A Crisis in Welfare Legitimacy? A review of citizens’ support for the welfare state in times of change
“Attitudes are coins that do not readily melt” (Schumpeter, 1942, p.12).
“Good Science is not futurology; we should always be humble about predicting the path ahead on the basis of what we know about the one already trodden” (Pawson, 2006. p. 1)
Trude Diesen Sundberg
PhD Social Policy 2013
Word count: 75,660
Acknowledgement
Completing a PhD is a chapter in life, full of challenges, valuable insights and above all, hard work. This thesis project would not have been possible without the guidance, patience and support of my supervisors, Professor Peter Taylor-Gooby and Professor Miri Song. They have both provided exceptional support and challenging discussions without which I could not have completed the project.
Of course, the professional side is one part, but I would also like to thank all my friends and loved ones who have supported and put up with me over the last four years. You know who you are, thank you each and every one of you for allowing me to see and experience all the unique, fascinating and curious aspects of life.
Abstract:
As governments in advanced capitalist countries respond to pressures on their welfare states by reforming their welfare systems, many scholars have asked how this impacts the legitimacy of welfare states.Pressures are both internal and external at social, political and economic levels. The pressures and governments’ responses to them have led many to ask how far citizens continue to support government intervention in areas of Social Policy, and to what extent these changes may impact the stability of welfare states. The number of cross-national surveys is steadily increasing and many studies examine developments in social provision and public attitudes towards welfare states. However, there is a lack of a clear and comprehensive overview of knowledge about tendencies in support for the welfare state in light of these pressures. This thesis uses tools from the systematic review tradition and draws on principles and tools from realist synthesis to achieve a comprehensive and systematic review of current knowledge on support for the welfare state in light of these pressures in Europe, the USA, Canada, Australia and New Zealand. Thus the project makes an original contribution to knowledge through assessing the use of an innovative method in Social Policy, as well as through shedding new light on the trends and dynamics at play in citizens’ support for the welfare state in a time where welfare state are facing many challenges.
The project maps current knowledge, and explores ways in which tools from systematic review and realist synthesis may be used in the Social Sciences. It is argued that welfare state attitudes as an area, is wide-ranging and diverse, something which causes problems for the emphasis in the Systematic Review tradition on replication, homogeneity and hierarchies. To illustrate ways in which systematic review tools may be used in a successful way in Social Policy, this thesis further include in-depth analyses of the area of immigration and redistribution. These illustrations show how systematic review tools may be used to further our understanding trends and explanations of support for the welfare state.
The thesis argues that the adapted and innovative approach to review taken here can help us not only identify what we know and don’t know, but can help inform and suggest ways in which we can further increase our understanding of the dynamics in the area. This is exemplified in the thesis through the finding that further attention to operationalization and to the difference between values and attitudes is needed. This is supported by findings across included studies, as there are differences in studies’ conclusions depending on how support for the welfare state is measured. These two aspects are identified as important contributions to further knowledge in the area. It is argued that this innovative approach also enhances our understanding of individual-level trends and their relationship with macro factors such as immigration. Findings suggest that trends at individual-level are different depending on which measure of support for welfare states is used, whilst changes and correlations with macro factors vary depending on different levels of generality of values and attitudes. Scholars’ arguments relating to the direction and impact of changes in macro factors such as the current economic recession must take into account the level of generality of values and attitudes.
Contents
Chapter 1.Introduction
Chapter 2.Theoretical framework
Chapter 3.Methodology: Research design and considerations
Chapter 4.Quality appraisal: Separating ‘bad’ quality research
from ‘good’ quality research
Chapter 5.Overall findings
Chapter 6.A question of redistribution
Chapter 7.Support for the welfare state in ethnically diverse welfare states
Chapter 8.Conclusion.
Bibliography
Appendices
Tables and Figures
Table 2.1Overview of factors influencing support for the welfare state
Table 3.1Realist synthesis: an introduction (Pawson et al., 2004, p. 29)
Table 3.2Quality criteria
Figure 3.1CRD (2009, p. 51).
Table 4.1Immigration studies with Quality issues
Table 4.2Quality criteria
Figure 5.1Top 10 countries covered in Web of Science across disciplines
Figure 5.2Comparisons of 3 large Social Science databases
Figure 5.3Top 10 countries covered by Scopus across disciplines
Table 5.1English, German and French Titles
Table 5.2Inclusions, exclusions and duplicates
Table 5.3First group’s outcomes and findings
Table 5.4Dependent variables - Pensions
Table 5.5Dependent variables - Unemployment
Table 5.6Mixed studies dependent variable measures
Table 5.7Government responsibility dependent variables
Table 6.1Redistribution measures used
Table 6.2Overview of Research questions
Table 6.3Directional effects of macro variables on redistribution
Table 6.4Micro variables directional effects on redistribution
Figure 7.1Immigration and ethnic diversity measures used
Figure 7.2OECD statistics on Foreign born population
Table 7.1Immigration measures
Table 7.2Directional Effects of key variables on welfare support
Table 7.3Findings on different levels of generality
Chapter 1. Introduction
Background
Governments in advanced capitalist countries are dealing with a range of changes stemming from internal and external pressures at social, political and economic levels. These include demographic and labour market changes as well as more contested trends such as increased immigration and family changes. As Taylor-Gooby (2004) claims, these “present formidable challenges to the current welfare settlement” (Taylor-Gooby 2004, p. 1). In light of these challenges, some argue that welfare states need to renegotiate policies and their social contracts with their citizens to ensure welfare states are sustainable. Parallel to, and maybe in response to the mounting pressures faced by welfare states and the increased importance given to understanding citizens’ support for welfare states, an increasing number of cross-national surveys have been established. As a result a vast amount of studies have been conducted, analysing both welfare states’ responses and attitudes towards welfare. At the same time there has been considerable technological developments enabling rapid structured systematic reviews. However, there is a lack of a systematic and comprehensive overview of tendencies in support for the welfare state in light of these pressures, an issue this project aims to address.
Through a critical evaluation of the use of tools from the systematic review this thesis takes an approach that also draws on principles from realist synthesis theory. Overall, the thesis takes a middle position between advocates and critics of the systematic review approach, and argues that systematic review is of value in providing an overview of research and also in tracing issues in the way knowledge develops within an academic field. In other words it is argued that tools from the systematic review tradition are useful in helping to achieve a comprehensive overview of current knowledge. This is the case if authors and scholars take into account and adapt these tools to ensure that a review values the heterogeneity and particular features of the Social Sciences. Furthermore, it cannot substitute theory, nor is it “a royal road to knowledge. It is simply a convenient technique for the review of large numbers of studies” (Sundberg & Taylor-Gooby, 2013, p. 421).
Introducing the project
The project contributes to knowledge in several ways. Firstly, it contributes to knowledge by embarking on a mapping exercise of what is known and not known, identifying causal mechanisms found in current knowledge about the relationship between support for welfare states and welfare pressures. Second, the use of systematic review tools enables a critical analysis of operationalization and a focus on the importance of distinguishing between the key concepts, of values and attitudes. The project includes studies focusing on Europe, and also the USA, Canada, Australia and New Zealand when compared to European countries. The project contributes to better theoretical explanations of micro-level variations and changes over timeby differentiating between levels of generality of values and attitudes, shortcomings of the field up until today. Third, the project is forward looking as it develops and critically assesses a new approach (adapting tools from the systematic review tradition) that may have potential for use by others. In other words, whilst criticising a traditional systematic review approach, the thesis can be argued to sit in the middle between a traditional systematic review and its critics. The thesis includes in-depth analyses of trends and explanations within the areas of redistribution and immigration. This is done so as to illustrate ways in which systematic review and realist synthesis tools can be used in the area, and build patterns of explanations expanding existing theoretical frameworks outlined in the theory chapter. Thus, the focus on redistribution and immigration are used to exemplify new methods that can help fill the gap in current research.
Although it is argued that systematic review techniques are relevant and may help enhance academic knowledge, a range of issues related to the use of systematic review in the Social Sciences have been identified based on the findings. These include aspects related to some of the basic methodological principles of systematic review such as homogeneity and a predetermined research processes, and practical issues related to the diversity of the Social Sciences. These needs to be taken into account to successfully carry out a systematic review in the area. Moreover, it is argued that by building on a fuller knowledge of what is known and not known within and across studies better theoretical perspectives can be constructed which can then be tested empirically.
As will be seen in the thesis, there are high levels of support for some government intervention in the area of welfare, but the level and explanations of support depend on the measure used. A need for more emphasis on and critical discussion of operationalization as well as greater importance given to the level of generality (by better differentiation between values, attitudes and beliefs) are thus two important findings of the thesis. In other words further value needs to be given to what attitudes are, and how they are measured. Furthermore, it is argued that the relationship between explanatory factors at micro and macro level and changes in support over time needs further exploration. Neither of these has been studied to sufficient level for this thesis to draw any significant conclusions about them.
Tools from the systematic review tradition are used with the purpose of achieving a comprehensive selection and review of conducted studies in a timely manner. However, realist synthesis approaches are also used as iteration and heterogeneity is argued to be important to appreciate the diversity of studies found in the area. It is clear that the scope of the project is wide ranging, an aspect that will be taken into account in every stage of the process to ensure that it is manageable within the timeframe without excluding too many studies so as to retain the coverage of studies within the area.
Aims and scope
Summary of the study
Overall this project aims to contribute to knowledge by identifying not only trends in current knowledge about support for welfare states in Europe, and USA, Canada, Australia and New Zealand when compared to European countries but also contribute to discussions on how systematic review and realist synthesis can be used in the Social Sciences. Due to the vast number of areas covered by attitudes to welfare, as well as the importance of understanding variations in citizens’ support for the welfare state, the project analyses support for the welfare state rather than welfare state attitudes more broadly. Only cross-national studies published/written in English between 1994 and 2010 will be included. A discussion of what challenges the welfare states face and approaches to support for the welfare state will follow in the literature review. The research aims of the project are:
Research objectives
- Gauge knowledge (what we know and don’t know) and identify the explanatory power of different theories found in the area of support for the welfare state related to the pressures developed welfare states in Europe, USA, Canada, Australia and New Zealand are facing.
- Identify and develop hypotheses and potential new approaches produced through the use of an adapted systematic review methodology (include knowledge produced by the quality appraisal of current studies)
- Analyse and provide new explanations of the relations between immigration and support for the welfare state as well as trends in attitudes towards redistribution as a measure of support for the welfare state, by using new approaches and hypotheses from the findings from research objectives 1 and 2 and carrying out in depth meta-analysis.
Time coverage
The restriction to studies published/written between 1994 and 2010 has been chosen based on various factors. Firstly, the year 1994 was chosen as it was the year the Commission of the European Community published their White Paper on Social Policy which “set out the main lines of action at the Union level for the coming years” (COM 1994, p.1). Both this White Paper and the European Council held in Essen the same year put unemployment as the top priority for the European Union’s Social Policy (COM 1994, p. 4; Taylor-Gooby 2004a, p. 188). The Commission “stressed its own role as an agent for mutual policy learning (Taylor-Gooby 2004a, 186) with the main focus being creation of employment. Thus, with Europe being amongst the main geographical spaces studied, EU policies are seen from 1994 to create a separate set of pressures on national Social Policy, and thus set the scene for understanding support for the welfare state within the context of specific countries, the region as a whole and the possible influence of regional institutions’ policies.
Secondly, by choosing 1994 rather than a later date we also allow for more waves of surveys to have been conducted, together with more areas covered as surveys may emphasise different topics in each wave. The International Social Survey Programme (ISSP) is an example of this approach, with surveys focussing on different themes in different years i.e. the first wave on the role of government, which is very important for this study, was conducted in 1985, the second wave in 1990 and the third in 1996. The time limit has also been chosen to allow for information having been accumulated, permitting comparisons of tendencies over time. This also enables more information to be included from surveys with infrequent waves such as the European Values Survey, which conducted the interviews for its first wave in 1981, second in 1990 and third in 1999. Fourthly, by choosing 1994 we allow time for authors and theories to react to Esping-Andersen’s (1990) canon within the welfare state area with his focus on the welfare state regimes which can be used as a starting point for evaluating whether or not type of welfare state institutions influence attitudes to social welfare. Lastly, a time limit had to be chosen to contain the amount of information in the study to a manageable amount. The end point was set to 2010. This year was chosen to ensure that enough time was left to assess changes over time from the starting point, but also to make sure that the study was able to include all studies up until the end point. If a later end-point had been chosen updated searches across all databases would have had to be carried out.
Geographical areas covered
The area of study of this project is limited to European countries, and USA, Canada, Australia and New Zealand when compared to one or more European countries. The inclusion of further areas would create problems of comparability since certain geographical areas (especially Latin-America, Africa and areas of Asia) are not covered the in major databases and included in the large cross-national surveys. There are further reasons for this choice of geographical coverage, all ensuring that the project will provide a secure academic foundation for later more extensive work.
Firstly, as the focus of this study is support for the welfare state the countries were chosen as the original focus of both theories of welfare states and the origin the welfare state institutions themselves that were Western Europe and the Anglo-Saxon countries. As an example, both Esping-Andersen, with his categorisation of welfare state regimes as liberal, corporatist and social democratic, (Esping-Andersen, 1990, pp.26-27) and the ‘varieties of capitalism’ approach (Hall & Soskice 2003) focus on these countries, underlining the importance given to these geographical areas by different schools of thought. There is substantial knowledge produced around welfare state regime categories, and also about attitudes within the welfare states. However the latter has been substantially less categorised. As a result, this project may compare the established knowledge about welfare state regimes in Europe and Anglo-Saxon countries with a new categorisation of attitudes to welfare based on the same geographical area. This might help create greater understanding of tendencies in the societies studied, by analysis of contextual factors such as welfare states regime types themselves as well as developing the theories and typology in the area further.