List of Figures and Tables. Foreword

Transcription

1 Contents List of Figures and Tables Foreword x xv 1 Introduction The health concept Health inequalities Social capital and health inequalities The welfare concept Welfare regimes Overview of the book 13 2 Theoretical Foundations of Social Capital in Health Research Theoretical definitions of social capital Classifications of social capital Social capital theory in health research Towards a definition of social capital in health research The contribution of a resource-based theory of social capital for health research Conclusions 40 3 Social Capital and Welfare: Do Universal Welfare States Crowd Out or Stimulate Social Capital? The welfare state and informal social contacts The welfare state and formal social contacts The welfare state and social trust The welfare state and social resources Conclusions 58 4 Social Capital and Health: Mechanisms and Empirical Findings Individual social capital and health Collective social capital and health 66 vii

2 viii Contents PROOF 4.3 Social capital as a health-promoting feature of welfare states The dark sides of social capital in relation to health Some examples of previous studies on individual and collective social capital and health Social capital and health in European welfare states: Ecological associations Conclusions 82 5 Health Inequalities by Education in European Welfare Regimes: The Contribution of Individual Social Capital Introduction Data and methods Results Conclusions Health Inequalities Between European Welfare Regimes: The Contribution of Collective Social Capital Introduction Data and methods Results Conclusions Social Capital and Health Inequality in the Social-Democratic Swedish Welfare State Introduction Data and methods Results Conclusions The Dark Sides of Social Capital: Homophily and Closure of Immigrants Social Networks in the Swedish Welfare State Introduction Data and methods Results Conclusions Concluding Discussion Welfare states and social capital Social capital and health in European welfare states Social capital and health inequality in European welfare states 191

3 Contents 9.4 The downsides of social capital in relation to health Policy implications Concluding remarks 199 ix Appendix 1: Methodological Concerns and Limitations 202 Notes 206 References 209 Index 227

4 1 Introduction Despite well-developed health care systems, advanced medical technologies and sufficient material living standards, there still exists a social gradient in health in Western societies. The unequal distribution of health between social groups was first acknowledged by the British Black Report at the beginning of the 1980s (Townsend and Davidson, 1982). Recently, the World Health Organization (WHO) Commission on Social Determinants of Health (CSDH) performed a global overview of health inequalities and found that large health inequalities, both between and within countries, still persist throughout the world at the beginning of the 21st century (CSDH, 2008). The commission and research into health inequalities emphasize that such disparities are largely due to differences in people s living conditions and life chances. Accordingly, the social determinants of health provide crucial knowledge of why health differs systematically between countries and between social groups within countries. Improving various social conditions among the disadvantaged is a key for reducing health inequalities and improving public health (Marmot, 2004; Siegrist and Marmot, 2006; Marmot and Commission on Social Determinants of Health, 2007). Nonetheless, the fact that health inequalities are stable and even tend to increase in relatively rich societies has continued to perplex scholars. A recent book The Spirit Level suggests that inequality in material and social circumstances between social groups in richer societies causes frustration, stress, and a wide variety of other adverse externalities among the disadvantaged, with adverse consequences for health and longevity (Wilkinson and Pickett, 2009). Accordingly, it is the psychological experience of inequality that causes ill health rather than the material conditions per se. Consequently, inequality makes people sick, irrespective of a country s overall economic circumstances. One of the 1

5 2 Social Capital and Health Inequality explanations for adverse health consequences resulting from inequality is that it crowds out social relationships, social cohesion, and social capital (Wilkinson, 1996; Wilkinson and Pickett, 2009; Vergolini, 2011; Lyte, forthcoming). However, research on the role of social capital in explaining health inequalities is still scarce (Pearce and Davey-Smith, 2003; Dahl and Malmber-Heimonen, 2010). This book deals with social capital, as a social determinant of health, and whether it contributes to the unfair distribution of health between and within societies. It has often been assumed that the Nordic or social-democratic welfare state reduces health inequalities and improves public health through its ability to combat poverty and social exclusion, while health inequalities are supposed to be larger in countries with less comprehensive welfare systems located in the Eastern and Southern parts of Europe (Bambra, 2007; Eikemo et al., 2008a; 2008b; Lundberg et al., 2008; Bambra and Eikemo, 2009). Recently, it has also been argued that the welfare state and its consequences could be crucial for the creation and maintenance of social capital. Consequently, levels of social capital might vary systematically between countries depending on welfare state characteristics and the generosity of welfare systems (van Oorschot and Arts, 2005; Rostila, 2007a). Although previous studies have shown that social capital is strongly related to health and wellbeing (for a review, see Islam et al., 2006), most previous research in the field of social capital and health has so far focused on pure associations and ignored the significance of the broader institutional and political context for the creation and maintenance of social capital and its potential health consequences. The overall objective of this book is to fill this gap in the literature by studying whether the consequences of social capital on health and health inequalities vary between countries with different institutional characteristics and welfare policies. This is important when considering that many interventions that potentially stimulate social capital and health, and also reduce health inequalities, take place at the state level. Global organizations, such as the Organisation for Economic Co-operation and Development (OECD) and the World Bank, have emphasized social capital as a potential strategy to improve the health of nations and communities (The World Bank, 1998; OECD, 2001). By studying the significance of the welfare state for the social capital health relationship, policy makers could learn much about how investments in equality and social goods (e.g. welfare) could influence social capital and its potential health consequences. The contribution of societal features for social capital and its health consequences has roots that go far back in the social sciences. Durkheim

6 Introduction 3 (1897/1997) argued that aspects of social capital can vary systematically between countries and that such differences could explain country-level suicide rates. Yet, it was the introduction of the concept of social capital by some sociologists (Bourdieu, 1986; Coleman, 1988), and especially the ensuing work by the political scientist Robert Putnam on the subject (Putnam et al., 1993; Putnam, 2000), that contributed to a dramatically increased interest in social capital in the health-related sciences. Accordingly, the number of published articles on social capital and health in the most influential journals in epidemiology and public health has increased enormously in recent years (Kawachi et al., 2008). Hence, the concept might be considered one of the most successful conceptual exports from sociology (Portes, 2000). The work of Robert Putnam is often considered the most influential in the field of social capital. His books Making Democracy Work Civic Traditions in Modern Italy (1993, together with Leonardi and Nanetti) and Bowling Alone: The Collapse and Revival of American Community (2000) received a lot of attention and introduced social capital as a phenomenon that makes societies work better through its ability to facilitate coordinated actions between individuals in society. In Making Democracy Work Putnam et al. examined regions of Italy in a historical perspective and argued that high levels of social capital explained the economic, social, and political success of North Italy, while the scarcity of social capital had contributed to the negative economic and political trend in the southern region of Italy. In Bowling Alone Putnam continued to study social capital in the American context and found declining rates of social capital in the United States during the post-war era a trend that Putnam also expected to be found in other Western societies. Accordingly, some studies have confirmed declining levels of social capital in other non-american Western societies (Hall, 2002) while others have found no such evidence (de Hart and Dekker, 1999; Torpe, 2003). However, even the declining levels of social capital in the United States have been challenged (Paxton, 1999). In Putnam s book Bowling Alone it is argued that welfare states and welfare state features have no major role in stimulating or perhaps destroying social capital in society. In its place, Putnam argued that increased levels of television watching, the middle-class movement to the suburbs, the increasing female labour market participation, and a new generation of less civic individuals were the most important factors underlying the decline of social capital in America (Putnam, 2000). Arguments that emphasize the significance of the welfare state and its

7 4 Social Capital and Health Inequality features for social capital have, however, become increasingly popular. Yet, there have also been controversies regarding the impact of welfare on social capital. Some argue that universal and comprehensive welfare states have negative consequences on social capital, as such societies crowd out social relationships, social trust, and civic activities (Wolfe, 1989; Fukuyama, 2000; Scheepers et al., 2002). The basic argument is that the need and incentives for the creation and maintenance of social contacts and civic activities decreased when the welfare state took over many of the responsibilities and duties that were previously located in people s social networks and associations (van der Meer et al., 2009). At the other end of the spectrum, others suggest that universal welfare states primarily have positive implications for various aspects of social capital (Klausen and Selle, 1995; Torpe, 2003; van Oorschot and Arts, 2005) and that people s ability and incentives to take part in social life are encouraged and supported by the welfare state and its institutions. However, the relationships between the welfare state and different dimensions or subtypes of social capital could also vary. It has, for instance, been suggested that the exchange of social resources in the networks of citizens is crucial in societies that lack comprehensive and generous welfare systems, while the success of universal welfare states builds on solidarity and trust among citizens (Rose, 1995; Rothstein, 2001; Völker and Flap, 2001; Jensen and Tinggaard Svendsen, 2011). Another important objective of this book is hence to scrutinize the relationship between welfare and different forms of social capital, as high social capital, in turn, could be a foundation for a healthy society. The concept of social capital has, however, also received a lot of criticism and could be considered one of the most questioned concepts in the social sciences (Portes, 1998). There is a present theoretical disagreement on whether social capital should be considered a property of individuals or social structures. Some researchers, for instance, adhere to the perspective that social capital is more than the aggregated characteristics of individuals, and that it is a feature of social structures rather than of individual actors within a social structure (Putnam et al., 1993; Lochner et al., 1999; Kawachi and Berkman, 2000; Putnam, 2000). Still, the concept has also been considered and defined as an individual good (Bourdieu, 1986; Coleman, 1988; Portes, 1998; Lin, 2000; 2001; Carpiano, 2006). The importance of structure versus individuals has, however, long been debated within the social sciences (e.g. Durkheim, 1897/1997; Parsons, 1951; Weber, 1983). One of the chapters within this book (Chapter 2) will discuss the theoretical foundations of social capital and tackle some of the conceptual confusion in the field.

8 Introduction 5 Disagreements about the appropriate level of analysis have also contributed to controversies over whether the health benefits of social capital lie on an individual level (e.g. Moore et al., 2005; Carpiano, 2006) or a collective level (e.g. Lochner et al., 1999; Kawachi and Berkman, 2000). The collective notion of the concept, in line with the work by Putnam and followers, has so far dominated the health-related sciences. Accordingly, this notion argues that social capital is a collective good and that social capital, as a feature of contextual units such as countries, states, or neighbourhoods, is decisive for health. Yet there have been some recent tendencies for a revival of individual-level notions of the concept in the health-related sciences (Altschuler et al., 2004; Moore et al., 2005; Carpiano, 2006; Stephens, 2008). These consider social capital as an individual resource that can be used by single individuals in order to achieve health and longevity. Despite the controversies, social capital will be regarded within this book as a concept including both individual and collective features (see Chapter 2 for a full description). Given that there is no consensus in previous research on whether the main health benefits lie on the individual or collective level, such an approach seems reasonable. It also seems plausible that individuals health and wellbeing could be affected by both their own personal social capital as well as the social capital that characterizes the surrounding social environment. Such a perspective seems particularly important in health research. Without considering social capital on different levels of aggregation we cannot understand the full health benefits of social capital. Moreover, there might be important interactions across levels. Low social capital in a society might, for instance, primarily influence the health of those with poor individual social capital. However, the social capital health relationship on different levels of aggregation may also depend on the specific dimension studied. High social trust might be a highly valuable resource for the health of societies, while citizens personal social relationships primarily influence individual health. Finally, most scholars accentuate the positive consequences of social capital, whereas they often ignore the dark sides of the concept. Nevertheless, the use of social capital might, in some instances, have negative consequences (Portes, 1998; Kunitz, 2004) and might rather contribute to poor health and larger health inequalities. Social capital may, for instance, reinforce unhealthy behaviours and norms leading to drug abuse, obesity, criminality, and risky sexual behaviour (Gambetta, 1993; Friedman and Aral, 2001; Christakis, 2004; Christakis and Fowler, 2007;

9 6 Social Capital and Health Inequality Maycock and Howat, 2007). Some also maintain that social networks can be coercive and constitute sources of strain as well as support (Rook, 1984; Due et al., 1999). Furthermore, it is also important to regard the unequal distribution of social capital in society as a downside of social capital (Pearce and Davey-Smith, 2003; Szreter and Woolcock, 2004) that, in turn, might contribute to health inequalities (Dahl and Malmberg-Heimonen, 2010). In a similar way to economic or human capital, it is reasonable to believe that the wealthier segments of the population and those with higher status have larger social networks and higher levels of trust, and receive social resources of higher quantity and quality from their networks (Lin, 2000; Dahl and Malmberg-Heimonen, 2010). Social capital can then be used by these groups in order to obtain additional advantages in society. Few studies have raised interest in the dark sides of social capital and the contribution to health inequalities. Therefore, one part of this book is devoted to examining this issue empirically. Next, I will give some background to the central concepts of health and health inequalities within this book, as well as the meaning of welfare and welfare regimes. The theoretical foundations of social capital will, however, be treated separately in Chapter The health concept This book is not merely focused on social capital and its development as such; it will also study the consequences of an individual s or a country s social capital on health and health inequalities. Health can be examined in many different ways, ranging from people s own feeling about their state of health to physicians diagnoses based on medical science. Hence, it is important to clarify the meaning of the concept of health and how health will be examined within this book. The WHO suggests that health is a state of complete physical, mental, and social wellbeing and not merely the absence of disease or infirmity (WHO, 2006). This very broad definition of health is, however, difficult to study empirically. Therefore, health research often focuses on states that deviate from good health in the form of disease, sickness, or mortality. The difference between disease and sickness is, however, of importance for this book. Disease concerns the medical or biological aspects of ill health, traditionally diagnosed by health care professionals, while sickness refers to the individual experience of health problems. Hence, it is possible that an individual can have a disease without being sick and vice versa. It is, however, reasonable to believe that disease and sickness coincide

10 Introduction 7 to a high extent. People usually go to the doctor when they experience some kind of health problem, which is then diagnosed by a physician. This book will primarily focus on social capital and how it relates to self-assessed health problems and therefore, to a large extent, it will examine sickness instead of disease among both individuals and societies. It will study how social capital relates to health outcomes such as overall self-assessed health problems and self-perceived indicators of psychological health and circulatory health problems. It will further examine whether social capital also contributes to inequalities in these health outcomes between social groups such as social classes, educational groups, and groups based on country of birth. Mortality could be considered an unproblematic dimension of health people live or die. However, many health problems and diseases do not necessarily lead to premature death. Yet, some of the analyses of this book will examine the associations between social capital and life expectancy in European countries. 1.2 Health inequalities This book will also examine whether social capital contributes to health inequalities both between and within welfare states. This makes it important to specify what is meant by health inequalities and how health inequalities differ from other types of differences in health. Health might be considered an individual phenomenon individuals fall ill and die. This is reinforced by the medical practice, as medical treatment is focused on the individual. Most of us accept that health varies between individuals and during different points in life. Some differences in health between larger groups within society are not even necessary health inequalities (Graham, 2007). Poor health among older adults when compared to young adults is a biological fact rather than a health inequality. Yet, there are large and systematic differences in health and longevity within society that may be considered unequal. These differences indicate that social structures and the processes that cause inequalities are more important for health inequalities than inherited or acquired individual risk factors. Consequently, poor health is not an individual problem alone but also a social problem. Health differences between social groups arise through systematic differences in living conditions, health behaviours, and/or vulnerability to disease between these groups, which ultimately lead to health inequalities between them. In order to separate unfair differences in health from acceptable and expected differences we might separate between the

11 8 Social Capital and Health Inequality concepts of health inequality and health inequity (Graham, 2007). Health inequality basically just refers to patterns of health differences within a population, while health inequity is a normative concept that refers to health differences that are politically, socially, and economically unacceptable. When health inequalities are mentioned within this volume, it refers to those types of inequalities that are unfair and unjust (e.g. health inequities). However, this book will not only examine unacceptable health inequities between social groups within countries. Health inequalities between countries could also be considered unfair and unjust, and might also be influenced by social determinants of health including social capital. 1.3 Social capital and health inequalities Different models explaining health inequalities between groups in society have been proposed and social capital may play an important role in these models. The psychosocial (Wilkinson, 1996; Marmot, 2004; Wilkinson and Pickett, 2009) and neo-material explanations (Muntaner and Lynch, 1999) are probably the most prominent ones and are often suggested as two contradictory explanations for health inequalities. The neo-material explanation suggests that health inequalities originate from differences in material circumstances such as income, living conditions, working conditions, place of residence, et cetera. Hence it is primary differences in such material circumstances that cause health inequalities between social groups. The psychosocial explanation acknowledges the unequal distribution of material resources. However, according to this notion differences in material circumstances between social groups cause stress and frustration among individuals in lower social positions, which, in turn, lead to ill health. Accordingly, it is the psychological experience of inequality that causes ill health rather than the material conditions per se. Social capital might, however, relate to both material and psychosocial explanations when considering the multidimensional view of the concept used within this book. Low social capital has, for instance, been suggested to follow feelings of marginalization and relative deprivation among disadvantaged groups in unequal societies (Wilkinson, 1996; Wilkinson and Pickett, 2009). People who experience long-term disadvantage may turn away from social and civic life, develop a distrustful attitude towards their fellow citizens, and be less willing to share social resources. Hence, reductions in social capital among individuals or in societies could be decisive for whether the psychological experience of inequality ultimately leads to health

12 Introduction 9 inequalities. Yet, social capital in the form of social resources embedded in social networks might also provide individuals with material resources available in social networks, and could hence be closely linked to the neo-material explanation. For instance, social capital can be useful when trying to find a job with high income and status, or a place to live in a good residential area that improves the material circumstances of individuals. However, social capital could also relate to another model explaining health inequalities; the model on lifestyles (Cockerham, 2005). This model suggests that health behaviours are socially patterned and that lower social groups are over-represented when it comes to adverse health behaviours. Individuals may adopt certain behaviours and lifestyles that are typical for their social group and that distinguish them from other social categories. Differences in such behaviours may account for social inequalities in health (Cockerham, 2005). Social capital could, however, influence people s lifestyles and health-related behaviour. It could either promote good behaviours or reinforce negative ones. Class-based social networks could, for instance, exert normative guidance over the healthrelated behaviours of other individuals in the same class. For instance, smoking was a common habit in the higher classes and was used as a marker of a high status in a historical perspective. However, the number of smokers in the higher classes dropped when the health risks of smoking became known, while people from the lower classes started to imitate the smoking habits of the higher classes. It has been suggested that social networks played an important role in the transformation of smoking from an upper-class to a working-class behaviour through their potential for spreading norms and behaviours (Dixon and Banwell, 2009). It should, however, be emphasized that the neo-material and psychosocial explanations, as well as the model on lifestyles, are not considered contradictory within this book. It is rather the combination of these that contributes to health inequalities. Consequently, social capital could influence people s material circumstances, lifestyles, and psychosocial experience of inequality, and thereby contribute to health inequalities. 1.4 The welfare concept Another aim of this book is to examine the contribution of social capital to health inequalities in a welfare state perspective. However, like the concept of health, the welfare concept also has different theoretical meanings.

13 10 Social Capital and Health Inequality It has been agreed that the welfare state should be understood as the state s involvement in the distribution and redistribution of welfare in a country, taking democracy and a relatively high standard of living as a basis for the welfare state (Korpi, 1983; Esping-Andersen, 1990; Berg-Schlosser and DeMeur, 1994; Huber and Stephens, 1996; Aidukaite, 2009). Social policy is also sometimes used synonymously with the welfare state concept. Skocpol and Amenta (1986), for instance, use the concept of social policies when they refer to state activities affecting the social status and life opportunities of families, individuals, and various social groups. Social policy is described as state activities that have redistributional effects upon the population of a given country through regulated mass education, social insurance, pension programmes, and the health care system. The Scandinavian welfare research tradition provides a more individual-oriented definition of welfare, and could therefore elucidate how welfare states influence citizens personal welfare resources, among the social capital. This definition is focused on the living conditions of individual citizens and their families, and was initiated by Sten Johansson and colleagues in the Swedish Level-of-Living Survey (LNU). Johansson defines level of living, or welfare, as the command over resources in terms of money, possessions, knowledge, psychological and physical energy, social relations, security and so on by means of which the individual can control and consciously direct her conditions of life. Further, it is suggested that welfare is an assembly term for living conditions in those areas where citizens seek influence through collective decisions and through commitments in institutional forms, i.e. through politics (Johansson, 1970, p. 138). According to Erikson (1993), the central element in this definition is the individual s command over resources, which was extracted from Richard Titmuss s writings on welfare (1958). Nevertheless, Erikson (1993) claims that it does not seem sufficient to restrict the concept of level of living or welfare to resources alone. Some circumstances, such as the quality of the work environment, cannot be regarded as a resource in a remote sense. Hence, he suggests that individuals resources and the characteristics of the arenas where they are to be used determine the scope of individuals for directing their own lives (Erikson, 1993, p. 74). The definition of welfare according to the resource perspective is important as it suggests that social capital makes up a crucial aspect of welfare, and that it might present one of the means by which individuals can control and consciously direct the conditions of life. Social capital could, however, also be important as it facilitates access to many other important welfare resources embedded in people s networks, such as knowledge, money,

14 Introduction 11 possessions, et cetera. Hence, social capital could also be considered a welfare resource, which facilitates the access to other welfare resources. 1.5 Welfare regimes The definitions of welfare given above are crude and describe general views of the welfare concept, as well as the types of welfare resources that individuals can use to control and consciously direct their conditions of life. Yet, this book will examine differences in social capital between European welfare states, and study whether differences in social capital contribute to health inequalities both between and within different welfare states. Hence, a model that describes different institutional and political contexts and how welfare is produced within these is crucial. In order to comprehend why some welfare states might possess higher levels of social capital than others, and why the contribution of social capital for health inequalities may differ between welfare states, it seems essential to elucidate how welfare states differ in some central aspects. The categorization of countries into welfare state regimes will permeate this book, as many of the chapters will use such a division. The welfare typologies introduced by Esping-Andersen (1990; 1999) clarify differences between various countries concerning welfare policy and its consequences. These variations may be significant for both the levels of social capital and its association with health. Esping-Andersen argues that welfare states have historically developed into systems with their own institutional logic, and that the relative importance of the market, family, and the state for citizens welfare varies from one country to another. The welfare regime concept hence stresses the various roles and the importance of these institutions in the production of welfare. However, even if there are similarities between countries belonging to the individual regime types, there are also differences, and the regime types should therefore merely be regarded as ideal types. The ideal typical social-democratic regime s policy of emancipation addresses both the market and the traditional family. These countries are characterized by the highest levels of social security, with mostly universal social benefits. The principle is not to wait until the family s capacity for aid is exhausted but to pre-emptively socialize the cost of family-hood. The ideal is not to maximize dependence on the family but the capacity for individual independence. The result is a welfare state that, compared with other regimes, largely takes direct responsibility for caring for children, the aged, and the marginalized (Esping-Andersen, 1990; 1999). In other words, this model is characterized by universalism and solidarity. Compared with the other two regime types, levels of

15 12 Social Capital and Health Inequality inequality and poverty are low (Fritzell, 2001). When citizens are dependent to some extent on the welfare state, and at the same time benefit from it, they probably feel more obliged to pay taxes and support state actions. Sweden and Denmark are examples of countries belonging to this regime type. In the conservative-corporatist type, preservations of status differentials predominate, and rights are therefore attached to class and status. The state only interferes when a family s ability to serve its members is exhausted, and it then provides social benefits based on previous earnings and status in society. This regime type is also largely shaped by the church and focuses on the preservation of the traditional family (Esping-Andersen, 1990). Further, the conservative-corporatist regime appears to create average levels of inequality and de-commodification in comparison with other regimes. De-commodification refers to activities and efforts, generally provided by the government, that reduce citizens reliance on the market (such as unemployment and sickness insurance). France and Germany are examples of countries belonging to the conservative-corporatist type. In the ideal typical market-dominated liberal regime, means-tested assistance, modest universal transfers, or modest social-insurance plans predominate. The state mainly encourages the market either passively, by guaranteeing only a minimum of benefits, or actively, by subsidizing private forms of the welfare system. This type of regime entails independence from the state and forces citizens to rely on family and friends for help and aid in situations of personal crisis. The consequences of this type of regime are high levels of income inequality, high levels of poverty and low levels of de-commodification compared with the socialdemocratic and conservative-corporatist regimes. The model also creates high levels of class dualism. Examples of countries belonging to this regime type are the United Kingdom and the United States. However, as several countries cannot be categorized into the three types of welfare regimes described above, two additional regime types have been suggested: the Mediterranean and the post-socialist types. The Mediterranean regime aims to produce even more dependence on family and friends. In this type of regime, a less developed system of social security exists, instead of an official level of security, accompanied by a very high degree of familialism (Ferrera, 1996; Bonoli, 1997). Spain and Italy can be regarded as belonging to the Mediterranean regime type. Finally, the post-socialist regime, which consists of some of the countries located in central and Eastern Europe, is still only moderately theorized and analysed. Aidukaite (2004; 2009), however, examines

16 Introduction 13 whether the Baltic countries, as examples of post-socialist states, have developed into a distinctive post-socialist model of social policy, or whether they fall into one of the models suggested by Esping-Andersen. The study shows that the Baltic states cannot be placed exactly in any model developed to study social policy. This supports the idea that the Eastern European countries constitute a separate kind of welfare regime. The results also indicate that the benefits of social security are very low in the Baltic countries and that this has resulted in high levels of income inequality and poverty. Moreover, the post-socialist countries are characterized by high coverage of the social security systems, but low benefits, and therefore citizens still, to a high extent, have to rely on family or the market for support (Aidukaite, 2009). However, it must be added that there are also large variations between the post-socialist countries (Deacon, 1993; Kangas, 1999), and it is therefore questionable whether we should regard them as one separate welfare regime type. The Czech Republic and Poland are examples of countries belonging to this regime type. Finally, it should also be mentioned that the welfare regime concept described by Esping-Andersen (1990) has been criticized because of the absence of a gender perspective that considers the relationship between unpaid as well as paid work and welfare (Lewis, 1992; Orloff, 1993). 1.6 Overview of the book With this introduction as a starting point, this book will consider social capital and its relationship with health/health inequality in a welfare state context, as shown in Figure 1.1. Consequently, the contribution of the book to the field of social capital and health is to fill Welfare state context Social capital Health Health inequality Figure 1.1 Social capital and health in a welfare state context

17 14 Social Capital and Health Inequality a current gap in the literature that has so far neglected the contribution of welfare states for the social capital health relationship. While health research has so far primarily been interested in the health consequences of social capital, most studies in the social sciences have merely focused on social capital in itself, and/or its determinants. Another contribution of this book is therefore to provide knowledge on both the determinants of (e.g. welfare) and consequences of (e.g. health) social capital. Following the work by Putnam and followers, this book will continue the tradition of studying trends in social capital during longer-term periods, as well as examining cross-national differences in social capital between European countries. The study of European countries is informative, as most of these countries could be considered welfare states, although they also vary considerably regarding types of welfare policies and the generosity of welfare systems. The focus on cross-national European comparisons is also common in contemporary welfare state research (Kautto et al., 2001; Kvist et al., 2012). Most of the chapters within this book (chapters 3 6) will use cross-national European data when studying the significance of the welfare state context for social capital and health. The final two chapters (chapters 7 and 8) will, however, use Sweden as an example of a social-democratic welfare state. Sweden is often considered among the most comprehensive welfare states in the world, with a long tradition of a welfare policy with universal and generous welfare benefits (Bergh, 2004). Although Sweden could be considered an interesting extreme case for the study of social capital and health, the high-quality data available are also necessary for some of the research questions proposed within this book. For instance, repeated measurement of social capital over longer time periods is needed when studying the development of social capital over time and how it relates to welfare. Moreover, in order to study the causal links between social capital and health, longitudinal data are necessary. The book has a multidisciplinary perspective on social capital, welfare, and health. Such a perspective is necessary for the understanding of how areas that are traditionally studied separately in different academic disciplines are related. For instance, many of the theories on the conceptual development of social capital and the significance of welfare state features for social capital are based on sociological knowledge and advancements in political science, while theories on the links between social capital, health, and health inequality are based on knowledge in medicine, epidemiology, and public health.

18 Introduction 15 The book chapters This book includes a mix of theoretical discussions and empirical evidence. However, Chapter 2 of this book is entirely devoted to the theoretical foundations of social capital. As suggested, some scholars argue that social capital should be chiefly regarded as a feature of the social structure, while others claim that it originates from individuals personal social networks. These theoretical uncertainties have contributed to the fact that social capital might be considered one of the most debated and criticized concepts within the social and healthrelated sciences. Chapter 2 discusses different theoretical perspectives of social capital, and relates these to the development of social capital within health research. It also aims to bridge the individual and collective notions of social capital. The chapter will naturally not be able to solve all the uncertainties surrounding the concept. Nevertheless, a resource-based theoretical definition of social capital guiding the chapters within this book will be provided, which aims to solve some of the antagonism between individual and collective notions of the concept and to clarify how various dimensions of social capital may be related. The theoretical definition provided in the chapter will guide the remaining empirical chapters within this book. The levels of social capital in a welfare state could influence the health of citizens and contribute to health inequalities between countries. The relationship between the welfare state and social capital has, however, been widely discussed. Some suggest that universal welfare states crowd out social capital, while others claim that welfare states of the social-democratic model promote various aspects of social capital. The theoretical arguments for such a crowding out effect by universal welfare states are discussed in Chapter 3, together with theories that emphasize the positive consequences of universal welfare states on levels of social capital. However, as the controversy between advocates of the crowding out hypothesis and those who support the opposite view remains, the chapter examines the trends and distribution of social capital in different welfare states. Such an analysis shows whether welfare states with different institutional characteristics differ in levels of social capital, which could support the presence or absence of a crowding out effect. Another empirical test on whether welfare stimulates or crowds out social capital is to study the association between welfare spending and levels of social capital in countries. Chapter 3 hence also provides knowledge on this relationship in order to show whether welfare effort is positively or negatively related to social capital.

19 16 Social Capital and Health Inequality Chapter 4 links social capital with health, and discusses it in light of the welfare state. It focuses on the previous theories and empirical evidence in the field of social capital and health/health inequality in order to give a general background to the field of social capital and health. The chapter scrutinizes why we should expect health consequences of social capital on health, how explanations may differ depending on whether individual or collective social capital is examined, and whether previous empirical studies support an association between social capital and health. The chapter also discusses how welfare state characteristics might contribute to the association between social capital and health, and specifies the circumstances in which negative health externalities by social capital may arise. Finally, the chapter shows some empirical evidence on the relationship between social capital and health in European welfare states. These findings show the overall ecological association between social capital and health. However, they also suggest whether the social capital health relationship follows a welfare state pattern; that is, whether clusters of countries belonging to certain welfare regimes have both low social capital and poor health. Chapters 5 and 6 are focused on social capital and health/health inequality in European welfare states. These two chapters aim to put the social capital health relationship in a political and institutional context, and examine whether social capital is more important for health inequalities in some European welfare states than in others. More specifically, Chapter 5 examines whether the associations between dimensions of individual social capital and health vary depending on the welfare state regime. The chapter further studies whether some forms of social capital are more important for health inequalities by education in some welfare state contexts than in others. Chapter 6 examines health inequalities by clusters of European countries grouped into welfare regimes. It further examines whether collective social capital explains health inequalities between regime types and whether there are cross-level, health-related interactions between individual trust and residence in different welfare regime types. The two final empirical chapters are devoted to in-depth analyses of the Swedish social-democratic welfare state concerning the social capital health relationship. The case of Sweden is interesting because it has long been considered one of the most equal societies, with a stable development of welfare, good population health, and low health inequalities (Fritzell and Lundberg, 2007). The use of unique Swedish long-term follow-up data could provide important knowledge on whether welfare development is positively or negatively related to

20 Introduction 17 trends in social capital. However, whether social capital also contributes to health and health inequalities in such a context is still somewhat unclear. Hence, Chapter 7 studies the development of social capital in Sweden during the past 40 years, in order to examine whether the comprehensive Swedish welfare state, with maturing welfare during the post-war period, seems to crowd out or stimulate the creation and maintenance of social capital. The chapter also empirically examines the association between individual social capital and health in Sweden, and whether social capital accounts for health inequalities between social groups in Swedish society, such as social classes and groups based on country of birth. An important focus in the chapter is the study of social capital and health with longitudinal data. This provides us with possibilities to get closer to causal inference. The final empirical chapter of this book, Chapter 8, focuses on the downsides of social capital in relation to health, and whether these can account for health inequalities. The chapter examines whether some negative network properties (homophily and closure) might account for health inequalities between immigrants and natives in Sweden. Such network features are thought to influence the quantity and quality of social capital embedded in social networks, which, in turn, have health consequences. The concluding discussion in Chapter 9 finally summarizes the findings of the book, links empirical results with previous theories and research, and provides policy implications from the results. Appendix 1 separately discusses the limitations and weaknesses of the empirical studies included in the book.

21 Index abuse drug, 5, 35, 68, 72, 162, 166, 170 physical, 71 psychological, 71 additivity, African American, 142 alcohol consumption, 144 5, 147, 152, 171 2, 176, 179, 181, 196 alienation, 65 American, 142, 162, 190 Asia, 74, 118 autonomy, 49, 115 bad social capital, 24, 34, 167 Baltic countries, 13 Belgium, 45, 48, 51, 55, 57, 76 82, 92, 120, 123, 163, 206 benefits biological, 6 7, 161 material, 138 purposive, solidarity, 25, 138 biological interaction, 119 black report, 1, 86 bonding social capital, 25 6 Bottleneck, 53 Bourdieu, P., 3 4, 18 20, 27 8, 32, 34 Brazil, 73 bridging social network, 25, 28 budget cut, 38, 67 9, 117, 194 Bulgaria, 57, 76 82, 92, 120, 123, 206 Canada, 73 4, 119, 140 Care elder-care, 44, 46, 50, 52, 56, 63, 188, 197 medical, 44 Caucasian, 142 causal link, 14, 39 priority, 60, 80, 133 causality, 60, 75, 84, 111, 130 1, 159, 173, 182 3, 202 Chicago, 73 chronic illness, 36 circulatory health problem, 147 8, civil society, 22, 46, class upper-class, 9, 195 working-class, 9, 195 closed network, see network closure, 17, 28, 33 5, 39 41, 70, 72, 163, 168 9, 177, 182, 198 cognitive component, 24, 26 collective arrangement, 43 decision, 10 social capital, 31 3, 36, 38 9, 42, 62 3, 66 8, 71 5, 82 4, , 116 collectivism, 115 community, 32, 36, 70 2, 138, confounder contextual, 73, 122, 131, 194 individual, 99, 122, 127, 143, 151 2, 158 conservative-corporatist, 12 contact formal, 159, 191 informal, 30, 76, 191 context, 2, 5, 17, 30, 54, 62, 68, 72, 75, 83, 89, 109, 111, 112, 123, 128, 132, 167, 182 contextual confounder, 73, 122, 131, 194 cooperation, 22 3, 25, 49, 115, 137, 194 coordinated action, 3, 22, 23, 24, 29, 31 3, 35, 36, 37, 113 correlation, 45 9, 52, 56, 59, 75 82, 134, 186 corruption, 55, 59, 115,

22 228 Index country of birth, 7, 17, 91 3, 96, 98, 101, 103, 105, 108, 140, 142, 144, , 152, differences, 122 of origin, 140 cox regression, 122, 140, 169, 173 crime rate, 37, criminality, 5, 19, 32, 68, 141 Croatia, 57, 76 82, 92, 120, 123 cross-level interaction, , 128, 129 cross-national, 14, 51, 55 6, 60, 118, 133, 192, 205 cross-sectional, 60, 111, 124, 130, 131, 143, 145, 147, 158 9, 169, 190 1, 202, 208 crowd out, 15, 17, 42 3, 45, 47, 58 9, 134, 157, 159, 185, 187, 199 cultural difference, 130, 204 Cyprus, 57, 76 82, 92, 120, 123, 203, 206 Czech Republic, 13, 57, 76, 82, 92, 120, 123 dark side, 5 6, 19, 21, 69 70, 160, 162 3, 165, 167, 169, 171, 173, 177, 183, 195, 196, 199 de-commodification, 12, 114 demand-control model, 66, 138 democracy, 10, 33, 115 Denmark, 12, 44 5, 47 8, 51, 55 7, 72, 76 82, 92, 120, 123 Depression, 74, 119, 144 developing countries, 64 difference absolute, 121 relative, 93 diffusion of innovation, 68, 117 discrimination, 161, 167 disease cardiovascular, 72, 139 distrustful attitude, 8, 130 1, individuals, downstream factor, 61 drug abuse, see abuse Eastern Europe, 12 13, 170, 174, 176, 180 East Germany, 53, 89 ecological association, 75 economic boom, 163 capital, 24 strain, 126, 171, 172, 176, 179, 196 educational group, 71, 91, 93, 96, 99, 106, 109, 142, 194 effect collective, 62 direct, 38 9, 42, 59, 88 indirect, 42, 59 individual, 61 2, 118 egalitarianism, 89, 109, 192 egalitarian value, 89, 192 ego, 26, 183 elite group, 35 employer organization, 47, 188 employment status, 171 2, 176, 179 England, 74, 118 equality, 2, 67, 69, 82 Esping-Andersen, G., 10 13, , 128, 133, 199 Estonia, 57, 76 82, 92, 120, 123 ethnic community, 167 enclave, 167 minority, 121, 126, 127, 141, 215 network, 167 residential segregation, 183, 198 ethnicity, 35, 86, 126, 142, 162, 165 European countries, , 129, 134, 137, 140 1, 180, 185 6, 190, 203, 209, 211 welfare state, 11, 16, 56, 59, 75, 109, 111, 128 9, 135, 184, 185, Eurostat, 121 expectations, 20, 22, 50, 167 of economic prosperity, 89, 193 and normative role, 63 and of reciprocity, 54, 189 explained fraction, 169, 171, 179 explanation compositional, 62 contextual, 62