نابرابری درآمد و همبستگی در اروپا

This paper studies the relationship between income inequality, a macro-level characteristic, and solidarity of Europeans. To this aim, solidarity is defined as the ‘willingness to contribute to the welfare of other people’. We rely on a theoretical idea according to which feelings of solidarity are derived from both affective and calculating considerations – we derive competing hypotheses relating the extent of income inequality to these ‘underlying’ motivations for solidarity. Using data from the 1999 European Values Study (EVS), we apply multilevel analysis for 26 European countries. Controlling for household income and a range of macro-level characteristics, we find evidence that in more unequal countries people are less willing to take action to improve the living conditions of their fellow-countrymen. This is true for respondents living in both low- and high-income households. According to our theoretical framework, this finding suggests that, at least when measured in terms of ‘willingness to contribute to the welfare of other people’, feelings of solidarity seem to be influenced more strongly by affective, rather than by calculating considerations.

The aim of this paper is to study how inequality within countries, in particular economic inequality, is related to solidarity in Europe. For this purpose, we use a measure that directly captures the core of solidarity – willingness to promote the welfare of other people. In the literature, this core element of solidarity is often poorly captured. Solidarity has, for instance, been confused with concepts like social cohesion, social trust, and social capital, or has been equated with ‘institutionalized’ or ‘formal’ solidarity – i.e. support for welfare state intervention. Although all these concepts refer to social relations and are in some way related to solidarity, they do not provide ‘direct’ information on what motivates people to support informal or ‘institutionalized’ forms of solidarity. For instance, in studies looking into popular support for welfare state intervention, it is difficult to disentangle to what extent respondents are willing to actively promote the well-being of other people based on feelings of solidarity, or whether support for the welfare state is rather motivated by self-interest. Our first contribution to the literature is hence to bring conceptual clarity and look at a measure, which captures more closely the general idea of solidarity – ‘willingness to contribute to the welfare of others’. More specifically, we look at solidarity in terms of support for the welfare of fellow countrymen: neighbors, older people, the sick and disabled, and immigrants. We hence look at feelings of solidarity as a determinant (among other determinants) of support for ‘institutionalized’ arrangements of solidarity. The concept of solidarity has been discussed ever since the origins of social theory. Durkheim (1893/1964) already emphasized the functional necessity of solidarity for the existence and survival of social systems. Solidarity binds a society together and is a foundation for realizing collective interests (van Oorschot & Komter, 1998). Classical social theorists not only recognized the importance of solidarity for society, but also worried about how to sustain solidarity in times of rapid social change. An important theme running through Durkheim's work is how to ensure collective morality, cohesion and solidarity as societies become more strongly characterized by an organic division of labor, resulting in specialization and anomie. More recently, it has been suggested that solidarity is threatened by individualism, the expansion of markets and market liberalism, and ethnic diversity (Alesina and Glaeser, 2004 and Stjernø, 2004). In recent years, a great deal of research has focused on the reversal of the long-term declining trend in economic inequality, a social fact that seems to characterize many welfare states since the late 1970s. According to the Growing Unequal-report (OECD, 2008, p. 15), this upswing in income inequality is ‘widespread and significant, but moderate’. Explanations have focused on large-scale trends on different levels, such as technological change, globalization and the internationalization of market economies, flexibilization of labor markets (which might or might not follow from the pressures of globalization), the declining impact of unions, welfare state restructuring, and changes in household size and structure. These changes are supposed to affect either the relative sizes of population groups with a lower and higher wage or (household) income – which makes for a ‘compositional’ effect – or the relative earnings received by these groups, and more specifically the high- and the low-skilled, in return for their labor (for overviews, see Alderson and Nielsen, 2002, Atkinson, 2008 and Neckerman and Torche, 2007). Recently, Wilkinson and Pickett (2009) have argued that growing inequalities may have important societal consequences. Their central argument is that the negative impact of income inequality on societal outcomes not only runs through absolute incomes. What matters is that people are relatively more unequal to each other. In more unequal societies comparing one's own situation to other people's, results in anxiety, and lower levels of security and self-esteem. Larger differences between people trigger status competition and rising aspirations, resulting in a range of undesirable outcomes, such as higher crime and violence rates, harsher criminal justice, worse physical and mental health, declining social trust, lower educational performance, and halted social mobility. Although their methodological approach is not uncontested (e.g. Saunders, 2010), several of Wilkinson and Pickett's results have been substantiated. Examples are Lancee and van de Werfhorst (2012) on social participation and Babones (2008) on population health. Overview articles are provided by Neckerman and Torche (2007) and Thorbecke and Charumilind (2002). Although there is an abundance of research on the impact of inequality on many outcomes, it is much more difficult to test which underlying mechanisms could be responsible for these outcomes (psychological effects, level of available resources and services in a society, and social distance). In fact, more research is needed here, and we believe that for the more ‘social’ outcomes, solidarity might play a mediating role. For instance, segregated lives and larger social distance might mean that both rich and poor have less feelings of solidarity, and hence care less about how visible and invisible crimes (e.g. tax fraud) affect the community and its members. Wilkinson and Pickett (2009) also suggest, for instance, that more imprisonment combined with harsher criminal justice regimes indicate less humane attitudes and less empathy toward fellow-countrymen in unequal societies. Such micro-mechanisms, however, are often not empirically tested. We thus argue that exploring the impact of inequality on solidarity (care for fellow countrymen) is actually part of the research into causal mechanisms. This is, however, a much broader research program, while in this paper we focus on solidarity as an outcome in the first place. A second aim of this paper is hence to find out how economic inequality – in particular within-country inequality of disposable household incomes – impacts on solidarity, operationalized in terms of the willingness to contribute to the welfare of other people. Durkheim (1893/1964) suggested that large social inequities compromise solidarity, while social justice and equality (of opportunity) are important conditions sustaining solidarity. The main foundation of solidarity is the feeling of a ‘shared fate’ (Mayhew, 1971 and van Oorschot and Komter, 1998). We thus argue that income inequality increases social distance and feelings of animosity between social groups, and hence erodes feelings of identification and a shared fate with fellow-countrymen. Thirdly, starting from the different motives of solidarity identified in the literature – affective and calculating considerations – we formulate competing hypotheses concerning the impact of inequality on solidarity. Willingness to help others is not only dependent on affective considerations but also on more rational motivations – people might support the welfare of others because they realize that this will benefit themselves, or society at large. The negative externalities of income inequality, such as increasing crime and societal problems, affect all members of society. Therefore, from a calculating perspective, the overall level of support for the welfare of fellow-countrymen might increase when inequality is higher. In this paper, we try to disentangle these opposite effects of income inequality on solidarity by referring to the theoretical distinction between affective and calculating solidarity. We start with a literature review, in which we formulate a number of hypotheses concerning the impact of income inequality on feelings of solidarity in Europe. We also pay attention to the fact that in order to establish a non-spurious association between income inequality and solidarity, a number of alternative explanations which might give rise to this association should be ruled out. Next, we discuss the data and methods. Given our interest in the impact of income inequality, a country-level characteristic, on feelings of solidarity, we estimate multilevel models. After presenting our empirical results, we conclude with a discussion and some avenues for future research.

The aim of this paper was to study the impact of economic inequality on solidarity in Europe. We defined solidarity in general terms as the willingness to promote the welfare of other people, as well as the wider community. We argued that the concept of solidarity has too often been mixed up with other concepts such as social cohesion, social trust, social capital and the redistribution of resources through welfare arrangements. Although these concepts, and their measures, say something about social relations, they do not capture the core idea of solidarity. We believe that our measure – ‘willingness to do something to improve the living conditions of other people’ – better reflects the core idea of solidarity. We also believe that our research provides conceptual clarity, as solidarity might be an important mediating variable between the extent of inequality in a society and other social outcomes. In fact, studying the impact of inequality on solidarity and linking our results to the so far mixed evidence concerning inequality trends and public support for the welfare state might well prove to shed some new light on earlier findings. Furthermore, in defining solidarity we rely on a theoretical idea according to which solidarity is founded on affective (caring) and calculating (self-interested) considerations. We hypothesized that while inequality should have a positive effect on calculating solidarity, the impact of on affective solidarity should be negative. Furthermore, as inequality rises, affective solidarity diminishes: the growing mental, social and physical distance between people limits their ability to ‘recognize’ the indirect benefits to themselves of helping other people, resulting in an ‘overall’ negative influence of higher inequality on our measure of solidarity. Although existing data do not allow us to distinguish between calculating and affective solidarity and hence to test all our hypotheses, our validity check (based on those respondents who indicate that they are willing to help) indicated that our measure of solidarity is indeed driven by mixed motivations: mainly feelings of moral duty and sympathy, but also a wish to contribute to the societal good and to reciprocate with others, as well as self-interest. We found that in general solidarity is higher toward people with whom there is less social distance – the sick and disabled, older people and community members. Solidarity is much lower toward immigrants. At the same time, we found that there is notable variation between countries in how much solidarity people express. Most of this variation is explained by individual-level characteristics. Women, older respondents, married and religious people, but also more educated and wealthier people tend to express more solidarity. Societal conditions also matter. We were particularly interested in the relationship between income inequality and solidarity in Europe. We found evidence that income inequality is negatively related with solidarity. The more inequality, the less people are willing to make a contribution to improve the living conditions of others in their community, of older people, and of the sick and disabled. It is worthwhile to note that the negative relationship between inequality and solidarity toward immigrants appeared only once we controlled for welfare regime. Further analyses indicated that this relationship was obscured by an opposite pattern in the Southern-European welfare regime, mainly caused by two countries – Italy and Spain. We believe that the particular migration regime in these countries makes a difference. In Italy and Spain immigrants play an important role as domestic workers, thereby facilitating the relationship between households and labor markets. Considering the specific (positive) role of immigrants in these countries, solidarity remains relatively high despite the generally high level of income inequality (in fact, higher inequality is sometimes related to a polarization between job-poor and job-rich households). Our results are generally in accordance with what we expected. Although it was empirically not possible to distinguish between affective and calculating motives, we expected income inequality to be negatively related to solidarity. We argued that although inequality might increase calculating solidarity, it is the effect on affective solidarity that is more straightforward and persistent. People might not necessarily recognize or believe that they could improve their own well-being by improving the welfare of others. The social distance that arises from inequality is much more easily recognizable and therefore has a more straightforward consequence in reducing solidarity toward fellow-countrymen. In fact, we showed empirically that people are indeed more motivated by affective considerations (moral duty and sympathy) to help others. Therefore, it is not surprising that when inequality increases, ‘overall’ solidarity suffers. The reason why the negative association between inequality and solidarity is not very strong could be explained by the fact that different motives (affective and calculating) push people toward different directions. Furthermore, we showed that the negative relationship between income inequality and solidarity remained after controlling for individual resources and societal characteristics. Also, we found that not only the poor, but also the wealthy become less solidary in unequal societies. The finding that income inequality might reduce solidarity toward fellow countrymen is relevant from different perspectives. It suggests that when economic disparities are larger and there are relatively more people in need, solidarity decreases. Furthermore, our finding provides some support for the Wilkinson and Pickett-hypothesis. Much of their argument relies on the psychosocial consequences of inequality and the way it negatively affects human relationships. Our study rather points at a specific mechanism related to social distance. Our findings are furthermore potentially relevant for the literature on public support for redistribution. It is generally accepted that support for redistribution is determined by material and solidary considerations. We know from the Meltzer–Richard model (1981) that income inequality should increase the direct financial incentive to support redistribution. At the same time, we show here that inequality decreases solidarity. Therefore, two important determinants of support for redistribution – direct material returns and solidarity – are moving in opposite directions. This might explain why empirical papers find inconsistent results when studying the relationship between income inequality and support for redistribution (Finseraas, 2009, Kenworthy and McCall, 2008 and Lübker, 2007). It might be that while there is a material incentive to support redistribution, people are less fond of the idea to share resources with their fellow-countrymen. Particularly the well-off who do not receive direct material benefits from the welfare state are more likely to support the system out of solidary considerations. Thus, the fact that inequality reduces solidarity among the well-off might be particularly important when we think of support for redistribution and collective welfare arrangements. Therefore, our paper points to a general need to pay more attention to different human motives. Support for redistribution clearly combines two important and yet different motivational aspects–purely material and solidary considerations. Given our difficulties to disentangle these effects with existing survey data, questionnaire designers might want to identify items allowing us to separate affective and calculating motives of solidarity. Lastly, we cannot ignore the possibility that causality is reversed. So far, we argued that increasing income inequality weakens solidarity toward fellow countrymen. However, it might also be true that in countries where people feel less solidary toward their fellow-countrymen, inequality is more likely to emerge and persist. We cannot exclude this option. However, our goal was to test the theoretical idea that social distance in the form of economic disparity would be related to a lower level of solidarity. Given the data available we can test empirically only those hypotheses derived from the literature. We find evidence that there is a negative relationship between income inequality and solidarity. Therefore, further research is needed to test the direction of causality. However, as others have suggested (Bénabou & Tirole, 2006), we believe that the causality runs in both directions.