. But these non-political associations, along with political associations, contribute to a democratic political system in another way – by encouraging and facilitating the participation of their members in the wider civic system. Voluntary associations do this by engaging members in relationships and concerns beyond the family or workplace; by providing members with a place to discuss politics; and by providing members with experience in organising, in working with others, in running meetings and in the art of compromise (Verba, Schlozman & Brady 1995). The role of voluntary associations in promoting political participation, and thus in underpinning democratic polities, was famously described in Almond and Verba’s Civic Culture (1965). In their comparative study of five countries (United States, United Kingdom, West Germany, Italy and Mexico), Almond and Verba sought to identify the characteristics of political culture most conducive to democracy. They found them in what they called a participant political culture, which was strongest in the United States and the United Kingdom. One of the features of a participant political culture was a high membership and active membership in voluntary associations. Fiftyseven per cent of the United States’ adult population belonged to at least one association, compared to 47 per cent in the United Kingdom and 44 per cent in Germany. In the United States, membership was far higher in religious organisations and in what Almond and Verba called civic political groups and lower in unions than in the United Kingdom. In most countries, 40 per cent to 45 per cent of members claimed that their organisation was engaged in political activity. These were described as ‘political’ organisations. From Almond and Verba we draw four main pointers for our Australian analysis. First, they found that in high-participant countries, membership in voluntary associations was highly correlated with level of education. In the United Kingdom, for example, while only 41 per cent of people with a primary school education belonged to a voluntary association, 92 per cent of those with a university education did. Second, in all countries members of voluntary associations were more confident in
AusSocAtts03
26/7/05
8:55 AM
Page 65
V O L U N TA R Y A S S O C I AT I O N S A N D P O L I T I C A L PA R T I C I PAT I O N
• 65
their ability to influence governments than non-members. This is what later political scientists have called ‘internal political efficacy’ (McAllister & Wanna 2001). Third is the distinction between political and non-political organisations, because internal political efficacy was found to be higher among members of political associations than among members of non-political organisations. Fourth, Almond and Verba found that active members of associations (which they defined as being or having been an officeholder) were markedly more confident in their ability to influence government. This increase in confidence held true for all levels of educational attainment, although it was greater among the less well educated. People who were members of two or more associations were also more likely to have a sense of political efficacy. Subsequent studies took the inequalities in American democracy as their starting point. Research found that people from high socioeconomic status (SES) groups were far more likely to participate in political activities and to express feelings of political efficacy than people from low SES groups. Active associational membership went some of the way to restoring the balance, because it made lower SES respondents more likely to participate in political activity. What made the difference was exposure to political discussion or activities in the organisation, and collective effort to address some community problem (Verba & Nie 1972). Later research pointed to other factors, such as workplace arrangements and practices, which also contributed to increased participation by low SES groups, but confirmed the importance of voluntary associations. Membership of religious congregations was shown to be particularly important in boosting political participation by African Americans (Verba, Schlozman & Brady 1995). The 1990s saw an expanded interest in the ways voluntary associations contributed to a strong society as well as to representative democracy. It was claimed that they helped build trust in society, and in this way, and via the networks they constituted, that they contributed to the production of social capital. A decline in associational membership, therefore, represented a threat to social and economic efficiency (Putnam 1993; 1995; 2000; see also chapter 8). The collapse of communist governments also sparked an interest in the contribution of voluntary associations to building new democracies and in contributing to civil society. Initially, the role of non-political associations was emphasised – Putnam (1993), for example, celebrated North Italy’s soccer clubs and choral groups. But soon some criticised the overemphasis on these groups to the neglect of explicitly political associations (Foley &
AusSocAtts03
26/7/05
66 •
8:55 AM
Page 66
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Edwards 1996). Other researchers drew attention to the different capacities of different types of associations to contribute to social capital, civic engagement or political activity (Stolle & Rochon 1998; Eastis 1998; Warren 2001). These are interesting and, arguably, important questions, but they have attracted little interest from Australian researchers. Few studies have even recognised that many Australians belong to voluntary associations. Apart from electoral studies, there has been little research exploring political participation. In their pioneering 1967 survey, Aitkin and Kahn (1974) noted that Australian associational membership was as high as that in the United States (citing Almond & Verba), but did not explore the implications of this for political activity. Kemp (1978) repeated the finding. Curtis, Grabb and Baer (1992) included Australia in a 15-country study of voluntary association membership, drawing on data from World Values Surveys in the early 1980s. They showed Australia with a membership rate of 61 per cent, dropping to 40 per cent when union and church memberships are excluded. This was at the high end of countries in the study. Occasional studies have reported levels of membership in particular kinds of associations: Peetz (1998), for example, studied unions, while Crook and Pakulski (1995) studied environment groups. Others have explored political participation by particular social groups: Bean (1991), for example, studied women; Pusey (2003) studied ‘middle Australia’, and Vromen (2003) studied Generation X. But, so far, there is no research on associational membership generally, and its impact on political participation. This question is central to the discussion that follows.
W H O J O I N S V O L U N TA R Y A S S O C I AT I O N S ? AuSSA 2003 shows that Australia is a nation of joiners. In table 5.1, we count associational membership for individual respondents (the full list is presented in table 5.2). We find that most AuSSA 2003 respondents belong to at least one voluntary association. Indeed, nearly half (48 per cent) belong to three or more associations, while 22 per cent belong to two associations, and 16 per cent belong to only one association. Just 14 per cent do not belong to any voluntary associations. We find that men and women are equally likely to belong to voluntary associations, and there is little variation by age (although those aged 65 and over are slightly less likely to be members) or by location (people living in rural areas are equally as likely to belong to at least one voluntary association as those living in cities). However, membership varies
AusSocAtts03
26/7/05
8:55 AM
Page 67
V O L U N TA R Y A S S O C I AT I O N S A N D P O L I T I C A L PA R T I C I PAT I O N
• 67
Table 5.1 Membership of voluntary organisation by demographic group, 2003, per cent Sex Female
86
Male
87
Age 18–34
86
35–49
88
50–64
88
65 and over
82
Education Year 12 or less
80
Trade/certificate or diploma
90
Bachelor’s degree and above
95
Occupation Manager/professional
94
Associate professional/trades
87
Advanced/intermediate clerical
86
Elementary clerical and labourers
79
Income $0 to $31 199 per year
79
$31 200 to $77 999 per year
90
$78 000 and over per year
93
Total (n = 2057)
86
NOTE Membership includes all respondents who are members of one or more organisations. SOURCE The Australian Survey of Social Attitudes 2003
markedly with income, level of education and occupational status. As table 5.1 shows, members of voluntary associations are more likely to be found among those who earn $78 000 or more per year, among those who have completed a university degree and among those in professional or managerial occupations. We discuss the importance of these differences later. Not surprisingly, as table 5.2 shows, some associations are far more popular than others. Membership in consumer or automobile associations, such as the NRMA or RACV, is the most typical reported by our respondents (54 per cent). The next most popular are sporting or recreation groups, of which 45 per cent of respondents are members. Almost one in three report being a member of a financial co-operative, such as a credit union, and 24 per cent are members of a religious organisation. At the other end of the scale, around 5 per cent or fewer reported
AusSocAtts03
26/7/05
68 •
8:55 AM
Page 68
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 5.2 Membership by type of voluntary association, 2003, per cent Consumer or automobile organisation
54
A sporting or recreation group (including registered clubs and service clubs)
45
Financial co-operatives (e.g. credit unions)
29
A religious group
24
A neighbourhood or community-based group
20
A union
19
Professional society
15
A group that helps people with special needs
14
An environmental group or aid organisation (e.g. Greenpeace, Oxfam)
10
Art, music or educational group
10
A self-help/consumer health group
8
A group working to improve the environment (e.g. Landcare)
7
A group that promotes rights (e.g. of women, refugees)
6
Lobby group to change specific government policies
4
A political party
4
(n≥ 2012) SOURCE The Australian Survey of Social Attitudes 2003
membership of groups promoting rights, lobby groups or political parties. There are two categories of associations that to some may not at first appear ‘voluntary’. These are unions and professional societies, of which 19 per cent and 15 per cent of respondents respectively report membership. However, with the almost total decline of closed shops, union membership can be treated as a voluntary decision (Peetz 1998). Professional associations are slightly different, since it is not possible to practise certain professions without being a member of the appropriate association. But such closed shops are relatively few in number. As well, analysis of the data reveals that only nine members of a professional association report no other associational membership. Hence while their membership of a professional society might not be entirely ‘voluntary’, almost all of these members are also members of another kind of voluntary association.
HOW ACTIVE ARE MEMBERS? The research we reviewed at the beginning of this chapter suggests that active members of associations are more confident in their ability to influence government than ordinary members or non-members. AuSSA 2003 distinguishes between these different groups of members by asking people to choose between three categories: member, active member and
AusSocAtts03
26/7/05
8:55 AM
Page 69
V O L U N TA R Y A S S O C I AT I O N S A N D P O L I T I C A L PA R T I C I PAT I O N
• 69
officeholder.1 Many members are content to pay their membership dues and enjoy the benefits that membership brings, such as access to (often subsidised) services or socialising. Here we call them ordinary members. Some members, however, may recognise that they need to be active in an association if it is to continue to operate. These members effectively perform a range of volunteer tasks for the association. We call them active members. There is a further group of ‘active’ members who take legal responsibility for their association, and perform governance activities by acting as officeholders. Unsurprisingly, we find that the majority of memberships (74 per cent) involve people acting as ordinary members, 20 per cent involve people acting as active members and 6 per cent involve people serving as officeholders. However, there are marked variations between different types of association (see table 5.3). The proportion of active members is over 30 per cent among art, music or educational groups, groups helping people with special needs, sporting or recreation groups, and reliTable 5.3 Voluntary membership by level of member commitment (ranked by per cent of active members) Active members
Officeholders
Ordinary members
Art, music or educational group
39
7
A group that helps people with special needs
38
9
54 53
A sporting or recreation group (including registered clubs and service clubs)
34
12
54
A religious group
34
11
55
Lobby group to change specific government policies
27
9
65
A group that promotes rights (e.g. of women, refugees)
26
10
65
A group working to improve the environment (e.g. Landcare)
25
6
69
A neighbourhood or community-based group
23
10
68
Professional society
22
6
73
A self-help/consumer health group
18
2
80
A political party
16
4
81
An environmental group or aid organisation (e.g. Greenpeace, Oxfam)
14
3
83
A union
86
10
3
Financial co-operatives (e.g. credit unions)
7
0
94
Consumer or automobile organisation
5
0
95
(n≥ 2012) SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
70 •
8:55 AM
Page 70
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
gious groups. These seem to be more ‘mass participation’ associations. It is likely that most of these organisations will rely solely on the efforts of their members in sustaining their activities. For the few associations that employ staff, there will be limited opportunities (and need) for members to become active as volunteers except as officeholders, and in large associations there are few such positions. Examples include consumer or automobile organisations, financial co-operatives, and unions, where 10 per cent or fewer of members report being active. Those who report being active members of financial co-operatives and consumer or automobile organisations are likely to be those people who vote in annual general meetings. Active membership of a union will involve taking part in meetings and holding union office in the workplace. However, it is clear that in these types of association active membership and officeholding is left to a relatively small cadre.
A R E M E M B E R S O F V O L U N TA R Y A S S O C I AT I O N S M O R E T R U S T I N G ? Although our main focus is on behaviour (membership of voluntary associations and political activity), much of the research we discussed above has examined the links between membership and political efficacy. The AuSSA 2003 questionnaire contains a small number of questions similar to measures of political efficacy, including trust in other people (often called social trust) and confidence in federal parliament (often Table 5.4 Two measures of trust by association membership, 2003, per cent Members
Nonmembers
Total
Confidence in federal parliament A great deal of confidence
4
4
4
Quite a lot of confidence
37
27
36
Not very much confidence
43
42
42
No confidence at all
13
21
14
3
6
Can’t choose n
4
(1738)
(276)
(2014)
Can be trusted
42
25
40
Can’t be too careful
53
68
55
5
7
5
Most people can be trusted …
Can’t choose n SOURCE The Australian Survey of Social Attitudes 2003
(1726)
(268)
(1994)
AusSocAtts03
26/7/05
8:55 AM
Page 71
V O L U N TA R Y A S S O C I AT I O N S A N D P O L I T I C A L PA R T I C I PAT I O N
• 71
seen as a measure of political trust). Like previous research, we do find that members of voluntary associations express higher levels of both types of trust, with 42 per cent agreeing that people can generally be trusted and 41 per cent having a great deal or quite a lot of confidence in the federal parliament (see table 5.4). Among non-members, by contrast, only 25 per cent express social trust, while 31 per cent express political trust. Chapter 8 looks at trust and participation in more detail.
WHICH AUSTRALIANS ARE POLITICALLY ACTIVE? Unlike much prior research, AuSSA 2003 asks people directly if they have undertaken certain types of political activity in the previous two years. This is because we are interested in the level and range of people’s political behaviour, rather than in their perceptions of whether this behaviour influences political processes (cf. Almond & Verba 1965). Some international comparisons use voter turnout in elections as a measure of political activity. These studies tend to show that in postindustrial countries similar to Australia voter turnout is at best stagnant and at worst falling. In Australia, voting is compulsory, so voter turnout is not a relevant measure. Instead, we need to measure other forms of political action to assess the level of political engagement. The four measures we use include formal and informal activity, some of which will be undertaken as individual citizens, some as part of groups of citizens. Overall, we find that 62 per cent of respondents have undertaken at least one of these activities in the past two years. The most common activity is of the exercise of consumer power – half the respondents say that they have either boycotted or bought particular products (see table 5.5). Some consumers exercise their consumer power positively, by purchasing ‘ethical’ products, such as those not tested on animals or by purchasing products from a particular Table 5.5 Political participation in previous two years, 2001–2003, per cent* Boycotted or bought certain products
50
Contacted a politician or government official
31
Taken part in a protest, march or demonstration
12
Participated in a strike or industrial action (n≥ 4156) SOURCE The Australian Survey of Social Attitudes 2003
8
AusSocAtts03
26/7/05
72 •
8:55 AM
Page 72
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 5.6 Number of political activities in previous two years, 2001–2003, per cent None
38
One
35
Two
20
Three
6
Four
2
(n = 2087) SOURCE The Australian Survey of Social Attitudes 2003
country or group of people. The negative exercise of consumer power in the form of boycotts may also be targeted in these ways. Some consumers, for example, refused to buy French cheese and wine during French nuclear testing in the Pacific. One in three respondents reports having contacted a government official or politician. The least common forms of political activity are in some ways the most costly (in terms of time and effort, and potentially financially and legally) and include protests (12 per cent of respondents) and strike action (8 per cent).2 We find, too, that participating in some form of political activity is the norm for Australians. Table 5.6 shows that only 38 per cent report having done none of these activities in the past two years. About one in three people have participated in one activity, and one in five have participated in two activities. More than 7 per cent of people have either done three or four activities – these are people who express their political interests in several different ways.
I S T H E R E A R E L AT I O N S H I P B E T W E E N MEMBERSHIP AND POLITICAL ACTIVITY? Previous research suggests that, in many nations, members of voluntary associations are more likely to take part in political activity than people who are not members. International evidence also suggests that membership of particular types of voluntary association correlates even more strongly with political activity. What does the evidence show for Australia? First, we find that members of associations are more likely than nonmembers to have engaged in political activity. Table 5.7 shows that 66 per cent report doing so, compared to fewer than 40 per cent of nonmembers. The more memberships people have, the more likely they are to be engaged in political activity. For example, just over half of those with just one association membership (51 per cent) have engaged in
AusSocAtts03
26/7/05
8:55 AM
Page 73
V O L U N TA R Y A S S O C I AT I O N S A N D P O L I T I C A L PA R T I C I PAT I O N
• 73
political activity, compared to 74 per cent of those who are members of three or more associations. In Australia, therefore, as in the United States, multiple memberships seem to be important in bringing people into political activity, either as recorded in people’s views of their political efficacy or through their own reports of political actions. Second, what about the level of commitment of members – might this have some kind of relationship with their political participation more widely? Table 5.8 compares rates of political activity among nonmembers of any voluntary association to ordinary members and active members and/or officeholders. The pattern is clear. Members, especially those who class themselves as active or as officeholders, are much more likely to be engaged in political activity compared to non-members. These people are active not just in their associations, but also more broadly in civic life. We conclude that membership has an effect on polit-
Table 5.7 Membership and political activity, 2001–2003, per cent Members
66
Non-members
39
Respondents with only one associational membership
51
Respondents with three or more associational memberships
74
SOURCE The Australian Survey of Social Attitudes 2003
Table 5.8 Political activity by level of member commitment, 2001–2003, per cent Nonmembers (n≥ 276) Contacted a politician or government official Taken part in a protest, march or demonstration Boycotted or bought certain products Participated in a strike or industrial action SOURCE The Australian Survey of Social Attitudes 2003
16
Ordinary Active members members/ (n≥ 1578) officeholders (n≥ 823) 34
42
5
14
17
32
55
55
3
9
9
AusSocAtts03
26/7/05
74 •
8:55 AM
Page 74
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
ical activity (although we cannot claim that it is the cause of political activity); and that active members and/or officeholders are more politically active than are ordinary members. So far, the evidence from Australia matches that from overseas. However, there is another important ingredient. International evidence suggests that, not only are more active members of voluntary organisations more likely to be politically active, but also that the kind of association people belong to makes a difference to their level of political activity. We explore this in the Australian case by comparing political activity rates of non-members with those of members for each of our associational categories. Here we make no distinction in member commitment; instead we are interested in the kind of association. We found that being a member of particular categories of association means that people are more likely to engage in political action (see table 5.9). This is most marked in associations geared towards particular causes, especially the environment and aid; or those that more broadly articulate and promote rights, including lobby groups and political parties. Among members of these types of association, more than 80 per cent of respondents have been engaged in political activity – these members are highly active in civic life. However, only a small proportion of respondents belong to such associations – typically 10 per cent or fewer. In a sense, these associations are specialised – relatively few people belong to them, but those who do report high political activity rates, leading us to speculate that these are the Australian equivalent of the ‘political associations’ that Almond and Verba identified in the United States. In contrast, respondents who are members of the kinds of associations that more people commonly join (such as financial co-operatives and sporting or recreation groups) are less likely to participate in political action. The same is true of professional associations, despite their being set up to promote specific interest groups. However, even in these types of association, members are more likely to be politically active compared to non-members. We know from international research that membership of religious organisations is strongly correlated with political participation, especially in the United States. In Australia, however, the evidence suggests no strong relationship, as the last row of table 5.9 shows. We can see that the proportion of members of religious organisations engaged in political activity is only 3 per cent higher than among non-members (this is the only difference that is not statistically significant). Explaining the difference between Australia and the United States requires further
AusSocAtts03
26/7/05
8:55 AM
Page 75
V O L U N TA R Y A S S O C I AT I O N S A N D P O L I T I C A L PA R T I C I PAT I O N
• 75
Table 5.9 Political activity ranked by association type, 2001–2003, per cent Member
Nonmember
An environmental group or aid organisation
89
59
A group that promotes rights
89
61
Lobby group to change specific government policies
89
62
A group working to improve the environment
87
61
A political party (e.g. Labor, Liberals, Greens)
83
62
Consumer or automobile organisation
71
52
A self-help/consumer health group
77
61
Art, music or educational group
75
61
A union (e.g. CFMEU, TWU, ASU, CPSU)
74
60
A group that helps people with special needs
74
61
A neighbourhood or community-based group
73
60
Financial co-operatives
71
60
Professional society
71
61
A sporting or recreation group
67
59
A religious group
65
62
(n = 2012) SOURCE The Australian Survey of Social Attitudes 2003
research, which would need to consider variation in the religious denominations between the two countries, as well as differences in rates of religious attendance.
W H AT A R E T H E I M PA C T S O F M E M B E R S H I P ON POLITICAL ACTIVITY? So far, we have found that members of voluntary associations are more likely to engage in political activity, and that the level of commitment of members, and the types of associations to which they belong, help to shape this interaction. But might not these effects be the consequence of other factors, such as higher education or a professional or managerial job? Can we more precisely measure the impacts of membership? To help us answer this question, we apply more sophisticated statistical testing in the form of regression modelling. Multiple regression allows us to determine the effect of several independent variables on the variable we want to explain, the so-called dependent variable. In our case, independent variables include membership of voluntary associations, age, gender, income, occupation and education, and the dependent variable is political activity.3 As table 5.10
AusSocAtts03
26/7/05
76 •
8:55 AM
Page 76
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 5.10 Comparing the impact of membership on political activity, 2001–2003 Odds ratio
Member of a voluntary association
2.66**
Qualified with Bachelor’s degree or higher
1.75**
Managerial or professional occupation
1.39**
(n = 2012) ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
reveals, three factors had significant and independent impact on the likelihood of political activity: membership of a voluntary association, level of education, and occupation. Because regression modelling shows that each has a separate impact, we can discuss their individual effects. We measure impact using an odds ratio, which expresses how much a variable contributes to the probability of taking the action expressed by the dependent variable (in our case, engaging in political activity). So, in our model, membership of a voluntary association had the biggest impact, increasing by about 2.7 times the likelihood of a person engaging in political activity. Having a Bachelor’s degree or higher increased the likelihood of political action around 1.8 times, and being in a managerial or professional occupation by 1.4 times. These latter two characteristics have been identified in other nations as important in increasing political activity and efficacy. While we found them to be important predictors of political action here in Australia, they are less important than membership of voluntary associations. Clearly, being a member of a voluntary association is the strongest predictor of political action. But do different types of association have any differential impacts on political action? Table 5.11 suggests that being a member of an environmental group or aid organisation increases almost four-fold the chances of a person taking part in political activity – an impact that is markedly larger than membership of more strictly ‘political’ organisations, such as groups promoting rights, lobby groups or political parties. Perhaps joining one of these ‘political’ associations is itself viewed as a political act by their members, when compared to membership of an environmental or aid organisation. That said, membership in these strictly ‘political’ associations remains among the most significant impacts on political activity. One further interesting finding is the impact of membership in a consumer or automobile organisation, which almost doubles the likelihood of political action when compared to all respondents, even after
AusSocAtts03
26/7/05
8:55 AM
Page 77
V O L U N TA R Y A S S O C I AT I O N S A N D P O L I T I C A L PA R T I C I PAT I O N
• 77
taking into account their membership of other associations. It would appear, then, that membership in these associations is not just about obtaining access to particular services or subscriptions to consumer magazines. We make this point partly because more than half of our respondents report membership in such associations, which means that the impact of membership on political action is likely to be wide ranging, simply because of the number of people involved. All this leads us to our final question. In Australia, are there combinations of association type and levels of member commitment that most impact on political activity? We cannot fully explore this question because of limitations in the data available to us. Accordingly, we focus on member commitment, and distinguish between people who hold some kind of active membership and/or officeholding position, and people who report only having ordinary memberships. We examine Table 5.11 Impact of membership on political activity by association type, 2001–2003 Odds ratio
Member of an environmental group or aid organisation
3.60**
Member of a group that promotes rights
2.58**
Member of a lobby group
2.49*
Member of a political party
2.40**
Member of a self-help / consumer health group
2.09*
Member of a group working to help improve the environment
1.91**
Member of a consumer or automobile organisation
1.86**
Qualified with Bachelor’s degree or higher
1.79**
Member of a union
1.68*
Member of a group that helps people with special needs
1.38**
(n = 2012) * p < 0.05; ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
Table 5.12 Impact of member commitment on political activity, 2001–2003 Odds ratio
Active member/officeholder* of a voluntary association
2.99
Ordinary member only of a voluntary association
2.43**
Qualified with Bachelor’s degree or higher
1.73**
Managerial or professional occupation
1.36**
(n = 2012) * Includes ordinary members of other voluntary associations; ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
78 •
8:55 AM
Page 78
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
whether the act of joining (of being an ordinary member) has less impact on political activity than being an active member. As in other countries, table 5.12 shows that active members of Australian voluntary associations are more likely to engage in political action than ordinary members, setting aside any educational or occupational differences between these groups. People who report being an active member or officeholder of a voluntary association are three times more likely to engage in political activity, compared to about 2.4 times for those reporting only ordinary membership(s). While both the possession of graduate level qualifications or being in high status occupations also increases this likelihood, they do so at a lower rate. Impacts on political activity therefore go beyond simply being a member of an association. Members who make the extra commitment needed to sustain their association are also more likely to make a commitment to participate in civic or political life.
CONCLUSION Australians are joiners – the rate of membership in voluntary associations in Australia is higher than for many other developed nations (Curtis, Grabb & Baer 1992). In Australia, members of voluntary associations are more likely to be involved in political action – not simply voting, but a broader range of activities. As the international research suggests, not all memberships are the same when it comes to assessing their impacts on political activity. In Australia, two important factors emerge, although it is likely that a combination of these increases the odds of political activity. First is the level of member commitment. Simply, more committed members are more engaged in politics. That much the international evidence suggested to us, especially the work of writers like Robert Putnam who lament the decline of active membership and its replacement by ‘cheque book’ membership. However, we should not over-egg this particular pudding. Our findings suggest that while active membership increases three-fold the likelihood of political activity, ordinary membership has a positive impact too, more than doubling the chances of political action. Second, the category of association has an impact – members of some types of association are more likely to be politically engaged than members of others, independent of any other characteristics these people might possess. In Australia this increase in impact is not confined to what some have termed ‘political’ associations – though they are impor-
AusSocAtts03
26/7/05
8:55 AM
Page 79
V O L U N TA R Y A S S O C I AT I O N S A N D P O L I T I C A L PA R T I C I PAT I O N
• 79
tant – but extends to environmental and aid organisations, and to selfhelp associations and those helping people with special needs, for example. It seems that certain associations do perform almost exclusively social functions, such as some clubs, societies and sporting organisations. Others we can identify as predominantly political – lobby groups and political parties, for example. However, many associations perform multiple functions – sometimes operating in the political sphere to represent their members’ interests, at other times providing these members with networking and socialising opportunities. We suggest, then, that trying to classify Australian associations into ‘political’ and ‘non-political’ organisations is not straightforward. Assessing how membership of voluntary associations might affect political activity across the nation is important at a time of heightened concern – raised in the media and elsewhere – about growing Australian disillusionment with, and disengagement from, politics and political processes (see, for example, Burchell & Leigh 2002). Our evidence suggests that membership of voluntary associations plays an important role in sustaining political engagement in Australia. Even the most common form of membership (consumer or automobile organisations) has a positive impact on the chances of a person engaging in political activity. Being a member is not simply about accessing services. Membership has wider consequences for Australian society, even though the majority of associations could not be described as overtly ‘political’. Instead, they seem to be acting as one route into forms of political action. At a time when there is evidence of a decline in the membership of many large traditional types of association, such as service clubs, unions and churches (Lyons 2001), the implications of these findings are important. If the proportion of Australians who claim membership in at least one association falls significantly, then we might expect levels of political participation to fall too. For those concerned with the quality of Australian democracy, this prospect warrants further study. NOTES
1
The questionnaire contains the following definitions: Being a member means you pay membership fees or a subscription or make donations, but you do not get any more involved. Being an active member means you are regularly involved in the group’s activities as well as paying membership fees or a subscription or making donations. Being an officeholder means you have a decision-making role in the group as well as participating in activities and paying your membership fees etc.
AusSocAtts03
26/7/05
80 •
2
3
8:55 AM
Page 80
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Strike action is obviously the preserve of workers who belong to unions, which raises the potential for circularity in our data. However, with the dismantling of the closed shop, this form of political activity is no longer compulsory for union members when a strike is called. Furthermore, examination of the data reveals that only 16 respondents report undertaking strike action and no other form of political activity, and hence any circularity is having only very marginal impacts on levels of political action (and as we see later, on the relationship between associational membership and political action). We began by entering all the independent variables (membership of voluntary associations, age, gender, income, occupation and education) and testing them for significance. This is the full model, and it allowed us to determine which independent variables were correlated with political activity, and which were not (see Appendix, table A5.1). Previous research suggested that a person’s age, gender and income were important determinants of the level of their political activity. However, our full regression model reveals that these factors were not significant in themselves. Instead, they were important only when they interacted with other characteristics. They were therefore excluded from our reduced model, which only included those factors identified as significant.
The authors gratefully acknowledge the statistical analysis undertaken by Ian Nivison-Smith for this chapter. Any errors are, however, the sole responsibility of the authors. REFERENCES
Aitkin, D & Kahn, M (1974) ‘Australia: Class politics in the New World’, in Richard Rose (ed.) Electoral Behaviour: A Comparative Handbook, The Free Press, New York. Almond, G & Verba, S (1965) The Civic Culture: Political Attitudes and Democracy in Five Countries, Little, Brown & Co, Boston. Australian Bureau of Statistics (ABS) (2002) Non-Profit Institutions Satellite Account 1999–2000, cat. no. 5256.0, Australian Bureau of Statistics, Canberra. Bean, C (1991) ‘Gender and political participation in Australia’, Australian Journal of Social Issues, 26(4): 276–93. Burchell, D & Leigh, A (eds) (2002) The Prince’s New Clothes: Why Do Australians Dislike their Politicians?, UNSW Press, Sydney. Crook, S & Pakulski, J (1995) ‘Shades of green: Public opinion on environmental issues in Australia’, Australian Journal of Political Science, 30(1): 39–55. Curtis, JE, Grabb, EG & Baer, DE (1992) ‘Voluntary association membership in fifteen countries: A comparative analysis’, American Sociological Review, 57: 139–52. Eastis, CM (1998) ‘Organisational diversity and the production of social capital’, American Behavioral Scientist, 42(1): 66–77. Foley, MW & Edwards, B (1996) ‘The paradox of civil society’, Journal of Democracy, 7 (2): 38–52. Gibson, R, Wilson, S, Meagher, G, Denemark, D & Western, M (2004) Australian Survey of Social Attitudes 2003, Data file, Australian Social Science Data Archive, The Australian National University, Canberra. Kemp, DA (1978) Society and Electoral Behaviour in Australia: A Study of Three Decades, University of Queensland Press, Brisbane.
AusSocAtts03
26/7/05
8:55 AM
Page 81
V O L U N TA R Y A S S O C I AT I O N S A N D P O L I T I C A L PA R T I C I PAT I O N
• 81
Lyons, M (2001) Third Sector: The Contribution of Non-profit and Co-operative Enterprises in Australia, Allen & Unwin, Sydney. —— & Hocking, S (2000) Dimensions of Australia’s Third Sector, Centre for Australian Community Organisations and Management, University of Technology, Sydney, Lindfield. McAllister, I & Wanna, J (2001) ‘Citizens’ expectations and perceptions of governance’, in Glyn Davis & Patrick Weller (eds) Are You Being Served? State, Citizens and Governence, Allen & Unwin, Sydney. Peetz, D (1998) Unions in a Contrary World: The Future of the Australian Trade Union Movement, Cambridge University Press, Melbourne. Pusey, M (2003) The Experience of Middle Australia: The Dark Side of Economic Reform, Cambridge University Press, Melbourne. Putnam, R (1993) Making Democracy Work: Civic Traditions in Modern Italy, Princeton University Press, Princeton, New Jersey. —— (1995) ‘Bowling alone: America’s declining social capital’, Journal of Democracy, 6 (1): 65–78. —— (2000) Bowling Alone: The Collapse and Revival of American Community, Simon & Schuster, New York. Stolle, D & Rochon, TR (1998) ‘Are all associations alike? Member diversity, associational type, and the creation of social capital’, American Behavioral Scientist, 42(1): 47–65. Verba, S & Nie, NH (1972) Participation in America: Political Democracy and Social Equality, Harper & Row, New York. Verba, S, Schlozman, KL & Brady, HE (1995) Voice and Equality: Civic Voluntarism in American Politics, Harvard University Press, Cambridge, MA. Vromen, A (2003) ‘“People try to put us down …”: Participatory citizenship of “generation X”’, Australian Journal of Political Science, 38(1): 79–99. Warren, ME (2001) Democracy and Association, Princeton University Press, Princeton, New Jersey.
AusSocAtts03
26/7/05
8:55 AM
Page 82
CHAPTER 6
A R E P O S T M AT E R I A L I S T S ENGAGED CITIZENS? M a r k We s t e r n a n d B r u c e Tr a n t e r
INTRODUCTION In June 2004, with an election due later in the year, then federal Labor leader, Mark Latham, endorsed and then secured preselection for Peter Garrett in the safe Labor seat of Kingsford Smith, despite opposition from local branch members. As well as being President of the Australian Conservation Foundation, Garrett had for 20 years been the lead singer of Midnight Oil whose iconic music dealt with the environment, human rights, indigenous issues and America’s military presence in Australia. Even the Prime Minister, when asked about Garrett’s move to Labor, replied that his favourite Midnight Oil song was ‘Beds are Burning’– a song the band played at the closing ceremony of the 2000 Olympic Games in Sydney and that calls for returning Aboriginal land. At the time of the Labor leader’s announcement, Garrett was not even a member of the Labor Party, but went on to win his seat with 59 per cent of the two-party preferred vote – a Labor victory that defied the national swing to the Liberal–National Coalition of almost 2 per cent. Why would the Labor Party leader invite opposition from his own party to support a high-profile environmental activist and popular musician with no experience in Labor politics? And why would the Prime Minister feel obliged to show some familiarity with Midnight Oil’s music and its activist stance? Some clues are found in Australian Election Studies (AES), which show that, since 1990, more than 80 per cent of voters have said that the environment was an ‘extremely important’ or
AusSocAtts03
26/7/05
8:55 AM
Page 83
A R E P O S T M AT E R I A L I S T S E N G A G E D C I T I Z E N S ?
• 83
‘quite important’ issue in the election (see figure 6.1). In 1998 and 2001, AES findings show that even more voters identified the environment as an election issue than in the 1990 poll, one widely remembered as the first election in which the environment was central (Bean, McAllister & Warhurst 1990).1 The rise of environmental issues and the emergence of the Australian Greens are significant in Australia’s political history. How can we explain their increasing importance, and what implications does the environment have for understanding political engagement in Australia? In this chapter, we answer these questions with the help of one influential account of these new political and cultural developments proposed by American political scientist Ronald Inglehart. We first examine evidence for Inglehart’s central argument – that a fundamental value change has transformed advanced democracies – and then draw out the implications of this change for political and civic engagement. We focus on finding out how widespread so-called ‘postmaterialist’ values are among Australians, and what they tell us about support for environmentalism and our willingness to become engaged citizens. Figure 6.1 Environment as an important election issue, federal elections 1990–2001, per cent 95
90
%
85
80
75
70 1990
1993
1996
Election year SOURCE Australian Election Study 1990, 1993, 1996, 1998, 2001
1998
2001
AusSocAtts03
26/7/05
84 •
8:55 AM
Page 84
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
T H E R I S E O F P O S T M AT E R I A L I S T VA L U E S According to Inglehart, the basic values held by citizens in advanced democracies are shifting away from ‘material’ concerns about economic and physical security towards ‘postmaterial’ concerns about quality of life and freedom of self-expression (Inglehart 1977, 1990a, 1997). These new values are, in turn, transforming politics, economics and social life. For Inglehart, this value change helps explain the emergence of new social movements, such as the environmental movement and the women’s movement, which are concerned with quality of life and selfrealisation, and which depend on the involvement of their supporters to reach their goals. Green political parties – and the responsiveness of major political parties to environmental issues – also underline the growing importance of postmaterial values among citizens. Scholars have shown that support for the environment – expressed by joining an environmental group, participating in environmental protest, or voting green – is much stronger among postmaterialists than materialists (Inglehart 1990a, 1990b; Tranter 1996, 2004; Papadakis 1993). Postmaterialists are also more likely to see environmental problems as urgent (Crook & Pakulski 1995; Pakulski, Tranter & Crook 1998), and even to accept cuts in their living standard to protect our natural heritage (Bean 1998). Postmaterial values also stress ‘sustainability’ as a desirable condition for economic growth, and inform concepts like ecologically sustainable development (Papadakis 2000). Wide acceptance of sustainability criteria in economic development suggests that mainstream political parties have had to adjust to new social and political awareness. In the past, economic development and environmental protection were seen as trade-offs, but increasingly, industry and business interests are having to respond to interests other than purely material ones (Papadakis 2000). In 1990, for instance, the Business Council of Australia asserted that ‘[W]hile material living standards figure prominently in the welfare of Australian families, how well off we feel depends not only on material things but also our quality of life’ (Business Council of Australia 1990, p. 1, cited in Papadakis 2000, p. 83). Postmaterial values may also prefigure in disenchantment that followed a long period of economic reform and greater competition in Australia (see Pusey 2003; see also chapter 10). According to Michael Pusey, quite a few ordinary Australians feel that their quality of life has been compromised by economic change and that Australian society as a whole is worse off. And several cultural changes might be seen as
AusSocAtts03
26/7/05
8:55 AM
Page 85
A R E P O S T M AT E R I A L I S T S E N G A G E D C I T I Z E N S ?
• 85
responses to these pressures, such as ‘downshifting’, which rejects a culture of overwork and overconsumption (see Schor 1998), and adopting a ‘seachange’ in favour of a gentler, more meaningful pace of life. Voting for minor political parties like the Australian Greens and Democrats is another possible sign of this change. In short, competitive economic conditions that hit quality of life may also grate against the new postmaterialist outlook, and encourage some to seek out new political and lifestyle choices.
W H O A R E T H E P O S T M AT E R I A L I S T S ? Inglehart understands materialism and postmaterialism as enduring value systems held by individuals. This distinction between values, attitudes and beliefs can be traced to Rokeach (1968), who described values as evaluative beliefs; that is, beliefs that entail some kind of judgment about worth or importance. In Inglehart’s scheme, these values communicate life priorities that are stable and enduring parts of people’s worldviews, and are not fleeting or transitory like other beliefs and attitudes. Inglehart (1977, 1990a, 1997) argues that values are learned in early adolescence from childhood socialisation and experiences and, from there, shape political preferences and behaviour. Those who grow up in circumstances of relative physical safety and economic affluence tend to prioritise quality of life, postmaterial issues to a greater extent than material or economic issues. By contrast, early experiences of economic hardship, war, or major social and political change all contribute to the development of material values (Inglehart 1977, p. 23). According to Inglehart (1997, pp. 34, 46), once values are formed, they tend to endure throughout people’s lives, so early formative experiences are particularly important. Younger generations, particularly those born after the Second World War, have more postmaterial values because they experienced relative affluence and safety during their ‘formative years’. Therefore, generational-based differences in value orientations should be apparent for different age cohorts, with citizens of advanced democracies becoming increasingly postmaterialist as younger generations replace older ones. We would therefore expect that younger Australians are more likely to take on postmaterial concerns like supporting environmental causes than older ones, and that postmaterialists would be more ‘proenvironment’ than materialists are. Inglehart (1997) bases his argument for a major cultural shift from material to postmaterial values on the results from the World Values Survey, an international study of political and cultural change using
AusSocAtts03
26/7/05
86 •
8:55 AM
Page 86
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
national findings from over 80 countries. Within these surveys, a set of 12 questions asks people to choose national priorities for their country over the next ten years. The 12 items are presented as three sets of four items each, and within each set of four, respondents nominate their first and second choice of national priorities or goals. The 12 items are shown in table 6.1, along with the percentage of respondents in the Australian Survey of Social Attitudes 2003 (Gibson et al. 2004) who nominate each one as either a first or second choice. The economy dominates respondent first choices, with over half of respondents indicating that maintaining a high level of economic growth should be first priority, and 50 per cent choosing a stable economy as their first choice. These are both clearly materialist objectives, as is ‘maintain order in the nation’, which is nominated by over one-third of respondents in the second set. These priorities highlight the material value orientation of many Australians – stability and economic growth still matter to citizens. However, for many, ensuring material wants is not a single priority. The equal third and fourth choices – giving people more say in important government decisions, and giving people more say in their jobs and community – are postmaterial priorities that confirm our desire for free expression and democratic engagement. When we turn to second choices, materialist priorities – the fight against crime, and maintaining a strong defence force – come out on top, nominated by 42 per cent and 31 per cent of respondents respectively. However, two postmaterialist priorities – freedom of speech and giving people more say in their jobs and communities – come equal second and third respectively. Overall, we see that Australians nominate materialist priorities first, but they also substantially endorse postmaterialist ones. These findings reflect the results found for other countries (Inglehart 1990). Table 6.1 does not tell us if some people consistently choose materialist goals while others choose postmaterialist ones or whether some individuals accommodate materialist and postmaterialist values at the same time. Inglehart suggests that people generally value both materialist and postmaterialist priorities, but younger people, with material goals satisfied, prioritise postmaterialist goals in a way that older generations could not. So, in order to classify individuals as materialist or postmaterialist, Inglehart uses two strategies. First, he identifies the postmaterialist items reported in Table 6.1 (C, F, H, J, K) and counts the number of these chosen by respondents from within each of the three sets of items (Abramson & Inglehart 1995, p. 121; Inglehart & Abramson 1999). Individuals are scored between 0 and 5, so that high
AusSocAtts03
26/7/05
8:55 AM
Page 87
A R E P O S T M AT E R I A L I S T S E N G A G E D C I T I Z E N S ?
• 87
scores reflect a higher number of postmaterial choices (Inglehart 1997, p. 389). We refer to this measure of postmaterialist values as the ‘12item index’. Accordingly, very few people are ‘pure’ materialists or postmaterialists: just over 30 per cent of respondents are at the materialist end of the spectrum (scores of 0 or 1), but 18 per cent are at the postmaterialist end, indicated by scores of 4 and 5 (see table 6.2). Over half of all respondents are mixed, indicated by scores of 2 or 3. These intermediate scores come about when a respondent chooses two or three Table 6.1 Preferences for national priorities over the next ten years, 2003, per cent, within columns and panels First choice
Second choice
25
A
Maintain a high level of economic growth
53
B
Make sure this country has strong defence forces
20
31
C
See that people have more say about how things are done at their jobs and in their communities
25
28
D
Try to make our cities and countryside more beautiful
2
15 17
E
Maintain order in the nation
37
F
Give people more say in important government decisions
37
25
G
Fight rising prices
12
26
H
Protect freedom of speech
14
31
I
A stable economy
50
21
J
Progress towards a less impersonal and more human society
21
19
K
Progress towards a society in which ideas count more than money
10
18
L
The fight against crime
19
42
(n > = 4150) SOURCE The Australian Survey of Social Attitudes 2003
Table 6.2 Materialists and postmaterialists in Australia, 2003 Values score on 12-item index
Per cent of respondents
0 Materialist
7
1
24
2
28
3
23
4
12
5 Postmaterialist (n = 4121) SOURCE The Australian Survey of Social Attitudes 2003
7
AusSocAtts03
26/7/05
88 •
8:55 AM
Page 88
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
postmaterialist items and two or three materialist ones. It seems that most people endorse both materialist and postmaterialist priorities, with fewer people at each pole. The second way that Inglehart (1977, p. 28) assesses values is by using the second set of four items shown in table 6.1 (items E to H). Respondents choosing options E and G (‘order in the nation’ and ‘fight rising prices’), regardless of the order of their choice, are deemed materialists, while those who choose any combination of options F and H (‘give people more say’ and ‘protect freedom of speech’) are postmaterialists. The third category consists of responses involving any other combinations of items. These respondents hold ‘mixed’ value orientations. This classification again shows that the mixed orientation category among AuSSA 2003 respondents is largest, but now pure postmaterialists outnumber materialists (table 6.3). The results reinforce the view that people with ‘pure’ materialist or postmaterialist values are comparatively few and that most people in fact endorse a mix of materialist and postmaterialist national priorities. These broad patterns are similar to those found in both other countries and in previous Australian research (Abramson & Inglehart 1995; Bean & Papadakis 1994a), although the number of postmaterialists identified in table 6.3 is somewhat higher than in earlier studies (Western & Tranter 2001; Tranter & Western 2003), a finding that may be due to slight differences in question order between AuSSA 2003 and earlier surveys. Our finding – that many people hold ‘mixed’ values – might be a good reason for not opposing materialism and postmaterialism as though citizens can be classified neatly into either. Most people are somewhat materialist and postmaterialist at the same time. Some criticism of Inglehart’s work addresses this very argument (see Bean & Papadakis 1994a, 1994b; Inglehart 1994; and Hellevik 1994 [in reply]). Critics maintain that since materialism and postmaterialism are both desired goals, and since satisfying material needs is a precondition for taking on postmaterialist priorities, it is a Table 6.3 Materialists, postmaterialists and ‘mixed value’ respondents, 2003 Values
Materialist
Per cent of respondents
14
Mixed
64
Postmaterialist
22
(n = 4161) SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
8:55 AM
Page 89
A R E P O S T M AT E R I A L I S T S E N G A G E D C I T I Z E N S ?
• 89
mistake to see postmaterialism and materialism as opposed (Bean & Papadakis 1994a). Inglehart strongly rejects this argument, and even in more recent work (see Inglehart 1997; Inglehart & Baker 2000) that otherwise involves some rethinking of his ideas. It is not important for us to resolve this debate here. What is important is that few people are unambiguously materialist or unambiguously postmaterialist. Most people value economic and physical security and quality of life and selfexpression. Fewer people choose one set of values over the other. Who holds these different values? Inglehart (1997, p. 4) predicts that the most important factor is formative experiences. As we reported earlier, high levels of economic security in advanced democracies after the Second World War contributed to a gradual shift from materialist values to postmaterialist ones as younger people worried less about economic and physical security. As younger generations replace older ones, the citizens of advanced societies become more postmaterialist. This line of reasoning suggests that age should be the key to explaining values across the population, but that other socialising forces, such as gender, religion and education (the latter two changing across generations), should also play some role. We can assess the dispersion of postmaterialist values by examining how they are distributed among different social and demographic groups. Moreover, values may play a role in occupational choice and union membership but, in these instances, values may function as a cause, and not an effect, of behaviour. Regardless of how values are measured or whether values are a cause or effect of group involvement, we find that major social and demographic groups have distinct materialist/postmaterialist profiles (see figures 6.2 and 6.3). We first show average scores on the 12-item postmaterialism index by: gender, age, education level, religion, political party identification, occupation, employment sector, and union membership (see figure 6.2). In figure 6.2, the horizontal line shows the overall average score on the postmaterialism index for AuSSA 2003 respondents, which is about 2.3. The average scores for different groups are also shown. Recall that scores of 4 or 5 indicate a strong postmaterialist value orientation, while a score of 0 or 1 is materialist. Most scores range from between 1.8 and 2.8, consistent with ‘mixed’ value positions. The factors that most differentiate respondents are age, education level, religion and, strongly, political party identification. We find that gender, occupation, employment sector and union membership are less important. AuSSA 2003 data only weakly supports Inglehart’s argument that younger people are more postmaterialist than older people. We find
AusSocAtts03
26/7/05
90 •
8:55 AM
Page 90
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
that 18–54 year olds are equally postmaterialist, but more postmaterialist than older people. Respondents with university degrees, and with no religion are both more postmaterialist. However, the largest differences are found for party identification. Minor party identifiers – Greens, Australian Democrats and supporters of other minor parties excluding One Nation – are the most postmaterialist, while Liberal Party and National Party identifiers are the most materialist. Australian Labor Party (ALP) identifiers are in the middle, about on a par with professionals, public sector employees, union members and younger respondents. We get a similar picture by examining the numbers of pure postmaterialists on the four-item index (items E to H from table 6.1). Overall, about 22 per cent of AuSSA 2003 respondents are postmaterialists, as we report in both figure 6.3 and table 6.3. Again, party identification most strongly divides the value groups, followed by education level and religion. Minor party identifiers are again most postmaterialist: over 45 per cent are pure postmaterialists, while Liberal and National identifiers are least postmaterialist (see figure 6.3). Respondents with higher educational qualifications are more likely to be postmaterialist than those with fewer qualifications, and the non-religious are more postmaterialist than those with a religion. There is less variation by age on this measure than on the previous one; though, again, younger cohorts stand out as most postmaterialist. Of course, the many social attributes that determine our values orientations frequently occur together; that is, professionals will also have degrees, younger people are more likely to have degrees than older people and less likely to be religious, and so on. By modelling these attributes simultaneously, multivariate analysis allows us to discover what explains these distinct value orientations. We include as independent variables: gender, age, education, religion, partisanship, ‘equivalised’ household income,2 occupation, sector, self-employment status and union membership. Whether we use the four-item or 12-item index to measure values, reliable differences in values are only found for education, party identification and religion. The other demographic variables – age, gender, income, occupation, sector, employment status and union membership – are not significant. Multivariate analysis shows that respondents with more education and university qualifications are more likely to be postmaterialist than people with fewer qualifications. People who are religious, which includes Christians and non-Christians, are less postmaterialist than the non-religious respondents, while Labor, Green, Australian Democrat, and One Nation identifiers (as well as
AusSocAtts03
26/7/05
8:55 AM
Page 91
• 91
A R E P O S T M AT E R I A L I S T S E N G A G E D C I T I Z E N S ?
Figure 6.2 Average scores on materialism/postmaterialism scale by demographic group, 2003 3.5 Minor
3.0
Mean Values Score
Degree None ALP
2.5
Other
18–54 Female Male
55–64
2.0
Profess.
Public
Yes
Other
Private
No
None HS&Trade
Catholic
Some HS
Protest
One Nat.
65+ Lib/Nat
Union member
Sector
Occupation
Party Identity
Religion
Education
Age
Gender
1.5
NOTE Scoring on vertical axis, 0 = Pure materialist, 5 = Pure postmaterialist. ABBREVIATIONS HS = High school; Protest. = Protestant; Lib./Nat. = Liberal or National Party;
One Nat. = One Nation; Profess. = Professional. SOURCE The Australian Survey of Social Attitudes 2003
Figure 6.3 Postmaterialists by demographic group, 2003, per cent Minor
0.45
Proportion Postmateralist
0.40
0.35 None
Degree
0.30
One Nat. ALP
Profess.
0.25 18–54 55–64
Female Male
0.20
Other Catholic
HS&Trade 65+
None
Other
Public
Yes
Private
No
Some HS Protest
0.15
Lib/Nat
ABBREVIATIONS HS = High school; Protest. = Protestant; Lib./Nat. = Liberal or National Party; One Nat. = One Nation; Profess. = Professional. SOURCE The Australian Survey of Social Attitudes 2003
Union member
Sector
Occupation
Party Identity
Religion
Education
Age
Gender
0.10
AusSocAtts03
26/7/05
92 •
8:55 AM
Page 92
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
non-identifiers) are all more likely to be postmaterialist than Liberal–National identifiers. (Green and Australian Democrats identifiers are particularly postmaterialist, followed by Labor identifiers, and finally One Nation and non-identifiers.) Genuine postmaterial values are found among Greens, Australian Democrats and Labor voters, while the apparently slight postmaterial orientation of One Nation voters and non-identifiers has different origins. Detailed studies of the attitudes of One Nation voters highlight their feelings of disenfranchisement, economic insecurity and dissatisfaction with ‘elite’ sponsored policies that favour immigration, Aboriginal rights and globalisation. So, these respondents may look ‘postmaterialist’ because they choose priorities such as protecting freedom of speech and giving people more say in government decisions, but these choices express their alienation from trade and immigration policies that they link to the ‘top-down’ decisions of major parties (see Bean 2000; Gibson, McAllister & Swenson 2002; Charnock & Ellis 2003, 2004; Goot & Watson 2001; Wilson & Turnbull 2001). When One Nation identifiers (and, by extension, respondents who reject all parties) choose these priorities, they perhaps reflect their sense of economic insecurity and personal grievance instead of underlying material security (see also Western & Tranter 2001). Inglehart and others believe that values are important because they shape political and cultural behaviour. But, as stable individual characteristics, do they also shape people’s personal identities, their understandings of who they are? AuSSA 2003 presents an opportunity to examine how values shape our personal identities, especially those parts of our identity embedded in social institutions and structures. AuSSA 2003 respondents were asked: We are all part of different groups. Some are more important to us than others when we think of ourselves. In general, which in the following list is most important to you in describing who you are? The options for identity sources are presented in table 6.4, ordered by their overall importance to respondents. We see that family and marital status are the most important source of identity overall, followed by occupation, nationality (Australian for most respondents), gender, religion and age. Social class and political party are quite unimportant sources of personal identity. These findings closely match earlier research, which used a slightly different wording (see Emmison & Western 1990, 1991; Phillips & Western 2005). The same identities also tend to matter
AusSocAtts03
26/7/05
8:55 AM
Page 93
A R E P O S T M AT E R I A L I S T S E N G A G E D C I T I Z E N S ?
• 93
Table 6.4 Values and personal identity, 2003, per cent Identity source
Materialist
Mixed
Postmaterialist
Family/marital status
49
55
47
Occupation
15
16
15
Nationality
10
11
8
Gender
6
6
8
Religion
5
5
4
Age group
6
4
6
Part of Australia where respondent lives
3
3
4
Race/ethnicity
3
2
2
Social class
1
2
1
<1
0
2
Political party n
(266)
(1311)
(502)
SOURCE The Australian Survey of Social Attitudes 2003
to people regardless of their value orientations, although there are minor variations. Family and marital status are more important to people with mixed values than to either materialists or postmaterialists, while nationality matters more to materialists and postmaterialists than to those with mixed values. Gender is more important to postmaterialists than others, which may reflect Inglehart’s argument that postmaterialist values drive many ‘issue-based’ movements, which emphasise identity like feminist and gender movements do. Religion is more important to materialists than others, probably reflecting that materialists are older, and that older people are more religious. But overall, differences are quite small, suggesting that there are common elements to identity that matter to all value groups.
HOW DO VALUES INFLUENCE POLITICAL ACTION? Social scientists argue that values and cultural characteristics are important because they influence people’s behaviour, which, in turn, shapes social and political institutions. An earlier tradition of political research established that the cultural values of citizens are particularly important in underpinning the legitimacy of democratic political institutions like free and fair elections (Almond & Verba 1963; see also chapter 5). More recently, Inglehart has argued that postmaterialist values, especially the concern with quality of life, democratic participation and self-expression, potentially translate into new forms of political expression (Inglehart
AusSocAtts03
26/7/05
94 •
8:55 AM
Page 94
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
1997, chapter 6). Established political behaviours, such as voting, he argues, are essentially elite-led , directed by parties, politicians and formal organisations such as business and industry groups as well as unions. As such, they are ‘top-down’ political behaviours requiring limited participation by individual citizens. By contrast, new forms of political behaviour are ‘bottom-up’; initiated by the participants themselves through direct action or as members of informal, issue-based groups and new political organisations. These forms of political involvement are consistent with postmaterial value orientations (Inglehart 1997, chapter 6). Here, we follow up this argument by finding out if postmaterialists are more likely to engage in these new forms of participation; in this case, particularly over environmental politics. These engagements include: membership of environmental organisations; willingness to pay higher taxes to protect the environment; contact with politicians and officials; and boycotts or purchases of products for ethical and environmental reasons. We use multivariate analysis to predict the probability with which respondents with materialist, mixed and postmaterialist values engage in these activities. The models control for: gender, age, education, religion, household income, occupation, employment sector, self-employment status, union membership, urban/rural location, and party identification. AuSSA 2003 asks respondents whether or not they are members of activist groups including Greenpeace, the Wilderness Society, Care Australia and Oxfam. These groups actively work to improve quality of life – eliminating poverty for others, and protecting the environment – and rely on direct participation through financial contributions and volunteer work. Overall, 10 per cent of AuSSA 2003 respondents belong to these organisations. Our model shows that the main factors predicting membership of these groups are education, party identification and postmaterial values. Respondents with more educational qualifications are more likely to belong to these groups, as are minor party identifiers (Greens, Democrats and others except for One Nation supporters). Respondents with strongly postmaterial values are also more likely to belong than those with mixed or material values. The first row of table 6.5 shows the probability of different respondents joining activist groups. There is only a 3 per cent probability of materialists belonging, but this probability rises to 5 per cent for respondents with mixed values, and to 10 per cent for postmaterialists. This means that, on average, postmaterialists are twice as likely as those
AusSocAtts03
26/7/05
8:55 AM
Page 95
A R E P O S T M AT E R I A L I S T S E N G A G E D C I T I Z E N S ?
• 95
Table 6.5 Predicted probability of environmental political engagement by value orientation, 2003, per cent Values
Materialist
Member of environmental group
Mixed
Postmaterialist
3
5
10
Willing to pay higher taxes to protect environment
45
56
72
Contacted a politician or government official to express views
26
32
41
Boycotted or bought products for environmental, ethical or political reasons
42
52
66
(589)
(2671)
(901)
n
SOURCE The Australian Survey of Social Attitudes 2003
with mixed values and more than three times as likely as materialists to belong to activist environmental organisations when we account for other factors. Another ‘hard’ measure of political engagement along these lines is respondent willingness to pay higher taxes to support spending in areas that they think are important. AuSSA 2003 asks respondents if they would be willing to pay higher taxes to allow the government to spend more on environmental protection. Over 53 per cent are willing to pay ‘quite a bit’ or ‘a little more’ tax to protect the environment (see also chapter 7). Several factors in the model predict willingness to pay more tax for the environment: occupation, education, party identification, and, again, values. Professionals and associate professionals, and clerical, sales and service workers are more willing to pay more tax to support the environment than are manual workers. Respondents with university degrees are also more willing than those without; and Labor and minor party identifiers are more willing than identifiers with the other parties (or those with no political party identification). Postmaterialists are also more willing than materialists and respondents with mixed values. The model predicts that, on average, 45 per cent of materialists are willing to pay higher taxes, compared to 56 per cent of people with mixed values, and 72 per cent of postmaterialists (see table 6.5). There is therefore a clear ‘values gradient’ to willingness to pay in the direction Inglehart would predict. Two final forms of political engagement we consider here are: contacting a politician or official to represent one’s interests; and boycotting or buying products for environmental, ethical or political
AusSocAtts03
26/7/05
96 •
8:55 AM
Page 96
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
reasons. Overall, just under one-third (32 per cent) of AuSSA 2003 respondents have contacted a politician or official in the previous two years. The most important factors that predict contacting politicians are: age, occupation, education, party identification, employment sector, region and values. Younger respondents are less likely than older ones to have contacted politicians and officials; managers and professionals are more likely than manual workers to have done so. Public sector workers, respondents living outside metropolitan areas, Liberal/National and minor party identifiers, and postmaterialists are also all more likely to have directly contacted politicians and officials. Again, we find that the probability of contacting politicians and officials varies by value orientation, with around 26 per cent of materialists doing so, compared to 41 per cent of postmaterialists (see table 6.5). Almost half of AuSSA 2003 respondents indicate that they have bought or boycotted products for ethical, political or environmental reasons in the previous two years (2001–2003). Multivariate analysis shows that gender, age, education, religion, party identification and values all influence this behaviour. Women, older people, respondents with more qualifications, those with no religion, minor party identifiers and postmaterialists are all more likely to buy or boycott for value reasons. Men, younger people, those with fewer qualifications, respondents with a religion, One Nation identifiers and supporters of both major parties as well as those with materialist or mixed values all are less likely. For the three value orientations, our model predicts that 42 per cent of materialists engage in boycotts or selective purchases, compared to about half of those with mixed values, and a high two-thirds of postmaterialists (see table 6.5).
CONCLUSION We have found here that most Australians are not easily classified as either materialists or postmaterialists. The largest group has a mixed value orientation that combines material and postmaterial priorities and, according to AuSSA 2003, makes up almost two-thirds of the adult population. Our findings would appear to contradict Inglehart’s argument that postmaterialism is above all else a generational shift that we can find among younger age groups. Instead, we find gender, religion, education and party identification most sharply define the boundaries of our three values orientations – and not age. However, if we ignore these factors, and simply compare values across the major age groups, younger generations are more postmaterialist, as figures 6.2 and 6.3 show, even if
AusSocAtts03
26/7/05
8:55 AM
Page 97
A R E P O S T M AT E R I A L I S T S E N G A G E D C I T I Z E N S ?
• 97
age differences in values are small. One way of overcoming this apparent contradiction is to point out that the value change that Inglehart identifies among younger cohorts is mediated by experiences like increased access to higher education (which, for many, influences value development into early adulthood) and the declining role of cultural institutions such as organised religion. The roles that education and religion play in these circumstances may help to explain how generational value change actually occurs rather than contradicting Inglehart’s argument. Indeed, social institutions and structures may have a greater influence on values than the relatively diffuse experiences of affluence or material hardship. People experience affluence or material and economic insecurity in concrete settings, most notably families, schools and peer groups as they grow up, and, as adults, they experience them as individuals, as men and women and as members of different classes and ethnic backgrounds. Our suggestions here are not intended to comprehensively address how values emerge but to highlight the role of ‘thick’ social institutions in value formation. AuSSA 2003 findings also suggest that values matter widely for political action, particularly new forms of direct political involvement. Postmaterialists are more attracted to join environmental and activist organisations, and are more willing to pay higher taxes to support the environment, to contact politicians and officials to represent a cause, and are more willing to choose goods and services on principle. These differences appear to be distinctly located in values that cannot be explained by other differences between postmaterialists and materialists like education or religion. Viewed broadly, these findings have significance beyond environmental politics – they point to engaged citizens who act on postmaterial beliefs and values that go beyond immediate material self-interest. It is also important to remember, however, that the connection between values and political and civic involvement does not only take the postmaterial forms that ‘progressive’ thinkers endorse. Both Australia and the United States returned conservative governments with expanded majorities in late 2004. Immediately after the US election, some analysts argued that President George W Bush’s second term success hinged on his ability to mobilise a ‘values-based’ constituency with roots in evangelical Christianity (see Goodstein & Yardley 2004). The values that mattered for Bush’s re-election were not of self-expression, self-realisation and tolerance of diversity, but traditional ones, which invest moral authority in God and religion, the nuclear
AusSocAtts03
26/7/05
98 •
8:55 AM
Page 98
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
family, and in a determined political leader who endorsed these values publicly. Moreover, it is not surprising that conservative politicians should draw on traditional moral values in appealing to voters who reject the humanism of postmaterial values. Inglehart himself explores the place of traditional values in his most recent work, suggesting that the distinction between traditional and secular values cuts across the distinction between materialist and postmaterialist ones (Inglehart & Baker 2000). How this new emphasis on traditional morality influences the dynamics of advanced democracies will tell us much about future political and civic engagement as well as conflicts over values. NOTES
1 2
The 1987 AES also asked about this issue, but we do not present these data because they are not directly comparable. Here we derive equivalent household incomes by dividing total household income by the size of the household.
The authors acknowledge the extremely helpful comments of their assigned editor, Shaun Wilson, and the other editors and researchers associated with this volume. The authors also thank Andrew Passey and Mark Lyons for allowing them to use their embargoed variables from the AuSSA data. Amanda Hosking provided able research assistance. Part of this chapter was written while Mark Western was on leave at the Sociology Department, University of Wisconsin-Madison. He acknowledges their generous support and hospitality during his stay.
REFERENCES
Abramson, P & Inglehart, R (1995) Value Change in Global Perspective, University of Michigan Press, Ann Arbor. Almond, G & Verba, S (1963) The Civic Culture, Princeton University Press, Princeton, New Jersey. Bean, C (1998) ‘Australian attitudes towards the environment in cross-national perspective’, in J Pakulski & S Crook (eds) Ebbing of the Green Tide? Environmentalism, Public Opinion and the Media in Australia, University of Tasmania, Hobart. ____ (2000) ‘Nationwide electoral support for One Nation in the 1998 federal election’, in M Leach, G Stokes & I Ward (eds) The Rise and Fall of One Nation, University of Queensland Press, Brisbane, pp. 136–52. Bean, C, McAllister, I & Warhurst, J (1990) (eds) The Greening of Australian Politics: The 1990 Australian Federal Election, Longman Cheshire, Melbourne. Bean, C & Papadakis, E (1994a), ‘Polarized priorities or flexible alternatives? Dimensionality in Inglehart’s materialism–postmaterialism scale’, International Journal of Public Opinion Research, 6: 264–88. ____ (1994b), ‘Polarized priorities and flexible alternatives: Response to Inglehart and Hellevik’, International Journal of Public Opinion Research, 6: 295–97.
AusSocAtts03
26/7/05
8:55 AM
Page 99
A R E P O S T M AT E R I A L I S T S E N G A G E D C I T I Z E N S ?
• 99
Charnock, D & Ellis, P (2003) ‘The Structure of the Australian party system and its strategic consequences’, Australian Journal of Political Science, 38: 423–43. ____ (2004) ‘Postmaterialism and postmodernization in Australian electoral politics’, Electoral Studies, March 2004, 23: 45–72. Crook, S & Pakulski, J (1995) ‘Shades of green: Public opinion on environmental issues in Australia’, Australian Journal of Political Science, 30: 39–55. Emmison, JM & Western, MC (1990) ‘Social class and social identity: A comment on Marshall et al.’, Sociology, 34: 241–53. ____(1991) ‘The structure of social identities’, in J Baxter, JM Emison, JS Western & MC Western (eds) Class Analysis and Contemporary Australia, Macmillan, Melbourne, pp. 279–306. Gibson, R, McAllister, I & Swenson, T (2002) ‘The politics of race and immigration in Australia: Race and One Nation voting in the 1998 election’, Ethnic and Racial Studies, 25: 823–44. Gibson, R, Wilson, S, Meagher, G, Denemark, D & Western, M (2004) Australian Survey of Social Attitudes 2003, Data file, Australian Social Science Data Archive, The Australian National University, Canberra. Goodstein, L & Yardley, W (2004) ‘The 2004 election: Faith groups: President benefits from efforts to build a coalition of religious voters’, The New York Times, 5 November. Goot, M & Watson, I (2001), ‘One Nation’s electoral support: Economic insecurity versus attitudes to immigration’, Australian Journal of Politics and History, 512–15. Hellevik, O (1994) ‘Measuring cultural orientation: Rating versus ranking’, International Journal of Public Opinion Research, 6: 292–95. Inglehart, R (1977) The Silent Revolution: Changing Values and Political Styles Among Western Publics, Princeton University Press, Princeton, New Jersey. ____ (1990a) Culture Shift in Advanced Industrial Societies, Princeton University Press, Princeton, New Jersey. ____ (1990b) ‘Values, ideology, and cognitive mobilisation in new social movements’, in R Dalton & M Kuechler (eds) Challenging the Political Order, Polity Press, London, pp. 43–67. ____ (1994) ‘“Polarized priorities or flexible alternatives? Dimensionality in Inglehart’s materialism–postmaterialism scale”: A comment’, International Journal of Public Opinion Research, 6: 289–92. ____ (1997) Modernization and Postmodernization: Cultural, Economic and Political Change in 43 Societies, Princeton University Press, Princeton, New Jersey. ____ & Abramson, P (1999) ‘Measuring postmaterialism’, American Political Science Review, 93: 665–77. ____ & Baker, W (2000) ‘Modernization, cultural change and the persistence of traditional values’, American Sociological Review, 65: 19–51. Pakulski, J, Tranter, B & Crook, S (1998) ‘The dynamics of environmental issues in Australia: Concerns, clusters and carriers’, Australian Journal of Political Science, 33: 235–52. Papadakis, E (1993) Politics and the Environment: The Australian Experience, Allen & Unwin, Sydney. ____ (2000) ‘Environmental values and political action’, Journal of Sociology, 36: 81–97. Phillips, T & Western, M (2005) ‘Social change and social identity: Postmodernity, reflexive modernisation and the transformation of social identities in Australia’, in F Devine, M Savage, R Crompton & J Scott (eds), Class, Consumption and Identity, Palgrave, London, pp. 163–85.
AusSocAtts03
26/7/05
100 •
8:55 AM
Page 100
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Pusey, M (2003) The Experience of Middle Australia: The Dark Side of Economic Reform, Cambridge University Press, Cambridge. Rokeach, M (1968) Beliefs, Attitudes and Values: A Theory of Organization and Change, Jossey-Bass, San Francisco. Schor, J (1998) The Overworked American: Why We Want What We Don’t Need, HarperCollins, New York. Tranter, B (1996) ‘The social bases of environmentalism in Australia’, Australian and New Zealand Journal of Sociology, August 32(2): 61–84. ____ (2004) ‘The environment movement: Where to from here?’, in R White (ed.) Controversies in Environmental Sociology, Cambridge University Press, Melbourne, pp. 185–203. ____ & Western, M (2003) ‘Postmaterial values and age: The case of Australia?’, Australian Journal of Political Science, 38: 239–57. Western, M & Tranter, B (2001) ‘Postmaterialist and economic voting in Australia, 1990–1998’, Australian Journal of Political Science, 36(3): 439–58. Wilson, S & Turnbull, N (2001) ‘Wedge politics and welfare reform in Australia’, Australian Journal of Politics and History, 47: 384–402.
AusSocAtts03
26/7/05
8:55 AM
Page 101
CHAPTER 7
W H E R E T O F O R T H E W E L FA R E S TAT E ? S h a u n Wi l s o n , G a b r i e l l e M e a g h e r a n d Tr e v o r B r e u s c h
INTRODUCTION Welfare states provide income support to individuals and families in and outside labour markets. Welfare states also provide health, education and social services, some on a universal basis; that is, as citizenship rights. The generosity of welfare states is closely linked to their governments’ ability to raise taxes to pay for benefits and services. Across the western OECD democracies, there has been a long debate about the sustainability of generous welfare programs supported by high taxes (see Castles 2004, pp. 1–9). Australia is not a highly taxed nation by OECD standards. At 31 per cent of GDP in 2000, taxes were lower than all other advanced democracies except for the United States and Japan (OECD 2002). However, since the 1980s, Australian governments have reformed the tax and welfare systems with the stated objectives of containing the size of government, meeting the public’s preference for lower income taxes, and encouraging greater participation in the workforce by welfare recipients. Lowering income tax and trimming welfare are often justified on the view that the Australian public prefers low taxes and is relatively ungenerous when it comes to welfare. It has been a surprise to many, then, that polls over the past two years have consistently shown that a majority of Australians actually prefer higher public spending on health and education to lower taxes (Newspoll 2004) and that they strongly support government provision of social services (Meagher 2004). So, do current attitudes to Australia’s welfare state support the view that Australians are preoccupied with lower taxes? How strongly do we support greater requirements on welfare recipients to actively seek employment? And how much private sector involvement in social service provision do
AusSocAtts03
26/7/05
102 •
8:55 AM
Page 102
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Australians think is best? The Australian Survey of Social Attitudes 2003 (Gibson et al. 2004) helps answer questions about how Australians are responding to changes in the welfare state.
A U S T R A L I A’ S W E L FA R E S TAT E E X PA N S I O N Before we consider public opinion about the welfare state, some background to recent changes to taxing and social spending is necessary. First to taxes. During the 1980s and 1990s, Labor governments cut personal and corporate income tax rates, and the Coalition government further reduced income taxes in July 2000 when it introduced the GST (Warren 2004). Yet further income tax cuts were announced in the federal budgets for 2004–05 and 2005–06 (Department of Treasury 2004a). Although income taxes have been cut, indirect taxes are now broader and government tax collection has improved, so that the overall tax take has remained stable.1 While governments have reformed the tax system, they have also sought to limit growth in the number of welfare beneficiaries in some programs (see Newman 1999; Whiteford & Angenent 2002) but to expand income support in others. Governments from both sides of politics have targeted programs to limit eligibility or contain expenditures: programs like the Age pension were means-tested in the mid-1980s, while unemployment benefits have been made increasingly conditional on work obligations over the last decade. Falling unemployment throughout the late 1990s and early 2000s helped reduce labour market spending further, as did cuts to Australia’s active labour market programs under the Howard government (OECD 2004). However, programs like the family payments system – which now includes the so-called Baby Bonus payment of over $3000 for new mothers – have expanded in both scope and generosity (Winter 2000; McNamara et al. 2004; see also Wade 2004). Table 7.1 shows social expenditure trends for Australia between 1980 and 2001. Averages for the six English-speaking OECD nations and 21 OECD countries are also included, which establish Australia’s position comparatively. Overall, we find something that might surprise many. Australia has gone from a low-spending nation at 11 per cent of GDP in 1980 to 18 per cent in 2001 – an expansion unrivalled among the English-speaking nations, and one that now leaves Australia in the lower-middle ranks of major OECD nations for public social spending. Most of this increase came during the Hawke and Keating governments. In particular, we see that income support to Australian families
AusSocAtts03
26/7/05
8:55 AM
Page 103
W H E R E T O F O R T H E W E L F A R E S TAT E ?
• 103
Table 7.1 Public social expenditure in Australia and OECD nations, 1980 and 2001, per cent of GDP Australia 1980
Australia 2001
Australia 1980–2001 change
Family benefits
1.0
2.8
1.8
1.7 (0.3)
1.9 (0.3)
Health
4.4
6.2
1.8
6.0 (1.0)
6.1 (1.0)
Old age and survivor benefits
3.8
4.9
1.1
5.5 (–0.1) 8.9 (2.1)
Incapacity-related benefits
1.0
2.3
1.3
1.8 (0.5)
Unemployment benefits and ALMPs**
0.7
1.4
0.7
1.1 (–0.1) 1.8 (0.4)
0.3
0.1
–0.2
11.3
18.0
6.7
Housing*** Total ****
EnglishOECD-21 speaking in 2001 OECD* in (change 2001 1980–2001) (change 1980–2001)
0.7 (0.3)
2.5 (0.0)
1.0 (0.4)
17.5 (2.3) 24 (3.2)
* Includes six nations – Australia, Canada, Ireland, New Zealand, United Kingdom and United States. ** Active labour market programs; excludes Ireland. *** Excludes United States. **** Includes ‘Other’ category of social expenditure. SOURCE OECD (2004)
contributed strongly to welfare state expansion, rising by 1.8 per cent of GDP over this time. And promises on family payments from both sides of politics featured prominently in the federal election campaign in late 2004. Australia now spends more than any other English-speaking nation on family income support, which now plays a major role in protecting low-income families in Australia’s post-reform labour market. Researchers have shown that, between 1997–98 and 2004–05, growth in family income in the bottom quintile of the income distribution kept up with average income growth through more generous family payments and not through increased wages and salaries (see McNamara et al. 2004, pp. 16, 30–31; Wade 2004). In addition to these developments in public social spending, the federal government has also promoted private provision of welfare goods. In health care, the government introduced a 30 per cent Private Health Insurance Rebate scheme in 1999 and has increased real spending on private education over its term in office as well as encouraged private sector entry into other social services, such as employment placement (the Jobs Network) and child care (see Meagher 2004). Given these changes, what are Australians’ priorities for policy on taxing and spending? This is a question that has certainly shaped political debate over the past few years (Gittins 2004). AuSSA 2003 asks respondents: ‘If the government had a choice between reducing personal income taxes or increasing social spending on services like health and
AusSocAtts03
26/7/05
104 •
8:55 AM
Page 104
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
education, which do you think it should do?’ Table 7.2 reports their answers, alongside responses to a similar question asked in the Australian National Political Attitudes Surveys (1967–79), National Social Science Surveys since 1984 and Australian Election Studies since 1987.2 To facilitate comparisons over time, we devise a ‘tax and spend’ scale, which is the proportion in favour of increased spending over the sum of support for increased spending and reduced taxes. This scale is displayed in figure 7.1. Table 7.2 and figure 7.1 show that Australians were more willing to forgo income to pay for major welfare state services in 2003 than they had been at any time in the previous two decades. We need to go back to the 1960s to find a stronger mood for social spending – a time that predated the expansion of the welfare state under the Whitlam Labor government of 1972–75. During the 1980s and 1990s, Australians apparently conformed to conventional political wisdom, strongly favouring tax cuts over increased spending. However, at the 1998 election, and for the first time since 1969, less than half of surveyed respondents (47 per cent) preferred tax cuts to maintaining or increasing spending. This election was held two years after the Howard government first came to office, two years during which the new government had cut real expenditures (Department of Treasury 2003, Appendix C). Support for higher social spending started increasing during these early Howard years, reaching about 30 per cent in 2001. Meanwhile, support for tax cuts fell away sharply, from a high point of 65 per cent at the 1987 federal election to 42 per cent in 2001. Two years later, in 2003, support for tax cuts had fallen to 28 per cent, and support for increased spending had increased to 48 per cent.3 Some of this increased support for spending might be attributed to slight changes to the question. The AuSSA 2003 question specifies two popular areas of spending – health and education – while the earlier question mentions only ‘spending on social services’ generally. But the question also specifies a popular option for lower taxes as well – lower personal income taxes. Published polls have confirmed strong support for social spending over lower taxes (Newspoll 2004; Grattan 2004). Although these slight changes may explain some of the rising support for spending, we argue that most of the increase continues the shift in public preferences evident in the Australian Election Studies since the Coalition’s first victory in 1996. Despite the shift towards higher social spending, some Australians are more likely than others to prefer spending increases to tax cuts.4 Age,
AusSocAtts03
26/7/05
8:55 AM
Page 105
W H E R E T O F O R T H E W E L F A R E S TAT E ?
• 105
Table 7.2 Lower taxes or more spending? 1967–2003, per cent Increase social spending
The same/ depends
Reduce taxes
Tax and spend scale*
1967
(n = 1898)
72
–
28
0.72
1969
(n = 1800)
73
–
27
0.73
1974
(n = 8084)
39
–
61
0.39
1979
(n = 1915)
38
–
62
0.38
1984
(n = 2909)
24
23
53
0.31
1987
(n = 1740)
15
20
65
0.19
1990
(n = 6084)
8
29
63
0.11
1993
(n = 2942)
17
27
56
0.23
1996
(n = 1727)
17
26
57
0.23
1998
(n = 1804)
26
27
47
0.36
2001
(n = 1951)
30
28
42
0.42
2003
(n = 4080)
48
24
28
0.63
NOTE See endnote 2 for groupings of original categories.
* Tax and spend scale is the proportion of respondents in favour of increased spending over the sum of support for spending and reduced taxes. SOURCES Australian National Political Attitudes Survey 1967, 1969, 1979; Morgan Poll, August 1974; National Social Science Survey 1984, 1990; Australian Election Study 1987, 1993, 1996, 1998, 2001; The Australian Survey of Social Attitudes 2003
Figure 7.1 Tax and spend scale, 1967–2003 0.8
0.7
Tax and spend scale
0.6
0.5
0.4
0.3
0.2
0.1
0 1967
1969
1974
1979
1984
1987
Year SOURCE Table 7.2
1990
1993
1996
1998
2001
2003
AusSocAtts03
26/7/05
106 •
8:55 AM
Page 106
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 7.3 Attitudes to tax and spend by age, 2003, per cent Increase social spending
Depends
Reduce taxes
18–34 (n = 714)
42
26
32
35–49 (n = 1238)
45
25
30
50–64 (n = 1233)
51
22
27
65 and over (n = 895)
51
25
24
SOURCE The Australian Survey of Social Attitudes 2003
for example, makes a difference, as do class identity and party affiliation. Older voters are now more solidly ‘pro-spending’, even if their voting behaviour strongly favours the Liberal–National Coalition. Table 7.3 shows that among voters over 50 years old, 51 per cent favour increased spending – around twice the proportion favouring reduced taxes. Young people are least likely to support social spending (42 per cent) and most likely to support cutting taxes (32 per cent). These findings make sense: young people are less frequent clients of the welfare state, and are at a stage of the life cycle where they are entering housing markets and having children. So we would anticipate that their support for lower taxes reflects their need for cash income. Lower unemployment rates – the product of expanded tertiary education participation and an improved labour market – also mean that young people are less likely to come into contact with the welfare state. In Australia, young people receive proportionately less in transfer payments than older cohorts (Whiteford & Angenent 2002, pp. 58–59). They are also understandably less likely to use medical services, high use of which prompts support among older voters for higher spending. Whatever the reasons for lower support for spending among young people, we should keep these differences in perspective: the 18–34 age group still prefers social spending over tax cuts by a 10 per cent margin, and full-time students, who are typically younger, are a constituency for more spending. Class identification also affects Australians’ preferences on taxation and spending. Table 7.4 shows that people who identify as middle-class are more likely to support higher social spending (52 per cent) than people who identify as working-class (44 per cent). In 2001, both broad class categories expressed similar support for tax cuts (43 per cent). Middle- and working-class respondents also expressed similar – and lower – levels of support for increased spending in the 2001 Australian Election Study, at 30 per cent and 28 per cent respectively (see Wilson & Breusch 2004, p. 105). However, it appears that middle-class respon-
AusSocAtts03
26/7/05
8:55 AM
Page 107
W H E R E T O F O R T H E W E L F A R E S TAT E ?
• 107
dents were more likely than working-class respondents to have changed their minds on this question between 2001 and 2003. This shift may also have something to do with change to the question: middle-class respondents may be more attracted to spending targeted to popular areas, such as health and education. But the middle class may have responded to substantial income tax cuts in mid-2000; with demands for lower taxes now satisfied, middle-to-higher income earning families may have turned their attention to the state of public services. Party affiliations are also important. Table 7.4 shows that Labor supporters are more likely to favour spending increases (53 per cent) than Liberal–National supporters (42 per cent), and less likely to favour tax cuts (24 per cent) compared to Liberal–National identifiers (35 per cent). Still, it is significant that more Coalition voters support spending increases than tax cuts.5 Green identifiers are the most ‘social-democratic’ and least equivocal respondents on this question: more than three-quarters support social spending, while only 10 per cent prefer cutting taxes. Green identifiers are also least likely to have responded ‘it depends’ to this question (13 per cent). By contrast, those without a party affiliation were most likely to have responded ‘it depends’ (30 per cent), perhaps indicating that uncommitted voters are also undecided on this question. Table 7.5 breaks down support for spending by perceptions of standards of health and Medicare and public education. Here, we uncover an important relationship – respondents who believe that the standard of health and education services has declined in the last two years are much more likely to prefer social spending (57 per cent and 59 per cent respectively). Perceptions of decline in both health and education are widely held (a clear majority for health and Medicare), with middle- and working-class identifiers expressing similar levels of dissatisfaction. So the large number of Australians who pay for and benefit from these Table 7.4 Attitudes to tax and spend by class and party identification, 2003, per cent Increase social spending
Depends
Reduce taxes
Middle-class (n = 1958)
52
22
26
Working-class (n = 1626)
44
25
31
Liberal–National identifiers (n = 1416)
42
23
35
Labor identifiers (n = 1379)
53
23
24
Green identifiers (n = 190)
77
13
10
No party identification (n = 801)
43
30
27
SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
108 •
8:55 AM
Page 108
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 7.5 Attitudes to tax and spend by perceptions of service standards in health and education, 2003, per cent Increase social spending
Depends
Reduce taxes
Health and Medicare same or better (n = 1531)
35
28
37
Health and Medicare declined (n = 2375)
57
20
22
Public education same or better (n = 2002)
41
25
34
Public education declined (n = 1696)
59
19
22
SOURCE The Australian Survey of Social Attitudes 2003
services are dissatisfied with current government provision and express support for more spending. We cannot assess whether respondents identify a direct link between lower taxes and fewer resources for major services. At the very least, the perception of declining services is enough to swing many voters to the spending option even if they remain attracted to lower taxes. So far, we have focused on the public’s choice between the taxing and spending options open to government. While both demographic and party-political factors matter when explaining the public’s choice, we find that those who are dissatisfied with services are more likely to support higher spending. These findings do not tell us if the public is willing to pay higher taxes so the government can spend more on social services. We must resolve a tricky question confronting almost all public opinion researchers who uncover high support for public spending. Critics of these findings argue that respondents are rarely made aware of the tax implications of higher spending (see, for example, Saunders 2003, pp. 33–34). They believe that when respondents are able to ‘optimise’ their preferences, in full awareness of the costs of their choices, they reduce their support for spending because the tax consequences are made apparent. A further AuSSA 2003 question enables us to establish whether voters are willing to pay the taxes necessary to increase spending. The question is: Here is a list of areas where the federal government spends money. Please tell us if you would be willing to pay higher taxes so the government could spend more in each of these areas … [Health and Medicare, Primary and secondary schools, Environmental protection, Defence and national security, Welfare benefits]
AusSocAtts03
26/7/05
8:55 AM
Page 109
W H E R E T O F O R T H E W E L F A R E S TAT E ?
• 109
Our findings do not confirm the critics’ argument: table 7.6 shows that respondents are willing to pay more tax for health and Medicare, and for public education, by large margins. Almost 70 per cent of voters are prepared to pay more tax to increase spending on health and Medicare, reflecting the service’s wide use and popularity. This finding contrasts remarkably with the finding from the Australian Election Study in 1990, which had one in five voters prepared to pay higher taxes for health and education services. This huge increase in support for paying higher taxes over time says something about the reliability of survey evidence. The fact that support for higher taxes for health and education has changed so markedly indicates that the Australian public have adopted a different view about tax increases. If we assume otherwise – that respondents always express ‘socially desirable’ spending preferences to pollsters – it would be hard to explain why so few voters wanted tax increases in 1990. We can also see in table 7.6 that a majority of voters are also willing to pay more tax for the environment (54 per cent) and defence and national security (54 per cent). Only welfare benefits attract minority support (34 per cent). These results provide compelling evidence about public support for raising taxes when tax increases are modest and when they are clearly marked for spending in areas of public demand. We can better understand the political implications of these findings if we classify the electorate according to their ‘tax and spend’ orientations to key services of the welfare state. To do this, we classify respondents who support higher taxes for the three welfare options (health, Table 7.6 Attitudes to tax increases to pay for services, 1990 and 2003, per cent (2003) Are you willing to pay more tax for …
Quite a bit more tax
A little more tax
No more tax
Health and Medicare
17
52
28
4
Primary and secondary schools
14
49
31
6
Environmental protection
11
43
40
6
Defence and national security
10
44
40
5
6
28
60
6
Agree
Disagree
Neither
Health care
18
51
31
Education
21
40
39
Welfare benefits
Can’t choose/ Neither
(n≥ 4062) (1990) The government should increase taxes and spend more on …
(n≥ 2002) SOURCES Australian Election Study 1990; The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
110 •
8:55 AM
Page 110
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
schools and welfare) as ‘Spenders’. We call respondents who support none of these three options ‘Anti-spenders’. The rest – those who support one or two areas for tax increases, but not all three – are ‘Conditional spenders’. Figure 7.2 shows that that 28 per cent of AuSSA 2003 respondents are ‘Spenders’, 19 per cent are ‘Antispenders’, and the remaining 52 per cent are ‘Conditional spenders’. These results reveal that a surprisingly small number of respondents consistently oppose tax increases. Although just over a quarter turn out to be ‘Spenders’ who consistently favour tax rises, a majority of Australians are ‘Conditional spenders’. This means a large number of voters might support a tax increase to fund higher spending, depending on the tax increases and spending proposed. In other words, many Australians are open to specifically justified tax increases. When we consider how key constituencies divide between ‘Spenders’, ‘Anti-spenders’ and ‘Conditional spenders’, we can identify the groups that have made up their mind about taxes and spending, and those whose preferences are more conditional. Table 7.7 shows that Green/Democrat voters have the highest number of ‘Spenders’ (45 per cent). More Labor identifiers are ‘Spenders’ than Coalition identifiers, but like Coalition identifiers, many more of them are ‘Conditional spenders’ than either outright ‘Spenders’ or ‘Anti-spenders’. Along with
Figure 7.2 ‘Spenders’, ‘Anti-spenders’ and ‘Conditional spenders’, 2003, per cent 60
50
40
30
20
10
0 Spenders
Conditional Spenders
SOURCE The Australian Survey of Social Attitudes 2003
Anti-spenders
AusSocAtts03
26/7/05
8:55 AM
Page 111
W H E R E T O F O R T H E W E L F A R E S TAT E ?
• 111
Table 7.7 Party identifiers by spending type, 2003, per cent in rows Spenders
Conditional Spenders
Anti-spenders
Liberal–National (n = 1484)
22
57
21
Labor (n = 1435)
35
49
17
Green/Democrat (n = 303)
45
44
11
SOURCE The Australian Survey of Social Attitudes 2003
the evidence we presented earlier, that shows a large movement in the electorate’s preparedness to accept tax rises, and the growing proportion of voters now prepared to conditionally support tax increases, these findings challenge the conventional wisdom that Australians are wedded to a low taxing and low spending model.
AT T I T U D E S T O W E L FA R E A N D REDISTRIBUTION We see in table 7.6 that respondents are less willing to pay higher taxes to fund spending on welfare benefits than they are to fund health or education expenditures. Weak support for welfare is not a new development. Historically, Australia’s ‘wage earners’ welfare state’ protected incomes by regulating the labour market, leaving income support outside work comparatively underdeveloped and lacking a firm base of public support (Castles 1985). Does this mean that Australians reject the idea of a social welfare system altogether? AuSSA 2003 asks respondents whether they agree with statements about the welfare system, people in receipt of benefits and the redistributive role of government (see table 7.8). Our findings show that, although most respondents are wary about the potential impact of welfare cutbacks and tend to support redistributing income, they are ungenerous to those they think are not making an effort to find work. AuSSA 2003 respondents agree with the statement that ‘cutting welfare would damage too many people’s lives’ – fewer than one in five disagree with this. Respondents with higher incomes (above $78 000 annually) and Liberal voters are least likely to think that cutting welfare would damage people’s lives. This is not surprising: Martin Gilens’s study of support for the welfare benefit system in the United States shows that support for welfare falls away among those who have little direct contact with the system (Gilens 1999, pp. 53–59). Although most respondents do not support cutting welfare benefits, most do support increasing work
AusSocAtts03
26/7/05
112 •
8:55 AM
Page 112
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
obligations on welfare recipients. This attitude has, no doubt, guided and reinforced welfare reform in recent years: both Labor and Liberal–National governments have increased obligations on the unemployed during the 1990s (Keating 1994; Newman 1999). Indeed, the Howard government has more recently attempted to extend ‘welfare-towork’ reforms to those receiving single-parent and disability benefits (Vanstone & Abbott 2001), especially in the 2005–06 federal budget. We have evidence that the Australian public broadly supports this approach – table 7.8 shows that 73 per cent of respondents agree that people who receive welfare benefits should be under more obligation to find work. Other research has shown that Australians are toughest on the young unemployed and immigrants when it comes to setting tough conditions for welfare (Eardley & Matheson 1999, p. 30; Eardley, Saunders & Evans 2000, p. 17). We still see a strong distinction between support for the ‘deserving poor’ and lack of support for the ‘undeserving poor’ that characterises ‘liberal’ welfare states (Arts & Gelissen 2001, p. 286). Successive governments have reformed the unemployment benefits system so that work tests are more stringent and penalties for breaches more significant (Whiteford & Angenent 2002, p. 16). Despite these policy changes, table 7.8 shows that only a small minority of respondents (14 per cent) think that ‘it is too hard to qualify for welfare benefits in Australia today’. However, more than a quarter of all respondents (28 per cent) choose the middle category, ‘Neither agree nor disagree’. This ‘fence-sitting’ could indicate that either respondents lack the knowledge Table 7.8 Attitudes to the welfare system and redistribution, per cent Agree
Neither agree nor disagree
Disagree
66
15
17
2
73
17
9
2
14
28
52
6
59
19
20
3
44
22
30
4
Cutting welfare benefits would damage too many people’s lives People who receive welfare benefits should be under more obligation to find work It is too hard to qualify for welfare benefits in Australia today Around here, most unemployed people could find a job if they really wanted to Government should redistribute income from the better off to those who are less well-off (n≥ 4119) SOURCE The Australian Survey of Social Attitudes 2003
Can’t choose
AusSocAtts03
26/7/05
8:55 AM
Page 113
W H E R E T O F O R T H E W E L F A R E S TAT E ?
• 113
to express a clear opinion about qualifying for welfare benefits or are registering that the government has ‘tightened up’ the benefits system and that, at least, they cannot actively disagree with the statement. One reason why respondents support tougher measures for the unemployed is the belief that most unemployment is voluntary. A majority of respondents (59 per cent) agree with the statement ‘around here, most unemployed people could find a job if they really wanted to’; only 20 per cent disagree. And there is some evidence that attitudes to the unemployed measured in this way have hardened; in 1996, 39 per cent of respondents in the Middle Australia Project survey conducted by Michael Pusey agreed with this statement (Wilson & Turnbull 2001, p. 399). One explanation for the large difference is the strong period of employment growth through the 1990s that has reduced headline unemployment rates to below 6 per cent nationally (ABS 2004). As Eardley and Matheson (1999, p. 8) suggest, the public perception that unemployment is voluntary is strongest when unemployment is low. Believing that it is easy to find work also strongly affects public support for imposing work obligations on welfare recipients. Table 7.9 shows that respondents who believe that the unemployed could find work if they wanted to strongly support work obligations, and that those who think that the unemployed could not easily find work do not support work obligations. Fully 89 per cent of those who agree that ‘around here, most unemployed people could find a job if they really wanted to’ also agree that people who receive welfare benefits should be under more obligation to find work. By contrast, only 40 per cent of those who disagree that most unemployed people could find work go on to agree with obligations on the unemployed. These findings lead us to speculate that tough economic times – when people believe that it is Table 7.9 Attitudes to work obligations by perceptions of the unemployed, 2003, per cent People who receive welfare benefits should be under more obligation to find work Around here, most unemployed people could find a job if they really wanted to
Agree
Neither agree nor disagree
Disagree
Agree (n = 2393)
89
8
3
Neither agree nor disagree (n = 788)
63
30
7
Disagree (n = 803)
40
30
30
SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
114 •
8:55 AM
Page 114
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
harder to find work – might undermine support for work obligations on the unemployed. In other words, public support for ‘workfare’ might be sensitive to the economic cycle. AuSSA 2003 respondents’ assessment of the welfare system is linked to their attitudes about the role of government in redistributing income. Overall, 44 per cent of respondents agree that ‘Government should redistribute income from the better off to those who are less well-off ’, while 30 per cent disagree (see table 7.8). This suggests that, on balance, Australians continue to approve of some redistributive role for the government, a basic principle of the welfare state (see also chapter 10). Table 7.10 shows that respondents who agree that the government should redistribute income are more likely to agree that cutting welfare would damage too many lives (79 per cent) than those who disagree with redistribution (52 per cent). Although the major-
Table 7.10 Attitudes to cutting welfare by redistribution, 2003, per cent Cutting welfare benefits would damage too many people’s lives Government should redistribute income from the better off to those who are less well-off
Agree
Neither agree nor disagree
Disagree
Agree (n = 1771)
79
10
11
Neither agree nor disagree (n = 886)
62
24
14
Disagree (n = 1210)
52
17
31
SOURCE The Australian Survey of Social Attitudes 2003
Table 7.11 Attitudes to getting welfare by redistribution, 2003, per cent It’s too hard to qualify for welfare benefits in Australia today Government should redistribute income from the better off to those who are less well-off
Agree
Neither agree nor disagree
Disagree
Agree (n = 1682)
23
30
47
Neither agree nor disagree (n = 853)
11
40
49
6
20
73
Disagree (n = 1191) SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
8:55 AM
Page 115
W H E R E T O F O R T H E W E L F A R E S TAT E ?
• 115
ity of those who disagree with redistribution remain opposed to cutting welfare, we find a very strong commitment to preserving welfare among those who think that the role of government is to boost the incomes of the less well-off. Similarly, we see in table 7.11 that respondents who agree the government should redistribute income are more likely to agree that it is too hard to qualify for welfare benefits (23 per cent) than those who disagree (just 6 per cent). In other words, believing that the government should provide extra resources for the less well-off translates into a more generous attitude to welfare provision and welfare clients.
WHO SHOULD DELIVER SOCIAL SERVICES? We round off this chapter by examining attitudes to another part of the welfare state – the provision of social services. Australia has a complex and evolving mix of social service provision to complement the informal services that families and communities provide for themselves. Public, not-for-profit, and for-profit organisations each deliver different proportions of different types of social services, subsidised to varying degrees from the public purse (Meagher 2004). And just as Australian governments have been modernising income support programs, so they have overhauled the way they fund, deliver and oversee social service provision. Significantly, both Labor and Coalition federal governments have promoted private businesses in services like nursing homes, child care, employment services and hospitals. Meanwhile, at the state level, some governments have been changing the way they subsidise the not-forprofit agencies that have long delivered the bulk of formal welfare and support services to people with disabilities and to vulnerable youths and families (see, for example, NSW Department of Community Services 2001). State governments now ‘purchase’ services from not-for-profit agencies under tight contracting arrangements rather than ‘funding’ services by ‘hand[ing] a lump sum to a reputable welfare agency and trust[ing] them to get on with the job’ (Healy 1998, p. 40). Current thinking is that private sector service delivery and stringent contracts between governments and not-for-profit agencies enhance ‘customer’ choice and improve efficiency and service quality. What kinds of organisations do Australians prefer to deliver social services? AuSSA 2003 asks respondents who is best suited to deliver a range of services – education, health, services for job seekers, child care, counselling and welfare services and care for the disabled and elderly. Their responses give us insight into the preferred mix of the three modes
AusSocAtts03
26/7/05
116 •
8:55 AM
Page 116
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
of welfare delivery: governments providing universal services, businesses providing private services, and families and communities providing personal support. Table 7.12 shows that more respondents believe that governments are better suited to deliver social services than any other form of organisation, including families and relatives. Public support for governments delivering health services (80 per cent) and education (83 per cent) is especially high. The education result is significant considering the comparatively high number of Australian students attending either Catholic or independent schools. Indeed, 80 per cent and 74 per cent of respondents who attended Catholic and independent schools respectively think that governments are best suited to deliver education. This may reflect that many respondents answer this question in a general sense (that is, the best service provider for the community as a whole) or that the ‘private business’ response category inadequately captures support for private education. For the remaining services, we find again that a majority of respondents also prefer governments to deliver care for the disabled (61 per cent), aged care (55 per cent) and services for job seekers (54 per cent). Support for government provision of child care is lower at 34 per cent, but exceeds support for delivery by private businesses (30 per cent) or families and relatives (27 per cent). Support for charitable provision of counselling and welfare services is considerable at 30 per cent, although more respondents prefer government provision (43 per cent). Respondents also prefer charities (14 per cent) over families (11 per cent) and private businesses (9 per cent) to provide services to disabled people. However, very few respondents believe that charities are best
Table 7.12 Attitudes about who is best suited to deliver human services, per cent Governments
Charities
Private business
Families and relatives
Can’t choose
Education
83
0
10
2
5
Health services
80
1
14
0
5
Services for job seekers
54
5
33
0
8
Child care
34
3
30
27
6
Counselling or welfare support
43
30
12
6
8
Care for the disabled
61
14
9
11
6
Care for elderly persons
55
10
11
18
6
(n = 2014) SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
8:55 AM
Page 117
W H E R E T O F O R T H E W E L F A R E S TAT E ?
• 117
suited to provide health services, education, child care, and services for job seekers (all 5 per cent or less). This pattern of responses suggests a complex distribution of attitudes about the services that Australians believe that governments should provide as a right to all, the services that communities should provide in the form of charity to the needy, the services that private businesses should supply to consumers in markets, and – finally – the services that family members should provide for each other. A large majority of respondents seem to adhere to a universalistic norm for provision of health and education as a right. A significant minority adhere to a norm of charitable provision of counselling and welfare support and (fewer) care for the disabled and the aged, although governments remain the preferred provider of all these services. Respondents are almost evenly divided between universalistic, business-oriented, and family-oriented norms when it comes to child care. Universalistic and business-oriented norms prevail for services for job seekers. Are there political differences about service delivery? We might expect to find that Liberal–National voters support private provision and Labor voters public. Surprisingly, for health, education and aged care, there is a fair degree of consensus about the dominant place of government. On who is best to deliver health services, 77 per cent of Liberal–National identifiers say governments, compared to 83 per cent of Labor identifiers. Among those who do not think that governments are best suited to deliver health, Coalition identifiers are more likely than Labor to favour private businesses (19 per cent versus 10 per cent). On who is best to deliver education, 81 per cent of Coalition identifiers say governments, compared to 86 per cent of Labor supporters. On aged care, 53 per cent of Liberal–Nationals prefer governments, compared to 59 per cent of Labor identifiers. Again, Coalition identifiers are slightly more in favour of private businesses in aged care than Labor (14 per cent versus 10 per cent). Where attitudes diverge on party lines is the contentious area of child care: Coalition identifiers actually favour private businesses over all other providers (39 per cent versus 28 per cent who favour governments). Conversely, Labor supporters favour government child care over private provision by 41 per cent to 24 per cent. Earlier, we found that many voters are willing to pay more tax for mainstay social services like health and education. But what form of delivery do these voters want to benefit from extra funds? Respondents willing to pay more tax to fund services are more likely to choose governments. However, support for government provision remains high among those
AusSocAtts03
26/7/05
118 •
8:55 AM
Page 118
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
who do not want to pay more tax. Among respondents willing to pay quite a bit more tax for health services, 87 per cent choose governments. But 72 per cent of respondents not willing to pay more for health still choose government provision. The same pattern emerges with education: among respondents who are willing to pay quite a bit more tax, 89 per cent choose governments as the best providers. Among respondents who were not willing to pay more tax, 78 per cent choose governments. Clearly a majority of voters who seek extra funds for health and education by raising taxes also want governments to provide additional services. If it is true that the public sector is inefficient in the eyes of the public, we might expect to find greater support for private provision among those who believe that the standards of major public services have declined. But this is not what we find: respondents who believe that the standard of health services, including Medicare, have decreased are just as likely to think that governments are best suited to deliver health services as respondents who feel that standards have stayed the same, or even those who believe that standards have increased (all 80 per cent). And respondents who feel that the standard of public education has decreased are only slightly less likely to think that governments are best suited to deliver education (84 per cent) than respondents who believe that standards have stayed the same or increased (82 per cent and 87 per cent respectively). We saw earlier in this chapter that a majority believe that health care standards are declining (59 per cent) and many believe the same of public education (42 per cent). But these negative perceptions do not indicate that the public has lost confidence in government services per se, and so prefer private alternatives. We might infer from these findings that the public still seeks the government to reverse declines in standards without necessarily opting for privatisation.6 In most social services, governments are expanding service delivery by subsidising private provision, a policy direction that might run into opposition at the ballot box in the future.
CONCLUSION So how have we answered our initial questions about Australians’ support for the welfare state in this chapter? Certainly, Australians have a history of enthusiastically endorsing tax cuts and they believe that people who receive welfare benefits should meet demanding work obligations. We do not find the strong pro-welfare state attitudes that researchers have consistently found among, for example, Scandinavian citizens (Svallfors 1997, pp. 288–89). Although its foundations are now eroding, Australia’s wage earner model has bequeathed only a limited institutional legacy to
AusSocAtts03
26/7/05
8:55 AM
Page 119
W H E R E T O F O R T H E W E L F A R E S TAT E ?
• 119
build welfare programs that protect incomes outside the labour market. However, Australia has done better in protecting the incomes of working families. The expansion of family payments over the past two decades is the most important example of this, and explains a good part of Australia’s recent welfare state expansion. In some ways, family income support is becoming the Australian Government’s major policy instrument for dealing with work and income in the new century as Australia’s traditional work and welfare model is transformed. This program design – which benefits many middle Australian families – also locks in public support in a way that reflects Australian welfare attitudes. Although the Australian model is quite unlike many higher spending European countries, it is also significantly different from the US model. Income support for the working-age population is higher in Australia than in the United States, and we do not encounter strongly individualistic attitudes to social service provision that researchers frequently attribute to the American public. A long history of public education, bipartisan support for protecting family incomes and increasing government involvement in health care in Australia has helped build a constituency for universal provision of some major social services. NOTES
1
2
Commonwealth budget data shows that receipts (revenues) were 26.5 per cent of GDP in 1999–2000, falling progressively to a projected 22.6 per cent in 2004–05 (Department of Treasury 2004b, table 1). However, in 2000, with the introduction of the New Tax System, several indirect taxes were abolished and the GST was introduced. Because the government pays the GST directly to the states, GST revenue is not reported as federal revenue in Budget Paper no. 1. Instead, GST receipts, which amount to 4 per cent of GDP annually and which effectively replace indirect taxes formerly reported as federal revenue, appear in Budget Statement no. 3 as part of Commonwealth–state financial relations. When we add GST revenue back in to the ratio of revenue to GDP, we get a rough and ready ratio of taxation to GDP that includes the GST. Calculated this way, the ratio has remained fairly stable at around 26 per cent since 1999–2000 (calculated using Department of Treasury 2004c, table 3). In AuSSA 2003 and the AES, the response options are ‘Strongly favour’ or ‘Mildly favour’ reducing taxes, with similar options for spending more on social services, and a middle category of ‘Depends’. The outer two options on each side of the question are grouped to produce the three categories in table 7.1. The options in NSSS are ‘Very strongly for’, ‘Fairly strongly for’ or ‘Mildly for’ reducing taxes, with parallel options for increasing spending. The outer two options on each side are again grouped, and the middle two options are grouped to become the middle category of three. This probably overstates the frequencies in the middle category in the grouped NSSS results compared with AES and AuSSA 2003, but the taxing and spending options are treated symmetrically for all surveys.
AusSocAtts03
26/7/05
120 •
3
4
5 6
8:55 AM
Page 120
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Weighting the AuSSA 2003 sample for demographic characteristics of the respondents reduces support for spending from 48 per cent to 44 per cent and increases support for lower taxes from 28 per cent to 31 per cent. The unweighted proportions in the table are more comparable with the earlier surveys, most of which use similar sampling methods to AuSSA 2003. Multivariate analysis shows that being over 50 years of age, having a degree, being a Labor, Green, or Democrat voter, and believing that the standard of health and education services is declining, and believing that the government should redistribute income are all significantly and independently correlated with support for higher social spending (see Wilson & Breusch 2004). After weighting, the gap in favour of higher spending over lower taxes among Coalition identifiers narrows but remains. Strong support for public provision of social services is not mirrored in the pattern of service provision. Meagher (2004) shows that private businesses, often in receipt of massive public subsidies, dominate provision of several types of health and social service.
REFERENCES
Arts, W & Gelissen, J (2001) ‘Welfare states, solidarity and justice principles: Does the type really matter?’, Acta Sociologica, 44(4): 283–99. Australian Bureau of Statistics (ABS) (2004) Labour Force Australia, cat. no. 6202.0, Australian Bureau of Statistics, Canberra. Castles, F (1985) The Working Class and Welfare: Reflections on the Political Development of the Welfare State in Australia and New Zealand, 1890–1980, Allen & Unwin, Sydney. —— (2004) The Future of the Welfare State: Crisis Myths and Crisis Realities, Oxford University Press, Oxford. Department of Treasury (2003) Final Budget Outcome, 2002–2003. [Online], Available: . —— (2004a) 2004–5 Budget Paper No. 1, Statement 1, Department of Treasury, Canberra. —— (2004b) 2004–5 Budget Paper No. 1, Statement 13, Department of Treasury, Canberra. —— (2004c) 2004–5 Budget Paper No. 3, Federal Financial Relations, Department of Treasury, Canberra. Eardley, T & Matheson G (1999) ‘Australian attitudes to unemployment and unemployed people’, SPRC Discussion Paper no. 102, Social Policy Research Centre, University of New South Wales, Sydney. Eardley, T, Saunders, P & Evans, C (2000) ‘Community attitudes towards unemployment, activity testing and mutual obligation’, SPRC Discussion Paper no. 107, Social Policy Research Centre, University of New South Wales, Sydney. Gibson, R, Wilson, S, Meagher, G, Denemark, D & Western, M (2004) Australian Survey of Social Attitudes 2003, Data file, Australian Social Science Data Archive, The Australian National University, Canberra. Gilens, M (1999) Why Americans Hate Welfare: Race, Media, and the Politics of Antipoverty Policy, University of Chicago Press, Chicago. Gittins, R (2004) ‘Hip-pocket nerve losing its appeal’, Sydney Morning Herald, 26 May. Grattan, M (2004) ‘Tax cuts no sure winner’, The Age, 28 April. Healy, J (1998) Welfare Options: Delivering Social Services, Allen & Unwin, Sydney.
AusSocAtts03
26/7/05
8:55 AM
Page 121
W H E R E T O F O R T H E W E L F A R E S TAT E ?
• 121
Jackson, A (2003) ‘ABC learning on acquisition alert’, Sydney Morning Herald, 15 September. Keating, P (1994) Working Nation: The White Paper on Employment and Growth, Australian Government Publishing Service, Canberra. McNamara, J, Lloyd, R, Toohey, M & Harding, A (2004) ‘Propserity for all? How low income families have fared in the boom times’, National Centre for Social and Economic Modelling, University of Canberra, 14 October. [Online], Available: Meagher, G (2004) ‘Do Australians want a private welfare state? Are they getting one anyway?’ The Drawing Board: An Australian Review of Public Affairs, Digest, May. [Online], Available: . Newman, J (1999) The Challenge of Welfare Dependency in the 21st Century, Commonwealth Department of Family and Community Services, Canberra. Newspoll (2004) Personal Income Taxes and Federal Government Spending, 21 February, [Online], Available: . NSW Department of Community Services (2001) Purchasing: A Partnership Model, Position Paper, Partnerships and Communities Directorate, NSW Department of Community Services, Sydney. Organisation for Economic Cooperation and Development (OECD) (2002) Revenue Statistics of OECD Member Countries, Organisation for Economic Cooperation and Development, Paris. —— (2004) Social Expenditure Database; [Online] available: . Saunders, P (2003) A Self-reliant Australia: Welfare Policy for the 21st Century, CIS Occasional Paper no. 86, Centre for Independent Studies, Sydney. Svallfors, S (1997) ‘Worlds of welfare and attitudes to redistribution: A comparison of eight western nations’, European Sociological Review, 13(3): 283–304. Vanstone, A & Abbott, T (2001) Australians Working Together – Helping People to Move Forward, A statement. [Online], Available: Wade, M (2004) ‘Handouts make welfare work for Howard’, Sydney Morning Herald, 6 November. Warren, N (2004) Tax: Facts, Fiction and Reform, Australian Tax Research Foundation Research Study no. 41, Australian Tax Research Foundation, Sydney. Whiteford, P & Angenent, G (2002) ‘The Australian system of social protection – An overview’, 2nd edn, Occasional Paper No. 6, Department of Family and Community Services, Canberra. Wilson, S & Breusch, T (2004) ‘After the tax revolt: Why Medicare matters more to middle Australia than lower taxes’, Australian Journal of Social Issues, 39(2): 99–116. Wilson, S & Turnbull, N (2001) ‘Wedge Politics and Welfare Reform in Australia’, Australian Journal of Politics and History, vol. 47, no. 3, September, pp. 383–402. Winter, G (2000) ‘Social security recipients, 1970 to 2000’, Parliamentary Library Research Note 15, 2000–01, Parliamentary Library, Australia.
AusSocAtts03
26/7/05
8:55 AM
Page 122
CHAPTER 8
IS THERE A CRISIS OF TRUST IN AUSTRALIA? Clive Bean
INTRODUCTION Announcing the date for the 2004 federal election, Prime Minister John Howard declared that ‘This election will be about trust’. With these words, Mr Howard invoked a catchword of modern social and political discourse: trust – or the lack of it. In newspapers and on talkback radio, we hear much talk of declining trust, the betrayal of trust and a crisis of trust from politicians, commentators and members of the broader public. This talk is fuelled by events on the political stage, such as the children overboard affair in 2001 and revelations about the poor quality of the intelligence leaders used to justify going to war in Iraq. But concern about trust is also fuelled by media reports of social events such as babies being snatched from their parents in broad daylight at what appears to be an ever-increasing rate. Does this anecdotal evidence reflect a newly emerging crisis of trust in Australia? Or has the ‘problem’ of trust always been around, with particular political and social events drawing our attention to it from time to time? Either way, the concept of trust has become the focus of academic, political and social debate over the last decade to an unprecedented degree. This focus on trust is due in no small part to the work of American political scientist Robert Putnam (1993; 1995; 2000), for whom social and political trust represent important indicators of the key theoretical concept of social capital. Social capital is what politicians are referring to when they talk about the need to revive and rebuild communities. It has become common for politicians from both sides of Australian politics to express concerns about
AusSocAtts03
26/7/05
8:55 AM
Page 123
IS THERE A CRISIS OF TRUST IN AUSTRALIA?
• 123
social capital, although the concept is most identified with the ideas of former Labor Party leader, Mark Latham (Latham 1998; but see also Costello 2003). In broad terms, social capital is the sense of connectedness and civic-mindedness that people feel in strong and close-knit communities. More specifically, Putnam has defined social capital as ‘features of social organization, such as networks, norms, and social trust that facilitate coordination and cooperation for mutual benefit’ (Putnam 1995, p. 67), or ‘social networks and the norms of reciprocity and trustworthiness that arise from them’ (Putnam 2000, p. 19). As is clear from the definition, trust is a central component of social capital. Putnam (1995, 2000) has argued that social capital is in decline in the United States and much of his evidence for this contention consists of measures of different aspects of trust. Other scholars have affirmed Putnam’s conclusions, both in the United States and in Australia (for example, Nye, Zelikow & King 1997; Burchell & Leigh 2002). As a general indicator of the health of social capital, measures of trust can also tell us how well we co-operate and maintain social networks. From the perspective of social capital theorists, if trust declines, economic growth and social cohesion may be undermined as citizens co-operate less at work and in their local and broader communities. Social capital is a highly complex phenomenon and the concept of trust is itself multifaceted (Braithwaite & Levi 1998; Warren 1999; Putnam 2000). In this chapter, I use data from the Australian Survey of Social Attitudes 2003 (Gibson et al. 2004) to consider two distinct elements of trust: social or interpersonal trust and political trust or trust in government. I also explore some related indicators, such as confidence in institutions, perceptions of the amount of power that different organisations should wield, and participation in social and political activities. Many see high rates of participation in social and political life as a sign of a community rich in social capital.
ARE AUSTRALIANS LOSING TRUST IN EACH OTHER? To find out whether Australians are losing trust in each other, we need to establish current levels of trust. AuSSA 2003 asks respondents: ‘Generally speaking, would you say that most people can be trusted or that you can’t be too careful in dealing with people?’ In response to this question, only a minority – 41 per cent – indicate that they think most people can be trusted, while 59 per cent are of the view that you cannot be too careful in dealing with people. To find that only four in ten give
AusSocAtts03
26/7/05
124 •
8:55 AM
Page 124
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
a trusting response may seem to some like a disturbingly low figure. However, only a slightly higher proportion of Americans gave a trusting response in recent surveys (Putnam 2000, p. 140), while in Britain the proportion of respondents giving trusting responses has been slightly lower than four in ten since the 1980s (Johnston & Jowell 2001, p. 182; see also Johnston & Jowell 1999). In the United States, interpersonal trust was higher in the 1960s and has been falling ever since, thus giving rise to the fears about declining social capital outlined above. In the 1960s, the proportion of Americans giving the trusting response was around 55 per cent (Putnam 2000, p. 140). Can we identify a similar trend in Australia? Unfortunately, we have nothing like the time series of observations that exist for the United States, or even for the United Kingdom. The best we can do is to track results for the 1983 and 1995 rounds of the World Values Survey and to compare these results with the 2001 Australian Election Study (AES) (Bean, Gow & McAllister 2002) and AuSSA 2003 findings. Figure 8.1 shows the percentage of respondents to these surveys who agreed that, generally speaking, most people can be trusted. In 1983, 48 per cent gave the trusting response. In 1995, the proportion of respondents giving the trusting response dipped to 40 per cent – almost the same as in 2003 – while in 2001 the trusting response was back up to close to the 1983 level, at 46 per cent. Having only four time points makes it difficult to discern a trend. If we take the first and the last time points, there might be evidence of declining interpersonal trust. But when we account for all four readings, a more reasonable interpretation seems to be one of trendless fluctuation. We will have to wait, though, for the results of several more surveys before we can confirm any trend. Of course, the real possibility remains that levels of social trust in Australia were higher further back in the 20th century. Another way of approaching the question of declining trust is to look at levels of trust within different age groups (or cohorts). Putnam (2000) was able to demonstrate, for example, that social trust was declining in the United States partly because older, more trusting, cohorts were being replaced with younger, less trusting cohorts. Johnston and Jowell (2001), on the other hand, found little evidence for this phenomenon in Britain, consistent with their ‘no change’ interpretation of two decades of British data on social trust. Is there a generational basis to trust in Australia? Table 8.1 shows that, to some extent, there is. The least trusting age cohort is those aged 18–34 years: only 36 per cent of this age group believe that most people
AusSocAtts03
26/7/05
8:55 AM
Page 125
• 125
IS THERE A CRISIS OF TRUST IN AUSTRALIA?
Figure 8.1 Agree that most people can be trusted, 1983–2003, per cent 60 50
48
46 41
40
%
40 30 20 10 0 1983
1995
2001
2003
Year
SOURCES World Values Survey 1983 Australian data (n = 1228), 1995 Australian data (n = 2048);
Australian Election Study 2001 (n = 2010); The Australian Survey of Social Attitudes 2003 (n = 4270)
can be trusted. This proportion jumps to 43 per cent among 35–49 year olds and to 45 per cent among the 50–64 year olds. However, among those aged 65 and over, the proportion of trusting respondents drops back to 37 per cent, almost the same level as for the youngest cohort. We certainly cannot say that the most trusting cohorts are ageing while the least trusting are coming of age. In Australia, interpersonal trust appears to peak in middle age and then declines thereafter. Overall, this pattern is more consistent with a life cycle account than a generational one, and provides no evidence to suggest that levels of trust might either decline or increase in the future. Table 8.1 also shows how trusting attitudes vary by gender, education, social class, income and area of residence. We can see a narrow difference between the sexes: men are a little more inclined to be trusting than women (43 per cent, compared to 39 per cent). A much wider gap, however, is evident between those with and without a university degree. Among respondents with fewer educational qualifications, 36 per cent give a trusting response. But among the fifth of the sample who are university graduates, fully 61 per cent express interpersonal trust. With education on the increase, this is surely one positive sign that social trust might increase in the future. There are similar class differences, with respondents who identify as middle-class more likely to express trust (50 per cent) than respondents who identify as working-class (30 per cent).
AusSocAtts03
26/7/05
126 •
8:55 AM
Page 126
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 8.1 Interpersonal trust by demographic groups, 2003, per cent Most people can be trusted
Can’t be too careful
Age 18–34 (n = 677)
36
64
35–49 (n = 1175)
43
57
50–64 (n = 1201)
45
55
65 and over (n = 893)
37
63
Female (n = 2081)
39
61
Male (n = 1824)
43
57
No university degree (n = 2997)
36
64
University degree (n = 839)
61
39
Sex
Education
Class identification Middle (n = 1948)
50
50
Working (n = 1580)
30
70
None (n = 319)
36
64
Income Under $10 400 (n = 744)
32
68
$10 400 to $25 999 (n = 1084)
35
65
$26 000 to $51 999 (n = 1123)
43
57
$52 000 and over (n = 746)
56
44
Region of residence Rural (n = 1427)
38
62
Urban (n = 2452)
43
57
SOURCE The Australian Survey of Social Attitudes 2003
Pusey (2003, p. 134) found that low income earners were less trusting than middle and high income earners. We can see differences between income groups among respondents to AuSSA 2003 too, such that trust increases progressively with rising incomes. Interestingly, and perhaps contrary to popular characterisations of city and country communities, urban residents are also more likely to express trust than rural residents, although the difference is modest.
ARE AUSTRALIANS LOSING TRUST IN GOVERNMENT? The evidence we have does not support the argument that Australians are losing trust in each other. What about the second element of trust: trust in government? AuSSA 2003 asks respondents: ‘Thinking about the federal government in Australia these days, would you say that it is
AusSocAtts03
26/7/05
8:55 AM
Page 127
IS THERE A CRISIS OF TRUST IN AUSTRALIA?
• 127
run for a few big interests looking out for themselves, or that it is run for the benefit of all the people?’ Four response categories are allowed: ‘entirely run for a few big interests’, ‘mostly run for a few big interests’, ‘mostly run for the benefit of all’ and ‘entirely run for the benefit of all’. Around 13 per cent believe that the federal government is run entirely for a few big interests and a further 47 per cent believe mostly, giving a total of 60 per cent of respondents expressing untrusting attitudes. On the other side, 38 per cent believe the government to be mostly run for the benefit of all and only another 2 per cent think that government is entirely run for the benefit of all, making a total of 40 per cent offering a trusting response. Interestingly, the respective proportions of trusting and untrusting responses are almost exactly the same for political trust as for social trust. But does 40 per cent represent a high or a low level of political trust and has this level changed over time? Again, it is difficult to find comparable data sources from the past, but the National Social Science Survey asked a closely comparable question in 1984 and again in 1995 (Kelley, Cushing & Headey 1987; Kelley et al. 1998). Figure 8.2 depicts the three data points. In 1984, 43 per cent of respondents gave a trusting response, falling to 35 per cent in 1995, only to rebound to 40 per cent in 2003. As with interpersonal trust, there is little evidence here of a secular decline in political trust. In fact, the data are quite consistent with the argument that political trust in Australia follows a cyclical pattern based on the age of the incumbent government (Goot 2002; Bean 2001). The 1984 survey was conducted when a new Labor government had recently been elected, the 1995 survey was conducted after this same government had been in office 12 years, and the 2003 survey was conducted when a Coalition government was seven years old. Can we get further insight into political trust from the same sociodemographic variables we explored for social trust? Table 8.2 provides the answer. Again, age does make a difference to levels of trust. Only 31 per cent of 18–34 year olds believe that the government is either mostly or entirely run for the benefit of all. Respondents aged 35–49 years are more likely to be trusting (36 per cent), and 50–64 year olds are even more trusting (44 per cent). Respondents aged 65 and over are the most trusting age group, with 46 per cent believing that the government is either mostly or entirely run for the benefit of all. For political trust, then, we may have evidence of generational replacement and a danger of secular decline. It is more likely, though, that this pattern
26/7/05
128 •
8:55 AM
Page 128
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Figure 8.2 Agree that the federal government is run for the benefit of all, 1984–2003, per cent
50 40
43
40 35
30
%
AusSocAtts03
20 10 0 1984
1995
2003
Year SOURCES National Social Science Survey 1984 (n = 3012), 1995 (n = 2338); The Australian Survey of Social Attitudes 2003 (n = 4270)
reflects the life cycle, with those who have had more experience of politics expressing greater faith that the government is run with the wider good in view, not just narrow interests. As for gender and education, men are only barely more trusting of government than women and although respondents with university education are more likely to express trust than the less educated, the ‘education gap’ is much smaller for political trust than it is for social trust. Mirroring the pattern for interpersonal trust, we can also see class differences in political trust, with the middle class more trusting. But unlike social trust, the class difference for political trust is greater than educational differences. With respect to income, there is a stark difference between those in the highest income band – who are more inclined to exhibit political trust – and all of the lower income groups. Finally, rural and urban dwellers are alike in trusting, or perhaps we should say not trusting, the government. Comparing tables 8.1 and 8.2, we can see that the kinds of people who are socially untrusting are also politically mistrustful. Does this mean that the same people who lack interpersonal trust also lack political trust? Perhaps surprisingly, when we measure the extent to which respondents who are socially untrusting are also politically untrusting, we find that, although social and political trust do vary together, many who express one kind of mistrust do not express the other.1
AusSocAtts03
26/7/05
8:55 AM
Page 129
IS THERE A CRISIS OF TRUST IN AUSTRALIA?
• 129
Table 8.2 Political trust by demographic group, 2003, per cent Federal government run Federal government run for a few big interests for the benefit of all
Age 18–34 (n = 683)
69
31
35–49 (n = 1231)
64
36
50–64 (n = 1231)
56
44
65 and over (n = 931)
54
46
Female (n = 2131)
61
39
Male (n = 1903)
59
41
No university degree (n = 3067)
61
39
University degree (n = 896)
55
45
Sex
Education
Class identification Middle (n = 2030)
52
48
Working (n = 1642)
69
31
None (n = 327)
69
31
Income Under $10 400 (n = 761)
63
37
$10 400 to $25 999 (n = 1119)
61
39
$26 000 to $51 999 (n = 1155)
64
36
$52 000 and over (n = 779)
51
49
Region of residence Rural (n = 1457)
60
40
Urban (n = 2553)
60
40
SOURCE The Australian Survey of Social Attitudes 2003
W H I C H O R G A N I S AT I O N S DO AUSTRALIANS TRUST MOST? Interpersonal trust and trust in government are particularly important when examining trust. However, it is useful to broaden the analysis by considering public attitudes to a range of social, political and economic institutions, all of which may be relevant to evaluating social capital. We can do this by examining measures of confidence (see Lipset & Schneider 1983). Whereas trust is about expressing faith in other people, confidence, at least in the public sphere, is more about expressing faith in how well organisations operate (Bean 2003). AuSSA 2003 asks: ‘How much confidence do you have in the following organisations?’ Twelve organisations or institutions are listed and they, together with the response categories for the question, are shown in table 8.3. The table
AusSocAtts03
26/7/05
130 •
8:55 AM
Page 130
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 8.3 Confidence in institutions, 2003, per cent within rows A great Quite a Not very No deal of lot of much confidence confidence confidence confidence at all
The defence forces
25
57
16
The police in my state (or territory)
13
59
23
2 5
The Australian Broadcasting Corporation (ABC)
16
54
25
4
Universities
11
59
25
5
Charities
9
52
29
10
Major Australian companies
2
40
47
11
The federal parliament
5
36
45
15
Churches or religious institutions
7
28
38
27
The public service
2
30
52
16
The courts and the legal system
4
25
47
24
The unions
3
25
48
24
Banks and financial institutions
2
24
45
29
(n = 4270) SOURCE The Australian Survey of Social Attitudes 2003
lists organisations and institutions in order of the confidence that respondents expressed in them. AuSSA 2003 respondents express very different levels of confidence in different organisations. The defence forces stand out as easily the most supported organisation, with more than four-fifths of respondents expressing either a great deal or quite a lot of confidence. A large majority of respondents also express confidence in the police (72 per cent) and the Australian Broadcasting Corporation (ABC) and universities (both 70 per cent). Charities are also fairly well supported, with 61 per cent expressing at least quite a lot of confidence in them. For no other organisation do a majority of respondents express confidence. At the low end of the scale, banks and financial institutions outscore a number of other close contenders for the organisation in which respondents have least confidence. Only 26 per cent of respondents have confidence in banks and financial institutions and fewer than 30 per cent express confidence in unions or the courts and the legal system. The public service and churches or religious institutions have 32 per cent and 35 per cent respectively. Forty-one per cent of respondents show confidence in the federal parliament and 42 per cent in major Australian companies. Has confidence increased or fallen over time? For most of the institutions AuSSA 2003 asks about, we have comparable data from the
AusSocAtts03
26/7/05
8:55 AM
Page 131
IS THERE A CRISIS OF TRUST IN AUSTRALIA?
• 131
1995 World Values Survey and from the 2001 AES (although in some cases from only one or the other).2 Before considering changes over time, we see immediately in figure 8.3 that the defence forces, police and universities stand out as organisations consistently attracting most confidence. At the other end, banks, financial institutions and the unions are consistently viewed with little confidence, as are, to a lesser degree, the courts and legal system and the public service. Focusing now on changing levels of confidence, only the defence forces appear to inspire a sustained increase in public confidence, and that is from a fairly high base of 68 per cent to over 80 per cent in 2001 and Figure 8.3 Confidence in institutions, 1995, 2001 and 2003, per cent 100 84 82 80 68 60
%
49 40
35 36
40
29
26 26 28
38
31
35
32
20
0 Defence forces
76
80
68 60
73
72
70
59 46
%
Courts and legal Federal parliament Public service system
Unions
43
42
35
40
23
26
20
0 Major Australian companies
1995
Police in my state Churches or (or territory) religious institutions
2001
Banks and financial institutions
Universities
2003
SOURCES World Values Survey 1995 Australian data (n = 2048); Australian Election Study 2001 (n = 2010);
The Australian Survey of Social Attitudes 2003 (n= 4270)
AusSocAtts03
26/7/05
132 •
8:55 AM
Page 132
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
2003. This support may reflect a response to the threat of terrorism in the new global environment. Then again, there are few institutions for which the data show consistent downward trends either. There appears to be lower confidence in the public service, from a modest starting point, and some decline in confidence in churches or religious institutions (although there are only two time points for this item). Only major Australian companies show a large and sustained fall, from 59 per cent of respondents displaying confidence in 1995 to 42 per cent in 2003. Declining confidence in major companies over the last few years may come as little surprise after recent corporate collapses such as OneTel and HIH. However, the recent decline actually continues a substantial decline from high levels of confidence in major companies expressed by the Australian public in the 1980s (Papadakis 1999, p. 76). Not that an earlier decline from the 1980s to the 1990s should be a surprise either, when we think of corporate scandals involving high-profile businessmen such as Alan Bond and Christopher Skase. In fact, the continuing fall in confidence in major Australian companies is a stark example of how the inappropriate actions of a few can taint the public image of a whole sector. In addition to these extreme examples, other factors may have contributed to declining confidence in major companies. Job losses from company restructurings, increasingly large pay packets for chief executives, and expedient relocations of businesses overseas may all have played a part. Confidence in other organisations, such as the unions, the courts and the legal system, the police, banks and financial institutions, and universities show relatively little change, while confidence in the federal parliament has risen and then fallen again. (As I noted earlier, when discussing trust in government, it is likely that these vacillations can be explained by the political cycle.) Another way of considering which organisations Australians trust most is to look at perceptions of the amount of power that they should have. AuSSA 2003 includes the question: ‘Thinking about the amount of power organisations have in Australia today, please say whether you think each of the following should have more power, less power or the same amount’. The organisations in the list are big business, the federal government, the mass media and unions. As table 8.4 reveals, very few respondents (6 per cent) think that big business should have more power, while 62 per cent feel that big business should have less power. An even higher proportion of respondents, however, believe that the mass media should have less power. Twenty-
AusSocAtts03
26/7/05
8:55 AM
Page 133
IS THERE A CRISIS OF TRUST IN AUSTRALIA?
• 133
Table 8.4 More or less power?, 2003, per cent Should the following have:
A lot more power
More power
Same amount of power
Less power
A lot less power
15
Big business
2
4
33
47
The federal government
4
20
56
16
4
The mass media
1
5
24
41
29
Unions
4
15
35
25
21
(n = 4270) SOURCE The Australian Survey of Social Attitudes 2003
nine per cent think that the media should have a lot less power and a further 41 per cent believe that the media should have less power – a combined total of 70 per cent. Views are more balanced about the federal government, with 24 per cent saying it should have more power – somewhat more than think it should have less power (20 per cent), although the majority (56 per cent) think that the federal government has about the right amount of power. With respect to unions, greater numbers favour them having less power (46 per cent) than more power (19 per cent), but over a third are content with the amount of power that unions currently have. It is interesting to compare the contrasting views of unions and big business with similar attitudes expressed in the 1970s and 1980s. Whereas now more people want to constrain the power of big business, then more people believed that unions had too much power (McAllister 1992, p. 90). Of course, there is ample evidence to suggest that unions actually do have less power now: both the proportion of employees who are union members and the working days lost in industrial disputes have fallen precipitously in recent decades (ABS 2004a; 2004b). Thus, the public’s changing perceptions as recorded in survey data reflect changing realities. Overall, table 8.4 depicts an Australian public reluctant to support major organisations having more power than they do already, with many believing that they should have less. These results are consistent with a picture of Australia as a society that is not prepared to place too much trust in its social and political leaders.
H O W D O E S T R U S T R E L AT E T O P O L I T I C A L A N D S O C I A L PA R T I C I PAT I O N ? Social capital is about action as well as attitudes. Indeed, politicians and social analysts are interested in levels of trust because they see trust as
AusSocAtts03
26/7/05
134 •
8:55 AM
Page 134
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
crucial to maintaining a healthy society and polity. AuSSA 2003 data allows us to examine this action in the form of participation in political groups and activities, and to explore any links between such participation and trust. AuSSA 2003 asks respondents: ‘Over the past two years or so, have you done any of the following things to express your views or represent your interests?’ The five activities are: ‘contacted a politician or government official either in person, or in writing, or some other way’, ‘taken part in a protest, march or demonstration’, ‘worked together with people who shared the same concern’, ‘boycotted or bought certain products for environmental, ethical or political reasons’ and ‘participated in a strike or industrial action’. To set the scene, figure 8.4 displays the percentages of AuSSA 2003 respondents who answer ‘yes’ to each of the activities. Clearly, the two activities we might regard as the least political of the five, working with people who share the same concern and boycotting or buying products, are those in which respondents are most likely to have participated. Around half of all respondents report participating in each of these activities in the two years up to 2003. Much smaller proportions of respondents have participated in overtly political activities, such as protest activity (12 per cent) and strike activity (8 per cent). Between these comes contacting politicians or government officials: 32 per cent of respondents report having done this in the past two years. Social capital theory anticipates a close relationship between participation and trust. Certainly, it is reasonable to expect that people who are interpersonally trusting would be more likely to participate in social life than those who are less trusting, since these activities involve co-operating with other people. The picture is less clear for political trust, since past research has shown that those expressing political trust are more likely to engage in system supporting political activities such as voting, but are less likely to engage in a range of other political activities that make some demand on the political system (Bean 1991). In other words, some people engage in some kinds of political activity partly because they do not trust politicians and want to change things, or at least to register their views. The activities that we are considering here have more of that ‘protest’ character and thus may depend on diminished political trust. Table 8.5 provides the answer. The pattern of responses is very consistent across the table and, on the surface, is similar for the two types of trust, though people expressing interpersonal trust are slightly more likely to participate in political activities than are those who express political trust. Respondents expressing interpersonal trust are 7 per cent
AusSocAtts03
26/7/05
8:55 AM
Page 135
IS THERE A CRISIS OF TRUST IN AUSTRALIA?
• 135
Figure 8.4 Participation in political activities in the past two years, 2003, per cent 60 51
48
50 40
%
32 30 20 12 8
10 0 Contacted Taken part in politician or govt protest, march or official demonstration
Worked with Boycotted or Participated in people sharing bought products strike or industrial some concern for environmental, action ethical or political reasons
SOURCE The Australian Survey of Social Attitudes 2003 (n = 4270)
Table 8.5 Participation by interpersonal and political trust, 2003, per cent Interpersonal trust
Political trust
Most people Can’t be can be too careful trusted
Federal Federal government government run for run for big benefit of all interests
Contacted politician or government official
36
29
31
33
Taken part in protest, march or demonstration
17
9
7
15
Worked with people sharing same concern
53
45
47
49
Boycotted or bought products for environmental, ethical or political reasons
59
47
48
55
8
7
5
9
Participated in strike or industrial action (n = 4270) SOURCE The Australian Survey of Social Attitudes 2003
more likely to contact a politician than the untrusting, 8 per cent more likely to take part in a protest, 8 per cent more likely to work with other people sharing the same concern, 12 per cent more likely to boycott or buy products for environmental, ethical or political reasons and perhaps slightly more likely to participate in industrial action. None of these
AusSocAtts03
26/7/05
136 •
8:55 AM
Page 136
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
differences are of great magnitude but, with the exception of industrial activity, they are all quite distinct. The pattern looks similar for political trust (note that the trusting response is in the far right column of the table). Respondents who express political trust display lower levels of participation than those who distrust the government, as we proposed above. The politically trusting are 2 per cent less likely to contact a politician than the politically untrusting, 8 per cent less likely to protest, 2 per cent less likely to work with others, 7 per cent less likely to boycott or buy products and 4 per cent less likely to engage in industrial action. Of course, it is important to emphasise the slight differences between the politically trusting and untrusting. Political trust does not seem to make much difference to whether people participate in political and related activities or not. In addition to questions about political participation, AuSSA 2003 also contains questions about participation in voluntary groups and organisations and it is this kind of social participation that is seen as a particularly vital indicator of social capital (Putnam 2000; see also chapter 5). AuSSA 2003 asks about 15 different groups and organisations, in two separate lists, introduced by the question: ‘We would like to ask you if you are a member of any of the groups and organisations listed below’.3 Table 5.2 in chapter 5 shows the membership levels of voluntary groups and organisations. Levels of participation range from 54 per cent of respondents reporting that they belong to a consumer or automobile organisation, to only 4 per cent being members of a lobby group or a political party. In between, 45 per cent belong to a sporting or recreation group, 30 per cent to a financial co-operative, 24 per cent to a religious group, 20 per cent to a neighbourhood or community-based group or to a union, 15 per cent to a professional society, 14 per cent to a group that helps people with special needs, 10 per cent to an environmental group or aid organisation, 10 per cent to an art, music or educational group, 8 per cent to a self-help/consumer health group, 7 per cent to a group working to improve the environment and 6 per cent to a group that promotes rights. Table 8.6 shows how membership of these organisations varies with interpersonal and political trust. Focusing first on interpersonal or social trust, we see that, despite the wide variation in levels of membership of the different groups, the pattern is the same as for participation in political activities. In all 15 cases, those who say most people can be trusted are more likely to participate in voluntary organisations than those who believe that you cannot be too careful in dealing with people. Again,
AusSocAtts03
26/7/05
8:55 AM
Page 137
IS THERE A CRISIS OF TRUST IN AUSTRALIA?
• 137
Table 8.6 Membership by interpersonal and political trust, 2003, per cent Interpersonal trust
Political trust
Most people Can’t be Federal Federal can be too careful government government trusted run for run for big benefit of all interests
Unions
22
18
16
23
Political party
4
3
5
3
Lobby group
6
3
4
4
Group that promotes rights
9
3
4
7
Environmental or aid group
15
7
11
11
Group working to improve the environment
10
5
8
7
Neighbourhood or community group
23
19
22
20
Sporting or recreation group
51
44
49
46
Self-help/consumer health group
9
8
10
7
Group that helps people with special needs
15
14
15
14
Art, music, or education group
13
8
10
11
Religious group
26
23
30
20
Professional society
23
10
20
13
Consumer or automobile organisation
64
51
61
54
Financial co-operative
33
29
28
32
(n = 2087) SOURCE The Australian Survey of Social Attitudes 2003
none of these differences are extraordinary – the largest is 13 per cent for membership of consumer or automobile organisations – and several of them are only 1 per cent or 2 per cent, but the tendency is consistent for the more trusting to be more inclined to participate in voluntary organisations. Results for political trust are more mixed. In the case of several groups with a more overt political character, such as unions and rights groups, the untrusting are more likely to participate than the trusting, mirroring results for political participation discussed above. But in most other cases where some difference is evident, table 8.6 indicates that political trust, as well as social trust, facilitates participation in voluntary groups and organisations. Those who are politically trusting are clearly more likely to participate in religious groups, professional societies and consumer or automobile groups, for instance. Again, however, as for political participation, the margins between the politically trusting and untrusting are relatively slim for the most part and in some cases there is no difference at all.
AusSocAtts03
26/7/05
138 •
8:55 AM
Page 138
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
CONCLUSION Is there a crisis of trust in Australia? And what does this tell us about the state of social capital in Australia? If a crisis is a decisive turning point, the empirical evidence reviewed in this chapter provides no indication of any current crisis. That is not to say, however, that Australia can be described as a healthy society with high levels of trust and, in turn, of social capital. The survey data we have considered indicate that only a minority of the adult population express social and/or political trust, a trend evident for some time. We might conclude, then, that there is something of a deficit of trust in Australian society. But the evidence that trust is in further decline is slim. In this sense, Australian findings better reflect the ‘no change’ scenario that Johnston and Jowell (2001) established for the United Kingdom rather than the darker picture that Putnam (2000) paints for the United States. Our findings on social trust among different age cohorts reinforce this conclusion. In America, older generations are consistently more trusting and the youngest generation is consistently the least trusting. But, like the United Kingdom, in Australia there is a curvilinear relationship between age and social trust: mistrust among the young seems to give way to greater trust among the middle-aged, with trust declining again among the older cohorts, leaving the old and the young with similar levels of trust. Perhaps the old and the young share a lesser sense of control over their lives than the middle aged. This suggests that social trust is not likely to decline in Australia simply because of generational change. The analysis of social trust by education is also very revealing. On the one hand, it is somewhat disturbing that those who lack a university education – the majority of the population – are less inclined to express trust in others. Among other things, this hints that socioeconomic divisions are behind differences in interpersonal trust, an observation reinforced by the correspondence between class identification and income differences with levels of trust. As long as such social divisions persist, it is likely those in more disadvantaged groups will continue to feel less trusting of others. Yet, on the other hand, the high level of social trust expressed by university-educated respondents provides cause for optimism about the future of social capital. University education is becoming more widespread and these findings imply that those who acquire greater knowledge and understanding through education are more likely to be comfortable putting faith in other people.
AusSocAtts03
26/7/05
8:55 AM
Page 139
IS THERE A CRISIS OF TRUST IN AUSTRALIA?
• 139
On the basis of the evidence that I have discussed here and elsewhere (Bean 2001), social or interpersonal trust seems to be a better indicator of social capital than political trust or trust in government. Social trust promotes greater levels of political and group participation, even if the differences between the trusting and the untrusting are relatively modest, while people who express political trust are in many instances less likely to participate.4 This chapter also shows that public confidence in different institutions varies widely, from the defence forces, which are highly regarded, to banks and financial institutions, in which the public manifestly lack confidence. Both the most and the least well regarded institutions tend to keep their rankings over time. The most disturbing finding about confidence in institutions is the sharp decline in public confidence in major Australian companies, a trend evident for 20 years and which may not yet have bottomed out. Misdemeanours of high-flying corporate figures, which may have provoked this downturn in public sentiment, do little to bolster the maintenance and regeneration of social capital in Australia. NOTES
1 2
3 4
This ‘overlap’ between the two groups can be measured using the Pearson product moment correlation, which gives a coefficient of 0.14 (on a scale where 0 indicates no association at all and 1 indicates a perfect association). There are minor wording variations in the descriptions of some of the items between AuSSA 2003 and the other two surveys (for example, the term used for ‘the defence forces’ in 1995 and 2001 is ‘the armed forces’ and ‘trade unions’ is used instead of ‘the unions’) but not to the extent that comparability is likely to be impaired. Note that these questions were asked of approximately half of the AuSSA 2003 sample. The data were supplied to the author by Andrew Passey and Mark Lyons (see also chapter 5). Political trust as measured here, however, is less about a broader sense of trust in the political system generally and more about public evaluations of the people who are running the government of the day. This makes political trust sensitive to the ups and downs of the political cycle and thus not a good indicator of social capital, which is clearly a concept with a broader referent.
REFERENCES
Australian Bureau of Statistics (ABS) (2004a) Australian Labour Market Statistics, cat. no. 6105.0, Australian Bureau of Statistics, Canberra, April. —— (2004b) Australian Labour Market Statistics, cat. no. 6105.0, Australian Bureau of Statistics, Canberra, July. Bean, C (1991) ‘Participation and political protest: A causal model with Australian evidence’, Political Behavior, 13: 253–83. —— (2001) ‘Party politics, political leaders and trust in government in Australia’, Political Science, 53 (1): 17–27. —— (2003) ‘Citizen confidence in social and political institutions in a changing world’, Proceedings of the Social Change in the 21st Century Conference, Queensland University of Technology, Brisbane, November.
AusSocAtts03
26/7/05
140 •
8:55 AM
Page 140
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
—— Gow, D & McAllister, I (2002) Australian Election Study, 2001: User’s Guide for the Machine-Readable Data File, Australian Social Science Data Archive, Australian National University, Canberra. Braithwaite, V & Levi, M (eds) (1998) Trust and Governance, Russell Sage Foundation, New York. Burchell, D & Leigh, A (eds) (2002) The Prince’s New Clothes: Why Do Australians Dislike their Politicians? UNSW Press, Sydney. Costello, P (2003) ‘Let’s revive a sense of society’, The Age, 17 July. Gibson, R, Wilson, S, Meagher, G, Denemark, D & Western, M (2004) Australian Survey of Social Attitudes 2003, Data file, Australian Social Science Data Archive, The Australian National University, Canberra. Goot, M (2002) ‘Distrustful, disenchanted and disengaged? Public opinion on politics, politicians and the parties: An historical perspective’, in David Burchell & Andrew Leigh (eds) Prince’s New Clothes: Why Do Australians Dislike their Politicians? UNSW Press, Sydney, pp. 9–46. Johnston, M & Jowell, R (1999) ‘Social capital and the social fabric’, in Roger Jowell, John Curtice, Alison Park & Katarina Thomson (eds) British Social Attitudes: The 16th Report – Who Shares New Labour Values? Ashgate, Aldershot, pp. 179–200. —— (2001) ‘How robust is British civil society?’, in Alison Park, John Curtice, Katarina Thomson, Lindsey Jarvis & Catherine Bromley (eds) British Social Attitudes: The 18th Report – Public Policy, Social Ties, Sage Publications, London, pp. 175–97. Kelley, J, Cushing, RG & Headey, B (1987) National Social Science Survey, 1984: User’s Guide, Australian Social Science Data Archive, Australian National University, Canberra. Kelley, J, Evans, MDR, Bean, C & Zagorski, K (1998) International Social Science Surveys/Australia 1995/96 Codebook: National Social Science Survey Module, Academic Research Modules, and the International Social Survey Programme’s National Identity, Round 1, Research School of Social Sciences, Australian National University, Canberra. Latham, M (1998) Civilising Global Capital: New Thinking for Australian Labor, Allen & Unwin, Sydney. Lipset, SM & Schneider, W (1983) The Confidence Gap: Business, Labor, and Government in the Public Mind, Free Press, New York. McAllister, I (1992) Political Behaviour: Citizens, Parties and Elites in Australia, Longman Cheshire, Melbourne. Nye, JS, Zelikow, PD & King, DC (eds) (1997) Why People Don’t Trust Government, Harvard University Press, Cambridge, Massachusetts. Papadakis, E (1999) ‘Constituents of confidence and mistrust in Australian institutions’, Australian Journal of Political Science, 34(1): 75–93. Pusey, M (2003) The Experience of Middle Australia: The Dark Side of Economic Reform, Cambridge University Press, Cambridge. Putnam, RD (1993) Making Democracy Work: Civic Traditions in Modern Italy, Princeton University Press, Princeton, New Jersey. —— (1995) ‘Bowling alone: America’s declining social capital’, Journal of Democracy, 6 (1): 65–78. —— (2000) Bowling Alone: The Collapse and Revival of American Community, Simon & Schuster, New York. Warren, ME (ed.) (1999) Democracy and Trust, Cambridge University Press, Cambridge.
AusSocAtts03
26/7/05
8:55 AM
Page 141
CHAPTER 9
PERCEPTIONS OF CRIME AND JUSTICE D a v i d I n d e r m a u r a n d Ly n n e R o b e r t s
INTRODUCTION Crime and justice are a mainstay of media and popular discussion in Australia – few topics reflect the moral sensibilities of a nation as much as crime and punishment. Beliefs and opinions about crime are shaped by widespread consumption of news and entertainment about crimes, police, courts and prisons. Attitudes about crime and justice reflect judgments about social threats and harms, and perceptions of personal safety as well as beliefs about criminal offenders. Over the last 20 years, advanced democracies have become preoccupied with crime and embraced greater levels of control.1 This movement has been described as a ‘culture of control’ by David Garland (2001), one of the foremost commentators in the field. Garland draws on contemporary research on attitudes to crime and justice, which reveal two competing emotional responses: anger and fear. As Roberts et al. (2003) point out, both media and politicians have a role to play in this culture, exploiting fear and misunderstandings about crime for votes and ratings.2 Gauging attitudes to crime and justice is a well-established area of study in both North America3 and Europe.4 Perceptions of crime and the criminal justice system in Australia have been the subject of a limited number of surveys and scholarly reviews since the 1970s (see Wilson & Brown 1973; Broadhurst & Indermaur 1982; Walker, Collins & Wilson 1988; Indermaur 1987, 1990). These studies have measured general punitiveness of citizens, the relative seriousness of crimes, and knowledge about crime and justice as well as testing participant response to additional information (Indermaur 1987, 1990).
AusSocAtts03
26/7/05
142 •
8:55 AM
Page 142
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Two main findings emerge out of Australian research on the public’s knowledge and perceptions of crime and criminal justice. First, research shows that most Australians are dissatisfied with current sentencing practices and are punitive about sentencing. Second, the public has a distorted view of crime and punishment, overestimating the level of violent crime and underestimating the current severity of sentences (Indermaur 1987, 1990; Weatherburn, Matka & Lind 1996; Weatherburn & Indermaur 2004). Although only a small proportion (less than 10 per cent) of crimes recorded by the police involve violence, most of the time respondents answer questions about sentencing with serious violent offenders in mind.5 These findings have also been made for other advanced democracies. In this chapter, we explore Australian perceptions of crime and criminal justice using questions in the Australian Survey of Social Attitudes 2003 (Gibson et al. 2004). The National Social Science Survey and the Australian Election Study (AES) are general surveys that also contain crime-related questions that track attitudes about crime and justice. Here, we use all these surveys to give us an idea of how national attitudes have changed over time. This chapter focuses on public perceptions of crime, and how the media influences these perceptions. We seek to discover how accurate public perceptions of crime rates are, and how respondents’ media use might influence the accuracy of perceptions. We then examine attitudes about integrity of key institutions of criminal justice, particularly confidence in the police force, the courts and current sentencing practices. We get some clear focus on the latter by looking at bellwether topics like decriminalising marijuana use and the death penalty.
ARE PERCEPTIONS OF C R I M E R AT E S A C C U R AT E ? AuSSA 2003 asks respondents if they believe that crime has increased or decreased or stayed the same over the past two years. This question allows us to test the accuracy of public knowledge about crime trends. The results (see table 9.1) indicate clearly that more than two-thirds of survey respondents (70 per cent) believe that crime has increased over the past two years, with more than a third overall (39 per cent) responding that it had increased ‘a lot’. We can compare these results with actual crime trends reported to the police over the two-year period from 2001 to 2003.6 The Australian Bureau of Statistics publishes national and state-level data on nine major
AusSocAtts03
26/7/05
8:55 AM
Page 143
PERCEPTIONS OF CRIME AND JUSTICE
• 143
Table 9.1 Perceptions of crime trends between 2001 and 2003 by state, per cent Increased Increased Stayed Decreased Decreased a lot a little the same a little a lot
Can’t choose
Western Australia (n = 380)
54
30
12
2
2
1
New South Wales (n = 1374)
44
32
19
2
1
2
Queensland (n = 681)
42
30
22
2
1
2
South Australia (n = 370)
41
37
17
2
1
2
ACT (n = 76)
34
30
25
5
0
5
Northern Territory (n = 15)
33
33
13
7
7
7
Tasmania (n = 119)
31
34
23
9
1
3
Victoria (n = 1065)
26
29
29
10
1
5
Total (n = 4123)
39
31
21
4
1
3
SOURCE The Australian Survey of Social Attitudes 2003
categories of crimes reported to the police. The national crime rate per 100 000 persons for each of the nine categories for 2001 and 2003 is presented in table 9.2. For six of the nine major categories of criminal offences, police-recorded rates of crime in Australia declined between 2001 and 2003. Recorded rates of crime fell for homicide and related offences, kidnapping/abduction, robbery, unlawful entry with intent, motor vehicle theft and other theft. Marginal increases in rates of crime were recorded for three categories of offences usually not reported to police: assault (2 per cent), sexual assault (5 per cent) and blackmail/extortion (6 per cent) (ABS 2004). The pattern that emerges, then, is one of decreasing crime: almost all categories recorded declines with significant falls in the high-incidence category of property crime.7 The relatively small number of cases of the most serious crime – homicide – also dropped.8 Given these trends, the most accurate response to the question would be to indicate a decrease in crime levels during the past two years. However, only one in 20 respondents hold this assessment of criminal activity. Using a slightly more generous measure of respondent accuracy – that crimes are falling or stable – we still find that only a quarter of respondents hold accurate views about crime trends.
AusSocAtts03
26/7/05
144 •
8:55 AM
Page 144
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 9.2 Police recorded crime rates (per 100 000 persons) in Australia by offence category, 2001 and 2003 Rate per 100 000 persons Offence category
Motor vehicle theft
2001
2003
720.7
497.1
–31
137
99.2
–28
2 244.9
1 777.9
–21
5.5
4.7
–12
3 607
3 214.3
–11
Robbery Unlawful entry with intent Homicide and related offences Other theft Kidnapping/abduction
4
3.6
–10
1.8
1.9
+6
87.1
91.7
+5
784.5
798
+2
Blackmail/extortion Sexual assault Assault
Change (per cent)
SOURCE Australian Bureau of Statistics (2004)
Perceptions of crime trends vary substantially across states, with Western Australians least accurate in their perceptions and Victorians most. One possible explanation for the difference between actual crime trends and the public perceptions may be differences in crime trends at the state level. We test this by comparing trends in crime rates by state. For each state with full crime data available, we calculated the number of major categories (out of a possible nine) with declines in recorded crimes. The results are presented in table 9.3, along with the corresponding percentage of respondents who think that crime trends are stable or decreasing. The results indicate no consistent relationship between actual and perceived crime trends across states. For example, Western Australia, one of the states with the most categories of crime decreasing over the two-year period, has the lowest proportion of respondents who believe that crime is stable or decreasing. The absence of any Table 9.3 Crime trends versus perceptions of crime trends over same period, 2001–2003
State
Categories of crime decreasing
Perception that crime is stable or decreasing
n
per cent
Western Australia
8
15
Victoria
8
40
Queensland
7
25
South Australia
6
20
New South Wales
6
22
SOURCES Australian Bureau of Statistics (2004); The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
8:55 AM
Page 145
PERCEPTIONS OF CRIME AND JUSTICE
• 145
relationship reinforces our earlier claim that public perceptions of crime rates are generally not based on actual crime rates. AuSSA 2003 also asks respondents whether they agree or disagree with the statement that ‘Immigrants increase crime rates’. This question is useful for tapping prejudices as well as the misperceptions in the community that drive attitudes to crime. The most authoritative and recent Australian research on this topic (Mukherjee 1999) reviewed the research in this area, finding no direct link between immigration and crime. Studies in the United States have also found no relationship between the number of immigrants and crime rates (Butcher & Piehl 1998). Indeed, in direct contrast to the view that immigrants increase crime rates, there is some evidence that immigrants are less likely to be criminally active than native-born Americans (Bucher & Piehl 1998; Mears 2001). Despite little evidence supporting a link between immigration and crime, almost a third (31 per cent) of respondents agree that immigrants increase crime rates, with 41 per cent disagreeing or strongly disagreeing (see table 9.4). (See also chapter 11 for attitudes to immigration.) Table 9.4 Agreement that immigrants increase crime rates by demographic characteristics, 2003, per cent Strongly agree
Agree
Neither Disagree agree nor disagree
Strongly disagree
Can’t choose
Sex Female (n = 1070)
10
19
24
28
16
4
Male (n = 956)
12
22
24
25
14
3
18–34 (n = 335)
9
11
30
27
20
2
35–49 (n = 625)
10
17
24
28
20
2
50–64 (n = 609)
10
24
20
29
14
3
65 and over (n = 449)
15
26
26
21
7
6
Less than Year 12 (n = 648)
15
24
25
23
8
4
Year 12 (n = 225)
9
18
25
29
16
3
17
24
27
25
5
3
Certificate/Diploma (n = 427)
9
23
26
25
13
3
Bachelor’s degree & above (n = 447)
3
11
18
33
33
3
11
20
24
27
15
3
Age
Education
Trade/Apprenticeship (n = 239)
Total (n = 2058)
SOURCE The Australian Survey of Social Attitudes 2003
26/7/05
146 •
8:55 AM
Page 146
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Respondent views about the relationship between immigrants and crime strongly correspond to general perceptions of crime rates. Those who believe that crime rates have remained stable or decreased are also less likely to agree that immigrants increase crime: just 14 per cent agree with this proposition. Among those who believe that crime has increased, 37 per cent agree that immigrants increase crime (see figure 9.1). We further investigate perceptions of crime by particular demographic groups in Australian society and find that men, younger respondents, and the more highly educated hold more accurate perceptions of actual crime trends (see table 9.5). This pattern is broadly similar to results shown in table 9.4, which reports agreement with the statement that immigrants increase crime rates by demographic groups. However, men, who generally hold more accurate perceptions of crime trends, are slightly more likely to agree that immigrants increase crime rates. Again, younger and more qualified respondents are less likely to link immigrants with crime, with only two in ten of those aged 18–34 agreeing with the proposition (compared to four in ten aged 65 and over), and just one in seven of those with a university
Figure 9.1 Agreement that immigrants increase crime rates by perceptions of crime rates, 2003, per cent
Increased
Crime rates
AusSocAtts03
Stayed the same
Agree Neutral Disagree
Decreased
Can't choose
0
20
40
60
80
Immigrants increase crime (%) SOURCE The Australian Survey of Social Attitudes 2003
100
AusSocAtts03
26/7/05
8:55 AM
Page 147
PERCEPTIONS OF CRIME AND JUSTICE
• 147
Table 9.5 Perception of crime by demographics, 2003, per cent Increased Increased Stayed Decreased Decreased a lot a little the same a little a lot
Can’t choose
Sex Female (n = 2164)
43
30
19
3
1
4
Male (n = 1928)
34
32
25
5
1
2
18–34 (n = 709)
25
35
30
5
1
4
35–49 (n = 1254)
36
32
23
4
1
3
50–64 (n = 1248)
41
29
21
5
2
3
65 and over (n = 870)
51
30
13
3
1
2
Less than Year 12 (n = 1279)
52
27
14
2
1
3
Year 12 (n = 471)
34
34
23
4
2
3
Trade/Apprenticeship (n = 493)
38
33
21
6
1
1
Certificate/Diploma (n = 874)
40
33
19
5
1
2
Bachelor’s degree & above (n = 903)
20
35
35
6
1
4
Age
Education
SOURCE The Australian Survey of Social Attitudes 2003
degree agreeing (compared to four in ten not completing Year 12 or holding a trade certificate). These results confirm findings from previous surveys conducted in New South Wales and Western Australia (Weatherburn & Indermaur 2004) that suggest the public is not well informed about actual trends in crime, either at a nation or state level. Older and less qualified respondents hold the most inaccurate perceptions of criminal activity in Australia. Evidence suggests that what drives these general perceptions is not the reality of crime but the reporting of crime (see Roberts & Stalans 1997; Chiricos, Eschholz & Gertz 1997). This is not surprising, since the media is the public’s main source of information on crime.9 So any understanding of public perceptions must equally examine the ways media treats crime, and how the public assimilates the reporting of crime.
AusSocAtts03
26/7/05
148 •
8:55 AM
Page 148
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
HOW DOES MEDIA USE AFFECT BELIEFS ABOUT CRIME TRENDS? To understand the interaction between the media and crime perceptions, we must first examine basic data about how respondents use the media. Previous research points to the influence of the media on perceptions of crime and justice.10 As we show in table 9.6, the accuracy of perceptions about crime varies according to the main source of media that respondents rely on. Those who rely on Internet sites for their news and information are most likely to believe that crime rates are stable or falling, while those who rely on talkback radio, family and friends or commercial television are least likely to believe this. Agreement that immigrants increase crime rates is highest among respondents who rely on talkback radio (50 per cent) and commercial television (37 per cent) as their main source of news and information. Television is likely to contribute to any public misperception about crime levels and criminals because of its overall popularity (see chapter 13) and is particularly so among respondents with fewer qualifications. Of course, unpacking the ‘cause and effect’ of media is complex. While the media may feed negative stories and images to viewers, individuals are likely to seek out media sources that accord with their preexisting opinions and beliefs. What is most likely, then, is that individual views and audience-maximising media programming amplify each other through synergistic selection and reinforcement. Table 9.6 Perceptions of crime by major source of news and information, 2003, per cent Media relied on most for news and information
Per cent with accurate perceptions of crime rates
Per cent disagree that immigration increases crime rates
Internet sites
44
67
ABC and or SBS television
34
55
Radio
31
41
Newspapers and news magazines
29
41
Talkback radio
21
25
Commercial television
20
33
Friends and family
17
55
Total
27
42
(4075)
(2034)
n SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
8:55 AM
Page 149
PERCEPTIONS OF CRIME AND JUSTICE
• 149
W H AT S H O U L D C O U N T A S A C R I M E ? Australians hold divergent views about what behaviour is acceptable, and defining behaviour as criminal is often fiercely contested. These debates are nowhere more intense than over ‘victimless’ crimes, such as those that involve the personal choice to consume drugs. One debate along these lines concerns the personal use of marijuana. AuSSA 2003 asks respondents whether they agree that smoking marijuana should not be a criminal offence. Almost a third (32 per cent) agree. This figure is particularly interesting: official statistics indicate that about one-third of Australians have used marijuana, with approximately 1.2 million Australians using the drug in any one month (Australian Institute of Health and Welfare 2002). A further one in five AuSSA 2003 respondents do not have a view either way on decriminalisation, and about half (49 per cent) disagree. Comparing this with earlier survey findings, these results indicate that support for decriminalising marijuana use has remained relatively steady over the past decade after a slight dip in support in the second half of the 1980s (see figure 9.2). One controversy in the marijuana decriminalisation debate is about whether prohibition deters new marijuana users and moderates consumption among existing users (Weatherburn & Jones 2001; Weatherburn, Jones & Donnelly 2003). Across Australia, all states Figure 9.2 Agree that smoking marijuana should not be a criminal offence, 1984–2003, per cent 100 90 80
Agree (%)
70 60 50 40 30 20 10 0 1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
SOURCES National Social Science Survey (Kelley et al. 1984, 1986–87, 1987–88, 1989–90); Australian
Election Study (McAllister et al. 1990; Jones et al. 1993, 1996; Bean et al. 1999, 2002); The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
150 •
8:55 AM
Page 150
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 9.7 Agree that smoking marijuana should not be a criminal offence by state, 2003, per cent State
Penalties
Agree
Neither agree nor disagree
Disagree
civil
48
13
39
diversion to treatment
34
19
47
civil*
34
17
49
Tasmania (n = 119)
cautioning
32
26
42
Victoria (n = 1083)
cautioning
31
21
48
civil
31
21
48
New South Wales (n = 1400)
cautioning
30
17
53
Northern Territory (n = 15)
civil
27
33
40
32
19
49
ACT (n = 77) Queensland (n = 700) Western Australia (n = 383)
South Australia (n = 375)
Australia (n = 4152) * cautioning and diversion at time survey was conducted SOURCE The Australian Survey of Social Attitudes 2003
prohibit marijuana use; however, the penalties for smoking marijuana vary widely. Four states and territories (South Australia, Australian Capital Territory, Northern Territory and Western Australia) now provide civil rather than criminal penalties for minor cannabis offences. Three states (Tasmania, Victoria and New South Wales) provide formal cautions for the first two or three offences if only small quantities of cannabis are involved. Queensland provides mandatory assessment and a brief intervention session (Lenton 2004). Are these differences reflected in state-level public opinion? When we examine support for the decriminalisation of marijuana use by state, we find no consistent relationship between penalties and public opinion (see table 9.7). It is likely, therefore, that reforms to these laws are a product of political leadership rather than the pressure of public opinion. Conversely, legal reform is perhaps yet to shift public attitudes.
H O W S AT I S F I E D A R E A U S T R A L I A N S WITH THE POLICE? Police play a leading role in the criminal justice system, responsible for protecting the community, maintaining public order, enforcing criminal
AusSocAtts03
26/7/05
8:55 AM
Page 151
PERCEPTIONS OF CRIME AND JUSTICE
• 151
laws and investigating crimes. Public confidence in the police is especially important because police rely on the public’s co-operation and assistance to undertake their work. But how satisfied are Australians with their police forces? As we shall see below, Australian attitudes to the police are quite contradictory. AuSSA 2003 asks two questions about attitudes to the police. First, respondents are asked how much confidence they have in the police in their state or territory. The results indicate that one in eight (13 per cent) have ‘a great deal’, and more than half (57 per cent) have ‘quite a lot’ of confidence in the police in their jurisdiction. A further two in ten (23 per cent) have ‘not very much’ confidence and only one in 20 (5 per cent) have ‘no’ confidence at all. The results of a second question asking about police corruption, however, provide a different perspective. Only one in four disagree that there is a lot of corruption in their state/territory police force. No doubt these findings reflect media coverage of numerous royal commissions and inquiries into police forces over the past two decades, including the Fitzgerald Inquiry into Police and Political Corruption in Queensland in 1987, the Wood Royal Commission into the New South Wales Police in 1997 and the Kennedy Royal Commission into corrupt or criminal conduct by Western Australian Police Officers in 2004. Findings of deep-seated corruption across commissions and enquiries led Kennedy (2004, p. 56) to remark that ‘police corruption is an inevitable and universal characteristic of a police service’. Perceptions of corruption vary by state, with respondents in Western Australia (44 per cent) and New South Wales (37 per cent) most likely to agree that there is a lot of corruption in the police force in their states. The same states also record the lowest levels of confidence in the police (63 per cent and 65 per cent respectively). The two territories record the highest confidence (80 per cent each), and the lowest levels of perceived corruption (the Northern Territory 13 per cent and the ACT 18 per cent).11 So overall, while only a quarter of respondents see the police as free of corruption, three-quarters maintain confidence in the police as a whole. Such findings indicate that, although people may now be primed to see police corruption as endemic, they do not consider police forces to be completely compromised. The idea that there is a reservoir of public trust in the police is confirmed by findings reported in chapter 8. There, Clive Bean reports that public confidence in police is exceeded only by confidence in the defence forces and is indeed
AusSocAtts03
26/7/05
152 •
8:55 AM
Page 152
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 9.8 Perceptions of corruption in police by confidence in police, 2003, per cent A lot of corruption in police Confidence in police
Agree
Neither agree nor disagree
Disagree
A great deal of confidence
15
35
50
Quite a lot of confidence
21
51
28
Not very much confidence
51
42
7
No confidence at all
79
18
4
Can’t choose
26
70
4
Total
30
46
24
(n = 4118) SOURCE The Australian Survey of Social Attitudes 2003
higher than the confidence enjoyed by universities or the ABC. Despite this, police forces should probably avoid complacency. When we examine the relationship between perceptions of corruption and confidence in the police, we find that these views are inversely related – as perceptions of corruption decline, confidence in police increase. Table 9.8 shows that while one in seven respondents with a great deal of confidence in the police believe that there is a lot of corruption in police ranks, this jumps to over half among those with ‘not very much’ confidence in the police.
A R E A U S T R A L I A N S S AT I S F I E D WITH SENTENCING? The public confidence enjoyed by the police contrasts strongly with respondent views of the courts. The majority (almost three-quarters) of respondents indicate that they have ‘not very much’ (46 per cent) or ‘no’ (24 per cent) confidence in the courts and legal system. This is on a par with banks and financial institutions, and indeed much less than major Australian companies (see chapter 8). This finding points to a broader confidence problem for the courts and the legal system, which casts a shadow over their legitimacy. We find the same low level of confidence in responses to statements about sentencing (see table 9.9). While the majority of respondents (63 per cent) agree that judges should reflect public opinion about crimes when sentencing criminals, seven out of ten believe that those who break the law should be given stiffer sentences, and almost half agree that the death penalty should be the punishment for murder. The public also
AusSocAtts03
26/7/05
8:55 AM
Page 153
PERCEPTIONS OF CRIME AND JUSTICE
• 153
Table 9.9 Attitudes to sentencing and punishment, 2003, per cent
Judges should reflect public opinion when sentencing People who break the law should be given stiffer sentences The death penalty should be the punishment for murder
Strongly agree
Agree
Neither Disagree agree nor disagree
Strongly disagree
Can’t choose
31
32
15
18
5
2
31
39
21
6
2
2
23
24
20
18
15
3
(n≥ 4195) SOURCE The Australian Survey of Social Attitudes 2003
Table 9.10 When big businesses break the law, they often go unpunished, 2003, per cent Strongly agree
40
Agree
41
Neither agree nor disagree
9
Disagree
5
Strongly disagree
3
Can’t choose
2
(n = 4203) SOURCE The Australian Survey of Social Attitudes 2003
questions treatment of institutional crimes: people are particularly suspicious of the leniency shown to illegal activities by big business, with a large majority believing that their crimes go unpunished (see table 9.10). This may be a by-product of falling confidence in major Australian companies reported in chapter 8. Those who agree that judges should reflect public opinion when sentencing are about twice as likely to agree with stiffer sentences (see figure 9.3), and that death should be the penalty for murder (figure 9.4). In other words, Australians who want sentencing to reflect public opinion tend to be the same people seeking harsher sentences and the death penalty. Not surprisingly, the relationship between stiffer sentences and the death penalty is even stronger: those seeking stiffer sentencing are five to six times more likely than those who do not seek stiffer sentencing to support the death penalty for murder.
AusSocAtts03
26/7/05
154 •
8:55 AM
Page 154
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Judges should reflect public opinion when sentencing
Figure 9.3 Agreement with stiffer sentences by agreement that judges should reflect public opinion in sentencing, 2003, per cent
Agree Agree Neutral
Neutral
Disagree
Disagree
0
20
40
60
80
100
People who break the law should be given stiffer sentences (%)
SOURCE The Australian Survey of Social Attitudes 2003
Judges should reflect public opinion when sentencing
Figure 9.4 Agreement with death penalty for murder by agreement that judges should reflect public opinion in sentencing, 2003, per cent
Agree
Agree Neutral
Neutral Disagree
Disagree
0
20
40
60
80
100
The death penalty should be the punishment for murder (%) SOURCE The Australian Survey of Social Attitudes 2003
In sum, our findings on sentencing and the courts and legal system paint a dim picture. About three-quarters of AuSSA 2003 respondents have little or no confidence in the legal system’s performance and about the same proportion want stiffer sentences. And half of AuSSA 2003 respondents would reinstate the death penalty for murder, and more than half believe that sentencing should reflect public opinion. However, neither the call for stiffer sentences nor the death penalty is new. Public opinion on stiffer sentencing and the death penalty has been sought on ten occasions over the past two decades. The trends
AusSocAtts03
26/7/05
8:55 AM
Page 155
PERCEPTIONS OF CRIME AND JUSTICE
• 155
over time are plotted in figures 9.5 and 9.6, and reveal something interesting: the proportion of Australians who agree with stiffer sentences has declined from a peak in 1987. And agreement that the death penalty should be the punishment for murder actually peaked in 1993 – a time Figure 9.5 Agree that people who break the law should be given stiffer sentences, 1984–2003 100 90 80
Agree (%)
70 60 50 40 30 20 10 0 1983
1985
1987
1989 1991
1993
1995 1997
1999 2001
2003
SOURCES National Social Science Survey (Kelley et al. 1984, 1986–87, 1987–88, 1989–90); Australian Election Study (McAllister et al. 1990; Jones et al. 1993, 1996; Bean et al. 1999, 2002); The Australian Survey of Social Attitudes 2003
Figure 9.6 Agree that the death penalty should be the punishment for murder, 1984–2003 80 70
Agree (%)
60 50 40 30 20 10 0 1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
SOURCES National Social Science Survey (Kelley et al. 1984, 1986–87, 1987–88, 1989–90); Australian Election Study (McAllister et al. 1990; Jones et al. 1993, 1996; Bean et al. 1999, 2002); The Australian Survey of Social Attitudes 2003
2003
AusSocAtts03
26/7/05
156 •
8:55 AM
Page 156
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
when political and media attention to crime across English-speaking countries was extraordinary (Roberts et al. 2003). However, in line with findings from other countries, it appears that Australians hold slightly less punitive attitudes to criminal behaviour compared with the 1980s and 1990s (Roberts et al. 2003). Surprisingly, this change has taken place despite a continuing emphasis on law and order in the media and politics. Perhaps the public has reached ‘saturation’ point with constant media attention to crime and sentencing, and has become somewhat desensitised to the issue. Many people might also have become cynical about the inflammatory tone of much of the political debate on this topic. What other factors can account for differences in opinion about sentencing? Responses to the three survey questions on stiffer sentencing, the death penalty and sentences reflecting public opinion are, in fact, so closely related that we can combine them in a scale that reflects the level of punitiveness or preference for punishment.12 Using this scale, we can analyse the overall factors that shape attitudes. The results show that a tendency to favour punishment increases with age – older age groups are most likely to favour tougher punishment. There are also minor gender differences: women are more likely to favour tougher punishment. University-educated respondents are less supportive of a tough approach. Punitive attitudes among the elderly and women may point to greater personal and social vulnerability experienced by these groups. Fear of crime is also reflected in greater levels of routine precautions taken by the elderly to protect themselves (Pinkerton James 1992). The link between vulnerability and punitiveness suggests that many community members think that increasing the sentences will reduce crime. However, research does not support this belief (Doob & Webster 2003). Indeed, for more punitive sentencing to deter criminals, a range of conditions must be met. The potential criminal must be aware of changes in punishment, believe that they are likely to be caught and that the new penalty will apply to them. The potential criminal must also consider the consequences of committing the offence, and calculate that an offence previously worth committing is no longer worth it (von Hirsch et al. 1999). So punitive sentences to reduce crime may hinder that goal, especially if this approach shifts resources from developing successful alternatives (Doob & Webster 2003).
AusSocAtts03
26/7/05
8:55 AM
Page 157
PERCEPTIONS OF CRIME AND JUSTICE
• 157
CONCLUSION AuSSA 2003 findings suggest that Australians are too pessimistic about crime trends and lack confidence in the courts. This pessimism rests on unsupported beliefs that crime rates are rising. While we are gloomy about crime and the courts, Australians still invest real confidence in the police – despite widespread acknowledgment of police corruption. This somewhat contradictory view of law and order, however, makes sense if we keep in mind that commercial television remains Australia’s main source of news and information about crime. Commercial television exploits crime as either sensational news or a source for fictional drama. Ultimately, the media blurs the boundaries between fact and fiction so that Australians are more familiar with scenes from police stations in the Bronx and Los Angeles than from their own cities. Crime is continually ‘news’, so it is not surprising that perceptions of crime generally exaggerate its incidence. This phenomenon is not confined to Australia. Research on this problem points to the impact of media on community perceptions (see Roberts et al. 2003 for an overview). To further understand public opinion and perceptions on crime, we need to better understand how media treats crime. The challenge is to provide quality information, which is accessible to a broad public, so that common understandings of criminal justice are less captured by media imperatives. NOTES
1
2 3
4
Although most of these studies were conducted in North America and Europe, available Australian evidence finds similar trends. Weatherburn, Matka and Lind (1996) note, for instance, the increasing place of ‘real concern’ in public opinion polls from 1986, 1992 and 1994. An AGB McNair Poll conducted in 1994 found that of a list of concerns ‘violence and crime’ was the most commonly mentioned concern. ‘Violence and crime’ scored ahead of hospitals and health, unemployment, education and the environment. See Beckett (1997) for a detailed treatment of this subject. American views on crime and justice are documented in a number of distinct sources. The US Bureau of Justice Statistics contains readily accessible data on public opinion and crime. Probably the best source for ready access to a range of public opinion data is be found in the sourcebook, Criminal Justice Statistics. Available online at . A number of British sources (for example, see Hough & Roberts 1998) draw on items now regularly surveyed in the British Crime Survey. These can be found at the United Kingdom’s Home Office Research and Development website. Available online at: . For European attitude studies, see Kury, Obergfell-Fuchs and Smartt (2002); and for international comparisons, see Mayhew and van Kesteren (2002).
AusSocAtts03
26/7/05
158 •
5
6
7
8 9 10 11 12
8:55 AM
Page 158
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
See Indermaur (1987), who replicated earlier work undertaken in Canada by Tony Doob and Julian Roberts. Indermaur asked members of the public, who had responded to a general question about sentencing, about what type of criminal they were thinking of when answering that question. Seven out of ten respondents said they thought of a murderer, rapist or some other type of violent criminal. The other commonly used source for assessing crime trends are population surveys that include self-reports of victimisation. It was not possible to use this data to establish crime trends for 2001–03 because the last national survey was conducted in 2002 (ABS 2003). However, information available in the 2002 survey indicates that crimes against the person and crimes against property were stable between 1998 and 2002 (ABS 2003). Out of approximately 1 662 000 offences with victims reported by the ABS in Recorded Crime Australia for 2003 (ABS 2004), 87 per cent were property crimes and this figure excludes robbery because this is usually classified as violent crime rather than property crime. Over the last 20 years, the homicide rate in Australia has been stable. David Denemark (see chapter 13) details the important place of commercial television as a main source of news and information; less than 2 per cent of respondents rely on a source outside the media for news and information. This research covers both the effect of types of media (radio versus newspapers, for instance) as well as forms of media (news versus entertainment) on perceptions. For a good overview of the literature, see Reiner (2002). The number for AuSSA 2003 respondents from the two territories is small, so these results should be treated with caution. The Cronbach’s alpha for the scale (a measure of reliability) is .70, which is satisfactory for our purposes.
REFERENCES
Australian Bureau of Statistics (ABS) (2003) Crime and Safety Australia 2002. cat. no. 4509.0, Australian Bureau of Statistics, Canberra. —— (2004) Recorded Crime, Australia 2003. cat. no. 4510.0, Australian Bureau of Statistics, Canberra. Australian Institute of Health and Welfare (2002) 2001 National Drug Strategy Household Survey: Detailed Findings, Australian Institute of Health and Welfare, Canberra. Bean, C, Gow, D & McAllister, I (1999) Australian Election Study 1998 [computer file], Australian Social Science Data Archive, Australian National University, Canberra. —— (2002) Australian Election Study, 2001 [computer file], Australian Social Science Data Archive, Australian National University, Canberra. Beckett, K (1997) Making Crime Pay: Law and Order in Contemporary American Politics, Oxford University Press, New York. Broadhurst, R & Indermaur, D (1982) ‘Crime seriousness ratings: The relationship between information accuracy and general attitudes in Western Australia’, Australian and New Zealand Journal of Criminology, 15: 219–34. Butcher, K & Piehl, A (1998) ‘Cross-city evidence on the relationship between immigration and crime’, Journal of Policy Analysis and Management, 17: 457–94. Chiricos, T, Eschholz, S & Gertz, M (1997) ‘Crime news and fear of crime: Toward an identification of audience effects’, Social Problems, 44: 342–55. Doob, AN & Webster, CM (2003) ‘Sentence severity and crime: Accepting the null hypothesis’, in M Tonry (ed.) Crime and Justice: A Review of Research, 30, University of Chicago Press, Chicago, pp. 143–96.
AusSocAtts03
26/7/05
8:55 AM
Page 159
PERCEPTIONS OF CRIME AND JUSTICE
• 159
Garland, D (2001) The Culture of Control: Crime and Social Order in Contemporary Society, Oxford University Press, Oxford. Gibson, R, Wilson, S, Meagher, G, Denemark, D & Western, M (2004) Australian Survey of Social Attitudes 2003, Data file, Australian Social Science Data Archive, The Australian National University, Canberra. Hough, M & Roberts, JV (1998) Attitudes Toward Punishment: Findings from the British Crime Survey (Home Office research study no. 179), Home Office, Research and Statistics Directorate, London. Indermaur, D (1987) ‘Public perception of sentencing in Perth, Western Australia’, Australian and New Zealand Journal of Criminology, 20:163–83. —— (1990) Perceptions of Crime Seriousness and Sentencing: A Comparison of Court Practice and the Perceptions of a Sample of the Public and Judges, Criminology Research Council of Australia, Canberra. Jones, R, McAllister, I, Denemark, D & Gow, D (1993) Australian Election Study, 1993 [computer file], Australian Social Science Data Archive, Australian National University, Canberra. Jones, R, McAllister, I & Gow, D (1996) Australian Election Study, 1996 [computer file], Australian Social Science Data Archive, Australian National University, Canberra. Kelley, J, Cushing, RG & Headey, B (1984) Australian National Social Science Survey, 1984 [computer file], Australian Social Science Data Archive, Australian National University, Canberra. Kelley, J, Bean, C & Evans, M (1986–87) National Social Science Survey 1986–1987: Role of Government [computer file], Australian Social Science Data Archive, Australian National University, Canberra. —— (1989–90) National Social Science Survey: Family and Lifestyles, 1989–1990 [computer file], Australian Social Science Data Archive, Australian National University, Canberra. Kelley, J Evans, M & Bean, C (1987–88) National Social Science Survey 1987–1988: inequality [computer file], Australian Social Science Data Archive, Australian National University, Canberra. Kennedy, GA (2004) Royal Commission into Whether there has been Corrupt or Criminal Conduct by any Western Australian Police Officer: Final Report. vol. 1. Kury, H, Obergfell-Fuchs, J & Smartt, U (2002) ‘The evolution of public attitudes to punishment in Western and Eastern Europe’, in J Roberts & M Hough (eds) Changing Attitudes to Punishment: Public Opinion, Crime and Justice, Willan, Cullumpton, Devon, pp. 93–114. Lenton, S (2004) ‘Pot, politics and the press: Reflections on cannabis law reform in Western Australia’, Drug and Alcohol Review, 23: 223–33. Mayhew, P & van Kesteren, J (2002) ‘Cross national attitudes to punishment’, in J Roberts & M Hough (eds) Changing Attitudes to Punishment: Public Opinion, Crime and Justice, Willan, Cullumpton, Devon, pp. 63–92. McAllister, I, Jones, R, Papadakis, E & Gow, D (1990) Australian Election Study, 1990 [computer file], Australian Social Science Data Archive, Australian National University, Canberra. Mears, D (2001) ‘The immigration–crime nexus: Toward an analytic framework for assessing and guiding theory, research, and policy’, Sociological Perspectives, 44: 1–20. Mukherjee, S (1999) Ethnicity and Crime: An Australian Research Study, Australian Institute of Criminology, Canberra. Pinkerton James, M (1992) ‘The elderly as victims of crime, abuse and neglect’, Trends & Issues in Crime and Criminal Justice, no. 37, Australian Institute of Criminology, Canberra. Reiner, R (2002) ‘Media made criminality: The representation of crime in the
AusSocAtts03
26/7/05
160 •
8:55 AM
Page 160
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
mass media’, in M Maguire, R Morgan & R Reiner (eds), The Oxford Handbook of Criminology, 3rd edn, Oxford University Press, Oxford, pp. 376–416. Roberts, JV, Stalans, LJ, Indermaur, D & Hough, M (2003) Penal Populism and Public Opinion: Lessons from Five Countries, Oxford University Press, New York. Roberts, J & Stalans, L (1997) Public Opinion, Crime and Criminal Justice, Westview Press, Boulder, Colorado. von Hirsch, A, Bottoms, AE, Burney, E & Wikstrom, P-O (1999) Criminal Deterrence and Sentence Severity, Hart Publishing, Oxford. Walker, J, Collins, M & Wilson, P (1988) ‘How the public sees sentencing: An Australian Survey’, in N Walker & M Hough (eds) Public Attitudes to Sentencing: Surveys from Five Countries, Gower, Aldershot, UK, pp. 149–59. Weatherburn, D & Indermaur, D (2004) ‘Public perceptions of crime trends in New South Wales and Western Australia’, Crime and Justice Bulletin: Contemporary Issues in Crime and Justice, no. 80, NSW Bureau of Crime Statistics and Research, Sydney. Weatherburn, D & Jones, C (2001) ‘Does prohibition deter cannabis use?’, Crime and Justice Bulletin: Contemporary Issues in Crime and Justice, no. 58, NSW Bureau of Crime Statistics and Research, Sydney. Weatherburn, D, Jones, C & Donnelly, N (2003) ‘Prohibition and cannabis use in Australia: Survey of 18–29 year olds’, Australian and New Zealand Journal of Criminology, 36:77–93. Weatherburn, D, Matka, E & Lind, B (1996) ‘Crime perception and reality: Public perceptions of the risk of criminal victimisation in Australia’, Crime and Justice Bulletin: Contemporary Issues in Crime and Justice, no. 28, NSW Bureau of Crime Statistics and Research, Sydney. Wilson, PR & Brown, JW (1973) Crime and the community, University of Queensland Press, Brisbane.
AusSocAtts03
26/7/05
8:55 AM
Page 161
CHAPTER 10
H AV E A U S T R A L I A N S E M B R A C E D ECONOMIC REFORM? M i c h a e l P u s e y a n d N i c k Tu r n b u l l
INTRODUCTION Anyone who turned 40 with the new millennium will have spent their adulthood living through a 25-year-long transformation of Australian society that we call ‘economic reform’ – or ‘economic rationalism’, ‘neo-liberalism’, ‘freeing up the markets’, ‘laissez faire’ or ‘structural adjustment’. All these terms refer to the same thing: the remodelling of Australia’s economy along free market lines. Many experts contend the process of structural adjustment has gone further in Australia and New Zealand than in many comparable countries (Castles, Gerritsen & Vowles 1996; Gwartney 2004, p. 17).1 As we shall see below, Australia’s economic reforms have prompted a variety of responses among individuals whose daily lives are affected by these policies. This chapter explores these attitudes and their social and political patterns. A key element driving reform was the change in orientation of Canberra policy bureaucrats away from Keynesian ideas towards neoclassical economics – the view that the operation of the economy is best left to markets (Pusey 1991).2 This meant reducing the interventionist role of Australia’s traditional governmental institutions and establishing free market mechanisms in their place. Reform was also driven by business economists, the ‘think tanks’ and credit rating agencies, international and Australian peak business organisations (most notably the Business Council of Australia), international economic organisations (the OECD, IMF, the World Bank and the WTO), the financial press, and the big four accounting and investment houses (see
AusSocAtts03
26/7/05
162 •
8:55 AM
Page 162
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Pusey 2003). Just as important was the rise of new mindsets, especially the managerialism applied to many public institutions in pursuit of efficiency (Castles 1994). While the nature, extent and causes of economic reform are disputed, some things are agreed on all sides. Despite many differences between the major political parties, economic reform was a product of bipartisan consensus pursued by both Labor and Coalition governments (Brennan & Pincus 2002). While a Coalition government established the Campbell Committee that recommended financial deregulation in 1979, the subsequent Labor government implemented its recommendations, floating the exchange rate when it came to office. Similarly, while the Hilmer Inquiry that prescribed competition reforms was initiated by a Labor government in 1992, subsequent Liberal federal and state governments implemented many of its recommendations (Brennan & Pincus 2002, p. 79). Labor was already committed to economic reform prior to its election in 1983, and Labor governments of the 1980s and early 1990s pursued an economic rationalist program, particularly through extensive tariff cuts, although this was moderated by interventionist policy measures in some areas (Easton & Gerritsen 1996). Similarly, since its election in 1996, the Coalition government has continued the reform program, notably the Workplace Relations Act of 1996 that introduced individual contracts, and the partial privatisation of Telstra. In short, economic reform in Australia has seemed all the more pervasive because it has been introduced and extended by both Coalition and Labor governments across the last 25 years.
MAIN ELEMENTS OF ECONOMIC REFORM What are the particular features of economic reform? First of all, public sector organisations providing major services were reformed internally, and either deregulated and made to operate along commercial lines or sold off, to varying degrees, to the private sector (see Pusey 2003, Appendix A, for a timetable of reform). Privatisation in Australia has probably cut further into the role of government as a provider than in any other country after Britain and New Zealand. In 1989, Telstra (previously Telecom) was corporatised, made to run on a ‘for profit’ basis, and thus made ready for sale and transfer from the public to the private sector. In November 1997, a third of Telstra was sold into private ownership. The mobile phone spectrum was also sold
AusSocAtts03
26/7/05
8:55 AM
Page 163
HAVE AUSTRALIANS EMBRACED ECONOMIC REFORM?
• 163
to create a vast pool of potential capital to fuel the telecommunications industry. By the end of 1999, the second stage of the Telstra privatisation meant that 49 per cent of Telstra was owned by private shareholders. The Commonwealth Bank was privatised in three stages from 1991, and the Qantas privatisation was completed in 1995. In 1993, Fred Hilmer delivered his report on national competition policy, which led to the deregulation and/or privatisation of electricity and transport services by many state governments. Another plank of the economic reform program has been to cut tariffs, lower import restrictions and generally liberalise trade. Until the early 1960s, about 60 per cent of Australian secondary industry and manufacturing was protected by tariffs. Both Labor and Coalition governments cut tariffs to expose Australian industry to greater competition from foreign producers. This was done first by 25 per cent across the board in 1973, then through another major reduction in tariffs on most goods from 1988 and again in 1991, with large reductions for passenger motor vehicles and textiles, clothing and footwear over the 1990s (Productivity Commission 2000, p. 17).3 From the early 1970s to 2001, tariffs supporting the Australian manufacturing sector were reduced from 35 per cent to 5 per cent (Productivity Commission 2000, p. 16). On one side of the coin, trade liberalisation offers consumers more product choice and, for some, new employment opportunities. On the other, it can send jobs offshore, and sometimes has devastating effects on the regions and towns that lose mainstay local industries (see Carroll 1992). By December 2004, free trade agreements were approved with Thailand and the United States to further open up Australia’s economy to foreign competition. All told, then, economic reform has meant significant change to the day-to-day lives of Australians. Australia’s protectionist economic policy was always closely linked to the arbitration system that set wages and the conditions of work. Compulsory arbitration had for a century established an egalitarian value in economic policy via the redistribution of income through the wages system (Castles 1985, 2002). Distributive justice, the ‘fair go’, was governed by the rule that the income of ‘no significant group shall fall if that of the others is rising’ (Corden, in Castles 1987, pp. 112–22). Industrial relations reform has partly dismantled the traditional system, giving the market a greater role in setting the price
AusSocAtts03
26/7/05
164 •
8:55 AM
Page 164
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
of labour, letting businesses have more say over wages and conditions by instituting enterprise bargaining in the 1990s and, under the Howard government, introducing individual employment contracts. However, reforms have not yet abandoned the award system or dismantled the role of the Australian Industrial Relations Commission (and its state equivalents) altogether. This new environment allows for more incentives, flexibility and wage dispersion, giving business greater freedom to improve productivity and respond more quickly to the demands of the global free market (Castles 2002). That said, these same labour market reforms coincided with a shift in national income, with the national accounts showing that, over 20 years from 1980 to the turn of the century, the total wages share of national income had fallen from 60 per cent to 54 per cent as the profit share increased from 17 per cent to 24 per cent; the government share stayed at about the same low level (ABS 2002). Such changes made possible greater earnings inequality and undermined the egalitarian ethic of the old system. At a more fundamental level, all these changes reflect the view that individual activity operating without restriction in the free market is most likely to realise a prosperous economy and an improved standard of living. But if these significant reforms have been led by government and big business, what do Australian citizens think of them? Have Australians truly embraced this transformation? With the Australian Survey of Social Attitudes 2003 data (Gibson et al. 2004), and in the light of 20 and some years of Australia’s experience with economic reform, this becomes an empirically testable proposition. This chapter, then, explores whether or not economic reform has built its own ‘neoliberal’ constituency in Australia.
G O V E R N M E N T O R P R I VAT E O W N E R S H I P ? Privatising government-owned service providers remains an important element of economic reform. After the 2004 elections, the Howard government is now able to fully privatise Telstra with Senate approval. The first Telstra float attracted huge media coverage and was promoted as a ‘democratisation’ of Australian shareholding that would give ordinary Australians an opportunity to become stakeholders in the new Telstra. But table 10.1 shows that public support for private ownership of public telecommunications as well as postal services has fallen over time and there appears to be little support for further privatisation of Telstra or privatisation of Australia Post. From this we might draw two
AusSocAtts03
26/7/05
8:55 AM
Page 165
HAVE AUSTRALIANS EMBRACED ECONOMIC REFORM?
• 165
conclusions. First, the table shows that only a minority of respondents ever supported full privatisation of these organisations. The government did not have widespread public support before the reforms to make Telecom and Australia Post more competitive and efficient, and even that support has now eroded to low levels. Secondly, the downward trend in support for privatisation evident in table 10.1 may indicate that some of these respondents have embraced economic reform. Those Australians supporting privatisation in 1987–88 may believe that the organisational changes and competition reforms had a good effect on the performance of Telstra and Australia Post, particularly with the partial privatisation of Telstra under the Coalition government, and consequently that full privatisation is unnecessary. On the other hand, these findings may indicate that some Australians who previously supported privatisation have changed their minds after experiencing it. We can further unpack judgments about attitudes to economic reform by examining the AuSSA 2003 data on public attitudes to privatisation of a range of major services, including Telstra, Australia Post, electricity, public transport, and prisons. Respondents are asked: Do you think the following enterprises or organisations should be in public ownership, private ownership, or a mix of public and private ownership? We present the results in table 10.2. If respondents were divided in their opinions about the ownership of services before the changes, they are much less so afterwards. Support for further privatising these services is very low, under 10 per cent in each case. In fact, for every service, a
Table 10.1 Support for fully private ownership of telecommunications and postal services, 1987–2003, per cent
Telstra in private ownership n Australia Post in private ownership n
1987*
1993*
1994*
39
33
30
(1638)
(2532)
(1493)
32
24
23
(1638)
(2532)
(1495)
2003
9 (2115) 5 (2116)
NOTE ‘Can’t choose’ responses excluded. NSSS asks respondents about privatising Telecom and postal
services; AuSSA 2003 asks whether Telstra and Australia Post should be in private ownership, public ownership, or a mix of both. SOURCES * National Social Science Survey (NSSS), 1987, 1993, 1994; The Australian Survey of Social Attitudes 2003.
AusSocAtts03
26/7/05
166 •
8:55 AM
Page 166
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 10.2 Preferred ownership of major services, 2003, per cent Preferred ownership
Telstra
Electricity
Public transport
Australia Post
Prisons
Public
57
60
63
67
67
Mix of public and private
31
31
28
24
19
Private
9
6
6
5
8
Can’t choose
4
4
3
3
6
n
(2115)
(2122)
(2118)
(2116)
(2114)
SOURCE The Australian Survey of Social Attitudes 2003
substantial majority of respondents support exclusive public ownership. On the other hand, the data in table 10.2 do show that there are an appreciable number of respondents who support a mix of public and private ownership, from 19 per cent for prisons to 31 per cent for Telstra and electricity services. This tells us that there is some support for the current arrangements, which may mean respondents are happy with the status quo – supporting limited private involvement, but not outright privatisation. But we also see judgments against market arrangements in responses to a question that asks whether AuSSA 2003 respondents agree or disagree with the following statement: For companies like AMP, the NRMA, and National Mutual, changing owners from members to shareholders has been a mistake. These demutualisations, most notably of the AMP Society in 1998, put bundles of shares into the hands of Australian households, giving them an unexpected windfall. Nonetheless, 40 per cent think that the demutualisations were a mistake, with a further 27 per cent not sure and only 16 per cent disagreeing. Can we detect an overall pattern across responses to these questions – is there a general pro- or anti-privatisation position? Figure 10.1 groups together responses from table 10.2 into three types of respondents: those who prefer the ‘old’ system of full public ownership for all services; those who prefer a mix of the old and new systems, supporting at least one service in full public ownership and at least one other in either a mix of public and private ownership or fully privately owned; and those who prefer fully privately owned services.4 Figure 10.1 shows that the largest group of respondents (47 per cent) do not make clear ideological judgments about the ownership of service organisations, appearing instead to make pragmatic assessments
AusSocAtts03
26/7/05
8:55 AM
Page 167
HAVE AUSTRALIANS EMBRACED ECONOMIC REFORM?
• 167
Figure 10.1 Preferences for the ownership of major services,* 2003, per cent
50 45 40 35
%
30 25 20 15 10 5 0 Only public
Mix of public and private
Only private
Ownership preferences * Services listed in table 10.2 SOURCE The Australian Survey of Social Attitudes 2003
about what they think is best for individual services. They support either full public ownership or a mixture of public and private ownership. Only a small proportion (13 per cent) of respondents support exclusively privately owned services. Interestingly, respondents who embrace some aspects of privatisation continue to support full public ownership for at least one service. Nonetheless, there is a large ‘anti-privatisation’ group (39 per cent) that would prefer these service organisations to be in full public ownership, supporting neither the status quo of a mixture of public and private ownership nor full private ownership of these services. This likely reflects deep-seated preferences for public ownership. However, it could also indicate a reaction against economic reform and towards a stronger role for government. In a similar vein, Wilson, Meagher and Breusch, in chapter 7, find greater support for government spending on social services than at any time in the past 30 years. In table 10.3, we break down these categories by respondent party identification to see where the pro- and anti-privatisation constituencies lie. The first point to note is that the constituency in support of further
AusSocAtts03
26/7/05
168 •
8:55 AM
Page 168
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 10.3 Ownership preferences of major services by party identification, 2003, per cent Labor
One Nation
Green
Only public
Liberal
29
32
32
38
48
53
54
Mix of public and private
54
53
54
48
42
45
39
Only private n
National Democrat No party
17
16
14
14
11
3
7
(663)
(57)
(50)
(401)
(721)
(40)
(100)
SOURCE The Australian Survey of Social Attitudes 2003. Excludes ‘other party’ due to small number of responses.
privatisation is small in each case, even among Liberal–National (Coalition) identifiers. At the same time, it also indicates widespread support for a mixture of ownership across the political spectrum. Of the established parties, we see an ideological division between the right, with Liberal and National identifiers most supportive of private ownership, and the left, with Labor identifiers supportive of public ownership. Among smaller party voters, matters are less ideologically divided. The conservative One Nation and the progressive Greens make up a heterogeneous anti-privatisation constituency alongside Labor identifiers. Where there is political division, it perhaps lies between pragmatic views and ideological views about ownership; Coalition and Democrat identifiers are perhaps the most pragmatic, with more than 50 per cent of each group supporting a mixed system.
FREE TRADE OR PROTECTIONISM? Have Australian attitudes shifted with free trade policy, or do protectionist attitudes still prevail? To see where Australians draw the line, AuSSA 2003 asks respondents if they agree that Australia should use tariffs to protect its industry. Table 10.5 shows that a majority of respondents (52 per cent) still believe that we should protect industry with tariffs – a pattern that has no discernible fluctuation since the mid-1990s (see table 10.4). We also have answers to a similar question on protection, asking if respondents agree that the import of foreign products should be restricted to protect our economy. Two-thirds of respondents agree with this, while a low 15 per cent disagree.5 One should not construe this as a ‘knee jerk’ response to all foreign economic incursions, given that, in response to a further question, only about 40 per cent of respondents opposed the right of foreigners to buy land in Australia.6 Can we conclude that Australians are protectionists; that they are
AusSocAtts03
26/7/05
8:55 AM
Page 169
HAVE AUSTRALIANS EMBRACED ECONOMIC REFORM?
• 169
against free trade altogether? No, because when they are asked if free trade has led to better products being available in Australia, almost 50 per cent agree. Only 18 per cent of the respondents disagree that free trade delivers better products to Australian consumers. Although many respondents support protectionist measures, they also enjoy the benefits of the global market. Their position is neither fully protectionist nor do they endorse more radical trade reform. They seem to take a pragmatic view that supports engagement in the global market economy concurrently with a continuing regulatory role for government (see also chapter 14). AuSSA 2005 will allow us to re-examine this question in light of the free trade agreement with the United States. We now group respondents by their answers to the two questions on protection and the question on free trade shown in table 10.5. This allows us to compare three groups of respondents classified by their
Table 10.4 Support for protection, 1995/96–2003, per cent Australia should continue to use tariffs to protect its industry
1996*
1998*
2003
59
57
58
(n = 1716)
(n = 1825)
(n = 1900)
Australia should limit the import of foreign products to protect its national economy
1995#
1996*
2003
78
71
66
(n = 2294)
(n = 1765)
(n = 2098)
# National Social Science Survey 1995; * Australian Election Study 1996, 1998; The Australian Survey of Social Attitudes 2003 (‘Don’t know’ and ‘Can’t choose’ responses excluded)
SOURCES
Table 10.5 Attitudes to free trade, 2003, per cent Australia should use tariffs to protect industry
Australia should limit import of foreign products to protect the economy
Free trade leads to better products available in Australia
Agree
52
65
49
Neither agree nor disagree
25
19
28
Disagree
12
15
18
Can’t choose
10
2
5
(2108)
(2142)
(2148)
n
SOURCE The Australian Survey of Social Attitudes 2003
26/7/05
170 •
8:55 AM
Page 170
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
views on international trade (see figure 10.2). The result indicates that few respondents (12 per cent) are pure ‘economic nationalists’ – those in favour of any measure of protection and who disagree that free trade brings better products to Australia. Even fewer are purely in favour of free trade – those who do not support any measure of protection and agree with the benefits of free trade (just 4 per cent). The large majority fall into the middle category of respondents who are neither emphatically pro- nor anti-free trade (83 per cent).7 When we break down the middle category, we find that 39 per cent of all respondents are pragmatic, agreeing that free trade brings better products to Australia and support at least one of the protectionist statements reported in table 10.5.8 Respondents have interests as consumers as well as workers, so they may see the benefits of liberal trade as well as the costs it represents for some employees. This is consistent with ‘Australian settlement’ values, which support a free market moderated by state regulation. But these findings also suggest that many respondents are still uncertain about markets and trade.
Figure 10.2 Free trade versus protectionism, 2003, per cent 90 80 70 60
%
AusSocAtts03
50 40 30 20 10 0 Economic nationalists
Mixed
Respondent group SOURCE The Australian Survey of Social Attitudes 2003
Support free market
AusSocAtts03
26/7/05
8:55 AM
Page 171
HAVE AUSTRALIANS EMBRACED ECONOMIC REFORM?
• 171
LABOUR MARKET INSTITUTIONS Along with deregulating trade and introducing competition to the public sector, deregulating the labour market has been a key component of economic reform. Have Australians embraced these reforms? Table 10.6 shows the results of the AuSSA 2003 question asking respondents whether they agree or disagree with the following two statements: Award wages are the best way of paying workers and setting conditions Employers and employees should negotiate pay directly What is striking here are the conflicting views on wage setting. Support for the award system remains very strong, with two-thirds of the respondents agreeing that award wages are the best way of paying workers and setting conditions. For the most part, they do not think that wages and working conditions should be set without the moderating hand of industrial relations commissions. At the same time, we find that 64 per cent of respondents agree that employees and employers should negotiate their pay directly. This is the case despite the fact that 46 per cent of respondents agree that one form of direct negotiation – individual contracts – favour the employer over the employee. So, even though support for the award system remains strong, people value some flexibility at work. These results may seem contrary. But they might reflect support for current arrangements that set pay conditions through both awards and bargaining at an enterprise level (ABS 2005). Again, the respondents are not strongly divided along ideological lines. Responses could reflect a lack of detailed knowledge about industrial relations, support for both new and old arrangements, or more nuanced opinions. Perhaps the responses simply reflect the broad values Table 10.6 Attitudes to regulating pay and conditions of work, 2003, per cent Award wages best way to pay workers/ set conditions
Employers and employees should negotiate pay directly
Individual contracts favour employer over employee
Agree
66
64
46
Neither agree nor disagree
17
16
29
Disagree
14
18
18
4
3
6
(4157)
(2578)
(2562)
Can’t choose n
SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
172 •
8:55 AM
Page 172
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
underlying these statements: people in general support both flexibility and collective guarantees of rights without perceiving a contradiction between the two. How and when Australians support different aspects of the industrial relations system presents a question for further research.
AT T I T U D E S T O I N C O M E E Q U A L I T Y, L I V I N G S TA N D A R D S , A N D D I S T R I B U T I O N O F W E A LT H In assessing current attitudes to inequality and living standards, we need to appreciate that the economic expectations of the nation were formed in the long post-war boom that lasted from 1945 until the mid-1970s. The broad Australian middle class that emerged at this time was formed in a period of steadily increasing prosperity – a time, as one noted historian puts it, of ‘getting and spending’ (Bolton 1990). Even with the very rudimentary income statistics of that time, it is clear that the average weekly incomes of a predominantly male workforce increased by about 4 per cent per annum, in real terms, and hence at something like five times the rate for the first 40 years of the last century. The cumulative rise in living standards during this period was very substantial. Moreover, tariffs and other arms of the ‘conservative’ social policy of those years were, as we have noted, used to maintain the principle that the income of one social group should not fall while the income of another rose. During the first 20 post-war years to 1965, Australia’s GDP increased by nearly 5 per cent per annum, faster than in the United States or the United Kingdom, and with far lower levels of unemployment and poverty.9 Incomes were, probably, more equally shared than in almost any other advanced democracy (Sawyer 1976). Real average weekly earnings for employed males grew by 52 per cent from 1945 to 1965 (Pusey 2003, p. 22). But with the emergence of stagflation this sustained growth stalled in the 1970s and set the scene for the economic reform that followed in the 1980s and 1990s. The picture that emerges from the first phase of extensive economic reform presents a different story. In the 1980s, Australia had, as Bob Gregory’s figures demonstrate, a very small increase in average male earnings of some 6 per cent for the whole decade. In Gregory’s words, ‘the slow down in real wage growth has been extraordinary. Since 1976 the total per capita increase over two decades has been equivalent to an average three-year increase during earlier decades’ (Gregory 1998, p. 7). However, from the mid-1990s, average real wages have begun to grow again, at a similar rate to the 1950s and 1960s (Gregory 2000).
AusSocAtts03
26/7/05
8:55 AM
Page 173
HAVE AUSTRALIANS EMBRACED ECONOMIC REFORM?
• 173
The long post-war run of improving standards of living is reflected in how our AuSSA 2003 respondents answer the question: Compared to your parents when they were the age you are now, do you think your standard of living is better, worse or the same as theirs was? These results are reported in table 10.7. Most respondents say that their standard of living is better than that of their parents when they were the same age. Fully 76 per cent say their living standards are better (with 44 per cent saying ‘much better’). Only 8 per cent believe their standard of living to be worse. However, as we can see from responses to the next question, the impact of declining expectations begins to show when respondents are asked to look forward into the economic future: When your children are the age you are now, do you think their standard of living will be better, worse or the same as yours is now? They are now less sanguine: 48 per cent of them believe that their children will be better off at the same age, 25 per cent believe that their children’s standard of living will be the same and 21 per cent see an unequal future. We did not have trend data available for this question, so we cannot know if economic reform has dampened expectations for future generations. However, these results confirm that we are less optimistic about the living standards of future generations than we are about our own living standards compared to our parents. After more than 20 years of reform, the English-speaking democracies have some of the largest income inequalities in the OECD (Whiteford Table 10.7 Perceptions of generational change in living standards, 2003, per cent Your standard of living compared to your parents at the same age
Your children’s standard of living compared to you at the same age*
Better
76
48
About the same
15
25
Worse
8
21
Can’t choose
1
7
n
(4229)
* Respondents without children excluded. SOURCE The Australian Survey of Social Attitudes 2003
(3856)
AusSocAtts03
26/7/05
174 •
8:55 AM
Page 174
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
2001, p. 58). The United States is the most unequal of these, with a larger proportion of its population living on less than 50 per cent of median income than any other comparable country (Mishel, Bernstein & Schmitt 1999, p. 371). Our targeted social security provisions have, in part, offset slow growth in incomes for the poorest 20 per cent of Australians (see chapter 7). However, the broader picture shows that economic reform may have been one major factor in widening the income gap that professional economists have estimated for Australia over the 1980s and 1990s (see Harding & Greenwell 2001, p. 22). Have Australians picked up on the widening gap? AuSSA 2003 asks respondents: Thinking of income levels generally in Australia today, would you say that the gap between those with high incomes and those with low incomes is … much too large, too large, about right, too small, much too small, can’t choose? Fully 84 per cent of the respondents agree that the income gap between high and low-income earners is too large and 42 per cent of them say that it is much too large. Only 10 per cent feel that the income gap is about right. Table 10.8 shows that this is a marked increase on the longterm trend. However, while we might attribute some of this recent increase to perceptions of incomes in a post-reform Australia, it may equally be due to changes in question wording between the surveys (see notes to table 10.8). Nonetheless, a clear majority agree in every case. Evidently Australians today are opposed to broad inequality in income, and they have been for a long time. Table 10.9 shows some fluctuation in responses over time in favour of the redistribution of income and wealth. Any growing perception among respondents of income inequality or unjust wealth distribution has not produced a counter-shift in favour of the government redistribTable 10.8 Agree that income differences in Australia are too large, 1984–2003, per cent 1984*
1987*
1994*
2003
62
61
66
84
(n = 2988)
(n = 1569)
(n = 1453)
(n = 4165)
NOTE NSSS asks respondents whether they agree that ‘there is too much of a difference between rich and
poor in this country’ and AuSSA 2003 asks respondents if the gap between those on high and low incomes is too large or too small. SOURCES * National Social Science Survey 1984, 1987, 1994; The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
8:55 AM
Page 175
HAVE AUSTRALIANS EMBRACED ECONOMIC REFORM?
• 175
Table 10.9 Agree that income should be redistributed, 1984–2003, per cent 1984#
1987*
1987#
1993*
1994#
1996*
2001*
2003
51
46
39
51
41
47
56
46
(n = 2979)
(n = 1778)
(n = 1631)
(n = 2948)
(n = 1449)
(n = 1725)
(n = 1949)
(n = 3986)
NOTE NSSS and AES ask respondents if income and wealth should be redistributed to ordinary working
people; AuSSA 2003 asks respondents if government should redistribute income from the well-off to the less well-off. SOURCES # National Social Science Survey 1984, 1987, 1994; * Australian Election Study 1987, 1993, 1996, 2001; The Australian Survey of Social Attitudes 2003 (‘Can’t choose’ excluded)
uting income and wealth directly from the better off to the less well-off. We should note that some fluctuations might result from changes in question wording across the surveys (see notes to table 10.9). However, the mean of all eight surveys is 47 per cent, which places the AuSSA 2003 result near the average. Despite a rising perception of inequality, redistribution of income and wealth is no more popular than it was, even if respondents now favour greater social spending in some areas (see chapter 7). Whatever their concerns with distributional fairness, economic growth remains an important priority for Australians. When AuSSA 2003 asks respondents about which would be their first priority for Australia – maintain a high level of ‘economic growth’, ‘make sure this country has strong defence forces’, ‘see that people have more say about how things are done at their jobs and in their communities’, or ‘try to make our cities and countryside more beautiful’ – 53 per cent identify maintaining economic growth as their first priority, and 25 per cent make it their second choice (see chapter 6). Renewed wages growth in recent years may be responsible for the change in responses to a question asking respondents ‘How proud are you of Australia’s economic achievements?’ Eighty per cent of respondents to AuSSA 2003 say they are proud of Australia’s economic achievements, up from 50 per cent in 1995 (see chapter 11).
HAVE AUSTRALIANS ADOPTED ‘ N E O - L I B E R A L’ VA L U E S ? What values underpin Australian attitudes about the economy? Responses in the tables thus far suggest that values hold to the institutions of the Australian settlement – government ownership of major services and utilities, award wages, and protected trade – while also adjusting to newer arrangements. But do these attitudes point to underlying ideological preferences for collective, state solutions and a rejection
AusSocAtts03
26/7/05
176 •
8:55 AM
Page 176
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
of individualism on the one hand, or have respondents become more individualistic and more accepting of liberal markets? To address this question, we examine two more responses from AuSSA 2003 – attitudes to big business and orientations towards either individual or community interests as the key to improving living standards. First, we examine responses to questions on the role of big business. Reformers argue that Australia benefits from a deregulated business sector and must let economic actors, whether they be corporations, small businesses or individuals, compete on the same level playing field – without distorting interference from unions, governments or other vested interests. One issue here is the core notion that corporations are merely ‘technical’ arrangements to co-ordinate economic activity and that deregulation should not be seen in political terms, but rather as a necessary means of setting big business free to do what it does best; namely to invest, create new business opportunities and generate wealth for shareholders. Do Australians have neutral views, then, about Australia’s main economic actors – its big businesses? AuSSA 2003 asks if people think that big business should have more power, less power or the same amount of power. Table 10.10 shows that a majority, 60 per cent of respondents, say that big business should have less power, and only 5 per cent of respondents believe that it should have more power. These sentiments might be inspired by indignation over corporate malfeasance, since 81 per cent of respondents agree with the statement that ‘when big businesses break the law they often go unpunished’ (and 40 per cent of the respondents strongly agree). Australians believe that big business has a major influence on society, but remain sceptical of it (see also chapter 8). Looking to the long-term trend in table 10.11, during the period of reform in the 1990s we see a rising proportion of people who thought Table 10.10 Attitudes to big business, 2003, per cent Big business should have
More power
Big businesses go unpunished when they break the law
5
Agree
81
Same amount of power
32
Neither agree nor disagree
Less power
60
Disagree
8
Can’t choose
2
Can’t choose n
3 (4110)
SOURCE The Australian Survey of Social Attitudes 2003
9
(4203)
AusSocAtts03
26/7/05
8:55 AM
Page 177
HAVE AUSTRALIANS EMBRACED ECONOMIC REFORM?
• 177
Table 10.11 Agree that big business has too much power, 1984–2003, per cent 1984#
1987*
1990*
1993*
1996*
1998*
2001*
2003
62
51
65
62
65
71
72
62
(n = 2960)
(n = 1765)
(n = 1998)
(n = 2918)
(n = 1734)
(n = 1829)
(n = 1949)
(n = 3978)
NOTE NSSS and AES asks respondents whether they agree big business has too much power; AuSSA 2003
asks respondents whether big business should have more or less power. The 2003 result is all respondents choosing ‘less power’. SOURCES # National Social Science Survey 1984; * Australian Election Study 1987, 1990, 1993, 1996, 1998, 2001; The Australian Survey of Social Attitudes 2003 (‘Can’t choose’ excluded.) Different question wording used in several cases
big business in Australia had too much power. However, the AuSSA 2003 results break with the upward trend. This may be, again, due to the change in question wording, which asks respondents to say not just whether big business has too much power, but whether its power should be reduced or increased (see notes to table 10.11). We will need to watch for further trends in AUSSA 2005. While this figure is lower in 2003, it still suggests that a majority of respondents might favour policy measures to keep the power of big business in check. A second AuSSA 2003 question examines the general orientations that inform respondents’ attitudes about the way we organise economy and society. Have Australians taken on a new ‘ideology’ of more competition and renewed individual effort and self-interest? Responses to the following question in table 10.12 give an initial clue: If you had to choose, which of the following would be closest to your view? Living standards in Australia would be best improved if each individual looked after his or her own interests first, or Living standards in Australia would be best improved if we looked after the community’s interests first. Table 10.12 The best way to improve living standards, 2003, per cent Each individual look after his or her own interests first
30
Look after the community’s interests first
49
Can’t choose
21
(n = 2141) SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
178 •
8:55 AM
Page 178
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
The answers are illuminating and show that many people understand that we are, in Michael Leunig’s words, ‘joined together by more than the weather’ and believe the general good to be primarily a product of community interest over self-interest. Almost half AuSSA 2003 respondents think that putting the community interest first is the best way to improve living standards in Australia, 30 per cent believe that selfinterest is the key, and one-fifth cannot decide between the two. This suggests a prevailing community orientation among respondents. At the same time, the high proportion of ‘Can’t choose’ responses leaves open the question of whether a broad value orientation exists that would predict the legitimacy of policy measures founded upon the individualistic values of reform against communitarian, collective responses. To examine this further, we performed a multivariate analysis to look for key factors that might predict a community-oriented response to this question against an individualist response (excluding ‘Can’t choose’ responses).10 One significant factor that increased the probability of answering in favour of community solutions were being a Green or Democrat party identifier. The characteristics that reduce the likelihood of favouring a community response were being a Liberal Party identifier, being employed in an advanced clerical and service occupation (that is, secretaries and personal assistants), and having less than a Year 12 education or a trade qualification. To draw a speculative inference, this result suggests two broad political constituencies, consistent with an educated, left-communitarian sentiment on the one hand, against a more individualistic orientation among conservative voters and elements of the blueand white-collar working class (the Howard battlers) on the other. These two constituencies probably view broader economic reform from different standpoints and do not clearly sit in either pro- or anti-reform camps.
CONCLUSION By examining contemporary opinions about the economy and institutions, we get some sense of whether attitudes have shifted in a direction that is supportive of economic reform. While we can make only impressionistic interpretations from the data we present here, AuSSA 2003 responses and past survey findings suggest that Australians hold a mixed view of the main elements of economic reform. If the reform of government service providers found some success, support for further sell-offs appears limited. Significant constituencies on the left and the right reject the privatised model. Protectionist policy measures continue to hold strong appeal, even though such interventionist policies have largely
AusSocAtts03
26/7/05
8:55 AM
Page 179
HAVE AUSTRALIANS EMBRACED ECONOMIC REFORM?
• 179
passed into history. On the other hand, the majority of respondents like the benefits of free trade and place a high value on economic growth. They believe that big business has too much power, and that income inequality is too great. However, they are no more willing to redistribute wealth than they were before many of the reforms began. On the whole, the evidence suggests that Australians live with the contradictions of the market economy – they see benefit in free markets and trade, but also are aware of their drawbacks, and so continue to believe in an active role for government. Australians do not generally accept a ‘hydraulic model’ that would have them believe that in order to have strong markets we must have weak states.11 In fact, the responses here are consistent with Braithwaite’s (1988) conclusions from the 1984 survey that Australian public opinion about the economy was neither ‘left’ nor ‘right’; they prefer a mixed economy, while perhaps grudgingly accepting the power of elite economic actors. The public continues to support the institutions of the past, even as they adjust to the new arrangements: AuSSA 2003 findings remind us of a heritage that is still actively shaping Australia’s attitudes to rights, entitlements and the role of the state as the principle agent of distributive justice. NOTES
1
2 3 4 5
6
A composite measure of economic reform can be found in the Annual Reports of the Fraser Institute in Canada. Using data on an individual country’s government, labour and credit markets, property laws, international trade rules, and access to money, the Institute derives an index of ‘economic freedom’. This index (between 0 and 10) is a useful measure for assessing the impact of ‘neo-liberal’ reforms cross-nationally. Between 1975 and 2002, Australia became more economically liberal by 2.1 index points, compared to New Zealand (2.8) and the United Kingdom (2.4). This exceeded the increase for most other advanced democracies, but even socialdemocratic countries like Sweden record substantial increases (2.1). Australia is equal sixth (7.9) on the rankings after Singapore (8.6), United Kingdom, United States, New Zealand and Switzerland (8.2). For an alternative view, see Brennan and Pincus (2002). Despite the long-term reductions, tariffs on passenger motor vehicles and textiles, clothing and footwear were markedly increased from the mid-1970s to the mid-1980s (Productivity Commission 2000, p. 16). We exclude ‘Can’t choose’ and missing cases, increasing the proportion of pro- and anti-privatisation positions. There are small differences in the percentages of AuSSA 2003 respondents shown agreeing with tariff protection and import restrictions in tables 10.4 and 10.5. Table 10.4 excludes ‘Don’t know’ responses when computing percentages to maintain comparability with other data sources. Table 10.5 includes them. These views on economic protectionism are consistent with the findings of the Middle Australia project showing increasing support over the late 1990s for government assistance for declining industries (Pusey 2003, p. 147).
AusSocAtts03
26/7/05
180 •
7 8
9 10
11
8:55 AM
Page 180
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Any ‘Neither agree nor disagree’ or ‘Can’t choose’ responses are included in the middle category. Our intention here is to identify the outliers on the trade question. The numbers in the middle are large in part because many respondents choose a ‘Neither agree nor disagree’ response to at least one of the statements; this may suggest some ambivalence or uncertainty about the benefits and costs of protectionism and the global economy. Bolton (1990) points out that in 1966, 4 per cent of Melbourne families lived below the Henderson poverty line, compared to 14 per cent in the United Kingdom and 20 per cent in the United States. We performed a simple binary logistic regression analysis on this choice of statements about living standards, which included as dependent variables: respondent age, gender, education, occupation, employment sector, party identification and region. For this inspired phrase, we acknowledge John Braithwaite.
REFERENCES
Australian Bureau of Statistics (ABS) (2002) Australian National Accounts, National Income, Expenditure and Product, ‘Other Selected Series’, table 41, Time Series, 2002, 5206.0, Australian Bureau of Statistics, Canberra. —— (2005) Employee Earnings and Hours, Australia, May 2004, 6306.0, Australian Bureau of Statistics, Canberra. Bolton, G (1990) Oxford History of Australia: 1942–1988, Oxford University Press, Melbourne. Braithwaite, J (1988) ‘Economic policy: What the electorate thinks’, in J Kelley & C Bean (eds) Australian Attitudes: Social and political analyses from the National Social Science Survey, Allen & Unwin, Sydney, pp. 26–35. Brennan, G & Pincus, J (2002) ‘Australia’s economic institutions’, in G Brennan & FG Castles (eds) Australia Reshaped: 200 Years of Institutional Transformation, Cambridge University Press, Cambridge, pp. 53–85. Carroll, J (1992) ‘Economic rationalism and its consequences’, in J Carroll & R Manne (eds) Shutdown: The Failure of Economic Rationalism and How to Rescue Australia, Text Publishing Company, Melbourne, pp. 7–26. Castles, F ( 1985) The Working Class and Welfare, Allen & Unwin, Sydney. —— (1987) ‘Australia and Sweden: The politics of economic vulnerability’, Thesis Eleven, 16: 112–22. —— (1994) ‘The wage earners’ welfare state revisited: Refurbishing the established model of Australian social protection, 1983–1993’, Discussion paper no. 39, March, Graduate Program in Public Policy, Australian National University, Canberra. —— (2002) ‘Australia’s institutions and Australia’s welfare’, in G Brennan & FG Castles (eds) Australia Reshaped: 200 Years of Institutional Transformation, Cambridge University Press, Cambridge, pp. 25–52. Castles, FG, Gerritsen, R & Vowles, J (eds) (1996) The Great Experiment: Labour Parties and Public Policy Transformation in Australia and New Zealand, Allen & Unwin, Sydney. Easton, B & Gerritsen, R (1996) ‘Economic reform: Parallels and divergences’, in F G Castles, R Gerritsen & J Vowles (eds) The Great Experiment: Labour Parties and Public Policy Transformation in Australia and New Zealand, Allen & Unwin, Sydney, pp. 22–47. Gibson, R, Wilson, S, Meagher, G, Denemark, D & Western, M (2004) Australian Survey of Social Attitudes 2003, Data file, Australian Social Science Data Archive, The Australian National University, Canberra.
AusSocAtts03
26/7/05
8:55 AM
Page 181
HAVE AUSTRALIANS EMBRACED ECONOMIC REFORM?
• 181
Gregory, R (1998) ‘Competing with Dad: Changes in the intergenerational distribution of male labour market income’, paper presented to the ‘Income Support, Labour Markets and Behaviour: A Research Agenda’ Conference, Centre for Economic Policy Research, Australian National University, Canberra. —— (2000) ‘A longer run perspective on Australian unemployment’, Discussion Paper 425, December, Centre for Economic Policy Research, Australian National University, Canberra. Gwartney, J D (2004) Economic Freedom of the World: 2004 Annual Report, Fraser Institute, Vancouver, British Columbia. Harding, A & Greenwell, H (2001) ‘Trends in income and expenditure inequality in the 1980s and 1990s’, NATSEM paper presented to the 30th Annual Conference of Economists, 24 September. Online available: Mishel, L, Bernstein J & Schmitt, J (1999) State of Working America 1998–99, ILS Press, Ithaca. Productivity Commission (2000) Review of Australia’s Tariff Arrangements, Inquiry Report, [Online] Available: [2000, 19 Dec.]. Pusey, M (1991) Economic Rationalism in Canberra: A Nation Building State Changes its Mind, Cambridge University Press, Cambridge. —— (2003) The Experience of Middle Australia: The Dark Side of Economic Reform, Cambridge University Press, Melbourne. Sawyer, M (1976) Income Distribution in OECD Countries, OECD Occasional Study, Organisation for Economic Cooperation and Development, Paris. Whiteford, P (2001) ‘Understanding poverty and social exclusion: Situating Australia internationally’, in R Fincher & P Saunders (eds) Creating Unequal Futures? Rethinking Poverty, Inequality and Disadvantage, Allen & Unwin, Sydney, pp. 38–69.
AusSocAtts03
26/7/05
8:55 AM
Page 182
CHAPTER 11
I M M I G R AT I O N , M U LT I C U LT U R A L I S M A N D N AT I O N A L I D E N T I T Y M u r r a y G o o t a n d I a n Wa t s o n
INTRODUCTION When the Howard government was elected in 1996, immigration and multiculturalism were important issues. Net migration was at a higher level than at any other time in the 1990s1 and aspects of Labor’s immigration policy, including its policy on family reunions, were generating vocal opposition. There were concerns, too, about some of the assumptions underlying Labor’s policies on multiculturalism. The Liberal Party’s promise to govern ‘For all of us’, as their campaign slogan put it, was interpreted by some as an implicit attempt, among other things, to mobilise these concerns; certainly some of Paul Keating’s advisers wanted him to make multiculturalism ‘less threatening’ (Williams 1997, p. 173). In the Australian Election Study (AES), conducted shortly after the election, more respondents said the Coalition was closer to their position on immigration than said Labor; indeed, on none of the other issues canvassed in the survey was the Coalition’s lead over Labor (25 percentage points) greater than it was on immigration (Goot 2004, p. 62). Swept into office, the government began to ring the changes: planned immigration was cut from 96 000 in 1995–96 to 86 000 in 1996–97; the proportion of the intake coming under the family reunion category, now more narrowly defined, fell from more than two-thirds in 1995–96 to less than a half in 1997–98 (Betts 2003); and while the Department of Immigration and Ethnic Affairs became the Department of Immigration and Multicultural Affairs, the word ‘multicultural’ was
AusSocAtts03
26/7/05
8:55 AM
Page 183
I M M I G R AT I O N , M U L T I C U L T U R A L I S M A N D N AT I O N A L I D E N T I T Y
• 183
not often to be read from the Prime Minister’s lips. That the government, in 2001, could turn back the Tampa and then be re-elected reinforced the view, among both the government’s supporters and its opponents, that public opinion had hardened not only towards asylum seekers – or ‘illegal immigrants’ as the government preferred to call them – but also to immigration in general (Hewett 2001; Devine 2002). But have Australian attitudes to immigration and multiculturalism become more hostile during the Howard years? Does our sense of national identity – the importance we attach to being Australian born or our pride in Australia’s achievements – influence the welcome we give to immigrants? And how does it affect support for multiculturalism? This chapter uses evidence from the Australian Survey of Social Attitudes 2003 (Gibson et al. 2004) to answer these questions and to show that, since the election of the Howard government, opposition to immigration has actually fallen, views about multiculturalism remain mixed, and our beliefs about how migrants affect the economy, jobs and crime have become more positive. Fewer respondents now say that Australia makes them feel ashamed and more take pride in Australian democracy and hold a positive view about Australia’s history. In addition, we show that attitudes to immigration are shaped by education and political affiliation, and by views about welfare, the death penalty and the economy. Importantly, ‘nativism’ – the belief that being ‘truly Australian’ means being born in Australia, having Australian ancestors, and living here most of our lives – explains our attitudes to immigration and multiculturalism much better than does national pride.
I M M I G R AT I O N A N D M U LT I C U LT U R A L I S M One of the most remarkable, if least noticed, recent changes in public opinion has been the turnaround in attitudes to immigration. When Labor left office in 1996, support for immigration was low: in the 1996 AES, almost two-thirds of respondents believed that immigration should be reduced, while one-third said the number of immigrants should be reduced ‘a lot’ (see table 11.1). However, after Howard’s election, the number of Australians saying that migrant numbers should be reduced started to fall. After the 1998 election, less than half the AES respondents wanted immigration reduced. This was the first time since the 1980s that the majority of respondents did not support lower migrant numbers. By the end of 2001, the numbers wanting lower immigration had fallen to little more than a third. AuSSA 2003 respondents confirm this trend: while more than a third still want immigration cut, a substan-
AusSocAtts03
26/7/05
184 •
8:55 AM
Page 184
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 11.1 Attitudes to immigration levels, 1995–2003, per cent 1995
1996
1998
2001
2003
Increase
11
8
13
25
26
Same
27
28
38
37
31
Decrease
58
61
47
36
38
4
1
2
2
5
Don’t know/ Can’t choose n
(2438)
(1795)
(1897)
(2010)
(2064)
SOURCES International Social Survey Program 1995; Australian Election Study 1996, 1998, 2001; The Australian Survey of Social Attitudes 2003; ‘Do you think the number of immigrants allowed into Australia nowadays should be reduced or increased?’ (1996, 1998, 2001, 2003). In 1995, the question asked: ‘Do you think the number of immigrants to Australia nowadays should be increased a lot, increased a little, remain the same as it is, reduced a little, reduced a lot, can’t choose?’
tial number (one-quarter) now want it increased. To find another period when polls indicated majority support for the size of the immigration program, we must go back to the years between 1958 and 1970, years of the White Australia policy (Goot 1999, table 1). Changes in public opinion under Howard are unlikely to have been driven by a decline in migrant numbers. True, the number of permanent arrivals declined in 1996–97, 1997–98 and 1998–99. But from 1999 to 2000 the number increased again, so that in 2000–01 it exceeded the level it was at when Howard won. In any event, few respondents would have known whether the number of arrivals had increased or decreased; asked to estimate the number of arrivals in any one year, the most common response is ‘Don’t know’ (Betts 1999, p. 103). It is possible that the Howard years were an exception and, as Betts suggests, that business criticism of reduced immigration ‘may have persuaded [the public] that the cuts were more dramatic than they actually were’. But her evidence of the ‘barrage of criticism from business groups, including the media’, is based largely on stories in newspapers read mostly by business and professional people. Other factors are more likely to explain changing attitudes to immigration. First, Howard has shifted the immigration mix away from family reunions to skilled migration (Jupp 2002, pp. 49, 150, 208). This reorientation, initially recommended by the FitzGerald Report in 1988, appears to be supported in the polls (Goot 2000, p. 55). Second, unemployment as officially measured has fallen under the Howard government and studies show unemployment levels correlate strongly with attitudes to immigration (McAllister 1993, p. 172). Third, changing attitudes may be related to the ‘lower volume setting’ on the government’s ‘multicultural sound machine’ (Betts 2000, p. 64). The idea that
AusSocAtts03
26/7/05
8:55 AM
Page 185
• 185
I M M I G R AT I O N , M U L T I C U L T U R A L I S M A N D N AT I O N A L I D E N T I T Y
the public linked immigration to multiculturalism, and that it saw multiculturalism as divisive, formed the centrepiece of the FitzGerald Report (FitzGerald 1988, pp. 10–11). This was an idea to which Howard already subscribed. As the Liberal Party President, John Elliot, said to him after the report was published: ‘It’s what you and I believe in’ (quoted in Kelly 1994, p. 422).
MULTICULTURALISM The claim that the ‘M’ word – multiculturalism – had fallen on bad times, either in 1988 or by the time Howard came to office, is difficult to establish. Depending on how it is presented to survey respondents, statements about multiculturalism can be widely accepted, widely opposed, or can divide opinion down the middle. In surveys conducted between 1988 and 1997, at least two-thirds of respondents held positive views about Australia as a multicultural society and supported the idea that multiculturalism should be government policy (Goot 1999, table 9). However, attitudes to some multicultural policies have been far from positive. The idea that ‘ethnic minorities should be given government assistance to preserve their customs and traditions’ was widely rejected in 1995 and again in 2003 (see table 11.2). Support for ethnic distinctiveness – as against ‘blending in’ – is equally low. And in both 1995 and 2003, almost three in four respondents preferred the statement ‘it is better if these groups adapt and blend into the larger society’ to the view that ‘it is better for a country if different racial and ethnic groups maintain their distinct customs and traditions’. At face value, these findings represent a rejection not just of ‘hard’ multiculturalism – John Hirst’s phrase for ‘government support for migrant cultures’ – but also of ‘soft’ multiculturalism – those attitudes of Table 11.2 Attitudes to maintaining ethnic distinctiveness, 1995 and 2003, per cent
Agree that it is impossible for people who do not share Australian customs and traditions to be fully Australian Agree that ethnic minorities should be given government assistance to preserve their customs and traditions It is better for society if groups … Maintain distinct traditions and customs, OR Adapt and blend into the larger society n
1995
2003
41
42
16
16
12 73
16 71
(2438)
(2128)
SOURCES International Social Survey Program 1995; The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
186 •
8:55 AM
Page 186
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 11.3 Impact of immigrants, 1995–2003, per cent agree 1995
1996
1998
2001
2003
Immigrants increase crime rates
31
51
46
45
34
Immigrants are generally good for Australia’s economy
63
49
55
52
69
Immigrants take jobs away from people who were born in Australia
36
40
36
34
25
Immigrants make Australia open to new ideas and cultures
86
78
78
73
74
(2438)
(1759)
(1897)
(2010)
(2160)
n
SOURCES International Social Survey Program 1995; Australian Election Study 1996, 1998, 2001; The Australian Survey of Social Attitudes 2003
‘tolerance’ and ‘satisfaction’ that Australians have ‘long displayed’ in ‘seeing migrants participate in Australian life’ (Hirst 1994, p. 2). But since there may be no contradiction between maintaining ‘distinct customs and traditions’ and ‘adapt[ing] and blending into the larger society’, the opposition of ethnic distinction versus ‘blending in’ may be overdrawn. Certainly, most AuSSA 2003 respondents do not regard ‘blending in’ as a pre-condition for being ‘fully Australian’. Asked whether they agree that ‘It is impossible for people who do not share Australian customs and traditions to be fully Australian’, respondents divide evenly. Some kinds of cultural differences, far from being feared, are welcomed. Table 11.3 shows that, in surveys conducted between 1995 and 2003, most respondents agreed that ‘Immigrants make Australia more open to new ideas and cultures’ (1995–2001) and that ‘Immigrants improve Australian society by bringing in new ideas and cultures’ (2003). For both statements, the proportion disagreeing failed to reach double figures. Even if the ‘new cultures’ are simply new cuisines, their significance should not be underestimated.
PERCEPTIONS OF THE ECONOMIC AND S O C I A L I M PA C T O F M I G R A N T S If attitudes to multiculturalism, as measured by AuSSA 2003, appear not to have shifted under Howard, attitudes to the impact of migrants on the economy, on employment opportunities and on crime rates have shifted – and in a positive direction. Australians now hold more favourable
AusSocAtts03
26/7/05
8:55 AM
Page 187
I M M I G R AT I O N , M U L T I C U L T U R A L I S M A N D N AT I O N A L I D E N T I T Y
• 187
views of migrants in relation to the economy, jobs and crime than at any time since the 1996 election. On the evidence of AuSSA 2003, changes in attitudes since 2001 have been dramatic. Immigrants are seen as good for Australia’s economy. About half the respondents in 1996, 1998 and 2001 agreed that immigrants are ‘generally good for Australia’s economy’; in 2003 more than two-thirds held this view (see table 11.3). While a substantial proportion (21 per cent) of respondents to AuSSA 2003 appear not to have a view one way or the other, relatively few think that immigrants are not ‘generally good for Australia’s economy’ (8 per cent). Respondents divide more evenly over whether ‘immigrants take jobs away from people who were born in Australia’. Agreement peaked at 40 per cent in 1996 at Howard’s first election victory and has since declined to 25 per cent. On whether immigrants increase crime, the pattern differs again. In 1995, a third of respondents agreed, rising to about one-half of respondents in surveys between 1996 and 2001, but falling again in 2003 to about one-third (see also chapter 9). The connection between the actions of the Howard government and changes in the perception of migrants is not entirely clear. Positive feelings about migrant contributions to the economy and the relatively low level of support for the view that migrants are a threat to other people’s jobs are likely to reflect, among other things, years of economic prosperity. But if there is a link between things for which the Howard government might take some of the credit and changing views about migrant crime – especially the dramatic drop, between 2001 and 2003, in the numbers who associate migrants with an increase in crime – it is far from obvious.
N AT I O N A L I D E N T I T Y In a society of migrants, what makes someone ‘truly Australian’? According to surveys taken since the mid-1990s, most respondents agree that what makes someone ‘truly Australian’ is ‘feeling’ Australian, having Australian citizenship, respecting Australia’s political institutions and laws, and speaking English (see table 11.4). Importantly, these are qualities that can be achieved; they are not merely ascribed characteristics (McAllister 1997, p. 14) like having lived in Australia, being Australianborn, or having English as one’s mother tongue. However, survey respondents also endorse attributes that would rule out many – perhaps most – of their compatriots. Two-thirds believe that living in Australia is at least ‘fairly important’ to being ‘truly’ Australian. Of the third that do not, many may recognise that nearly a million
AusSocAtts03
26/7/05
188 •
8:55 AM
Page 188
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 11.4 Truly Australian, 1995–2003, per cent agree 1995
1996
1999
2001
2003
Speak English
86
90
88
89
92
Feel Australian
93
89
88
88
91
Have Australian citizenship
87
88
90
87
89
Respect Australia’s political institutions and law
93
91
90
90
89
Life mostly in Australia
60
67
67
66
68
Born in Australia
55
57
54
56
58
–
–
–
–
37
Have Australian ancestry Be Christian n
31
35
29
32
36
(2438)
(1795)
(2311)
(2010)
(2125)
QUESTION ‘Some people say the following things are important for being truly Australian. Others say they
are not important. How important do you think each (thing/of the following) is?’ SOURCES International Social Survey Program 1995; Australian Election Study 1996, 2001; Australian
Constitutional Referendum Study 1999; The Australian Survey of Social Attitudes 2003
Australians live overseas (Fullilove & Flutter 2004). More than half (58 per cent in 2003) think that it is ‘fairly important’ for ‘true’ Australians to be born in Australia. About a third (37 per cent) think that Australian ancestry is ‘fairly important’; the others (63 per cent) may recognise that many Australians have a migrant heritage. And one-third say it is important to be Christian; the rest may recognise that the proportion of Australians identifying themselves as Christian in the 2001 Census had dropped to 68 per cent (ABS 2004).2 In addition to giving us an idea about what makes us ‘truly’ Australian, the same surveys give us an idea of how positively we value our country. Almost all respondents agree that they would ‘rather be a citizen of Australia than of any other country in the world’, that ‘Australia is a (much) better country than most other countries’, and that when ‘my country does well in international sports, it makes me proud to be an Australian’ (see table 11.5). Other national evaluations are more contentious. Opinion divides over whether ‘the world would be a much better place if people from other countries were more like Australians’ – table 11.5 shows that less than half (43 per cent) of respondents to AuSSA 2003 agree. Although no more than 24 per cent agree that they are ‘often less proud of Australia’ than they ‘would like to be’, only 27 per cent of respondents agree that ‘people should support their country even if the country is in the wrong’. And opinions also divide over whether ‘there are some things about Australia today that make me feel ashamed of Australia’. In
AusSocAtts03
26/7/05
8:55 AM
Page 189
I M M I G R AT I O N , M U L T I C U L T U R A L I S M A N D N AT I O N A L I D E N T I T Y
• 189
Table 11.5 Supporting Australia, 1995–2003, per cent in agreement
Rather be a citizen of Australia than of any other country in the world Generally speaking, Australia is a (much) better country than most other countries There are some things about Australia today that make me feel ashamed of Australia The world would be a much better place if people from other countries were more like Australians People should support their country even if the country is in the wrong I am often less proud of Australia than I would like to be When my country does well in international sports, it makes me proud to be an Australian n
1995
1996
1999
2001
2003
87
88
86
87
85
83
82
77
–
82
61
–
–
–
47
38
47
54
–
43
23
–
–
–
27
–
–
–
–
24
85
87
84
–
78
(2438)
(1795)
(2311)
(2010)
(2125)
SOURCES International Social Survey Program 1995; Australian Election Study 1996, 2001; Australian
Constitutional Referendum Study 1999; The Australian Survey of Social Attitudes 2003
1995, almost two-thirds of respondents (61 per cent) agreed with that view, but by 2003 less than half agree (47 per cent). It may be that Prime Minister Howard’s own campaign for greater national pride, and a less critical view of Australian history, has had an influence here. What of national pride? There is widespread pride (above 80 per cent) in ‘Australia’s achievements in sports’; more surprisingly, in its ‘scientific and technological achievements’; and most surprisingly, in its ‘achievements in the arts and literature’ (see table 11.6). Not far behind (above 75 per cent) in public esteem are Australia’s ‘armed forces’, ‘the way [its] democracy works’ – one in the eye for those who insist that Australians are highly cynical about their politics and politically disconnected3 – and Australia’s history. Immigrants, it seems, are just as proud as others (Evans & Kelley 2002, p. 325). But these figures hide some important shifts. The proportion feeling ‘very proud’ of the armed forces jumped from 27 per cent in 1996 to 45 per cent in 1999, after Australia’s intervention in East Timor. The number of respondents ‘very proud’ of the way Australian democracy
AusSocAtts03
26/7/05
190 •
8:55 AM
Page 190
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 11.6 Pride in Australia, 1995–2003, per cent agree Pride in …
1995
1996
1999
2003
Its achievements in sports
91
92
90
92
Its scientific and technological achievements
93
89
86
91
Australia’s armed forces*
76
77
81
84
Its achievements in the arts and literature
85
86
83
82
Australia’s economic achievements
48
60
73
80
Way democracy works
76
85
80
78
Its history
72
78
73
76
Its fair and equal treatment of all groups in society
53
–
–
58
Its social security system
50
56
53
57
Its political influence in the world
50
64
60
55
(2438)
(1795)
(2311)
(2125)
n
* Response category in 2003 is ‘Defence forces’. SOURCES International Social Survey Program 1995; Australian Election Study 1996; Australian Constitutional Referendum Study 1999; The Australian Survey of Social Attitudes 2003
works almost doubled (from 15 per cent to 29 per cent) between 1995 and the Coalition’s victory in 1996. The proportion feeling ‘very proud’ of Australia’s history also rose following the debate about Australia’s past in which the Prime Minister (among others) criticised the ‘black armband’4 view of history. Intriguingly, there is now much greater pride in ‘Australia’s economic achievements’. Between 1995 and 2003, the proportion of respondents proud of our ‘economic achievements’ rose steadily from 48 per cent to 80 per cent. What underlies this remarkable trend – low inflation and interest rates, strong economic growth, or something else – is unclear.5 Less widespread was pride in Australia’s ‘political influence in the world’, in ‘its fair and equal treatment of all groups in society’, and in ‘its social security system’. Nonetheless, the majority of respondents express pride in each. Again, trends are revealing. Pride in Australia’s ‘political influence in the world’ rose from 50 per cent in 1995 to 64 per cent in 1996, but (East Timor notwithstanding) has declined modestly since to 55 per cent in 2003. Pride in the ‘fair and equal treatment of all groups’ appears to be greater in 2003 than it was in 1995 – a reflection, perhaps, of the government’s pledge to govern ‘For all of us’. And pride in Australia’s ‘social security system’ appears to have been slightly higher since Howard came to office; perhaps because of the way the government has targeted its welfare spending, including its expanded system of family payments.
AusSocAtts03
26/7/05
8:55 AM
Page 191
I M M I G R AT I O N , M U L T I C U L T U R A L I S M A N D N AT I O N A L I D E N T I T Y
• 191
I M M I G R AT I O N , M U LT I C U LT U R A L I S M A N D N AT I O N A L I D E N T I T Y Does a strong sense of national identity imply a resistance to immigration or wariness towards outsiders? The answer depends on how we define national identity and on prevailing economic and social conditions. Nationalism has as many dimensions as it has synonyms – patriotism, jingoism, chauvinism, and so on. To distinguish these dimensions, we use factor analysis. This technique establishes which items in tables 11.4, 11.5 and 11.6 are highly correlated and reduces them to a small number of factors – in this case, two – that are uncorrelated (see Appendix, table A11.1). The first factor includes three items about being ‘truly Australian’ – being born in Australia, having lived in Australia for most of one’s life, and having an Australian ancestry. Following Jones (1996; 1997) we call this a ‘nativism’ factor.6 The second factor combines pride in Australia’s political influence in the world and pride in Australia’s achievements in sport. We call this factor ‘national pride’. How do attitudes to immigration relate to nativism and national pride? Views about immigration levels divide respondents quite sharply on the nativism scale, but less sharply on the national pride scale (see table 11.7). On the nativism scale, low scorers are much more inclined to support immigration, while high scorers believe that immigration should be reduced.7 The difference between those who score low and high on the national pride scale is only a third as great. Some studies have found that both education and occupation are important demographic factors that shape attitudes to immigration; indeed, a whole theory about ‘a new class’ – or status group – of universityeducated workers in knowledge industries pitted against other Australians builds on this (Betts 1999). Other studies have highlighted different factors: unemployment, presumably on labour market grounds Table 11.7 Mean scores on nativism and national pride scales by attitudes to immigration, 2003 Attitudes to immigration
Nativism
National pride
Increase a lot
–0.65
–0.39
Increase a little
–0.41
–0.06
Remain the same
–0.12
0.01
Reduce a little
0.25
0.13
Reduce a lot
0.62
0.02
(n = 1980) SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
192 •
8:55 AM
Page 192
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
(McAllister 1993; see also Betts 1996, pp. 12–13); whether one is of non-English-speaking background (Betts 1996, pp. 16–17; Betts 1999, p. 356), presumably on solidarity grounds; and country versus city residence (Betts 1996, p. 11), presumably on cultural grounds. We certainly find a strong association between support for immigration and university education. Table 11.8 shows that, while 18 per cent of universityeducated respondents say that immigration should be reduced, 49 per cent of those who had not completed Year 10 want it reduced. However, because education and occupation tend to vary together, we may not learn much more by examining how attitudes to immigration vary with occupation. And because both education and occupation are related to place of residence, we may not learn anything more from examining how attitudes to immigration vary with where respondents live. Instead, we use multivariate analysis to reveal the net impact of a range of variables on attitudes, so that we can determine which are the most important. In other words, we can examine the association between, say, education and attitudes to immigration, with all other factors (such as occupation and place of residence) held constant. The results of this modelling (see Appendix, table A11.2) confirm the pre-eminent importance of education: the less formal education respondents have, the more likely they are to want immigration reduced. Occupational background has a much weaker (and not statistically significant) effect, with blue-collar workers (except labourers) more inclined than professionals to support reduced immigration. Unemployed respondents are more inclined to support lower immigration, though again the difference is not statistically significant. Area of residence is important: inner-metropolitan residents are less likely to support reducing immigration than residents in rural and outer metroTable 11.8 Attitudes to immigration by education, 2003, per cent Increase
A lot
Same
A little
Reduce
A little
A lot
Less than Year 10 (n = 279)
2
10
39
24
25
Year 10 (n = 307)
2
13
43
18
25
Year 12 (n = 241)
7
15
41
23
14
Trade (n = 240)
3
13
38
25
20
Certificate/Diploma (n = 431)
6
15
42
20
16
University degree and above (n = 449)
10
27
46
11
6
SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
8:55 AM
Page 193
I M M I G R AT I O N , M U L T I C U L T U R A L I S M A N D N AT I O N A L I D E N T I T Y
• 193
Table 11.9 Predicted agreement with lower immigration by nativism and national pride, 2003, per cent Nativism scale
National pride scale
Low
16
33
Moderately low
23
34
Middle
33
34
Moderately high
44
34
High
56
35
(n = 1980) SOURCE The Australian Survey of Social Attitudes 2003
politan areas. Compared to Australian-born respondents, firstgeneration migrants are opposed to reducing immigration and secondgeneration migrants also lean this way (though the difference is not statistically significant). Speaking a language other than English at home also predicts support for immigration. However, age produces a few surprises, with those aged under 25 more inclined than older respondents to want immigration reduced. Gender does not significantly affect support. How do orientations towards nativism and national pride shape attitudes to immigration? Nativism is a good predictor of hostility to immigration; national pride is not (see Appendix, table A11.2).8 As table 11.9 shows, as we move up the nativism scale, predicted agreement with reduced immigration rises from 16 per cent to 56 per cent. In other words, respondents who are attached to the idea that being truly Australian involves being born here and living here are those most likely to resist immigration. However, shifting along the national pride scale yields no significant change: how proud respondents are of Australia’s influence in the world or of Australia’s sporting achievements does not affect their attitudes to immigration.
T H E E N D U R I N G I M P O R TA N C E O F N AT I V I S M While nativism is strongly associated with hostility to immigration, is it reducible to other attitudes and value orientations that we have not yet examined? To answer this question, we measure whether economic and political attitudes and value orientations reduce the strength with which nativist attitudes shape attitudes to immigration. In a three-step process, we measure first the impact of economic attitudes. Previous research suggests that being unemployed affects attitudes to immigration (McAllister 1993). More generally, it is commonly taken
AusSocAtts03
26/7/05
194 •
8:55 AM
Page 194
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
for granted that attitudes to immigration are affected by economic security. That is, people who feel that their own livelihood or standard of living is threatened are more likely to be hostile to immigration (see, for example, Birrell & Birrell 1987, pp. 238ff; Betts 1999, pp. 320–22). To test this argument, we include three economic variables: respondents’ perceptions about changes in unemployment over the previous two years; respondents’ reports about how they are managing on their current household incomes; and respondents’ perceptions of economic mobility – that is, whether they feel that people like them and their families have a good chance of improving their standard of living. We combine items two and three to create an ‘economic alienation’ variable.9 Research also suggests that political attitudes are linked to attitudes to immigration. An analysis of the 1984–85 National Social Science Survey (NSSS), found that ‘Australians who are right-wing in orientation’, and ‘pay little attention to politics’, were most likely to feel ‘a deep and profound attachment to the symbolic boundaries of the national community’ (Phillips 1996, p. 128). An analysis of the 1995 NSSS found ‘ethno-nationalists’ disproportionately likely to vote for the Coalition (Pakulski & Tranter 2000a, p. 43). And an analysis of the 1996 AES noted that, while ‘conservative opinions’ about immigrants was ‘not confined to one side of politics’, they were ‘significantly more likely to be found “on the right”’ (Jackman 1998, p. 182). So we next include two political variables: respondent voting intentions, and perceptions of how important respondents rate politics to be in their lives. Class self-identification (middle-class or working-class) is also included. Finally, previous research suggests that those who hold socially conservative attitudes are likely to be more hostile to immigration. For example, in the 2001 AES, those respondents who agreed that the number of migrants coming to Australia should be reduced were much more likely to endorse a raft of conservative positions – that there should be fewer migrants from the Middle East, that Aboriginal land rights had gone too far, that stiffer sentences were needed for people who broke the law, and that the death penalty should be reintroduced for murder (Goot & Sowerbutts 2004). So, we add three variables that measure value orientations: attitudes to capital punishment, to welfare (we combine attitudes to cutting welfare and perceptions of difficulty of qualifying for welfare benefits, to create a ‘welfare hostility’ variable) and a dummy variable that controls for regular attendance at religious services (at least once a week). The result is clear: the nativism scale is very robust, barely shifting
AusSocAtts03
26/7/05
8:55 AM
Page 195
I M M I G R AT I O N , M U L T I C U L T U R A L I S M A N D N AT I O N A L I D E N T I T Y
• 195
as we add the economic, political and values variables to the model (see Appendix, table A11.3). In other words, respondents holding strongly nativist attitudes still do so, even when we take into account their economic and political attitudes and their value orientations. But the new variables do add further insight into anti-immigration attitudes. Perceptions of unemployment matter less than respondents’ personal financial circumstances. Respondents suffering financial hardship and those without hope of improving living standards are twice as likely to want lower immigration as respondents who think that their financial circumstances and living standards are likely to improve. On the political variables, left-of-centre respondents (especially Australian Democrats) are less likely than Coalition voters to want immigration reduced. By contrast, One Nation voters are nearly two and a half times as likely as Coalition voters to prefer reductions. Those who do not feel that politics is very important in their lives are more likely to want immigration reduced; and so are those who identify as working-class. Value orientations also make a difference, with conservatism associated with hostility to immigration. Respondents who support capital punishment are nearly twice as likely to want immigration reduced,10 and hostility to welfare also predicts support for lower immigration. But respondents who attend church at least once a week are about half as likely to want lower immigration compared to less actively religious and non-religious respondents. And what among these factors ranks as most powerful in explaining support for lower immigration? Voting One Nation is strongest, followed by economic alienation, support for capital punishment, and being hostile to welfare.11 Economic alienation remains important, even after we control for nativism. At every level of nativism, economically Table 11.10 Predicted agreement with lower immigration by nativism and economic alienation, 2003, per cent Nativism score (quintiles)
Economically alienated Not economically alienated
Low
27
15
Moderately low
37
22
Middle
48
30
Moderately high
58
40
High
68
51
(n = 1980) SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
196 •
8:55 AM
Page 196
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
alienated respondents are more likely to support lower immigration. Even at the lower end of the nativism scale, we predict that 27 per cent of economically alienated respondents agree with lower immigration, compared to 15 per cent who are not (table 11.10). Support for higher immigration is predicted most strongly by voting Democrat, followed by religiosity.
N AT I V I S M , I M M I G R A N T S A N D MULTICULTURALISM While nativism matters more than national pride when it comes to explaining support for the level of immigration, does the same contrast operate when it comes to explaining attitudes towards immigrants themselves? To answer this question, we examine the impact of the nativism and national pride scales on two positive and two negative statements about migrants. The results are striking. As table 11.11 shows, few at the lower end of the nativism scale agree that immigrants increase crime rates or take jobs from people born in Australia; but a majority at the upper end hold this view. For both items, the gap between the lowest and highest group on the nativism scale is large – around 45 percentage points. By contrast, more than half of those who score high on the nativism scale believe that migrants are generally good for the economy, and that they improve Australian society by bringing in new ideas and cultures. While these figures fall short of those recorded by respondents who score low on nativism – the gap between lower and upper groups is 28 and 36 percentage points respectively – they suggest, nevertheless, that even respondents who score high on this scale acknowledge the benefits that immigrants bring.12
Table 11.11 Predicted agreement with statements about immigrants by nativism, 2003, per cent Nativism score (quintiles)
Increase crime rates
Take jobs Good for Bring new ideas from Australians Australia’s economy and cultures
Low
14
7
81
89
Moderately low
22
11
75
80
Middle
30
21
70
74
Moderately high
45
35
63
69
High
59
51
53
53
(n = 1980) SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
8:55 AM
Page 197
I M M I G R AT I O N , M U L T I C U L T U R A L I S M A N D N AT I O N A L I D E N T I T Y
• 197
Table 11.12 Predicted agreement with statements about multiculturalism by nativism, 2003, per cent Nativism score (quintiles)
Share customs and traditions to be fully Australian
Better if distinct Better if groups customs and adapt and traditions blend into maintained society
Government assistance for ethnic minorities
Low
21
33
52
28
Moderately low
33
18
65
18
Middle
43
15
73
16
Moderately high
52
9
80
12
High
67
6
85
7
(n = 1980) SOURCE The Australian Survey of Social Attitudes 2003
Does the same relationship hold for nativism and attitudes to ‘soft’ multiculturalism? Apparently it does. For the statement it is ‘impossible for people who do not share Australia’s customs and traditions to become fully Australian’, we find a 46 percentage point difference between respondents in the lower and upper groups on the nativism scale – about the same difference there is on whether migrants increase crime or take jobs from people born in Australia (compare tables 11.11 and 11.12). In other words, those who believe that being ‘truly Australian’ means being born in Australia and living here are much more likely to agree that it is impossible to become fully Australian without sharing Australia’s customs and traditions. We also find a large difference (27 percentage points) between ‘nativists’ and ‘non-nativists’ on whether it is ‘better if groups maintain their distinct customs and traditions’ – a margin similar to responses to whether migrants are good for the economy. And the difference (33 percentage points) between the proportions high on nativism and low on nativism who agree that it is ‘better if groups adapt and blend into the larger society’ is just as great as the corresponding difference over whether migrants ‘improve Australia by bringing in new ideas and cultures’. When it comes to ‘hard multiculturalism’, nativism appears to be a weaker predictor of variance. The difference between respondents in the lower and upper groups on the nativism scale on whether ‘ethnic minorities should be given government assistance to preserve their customs and traditions’ was a relatively low 21 percentage points.
AusSocAtts03
26/7/05
198 •
8:55 AM
Page 198
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
N AT I O N A L P R I D E , I M M I G R A N T S AND MULTICULTURALISM Does national pride explain as much about attitudes to immigrants as it explains about attitudes to multiculturalism? The answer is no. This confirms the findings of our earlier modelling. Whether or not respondents are proud of Australia’s influence in the world or of its sporting achievements has little bearing on their attitudes to immigrants. As table 11.13 shows, those who score relatively low on national pride are less likely than those who score high to believe that migrants increase crime. But national pride makes little difference, except at the extremes, to beliefs about whether migrants are good for the Australian economy or whether they bring new ideas and cultures, and no significant difference to beliefs about the effect of migrants on job opportunities. On questions about multiculturalism, the national pride scale generally yields weak findings as well. One exception is responses to the statement that it is ‘Better if groups maintain their distinct traditions and Table 11.13 Predicted agreement with statements about immigrants by national pride, 2003, per cent National pride score (quintiles)
Increase crime rates
Take jobs from Good for Bring new ideas Australians Australia’s economy and cultures
Low
25
21
64
72
Moderately low
32
26
62
68
Middle
33
28
69
73
Moderately high
38
25
70
73
High
40
22
77
89
(n = 1980) SOURCE The Australian Survey of Social Attitudes 2003
Table 11.14 Predicted agreement with statements about multiculturalism by national pride, 2003, per cent National pride score (quintiles)
Share customs and traditions to be fully Australian
Better if distinct Better if groups customs and adapt and traditions blend into maintained society
Government assistance for ethnic minorities
Low
34
24
56
24
Moderately low
38
17
69
15
Middle
44
14
75
16
Moderately high
45
14
74
13
High
45
11
81
13
(n = 1980) SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
8:55 AM
Page 199
I M M I G R AT I O N , M U L T I C U L T U R A L I S M A N D N AT I O N A L I D E N T I T Y
• 199
customs rather than adapt and blend into the larger society’. While respondents at all steps in the national pride scale are in support, the gap between those in the highest and lowest scoring groups is substantial at 25 percentage points (see table 11.14).
CONCLUSION The events surrounding Tampa led many to believe that the public is now more wary of immigration. This view turns out to be mistaken. The government’s strong distinction between ‘legal’ and ‘illegal’ immigrants may not have boosted support for its immigration program, as Hirst (2002, p. 91) argues; but there is certainly no evidence that it has reduced support. The Howard years have seen not just diminishing opposition to immigration, but the longest period of majority support for the program since the heyday of immigration between the late 1950s and the early 1970s. And increased support for immigration has been accompanied by more positive attitudes to the contribution of migrants, especially to the economy, and less negative attitudes about things that are often feared – migrants increasing crime and taking jobs. Nor have the Howard years been marked by a souring of attitudes on the broader cultural front. Our evidence does not point to lower levels of pride in Australian democracy, in Australian history, or in the sense that Australia gives ‘fair and equal treatment’ to ‘all groups’. Fewer respondents feel ashamed of Australia. If the country is going through ‘a radical period’ of ‘redefinition of its cultural institutions and conventions’ and is ‘still a very long way from being able to define’ its ‘national identity’, as social researcher Hugh Mackay (1993, p. 25) argued a few years before Howard came to office, there is no sign of it in these findings. The call for a new ‘national identity that is relevant for all of us’ (Duncan et al. 2004, p. 11) is likely to be ignored both by those satisfied with Australian identity as it stands and by those who realise that a new identity embraced by ‘all of us’ that ‘resolves the tensions’ between ‘urban elites’ and ‘ordinary Australians’ is a fantasy. Our findings show the powerful effects on attitudes to immigration, not only of education, economic alienation and social values, but also of party choice (with the Democrats at one end of the spectrum, and One Nation at the other). In an age in which party differences and the power of parties to shape public opinion are widely discounted, this is worth stressing. One final point is that, when it comes to explaining attitudes to immigration and multiculturalism, nativism trumps national pride.
AusSocAtts03
26/7/05
200 •
8:55 AM
Page 200
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Reporting on their study of popular understandings of the term ‘unAustralian’ conducted in Queensland in 1997, Philip Smith and Tim Phillips observe that participants frequently singled out Pauline Hanson’s ‘divisive politics’ as ‘unAustralian’, even though their own opinions on ‘Chinatowns, the Gold Coast and strongly ethnic suburbs’ were ‘broadly consistent’ with those of Hanson (Smith & Phillips 2001, pp. 327, 337). This reinforces our finding that nativism scores are better at predicting attitudes to items on multiculturalism than whether the kind of multiculturalism in question was ‘soft’ or ‘hard’. While we have pointed to clear differences between the attitudes of ‘nativists’ and ‘non-nativists’, we should also note that on some items to do with migrants (that they are good for Australia’s economy; that they bring new ideas) and multiculturalism (that groups should adapt and blend; that government should assist minorities to preserve their customs and traditions) both nativists and non-nativists are on the same side. That is, both nativists and non-nativists are pro-migrant and antimulticulturalism. This suggests, as was the case before Howard came to office, that nativism or ethno-nationalism should not necessarily be equated with the views often associated with the ‘extreme right’ (Pakulski & Tranter 2000b, p. 217). NOTES
1 2 3 4 5
6
There are different ways of measuring this; compare Betts (1999, p. 328) and Goot (2000, p. 55). For a very different list of ‘Australian’ values, volunteered by respondents in a qualitative Queensland study, see Phillips and Smith (2000, table 6). For an extended discussion, see Goot (2002). Geoffrey Blainey coined the expression ‘black armband’ to describe what he considered to be an undue focus on negative aspects of Australian history; see Macintyre and Clark (2003, pp. 2–4, 13) Whether any of these changes worked to the government’s electoral advantage is another matter. In 1996, according to McAllister (1997, p. 15), ‘institutional nationalism’ as he calls it – pride in Australia’s armed forces, the way Australian democracy works, Australia’s economic achievements, and so on – worked to Labor’s advantage. Pride in Australia’s cultural achievements or views about what made one ‘truly’ Australian counted for nought. In their analysis of the 1995 data, Pakulski and Tranter (2000a; 2000b) prefer the term ‘ethno-nationalism’. According to the Oxford English Dictionary, ‘Nativism’ emerged in the United States in the 1850s and meant ‘prejudice in favour of natives against strangers’; during the first election for the New South Wales Legislative Assembly, Henry Parkes lamented the absence of ‘an Australian native party’ (An Elector 1856). The Australian Natives’ Association, formed in 1871, saluted the Australian-born sons of immigrants as the ‘true Australians’; it was a progenitor of the White Australia policy and a defender of the country’s ‘racial heritage’ to the end (Menadue 1971, pp. 252–54). The respondents we designate as ‘nativists’ are closer to Evans, Jones
AusSocAtts03
26/7/05
8:55 AM
Page 201
I M M I G R AT I O N , M U L T I C U L T U R A L I S M A N D N AT I O N A L I D E N T I T Y
7 8
9
10 11 12
• 201
and Kelley (1988, p. 112), for whom Australia’s ‘native stock’ consists of residents who are ‘Australian-born of Australian-born parents.’ In his paper dealing with this theme, Barkan (2003, p. 271) draws on Higham’s definition of nativism as ‘an aroused and defensive nationalism’. Both scales are centred on 0, but their distributions differ, with the national pride scale much more skewed (scales range from –2.4 to 2.0 for nativism and from –4.2 to 1.8 for national pride). Predicted probabilities are calculated by setting the scales at varying values (–1.0, –0.5, 0, 0.5 and 1.0) and leaving all other variables set to their mean values. The categories (low through to high) correspond to the varying values (–1.0 through to 1.0). These categories represent quintiles of respondents. We did not add some of the other economic variables – concern about losing one’s job and perceived difficulty of getting another one, for example – because they only applied to people in employment and would have reduced the number of observations too much. In 2001, attitudes to capital punishment helped drive attitudes to asylum seekers (Goot & Sowerbutts 2004). This, together with evidence from the 1998 AES (Goot & Watson 2001), argues against the idea that support for One Nation was largely a revolt of the economically discontented. For a different analysis of the 1995 data, but similar results, see Pakulski and Tranter (2000a, p. 47).
The authors thank Katherine Betts and Kylie Brass for providing information at various points and John Buchanan, along with the editors, for their comments. REFERENCES
Australian Bureau of Statistics (ABS) (2004) ‘Culture and recreation: Religion’, in Year Book Australia. Available at: (accessed 15 July 2004). An Elector (1856) ‘The Coalition candidates’, Freeman’s Journal, 1 March. Barkan, E (2003) ‘Return of the nativists? Californian public opinion and immigration in the 1980s and 1990s’, Social Science History, 27(2): 229–83. Betts, K (1996) ‘Immigration and public opinion in Australia’, People and Place, 4(3): 9–20. —— (1999) The Great Divide: Immigration Politics in Australia, Duffy & Snellgrove, Sydney. —— (2000) ‘Immigration: Public opinion and opinions about opinion’, People and Place, 8(3): 60–67. —— (2003) ‘Immigration policy under the Howard government’, Australian Journal of Social Issues, 38(2): 169–92. Birrell, R & Birrell, T (1987) An Issue of People: Population and Australian Society, 2nd edn, Longman Cheshire, Melbourne. Devine, M (2002) ‘When allegations go overboard’, Sydney Morning Herald, 27 June. Duncan, M, Leigh, A, Madden, D & Tynan, P (2004) Imagining Australia: Ideas For Our Future, Allen & Unwin, Sydney. Evans, MDR, Jones, FL & Kelley, J (1988) ‘Job discrimination against immigrants? Perceptions, personal experience, norms and reality’, in J Kelley & C Bean (eds) Australian Attitudes: Social and Political Analyses from the National Social Science Survey, Allen & Unwin, Sydney, pp. 111–27. Evans, MDR & Kelley, J (2002) ‘National pride in the developed world: Survey data from 24 nations’, International Journal of Public Opinion Research, 14(3): 303–38.
AusSocAtts03
26/7/05
202 •
8:55 AM
Page 202
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
FitzGerald, S (1988) Chair. Immigration: A Commitment to Australia, The Report of the Committee to Advise on Australia’s Immigration Policies, Australian Government Publishing Service, Canberra. Fullilove, M & Flutter, C (2004) Diaspora: The World Wide Web of Australians, Lowy Institute, Sydney. Gibson, R, Wilson, S, Meagher, G, Denemark, D & Western, M (2004) Australian Survey of Social Attitudes 2003, Data file, Australian Social Science Data Archive, The Australian National University, Canberra. Goot, M (1999) ‘Migrant numbers, Asian immigration and multiculturalism: Trends in the polls, 1943–98’, Australian Multiculturalism in a New Century: Towards Inclusiveness, Statistical Appendix, National Multicultural Advisory Council, Canberra, pp. 28–60. —— (2000) ‘More “relaxed and comfortable”: Public opinion on immigration under Howard’, People and Place, 8(3): 46–60. —— (2002) ‘Distrustful, disenchanted and disengaged? Public opinion on politics, politicians and the parties: An historical perspective’, in David Burchell & Andrew Leigh (eds), Prince’s New Clothes: Why Do Australians Dislike their Politicians? UNSW Press, Sydney, pp. 9–46. —— (2004) ‘Party convergence reconsidered’, Australian Journal of Political Science, 39(1): 49–73. —— & Sowerbutts, T (2004) ‘Dog whistles and death penalties: The ideological structuring of Australian attitudes to asylum seekers’, Australasian Political Studies Association Conference, Adelaide: September 20 – October 1. —— & Watson, I (2001) ‘One Nation’s electoral support: Where does it come from, what makes it different, and how does it fit?’, Australian Journal of Politics and History, 47(2): 159–91. Hewett, J (2001) ‘Skipper Howard all at sea without a moral compass’, Sydney Morning Herald, 31 August. Hirst, J (1994) ‘National pride and multiculturalism’, People and Place, 2 (3): 1–7. —— (2002) ‘Correspondence’, Quarterly Essay, no. 6: 89–91. Jackman, S (1998) ‘Pauline Hanson, the mainstream, and political elites’, Australian Journal of Political Science, 33(2): 167–86. Jones, FL (1996) ‘National identity and social values’, People and Place, 4(4): 17–26. —— (1997) ‘Ethnic diversity and national identity’, Australian and New Zealand Journal of Sociology, 33: 285–304. Jupp, J (2002) From White Australia To Woomera, Cambridge University Press, Cambridge. Kelly, P (1994) The End of Certainty, rev. edn, Allen & Unwin, St Leonards. Long, JS (1997) Regression Models for Categorical and Limited Dependent Variables, Sage, Thousand Oaks. McAllister, I (1993) ‘Immigration, bi-partisanship and public opinion’, in J Jupp & M Kabala (eds), The Politics of Australian Immigration, Australian Government Publishing Service, Canberra, pp. 161–78. —— (1997) ‘Political culture and national identity’, in B Galligan, I McAllister & J Ravenhill (eds) New Developments in Australian Politics, Macmillan, South Melbourne, pp. 3–21. Macintyre, S & Clark, A (2003) The History Wars, Melbourne University Press, Carlton. Mackay, H (1993) ‘A national identity? Wait and see’, in J Beaumont (ed.) Where to Now? Australia’s Identity in the Nineties, Federation Press, Sydney, pp. 12–25.
AusSocAtts03
26/7/05
8:55 AM
Page 203
I M M I G R AT I O N , M U L T I C U L T U R A L I S M A N D N AT I O N A L I D E N T I T Y
• 203
Menadue, JE (1971) A Centenary History of the Australian Natives’ Association 1871–1971, Horticultural Press, Melbourne. Pakulski, J & Tranter, B (2000a) ‘Civic identity in Australia’, Australian Journal of Social Issues, 35(1): 35–51. —— (2000b) ‘Civic, national and denizen identity in Australia’, Journal of Sociology, 36(2): 205–22. Phillips, TL (1996) ‘Symbolic boundaries and national identity in Australia’, British Journal of Sociology, 47(1): 113–34. Phillips, T & Smith, P (2000) ‘What is “Australian”? Knowledge and attitudes among a gallery of contemporary Australians’, Australian Journal of Political Science, 35(2): 203–24. Smith, P & Phillips, T (2001) ‘Popular understandings of “unAustralian”: An investigation of the un-national’, Journal of Sociology, 37(4): 323–40. Williams, P (1997) The Victory, Allen & Unwin, St Leonards.
AusSocAtts03
26/7/05
8:55 AM
Page 204
CHAPTER 12
KNOWING YOUR GENES K r i s t i n e B a r l o w - S t e w a r t , S a n d r a Ta y l o r a n d M a r g a r e t O t l o w s k i
INTRODUCTION April 25th is not only Anzac Day. It is also the day when, in 1953, Watson and Crick published the basic chemical structure of life, DNA.1 And on 25 April 2003, the 50th anniversary of that publication, the international Human Genome Project finished a complete map of the 30 000 odd genes that comprise our genetic make-up. Using discoveries from the Project, genetic tests can now detect the information contained in around 1000 genes. There are already tests to determine whether people are at risk of developing various cancers, haemochromatosis,2 and neurological conditions such as early-onset Alzheimer’s disease. And for some conditions, such as Huntington’s disease,3 testing can definitely determine whether a person will develop the condition later in life. Although the range of genetic tests is still quite small, this situation is rapidly changing. Until now, in Australia, almost all genetic testing has been conducted for medical reasons. However, genetic applications are being extended to include DNA fingerprinting in forensic investigations, paternity testing and confirming family connections for immigration purposes (ALRC 2001). With this greater range of testing, the potential for misuse of genetic information is also growing. Given the increasing importance of genetic information in our community, policy-makers need to know more about public awareness of, and support for, genetic technologies. The Australian Survey of Social Attitudes 2003 (Gibson et al. 2004) provides the first comprehensive insight into how much Australians know about genetic technology and their attitudes to its use. This chapter reveals what Australians really think about genetics, and about the possible applications of genetic test-
AusSocAtts03
26/7/05
8:55 AM
Page 205
KNOWING YOUR GENES
• 205
ing in insurance, employment and immigration. Finally, we draw out the policy implications of our findings.
T H E VA L U E O F G E N E T I C I N F O R M AT I O N Genetic testing offers significant information to individuals and families about their health. Even when people find they have a condition for which there are few or no current treatments, genetic tests still enable them to make better decisions. For conditions such as bowel and breast cancer, a positive genetic test might encourage early detection strategies. And a positive genetic test for a predisposition to haemochromatosis might prompt blood donation as a preventative strategy. Testing may also reassure those with a high family risk of developing a genetic condition if they find they have avoided this risk. Genetic testing can be said to offer people new freedoms to make better health and lifestyle choices. Yet, along with the freedoms brought about by new knowledge of our genetics come burdens as well; for instance, the burden of deciding to terminate a pregnancy or making radical lifestyle changes. Someone might discover that they have an untreatable condition or one that requires intensive surveillance to enable early detection. Another possible burden is perhaps becoming a member of a new patient group, the ‘worried-well’, who experience increased anxiety whether or not they actually go on to develop any condition. Frequently, genetic information has implications beyond the individual. Genetic conditions are in effect family health problems because identifying one family member with a genetic condition or risk may mean others are also at risk. Individuals might refuse to share their genetic information or be selective in how it is shared. Genetic knowledge may also impact on family dynamics: those who have a positive test result may find themselves having to communicate complex and difficult information to their relatives. Genetic information may have even wider institutional implications. Commercial organisations, such as insurance companies and employers, could have a strong interest in the future health of a currently asymptomatic person who has had a genetic test and in certain circumstances be able to exert power over that person. While genetic information is currently used by life insurance companies in Australia, there is little evidence that this country is following overseas trends and using it in workplaces (Center for Genetics & Public Policy 2004). The Australian Government already uses genetic testing to check the veracity of claimed family membership when deciding some immigrant family reunion cases.
AusSocAtts03
26/7/05
206 •
8:55 AM
Page 206
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
The Department of Immigration and Multicultural and Indigenous Affairs (DIMIA) is also interested in using genetic testing to ensure that immigrants meet health requirements in the future if such tests become more reliable (ALRC 2003, chapter 37). The value of genetic information raises the spectre of ‘genetic discrimination’ – unjustified treatment of an asymptomatic person because of their genetic make-up. The prospect of using genetic information to discriminate has been comprehensively addressed by the Australian Law Reform Commission (ALRC) and the Australian Health Ethics Committee (AHEC) in their joint report, Essentially Yours (ALRC 2003), which argues that genetic discrimination raises problems that must be addressed by public policy.
HOW AWARE ARE AUSTRALIANS OF G E N E T I C I N F O R M AT I O N ? There are now extensive investigations in other countries of public opinion about genetic technologies and discrimination (for Europe, see Gaskell, Allum & Stares 2003; for the United Kingdom, see Stratford, Marteau & Richards 2001 and Park et al. 2001; and for the United States, see VCU 2001). But less is known here. A recent government report notes that little research has been done on Australian awareness of, and attitudes to, science and technology (Australian Government 2003). However, we do know that, in comparative terms, Australians are both aware and tolerant of technological and scientific change. A 1995 survey placed Australians top overall for accurately answering a science quiz that included two items on genetics and inheritance (Bureau of Industry Economics 1995). And recent opinion polls suggest that interest in science, technology and medical research is high and even increasing slightly (Newspoll 2001; Research Australia 2003). Australians are comfortable with the rate of technological change and have comfort levels similar to those of Americans (Australian Centre for Emerging Technologies and Society 2003). A government agency, Biotechnology Australia, has examined public awareness and attitudes and, although these studies largely focused on attitudes to plant biotechnologies, two surveys have touched on the use of genetic information in medicine, finding that about six in ten people (58 per cent) are concerned that genetic screening for diseases may lead to discrimination (Biotechnology Australia 2003). AuSSA 2003 prompts respondents about their genetics awareness using a list modelled on results from a MORI survey in the United Kingdom (see Park et al. 2001).4 Respondents are asked:
AusSocAtts03
26/7/05
8:55 AM
Page 207
KNOWING YOUR GENES
• 207
When you hear the phrase ‘human genetic information’ which, if any, of the following come to mind? DNA; Inherited health problems; Having healthy babies; Crime scene investigations; The uniqueness of a person; Medical research; Paternity testing; or Other? By far the most common association with ‘human genetic information’, reported by 77 per cent of AuSSA 2003 respondents, is ‘DNA’ (see table 12.1). Around half of respondents nominate ‘medical research’ or ‘inherited health problems’, while between a third and a quarter select ‘the uniqueness of a person’, ‘crime scene investigations’ and ‘paternity testing’. Only 19 per cent include ‘having healthy babies’, which is interesting, since the original focus of genetic testing some 20 years ago was prenatal testing and pediatric conditions. Current understanding links genetic testing to a wider range of health conditions. Other items not on the list, such as ‘cloning’ and ‘stem cells’, are mentioned by less than 1 per cent of Australian respondents, while these items are mentioned by 9 per cent of British respondents in the MORI study (Park et al. 2001). In summary, fully 92 per cent of AuSSA 2003 respondents associate ‘human genetic information’ with one or more of the following: ‘DNA’, ‘medical research’, ‘inherited health problems’ and ‘crime scene investigations’. Although AuSSA 2003 provides respondents with a list of possible answers to choose from, the pattern of their responses indicates that a large majority of Australians are able to identify applications of human genetic information.5 What is the social profile of respondents who are ‘genetically aware’? One way of measuring this is to define the 45 per cent of respondents who mention ‘inherited health problems’ as being genetically aware, Table 12.1 Awareness of ‘human genetic information’, 2003, per cent Items selected
DNA
77
Medical research
56
Inherited health problems
45
Uniqueness of a person
30
Crime scene investigations
29
Paternity testing
25
Having healthy babies
19
(n = 2038) SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
208 •
8:55 AM
Page 208
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
since choosing this item indicates a clear recognition of the main focus of current genetic applications. We find that these respondents tend to be female, work in skilled jobs, are university-qualified, and have the Internet at home (or another access point besides work). Similar demographic factors have been identified in overseas studies (Stratford, Marteau & Richards 2001).
OVERALL SUPPORT FOR GENETIC TESTING AuSSA 2003 asks respondents to choose between the following statements: Genetic testing: Should be allowed because it gives new hope in detecting, treating and even reversing serious medical conditions. Should not be allowed because of the potential misuse of the personal information that the tests reveal. Overall, 66 per cent of the respondents think that human genetic testing should be allowed (see table 12.2). Seventeen per cent believe that testing should not be allowed, although an equal number say they ‘Can’t choose’. A high level of non-responses may indicate some uncertainty about what is at stake ethically, or that the issue of genetic testing is too complex for some respondents. We find that support increases for genetic testing among the group we have defined as genetically aware: 72 per cent of this group support testing. The finding that most people think that the benefits of genetics technologies outweigh the harms is consistent with an earlier finding (see Biotechnology Australia 2003). Similar figures are seen in Europe – 50 per cent (Gaskell, Allum & Stares 2003); the United Kingdom – 90 per cent (Park et al. 2001); and the United States – 57 per cent (VCU 2001). One particularly interesting finding is the variation in responses by age. The Australian Centre for Emerging Technologies and Society (2003) found that younger Australians are significantly more comfortable than older Australians with new technology. But we find here that quite a few younger respondents express uncertainty about genetic testing: 26 per cent aged 29 or under say they ‘Can’t choose’, which falls to just 16 per cent of those 30 and over. We have established that there is broad support across the community for genetic testing, but what factors best explain this support? Multivariate analysis provides further insights (see Appendix, table
AusSocAtts03
26/7/05
8:55 AM
Page 209
KNOWING YOUR GENES
• 209
Table 12.2 Genetic testing or not? 2003, per cent Choices about human genetic testing
Total
With good genetic awareness
Should be allowed because it gives new hope in detecting, treating and even reversing serious medical conditions
66
72
Should not be allowed because of the potential misuse of the personal information that the tests reveal
17
13
Can’t choose
17
15
(n = 2003) SOURCE The Australian Survey of Social Attitudes 2003
A12.1). We use a logistic regression model to test support for genetic testing (the dependent variable) against a set of demographic factors and attitude scales (trust in people, and personal health rated between excellent and poor) as independent variables. We find that men are more likely to support genetic testing than women, and respondents who agree that most people can be trusted are more likely to support testing than those who disagree. There are weak age effects in the model, with 18–34 year old respondents less likely to support testing. So, although we have established that a good majority have general support for testing, we also see a gender and trust divide in support. But does this support translate into support for specific applications of genetic testing; particularly in the areas of insurance, employment and immigration?
G E N E T I C I N F O R M AT I O N A N D I N S U R A N C E In Australia, health insurance is governed by the National Health Act 1953, and this legislation prevents health insurers from using genetic information to rate individual risk. Although a pre-existing illness may be taken into account by insurers, premiums are ‘community rated’ so that everyone pays the same premium for a policy (ALRC 2001). By contrast, life insurance products that cover death, disability, income protection and business expenses are ‘underwritten’, so that factors that may impact on the risk of making a claim, such as current health, family health history, age and lifestyle, are taken into account. Applicants actually have a legal duty to disclose these risk factors under the federal Insurance Contracts Act 1984, where duty of disclosure refers to a ‘matter that a reasonable person in the circumstances could be expected to know’. Assessing such insurance products takes into account human genetic information imputed from family health history, and in some
AusSocAtts03
26/7/05
210 •
8:55 AM
Page 210
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
cases (for example, where there is a history of Huntington’s disease), this impacts on insurability or premiums charged. And if an applicant (or parent or sibling) has had a genetic test, the result, if known to the applicant, must also be declared because it too is considered relevant to risk assessment (Barlow-Stewart 2003). These risk factors are determined by statistical and actuarial evidence, and their use in underwriting has been specifically exempted from the provisions of the federal Disability Discrimination Act 1992. Despite this exemption, the Investment and Financial Services Association (IFSA) says that around 93 per cent of Australian applicants receive a policy at standard rates, 3 per cent have to pay a higher premium or have some conditions excluded, and about 2 per cent are refused (IFSA 2002). Nevertheless, professional groups such as the Human Genetics Society of Australia (HGSA 2001) as well as eminent scientists (see Sample 2004) remain opposed to the use of genetic tests in underwriting because not enough is yet known about the health implications of genetic make-up, and test results lack precision.6 While a negative result frees an applicant from additional costs, a positive result not only exposes an individual and their family to higher premiums or refusal, it may also burden these individuals with worrying health information. One aspect of that burden may be disclosing test results to family (Barlow-Stewart 2003). These burdens may discourage some from seeking genetic testing in the future, even where there are possible benefits. Given current law and insurance practice in Australia, it is surprising that only one in ten of AuSSA 2003 respondents say that genetic information is ‘definitely’ used by insurance companies to assess applicants’ future health problems, while another 40 per cent think that it is probably used (table 12.3). Genetic awareness makes very little difference – again, around half of those who are genetically aware think that it is definitely or probably used. We further use a logistic regression model to determine what demographic and attitudinal factors predict awareness of use of genetic test results in insurance (the dependent variable). We find that few of the independent variables predict awareness – older respondents (65 and over) appear to be more likely to think that testing is used in insurance than younger respondents (see Appendix, table A12.2). Curiously, we find no additional awareness among the socioeconomically advantaged – those with a university degree or a professional or managerial employment background – who are well represented among the 30 per cent or so of Australians who access life insurance products.
AusSocAtts03
26/7/05
8:55 AM
Page 211
KNOWING YOUR GENES
• 211
Table 12.3 Genetic testing in insurance, 2003, per cent Awareness of current use in insurance
Total
Respondents with good genetic awareness
Definitely used
10
11
Probably used
40
41
Probably not used
25
26
Definitely not used
12
13
Can’t choose
13
9
(n = 2019) SOURCE The Australian Survey of Social Attitudes 2003
Current policy in Australia governing the use of genetic test information in underwriting states that, while insurers may request the results of previously undertaken genetic tests to assess risk, they cannot require an applicant to have genetic testing. Additionally, the policy states that genetic tests will not be used in ‘preferred risk underwriting’; that is, to offer individuals insurance at lower than the standard premium rate (IFSA 2002). However, with improvements in the predictive value of genetic testing, insurance companies may seek routine genetic testing, just like other health information, as part of assessing applications. Indeed, as Stratford, Marteau & Richards (2001) suggest for the United Kingdom, widespread opposition to the use of genetic information for insurance purposes may lessen if the tests become more reliable. They point out, however, that current opposition may also be underpinned by normative values, which may not shift. Among AuSSA 2003 respondents, almost four in five (77 per cent) disagree with the statement that ‘insurance companies should be able to genetically test policy applicants to determine their risk of developing health problems’ (see table 12.4). Moreover, 75 per cent of those who think that insurance companies definitely use genetic test results in underwriting still disagree with any requirement for genetic testing. This means that knowing about the reality of genetic testing for insurance purposes does not increase support for a testing requirement. In fact, 90 per cent of those who are genetically aware oppose such a requirement. Similar results are reported elsewhere: over half (56 per cent) of British citizens disagree with testing requirements, although 77 per cent think that it is likely that genetic information will be required in the next 25 years for health and life insurance (Stratford, Marteau & Richards 2001). We use a logistic regression model that predicts disagreement with the use of genetic testing in insurance (see Appendix, table A12.2). We
AusSocAtts03
26/7/05
212 •
8:55 AM
Page 212
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 12.4 Insurance companies should genetically test applicants to determine health risks, 2003, per cent Attitude towards use in insurance
Total
Respondents who believe that tests are definitely used by insurers
Strongly agree
1
2
Agree
8
14
Neither agree nor disagree
10
10
Disagree
37
36
Strongly disagree
40
39
3
1
Can’t choose (n = 2040)
SOURCE The Australian Survey of Social Attitudes 2003
find that women, respondents aged less than 65, and respondents who say they have little or no confidence in banks and financial institutions (which include insurance companies) are all more likely to disagree with testing. Again, low trust appears to be an important factor in explaining disagreement with testing: fully 76 per cent of respondents who disagree with insurance companies using tests have little or no confidence in financial institutions, broadly defined.
G E N E T I C I N F O R M AT I O N A N D E M P L O Y M E N T Genetic testing can be used to assess applicants for employment, as well as for determining the occupational health and safety risks of employees (ALRC 2003, chapters 29 and 32). The recent ALRC/AHEC Inquiry on protection of genetic information reports that genetic testing is probably rare in Australian workplaces at the moment, but suggested that the use of such tests may increase as their reliability improves and as new tests are developed that apply specifically to work (ALRC 2003). As with insurance, there is the potential for new freedoms for individuals, as well as new burdens and the misuse of power. For example, identifying employees who are genetically predisposed to a condition if exposed to a particular environmental factor could be seen as a freedom if that factor could be removed from the workplace or could be avoided. On the other hand, employees may be burdened by the use of genetic information if it meant job loss. For employers, it may be easier and cheaper to employ only those individuals who would be unaffected by the specific environmental factor. The majority of respondents (58 per cent) think that employers probably or definitely do not use genetic testing in the workplace to
AusSocAtts03
26/7/05
8:55 AM
Page 213
KNOWING YOUR GENES
• 213
assess future health problems of job applicants and current employees – a reasonably accurate perception, given its low occurrence (see table 12.5). Sixty-four per cent of this group are genetically aware. Nevertheless, a sizeable minority (29 per cent) think testing is being used in the employment setting, but only 27 per cent of these are genetically aware. A logistic regression model reported in Appendix table A12.3 reveals the factors that predict awareness of the use of testing in employment. We find that women, respondents with a managerial or professional employment background, and respondents who trust others are all more likely to think that genetic testing is not used in the workplace. The MORI study reveals that about seven in ten British respondents think that it is inappropriate for an employer to see genetic test results of current and potential employees if they are used to determine future health or disability (Park et al. 2001). However, other research has established that support for testing in the workplace is closely linked to its purpose (see Stratford, Marteau & Richards 2001). If the test is conducted for safety purposes (for example, to identify those sensitive to workplace chemicals), then four-fifths of British respondents believe that employers have the right to make employees take a genetic test. Australians appear more opposed than British respondents when it comes to testing for occupational health and safety purposes. About three in four Australian respondents oppose workplace testing when they are asked to agree with the following statement: ‘Employers should be able to genetically test employees to determine their risk of developing health problems that might affect their work or the safety of others’ (see table 12.6). Although there are slight differences between the questions asked in Australia and the United Kingdom, the differences in response are substantial and may warrant further research. Perhaps Australian Table 12.5 Genetic testing current and prospective employees, 2003, per cent Awareness of use in employment
Definitely used
Total
Respondents with good genetic awareness
4
3
Probably used
25
24
Probably not used
37
41
Definitely not used
21
23
Can’t choose
13
9
(n = 2016) SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
214 •
8:55 AM
Page 214
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 12.6 Employers should be able to genetically test employees, 2003, per cent Attitude towards use in employment
Total
Respondents who believe that tests are definitely or probably used by employers
Strongly agree
2
2
Agree
9
14
Neither agree nor disagree
12
10
Disagree
37
36
Strongly disagree
38
39
3
1
Can’t choose (n = 2041)
SOURCE The Australian Survey of Social Attitudes 2003
workplaces have something to do with it. Indeed, a logistic regression model predicting disagreement with employment testing shows that respondents who have poor relations in their workplaces are more likely to disagree with testing than those with good relations (see Appendix, table A12.3). We again find women and younger respondents (aged 18–34 and 35–49 years) are more likely to disagree.
G E N E T I C I N F O R M AT I O N A N D IMMIGRANTS TO AUSTRALIA Governments can use genetic technologies for immigration purposes. They can screen the health of prospective immigrants and test the genuineness of applications for family reunion. About 200 genetic tests per year are conducted by DIMIA in family reunion cases (ALRC 2003; DIMIA 1994–2002). However, at present, genetic testing is rarely used for assessing compliance with the Migration Regulations because DIMIA judges most tests to be insufficiently accurate (ALRC 2001). The Regulations include the requirement that potential immigrants not have a disease or condition such that they would be likely to impact on health care or community services (ALRC 2001). Yet, the recent ALRC/AHEC Report predicts that, as genetic testing becomes cheaper and the range and accuracy of tests improve, their use to determine the health status of applicants may become more frequent (ALRC 2003). AuSSA 2003 asks respondents how much they agree with the following statement: ‘the government should be able to genetically test immigration applicants to determine their risk of developing health problems’. We find that 37 per cent agree, 11 per cent neither agree nor disagree, and 49 per cent disagree (see table 12.7). We can certainly see
AusSocAtts03
26/7/05
8:55 AM
Page 215
KNOWING YOUR GENES
• 215
Table 12.7 Government should be able to genetically test immigration applicants, 2003, per cent Strongly agree
11
Agree
26
Neither agree nor disagree
11
Disagree
26
Strongly disagree
23
Can’t choose
3
(n = 2044) SOURCE The Australian Survey of Social Attitudes 2003
that Australians are more attracted to testing immigration applicants than they are to allowing the use of genetic information by employers and insurers, situations with personal implications. Indeed, 54 per cent of those who agree with immigration use disagree with both the two other uses. When we turn to the results of a logistic regression model predicting disagreement with testing in immigration (see Appendix, table A12.4), we note the important role played by views about immigration. Those respondents who prefer lower immigration are much less likely to be opposed to testing. For them, testing may be seen as one way of deterring immigrants. On the other hand, we find again that women, younger respondents and those with university education or a managerial or professional employment background are all more likely to oppose testing.
CONCLUSION Overall, we find some recognition of genetic technologies and their applications. About 45 per cent of AuSSA 2003 respondents link genetics to testing for inherited problems, which indicates a good understanding of the main focus of current applications. The ‘genetically aware’ on this score are more likely to give accurate responses about current uses of genetic testing in the workplace, but are no more likely to give accurate responses about insurance use. This demonstrates that, while community knowledge about genetic technologies may be adequate, the same cannot be said for their applications. Clearly, the community needs to be well informed about the impact of the rapidly expanding genetic technologies, so that their benefits are harnessed and their risks minimised (ALRC 2003). Making informed choices about using or undertaking genetic tests requires that everyone, including third parties, understands their implications and trusts their use. For example, research shows that only 7 per
AusSocAtts03
26/7/05
216 •
8:55 AM
Page 216
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
cent of those individuals attending genetic counselling to discuss tests for the risk of future health problems raise questions about insurance. This compares to 97 per cent of genetic counsellors who raise this problem (Taylor et al. 2004), a finding that underlines the importance in the genetic testing context of counselling and education. The larger role of education is affirmed in the final report from the ALRC/AHEC Inquiry, which proposes the establishment of a human genetics commission to ensure this happens (ALRC 2003). Overall, we find solid support in the Australian community for genetic testing because it gives new hope of detecting, treating and even reversing serious medical conditions. But this does not necessarily hold up for specific applications of this new technology. Women and respondents with low trust in others are more likely to express opposition. The importance of trust appears to be critical in explaining attitudes, and may be key to regulation. The 2001 Eurobarometer survey of the European Union (Gaskell, Allum & Stares 2003) rightly notes that genetic information is not merely of medical importance, but of personal and social interest, and that the boundaries between the clinic and society and between the patient and the citizen are becoming blurred. All of these relationships deal with trust. Among AuSSA 2003 respondents, there is strong opposition to the use of genetic information in key institutional environments – insurance and the workplace. In the latter, this is true even when tests are designated for occupational health and safety purposes. Indeed, 75 per cent of those who generally support genetic testing for health benefits go on to disagree with its use to assess the health of current or prospective employees. We find that low trust between management and employees may be part of the reason for this opposition. The same concerns about employment testing are found internationally. A survey conducted by the Center for Genetics & Public Policy (2004) in the United States suggests that American opposition to the use of genetic test results in this area has increased since 2002. According to these findings, more than 90 per cent of Americans now oppose employer access to test results, and eight in ten oppose such access by insurers. In contrast, the Australian community is more comfortable with using genetic tests for immigration purposes, presumably because there are fewer personal implications and because some respondents seek lower immigration. There is no international comparative data on this application of genetic technology. Certainly, in all of the three cases examined here – insurance, employment and immigration – Australians need more information to make
AusSocAtts03
26/7/05
8:55 AM
Page 217
KNOWING YOUR GENES
• 217
informed decisions about their own interests and about those of the community at large. Clearly, a body like a new human genetics commission, as proposed by the ALRC, could take up this important role, monitoring developments, educating Australians and ensuring that the law and practice involved in genetic testing balances freedoms, burdens and power. One important problem for any such commission would be to address low trust in financial institutions and in the workplace. When the draft Human Genome Map was announced in June 2000, there was much hyperbole. President Bill Clinton, for example, said that ‘we have deciphered the Book of Life’. Yet science is really only at the beginning of the ‘Book’. And as genetic technologies become more sophisticated, balancing the freedoms, burdens and power that these technologies bring will also become even more difficult. Public policies must address these issues if Australians are to maximise the benefits of the ‘New Genetics Revolution’, minimise the potential harms and limit the misuse of power. NOTES
1
2 3 4
5 6
DNA (deoxyribonucleic acid) is the chemical that makes up the genes in the cells of our body and, through coded messages sent to the cell, instructs our development and growth and our physical, intellectual and mental health directly or in conjunction with environmental factors. Haemochromatosis is a condition in which iron accumulates in the body, damaging organs. It is also known as ‘iron overload syndrome’. Symptoms are not usually seen until middle age. Huntington’s disease is a neurological degenerative condition that leads to memory loss and difficulties with swallowing, speaking and walking. Symptoms are not usually seen until middle age. The study of public attitudes to human genetic information for the UK Human Genetics Commission (Park et al. 2001) was used to guide the development of AuSSA 2003 questions asked about human genetic information. Analysis of the data was modelled on the British Social Attitudes Survey (Stratford, Marteau & Richards 2001) and guided by findings from the Eurobarometer survey (Gaskell, Allum & Stares 2003) and the Australian Centre for Emerging Technologies and Society report (2003). Three per cent of the total respondents to the questionnaire did not answer these questions. Data about the experiences of selected individuals in the Australian community who have genetic information is currently being collected through the Australian Research Council-funded Genetic Discrimination Project (see Otlowski, Taylor & Barlow-Stewart 2002). But more research is needed on broader community views.
The authors thank Dr Susan Treloar, Dr Mark Stranger, Ms Mona Saleh and Mr Ian Muchamore for their contributions to this study.
AusSocAtts03
26/7/05
218 •
8:55 AM
Page 218
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
REFERENCES
Australian Centre for Emerging Technologies and Society, Swinburne University of Technology (2003) The Swinburne National Technology and Society Monitor. Available at: (accessed May 2004). Australian Government (2003) Mapping Australian Science and Innovation: Main Report, section 3.2.1.10, Public Attitudes to Science. Available at: (accessed May 2004). Australian Law Reform Commission (ALRC) (2001) Protection of Human Genetic Information, Issues Paper 26, chapter 11, Insurance, ALRC 96, Canberra. p. 318. Available at: (accessed May 2004). —— (2003) Essentially Yours: The Protection of Human Genetic Information in Australia, ALRC 96, Canberra. Available at: (accessed May 2004). Barlow-Stewart, K (2003) Genetics Information and Life Insurance Products in Australia, The Centre for Genetics Education, NSW, Australia. Available at: (accessed May 2004). Biotechnology Australia (2003) Biotechnology Public Awareness Survey: Final Report. Available at: (accessed May 2004). Bureau of Industry Economics (1995) Science Understanding and Awareness: How Australia Ranks in International Surveys, Occasional Paper 30, Canberra, Australian Government Publishing Service. Center for Genetics & Public Policy (2004) ‘Statement of Kathy Hudson’, Employer–Employee Relations Subcommittee of House Education and the Workforce Committee hearing on ‘Genetic Non-Discrimination: Examining the Implications for Workers and Employers’, 22 July. Available at: (accessed May 2004). Department of Immigration and Multicultural and Indigenous Affairs (DIMIA) (1994–2002) Procedures Advice Manual (PAM3), Div 1.2 r 1.12, Member of the Family Unit, DIMIA, Canberra. Gaskell, G, Allum, N & Stares, S (2003) Europeans and Biotechnology in 2002, Eurobarometer 58.0 (2nd edn). Available at: (accessed May 2004). Gibson, R, Wilson, S, Meagher, G, Denemark, D & Western, M (2004) Australian Survey of Social Attitudes 2003, Data file, Australian Social Science Data Archive, The Australian National University, Canberra. Human Genetics Society of Australasia (HGSA) (2001) Genetic Testing and Insurance Policy. Available at: (accessed May 2004). Investment and Financial Services Association (IFSA) (2002) Life Insurance and Genetic Testing in Australia. Available at: (accessed May 2004). Newspoll (2001) Science and Technology Attitudes 2001, Study commissioned by the National Innovation Awareness Strategy, Department of Industry, Science and Resources, Canberra. Otlowski, MF, Taylor, SD & Barlow-Stewart, KK (2002) ‘Australian empirical study into genetic discrimination’, Genetics in Medicine, 4: 392–95. Park, A, Curtice, J, Thomson, K, Jarvis, L & Bromley, C (2001) ‘Public attitudes to human genetic information’, People’s Panel Quantitative Study Conducted for the Human Genetics Commission by MORI. Available at:
AusSocAtts03
26/7/05
8:55 AM
Page 219
KNOWING YOUR GENES
• 219
(accessed May 2004). Research Australia (2003) Health and Medical Research. Public Opinion Poll. Available at: (accessed May 2004). Sample, I (2004) ‘Why limits must be set on the use of science’, The Guardian, London, reported in Sydney Morning Herald, 17 May 2004. Stratford, N, Marteau, T & Richards, M (2001). ‘Genetic research: Friend or foe?’ British Social Attitudes: The 18th Report – Public Policy, Social Ties, London, pp. 103–30. Taylor, S, Barlow-Stewart, K, Manning, D, Treloar, S, Stranger, M, & Otlowski, M (2004) Genetic Discrimination in Australia: The Perceptions of Genetic Counselors, 27th Annual Meeting of the Human Genetics Society of Australia, Fremantle. Virginia Commonwealth University (VCU) Centre for Public Policy (2001) ‘Americans welcome scientific advancements with caution’, VCU Life Sciences Survey, Richmond. Available at: (accessed May 2004).
AusSocAtts03
26/7/05
8:55 AM
Page 220
CHAPTER 13
MASS MEDIA AND MEDIA POWER IN AUSTRALIA David Denemark
INTRODUCTION Australians in the 21st century live in a complex society in which the mass media plays a vital and changing role. The mass media, as it has always done, connects individuals with the larger world – providing all of us, as Walter Lippmann proposed long ago, with the ‘pictures in our heads’ that are the only way we can truly perceive a global reality (Lippmann 1922). Never has this mediating role been more important than in advanced democracies such as Australia, where the media now represents the primary window on the world of public events and politics (see Ansolabehere, Behr & Iyengar 1993, p. 2; and Ward 2002, p. 401). Australians live amid the contradictory and controversial functions performed by the media. Televised ‘news as entertainment’ (Marshall & Kingsbury 1996, p. 6) is an obvious example. Driven by advertising profits, commercial television combines sensational stories and images with news and information. In so doing, it blurs the distinction between entertainment and the media’s more serious watchdog role to provide citizens with news and information that keeps government and politicians accountable (Cook 1998). While most Australians say that television is now their main source of news and think that it is the most trustworthy news source (Craig 2004, pp. 94, 15), they are also increasingly uneasy about most of the country’s media being privately owned and controlled by a few powerful, profit-seeking families. Their concerns have basis in fact: Australia has the most concentrated pattern of media ownership of any western country – making it, some argue, akin to the ‘press domination found in Central American banana republics’ (Windshuttle 1988, pp. 86, 40).
AusSocAtts03
26/7/05
8:55 AM
Page 221
MASS MEDIA AND MEDIA POWER IN AUSTRALIA
• 221
Five questions help to frame this chapter about the role and impact of the mass media in Australia: (i) does the mass media dominate Australians’ lives?; (ii) does violence in television encourage violence in Australian society?; (iii) is mass media ownership too concentrated in the hands of a few individuals?; (iv) does the media play an effective watchdog role in keeping governments honest and on their toes?; and (v) is the media’s influence too powerful in Australian society? Using the Australian Survey of Social Attitudes 2003 (Gibson et al. 2004), we explore who relies on which media for their news and information and for their home entertainment. We also look at respondent attitudes to the media they use – their confidence in the media as well as their views of its power, ownership and roles. As we will see, while most respondents depend on the media in their daily lives, and trust it to play important political functions, they also have serious reservations about its power and concentration. These paradoxes of media power – especially the ability for its power to be used for both harm and good – are central to our discussion.
USING THE MEDIA FOR NEWS A N D E N T E R TA I N M E N T Australians rely on a variety of different media for their news and information: television, radio, newspapers, magazines and, increasingly in the last decade, the Internet. Despite the concentration of media ownership in Australia, these media provide diverse points of access to the world. They differ in their detail and their graphic power, in their accessibility to the public, and in their demands for active observation. They also differ in their dependence on advertising revenue for their viability. Australia’s mainstream media are primarily commercial outlets, which fuse advertising with information and entertainment. They include the capital city newspapers – with two large corporations controlling 88 per cent of the market (Ward 1995, p. 126) – as well as three commercial television networks and the capital city radio stations, equally dominated by a few corporate owners (Craig 2004, p. 11). Australia’s mainstream media also include publicly funded sources, such as the ABC and SBS television and radio, which have limited advertising content and reflect, some contend, substantial control by the national government (Windshuttle 1984, p. 141; also see Marshall & Kingsbury 1996, p. 109). All told, then, Australia’s diverse population can select from a wide range of mass media alternatives in its pursuit of news, information and entertainment, even though most outlets are (except for publicly funded
AusSocAtts03
26/7/05
222 •
8:55 AM
Page 222
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 13.1 Media use for news and information, 2003, per cent Daily
Several times Once a a week week or less
Never
Commercial television
65
20
9
4
ABC and/or SBS television
41
28
21
10
Radio
63
21
12
5
Newspapers
40
26
30
5
Internet sites
11
16
26
47
Talkback radio
14
12
18
57
4
8
40
48
News magazines (n ≥3860) SOURCE The Australian Survey of Social Attitudes 2003
television and radio) owned and controlled by a few large corporations. But, where in fact do we get our information about the news and events in the world around us? Table 13.1 presents respondent use of media for their news and information: commercial television, ABC and/or SBS television, radio, newspapers, Internet sites, talkback radio and news magazines. Leading the way in daily usage is commercial television – 65 per cent of respondents say they watch commercial television daily as a source for news and information. Clearly, the image-centred power of television news exceeds the detailed and broader coverage of newspapers, which only 40 per cent of respondents consume daily. These patterns confirm the argument that television in Australia has ‘long usurped newspapers as the chief source of news and information for most Australians’ (Ward 1995, p. 2). But the supremacy of television news is most sharply seen in that more Australians rely on the publicly funded ABC and SBS television news (41 per cent) than on newspapers. Close behind commercial television is radio with 63 per cent of respondents saying they listen to it as a daily news and information source. Though radio primarily offers listeners a ‘headline service’ (Tiffen 1989, p. 26; also see Ward 1995, p. 2), it remains an important medium for Australians – nearly as many use radio as television as their daily source of news and information. Much further down on the scale is talkback radio, used by only 14 per cent of Australians daily for news and information. While it has been argued that talkback radio is ‘the greatest source of political influence emanating from the radio today’, (Craig 2004, p. 102), these figures serve as a reminder that talkback radio is used by a minority of Australians.
AusSocAtts03
26/7/05
8:55 AM
Page 223
MASS MEDIA AND MEDIA POWER IN AUSTRALIA
• 223
The Internet in Australia is a source of news and information for only a small portion of the population – 11 per cent use the World Wide Web on a daily basis, about the same as in the United States and Britain (Davis 1999; Curran & Seaton 2003). Clearly not yet a universal medium, the Internet may well appeal to Australians, like Americans, who are already highly informed and seek additional political and other information (see Farnsworth & Lichter 2003, p. 193). It is also still largely limited to those who can afford a computer at home or who have online access as part of their jobs. The least used source of daily news and information is news magazines – only 4 per cent of respondents say they read a news magazine daily, and nearly half (48 per cent) say they ‘never’ use magazines for news and information. All told, the figures in table 13.1 point to the ‘news dominance’ of the electronic media. Australians’ reliance on the electronic media is further evident in table 13.2, which shows those media sources respondents rely on most for their news and information.
Table 13.2 Source most relied on for news and information, 2003, per cent ABC SBS television
Radio
Newspapers
34
22
18
19
4
1
0
1
40
20
19
15
2
2
1
2
44
13
18
14
6
1
0
4
41
20
19
13
3
2
0
2
34
24
18
19
2
2
0
1
30
27
18
20
1
2
0
1
41
20
18
16
2
2
0
2
23
28
21
20
5
1
1
1
37
21
19
17
3
2
0
2
Commercial
television
Sex Male
and/or
Internet Talkback News Friends & sites radio magazines family
(n = 1952)
Female (n = 2216)
Age 18–34 (n = 721)
35–49 (n = 1266)
50–64 (n = 1272)
65 and over (n = 898)
University degree No (n = 3182)
Yes (n = 912)
Total (n = 4212)
SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
224 •
8:55 AM
Page 224
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Here again, we see that commercial television is cited by 37 per cent of respondents as the source they rely on most for their news and information – far exceeding the ABC and/or SBS television (21 per cent) and radio (19 per cent). While the top three sources are all electronic broadcast media, commercial television clearly rules the roost. Newspapers have become poor cousins in Australia’s news and information stakes – only 17 per cent of AuSSA 2003 respondents cite this medium as the one they rely on most. If there was a time in a previous generation when metropolitan dailies dominated Australia’s news landscape (see Windshuttle 1984, p. 33), these figures confirm that that era has gone, to be replaced by broadcasting. And, indeed, the reason for the decline of Australia’s newspapers across the last two or three decades is closely linked to the rise of the electronic media, and their ability to provide Australians with cheap, readily available news and entertainment (Windshuttle 1984, p. 37). As for the remaining media, the figures in table 13.2 show that only tiny percentages of Australians cite the other few sources of news and information as important on a daily basis: Internet sites (3 per cent), talkback radio (2 per cent), and news magazines (less than 1 per cent). Clearly, while many Australians tune in to talkback radio, surf the Internet, or leaf through news magazines, these three media do not figure importantly as major sources of news and information. Again, the figures in table 13.2 suggest that Australians turn to, and depend on, the electronic mass media, as their window on the world. Table 13.2 also breaks down reliance on media for news and information by key demographic distinctions: sex, education and age. We see that 40 per cent of women rely most on commercial television for their news and information, while only 34 per cent of men do. At the same time, men are 4 per cent more likely than women to rely on newspapers. Women’s higher reliance on commercial television is likely a reflection of their being less likely to have a university education, and to be full-time members of the workforce. In the second section of table 13.2, we can see that university education delineates respondent use of different media for news and information. Those with a university degree are 18 per cent less likely than those without one to cite commercial television as their most important source for news. At the same time, university-educated respondents are 8 per cent more likely to rely on the ABC or SBS television for their news and information, 4 per cent more likely to rely on newspapers, and 3 per cent more likely to rely most on the Internet for news and information.
AusSocAtts03
26/7/05
8:55 AM
Page 225
MASS MEDIA AND MEDIA POWER IN AUSTRALIA
• 225
Age also highlights distinct patterns. Younger respondents are more likely than older respondents to rely on commercial television, while they are less likely to rely on public television – the ABC and SBS. At the same time, we see that younger respondents are less likely to use newspapers for their news and information, but more likely to rely on the Internet – an age-specific pattern we investigate in more detail below. And, finally, in the last column of table 13.2, we see that young respondents are the most likely to turn to friends and family for their news and information.
D O E S T H E I N T E R N E T M A K E I N F O R M AT I O N AVAILABLE TO ALL? The rise of the Internet has prompted intense speculation about its potential for expanding individual access to information (Norris 2000, p. 120). The Internet, which came into existence as the World Wide Web in 1992, is a highly interactive medium – the individual seeks out websites because of their interest – and it also enables the individual both to select and format the information they desire and, in many cases, to respond to and communicate with information providers. The Internet, then, offers individuals ‘less mediated information’ than the other mass media (Farnsworth & Lichter 2003, p. 10). But, perhaps more importantly, it provides a two-way communication alternative to the broadcast media’s and newspapers’ one-way flow. One problem with pronouncements about the ‘Internet revolution’ is that, at least at the present time, the Internet is not available to everyone. In the year 2000, only 6 per cent of the world’s population used the Internet, while 66 per cent of that use was within North America and Western Europe, and only 5 per cent in the whole of Latin America and Africa (Curran & Seaton 2003, p. 260; also see Norris 2000, p. 123). At the same time, within advanced democracies, Internet use is concentrated among affluent communities, while its interactive capacities mainly benefit young, educated and informed individuals seeking to expand their knowledge (see Farnsworth & Lichter, 2003 p. 193; Curran & Seaton 2003, pp. 193, 280; Norris 2000, p. 129). The question remains, though: how do Australians use and think about the Internet? Who has access to the Internet in this country, and is it viewed as a means by which information can be made available to more people worldwide? Overall, 56 per cent of respondents report that they have access to the Internet at home – a figure that puts Australia ahead of the United Kingdom, where 42 per cent of homes in 2002 had access to the World Wide Web (Curran & Seaton 2003, p. 279).
AusSocAtts03
26/7/05
226 •
8:55 AM
Page 226
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Additionally, 35 per cent of AuSSA respondents say they could access the Internet at work, and 10 per cent say they could use the Internet in other public places. At the same time, however, fully 30 per cent of respondents say they have no access to the Internet. If European and British research is any guide, patterns of Australian access to the Internet should depend importantly on the individual’s socioeconomic background (see Norris 2000, p. 129; and Curran & Seaton, 2003 pp. 279–80). Most important is income – affluence allows families to buy computers and the connections necessary for access to the Internet in the home. And, indeed, respondent access to the Internet is similar to that of British and European citizens. Only 7 per cent of respondents in households with incomes over $78 000 have no access to the Internet, while a majority (53 per cent) of households with incomes less than $31 199 have no access. Those in the middle income bracket split the difference: 21 per cent say they have no access at all to the Internet. As income and education tend to go hand in hand in most western countries, it is no surprise that Internet use by university education parallels that for income. Only 7 per cent of those with a Bachelor’s degree or above say they have no access to the Internet, while five times that number – 36 per cent – of those with no university degree say they have no access. Another important factor affecting respondent access to the Internet is age. For those Australians aged 65 and older, two-thirds have no access to the Internet. But, for those under 64 years of age, the vast majority have access to the Internet, while for those under 34 years of age fewer than 12 per cent say they have no access. Clearly, a new generation of Australians is growing up with near-universal access to the Internet, while older generations retire, unconnected to the world through the Internet. While income will continue to be an important factor in determining access to the Internet, perhaps the sharpest disparities seem destined to fade with time, as school children raised with the World Wide Web bring greater computer skills into their adult, working lives.
T E L E V I S I O N A S E N T E R TA I N M E N T As we have seen so far in this chapter, television – especially commercial television – is relied on more than any other medium for news and information. But does television also dominate as a source of entertainment in Australian lives? And what do people think about television violence? This section takes up these two issues as a way to examine the positive and negative impact of the media in the realm of entertainment.
AusSocAtts03
26/7/05
8:55 AM
Page 227
MASS MEDIA AND MEDIA POWER IN AUSTRALIA
• 227
Television came to Australia in 1956 and had been embraced almost universally by Australians only a decade later – 91 per cent of households own a television set by 1965 (Given 2003, pp. 46–47). By 1981, over 97 per cent of households owned at least one television set, and television programming had begun to reach essentially ‘the entire Australian population’ (Windshuttle 1984, p. 59). Today, in nearly every Australian home, television viewing is a significant part of daily life. Australians, on average, watch over three hours of television a day – or approximately 24 hours a week (Given 2003, p. 192). Table 13.3 reports the number of respondents for whom television is the main source of household entertainment. It also breaks down these results for three main demographics – sex, university education and age – and for respondents grouped by their primary source of news and information. Overall, we can see that 67 per cent agree that television is their main form of entertainment. We find little difference between the sexes in their reliance on television, but respondents with a university degree are less likely than those without a university education to nominate television as their most important household entertainment (57 per cent versus 70 per cent). Thus, echoing the patterns in table 13.2 above, university-educated Australians appear more likely to rely on public, not commercial television, for their news and information and are less likely than those with fewer qualifications to view television as the primary form of home entertainment. Age does not have an important effect in distinguishing respondents. While those who are 65 years of age or older are the most likely to regard television entertainment as most important, the percentages of every other group hover around two-thirds. Television increasingly blurs the line between news and entertainment (Marshall & Kingsbury 1996, p. 6), with Australian commercial broadcasters since the 1980s using the evening news to entice an audience for subsequent programs (Tiffen 1989, p. 26). While even the ABC and SBS have had to compete for television ratings to maintain their share of the viewer market – lest government funding be cut for failing to attract viewers (Ward 1995, pp. 143–44) – respondents who use non-commercial media for their news and information are clearly less reliant on television for their home entertainment. And yet, the figures show that a majority of respondents tell us that television is their most important source of home entertainment, no matter which medium they use for their news and information. Still, those who rely on commercial television for news and information are clearly most likely to view television as their main source of entertainment (76 per cent), while the
AusSocAtts03
26/7/05
228 •
8:55 AM
Page 228
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 13.3 Television is the main source of household entertainment, 2003, per cent Sex Male (n = 1920) Female (n = 2183) Age 18–34 (n = 718) 35–49 (n = 1257) 50–64 (n = 1250) 65 and over (n = 868) University degree No (n = 3149) Yes (n = 908) Source of news and information Commercial television (n = 1531) ABC and/or SBS television (n = 877) Radio (n = 761) Newspapers (n = 702) Internet sites (n = 104) Talkback radio (n = 81) News magazines (n = 17) Friends and family (n = 64) Total (n = 4146)
Agree
Disagree or neither
69
31
66
34
67
33
64
36
67
33
72
28
70
30
57
43
76
24
61
39
60
40
65
35
51
49
67
33
71
30
61
39
67
33
SOURCE The Australian Survey of Social Attitudes 2003
small number of respondents who rely on Internet sites for news are the least likely to view television as their main form of entertainment (51 per cent). Those who cite other media as their main source of news and information range between 60 per cent and 71 per cent. Overall, although respondents use a variety of media for news and information, the majority still rely on television in their leisure time at home. Television, in short, dominates Australians’ lives at home.
AusSocAtts03
26/7/05
8:55 AM
Page 229
MASS MEDIA AND MEDIA POWER IN AUSTRALIA
• 229
DOES TELEVISION VIOLENCE MAKE AUSTRALIA MORE VIOLENT? A significant part of television entertainment portrays violence. From children’s cartoons to prime-time crime shows and movies, violence permeates television programming, especially on commercial networks. In America, research has shown that approximately 70 per cent of prime-time programs and 90 per cent of children’s programs contain some violence – with, on average, five violent actions per hour in the former, and 20 violent actions in the latter. American public broadcasting has the smallest proportion of programs with violent content (about 20 per cent). But, all told, the average American watches 10 000 murders, rapes and aggravated assaults per year on television. Not surprisingly, 75 per cent of Americans believe that televised violence contributes to violence in society (Strasburger & Wilson 2002, pp. 74–76). While social research suggests that actual violence is not caused purely by television violence (see Van Evra 1990, p. 81), many television viewers worry about the ceaseless violence finding its way into their homes through the airwaves. With the large number of American programs on Australian television, and violent programs of its own, Australians are likely to have comparable concerns. Table 13.4 reports the proportion of those surveyed who believe that television violence encourages violence in Australian society – the percentages grouped, as in previous tables, by respondents’ sex, education, age and their main source of news and information. With television content becoming increasingly violent across the last three decades (see Van Evra 1990, p. 80) and with violent programming routinely geared to men, attitudes to violence in television differ for viewers, depending on their age and their sex. We can see that, while majorities of both men and women agree that television violence encourages social violence, women are more likely than men to agree (75 per cent versus 67 per cent). At the same time, university education has little effect on these sentiments: those with a university degree are just 2 per cent less likely than those without to feel that television violence causes social violence. Age most sharply delineates attitudes about television violence – older respondents are more likely to feel that television violence causes social violence (85 per cent, compared to 50 per cent of 18–34 year olds). While the young are just as likely as other age groups to rely on television for news (see table 13.2) and for household entertainment (see table 13.3), they seem least worried about any possible negative impact of television violence on society.
AusSocAtts03
26/7/05
230 •
8:55 AM
Page 230
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table 13.4 Television violence encourages social violence, 2003, per cent Agree
Disagree or neither
Male (n = 1914)
67
33
Female (n = 2181)
75
25
18–34 (n = 714)
50
50
35–49 (n = 1253)
68
32
50–64 (n = 1243)
75
25
65 and over (n = 875)
85
15
No (n = 3120)
71
29
Yes (n = 907)
69
31
Commercial television (n = 1525)
71
29
ABC and/or SBS television (n = 879)
74
26
Radio (n = 763)
70
30
Newspapers (n = 698)
71
29
Internet sites (n = 102)
48
52
Talkback radio (n = 81)
81
19
News magazines (n = 17)
77
23
Friends and family (n = 65)
60
40
Total (n = 4139)
71
29
Sex
Age
University degree
Source of news and information
SOURCE The Australian Survey of Social Attitudes 2003
The last section of table 13.4 reports the percentage of respondents who agree with the statement that television violence is a cause of social violence, grouped by respondents’ main source of news and information. Respondents who cite the Internet as their main information source are
AusSocAtts03
26/7/05
8:55 AM
Page 231
MASS MEDIA AND MEDIA POWER IN AUSTRALIA
• 231
the only group for which a majority did not agree (48 per cent). Those who rely on friends and family are the group with the next lowest level of agreement (60 per cent). Both groups, as we saw in table 13.2, have disproportionate numbers of young respondents – confirming strong age distinctions. For the remaining groups of respondents, we see strong majority support for the statement that television violence encourages social violence, with strongest agreement among those who use news magazines for their news and information (77 per cent) and those who listen to talkback radio (81 per cent). Although a clear majority believe that television violence causes real violence, we cannot determine why respondents continue to watch violence-filled television programming if they feel this way. One likely answer is that they are dependent on television and its news, information and entertainment – good, bad or indifferent.
MEDIA POWER The idea of ‘media power’ has become ‘almost a commonplace of modern commentary’ (Street 2001, p. 231). As we have seen, the power of mass media results from our almost universal reliance on it for news, information and entertainment. But, in turn, newspapers, radio and television are able to demand massive sums from advertisers seeking to access that mass audience. This pursuit of commercial profit serves not only to cement a close relationship between mass media and the business community – it means that mass media is big business, with substantial economic and political leverage. The question, as we will see, is whether Australians feel that the media uses its power for public good by keeping a close and critical watch over government. In Australia, a handful of families dominate the ownership and control of newspapers, radio and television, making them among the most powerful economic actors in the nation’s politics and economy. Companies associated with the Packer family now own the Channel Nine television network and a large number of magazines, while Rupert Murdoch’s vast global empire, News Corporation, controls over twothirds of Australia’s newspapers. A third group, John Fairfax, owns several important metropolitan daily newspapers, which have not been absorbed into the Packer and Murdoch groups, if only because Australian law currently prevents further cross-media ownership (Ward 2002, pp. 405–06; for a detailed chronology of the rise of Australian media concentration, see Chadwick 1989). All told, then, Australia’s mass media is in ‘monopoly or semi-monopoly positions’ (Tiffen 2000,
AusSocAtts03
26/7/05
232 •
8:55 AM
Page 232
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
p. 180), and a focused source of economic and political power. A second aspect of media power sits uneasily with media concentration: the media’s role in keeping governmental power accountable to the people. Advanced democracies have long championed a strong, free press (and, more recently, mass media) as vital for sustaining a free society. Indeed, for centuries, an independent press has been seen as ‘necessary to check abuse of state power’ (Ward 1995, p. 8). Since democracies depend on an informed electorate keeping popular control of governmental power, the media plays a vital ‘watchdog role’ in providing people with information to keep government answerable to the masses (Ansolabehere, Behr & Iyengar 1993, p. 221). But, of course, a free press, actively playing a ‘watchdog’ role, rests on a key assumption: that the commercial media uses its power to provide a critical perspective on government. The problem with this is that big business and government depend on each other. In the end, despite a growing unease about media ownership concentration, Australians have come to rely on political information provided by privately owned media corporations. It is, as Tiffen reminds us, an undesirable trade-off: ‘We would not leave our political liberties to the noblesse oblige of the state, but we are asked to leave our most vital processes of political communication to the noblesse oblige of media corporations’ (Tiffen 2000, p. 180). The tables in this last section report attitudes on these issues. Australians have clear views about the concentration of media ownership. Overall, a total of 81 per cent either agree or strongly agree that media ownership in Australia is too concentrated among a few rich families. At the same time, 70 per cent of those surveyed agree that the media should have less power. Thus, while a strong majority of respondents believe that the media should have less power, this is a smaller percentage than those who feel that media ownership is too concentrated. It seems clear that, although many respondents look to the mass media as a means for keeping a watchful eye on government, they are wary of the watchdog itself. These sentiments are not different for men and women. The same sentiments are found for respondents with and without university education, although university-educated respondents are slightly more likely than those who are not university-educated to feel that the media should have less power (74 per cent versus 69 per cent). Once again, age sharply divides attitudes: younger respondents are least likely to agree that media ownership is too concentrated and
AusSocAtts03
26/7/05
8:55 AM
Page 233
MASS MEDIA AND MEDIA POWER IN AUSTRALIA
• 233
Table 13.5 Attitudes to media concentration and power, 2003, per cent Media ownership is too concentrated
Media should have less power
Agree
Disagree or neither
Agree
Disagree or neither
Male
82
18
70
30
Female
80
20
69
31
18–34
73
27
62
38
35–49
79
21
69
31
50–64
83
17
72
28
65 and over
87
13
74
26
Sex
Age
University degree No
81
19
69
31
Yes
82
18
74
26
Source of news and information Commercial television
79
21
66
34
ABC and/or SBS television
87
13
74
26
Radio
82
18
74
26
Newspapers
79
21
69
31
Internet sites
80
20
71
29
Talkback radio
77
23
66
34
News magazines
80
20
53
47
Friends and family
72
28
69
31
81
19
70
30
Total n
(4036)
(3984)
SOURCE The Australian Survey of Social Attitudes 2003
should have less power, while older respondents are most likely to agree. Thus, while we have seen that young respondents are least likely to feel that the media has a negative influence in causing social violence, they are also – as we will see in table 13.6 – least likely to feel that it plays a positive role in keeping a watchful eye on government. Not surprisingly, the more benign view of the media among young respondents coincides with a lower likelihood of calls for weakening its power. By contrast, older respondents, who are most likely to think that media causes violence, are also most likely to believe that the media plays a positive watchdog role, but also most believe that media concentration and power are too high. In short, respondents’ attitudes reflect the fact that media power brings with it both positive and negative contributions to society.
AusSocAtts03
26/7/05
234 •
8:55 AM
Page 234
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
As mentioned above, the Australian mass media includes noncommercial media – publicly funded television and radio (ABC and SBS) – and the Internet. If Australians are wary of the commercial media’s concentration, we might expect significant differences in views of the media for those who turn to non-commercial sources for news and information. The last section of table 13.5 reports on those who agree that Australia’s mass media is too concentrated and should have less power, grouped by respondents’ main source of news and information. We find that those who rely most on the ABC and SBS television for news are also most likely to see the media as too concentrated (87 per cent), and equal most likely to feel that the media should have less power (74 per cent). At the same time, those who rely most on commercial television news and information are about as likely as other media users to think that media ownership is too concentrated, but they are least likely to see the media as too powerful. Nonetheless, overall, two sharp patterns stand out here: no matter which medium Australians use for their news, a vast majority feel that the country’s media ownership is too concentrated, and a majority believe that the media is too powerful. AuSSA 2003 also asks respondents about two aspects of the media’s watchdog role. The first question taps respondent attitudes about the media as a critic of governmental power – a thorn in its side: ‘Is the mass media effective in keeping governments on their toes?’ The second question taps respondent attitudes about the media’s ability to keep government true to its principles: ‘Is the mass media effective in keeping governments honest?’ Though related, the first role for the media is less demanding, requiring only that the media ask probing questions of the government. The second role asks whether the media can deliver an important political result by keeping governments honest. Of interest to us is whether respondents – many of whom think that media ownership is too concentrated and the level of media power is too high – believe that the media uses its power in a positive way, as a watchdog over Australia’s governments. Table 13.6 presents responses to these two statements. We find that respondents are far more likely to agree that the media is effective in keeping government on its toes (60 per cent) than to agree that it is an effective mechanism for keeping governments honest (38 per cent). These perceptions are not substantially different for men and women. At the same time, those with a university degree (58 per cent) are only slightly less likely than those without a degree (60 per cent) to feel that the media keeps governments on their toes, but are more sceptical about
AusSocAtts03
26/7/05
8:55 AM
Page 235
MASS MEDIA AND MEDIA POWER IN AUSTRALIA
• 235
Table 13.6 Attitudes to media and government, 2003, per cent Media keeps governments on toes
Media keeps governments honest
Agree
Disagree or neither
Agree
Disagree or neither
Male
60
40
40
60
Female
59
41
36
64
18–34
55
45
25
75
35–49
56
44
38
62
50–64
62
38
44
56
65 and over
65
35
40
60
Sex
Age
University degree No
60
40
39
61
Yes
58
42
33
67
Source of news and information Commercial television
59
41
37
63
ABC and/or SBS television
58
42
37
63
Radio
59
41
37
63
Newspapers
65
35
41
59
Internet sites
48
52
23
77
Talkback radio
63
37
54
46
News magazines
78
22
50
50
Friends and family
30
70
36
64
60
40
38
62
Total n
(1990)
(2085)
SOURCE The Australian Survey of Social Attitudes 2003
the media’s ability to keep governments honest (33 per cent versus 39 per cent). While each age group feels that the media is more able to keep governments on their toes than to keep them honest, younger respondents are more doubtful that the media can keep governments honest. Because each of the two media watchdog questions was asked of only one-half of the pool of respondents, the numbers of respondents in several categories for ‘source of news and information’ in table 13.6 are smaller, making inferences less reliable than previous results. We can see, nonetheless, that, except for those who rely on the Internet – most widely used by the young – and those relying on friends and family for news and information, a strong majority feel that the media keeps governments on their toes. At the same time, strong majorities of respondents disagree that
AusSocAtts03
26/7/05
236 •
8:55 AM
Page 236
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
the media is effective in keeping governments honest, except for those reliant on talkback radio and news magazines. So we see that, while the media is acknowledged as a thorn in politicians’ sides, it is less clear that Australians trust the media to keep governments honest. Table 13.7 explores the links between Australians’ attitudes about media power and their perceptions of the media’s effectiveness in keeping a watchful eye on government, thus allowing us to see how respondents understand any trade-off between the media’s negative and positive influences. In the top half of the table, we see that respondents’ attitudes about media ownership concentration do not sharply separate their views about the media’s effectiveness in keeping governments on their toes or keeping governments honest. Roughly three-fifths of respondents agree that the media keeps governments on their toes, whether or not they feel that media ownership is too concentrated. At the same time, just above one-third of both groups believe that the media keeps governments honest. So, regardless of what respondents think about media concentration, a strong majority feel that the media is a thorn in the side of government – though this does not extend as far as believing that media can keep government to task. In the lower half of table 13.7, we see that attitudes to the level of media power have a much more important impact on attitudes about the media’s watchdog role than do attitudes about ownership. Those who agree that the media should have less power are 10 per cent less likely than those who disagree to believe that the media keeps governments on their toes. The same group who agree that the media should have less power are 17 per cent less likely than those who disagree to feel that that the media keeps governments honest. Table 13.7 Does a powerful media make a better watchdog? 2003, per cent. Media keeps governments on toes
Media keeps governments honest
Agree
Disagree or neither
Agree
Disagree or neither
Agree
60
40
38
62
Disagree
58
42
34
66
Agree
57
44
33
67
Disagree
67
33
50
50
Media ownership too concentrated
Media should have less power
n SOURCE The Australian Survey of Social Attitudes 2003
(1990)
(2085)
AusSocAtts03
26/7/05
8:55 AM
Page 237
MASS MEDIA AND MEDIA POWER IN AUSTRALIA
• 237
Clearly, Australian attitudes to the media range from supportive to cynical. Those who believe that the media plays an important watchdog role are least likely to feel that it should shed its power, while those who think that the media is not a watchdog want it to have less power. One explanation for this is that respondents may be distrustful of power in government and in the media – some see both as institutions with too much power. If so, they may be inclined to want to limit the power of both institutions, perhaps because they are seen as linked. Another explanation is that the media is an ineffective watchdog because respondents believe that government is too strong to be seriously checked by the media or that the media’s information role to the public is ineffective in curbing governmental abuses of power. Without additional survey items, we cannot be certain what drives widespread cynicism about the media’s ability to check government. However, as we have seen, nearly four-fifths of respondents feel that the media is too concentrated among a few rich families, while around six in ten believe that the federal government is either run ‘entirely’ or ‘mostly’ for the benefit of a few big interests, and not run for the benefit of all. It seems likely, then, that many respondents have concluded that to expect media barons to guard the government henhouse may be placing too much confidence in the fox.
CONCLUSION Attitudes to the mass media in this country seem to reflect something of a love–hate relationship. On the one hand, we have seen strong indications that the majority of AuSSA 2003 respondents depend on the media, especially the commercial media, for their news and information. Commercial television emerges the overwhelming leader in the news and information stakes. Similarly, we have seen that television is the main source of entertainment in the majority of respondent households. Only one in five say that television is not the main form of entertainment at home. However, within this reliance on commercial electronic media for news, information and entertainment, we find signs of distrust and unease with the media. A strong majority believe that television violence – whether television was relied on for home entertainment or not – encourages social violence. These feelings, though especially strongly held by women, are held by a strong majority of both men and women and by all but the youngest age group. We have also seen that, though a majority of respondents feel that the media keeps governments on their
AusSocAtts03
26/7/05
238 •
8:55 AM
Page 238
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
toes, less than half believe that it can keep governments honest. In turn, we find sharply different perceptions of the power of the media, depending on whether the respondent believes that the media is able to play an effective government watchdog role. It seems likely that many respondents are sceptical that the media – itself viewed as too concentrated among a few corporate powerbrokers – can play an important role in checking governmental power, which a majority believe primarily or entirely benefits big interests. A majority of Australians seem unwilling to place their confidence in corporate-dominated media to control the power of government, which they believe benefits the country’s big interests. Nonetheless, that does not mean they have reached for the off button on their television sets. For better or worse, though they harbour significant unease over media concentration, and scepticism about its ability to wrestle with government power, Australians remain dependent on the mass media for the pictures in their heads. REFERENCES
Ansolabehere, S, Behr, R & Iyengar, S (1993) The Media Game: American Politics in the Television Age, Macmillan, New York. Chadwick, P (1989) Media Mates: Carving Up Australia’s Media, Macmillan, Melbourne. Cook, T (1998) Governing with the News: The News Media as a Political Institution, University of Chicago Press, Chicago. Craig, G (2004) The Media, Politics and Public Life, Allen & Unwin, Sydney. Curran, J & Seaton, J (2003) Power Without Responsibility: The Press, Broadcasting and New Media in Britain, 6th edn, Routledge, London. Davis, R (1999) The Web of Politics: The Internet’s Impact on the American Political System, Oxford University Press, New York. Farnsworth, S & Lichter, SR (2003) The Nightly News Nightmare, Rowman & Littlefield, Lanham. Gibson, R, Wilson, S, Meagher, G, Denemark, D & Western, M (2004) Australian Survey of Social Attitudes 2003, Data file, Australian Social Science Data Archive, The Australian National University, Canberra. Given, J (2003) Turning Off the Television: Broadcasting’s Uncertain Future, UNSW Press, Sydney. Lippmann, W (1922) Public Opinion, Macmillan, New York. Marshall, I & Kingsbury, D (1996) Media Realities: The News Media and Power in Australian Society, Addison Wesley Longman, Melbourne. Norris, P (2000) A Virtuous Circle: Political Communications in Postindustrial Societies, Cambridge University Press, Cambridge. Strasburger, V & Wilson, B (2002) Children, Adolescents, and the Media, Sage, Thousand Oaks. Street, J (2001) Mass Media, Politics and Democracy, Palgrave, Houndmills. Tiffen, R (1989) News and Power, Allen & Unwin, Sydney. —— (2000) ‘The news media and Australian politics: Contemporary challenges for Australian democracy in the information age’, in Paul Boreham, Geoffrey Stokes & Richard Hall (eds) The Politics of Australian Society,
AusSocAtts03
26/7/05
8:55 AM
Page 239
MASS MEDIA AND MEDIA POWER IN AUSTRALIA
• 239
Longman, Melbourne. Van Evra, J (1990) Television and Child Development, Lawrence Erlbaum, Hillsdale. Ward, I (1995) Politics of the Media, Macmillan, Melbourne. —— (2002) ‘Media power’, in John Summers, Dennis Woodward & Andrew Parkin Government, Politics, Power and Policy in Australia, 7th edn, Longman, Melbourne. Windshuttle, K (1984) The Media, Penguin, Melbourne. —— (1988) The Media, new edn, Penguin, Melbourne.
AusSocAtts03
26/7/05
8:55 AM
Page 240
CHAPTER 14
ARE AUSTRALIANS OPEN TO G L O B A L I S AT I O N ? I a n M a r s h , G a b r i e l l e M e a g h e r a n d S h a u n Wi l s o n
INTRODUCTION Globalisation is an elusive – and much contested – phenomenon. Some analysts highlight economic integration, while others emphasise deepening linkages across cultures, societies and polities. Some see regional institutions like the European Union or the Association of South East Asian Nations (ASEAN) as a means for globalisation, while others see these same institutions as a defence against globalisation (Fligstein & Merand 2001). One thing seems certain: the reach of supranational actors, institutions and decisions is expanding, through trade, financial markets, global production networks, multinational corporations, migration, international crime and terrorism, cultural diffusion and environment change (Held et al. 1999; see also Keohane & Nye 2000). Many fear that this will produce a ‘race to the bottom’ as global trade transfers production and wealth to cheaper producers, which will in turn require cuts to the well-established welfare states and high living standards of advanced democracies (see, for example, Stiglitz 2003; Tonelson 2002). Others argue that globalisation is having less harmful effects, even promoting social spending and faster growth (see Swank 2002; Garrett & Mitchell 2001). Whether these developments mean that nation-states are now constrained in their domestic policy capacity or adapting to the new environment is also widely debated (see, for example, Conley 2002; Marsh 2005; Weiss 2003). Australia accelerated its engagement with the global economy in the early 1980s when the protectionist and regulated framework of the Australian settlement was relaxed (see Conley 2001; Edwards 1996;
AusSocAtts03
26/7/05
8:55 AM
Page 241
A R E A U S T R A L I A N S O P E N T O G L O B A L I S AT I O N ?
• 241
Evans 1997). The main changes are familiar: tariff reductions, a free floating Australian dollar, relaxation of capital flows, and some deregulation of the nation’s labour and product markets – all designed to promote competitiveness. Reformist energy peaked under the Keating government in the 1990s, which endorsed the Garnaut report in stressing Asia’s importance to Australia’s future trade and economic prospects (Garnaut 1989). Although the Howard government has since been more identified with the United States in politics and trade, Australia has recently signed free trade agreements with Singapore and Thailand, and, together with New Zealand, has begun negotiating a free trade agreement with ASEAN (see DFAT n.d.). Culturally and politically, Australia has also become more globally integrated (see chapter 11). Australia has benefited from a new wave of migration: Asian-born Australians are now a substantial minority population, and Australia is host to large numbers of Asian students who benefit from Australia’s higher education system. Tourism is a major industry, and visitors from Japan and other regional states now make up almost half of all international holiday arrivals (Tourism Australia 2004, p. 24). Australians are also enthusiastic travellers, and young people are, in increasing numbers, spending extended periods working overseas (Hugo, Rudd & Harris 2003). And Australians have been early adopters of new Internet technologies, which have driven the expansion of microexporters (Ham 2004). Some Australians believe that Australia’s global engagement has been rapid. But key indicators like trade suggest more gradual change. Take one measure of Australia’s trade exposure – the total value of exports and imports as a percentage of GDP. On this measure, Australia’s trade exposure has risen from 27 per cent in 1973 to 30 per cent at the start of major reforms in 1983, and is now 39 per cent (see also figure 14.1). But, of the 14 advanced democracies surveyed in this chapter (see table 14.6), Australia’s trade exposure remains among the lowest (12th after Japan and the United States). Still, One Nation’s emergence in the late 1990s signalled some latent resistance to Australia’s international and Asian engagement. Many who voted for One Nation linked new economic insecurities to the open economy and wanted a return to economic nationalism, and/or worried about Asian influence and immigration (Goot & Watson 2001; Wilson & Turnbull 2001; Mughan, Bean & McAllister 2003). This chapter uses the Australian Survey of Social Attitudes 2003 (Gibson et al. 2004) to undertake a more detailed study of the public’s
26/7/05
242 •
8:55 AM
Page 242
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Figure 14.1 Imports and exports as a percentage of GDP, Australia, 1972–2003
25
20
15
GDP %
AusSocAtts03
Imports Exports 10
5
0
1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 SOURCE OECD Economic Outlook no. 75, 2004
reaction to globalisation. Are Australians embracing economic, political and cultural engagement with the rest of the world – in particular, with Asia? Or are we still attached to protectionism, both economic and cultural? Are Australians divided about the benefits of global trade, depending on their own economic fortunes? Do we see international bodies like the United Nations as too interfering? Investigating these questions results in some surprising answers. We conclude by considering how Australia’s global experience compares with other advanced democracies and by drawing out some implications of our findings.
HAVE AUSTRALIANS EMBRACED ECONOMIC OPENNESS? AuSSA 2003 asks respondents about their perceptions of the risks of economic openness posed by the impact of: foreign competition on job security, large international companies on local businesses, and free trade on the availability of products. Attitudes about appropriate policy responses are addressed by two questions about industry protection (see table 14.1). Overall, table 14.1 shows that many Australians see increased economic openness as risky for both workers and businesses, and that many support government intervention to protect local industry. One much-cited risk of globalisation is increased job insecurity – and nearly half
AusSocAtts03
26/7/05
8:55 AM
Page 243
A R E A U S T R A L I A N S O P E N T O G L O B A L I S AT I O N ?
• 243
of all respondents (49 per cent) agree that opening up Australia’s economy to foreign competition has had a bad effect on job security. Only 22 per cent disagree with this, and around a quarter are in the middle. Economies can open to foreign competition through trade liberalisation, or through the relaxation of foreign investment controls. In recent decades, Australian governments pursued both. Foreign investment rules in Australia have become more liberal (Commonwealth Treasury of Australia 1999) and company tax rates have been reduced (Warren 2004), making Australia more attractive for foreign investment. Table 14.1 shows that a large majority of Australians are anxious about the impact of multinational companies (MNCs): three-quarters of all respondents to AuSSA 2003 agree that ‘Large international companies are doing more and more damage to local businesses in Australia’. A small minority (6 per cent) disagree. Respondents with higher education, those with higher incomes, and those living in inner cities are less likely to agree that MNCs are damaging. According to the Australian Treasury (1999, p. 65), the vast majority of foreign investment proposals involve the purchase of real estate. Surprisingly, respondents are less worried about such transactions than they are about the damaging impact of large international companies; 42 per cent of respondents agree that ‘Foreigners should not be allowed to buy land in Australia’, while 31 per cent disagree. Table 14.1 Attitudes to economic openness, 2003, per cent Agree
Neither agree nor disagree
Disagree
Opening up Australia’s economy to foreign competition has a bad effect on job security in this country
49
23
22
6
Large international companies are doing more and more damage to local businesses in Australia
75
16
6
3
Foreigners should not be allowed to buy land in Australia
42
24
31
3
Australia should limit the import of foreign products in order to protect its national economy
65
19
14
2
Australia should use tariffs to protect its industry
52
25
13
10
Free trade leads to better products becoming available in Australia
49
28
18
5
(n ≥ 2108) SOURCE The Australian Survey of Social Attitudes 2003
Can’t choose
AusSocAtts03
26/7/05
244 •
8:55 AM
Page 244
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Governments can manage the risks of foreign competition through trade protection, and Australia has traditionally pursued this strategy with tariffs. AuSSA 2003 asks respondents how much they agree with the statement that ‘Australia should limit the import of foreign products in order to protect its national economy’. Support is high: 65 per cent of respondents agree, and only 14 per cent disagree (see table 14.1). Interestingly, however, when offered a concrete proposal for trade protection, we see that agreement drops somewhat: 52 per cent of all respondents agree that Australia should use tariffs to protect industry, and only 13 per cent disagree. We have shown that many AuSSA 2003 respondents believe that economic openness threatens job security and are worried about MNCs. But many also believe that economic openness enhances consumer choice. Table 14.1 shows that nearly half AuSSA 2003 respondents (49 per cent) agree that ‘Free trade leads to better products becoming available in Australia’. Indeed, only 18 per cent disagree with this proposition. This suggests that Australians hold mixed views about the benefits of global economic integration, or perceive its impacts differently. We use multivariate analysis to uncover relationships between perceptions of the risks and opportunities of economic openness, attitudes to protection, and demographic characteristics. Appendix table A14.1 reports the results of three logistic regression models. The first model predicts disagreement with the proposition that opening the economy to foreign competition is bad for job security. In addition to demographic variables (age, income, education, occupation, and region of residence), we include expectations about living standards and a general measure of job security to capture subjective influences. Results show a strong divide between winners and losers. Respondents who disagree that economic openness threatens job security are more likely to have university qualifications, a higher income, and to work in management or the professions. Respondents who live outside cities, who believe that they are likely to lose their job in the next year, and who feel that they cannot improve their living standards are all less likely to disagree. In other words, these respondents are more inclined to believe that an open economy is bad for job security. Second, we model disagreement with the proposition that Australia should limit the import of foreign products to protect the economy. On this question, education matters most: we find that university-educated respondents are two and a half times more likely than those without a university degree to disagree with import restric-
AusSocAtts03
26/7/05
8:55 AM
Page 245
A R E A U S T R A L I A N S O P E N T O G L O B A L I S AT I O N ?
• 245
tions (see Appendix, table A14.1). Age, income and occupation do not significantly affect attitudes to trade protection, but those in rural areas are less likely than those in cities to disagree. Importantly, we also find that respondents who feel that they cannot improve their living standards are much less inclined to oppose import restrictions. Those who think that opening the economy to foreign competition has been bad for job security are also less opposed. This suggests that respondent support for an open economy depends on perceptions of their own economic fortunes. Third, we model agreement with the proposition that free trade has led to better products being available in Australia. We find that income, education, occupation and region of residence do not make a significant difference to attitudes (see Appendix, table A14.1). Age does make a difference, although a small one: younger respondents are slightly less likely to agree. Perceptions of economic mobility also make a significant difference: we find that those who do not think that they can improve their living standards are nearly 30 per cent less likely to agree. This suggests that Australians who feel excluded from the benefits of economic growth and mobility also feel that free trade does not offer them much. How do Australian attitudes compare? In 1995, the International Social Survey Program (ISSP) asked respondents from 14 advanced democracies for their views on limiting imports to protect the economy.1 This data allows us to compare Australian attitudes then and now, and Australian attitudes with those of other rich countries. The first thing to note is that support for import restrictions among Australian survey respondents fell from a high of 78 per cent to 65 per cent between 1995 and 2003.2 Second, in 1995, Australians were more likely to support import restrictions than citizens of any other country mentioned (see table 14.6). Spaniards (77 per cent) and Austrians (72 per cent) came close. Next were the world’s major English-speaking economies, the United States (69 per cent) and the United Kingdom (66 per cent). By contrast, less than half of Swedish, (West) German, Norwegian, Japanese and Dutch respondents agreed – all countries with high trade exposure and most with low tariffs.
N AT I O N A L V E R S U S S U P R A N AT I O N A L GOVERNMENT Keohane and Nye (2000) argue that one important difference between globalism in the 19th century and globalism in the 20th and 21st centuries is that organisations such as the United Nations and the WTO
AusSocAtts03
26/7/05
246 •
8:55 AM
Page 246
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
represent a truly supranational sphere. Australian governments have not responded consistently to these developments, reflecting different approaches to economic and political institutions and conceptions of Australia’s interests. In 1995, the Keating Labor government ceded binding arbitral authority to the WTO, in line with its commitment to a global multilateral free trade regime. We can speculate that, had the WTO dispute settlement system been established during the Howard government, it too would have ceded this authority. However, the Howard government has been more reluctant than Labor governments to recognise the authority of other international institutions. For example, the ongoing controversy about Australia’s treatment of refugees, particularly the mandatory detention of children, has raised questions about Australia’s failure to meet its international human rights obligations under several UN Conventions (Human Rights Watch 2002). The Howard government has also refused to recognise the jurisdiction of the World Court in its dispute over sea boundaries with East Timor (Joint Standing Committee on Treaties 2002). AuSSA 2003 asks several questions about the appropriate balance of power between international organisations and the Australian Government (see table 14.2). On the question of whether, ‘for certain problems, like environmental pollution, international bodies should have the right to enforce solutions’, nearly two-thirds of AuSSA 2003 respondents agree (63 per cent). Thus, it seems, a majority of Australians hold internationalist attitudes: despite the government’s reluctance to sign the Kyoto Protocol, and to meet its obligations to refugees, many Australians accept international authority. But a substantial minority of the Australian public are wary of this authority. Almost two in five agree that ‘international organisations are taking away too much power from the Australian Government’, while only a quarter (27 per cent) agree with generally accepting international authority. We find high numbers of respondents choosing ‘Neither agree nor disagree’ with these questions, which suggests that many Australians are ambivalent about the power of international organisations, or that the problems raised here are too complex for firm responses. How do these views compare to those of the citizens of other advanced democracies? ISSP findings in 1995 show that Australians are the least internationalist. Among Australian respondents in 1995, 67 per cent agreed that international bodies should have the right to enforce solutions, compared to 63 per cent in 2003 (see table 14.6). By contrast, 82 per cent of Swedes, 86 per cent of (West) Germans, and 92
AusSocAtts03
26/7/05
8:55 AM
Page 247
A R E A U S T R A L I A N S O P E N T O G L O B A L I S AT I O N ?
• 247
Table 14.2 Attitudes to international authority, 2003, per cent Agree
Neither agree nor disagree
Disagree
Can’t choose
For certain problems, like environmental pollution, international bodies should have the right to enforce solutions
63
17
18
3
International organisations are taking away too much power from the Australian Government
39
32
23
6
In general, Australia should follow the decisions of the international organisations to which it belongs, even if the government does not agree with them
27
32
36
6
(n ≥ 2134) SOURCE The Australian Survey of Social Attitudes 2003
per cent of Italians agreed. Support for the authority of international bodies was lower among respondents from the English-speaking countries with which Australia is most frequently compared, but Australia still stands out. Among New Zealanders, 72 per cent agreed; among Canadians, 78 per cent agreed; among respondents from the United States, 73 per cent agreed, and Britons agreed at a rate of 74 per cent. Again we turn to multivariate analysis to deepen our understanding. We apply ordinary least squares regression to responses on the most interesting of these statements for contemporary Australian politics: that international organisations are taking too much power away from the Australian Government (see Appendix, table A14.2). The model tests the effect of demographic variables and party identification as well as the influence of attachment to an Australian identity. On the latter, we construct a ‘nativism’ scale, which is similar to that used in chapter 11. It includes three measures of identity – the importance of being born in Australia, living mostly in Australia, and having Australian ancestry. The results show that respondents with a university education and those who identify as Green or Democrat are less likely to agree that international organisations are taking power. But Liberals, Nationals and One Nation supporters are inclined to agree, and so are respondents without a party attachment. Having a strong sense of Australian identity also makes a difference. Strongly ‘nativist’ Australians think that international organisations are taking power away from the Australian Government. Not all of this may have to do with the United Nations and the WTO; it might
AusSocAtts03
26/7/05
248 •
8:55 AM
Page 248
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
equally have to do with the power of multinationals in the economy. But one thing is clear: ignoring international agencies, as the Howard government has occasionally done over asylum seekers, trade and the environment, will be seen by some as reasserting the authority of the Australian Government.
W H AT R E L AT I O N S H I P D O A U S T R A L I A N S WANT WITH ASIA? In practice, leaders of both major political parties in Australia have actively supported closer trade and political ties with Asia, but the Howard government’s rhetoric has been more ambiguous. During the mid-1990s, Prime Minister Keating enthusiastically championed turning to the Asia–Pacific region (see Keating 2000). Some believe that, in contrast, John Howard’s approach to Australia–Asia relations has played on traditional Australian insecurities (Brett 2003). But the government may also be telling one story to its Australian audience, and quite another to our nearest and most important trading partners. As we mentioned earlier, the Howard government has concluded free trade agreements with Singapore and Thailand, and, in late 2004 revitalised its relationship with ASEAN. However, the Prime Minister continues to describe Australia as ‘an essentially European derived nation, but located in the Asian Pacific region, [which] means of course that we share the culture and the inheritance of western civilisation’ (Howard 2002). He emphasises Australia’s role as a ‘political citizen of the world’ rather than ‘just a citizen of our own region’ (Howard 2000). Australian attitudes to economic engagement with Asia have become less supportive in the last decade or so, suggesting that the transfer of leadership from Keating to Howard has influenced how Australians feel
Table 14.3 Attitudes to Asian engagement, 2003, per cent Agree
Neither agree nor disagree
Disagree
Australia should pursue greater economic ties with Asia
56
27
13
5
Australia should pursue greater cultural and political ties with Asia
43
31
22
4
(n ≥ 4165) SOURCE The Australian Survey of Social Attitudes 2003
Can’t choose
AusSocAtts03
26/7/05
8:55 AM
Page 249
A R E A U S T R A L I A N S O P E N T O G L O B A L I S AT I O N ?
• 249
about the region. Australian Election Studies (AES) have sought responses to the statement ‘Australia’s trading future lies in Asia’. In 1993, 68 per cent agreed. Support for the statement fell to 66 per cent in 1996 and to 55 per cent in 1998. In 2001, 57 per cent of respondents agreed – almost the same proportion of AuSSA 2003 respondents who agree with a similar statement, that Australia should pursue greater economic ties with Asia. Overall, it seems that ambivalence about engaging with Asia is replacing earlier agreement: undecided responses in the AES increased from 24 per cent in 1993 to 33 per cent in 2001. Table 14.3 shows that, while a majority of Australians support closer economic ties with Asia (56 per cent), fewer agree that Australia should pursue greater political and cultural ties with the region (43 per cent). On both questions, a large minority of respondents neither agree nor disagree (around 30 per cent). AuSSA 2003 also asks ‘How close do you feel to Asia?’, and responses to the latter question strongly condition attitudes to economic, political and cultural ties to Asia. Among the minority (15 per cent) of respondents who feel close or very close to Asia, 80 per cent agree that Australia should pursue closer economic ties, and 70 per cent agree that we should pursue closer political and cultural ties. By contrast, among the majority (80 per cent) of respondents who feel not very or not at all close to Asia, 52 per cent agree that Australia should pursue closer economic ties with Asia, while just 35 per cent agree that we should pursue closer political and cultural ties. We use ordinary least squares regression to analyse attitudes to economic and cultural/political engagement with Asia (see Appendix, table A14.3). We find similar factors influence responses for both (further research might usefully discover the identity of respondents who support economic engagement but do not prefer cultural and political engagement). Factors that predict agreement with both types of engagement are: being male, being older and having a university degree. Disagreement is higher among women, younger respondents, those who think that an open economy is bad for job security, those who score highly on the nativism scale, and voters who identify as Liberal or National, or have no party identification. One Nation identifiers register strong resistance to cultural/political engagement but not to economic engagement – but this difference may be a product of a small subsample. Fears of job insecurity and a strong attachment to national identity may hamper Australia’s efforts to engage with Asia: the domestic audience is clearly still receptive to ‘Australia first’ messages.
AusSocAtts03
26/7/05
250 •
8:55 AM
Page 250
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
G L O B A L I S AT I O N A N D C U LT U R E So far we have focused on attitudes to economic and political dimensions of globalisation. AuSSA 2003 also asks about cultural aspects of globalisation in questions about the impact of exposure to foreign cultural products and the need to protect Australian content in film and television. Table 14.4 shows that less than a quarter of all respondents (24 per cent) agree that increased exposure to foreign films, music and books is damaging our national and local cultures, and more than half (53 per cent) disagree. Older respondents and those with lower incomes tend to be more protectionist. Among respondents aged 65 and over, 40 per cent agree, compared to less than a quarter of all those younger. Among respondents with annual income of less than $16 000 per year, nearly a third agree, compared to 13 per cent of those with an income of more than $78 000. We often see a strong link between ‘cosmopolitan’ attitudes and education. This time, however, university education does not draw the crucial dividing line. Less than a fifth of all respondents who completed Year 12 (18 per cent), a certificate or diploma
Table 14.4 Exposure to foreign films, music and books causes damage to national and local cultures by age, education and income, 2003, per cent Agree
Neither agree nor disagree
Disagree
Can’t choose
18–34
15
18
65
2
35–49
17
21
60
2
50–64
23
23
51
3
65 and over
40
21
35
3
Less than Year 12
30
25
40
5
Year 12
18
20
60
1
Trade/apprenticeship
34
22
42
2
Age
Education
Certificate/Diploma
17
25
57
2
Bachelor’s degree and above
14
13
72
0
Annual individual income $0 to $15 599
30
22
45
3
$15 600 to $36 399
23
22
52
2
$36 400 to $77 999
17
20
62
2
$78 000 and over
13
16
70
1
24
21
53
Total (n ≥ 2023) SOURCE The Australian Survey of Social Attitudes 2003
2
AusSocAtts03
26/7/05
8:55 AM
Page 251
A R E A U S T R A L I A N S O P E N T O G L O B A L I S AT I O N ?
• 251
Table 14.5 Australia’s television should give preference to Australian films and programs by age, education and income, 2003, per cent Agree
Neither agree nor disagree
Disagree
Can’t choose
18–34
38
33
28
1
35–49
42
31
24
2
50–64
47
27
25
2
65 and over
59
20
20
1
Less than Year 12
46
28
23
4
Year 12
38
31
30
1
Trade/apprenticeship
55
24
21
1
Age
Education
Certificate/Diploma
46
27
26
1
Bachelor’s degree and above
46
30
24
0
Annual individual income $0 to $15 599
52
27
19
2
$15 600 to $36 399
44
27
26
3
$36 400 to $77 999
44
29
26
1
$78 000 and over
36
32
33
0
46
28
24
Total (n ≥ 2011)
2
SOURCE The Australian Survey of Social Attitudes 2003
(17 per cent), or a Bachelor’s degree or higher (14 per cent) agree, compared to 30 per cent of those who did not complete Year 12 and 34 per cent of those with a trade qualification. Just as governments can protect local industries with import restrictions, they can also protect cultural industries with ‘local content’ rules. AuSSA 2003 asks respondents whether they agree or disagree with this statement: ‘Australia’s television should give preference to Australian films and programs’. Nearly half (46 per cent) agree (see table 14.5). Of the remainder, more are undecided (28 per cent) than disagree (24 per cent). We see a familiar pattern in responses. Age makes a difference, with the young much less likely to agree (38 per cent of those under 35 years) than older respondents (59 per cent of those aged 65 and over). Income also matters, with those earning higher incomes less likely to agree with this form of cultural ‘protectionism’ than those with lower incomes. Education does not seem to make much difference here, although respondents with different educational backgrounds may support protection for different reasons: ‘cosmopolitans’ who are liberal on other measures of globalisation may have incentives to protect and nurture local culture.
AusSocAtts03
26/7/05
252 •
8:55 AM
Page 252
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
In 1995, the ISSP asked respondents whether they agreed that television should give preference to local films and programs. Again, Australia stands out as the population with the most protectionist attitudes among the rich democracies surveyed (see table 14.6). In 1995, 58 per cent of Australian respondents agreed that Australian television should give preference to Australian material, compared to 21 per cent of Swedes, 29 per cent of (West) Germans, and 36 per cent of Italians. Among the advanced democracies included here, only Spain and Ireland also express majority support (56 per cent) for preferential treatment of local films and programs. And among the English-speaking countries, Australia continues to stand out: 30 per cent of New Zealanders, 49 per cent of Canadians, 37 per cent of Americans and 38 per cent of Britons agreed.
G L O B A L C O M PA R I S O N S A N D P O L I C Y A LT E R N AT I V E S Although the effect of global capital flows on welfare states is ambiguous, one finding about globalisation is that advanced economies with high trade exposure tend to develop larger welfare states, which are required to protect domestic constituencies from adverse risks of free trade (see Rodrik 1997; Swank 2002, p. 33). Australia, by contrast, is characterised by a relatively low reliance on overseas trade and a lean social welfare effort. Do we find any link between attitudes, trade and welfare effort across nations that helps us to understand Australia’s response to global challenges? Table 14.6 compares attitudes to three components of globalisation – import limits, international enforcement, and cultural protection – with OECD trade and social expenditure data from 14 advanced democracies, including the six English-speaking welfare states. It turns out that Australians were more supportive of limiting imports than any other nation, and had the third lowest social expenditure and trade exposure. The two countries with lower trade exposure are the United States and Japan – both with huge internal markets. Of the smaller countries, Australia is the leanest social spender, and has the lowest exposure to trade. On international bodies enforcing solutions, Australians were, again, least likely to agree and most protectionist on local film and programs. By 2003, support for protecting industry and local content had fallen somewhat and how this compares with changes in other countries will be a subject for further research as new ISSP data become available.
AusSocAtts03
26/7/05
8:55 AM
Page 253
A R E A U S T R A L I A N S O P E N T O G L O B A L I S AT I O N ?
• 253
Table 14.6 Australia compared – attitudes to globalisation, trade exposure and public social expenditure for 14 countries, 1995 (2003), per cent agree* Total trade** (% GDP)
Public social Imports of International Television expenditure*** foreign products bodies should should give (% GDP) should be have the right preference limited to to enforce to local protect the solutions films and economy programs
Australia
39.9 (39.2)
17.8 (n/a)
78 (65)
67 (63)
58 (46)
Spain
45.4
20.9
77
81
56
Austria
74.4
27.9
72
76
47
United States
23.2
15.4
69
73
37
United Kingdom
57.1
25.8
66
74
38
141.4
19.6
66
68
56
Italy
50.0
23.7
62
92
36
New Zealand
58.7
19.3
54
72
30
Canada
71.5
19.2
48
78
49
Sweden
71.8
33.0
43
82
21
(West) Germany
48.3
26.7
42
86
29
Norway
69.7
27.6
41
71
24
16.8
13.5
33
70
33
109.0
25.9
31
72
17
Ireland
Japan Netherlands
* Countries are ranked by agreement scores in column 3. SOURCES International Social Survey Program 1995; ** OECD Economic Outlook no. 75 (2004); includes East and West Germany; *** OECD Social Expenditure Database, 2001
Both in social attitudes and on key indicators of global engagement, Australians are comparatively wary of the benefits of global and regional integration – and more so than many Australians might suspect. Part of this wariness may have something to do with the Australian political system itself. Despite a bipartisan consensus in support of economic reform that has now entered its third decade, public opinion has not always followed along with it (see also chapter 10). The gap in public opinion – most powerfully expressed in the 1990s by One Nation’s attack on political elites – indicated a latent conflict between reformers and the public. As Marsh and Yencken (2004) argue, this ‘democracy gap’ emerges out of inadequate links between the political system and the public that might also be reformed to allow for greater public involvement in policy development. And economic reforms themselves might be contributing to public resistance to global integration. If Australian governments are right in pursuing an open economy more dependent on international trade, then they may need to deal with the ‘new insecurities’ for Australian workers in the global economy who must cope with rapidly
AusSocAtts03
26/7/05
254 •
8:55 AM
Page 254
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
changing industries and market developments. The OECD (2004b) itself recognises this in calling for greater commitment to active labour market programs that retrain workers, and for the creation of new job opportunities. At the moment, Australians identify ‘traditional’ forms of protection as a means to job security and a viable national economy. If greater integration is to be sustained, innovative social and employment protection may be required so that Australians believe that dealing with the global economy is worth it. In turn, this would require a return to active government and the kind of spending on social and employment programs that has been adopted by some of the world’s most successful open economies. Our final contribution in this chapter looks at the size of the potential constituencies for different policy directions. Here, we contrast the open economy/high social protection model of some of the most successful nations with the traditional Australian policy orientation of high industry protection/low social spending. We divide the AuSSA 2003 sample into groups by their support for tariffs to protect industry, and their support for higher social spending. This produces four constituencies: a pro-global group (who are against tariffs, but support spending), a neo-liberal group (who are against both), an ‘Australian settlement’ group (who support tariffs, but are against spending) and a final group of social and economic protectionists who support both tariffs and spending. In 2003, the smallest group are neo-liberals, who make up about 20 per cent of respondents, and who are followed by the pro-global group with 22 per cent of respondents. The largest group are the Australian settlement group, who support both protection and lower taxes (30 per cent). This group prefers the Australia of old, with tariffs offering both industry and social protection, and table 14.7 shows that they are more likely to identify with the Coalition than with Labor. Convincing this group that there is much to be gained from trade with Asia and that global engagement will improve living standards may depend in part on the new jobs, industries and income support that can be guaranteed in return. But, unlike many small European economies that have increased social spending in response to greater trade, the ‘liberal’ welfare states have been most likely to cut back social provision in response to increased global capital mobility (see Swank 2002, pp. 118–19). The pro-global group includes the highest number of Green and Democrat identifiers (13 per cent), which gives some clues about the profile of the constituency most attracted to global engagement.
AusSocAtts03
26/7/05
8:55 AM
Page 255
A R E A U S T R A L I A N S O P E N T O G L O B A L I S AT I O N ?
• 255
Table 14.7 Constituencies for trade and social protection, 2003, per cent
Party Identification ALP Liberal/National One Nation Green/Democrat Other/none Total n
Pro-global
Neo-liberals
Social and economic protectionists
Australian settlement
41 28 1 13 17
30 43 2 3 22
35 34 3 11 17
35 40 2 2 21
22
20
28
30
(398)
(368)
(509)
(535)
SOURCE The Australian Survey of Social Attitudes 2003
CONCLUSION This chapter has shown that Australians are comparatively ‘closed’ to globalisation on key subjective and objective measures, and that Australian attitudes to global engagement are shaped by respondents’ economic security and views about the effect of an open economy on jobs. Many respondents have a strong sense of Australian identity, which shapes their attitudes to the influence of international organisations on Australian life. Overall, we find that many Australians are wary of greater economic, political and cultural engagement. If these attitudes are to change, one policy response is to manage better the risks of global engagement. Economist Dani Rodrik (1997, pp. 6–7), for example, has warned that: the greatest risk of all [is] … that the cumulative consequences of [globalisation] will be the solidifying of a new set of class divisions between those who prosper in the globalized economy and those who do not, between those who share its values and those who would rather not, and between those who can diversify away its risks and those who cannot. Our findings suggest that the constituency for policy that embraces global trade and greater social spending to manage the attendant employment and income risks is limited. Australians remain protectionist, both culturally and economically, and although protectionist sentiment may change over time, as it has done already, it will continue to provide political opportunities for disengagement.
AusSocAtts03
26/7/05
256 •
8:55 AM
Page 256
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
NOTES
1
2
The ISSP included 24 countries in this study; of these, 14 were what Harold Wilensky (2002) has called rich democracies: Australia, Austria, West Germany, Ireland, United Kingdom, the Netherlands, Norway, Sweden, Italy, Spain, United States, New Zealand, Japan, Canada. The other rich democracies usually included in comparative studies, but absent here, are Belgium, Denmark, Finland, France and Switzerland. The remaining nations included in the ISSP in 1995 were mostly transition economies: East Germany, Hungary, the Czech and Slovak Republics, Bulgaria, Poland, Russia, the Philippines, Latvia and Slovenia. This finding is corroborated by data from Australian Election Studies of 1993, 1996 and 1998 on responses to the statement ‘Australia should continue to use tariffs to protect its industry’. In 1993, 63 per cent of respondents agreed; in 1996, 69 per cent agreed; in 1998, 57 per cent agreed, and, as we have noted, 52 per cent of respondents to AuSSA 2003 agree to the slightly different statement that Australia should use tariffs to protect industry.
REFERENCES
Brett, J (2003) Australian Liberals and the Moral Middle Class, Cambridge University Press, Melbourne. Commonwealth Treasury of Australia (1999) ‘Foreign investment policy in Australia – A brief history and recent developments’, Economic Roundup, Spring: 63–70. Conley T (2001) ‘The domestic politics of globalisation’, Australian Journal of Political Science, 36(2): 223–46. —— (2002) ‘Globalisation as constraint and opportunity: Reconceptualising policy capacity in Australia’, Global Society, 16(4): 377–99. Department of Foreign Affairs and Trade (DFAT) (n.d.) ‘Free trade agreements’, Department of Foreign Affairs and Trade. Available at: (accessed March 2005). Edwards, J (1996) Keating: The Inside Story, Viking Books, Melbourne. Evans, P (1997) ‘The eclipse of the state: Reflections of stateness in an era of globalization’, World Politics, 50(1): 62–87. Fligstein, N & Merand, F (2001) ‘Globalization or Europeanization? Evidence on the European economy since 1980’, paper prepared for the conference ‘Shareholder Value–Capitalism and Globalization’, Hamburg, 10–12 May. Garnaut, R (1989) Australia and the Northeast Asian Ascendancy: Report to the Prime Minister and Minister for Foreign Affairs and Trade, Australian Government Publishing Service, Canberra. Garrett, G & Mitchell, D (2001) ‘Globalization, Government Spending and Taxation in the OECD’, European Journal of Political Research, 39(2): 145–77. Gibson, R, Wilson, S, Meagher, G, Denemark, D & Western, M (2004) Australian Survey of Social Attitudes 2003, Data file, Australian Social Science Data Archive, The Australian National University, Canberra. Goot & Watson (2001) ‘One Nation’s electoral support: Where does it come from, what makes it different and how does it fit?’, Australian Journal of Politics and History, 47(2): 159–91. Ham, M (2004) ‘Small exporters making it big’, Sydney Morning Herald, 13 October. Held, D, McGrew, A, Goldblatt, D & Perraton, J (1999) Global Transformations: Politics, Economics and Culture, Stanford University Press, Stanford.
AusSocAtts03
26/7/05
8:55 AM
Page 257
A R E A U S T R A L I A N S O P E N T O G L O B A L I S AT I O N ?
• 257
Howard, J (2000) Transcript of the Prime Minister, the Hon John Howard MP, Address to Aston electorate luncheon, Melbourne, 10 August. —— (2002) Transcript of the Prime Minister, the Hon John Howard MP, Address to Representatives of the Federation of German Industry and the German Council of Foreign Relations Deutsche Bank, Berlin, 2 July. Hugo, G, Rudd, D & Harris, K (2003) Australia’s Diaspora: Its Size, Nature and Policy Implications, Committee for Economic Development of Australia, Melbourne. Human Rights Watch (2002) ‘By Invitation Only’ Australian Asylum Policy, December, 14(10). Available online: . Joint Standing Committee on Treaties (2002) Report 47: Treaties tabled on 18 and 25 June 2002, Parliament of the Commonwealth of Australia. Keating, P (2000) Engagement: Australia Faces the Asia–Pacific, Macmillan, Sydney. Keohane, R & Nye, J (2000) ‘Globalization: What’s new? What’s not? (And so what?)’, Foreign Policy, Spring: 104–19. Marsh, I (2005) ‘Neo-liberalism and the decline of democratic government: A problem of institutional design’, Political Studies, 53(1): 22–42. —— & Yencken, D (2004) Into the Future: The Neglect of the Long Term in Australian Politics, Australian Collaboration/Black Inc., Melbourne. Mughan, A, Bean, C & McAllister, I (2003) ‘Economic globalization, job insecurity and the populist reaction’, Electoral Studies, 22(4): 617–33. Organisation for Economic Cooperation and Development (OECD) (2004a) Economic Outlook 2004, no.75, Organisation for Economic Cooperation and Development, Paris. —— (2004b) Employment Outlook 2004, Organisation for Economic Cooperation and Development, Paris. Rodrik, D (1997) ‘Has globalisation gone too far?’, California Management Review, 39(3): 29–53. Stiglitz, J (2003) Globalization and Its Discontents, Norton, New York. Swank, D (2002) Global Capital, Political Institutions and Policy Change in Developed Welfare States, Cambridge University Press, New York. Tonelson, A (2002), Race to the Bottom: Why a Worldwide Worker Surplus and Uncontrolled Free Trade Are Sinking American Living Standards, Westview Press, Boulder, Colorado. Tourism Australia (2004) Inbound Tourism Trends, Year Ended 30 June 2004, Tourism Australia, Sydney. Warren, N (2004) Tax: Facts, Fiction and Reform, Australian Tax Research Foundation Research Study no. 41, Australian Tax Research Foundation, Sydney. Weiss, L (ed.) (2003) States in the Global Economy: Bringing Domestic Institutions Back In, Cambridge University Press, Cambridge. Wilensky, H (2002) Rich Democracies: Political Economy, Public Policy, and Performance, University of California Press, Berkeley. Wilson, S & Turnbull, N (2001) ‘The two faces of economic insecurity: Reply to Goot and Watson on One Nation’, Australian Journal of Politics and History, 47(4): 507–10.
AusSocAtts03
26/7/05
8:55 AM
Page 258
APPENDICES CHAPTER 2 Table A2.1 Logistic regression results predicting agreement with living arrangements that make a family, 2003 Unmarried couple with no children
Unmarried couple with children
B
B
S.E
%
S.E
%
Single parent with children
B
S.E
%
Sex Male
ref.
Female
0.33
27 0.11 ** 35
ref. 0.46
80 0.14 ** 86
ref. 0.36
76 0.13 ** 82
0
Age 18–34
ref.
35–49
0.23
0.17
34 –0.34
0.28
50–64
0.21
0.18
34 –0.96
–0.20
0.21
25 –1.92
65 and over
29
ref.
92
ref.
89 –0.56
92 0.27
*
87
0
0.28 ** 82 –1.43
0.27 ** 73
–0
0.29 ** 63 –2.27
0.28 ** 54
–0
Marital status Single, never married
ref.
Living with partner
0.60
0.24
–0.12
Married Separated/divorced Widowed
32
ref.
47
1.83
0.56 ** 95
0.71
0.38
86
0
0.17
30
0.43
0.24
84
0.29
0.24
80
–0
0.22
0.24
37
0.16
0.31
80 –0.09
0.31
73
0
–0.43
0.33
24
0.28
0.34
81
0.34
76
–0
*
77
ref.
0.07
75
Highest level of education Less than Year 12
31
ref.
77
ref.
Year 12
–0.09
0.19
29
0.37
0.23
83
0.09
0.22
75
0
Trade/apprenticeship
–0.17
0.19
27
0.31
0.21
82 –0.01
0.19
74
0
Certificate/diploma
–0.18
0.15
27
0.31
0.17
82
0.21
0.16
78
–0
0.40
0.15
*
40
1.13
0.21 ** 91
1.08
0.19 ** 89
0
36
ref.
*
31 –0.36
Bachelor’s degree and above
ref.
74
Religious attendance Never attends
ref.
Sometimes attends
–0.23
0.12
Often attends
–0.63
0.16 ** 23 –1.29
88 0.15
*
ref.
81
84 –0.22
0.14
77
–0
0.18 ** 68 –0.15
0.17
79
–1
Attitude to single parents As good as couple parents
n/a
n/a
1.23
Not as good as couple parents Constant n
0.14 ** 88
ref. –0.80
0.20 ** (1798)
1.80
0.27 **
1.35
(1800)
69 0.28 ** (1794)
NOTE These models predict agreement with living arrangements that make a family. Agreement is a
combination of ‘Strongly agree’ and ‘Agree’. % = the percentage of respondents who agree predicted by the logistic regression model. * p < 0.05; ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
–1
AusSocAtts03
26/7/05
8:55 AM
Page 259
APPENDICES
Same-sex couple with no children
%
B
76
ref.
82
0.75
S.E
Same-sex couple with children
% 11
0.14 ** 21
18
B
S.E
%
ref. 0.85
28 0.12 ** 47
92
ref.
87
0.23
0.19
22 –0.17
0.17
73
–0.05
0.21
17 –0.87
0.19 ** 34
54
–0.92
0.29 **
75
ref.
86
0.65
0.26
80
–0.14
0.20
*
8
ref.
–1.84
56 51
0.23 ** 17
17
ref.
28
0.15
0.26
47
43
15 –0.29
0.18
36
73
0.32
0.27
22 –0.14
0.25
40
76
–0.56
0.49
10 –0.71
0.38
27
14
ref.
0.22
17
0.32
0.19
35
74
ref.
75
0.28
28
74
0.00
0.24
14
0.13
0.20
78
–0.06
0.19
13
0.36
0.15
89
0.72
0.18 ** 24
1.42
0.16 ** 61
81
ref.
77
–0.33
0.14
21
79
–1.09
0.21 **
88
n/a
*
ref.
16 –0.39 8
–1.69
50 0.18 ** 15
n/a
0.24 ** (1801)
0.06
36
0.12 ** 40
69 –1.67
30 *
0.20 (1803)
• 259
AusSocAtts03
26/7/05
260 •
8:55 AM
Page 260
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table A2.3 Logistic regression results predicting agreement with women’s reproductive rights by sex, 2003 Right to choose abortion Men
B
S.E
IVF for
Women
%
B
87
ref.
un
Men
S.E
%
B
92
ref.
S.E
%
Age 18–34
ref.
35–49
–0.08
0.23
86 –0.07
0.24
92 –0.56
0.17 ** 39
–0
50–64
0.02
0.24
87 –0.31
0.25
90 –0.58
0.18 ** 38
–0
65 and over
0.13
0.25
89 –0.52
0.26
88 –0.76
0.19 ** 34
–1
Single, never married
–0.03
0.23
88 –0.08
0.25
90
0.34
0.16
Currently married or cohabiting
ref.
90
ref.
Divorced, separated, widowed
–0.52
92
0.07
92
ref.
*
53
Marital status 88 0.21 ** 81
ref. 0.14
0.18
*
47
0
38 0.18
40
0
Religion No religion Yes, a religion
ref. –0.68
89
ref.
0.20 ** 81 –0.53
0.26
*
87 –0.65
44 0.13 ** 29
0
Religious attendance Never attends
2.05
0.20 ** 92
2.35
0.22 ** 95
0.68
0.19 ** 42
1
Sometimes attends
1.58
0.16 ** 88
2.03
0.16 ** 93
0.67
0.18 ** 42
0
Often attends
ref.
60
ref.
64
ref.
27
ref.
88
ref.
90
ref.
38
86
0.37
0.18
93
0.39
0.28
0.25
Highest level of education Less than Bachelor’s degree Bachelor’s degree and above Constant
n
–0.10
0.16
0.00
0.25
(1718)
* p < 0.05; ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
(1934)
*
–1.06
0.13 *** 47 0.22 **
(1698)
0
–0
AusSocAtts03
26/7/05
8:55 AM
Page 261
APPENDICES
• 261
Table A2.2 Logistic regression results predicting agreement with recognition of same-sex relationships and life without children is not fully complete, 2003 VF for
unmarried women
Law should recognise same-sex relationships
Women
%
B
S.E
B
%
S.E
Life without children not fully complete
%
B
24
ref.
S.E
%
Sex 53
ref.
56
39
–0.21
0.15
38
–0.94
0.16 ** 33
34
–1.36
0.19 ** 25
47
0.51
38
ref.
40
0.29
51
Male
ref.
Female
0.66
0.11 ** 38 –0.49
45 0.07 ** 34
Age
0.16 ** 50 38
18–34
ref.
35–49
–0.13
0.16
38
ref.
35
0.10
0.12
50–64
–0.42
0.17
65 and over
–0.68
29
0.05
0.12 36
0.20 ** 24
0.62
0.13 ** 49
*
34 37
Marital status 0.14
*
44
Single, never married
ref.
Never married 44
ref.
29
0.15
0.13
41
Living with partner
38
Married
0.23 0.16
*
30
n/a
Separated/divorced –0.54
0.23
*
27
n/a
0.32
*
23
n/a
1.09
0.17 ** 49
Widowed
0.86
0.15 ** 43
27
ref.
Highest level of education
47
ref. 0.48 –0.85
38 0.12 ** 50 0.19 **
(1718)
–0.73
43
27 0.11 ** 42
0.18
42
38
ref. 0.66
–0.40
42
24
38 n/a
26
n/a
Less than Year 12
ref.
Year 12
0.14
0.18
28 –0.07
ref. 0.12
40
Trade/apprenticeship –0.15
0.20
23
0.12
43
Certificate/diploma
0.20
0.15
30 –0.25
0.10 ** 36
Bachelor’s degree and above
0.96
0.15 ** 47 –0.30
0.10 ** 35
0.04
42
Religious attendance Never attends
ref.
42
ref.
35
Sometimes attends –0.46
0.11 ** 31
0.23
0.08 ** 41
Often attends
–1.40
0.18 ** 15
0.27
0.10 ** 42
0.89
0.28 ** 30
Sexual identification Heterosexual Homosexual Constant n
n/a
51 –0.32
0.20 (1979)
* p < 0.05; ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
–0.96
0.13 ** (3880)
AusSocAtts03
26/7/05
262 •
8:55 AM
Page 262
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
CHAPTER 5 Table A5.1 Logistic regression model specifications List of dependent variables used Variable name
Description
POLB1245
Indicates political activity for items v61, v62, v64 or v65
Coding
1 = yes, 0 = otherwise
MTALLACT
Active member / officeholder of a voluntary association
0–1
MTORD
Ordinary member only of one or more voluntary associations
0–1
DEGREEP
Respondent qualified with Bachelor’s degree or higher
0–1
MANPROF
Whether respondent’s current or previous occupation is managerial or professional
0–1
INDINC36
Respondent’s annual individual income is $36 000 or higher
0–1
AGE1834
Indicates that respondent is aged 18–34
0–1
AGE3549
Indicates that respondent is aged 35–49
0–1
AGE5064
Indicates that respondent is aged 50–64
0–1
FEMALE
Whether respondent is female
0–1
WORKING
Indicates that respondent is currently working
0–1
Results from full model Variable name
Description
Odds ratio
p-value
MTALLACT
Active member / officeholder of a voluntary association
2.737 <0.0005
MTORD
Ordinary member only of one or more voluntary associations 2.206 <0.0005
DEGREEP
Respondent qualified with Bachelor’s degree or higher
1.811 <0.0005
MANPROF
Whether respondent’s current or previous occupation is managerial or professional
1.292
0.057
INDINC36
Respondent’s annual individual income is $36 000 or higher
1.205
0.15
AGE1834
Indicates that respondent is aged 18–34
0.879
0.5
AGE3549
Indicates that respondent is aged 35–49
1.118
0.5
AGE5064
Indicates that respondent is aged 50–64
1.127
0.5
FEMALE
Whether respondent is female
1.274
0.03
WORKING
Indicates that respondent is currently working
0.999
0.9
R-square = 0.10 (n = 2012) SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
8:55 AM
Page 263
APPENDICES
• 263
CHAPTER 11 CONSTRUCTION OF THE SCALES
An analysis of the items in tables 11.4, 11.5 and 11.6 yields two factors; together they account for 94 per cent of the variance in the responses (table A11.1). The first factor loads very heavily on three items related to being ‘truly Australian’ – being born in Australia, having lived in Australia for most of one’s life, and having an Australian ancestry. This factor is scored to produce a scale that ranges from –2.4 to 2.0 (with a mean of 0.0). Following Jones (1996, 1997, table 4), who analysed the 1995 survey in which two of these three items appeared, we call this a ‘nativism’ scale. The second factor loads most heavily on two items to do with pride in Australia – pride in Australia’s political influence in the world, and Table A11.1 Factor loadings Factor 1
Factor 2
Nativism
National pride
Important to have been born in Australia
0.771
0.096
Important to have Australian citizenship
0.406
0.279
Important to have lived in Australia for most of one’s life
0.730
0.099
Important to be able to speak English
0.453
0.196
Variable
Important to be Christian
0.483
0.096
Important to have Australian ancestry
0.737
0.080
Rather be Australian citizen than any other
0.360
0.398
Better world if others were like Australians
0.378
0.214
Success at sport makes me proud Australian
0.338
0.483
Proud of the way democracy works
0.029
0.482
Proud of Australia’s political influence
0.164
0.607
Proud of Australia’s economic achievements
0.077
0.569
Proud of Australia’s science and technology
–0.018
0.479
Proud of Australia’s achievements in sports
0.183
0.600
–0.019
0.483
Proud of Australia’s achievements in arts and literature Proud of Australia’s defence forces
0.177
0.587
Proud of Australia’s history
0.345
0.483
Proud of Australia’s fair and equal treatment of all groups
0.224
0.411
(n = 1990) NOTE Only factors that achieved a loading of 0.3 or above (on one item) were retained. Scales are created
by a statistical procedure, which employs a linear combination of each item’s (weighted) factor loading. This procedure also standardises the scales. SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
264 •
8:55 AM
Page 264
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Australia’s achievements in sport. It is scored to produce a scale that ranges from –4.2 to 1.8 (with a mean of 0.0). Jones (1996, 1997 tables 5 and 6), who analysed the 1995 survey in which all but one of these items appeared, called this an ‘Australian chauvinism’ factor. However, as the Macquarie Dictionary notes, chauvinism implies a certain aggression or belligerence. We therefore prefer the term ‘national pride’. This scale is clearly more skewed than the nativism scale. CONSTRUCTION OF THE MODEL
The models fitted to the data here are binomial logit models, in which the dependent variable is attitudes to immigration. The three most common approaches to understanding the results of logit modelling in this case are: first, a change in the factor odds of wanting to see a reduction in immigration, compared to a reference category; second, a change in the standardised coefficient of wanting to see a reduction in immigration, compared to a reference category; and third, a change in the predicted probability of wanting to see a reduction in immigration, compared to a reference category. The first and third approaches lend themselves more easily to discussion (since they are directly interpretable), while the second approach (while less interpretable) lends itself to graphing, since the standardisation allows for direct comparisons of magnitude. It is worth noting that the reference category chosen for the factor odds approach is usually the most common group, or a reference group that is most meaningful (for example, university education in relation to attitudes to immigration). The reference categories are also referred to as omitted categories. The factor odds approach uses the language of ‘odds ratios’, which express how much greater the odds of wanting to see a reduction in immigration is for one category compared to its reference category. By contrast, the predicted probability approach is much more straightforward. It provides a simple percentage of respondents predicted to want to see immigration reduced, with all other variables set to their mean. The reason for this last part is technical, and is necessary because the probabilities vary according to different levels of the variables in the model (Long 1997).
AusSocAtts03
26/7/05
8:55 AM
Page 265
APPENDICES
• 265
Table A11.2 Logit model predicting agreement with reduced immigration (odds ratios), 2003
Sex Male Age (ref: 26–44) 18–25 45–64 65 and over Education (ref: Bachelor’s degree or above) Less than Year 10 Year 10 Year 12 Trade TAFE/Diploma Employment background (ref: Professionals) Managers and administrators Associate professionals Tradespersons Advanced clerical and service Intermediate clerical, sales and service Intermediate production and transport Elementary clerical, sales and service Labourers Location (ref: Inner metropolitan) Outer metropolitan Regional Rural Birthplace (ref: Born in Australia) First-generation migrant Second-generation migrant Speaks only English at home Unemployed Nativism scale By males By age 18–25 By age 45–64 By age 65 and over National pride scale Constant McFadden’s pseudo R squared Log likelihood n
Demographics
Nativism
Pride
0.92
0.94
0.92
1.22 0.89 0.80
1.43 0.77* 0.62**
1.23 0.88 0.79
4.98*** 3.49*** 2.52*** 3.60*** 2.57***
3.48*** 2.83*** 2.21*** 2.98*** 2.35***
4.94*** 3.46*** 2.50*** 3.57*** 2.56***
0.91 0.91 1.37 1.42 1.27 1.44 1.11 0.95
0.97 0.89 1.13 1.05 1.12 1.23 1.02 0.83
0.91 0.91 1.37 1.41 1.27 1.45 1.10 0.95
1.38* 1.18 1.87***
1.42* 1.14 1.90***
1.38* 1.18 1.87***
0.51*** 0.90 1.51* 2.05
0.92 1.05 1.57* 2.36 2.61*** 0.83 2.24* 0.99 0.74
0.51*** 0.90 1.51* 2.06
0.14*** 0.09 –1046
0.15*** 0.16 –964
1.03 0.14*** 0.09 –1046
(1758)
(1751)
(1758)
* p < 0.05; ** p < 0.01; *** p < 0.001 SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
266 •
8:55 AM
Page 266
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
Table A11.3 Further logit model results predicting agreement with reduced immigration (odds ratios), 2003 Economic
Political
Values
0.95
0.96
0.95
18–25
1.38
1.15
1.22
45–64
0.75*
0.77
0.78
65 and over
0.61**
0.68
0.83
Less than Year 10
3.36***
3.13***
2.81***
Year 10
2.70***
2.34***
2.21**
Year 12
2.14**
1.84*
1.70*
Trade
2.85***
2.40***
2.04**
TAFE/diploma
2.31***
1.97**
1.87**
Managers and administrators
1.00
0.96
0.94
Associate professionals
0.93
0.85
0.84
Tradespersons
1.10
0.83
0.82
Advanced clerical and service
1.07
0.92
1.00
Intermediate clerical, sales and service
1.14
1.07
1.12
Intermediate production and transport
1.21
0.94
0.94
Elementary clerical, sales and service
1.04
0.87
0.87
Labourers
0.81
0.56
0.62
Sex Male Age (ref: 26–44)
Education (ref: Bachelor’s degree or above)
Employment background (ref: Professionals)
Location (ref: Inner metropolitan) Outer metropolitan
1.43*
1.37
1.32
Regional
1.13
1.14
1.01
Rural
1.85***
1.79**
1.66**
Birthplace (ref: Born in Australia) First-generation migrant
0.93
0.80
0.80
Second-generation migrant
1.03
1.11
1.10
Speaks only English at home
1.66*
1.60*
1.67*
Unemployed
2.12
2.08
1.97
Nativism scale
2.58***
2.71***
2.39***
By males
0.84
0.79
0.75
By age 18–25
2.02
1.87
1.60
By age 45–64
0.97
0.93
1.01
By age 65 and over
0.73
0.65*
0.68
National pride scale
AusSocAtts03
26/7/05
8:55 AM
Page 267
APPENDICES
• 267
Table A11.3 Further logit model results predicting agreement with reduced immigration (odds ratios), 2003 continued Economic
Political
Values
Think unemployment has risen in past two years
1.23
1.18
1.18
Economic alienation
1.86**
1.97**
2.08**
Vote Labor
0.71*
0.80
Vote Democrats
0.24***
0.26**
Vote Greens
0.62
0.71
Vote One Nation
2.37**
2.63**
Vote Other
0.93
0.84
Informal
1.16
1.34
Vote (ref: Coalition)
Low political engagement
1.68**
1.61**
High political engagement
1.17
1.17
Class identification (ref: Middle)
0.76
0.69
Upper
0.76
0.69
Working
1.43*
1.40*
None
1.10
1.09
Agree with death penalty
1.91***
Welfare hostility
1.89***
Religious observance (weekly)
0.55**
Constant
0.13***
0.15***
0.10***
McFadden’s pseudo R squared
0.16
0.19
0.22
Log likelihood
–938
–841
–794
(1713)
(1600)
(1556)
n * p < 0.05; ** p < 0.01; *** p < 0.001 SOURCE The Australian Survey of Social Attitudes 2003
AusSocAtts03
26/7/05
268 •
8:55 AM
Page 268
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
CHAPTER 12 Table A12.1 Logistic regression results predicting agreement with genetic testing for health benefits, 2003 Odds ratios
Male Age (ref: 65 and over) 18–34 35–49 50–64 Education (ref: No university) Bachelor’s degree and above Occupation (ref: All other) Managers and professionals Born in Australia No religion Trusts people Good personal health Constant Nagelkerke R-square = 0.036
1.252* 0.691* 1.095 1.047 1.201 1.115 1.147 1.210 1.505** 1.188 0.911
(n = 1690) * p < 0.05; ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
Table A12.2 Logistic regression results predicting awareness and attitudes to genetic testing in insurance, 2003 Currently used
Disagree with insurers using genetic tests
Odds ratios
Odds ratios
Male Age (ref: 65 and over) 18–34 35–49 50–64 Education (ref: No university) Bachelor’s degree and above Occupation (ref: All other) Managers and professionals Born in Australia No religion Trusts people Low confidence in financial institutions Good personal health Constant Nagelkerke R-square n * p < 0.05; ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
1.090
0.717**
0.713* 0.637** 0.769
2.153** 2.406** 2.130**
0.853
1.382
0.927 0.924 0.956 0.898 1.145
1.070 1.249 1.611** 1.177 1.603**
0.817 1.490 0.036
1.260 1.436 0.079
(1705)
(1714)
AusSocAtts03
26/7/05
8:55 AM
Page 269
APPENDICES
• 269
Table A12.3 Logistic regression results predicting awareness and attitudes to genetic testing in employment, 2003 Currently used Disagree with in employment employers using genetic tests
Male Age (ref: 65 and over) 18–34 35–49 50–64 Education (ref: No university) Bachelor’s degree and above Occupation (ref: All other) Managers and professionals Born in Australia No religion Trusts people Poor relations at work Good personal health Constant Nagelkerke R-square n
Odds ratios
Odds ratios
1.372**
0.563**
0.744 0.599** 0.803
2.657* 2.995** 2.206
0.816
1.559*
0.695** 0.781 0.838 0.766** – 0.731** 0.671 0.055
.993 1.266 1.497* 0.985 1.596* 1.139 2.140 0.064
(1711)
(1113)
* p < 0.05; ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
Table A12.4 Logistic regression results predicting attitudes to genetic testing in immigration, 2003 Disagree with testing immigration applicants Odds ratios
Male Age (ref: 65 and over) 18–34 35–49 50–64 Education (ref: No university) Bachelor’s degree and above Occupation (ref: All other) Managers and professionals Born in Australia No religion Trusts people Reduce immigration Good personal health Constant Nagelkerke R-square (n = 1720) * p < 0.05; ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
0.600** 3.108** 3.399** 1.960** 1.721** 1.410** 0.983 1.347* 1.405** 0.304** 1.367* 0.772 0.242
AusSocAtts03
26/7/05
270 •
8:55 AM
Page 270
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
CHAPTER 14 Table A14.1 Logistic regression results predicting attitudes to economic openness, 2003 (odds ratios) Disagree that economic openness increases job insecurity
Age (year born) Respondent income Bachelor’s degree or higher (ref: Other qualifications) Manager or professional (ref: All other occupations) Lives outside metro area (ref: Metropolitan area) Cannot improve living standards Agree job loss in next 12 months Agree that opening the economy is bad for job security Constant Nagelkerke R-squared n
Disagree that Agree that government free trade leads should limit to better imports to protect products being the economy available
0.993 1.084** 1.762**
0.999 1.000 2.570**
0.990* 1.028 1.229
1.397**
1.162
1.070
0.678**
0.653*
0.936
0.598** 0.692*
0.622* 1.257
0.725* 1.119
0.194** 0.106
0.363** 0.221** 0.128
0.825 1.253 0.025
(2312)
(1210)
(1210)
* p < 0.05; ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
Table A14.2 OLS results predicting attitudes about the power of international organisations, 2003 International organisations are taking away too much power from the Australian Government
Male Age (year born) Respondent income Bachelor’s degree or higher (ref: Other qualifications) Manager or professional (ref: All other occupations) Lives outside metro area (ref: Metropolitan area) Agree that opening the economy is bad for job security Nativism scale Identify as Liberal–National (Ref: Labor) as Green or Democrat as One Nation as No Party or other Constant Adjusted R-squared (n = 1578) * p < 0.05; ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
0.020 0.002 0.025** 0.282** 0.052 –0.007
–0.290** –0.046** –0.107* 0.199* –0.708** –0.182** 2.862** 0.157
AusSocAtts03
26/7/05
8:55 AM
Page 271
APPENDICES
Table A14.3 OLS results predicting attitudes to Asian engagement, 2003 Australia should Australia should pursue greater pursue greater economic ties cultural and with Asia political ties with Asia
Male Age (year born) Respondent income Bachelor’s degree or higher (ref: Other qualifications) Manager or professional (ref: All other occupations) Lives outside metro area (ref: Metropolitan area) Agree that opening the economy is bad for job security Nativism scale Identify as Liberal–National (Ref: Labor) as Green or Democrat as One Nation as No Party or other Constant Adjusted R-squared
–0.214** 0.009** 0.004 –0.249**
–0.102* 0.011** 0.000 –0.321**
–0.076
–0.072
0.038
0.054
0.210** 0.047** 0.141** –0.102 0.156 0.211 1.836** 0.119
0.204** 0.058** 0.250** –0.165 0.467** 0.198** 1.849** 0.142
(1602)
(1608)
n * p < 0.05; ** p < 0.01 SOURCE The Australian Survey of Social Attitudes 2003
• 271
AusSocAtts03
26/7/05
8:55 AM
Page 272
INDEX [References to tables are shown in bold text.] ABC 130, 130, 234 abortion 3, 22–24, 23, see also reproductive choice active membership of organisations 65, 68–70, 69 political participation and 72–78, 73 activism, see political participation Afrobarometer vii age differences and pressure to work harder 54 assessment of crime trends 146, 147 attitudes to abortion 22–24, 23 children and marriage 15, 16 defining a family 13, 13–15 punishment 156 reproductive choice 22–25, 23, 25 social spending 106, 106 the media 228, 229, 230, 232–36, 233, 235, 251 cosmopolitan attitudes 250–51 in material/postmaterial values 85–86, 89–90, 91, 96–97 in political trust 127, 129 Internet access 226 job insecurity 48–49 job satisfaction 45, 46 skill marketability 48–49 social trust 124–25, 126 sources of news and information 225 voluntary association membership 66, 67 aged care, delivery of 116 ALRC/AHEC Inquiry 216 appendices 258–71 arbitration of industrial disputes 163 armed forces, see defence forces Asia East Asian Social Survey vii immigration from 241 engagement with 5, 248, 248–49, 271 asylum seekers, see immigration
AuSSA vii–viii, 2, 6–10, 9 Australia, see also Howard government as welfare state 102–3, 103 attitudes towards 175, 187–90, 189, 190, 191 being born in 187–88, 188 changes in 2–6 crime rates, perceived and actual 142–47, 143, 144, 147 economic reform 161–81 GDP growth 172 globalisation in 240–41, 253 international organisations and 246, 247 Internet access in 225–26 job satisfaction in 43–45 local media support 250–52, 251 mass media in 221, 227, 228 national priorities for 87 political participation in 71–72, 72 regional loyalties 31 sense of national identity 182–203 social spending 102–3, 103 trust 122–40 taxation levels 101 voluntary associations in 62–63, 66 Australian Broadcasting Corporation 130, 130, 234 Australian Democrats supporters attitudes to social spending 110, 111 immigration 196 privatisation 168 Australian Election Study 142, 249 Australian Greens supporters 84 attitudes to privatisation 168 social spending 107, 110, 111 postmaterial values 90, 92 Australian Labor Party governments, see also Australian Labor Party supporters economic reform and 162 immigration policy of 182 Australian Labor Party supporters attitudes to privatisation 168 social spending 107, 110, 111 material/postmaterial values 90, 92 Australian National University 2 Australian settlement values 170, 175
AusSocAtts03
26/7/05
8:55 AM
Page 273
INDEX • 2 7 3
Australian Social Science Data Archive 7 Australian Survey of Social Attitudes vii–viii, 2, 6–10, 9 automobile associations 67, 68, 69, 76–77 award wages 171, 171–72 banks, trust in 130, 130, 131 bias in social surveys 8–9 big business, see businesses Biotechnology Australia 206 births, see reproductive choice boycotting products, see consumers breadwinner families 30–41 attachment to family 32, 32 attitudes to working mothers 33, 34 optimism of 37–40, 39 Britain, see UK British Social Attitudes vii Bush, George W 97–98 Business Council of Australia 84 businesses, see also media; multinational companies attitudes towards big 176–77, 176, 177 role in child care 35, 35 crimes committed by 152–53, 153 criticise immigration policy 184 genetic testing by 205, 209–12, 211, 212, 216, 268 media outlets as 231–37, 233 power of 132–33, 133, 176, 176–77 trust in 130, 130, 131, 132, 139 welfare services provided by 103, 115–18, 116 Campbell Committee 162 Canada, attitudes to international organisations in 247 caregiving, see child care Census 2001, AusSSA compared with 9 Centre for Social Research 2 charities government subsidies of 115 role in child care 34 service provision by 115–18, 116 child care attitudes to 33–36, 34 fathers’ role in 26–27 delivery of 34, 35, 116–17 children, see also families
in defining a family 13–15 a complete life and 20–21, 261 marriage and 15–16, 16 of same-sex couples 18–20, 19 of single parents 17–18, 17 of working mothers 33–35, 34, 57–59, 59 work-family balance and 56 shared parenting of 26–27, 27, 57–58 Christianity, see religiosity; religious organisations churches, see religiosity; religious organisations citizenship rights 101 civic engagement 4–5, 82–100, see also political participation; voluntary associations levels of trust and 133–37 of active association members 73–74 right to, valued 86 civil society, see civic engagement; voluntary associations class identification and attitudes to social spending 106–7, 107 and political trust 129 and social trust 125–26, 126 Coalition, see Liberal/National party cohabitation without marriage 12, 13 commercial television crime as entertainment on 157 information on crime rates 148 news and information from 220, 221–22, 222, 224 community, priority of 178 Comparative Study of Electoral Systems vii competition, ideology of 177–78 confidence in Australian institutions 129–33, 130, 131, see also trust conservatism in Howard government 2–3 and attitudes to immigration 194–5 consumers associations of 67, 68, 69, 76–77, 77 boycotts 63, 71, 71–2, 73, 94, 95, 96, 134, 135, 136 exercise of power by 71–72, 95–96, 244 corporations, see businesses
AusSocAtts03
26/7/05
274 •
8:55 AM
Page 274
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
cosmopolitanism 250–51 courts, see legal system crime 141–60 actual rates of 142–44, 144 committed by big business 152–53, 153, 176, 176 impact of immigration on 145, 145–46, 186, 187, 196, 198 marijuana smoking 149, 149–50, 150 perceived trends in 5, 142–46, 143, 144 victimless 149–50 culture exposure to overseas media and 250, 250–51 impact of immigration on 186, 186 response to materialism in 84–85 de facto marriage 12, 13, 22 death penalty, support for 152–56, 153, 154, 155, 155–56 defence forces attitudes to higher spending on 109, 109 trust in 4, 130, 130–32, 131, 189–90, 190 Democrats see Australian Democrats supporters democratic engagement, see civic engagement demographics of AuSSA sample 8–9 demutualisations 166 deregulation, see labour market deregulation; privatisation; trade deregulation deserving poor 112 Disability Discrimination Act 1992 210 disability services, delivery of 116 disease, genetic detection of 204–6, 207, 210, 213–14, 268 distributive justice 163 divorce attitudes to children and 15–16, 16 fathers’ role after 26–27 initiated by female partner 17 joint custody after 26 rise in rates 12 DNA fingerprinting 204 donations to voluntary associations 63 downshifting 85 East Timor 189–90, 246
economic alienation 194–96, 195 economic issues, see also globalisation; labour market deregulation; taxation; trade deregulation attitudes to immigration and 193–94 contribution of voluntary associations 63 economic reform 161–81 GDP growth 172 growth and stability as 86, 87 hydraulic economic model 179 impact of immigration 186, 186–87, 196, 198 importance of 175 income and wealth distribution 172–75, 174, 175 openness of the Australian economy 242–45, 243, 270 pride in achievements 190, 190 privatisation 164–68, 165, 166, 167, 168 redistribution of income and wealth 111–15, 112, 114 ties with Asia 248, 248–49 work obligations for unemployed 111–15, 112, 113 economic nationalists 170 education attitudes to higher spending on 107–9, 109 delivery of 116, 116 standards in 107–8, 108, 118 education levels assessment of crime trends and 146–47, 147 attitudes to defining a family and 14 immigration and 191–92, 192 punishment and 156 reproductive choice and 22, 24, 25 cosmopolitan attitudes and 250–51 Internet access and 226 material/postmaterial values and 90, 91 political participation and 64–65, 76, 76 political trust and 128, 129 protectionism and 244–45 social trust and 126, 138 sources of news and information and 223, 224 television viewing and 227, 228
AusSocAtts03
26/7/05
8:55 AM
Page 275
INDEX • 2 7 5
trust in media and 232, 233 voluntary association membership and 67, 67 Elliot, John 185 employment, see also labour market deregulation; unemployment; working hours and conditions employee relations with management 43, 44, 47 growth in 113 job satisfaction 43–48, 44, 46, 56–57 engagement, see civic engagement; political participation; voluntary associations entertainment, news and information as 220, 226–28, 228 environmental issues, see also postmaterial values activist groups 74 attitudes to higher spending on 109, 109 importance of 82–83, 83 political participation in 94–95, 95 sustainability 84 Esping-Andersen, Gøsta 1 ethnic minorities 185, 185–86 Europe attitudes to genetic testing 208, 216 international organisations in 246–47 local media content 252 trade deregulation 245 Eurobarometer surveys vii, 216 European Social Survey vii evidence in policy-making 1 exnuptial births 12 export levels 242 families 12–41 central to identity 3, 31–33, 32 definition of 13–21, 13, 16, 17, 19, 27–28 father’s role in 26–27, 27 genetic testing and 205, 210 living standards and quality of life 37–40, 39 long working hours and 31, 36–37, 42, 51-55, 52, 53, 56 role in child care 33–36, 35 work-life balance and 30, 33–34, 57–59 family payments 102–3
family reunion immigration 182, 214 fathers, shared parenting by 26–27, 27, 57–58, see also families Federal Government, see governments; Howard government fertility treatments 21–22, 24–26, 25 financial co-operatives 67, 68, 69 financial institutions, trust in 130, 130, 131 financial issues, see economic issues; income FitzGerald Report 184–85 foreign investment 243 free expression, right to 86 free market, see economic issues; trade deregulation free press 232 free trade, see trade deregulation Garnaut Report 241 Garrett, Peter 82 gay marriage, see same-sex couples gender differences attitudes to abortion 23–24, 23 children in marriage 15 genetic testing 209 punishment 156 single-parent households 17 violence on TV 229, 230 working mothers 58–59, 59 hours worked 51 in assessing crime trends 146, 147 in defining a family 13–15 in job insecurity 48 in job satisfaction 45 in political trust 128, 129 in skill marketability 48 in social trust 125 in sources of news and information 224 in voluntary association membership 66, 67 postmaterialist view of 93 General Social Survey (USA) vii genetic testing 6, 204–19, 209, 268, 269 globalisation 5, 240–57, 253, see also economic issues; trade deregulation governments, see also Howard government; social services economic reform and 161–81 genetic testing by 205–6, 214–15, 215
AusSocAtts03
26/7/05
276 •
8:55 AM
Page 276
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
income and wealth redistribution by 114–15, 114 power of 132–33, 133 role in child care 34, 35 services delivered by 101–3, 103, 115–18, 116 supranational controls on 245–48, 247 trust in 126–28, 129, 128, 132 watchdog role of media and 234–37, 235, 236 Great Britain, see UK Greens party supporters, see Australian Greens supporters Gregory, Bob 172 group identification 31, 31–33, 32, 92–93, 93 health care attitudes to higher spending on 107–9, 109 standards of 107–8, 118 genetic detection of disease 204–6, 207, 210, 213–14, 268 delivery of 116, 116 health insurance 103 Hilmer Inquiry 163 Hochschild, Arlie 55 homosexuality, see same-sex couples Howard government 2–6, see also labour market deregulation; trade deregulation Asian engagement by 248 immigration policy of 182–85, 199 social spending during 102–4 same-sex couples, recognition by 19–20 family payments system 102–3 reluctance to cede to supranational power 246 work obligations for welfare recipients 112 Howard, John, see also Howard government attitude to Asian engagement 248 attitude to multiculturalism 185 election about ‘trust’ 122 favourite Midnight Oil song 82 Human Genetics Commission (proposed) 217 Human Genome Project 204 human services, see social services immigration 182–203
asylum seekers 183 attitudes to level of 3, 184 from Asia 241 perceived effect on crime rates 145, 145–47, 146 perceived impact of 186, 186–87 use of genetic testing in 205–6, 214–15, 269 import levels 242 import restrictions, see trade deregulation income, see also taxation; wage setting systems; attitudes to immigration and 195 household types and 37–40 inequality in 172–75, 174 levels of, and Internet access 226 national shift from wages to profit 164 political trust and 128 redistribution of wealth and 175 skill use and 47, 47 slowdown in growth 172 social trust and 126 voluntary association membership and 67, 67 individualism 30 industrial relations, see labour market deregulation Inglehart, Ronald 82–100 institutions, see businesses; organisations insurance companies, genetic testing by 205, 209–12, 211, 212, 216, 268 internal political efficacy 65 international comparisons attitudes to globalisation 253 social services spending 103, 253 trade exposure 253 international organisations, see supranational controls on governments International Social Survey Program (ISSP) 6 Internet as source of information 148, 222, 223, 223, 225–26 interpersonal trust, see social trust IVF (in-vitro fertilisation) 21–22, 24–26, 25 job insecurity 4, 43–44, 44, 48–49 Asian engagement and 249 and globalisation 242–43 by occupation 50 job satisfaction 43–48, 44 working hours and 56–57, 57
AusSocAtts03
26/7/05
8:55 AM
Page 277
INDEX • 2 7 7
John Fairfax Ltd 231 judges, see legal system justice, see crime; legal system Kennedy Royal Commission 151 Labor party, see Australian Labor Party governments and Australian Labor Party supporters labour market deregulation 6, 42, 163–64, 171–72 active labour market programs and 254 land purchases by non-Australians 243 Latham, Mark 82, 123 legal system sentencing by 142, 151–56, 153, 154, 155 trust in 130, 130, 131, 150–54 Liberal/National parties, see also Howard government attitudes to economic reform 162 Liberal/National party supporters attitudes to higher social spending 107, 110, 111 material values 90, 91 living standards 37–40 best way to improve? 177, 177–78 concern for quality of life 84–85 inequality in 172–75 perceived, by family type 37–40, 39 local content rules in media 250–52 long working hours, see working hours and conditions mail surveys 8 management, employee relations with 43, 46–48, 46, 47, see also working hours and conditions managerialism 162 marijuana, attitudes to use 149, 149–50, 150 marketability of skills 43–44, 49–51, 50 marriage attitudes to, by gender 13, 13–14 attitudes to children and 15–16, 16 in defining a family 13, 20, 28 marital status and identity 31–33 median age at 12 of same-sex couples 19–20 reproductive choice and 24–26 mass media, see media
mass participation organisations 70 material values 84–92, 87, 88, see also postmaterial values McBain, John 21–22 media accuracy of information on crime rates 148, 148 as a source of information on crime 147–48 attitudes to 5 local, attitudes to 250–52, 251 power of 132–33, 133, 231–37, 233, 236 watchdog role and government 234–37, 235, 236 medical research, awareness of 206–7, 207 Medicare 107–8, see also health insurance membership of organisations 62–81 of environmental organisations 94–96, 95 trust and 130–37, 137 men, see fathers, shared parenting by; gender differences migration, see immigration mothers 12, 33–36, see also families multiculturalism 3–4, 182–203, see also ethnic minorities multinational companies 243 National Health Act 1953 209 national identity 182–203, 191 National party, see Liberal/National parties national pride 189–90, 190, 198–99 attitudes to immigration and 193 national identity and 189–90 national security, see defence forces nativism 183, 191 attitudes to immigration and 193–97, 195 multiculturalism and 196–97, 197 international organisations and 247–48 neo-liberal attitudes 6, 175–78, 254, 255 see also economic issues neo-traditional households 31–33, 36–38 neoclassical economics 161, see also economic issues New Zealand, attitudes to international organisations in 247
AusSocAtts03
26/7/05
278 •
8:55 AM
Page 278
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
news and information as entertainment 220, 226–28, 228 sources of 148, 221–25, 222, 223, see also media News Corporation Ltd 231 news magazines 222, 223 newspapers 221, 222, 224 non-responses in social surveys 10 not-for-profit organisations, see charities, voluntary associations occupational differences, see also professionals and working hours problem 53–55, 54 contribution to identity 31–33, 32 in attitudes to immigration 191–92 in attitudes to working mothers 58, 59 in job insecurity 49, 50 in job satisfaction 45–46, 46 in material/postmaterial values 90, 91 in political participation 76 in skill marketability 49–50, 50 in voluntary association membership 67, 67 in workplace relations 47 occupational health and safety, see working hours and conditions OECD nations, see international comparisons Offe, Claus 1 officeholders, see active membership of organisations One Nation party, see also One Nation party supporters attacks political elites 253 trade policy 241 One Nation party supporters attitudes to privatisation 168, 168 hostility to immigration 195 postmaterial values 90, 91, 92 optimism and pessimism, see also confidence; trust about living standards 173, 173 economic confidence 4 feelings of prosperity 4 job insecurity 43–44, 44 of families 38, 39 organisations, see also businesses; charities; unions; voluntary associations trust in 130, 130–33, 131 overwork syndrome 53
Packer family 231 parenting, shared 26–27, 27, 57–59 part-time workers 52, see also neotraditional households party identification, see also names of parties, e.g. Australian Labor Party identifiers attitudes to Asian engagement 249 community 178 globalisation 254, 255 immigration 195 privatisation 167–68, 168 social spending 107, 107, 110–11, 111 international organisations 247 material/postmaterial values 89–92, 91 political participation 194 shift to minor parties 85 social service delivery 117 Pell, George 19 personal life, work interfering with 55–56, 56, see also families Phelps, Kerryn 19 Pocock, Barbara 55 police confidence in 130, 130–32, 131 crime rates recorded by 144 perceptions of corruption in 150–52, 152 satisfaction with 150–52 political associations 63–64, see also party identification political participation 4–5, 62–81, 71, 72, 135, see also civic engagement attitudes to immigration and 194 by members of voluntary associations 72–78, 75, 76 levels of trust and 133–37, 135 material/postmaterial values and 93–96, 95 new approaches to 97 political parties, see also party identification; political party names e.g. Australian Greens political trust 70, 126–29 by demographic group 129 levels of 127–28, 128 political participation and 134–37, 135, 136 voluntary group membership and 136–37, 137 postmaterial values 5, 82–100, 87, 88
AusSocAtts03
26/7/05
8:55 AM
Page 279
INDEX • 2 7 9
power, of major Australian organisations 132–33 pressure to work harder 44, 44, 52, 54 pride, see national pride; optimism prisons, attitudes to privatisation 166 privatisation attitudes to 6, 164–68, 165, 166 effects of 162–63 of welfare delivery 115–18 Telstra and 162–63, 164–65, 165, 166 professionals job security of 51 membership of professional societies 68, 68, 69, 74 and work pressure 53 protectionism, see trade deregulation protest activity 134, 135 public policy and social research 1–2 public sector, see also privatisation; social services; welfare state broadcast media in 221, 234 standards in 107–8, 108 deregulation of 162 internal reform of 162 trust in 130, 130, 131, 132 workers in 49–50, 50, 91 public transport 166 punishment, attitudes towards 152–56, 153, 154, 155 Putnam, Robert 78, 122–23 quality of life, see living standards radio stations 221–22, 222 recreational associations 67, 68, 69 redistribution of income and wealth, see economic issues regional location attitudes to decriminalising marijuana 150, 150 immigration and 192–93 the police by State 151 contribution to identity 31, 31 perceived crime trends by State 143, 144–45 political trust and 129 protectionism and 245 social trust and 126 voluntary association membership and 66, 67 religiosity attitudes to
abortion and 22–23 children and marriage and 15–16, 16 defining a family and 14–15 immigration and 195, 196 reproductive choice and 22–26 Bush supporters 97–98 material/postmaterial values 90–91, 91 political participation and 65 religious organisations and national identity 188, 188 membership of 67, 68, 69 political participation by members 74–75, 75 trust in 130, 130, 131 reproductive choice 21–26, see also genetic testing abortion 3, 22–24, 23 access to IVF for married and unmarried women 24–26, 25 exnuptial births 12 genetic testing and 207 rights to 260–61 response rates 7 retrenchment, rates of 43 right to choose, see reproductive choice rights promotion groups 74 Rodrik, Dani 255 same-sex couples 18–20, 19, 21, 261 sampling methods in AuSSA 2003 6–7, 9 SBS 234 science, interest in 206 sentencing, see legal system Sex Discrimination Act amended 22 shared parenting 26–27, 27, 57–58 single-parent households 17, 17–18 attachment to family 32 attitudes to defining a family and 17–18, 17 less time pressure on 36 pessimism of 38–39 skilled migrants 184 skills marketability of 43–44, 49–51, 50 not used at work 43, 44, 46–47, 47 pressure to work harder and 53–54 social capital 122–23, 134, see also trust social complexity 10 social conservatism and attitudes to immigration 194–95
AusSocAtts03
26/7/05
280 •
8:55 AM
Page 280
A U S T R A L I A N S O C I A L AT T I T U D E S : T H E F I R S T R E P O R T
social economy organisations, see voluntary associations social research and public policy 1–2 social services, see also welfare state attitudes to spending on 103–11, 105–111, 167 international comparisons 253 preferred deliverers of 115–18, 116 social trust 70, 71–72 levels in Australia 123–26, 125, 126 political participation and 134–35, 135 voluntary group membership and 136–37, 137 sporting or recreational associations 67, 68, 69 sports achievements, pride in 190 States, see regional location status, see also class; income strike activity 134, 135 supranational controls on governments 245–48, 247, 270
by Howard government 241 import and export levels 242 international comparisons 253 local media content rules 250–52 party identification and 254, 255 tariff reduction 163 ties with Asia 248, 249 vs. protectionism 244–45, 270 tradespeople, attitude to working hours 54 transfer payments, see welfare state trust 122–40, see also confidence; political trust; social trust attitudes to genetic testing and 212, 216 during Howard government 4 in governments, media’s role in 234–37, 235, 236 in media outlets 232 social capital and 122–23, 134 voluntary association membership and 70–71, 70 two-earner families, optimism of 38
talkback radio 148, 148, 222, 222 tariffs, see trade deregulation tax and spend scales 104, 105, see also social services; welfare state attitudes to social spending and 109, 109–10 attitudes to service delivery and 117–18 party identification and 111 taxation, see also economic issues; tax and spend scales changes in 102 willingness to pay higher taxes 108–9, 109 cuts in vs. social spending 104–8, 105, 106, 107 technology, interest in 206 television, see also commercial television ABC 130, 130, 131, 234 attitudes to 5 entertainment and 226–29, 228 local content rules 251–52, 251 SBS 234 violence on 5, 229–31, 230 Telstra privatisation 162–63, 164–66, 165, 166 third sector, see voluntary associations trade deregulation 163, 243, 243 attitudes to 5, 168–70, 169, 170, 244
UK attitudes to genetic testing 208, 211, 213 import restrictions 245 marriage 14 international organisations in 247 Internet access in 225–26 participant political culture 64 British Social Attitudes vii social trust in 124 working mothers and childcare 35–36 UN Conventions, failure to meet 246 unemployment attitudes to immigration and 184, 193–94 effect of immigration on 186, 186–87, 198 perceived as voluntary 113, 113–14 unions, see also wage setting systems membership of 68, 68, 69 postmaterial values 91 power of 132–33, 133 trust in 130, 130, 131 United Kingdom, see UK United States, see USA universities, trust in 130, 130, 131 unmarried couples, see cohabitation
AusSocAtts03
26/7/05
8:55 AM
Page 281
INDEX • 2 8 1
without marriage unmarried (single) parents, see singleparent households unmarried women, access to IVF 24–25 USA attitudes to genetic testing in 208, 216 attitudes to import restrictions in 245 attitudes to welfare in 111 General Social Survey vii immigration and crime rates 145 income inequalities in 174 attitudes to international organisations in 247 political culture 64–65, 74–75 religiosity and politics in 97–98 social trust within 124 values, learning 85, 89, see also postmaterial values violent crimes actual incidence of 142 on television 229–31, 230 voluntary associations 4–5, 62–81, see also charities; civic engagement government subsidies of 115 member commitment 69 membership by type 68 political participation by members 72–79, 73, 75, 76, 77 postmaterialist values and 94–95, 95
trust levels and 136–37, 137 wage setting systems 171, 171–72 Wallerstein, Immanuel 2 watchdog role of media 232, 234, 235, 236, 237 weighting in AuSSA 2003 9 welfare state 101–21, see also social services attitudes to welfare 101–21, 109, 112, 114, 194 offsets income inequality 174 pride in social security system 190 work obligations for welfare recipients 112, 112–14, 113 women, see gender differences; mothers work, attitudes to 42–61 work obligations for welfare recipients 112, 112–14, 113 working hours and conditions 4, 37, 42, 53, see also occupational differences conflict with family life 30, 33–34, 36–37, 42, 55–59, 56 excessive hours 51–55, 52, 53, 54 genetic testing at work 205, 212–14, 213, 214, 216, 269 job satisfaction 46, 45–48, 56–57, 57 Workplace Relations Act 1996 162 World Court, refusal to recognise jurisdiction of 246 World Trade Organization 246 World Values Survey vii, 6, 85–86, 124
AUS T R AL IAN S OCIAL AT T I T UDE S
Australian Social Attitudes: The First Report offers a new level of insight into the dynamics of Australian social life. Its 14 chapters develop an in-depth and accessible understanding of how Australia is responding to the new realities of work, globalisation, family and community life. The book draws on the latest social research and analysis from many of Australia’s leading social scientists to challenge conventional wisdom about Australia, and assesses the impact of the Howard government on the attitudes and behaviour of Australians of all ages and social backgrounds. Australian Social Attitudes: • is the fi rst major report on what Australians think about topics as diverse as the use of human genetic information, what makes a family, the costs and benefits of immigration, and crime and punishment • combines analysis of public opinion and social attitudes with reported behaviour and policy developments • compares Australian attitudes to those in other countries. Australian Social Attitudes will prove an excellent resource for students, teachers, researchers and policy makers, and anyone interested in understanding the social dynamics of contemporary Australia.
AUS TR ALIAN SOCIAL AT TIT UDES THE FIRST R E P O R T
edited by WILSON
UNSW PRESS ISBN 0-86840-671-6
MEAGHER GIBSON DENEMARK WESTERN
� 9 780868 406718 �
ASAcover2.indd
1
Shaun WILSON Gabrielle MEAGHER Rachel GIBSON David DENEMARK Mark WESTERN EDITED BY
UNSW PRESS
22/7/05
11:56:42 AM