The Institutional Logic of Welfare Attitudes

How Welfare Regimes Influence Public Support

Christian Albrekt Larsen

• Publisher: Routledge
• Year: 2016
• Topic: Social Sciences
• eBook ISBN:
  9781317027485
• Subtopic: Sociology
• Index: Social Sciences

Why are people who live in liberal welfare regimes reluctant to support welfare policy? And conversely, why are people who live in social democratic welfare regimes so keen to support it? These core questions lie at the heart of this intriguing book. By examining how different welfare regimes influence public support for welfare policy, the book explores the institutional settings of different regimes and how each produces its own support. While previous studies in this field have failed to link the macro-structure of welfare regimes and the micro-structure of welfare attitudes, this book redresses this problem by combining welfare regime theory and literature on deservingness criteria alongside empirical evidence from national and cross-national data. While recent trends in welfare state development such as cuts in benefit levels and increased use of targeting, combined with increased immigration, might very well influence our perceptions of the deservingness of the needy, this book provides a strong, convincing and provoking argument that challenges the micro-foundation of present comparative welfare state theory. The result is an important work for all studying and working in the fields of public policy and social welfare.

Chapter 1

Introduction

The development of large and generous welfare states is one of the most remarkable characteristics of Western countries in the period from the end of the Second World War to the mid 1970s. One can speak of a golden-age where the combination of capitalism and social protection almost came to represent the culmination of Western civilisation. Beneath this overall trend – in the 1970s, it was sometimes described as a necessary condition for advanced capitalism – one finds substantial variations between the Western countries. One of the most accepted explanations of this variation in the size and the principle of welfare states is the so-called power resource theory (Korpi 1983). The claim is that it very much depends on the strength of the labour movement and its ability to form coalitions with the rising middle class. However, the welfare optimism of the golden-age was interrupted by the oil crises in the 1970s and the following decades, with high levels of unemployment in most Western countries. The immediate result was reduced tax revenues and increased expenditures, which led to fiscal pressure on the welfare states. Combined with increased economic integration, which limited the possibilities to pursue expansive macro-economic policy, low productivity growth, and an increased number of old people, scholars came in the 1990s to speak of a period of permanent austerity (e.g. Pierson 2001b). In this present phase of welfare state retrenchment or restructuring, a number of scholars have argued, Pierson being the pioneer, that we are witnessing a so-called ‘new politics’ of the welfare state. This ‘new politics’ is centred around a conflict between a constituency in favour of the welfare state and a policy elite forced to take necessary decisions (1994, 1996, 2000, 2001b). The resistance to change is basically explained by the classic argument of concentrated costs (those who are to lose benefits, services or public employment) versus dispersed benefits (a balanced public budget). The end result is that politics under these conditions becomes an art of ‘blame avoidance’.

There is, without doubt, a lot of truth in these ‘macro stories’ of welfare state development and this book will follow the comparative, and especially the institutional, approach taken in this literature. However, one of the major problems with these ‘macro stories’ is that they have rarely paid serious attention to the micro-level foundation of their causal reasoning. Basically, the power resource theory claims that the electorate is guided by (long-term) class-interest and the ‘new politics theory’ claims that the electorate is guided by (short-term) self-interest; but, as we shall see, the empirical evidence actually points in another direction. Respectively modelled around class-interest and self-interest, both theories also claim that the welfare state institutions in place influence the future preferences of the electorate. This line of reasoning helps to explain why the restructuring of welfare states (even in a situation where they face very similar external pressures) followed distinct paths. Many current discussions of welfare state development relate to the notion of path-dependency. But again we shall see that empirical studies have had difficulties in finding the linking mechanisms between the macro and micro levels, which are suggested by the power resource and the new politics theory. One solution to these anomalies is to question the relevance and the quality of the empirical studies we have. In 1998, Korpi & Palme – being the primary defenders of the power resource theory – argued that, for example, ‘…the empirical testing of the macro-micro-links among institutions and the formation of interest and coalitions provides a major challenge for social scientists, but comparative micro-data currently are lacking’ (Korpi & Palme 1998, p. 682). Another solution is to gather the empirical evidence we actually have, try to find a theoretical explanation of the anomalies, and test the new theories on existing and new data.

It is this latter research strategy that will be followed in this book. Based on available cross-national surveys, a number of studies have looked in particular for a connection between the famous distinction between social democratic (Scandinavian), conservative (Continental European), and liberal (Anglo Saxon) welfare regimes (Esping-Andersen 1990, 2000) and public support for welfare policy. Our review of these studies, in Chapter 3, leads to three overall conclusions. First, the previous studies clearly show that we sometimes find substantial cross-regime differences in public support for welfare policy. Secondly, the studies that find the expected pattern of high support in social democratic regimes, medium support in conservative regimes, and low support in liberal regimes, primarily measure attitudes towards policies that concern the living conditions of the poor and unemployed. In the liberal regimes, this taking care of the (potentially) poor is basically what welfare policy is all about. For Americans, welfare policy only refers to need-tested programmes targeted at those at ‘the bottom’ of society. For Europeans, welfare policy refers to a much broader range of policies that also include benefits and services given to all citizens. Nevertheless, the point is that when it comes to retrenchment or restructuring of welfare policy in the narrow or American meaning we could actually expect the policy elite in social democratic regimes to be ‘blamed’ and the policy elite in liberal regimes to be ‘blessed’. Thirdly, the previous studies have not been able to find the mechanisms that link the macro- and micro-structure. The substantial cross-regime differences in support for welfare policy cannot simply be explained by differences in the attitudes of the middle class, the number of people who receive welfare benefits and services, or the number of public employees. Thus, when it comes to welfare attitudes, the previous studies leave us with a regime pattern at the aggregated level on the one hand and on the other hand demonstrate an inability to explain this pattern. With this research frontier as our point of departure, the aim of this book is to search for the mechanisms that link the macro-structures of welfare regimes to the micro-structures of public opinion towards welfare policy – call it a search for the missing link or the intervening variables.

Even though we are sceptical about the micro foundation of the power resource theory, the welfare regime theory, and the new politics theory, we borrow many insights from this literature and basically end up explaining cross-national differences in attitudes with cross-national differences in welfare state institutions. By pursuing this institutional line of reasoning we question two other popular explanations of cross-national differences in welfare attitudes.

The first popular explanation basically claims that low support (by comparative standards) for welfare policy primarily has to do with the presence of ethnic heterogeneity. The argument is primarily inferred from the American experience. The claim is that the presence of Indians and the import of slaves simply gave, and continues to give, fundamentally different preconditions for the welfare discussion in USA. This idea is widespread among American scholars and is supported by studies that show a strong correlation between attitudes towards the race issue and the welfare issue (e.g. Quadagno 1994; Gilens 2000). Moreover, this explanation gains increasing influence as recent European discussions about immigration sometimes also link-up with the welfare discussions. From their study of so-called ethnic fractionalization, Alesina & Glaeser (2004) infer that eventually increased ethnic heterogeneity will lower the Europeans’ passion for welfare policy. However, simply by turning the argument around, i.e. to claim that higher support for welfare policy in the European countries is caused by ethnic homogeneity, one becomes more uneasy about the argument. Naturally, the USA has had a large minority with a black skin but European history is also filled with clashes between different ethnic groups. The building of nation states had a homogenising effect on the one hand; a process that has not been given the same time in the colonies. However, this long process of turning inhabitants of a given area into national citizens also established new and persistent divides between the majority and ethnic minorities who wanted their own nation state or wanted to belong to another nation state (e.g. Bommes & Geddes 2000). Furthermore, Alesina & Glaeser (2004) and others may argue that the degree of ethnic homogeneity was crucial for establishing different kinds of welfare institutions, but still they need to take into account that once welfare institutions are established they have an impact on their own. The impact institutions have on public welfare attitudes is the topic of this book. And we will demonstrate that the institutional line of reasoning indeed helps to explain the pattern of low support in the USA and other liberal regimes, medium support in the conservative regimes, and high support in the social democratic regimes.

The second popular explanation of cross-national differences, which we will question, is the ‘culture thesis’. It basically claims that lower support in the USA and other liberal regimes is caused by a ‘passion for freedom over inequality’. The argument is in line with Lipset’s thesis of an American exceptionalism (e.g. Lipset 1996). The other way around, however, the ‘culture thesis’ claims that the high support in the social democratic regimes is caused by a ‘passion for equality’ (Graubard 1986). This explanation is widespread among Europeans; probably because, from a European perspective, it reproduces the ‘nice’ idea about a socially responsible Europe and a socially irresponsible USA. Not being able to find the expected effects from class-interest and self-interest, academic scholars sometimes also end up using different degrees of ‘egalitarianism’ as a cultural and residual explanation of cross-national differences in support for welfare policy (e.g. Andress & Heien 2001). However, the argument contradicts one of the very first and strongest impressions Europeans got after crossing the Atlantic. Coming from France, Tocqueville described in detail, and was indeed somewhat worried about, the ‘passion for equality’ that prevailed in the USA. Naturally, it was first of all a call for equality of opportunities and naturally much has happened in Europe since then. However, the comparative studies that more positively try to measure justice beliefs or level of egalitarianism do not find a pattern that neatly coincides with the described regime pattern. In Chapter 3 we will actually show that, measured in terms of the perception of just wage differences, the most anti-egalitarian attitudes are found in the conservative welfare regimes and not in the USA or the other liberal welfare state regimes (as Tocqueville would predict if he were still alive). Therefore, we argue that, in terms of egalitarian values, what distinguishes Americans and others who live in liberal welfare state regimes is not a general anti-egalitarian attitude. Instead, we find a specific anti-egalitarian attitude towards ‘the bottom’, which can be given an institutional interpretation.

The main contribution from this book is to develop a theoretical framework that enables us to explain how the institutional structure of the different welfare regimes influence public support for welfare policy. The theory is established by combining the welfare regime theory, which is presented and critically discussed in Chapter 2, with the theory of deservingness criteria (Cook 1979; De Swaan 1988; Will 1993; Oorschot 2000). For some reason these two lines of reasoning have lived rather separate lives. The basic idea of our synthesis is that the structures that characterise the different welfare regimes influence the way the public perceives the poor and the unemployed, which again influences the judgement of deservingness and thereby support for welfare policy. The theoretical framework is inspired by the classic thesis that means testing in contrast to universalism gives the public a negative perception of recipients (‘them’ versus ‘us’ logic). But the combination of welfare regime theory and deservingness theory enables us to specify this classic argument and open a broader theoretical perspective that includes the influence from the generosity of the welfare state, and the job opportunities provided by the (regime specific) labour market. Even though the ethnic fractionalisation thesis (Alesina & Glaeser 2004) is treated as a competing theory our theoretical framework also allows us to specify through which mechanisms the ethnic heterogeneity potentially can influence support for welfare policy. This theoretical framework is presented in Chapter 4.

Our line of reasoning has a lot in common with normative institutionalism, which in political science often is associated with March & Olsen (1984, 1989), but actually goes back to Talcott Parson and the old classic sociologists (Peters 1999). The idea that the political preferences of individuals are not exogenous, as in rational choice theory, but highly influenced by the institutional structures is shared by this book. Therefore we are sceptical about the rational choice institutionalism and its assumptions about an ‘economic man’, which has dominated the ‘new politics theory’. However, we are also sceptical about the variants of normative institutionalism where the attitudes of the ‘sociological man’ in a deterministic way are linked to a national cultural history. In contrast, we suggest what we shall call a ‘political man perspective’ on the formation of public support for welfare policy. This implies a reflective individual, where attitudes towards welfare policy are not only guided by self-interest or abstract societal values and norms but also based on concrete perceptions of the reality; in this case the perceptions of the poor and unemployed. Such a position fits nicely with recent studies that have shown that attitudes towards concrete policy proposals are highly dependent on the framing of the political issues (e.g. Gamson & Modigliani 1987; Kinder & Sanders 1996; Nelson et al. 1997; Kangas 1997).

As the dependent variable we will only discuss attitudes related to policies that concern the living condition of the poor and unemployed. Above, we called it ‘support for welfare policy’ in the narrow meaning or American meaning. Thereby this book primarily helps to explain why the living conditions of the poor and unemployed vary – and continue to vary – so tremendously across the Western capitalist countries. If, for example, we take the ideal type countries used by Esping-Andersen and the extent to which the income distribution generated in the market is altered (measured as the degree of reduction in the so-called Gini-coefficient), we find a clear pattern with the United States at the one end, Germany somewhere in the middle and Sweden at the other end (e.g. Ferrarini & Nelson 2002). If we look at the share of the population that lives below the relative poverty line, we find the same pattern. In 2000, 17 per cent of Americans had an income below the relative poverty line (less than 50 per cent of the equivalent median income) compared with 8.3 per cent in Germany and 6.5 per cent in Sweden.1 Such findings support the old saying in comparative welfare state research that ‘welfare for the poor becomes poor welfare’. This book will show how this process can be partly explained by regime-dependent welfare attitudes.

Naturally, the focus on welfare policy in the narrow sense puts limitations on the conclusions we can draw but we have a number of reasons for taking this focus. First, from previous studies we have indications of a regime effect when it comes to welfare policy in the narrow sense. Second, even though good comparative data on welfare attitudes are unavailable we do have some comparative data on attitudes towards welfare policy in the narrow sense. And actually some of these data are little explored. Third, the management of ‘bad risks’ is a crucial variable in Esping-Andersen’s argument for the existence of three, and only three, distinct regimes (Esping-Andersen 2000, pp. 77–81). Fourth, welfare policy in the narrow sense is also the crucial variable in the American literature on the connection between race and public support which, as already mentioned, will be one of the main opponents or ‘discussion partners’ in this book.

The empirical backbone of the book is presented in Chapters 5, 6, 7 and 8. In these chapters we will prove the soundness of the causal reasoning in our synthesis of the welfare regime theory and the deservingness literature. In Chapter 5, we analyse how the three regime dimensions deduced in Chapter 4 (job opportunities, degree of selectivism, and degree of generosity), and the differences in ethnic fractionalisation influence public explanations of causes of poverty. Based on the World Values Studies (WVS) and Eurobarometer surveys, we are able to analyse cross-regime data over a long period. Still one cannot totally avoid the so-called ‘small n-problem’ – having too few cases and too many independent variables – that haunt all comparative research. In our case, the problem is that even though we find the expected connection between welfare regimes and the perception of poor and unemployed, one cannot totally dismiss the possibility that the pattern could be caused by differences in egalitarianism or ethnic homogeneity rather than differences in welfare institutions.

Therefore, Chapters 6 and 7 supplement the ‘traditional’ comparative analysis in Chapter 5 with two so-called focused comparisons. The task is to ‘isolate’ the institutional effects derived respectively from the degree of selectivism and degree of generosity. By comparing the Nordic countries, we hold the size of the countries, ethnic fractionalisation, social trust, political systems, cultural legacy etc. more or less constant and look at the effects of differences in welfare policy (which actually can be found across the Nordic countries). The data for these ‘most similar designs’ come from a Nordic module to the International Social Survey Program (ISSP) from 1999 and a large Danish and Finnish dataset. The latter dataset both includes a general population sample and a large sub-sample of long-term unemployed.

These focused comparisons enable us to conclude with very high certainty that the institutional effects suggested by our theoretical framework are present. However, one of the main problems with ‘most similar designs’, i.e. everything other than the independent variable is held constant, is that one cannot assess the relative importance of the independent variable in play. In our case, one would for example like to know how strong the institutional effects are compared with the effects-derived from ethnic divides. Therefore, Chapter 8 applies our theoretical framework on a national sample that has been established exclusively for this purpose. At this individual level within one nation we do not have ‘real’ cross-national institutional variation. But the sample enables us to analyse, in one model, how public perceptions of (1) immigrants, (2) the poor and unemployed in general, (3) the generosity of the welfare state, and (4) job opportunities, affects public support for welfare policy. The analysis shows that the perceptions, which we argue in Chapter 4 are likely to be regime-dependent, are highly relevant in order to explain variations in individuals’ support for welfare policies. From that we infer that they may also be highly relevant in explaining cross-national variations in public support for welfare policies.

Besides summarising the theoretical argument and the main empirical findings, Chapter 9 includes four discussions. First, we discuss the thesis’ implication for the welfare regime theory. We have only analysed public support for welfare policy in the narrow sense, and we admit that what is most needed is better comparative data. However, we still believe that the book challenges the micro-foundation of present comparative welfare state theory and succeeds in presenting an alternative explanation. Secondly, we discuss to what extent regime-dependent public support for welfare policy actually influences the pursued welfare policy. Thirdly, we discuss the notion of path-dependency in relation to public support for welfare policy. And finally we discuss the future public support for welfare policy in the light of the recent trends in welfare state development and immigration. It leads to the prediction that, due to the impact of welfare institutions, the European future might not be American.

1 The estimates of the proportion of citizens living below the poverty line are heavily influenced by the method of calculation. Here we have simply taken the overall figures provided on the webpage of the Luxembourg Income study.

Chapter 2

Welfare Regimes and the Attitudes of their Inhabitants

This chapter introduces the main concepts of the regime theory and describes how various researcher...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Contents
• List of Figures and Box
• List of Tables
• Preface
• 1 Introduction
• 2 Welfare Regimes and the Attitudes of their Inhabitants
• 3 The Puzzle of Public Opinion
• 4 Welfare Regimes and Deservingness
• 5 Welfare Regimes and Perceived Causes of Poverty
• 6 Selectivism and Stigmatisation
• 7 Generosity and Stigmatisation
• 8 Regime-dependent Perceptions and Social Assistance
• 9 Deservingness and Welfare State Development
• References
• Index