Globalisation and trust in Europe between 2002 and 2018

Verhoeven, Loesje; Ritzen, Jozef M. M. Article Globalisation and trust in Europe between 2002 and 2018 Research in Globalization Provided in Cooperation with: Elsevier Suggested Citation: Verhoeven, Loesje; Ritzen, Jozef M. M. (2023) : Globalisation and trust in Europe between 2002 and 2018, Research in Globalization, ISSN 2590-051X, Elsevier, Amsterdam, Vol. 7, pp. 1-15, https://doi.org/10.1016/j.resglo.2023.100142 This Version is available at: https://hdl.handle.net/10419/331078 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Research in Globalization 7 (2023) 100142 Available online 8 July 2023 2590-051X/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Globalisation and trust in Europe between 2002 and 2018 Loesje Verhoeven , Jo Ritzen * Maastricht University, Boschstraat 24, 6211 AX Maastricht, The Netherlands ARTICLE INFO Keywords: Globalisation Institutional trust Interpersonal trust Diversity Inequality Economic decline Government expenditure Government intervention ABSTRACT Are institutional trust and interpersonal trust threatened by globalisation? For nineteen countries in Europe, using a fixed effects model for a panel data set relating globalisation to several economic and social macro variables, like income inequality and diversity, to average institutional and interpersonal trust derived from responses in European Social Surveys, we do not find any significant relation between the relatively moderate globalisation of the first two decades of the 21st century on average interpersonal and institutional trust. At the same time, occurrences of economic decline in a country are negatively related to institutional trust. GDP has a positive effect on both institutional and interpersonal. Combining the macro factors with the individual traits of respondents using pooled repeated cross-sectional data demonstrate the dominance of personal characteristics in individual levels of trust, with only institutional quality emerging as a macro variable which is significantly and positively related to trust, especially for the Socio-Economic Groups 3 to 7 (of the eight groups distinguished). Those who are born in the country exhibit higher levels of interpersonal trust, in particular in the higher SES groups 4–7, but show significantly lower institutional trust for the SES groups 0–2. Age is negatively related to institutional trust for all SES groups, but positively related to interpersonal trust for SES groups 4–7. These findings appear to imply that those who are concerned with the level of institutional trust in the population as a basic requirement for democracy in Europe should focus on the quality of institutions and not on globalisation. Introduction We aim at deepening the understanding of the relationship between globalisation and trust of the population in institutions and in each other for a relatively homogeneous set of high-income countries in a rather recent period. Globalisation has often been viewed as threatening trust in institutions and reducing interpersonal trust. “Social theories tend to equate globalisation with the general erosion of societal bonds at the national level. Social cohesion is undermined, they claim, by a variety of social trends, including the erosion of national/state identities, the rise of individualism and increasing structural inequalities in societies.” (Green, Janmaat & Chen, 2011, p. R6). We understand trust as a “state variable” in the state space approach (Zadeh and Desoer, 1963). Trust can accrue or decrease due to external circumstances and due to time (age). The loss of interpersonal trust and trust in institutions is often – rightly or wrongly – recognised in higher income countries in the surge of populism and political polarisation (Putnam, 2020; Swank & Betz, 2018; Grechyna, 2016). The importance of trust for social cohesion and therefore for transaction and administration costs in a society is widely recognised (see for example: Easterley, Ritzen & Woolcock, 2006, p. 105 or Putnam, 2000). Social cohesion has been a popular concept, ever since its introduction by Durkheim (Durkheim, 1893). Nowadays, the relation between social cohesion, trust and economic growth is the subject of substantial discussion and extensive empirical research. Bjørnskov and M´ eon (2013) report evidence that trust is the missing link relating education, the quality of institutions, and economic development. Both the quantity and the quality of investment in the economy may be raised by social trust. In accordance with that notion, Dearmon and Grier (2009) observe, using growth regressions, that the marginal impact of investment on growth is larger in more trusting economies. Our question is whether globalisation might have impacted trust if most other major potential macroeconomic or social contributors to trust are statistically kept constant, while understanding the individual contributors to trust, like individual income, education, age, gender, religion, being born in the country or not, and gender, as well as the country level contributors of diversity, inequality, economic decline, institutional quality, GDP per capita and Government expenditure. We are interested in globalisation being an exogenous force in relation to * Corresponding author. E-mail address: [email protected] (J. Ritzen). Contents lists available at ScienceDirect Research in Globalization journal homepage: www.sciencedirect.com/journal/research-in-globalization https://doi.org/10.1016/j.resglo.2023.100142 Received 23 November 2022; Received in revised form 26 May 2023; Accepted 7 July 2023 Research in Globalization 7 (2023) 100142 2 trust, even though we recognise that Governments in principle can step out of trade agreements or international trade in general as well as deciding on new agreements in the World Trade Organization (WTO). Globalisation increases the interconnectedness of countries, through the increasing flow of goods, services, capital, and labour. This may be associated with increasing diversity – as a result of migration – as well as increasing economic inequality (Ritzer & Dean, 2015). In turn, inequality and diversity may well decrease trust to the extent of connectedness, solidarity, and sense of belonging of society and its citizens (Paskov & Dewilde, 2012; Manca, 2014; Beugelsdijk and Klasing, 2016), as they contribute to social and economic divisions within society (Pervaiz, Chaudhary & van Staveren, 2013). Yet, the causality between trust and inequality is more complicated, as inequality is endogenous to the system: in high trusting societies income inequality is lower (Bergh and Bjornskov, 2014), presumably reducing the effect of globalisation on trust through inequality. We have set ourselves the task to disentangle the different forces in the dynamics of globalisation and trust, providing deeper insight in the analysis of previous writers on the topic, like Putnam (2020). He finds for the US a positive correlation between trends in social cohesion, economic growth, political harmonisation, and communitarianism, and sees all these four trends decline since 1970 – the start of a new stage of globalisation as a result of the abolition of the gold standard and the increase in trade agreements. Snower & Bosworth (2021) highlight how individualism (as contrary to communitarianism) has a negative effect on trust, leading to fragmentation and increased polarisation in society between citizens with different education levels. They argue that unevenly distributed productivity is the driving force of fragmentation. This thinking fits into the notion that globalisation has often been embraced as a source of productivity increases (Heimberger, 2020), even though it ends up in unevenly distributed levels of productivity. Our paper provides an assessment of the relation between trust and globalisation for nineteen European countries for the period 2002–2018. The concurrence of increasing globalisation with a decline in trust, found for the US, is often suggested to be also present in European countries. In consequence, globalisation is claimed to be one of the determinants for polarisation (Swank & Betz, 2018). However, by focusing solely on the group of rather homogeneous European countries we hope to find marked differences between developments which have been analysed for a wide range of market countries (as for example by Zak and Knack, 2001) and for our sample and time, for a period when globalisation first increased (2002–2010) and subsequently stalled or decreased (2002–2018) for the countries concerned. We also seek evidence to establish whether Government intervention has indeed been a factor in the relation between globalisation, its impact on society, and trust, as is suggested by, for example Rodrik and Stantcheva (2021a/b). They have developed a framework consisting of nine different ways in which governments can intervene, shaped by choices of when to intervene and who to target with the intervention. Their main argument is that intervening in the so-called production stage will ensure most inclusivity and equality within society, which in turn can foster trust (Ozili, 2020). In section 2 we elaborate on the relation between trust and globalisation, taking diversity, inequality and Government intervention into account, leading to the model used for the econometric analysis. In section 3 we present the data used with a cursory examination that might have an effect in the relation between trust and globalisation. In section 4 the results are presented, with in section 5 our conclusions and a summary. Trust in each other and institutional trust Trust in each other and institutional trust We are interested in trust as a cornerstone of social cohesion which in turn may have a strong impact on (sustainable) economic growth and happiness in the population (Easterly, Ritzen & Woolcock, 2006). Social cohesion rests on a communal agreement that exists without any force (Green, Janmaat & Chen, 2011). Easton (1975) argues that trust is the component that ensures support in this communal agreement which forms society. Trust in each other and the feeling of belonging appear to be two sides of the same coin of social cohesion (Soroka, Kanting & Johnston, 2004). A certain level of trust in each other is needed for governments to be able to rule and provide as well as enforce the institutional framework of a society. When this level of institutional trusts is not reached, governments would need to use force to actually govern (Easton, 1975; Marien & Hooghe, 2011). Trust in each other and trust in the institutions in a society are, as a result, intertwined issues as we shall further investigate for our set of countries. There is a distinction between interpersonal trust and trust in institutions (Seligman, 1997; Besic, 2021). Sometimes this distinction is interpreted as the difference between ‘particularised’ and ‘generalised’ trust (Putnam, 2000; Newton, 1999; Uslaner, 2002). Where particularised trust entails trust in “known others”, generalised trust involves “unknown others”, like the government, “out-groups” in society and so on. Yet, it is argued that particularised trust extends to generalised trust (Glanville & Shi, 2020), whilst the latter is particularly necessary in the globalised society (Eriksen, 2007). Both “unknown” interpersonal and institutional trust are forms of generalised trust. Measuring trust The “grandfather” of trustmeasurement is surely Inglehart who in the 1970 s introduced the kinds of measures later found in the WVS (World Value Survey), with simple questions. Social trust is measured with the standard question “In general, do you think most people can be trusted?” The WVS specifies trust in ‘people’ with the following categories of ‘people’: your family; people in your neighbourhood; people you know personally; people you meet for the first time; people of another religion; people of another nationality. The WVS has received a number of followers, like the Organisation for Economic Co-operation and Development’s (OECD) value survey and the European Social Survey (ESS), with minor differences. The ESS (which we will use) employs the following three questions: – “Most people can be trusted or you can’t be too careful” – “Most people try to take advantage of you, or try to be fair” – “Most of the time people are helpful or mostly looking out for themselves” There are also differences between the institutional trust measurement of WVS and ESS. The ESS uses the word ‘trust’, while the WVS uses ‘confidence’. Both surveys use questions on the confidence in parliament, judiciary system, police and political parties (Inglehart and Welzel, 2005). Additionally, in the ESS one of the items is about the trust in the ‘legal system’, while in the WVS the corresponding item is about confidence in the ‘justice system’. In 25 European countries both the WVS and ESS were held between 2017 and 2018. Besic (2021) demonstrates that for these years the relation between trust and socio-demographic variables is similar for both surveys. From a factor analysis he concludes that the different responses on the trust question have in both surveys two different factors which can well be labelled “interpersonal trust” and “institutional trust”. However, limitations of these trust measurements have been recognised and underlined by Sapienza, Toldra-Simats and Zingales. (2013): they argue that the responses to these questions are more driven by “beliefs” rather than by preferences. Globalisation We see many different aspects of globalisation at work. The past fifty years or so have seen a tremendous growth in trade and cross-border L. Verhoeven and J. Ritzen Research in Globalization 7 (2023) 100142 3 finance, including foreign direct investment. Migration flows have been increasing, while informational diffusion has skyrocketed, and cultural exchanges have increased (Ritzer & Dean, 2015). These processes of globalisation have left their imprint on societies (Eriksen, 2007). Trade openness has been viewed as an important incentive to raise productivity as a means to become world-wide competitive (Heimberger, 2020). Although formally it is Governments that have decided on increased economic openness and international trade agreements, we can safely assume that globalisation is an external force to the country: it is exceedingly hard and requires a highly autocratic regime to enforce closeness to external trade and finance (Ritzer & Dean, 2015; Thompson & Hirst, 2002; Eriksen, 2016). Globalisation in trade and finance has brought economic gains. The integration into a world-system as a result of globalisation has made the world, its countries, regions and people become more interdependent (Frenk, G´ omez-Dant´ es & Moon, 2014). Early hopes in the post second world period that interdependency would contribute to world-wide peace (Lewin, 1948), seemed to indeed to materialise, until Russia invaded Ukraine, preceded by sanctions in trade and finance in the Middle East and Far East. Trust and globalisation: The model “Trust arrives on foot but leaves on horseback”. This popular expression captures that trust is a state of mind of a person which is in flux under the influence of experiences by that person. People experienced “globalisation” in the period 2002–2018 in many different ways presumably without making the direct connection between globalisation and their trust in each other or in institutions. They experience the changes in the labour market, they witness increasing diversity as a result of migration and notice the relative changes in the income and wealth distribution. They also may experience economic decline and may respond by lowering their level of trust (Habich and Nowotny, 2017). These are the four explanatory variables we hypothesise to have affected trust in our sample in the period 2002–2018. As a fifth variable we introduce Government expenditures. Can they mitigate the impact of globalisation, income inequality, diversity and economic decline similar to the theory of Rodrik and Stantcheva (2021a/b), who argue that Government intervention can increase inclusivity – and its subsequent positive effects on society? Direct effect of globalisation on trust Globalisation may cause people to feel less connected to the specific geographical territory that represents “their” nation, and more connected to a cosmopolitan identity (see, for example, Eriksen, 2007). This may harm the social bond between state and citizens and may undercut the social bond that ensures citizens to comply to the unwritten rules of society (Anderson, 1996). Simultaneously, globalisation reduces the role governments have in their administration and the political sphere (Magalh˜ aes, 2014). Governments are becoming more subservient to transnational powers in economics, trade, and governance. This refers to transnational agreements on trade and monetary issues, but also to humanitarian and development affairs. Global and collective challenges like adherence to human rights and unbalanced development, require global (or at least transnational) action, and therefore, states might be more productive in a global scheme contrary to a national or local scheme. However, these agreements can reduce the national autonomy of states, as is often remarked in European Parliaments as a complaint about the EU (Bulmer, 2020). They are becoming a part of an international governance scheme in the globalised era (Ritzer & Dean, 2015). This might negatively impact the level of institutional trust of citizens, depending on other factors like the level of education of an individual as was empirically established by Fischer (2012) for the period 1981 to 2007. Furthermore, it has been argued that globalisation has challenged the welfare state. The reason behind this is that welfare states are built on trust and the feeling of belonging specifically (Soroka, Johnston & Banting, 2004). The willingness of citizens to tax themselves for an extensive welfare state may then decline. In principle, the modelling of this mechanism is one where changes in “globalisation” (in whatever measure) bring about changes in trust. Inequality and trust The first mechanism we recognise in our model is that income inequality may impede trust, while at the same time globalisation is likely to increase inequality (Rodrik & Stantcheva, 2021b; Heimberger, 2020). Globalisation has been accompanied by increased international competition with a substantial potential income inequality as a result (as was predicted by the Stolper-Samuelson theorem of international trade (Davis, 1996)). Globalisation implies growing competition on the world market, creating the so-called “winners” and “losers” of globalisation, where the poorer cannot reap the rewards of globalisation due to a shortage of resources, and the wealthy can make use of transnational benefits to accumulate more wealth (Ritzer & Dean 2015; Stiglitz, 2017). This results in a decrease of social mobility and an increase in inequality (Kriesi et al., 2008), as well as in an increase in vulnerable employment. In our model we introduce globalisation and inequality separately, recognising that globalisation’s impact on trust is far more than only that of inequality. Trust and (social) inequality have been widely discussed in literature (e.g., Stiglitz, 2017; van Staveren & Pervaiz, 2015; Wilkinson & Pickett, 2017; Pervaiz, Chaudhary & van Staveren, 2013). Welfare states aim at keeping the growth of income inequality in check but may not always be successful in this respect (also not within the group of countries considered in this paper). Income and social inequality harms trust through social stratification (Wilkinson, 1997; Vergolini, 2011) in other people and in institutions such as the government (Boarini, Causa, Fleurbaey, Grimalda & Woolard, 2018). Bergh and Bjørnskov (2014) have examined the direction of the causality between trust and income inequality (without explicitly considering globalisation). They show that under plausible circumstances, trust aids the creation of welfare states, which reduce inequality. This is line with the “compensation hypothesis”: Governments may take on a bigger role in times of globalisation to compensate for the heightened vulnerability and risks (for some) as a result of the increased competition on the labour market (Heggem & Jakobsen, 2016). All in all, we would expect that – in line with the literature – inequality has a negative impact on trust, even if globalisation is already taken into account. Trust and diversity Our second assumed aspect through which globalisation affects trust, is diversity. Migration has become easier in the recent past as a result of increasing globalised (social) networks, more and easier global communication, and the spread of modernity and development in general. In particular, people have become abler at migrating over longer distances. People do not merely move to neighbouring countries, but also intercontinentally. Where people from the same regions may be relatively alike, this is not as true for people who originate from different continents. Diversity, rather than migration itself, is increasing due to globalisation (de Haas, Castles & Miller, 2020). Increased diversity has implications for social cohesion and trust, as has been proposed by the so-called “diversity thesis” (Pervaiz, Chaudhary & van Staveren, 2013). Putnam (2007), for example, describes how ethnically heterogeneous communities are less connected and have lower levels of trust: in diverse neighbourhoods, people would retreat from social life. This could be explained by social-psychological processes natural for human beings. As the world is inherently too complex for individuals to grasp completely and perfectly, basic cognitive mechanisms help individuals to comprehend their (social) L. Verhoeven and J. Ritzen Research in Globalization 7 (2023) 100142 4 surroundings (Gray & Bjorklund, 2014), by using social categorisation and identification. These processes refer to one’s need of an “other” to be able to understand one’s own identity. By comparing ourselves to others, we can attribute characteristics that do not belong to ourselves (Simmel, 1908/1971). This is often followed by attaching negative attitudes to the different characteristics of the “other” (Eriksen, 2007; Taylor, Moghaddam, Gamble & Zellerer, 1987). This is in line with the social identity theory (SIT), which shows how divisions in society based on selfcategorisation produces positive in-group attitudes and negative outgroup attitudes. Such attitudes cause stereotyping and discrimination (Tajfel, 1970). It can also result in scapegoating migrants (Komisarof & Leong, 2020), which is often observed in the polarised political landscape (de Haas, Castles & Miller, 2020). Hence, globalisation can affect trust through increased diversity and cognitive processes of social categorisation and identification. In other words, it can result in negative attitudes towards ‘others’. The coefficient for the relation between diversity and trust is then expected to be negative. However, although the direct theorised relation between diversity and trust is found to be negative, research has shown some mediators that can neutralise this negative relationship: intercultural contact and social ties between people can decrease the initially negative effect of diversity on trust (Pervaiz, Chaudhary & van Staveren, 2013), as is also well known as the contact theory (Pettigrew & Tropp, 2008). Trust, the labour market and economic decline The third mechanism for globalisation to impact trust is the labour market. The countries we study have chosen for opening up in order to increase productivity and the competitive position of their production (Heimberger, 2020). They have chosen to privatise public production wherever possible. Globalisation has caused a ‘new geography of jobs’, changing compositions of low or high-skilled jobs (Iammarino, Rodriguez-Pose & Storper, 2019). Therefore, we incorporate manufacturing jobs as a % of all jobs as a control variable in our model. Snower & Bosworth (2021) show that these forms of globalisation are likely to result in economic and social fragmentation as a result of diverging capital caused by an uneven distribution of productivity. It is equally likely that social fragmentation is associated with less interpersonal trust and less institutional trust (Soroka, Johnston & Banting, 2004). Consequently, we have also included economic decline in the model as potential contributor to lower levels of trust. The relationship between institutional trust and economic decline is dependent on an individual’s position (Dotti Sani & Magistro, 2016). The so-called “losers” of globalisation, who run a higher risk of unemployed in times of crises, might have lower levels of trust. This causes an asymmetrical decline in trust, where citizens with higher risks have decreasing levels of trust compared to those exposed to less risk (in the labour market) in case of an economic decline (Foster & Frieden, 2017). This is in line with the finding that negative attitudes towards globalisation can be explained by the length and depth of the economic crisis (and increasing income inequality over time) (Harms & Schwab, 2020). A concrete example is the economic crisis in 2007, which negatively impacted many people. Here, increased inequality was associated with a negative effect on both institutional and interpersonal trust. Especially tensions between social groups – rich and poor; young and old; different ethnic groups – were observed to be rising (Andrews, Downe, Guarneros-Meza, Jilke & Van de Walle, 2013). Perhaps this aspect is measured through inequality in our model. To be sure we have added “economic decline” as a separate measure in statistically “explaining” trust. We expect economic decline to negatively affect trust. Furthermore, we have added GDP to our model, as this is an important measure in economic-political research as well as has been shown to have a relation to trust (Dincer & Uslaner, 2010). Government intervention The welfare state is a mechanism in which people pay taxes for the government to be able to provide the services they do, e.g., social protection, education, healthcare, housing, and community building. As it is based on a redistributive system, people with higher incomes contribute relatively more compared to people with lower incomes. This mechanism heavily relies on trust; trust from people in others and trust in the government that the mechanism works in a proper, equal, and efficient way (Dethlefsen, Emmanouilidis, Mitsos, Primatarova & ˇ Spok, 2014). Without this trust and support in the government, their efficiency declines, hampering the welfare system (Bjornskov & Svendsen, 2013). So, as argued, trust is needed in order for Governments to govern. At the same time, Governments can possibly have a mediating impact on threats of decreasing social cohesion and trust by implementing policies focused on building trust and cooperation between citizens (Cilingir, 2016). We hypothesise that an increase in government expenditure can lead to increases in the quality and inclusivity of the public service system. This might bring people to become more trusting (Alan, Baysan, Gumren & Kubilay, 2021; Saint-Sup´ ery Ceano-Vivas, Rivera Lirio & Mu˜ noz-Torres, 2014). However, it appears that low-income citizens demand less income redistribution from the welfare state, when income inequality increases (Kim, 2019). This might be explained by the (low) levels of trust in the government, as well as the absence of specific policy interventions associated with inclusion (as defined by Rodrik and Stantcheva (2021a)). This shows the role Governments can play in building trust. It is therefore relevant to consider that besides government expenditure as such, the quality of governments could also make a significant difference (Rodríguez-Pose & Garcilazo, 2015; Heimberger, 2020). In particular, government expenditure on education could improve the level of trust. Intervention of Governments can prevent market failures and increase efficiency due to the positive externalities related to education (Gruber, 2015). Education has more positive effects for society than merely the return of education for the students themselves (Gruber, 2015). Government intervention in education can enhance the returns of education in two ways: by improving the quality of education and by making education accessible for students of all different backgrounds. It enables the development of talents and skills, which causes innovation and creates individuals who are a “more responsible and participating citizen who is better able to contribute to the economic production process.” (Ritzen, 2017, p. 1). Well-developed social skills, as trust, reciprocity, and cooperation, enable individuals to communicate and interact efficiently in a diverse context, and thus are essential to building inclusive and trusting communities (Putnam, 1993; Alan, Baysan, Gumren & Kubilay, 2021). Besides education, socio-cultural and redistributive policy can likewise negate the hypothesised negative effect of globalisation on trust. Socio-cultural policies aimed at building trust and cooperation in diverse communities can, as argued before, overcome the initial negative effect of heterogeneity in society on trust. Social services are meant to support the redistributive function of the state and are therefore assumed to contribute to social cohesion and trust. Thus, fiscal policy can play an important role to enhance these aims (Saint-Sup´ ery Ceano- Vivas, Rivera Lirio & Mu˜ noz-Torres, 2014). Rodrik and Stantcheva (2021a/b) suggest that Government intervention can decrease inequality and increase inclusivity – which, according to them, relates to various other economic, political and social dimensions, by general and specific expenditures. More specifically, they identify three target groups: low-, middle- and high-income individuals, as well as three stages of intervention: pre-production, during production and post-production. Education is an example of a preproduction intervention, whereas social benefits are post-production interventions. With these different kinds of interventions, Government expenditure becomes more nuanced and can have different effects depending on the moment and direction as identified in this matrix. As L. Verhoeven and J. Ritzen Research in Globalization 7 (2023) 100142 5 argued by the authors (Rodrik and Stantcheva, 2021a/b), purposefully aimed Government intervention can e.g., smoothen the effects of globalisation on social issues. Spending on public services may contribute to feelings of citizenship (Delhey and Newton, 2005). Individual characteristics We shall explore how individual trust responds to globalisation and other macro-economic factors, by considering individual trust itself while taking the following variables into account: - Born in country - Individual income group of respondent (decile) - Years of full-time education completed - Degree of religiousness (Scale 0–10). - Age of respondent - Gender - Political position (a left-to-right scale) The model Our model is an extension of a socio-economic profile of trust, expanded with macro variables on globalisation, inequality, diversity and economic decline. The data on variables referred to in the models below will be further specified in paragraph 5.5.2. The socio-economic profile is derived from the simple fixed effects regression model (Stock & Watson, 2015): Yijt =β0+β1Zijt + ε ij (1) Where:i denotes the country, j the individual, t the yearY ijt denotes the level of trust; and Z ijt denotes a vector of characteristics of individual j in country i at time t (as indicated in section 3); and ε ij denotes the disturbance term. As argued, we expect country level determinators to also play a role in the level of trust, so that: Yijt =γ0+γ1X1it +γ3X2it +γ4Zijt +ζij (2) WhereX 1it denotes globalisation;X 2it denotes a vector of inequality, diversity, economic decline, and government expenditure;γ 0 denotes the intercept term which encompasses country-specific variables;and γ 4 denotes (vectors of) coefficients, while ζ ij denotes the error term. This regression model will allow us for testing the relationship between globalisation and trust, in the presence of an attribution of individual trust to individual characteristics, and furthermore to analyse whether income inequality, diversity, economic decline and government intervention are relevant. Methodology and data Research strategy Our method of analysis is a fixed effect regression model. The fixed effect model is preferred over ordinary least square (OLS), as fixed effects could mitigate a possible omitted variable bias (OVB). OVB is problematic, as this implies that some of the unobserved variances of the dependent variable will be explained by the error term (Stock & Watson, 2015). In our research strategy we make three different steps: - An analysis with country panel data (averages on trust from the ESS) and country data on globalisation, inequality, diversity, economic decline, and government expenditure. This dataset is a panel data set – where i =countries and t =year. This allows us to control for both entity and time fixed effect, i.e., we control for average differences across our panels for any (un)observable predictors, like country or time specific events (Stock & Watson, 2015). - An analysis of the profile of individual trust, using only ESS data. As ESS data is collected every other year with a different group of respondents, this constitutes a pooled repeated cross-sectional data. Observations are likely to be correlated within a unit (countries in our case) as a result of an unobserved cluster effect, so in order to control for heteroskedasticity over time and to correct the error term to these different pools (Cameron & Miller, 2015; Wooldridge, 2010), we cluster our model by country. This way, we control for heterogeneity across units. We also incorporate a year dummy in our model and take those fixed effect into account to reduce the potential of OVB (Stock & Watson, 2015). - An analysis with a mix of individual responses on trust combined with the country data on the macro level we have analysed in our SEM. Here, we use two levels of measurement, i.e., we conduct a multilevel analysis, when working again with a pooled repeated cross-sectional dataset, and using the same measures to control for fixed effects as described for the analysis of individual trust. We also introduce socio-economic classes to regroup the population in terms of income and education level. Data The data cover nineteen European countries: Austria, Belgium, Czech Republic, Denmark, Estonia, Finland, France, Germany, Hungary, Ireland, Netherlands, Norway, Poland, Portugal, Slovenia, Spain, Sweden, Switzerland, and the United Kingdom, over a period from 2002 until 2018 (every other year). As the aim is to create a sample with a relevant coverage, we have included as many countries as available from the ESS data that were included in the survey at least 8 out of 9 rounds (to prevent missing data). This formed the basis of our sample, to which we have adjusted the data from other sources. Trust and individual characteristics: ESS data The ESS (European Social Survey, 2018, a,b,c,d,e,f,g,h,i) has conducted a cross-sectional survey in European countries every other year from 2002 onwards. These rounds will be used for the dependent variable of trust. The ESS data has different questions related to trust (see Table 1), which are all coded in the same way: they are scaled from 0 to 10, where 0 represents the lowest score on trust – i.e., no trust at all, and 10 the highest score on trust. We use the answers to the questions: To reduce the number of variables on institutional trust, we use a principal component factor analysis. A principal component factor (PCF) analysis is chosen rather than a normal factor analysis because the values of the different variables do not vary as much (Di Franco, 2013; Acock, 2016). Moreover, the rotated factor analysis, inherent in PCF, corrects a possible bias that can be present in a general factor analysis (Bryman, 2012). When conducting the PCF on the above listed variables – excluding trust in people, we see an outcome of one factor with an eigenvalue of 3.73. The loadings of the variables are higher than 0.7. This passes the minimal threshold of 0.3, which is needed to ensure the loadings on the factor capture the variance of each variable into the factor (Bryman, 2012). The alpha of these variables combined is 0.8751, Table 1 Measures for trust as derived from the ESS data. Variable name Variable label ppltrst Most people can be trusted or you can’t be too careful trstprl Trust in country’s parliament trstlgl Trust in the legal system trstplc Trust in the police trstplt Trust in politicians trstep Trust in the European Parliament trstun Trust in the United Nations L. Verhoeven and J. Ritzen Research in Globalization 7 (2023) 100142 6 which passes the threshold of 0.7 (Hair, Black, Babin, Anderson & Tatham, 2006). This means the six different measures of institutional trust can well be merged into one variable. This is also found for the European Value Survey, leading to the suggestion that in future research confidence in police OR in legal system should be removed from the scale construction as they are highly collinear (Besic, 2021). The individual data used in the socio-economic profile are previously described in paragraph 4.6. Globalisation: KOF index We use data from the KOF Swiss Economic Institute (Gygli, Haelg, Potrafke & Sturm, 2019) as the index of globalisation (in line with for example, Potrafke, 2015; Villaverde & Maza, 2011; Anderson & Obeng, 2021; Sangha & Riegler, 2020). The different components of the KOF- index are economic, social, and political globalisation. Economic globalisation incorporates trade and financial globalisation and measures, for example, foreign direct investment, while informational and cultural globalisation are used to index the social dimension of globalisation. Political globalisation is indexed by, amongst others, the number of embassies and international non-governmental organisations (NGO), international treaties and United Nations peace keeping missions. These three main dimensions all have the same weights of 33.3% in the composite KOF index. As our sample is relatively homogenous, we firstly have to establish whether they indeed have a similar experience with globalisation, or where specific differences in the level of globalisation occurs. There is quite some between group variation: countries as Poland (76.882), Slovenia (77.508) and Estonia (80.364) have a relatively lower score on the globalisation index, whilst countries as Belgium, the Netherlands, Sweden, Switzerland and the United Kingdom have higher levels (>88). Within-group variation also differs from case to case, where Slovenia and Poland have the highest standard deviation. Plotting this to explain this deviation shows that this high standard deviation represents an increase in the level of globalisation (Appendix 1). A plot of the level of globalisation over the years for all countries, shows that in general, globalisation has increased (Fig. 1, Appendix 1). However, the increase takes place in the years 2002 until 2010/2012 and is less, or even becomes negative in later years. Inequality Inequality can be captured with different statistical measures. In this paper, inequality will be measured with the so-called Gini-coefficient (as is done in e.g., Faustino & Vali, 2013; Almas & Sangchoon, 2010; Neutel & Heshmati, 2006). The Gini-coefficient used in this research is derived from the World Bank (2022). As the data had some gaps (10/106 missing values), we interpolated the variable to be able to use a complete panel for our macro analysis. Diversity Diversity is measured by the percentage of foreign-born individuals in a country, using data from the OECD (2022). As described in the literature, because of an increase in social globalisation, diversity increases as migrants are moving further away. Using foreign-born population as a proxy for diversity has been done before in socio-economic (e.g., Ottaviano & Peri, (2006) and political research (e.g., Wright, 2011). A limitation to our study is that presumably the country of origin of the foreign-born population may matter. Most studies show a difference in the appreciation of diversity between persons born inside Europe and outside of Europe. We interpolated the variable of diversity as well, along the same line of reasoning for the Gini variable. For diversity, there were 5 missing values in the panel. Economic decline The Quality of Governance (QoG) dataset gives us an indicator on economic decline, which measures different factors: “per capita income, gross national product, unemployment rates, inflation, productivity, debt, poverty levels, or business failures, […] sudden drops in commodity prices, trade revenue, or foreign investment, and any collapse or devaluation of the national currency”. (Teorell, Sundstr¨ om, Holmberg, Rothstein, Alvarado & Dalli, 2021, p. 270). Moreover, the QoG dataset has a variable on manufacturing: employment in manufacturing as percentage of total employment. This control variable will be included as a measure of the changes in labour market caused by globalisation as described in 4.4. Government expenditures The data on Government expenditure is derived from Eurostat (2021). The data represents total Government expenditure, expressed as a percentage of GDP. Eurostat data is a recognised dataset that is widely used, also within research of the relation between expenditure and trust and social cohesion (e.g., Wishlade, Gross, Yuill, Gorzelak, Kozak & Mendez, 2010; Alonso, 2015), which is why Eurostat data will be used in this research to investigate whether government expenditure mitigates the relation between globalisation and trust. Quality of institutions Besides government expenditure, the quality of expenditure could also play a role. We use a measure found from the QoG dataset (Teorell, Sundstr¨ om, Holmberg, Rothstein, Alvarado & Dalli, 2021), but originally derived from the International Country Risk Guide (PRS group, 2021). It represents the mean value of the International Country Risk Guide variables on corruption, law and order and quality of bureaucracy, and is scaled from 0 to 1, where higher values indicate a higher quality of institutions. Again, this measure has been chosen to work with in this research, as it is used as a variable to measure institutional quality in relation to social cohesion, for example in Easterly, Ritzen and Woolcock (2006) and Dragolov, Ign´ acz, Lorenz, Delhey and Boehnke (2013). Statistics In Table 2 we present the maximum, minimum and average scores of the variables used in our analysis of the socio-economic profile of trust, Table 3 presents the summary statistics on the individual measures with summary statistics on the country level in Table 4. Table 2 Differentiation of globalisation between and within countries. Level of globalisation (KOFGI index) Mean Standard deviation Min. Max. Austria 87.681 0.933 85.777 88.709 Belgium 88.858 1.447 86.168 90.463 Czech Republic 82.291 2.331 77.571 84.883 Denmark 87.428 0.663 86.266 88.273 Estonia 80.364 2.417 72.615 82.905 Finland 86.198 1.231 83.618 87.701 France 85.888 1.673 82.838 87.692 Germany 86.660 1.662 83.245 88.829 Hungary 83.210 2.335 77.676 85.361 Ireland 85.025 1.079 83.477 86.240 Netherlands 88.118 1.952 84.669 90.683 Norway 84.481 1.065 82.512 85.965 Poland 76.882 3.236 69.933 80.582 Portugal 81.162 1.922 77.421 84.881 Slovenia 77.508 3.465 70.050 81.207 Spain 83.210 2.335 77.676 85.361 Sweden 88.467 0.874 86.996 89.720 Switzerland 88.751 1.642 86.792 90.984 United Kingdom 88.026 1.233 85.627 89.390 L. Verhoeven and J. Ritzen Research in Globalization 7 (2023) 100142 7 Results Country level measures influencing trust Table 5 presents the output of our fixed effects model with annual country averages of different macro variables. The measures of trust are also averages per year per country, estimated from the individual responses from the ESS. The results can be interpreted quite straightforwardly. They show that: – GDP has a positive and significant effect on both institutional and interpersonal trust, where the effect is bigger on institutional trust. This is in line with what is often argued in existing literature. –Economic decline is negatively and significantly related to institutional trust but has no relation to interpersonal trust. – Total expenditure has a negative effect on interpersonal trust, while it does not seem to affect institutional trust. Expenditure in education has no relation to either measure on trust. The other variables have no relation to the average level of trust in a country. Table 3 Summary statistics – individual measures. Variable name Variable label Source N Mean Standard deviation Minimum value Maximum value Individual measures inst_trust(individual) Institutional trust (factor variable) ESS 275,377 5.011 1.919 0 10 ppltrst(individual) Interpersonal trust ESS 317,986 5.243 2.381 0 10 income Income in deciles ESS 246,613 5.398 2.665 1 10 eduyrs Years of full-time education completed ESS 315,558 12.528 4.098 0 60 rlgdgr How religious are you? Scale 0–10. ESS 316,398 4.425 3.019 0 10 agea Age of respondent ESS 317,700 48.306 18.598 14 123 gndr Gender (1 =male) ESS 318,709 0.472 0.499 0 1 brncntr Born in country ESS 318,598 0.907 0.290 0 1 lrscale Political ideology (scale 0 =left −10 =right) ESS 283,912 5.075 2.132 0 10 Table 4 Summary statistics – country measures. Variable name Variable label Source N Mean Standard deviation Minimum value Maximum value Country measures inst_trust(mean) Institutional trust (factor variable) ESS 167 5.015 0.782 3.421 6.634 ppltrst (mean) Interpersonal trust ESS 167 5.254 0.911 3.596 7.063 KOFGI Globalisation index KOF 171 84.715 4.178 69.933 90.984 diversity Foreign-born population OECD 171 8.631 4.848 1.600 24.119 gini Gini index World Bank 171 30.506 3.533 23.700 41.300 ecodecline Economic decline QoG 171 2.951 1.284 −0.800 6.000 total_exp Government expenditure as % of GDP Eurostat 171 45.892 6.745 25.300 64.900 educ_exp Government expenditure in education as % of GDP Eurostat 171 5.413 0.827 3.200 7.100 quality Institutional quality QoG 171 0.822 0.126 0.583 1.000 GDP Natural log of GDP QoG 171 26.723 1.184 23.502 29.142 Table 5 Country averages on institutional and interpersonal trust in a fixed effects regression with standardised variables (panel data). Institutional trust(1) Institutional trust(2) Interpersonal trust(3) Interpersonal trust(4) Globalisation −0.376 −0.311 0.110 0.141 (0.200) (0.210) (0.0754) (0.0867) Inequality −0.0966 −0.105 0.00300 0.00159 (0.0769) (0.0743) (0.0362) (0.0338) Diversity 0.176 0.242 −0.0595 −0.0182 (0.103) (0.118) (0.0680) (0.0685)\ Economic decline −0.166** −0.141* −0.0449 −0.0300 (0.0575) (0.0576) (0.0494) (0.0461) Institutional quality 0.0348 0.0610 −0.219 −0.202 (0.162) (0.175) (0.126) (0.133) GDP 1.267* 1.037* 0.552*** 0.412* (0.446) (0.479) (0.134) (0.154) Government expenditure in education −0.0661 −0.0131 (0.0580) (0.0412) Government expenditure −0.123 −0.0634* (0.0635) (0.0226) Constant −0.0175* −0.0125 −0.0115*** −0.00868* (0.00826) (0.00902) (0.00278) (0.00312) N 167 167 167 167 r2 0.327 0.348 0.323 0.341 Standard errors in parentheses. * p <0.05, ** p <0.01, *** p <0.001. L. Verhoeven and J. Ritzen Research in Globalization 7 (2023) 100142 8 We will take these results into consideration when moving on to our second research strategy. We will use the same macro effects combining them with individual measures to test the individual level of trust. But, first, we will zoom into the individual profile. The profile of individual trust Table 6 presents the profile of individual trust. This profile is easily interpretable. Education and income reinforce trust in each other and in institutions significantly, signalling a potential for social and political polarisation, as suggested by several authors (Foster & Frieden, 2017; Eriksen, 2007; Nowakowski, 2021). Moreover, a rightist political orientation increases institutional trust. This sounds contradictory, but has previously been found e.g., Foster and Frieden (2017). Religion also increases institutional trust (significantly). Age is interesting: older people are more trusting each other but does not have an effect on trusting institutions. Being employed decreases trust in institutions. Again, this seems contradictory. When zooming into this relation, it becomes visible that until 2010, the relation between being employed and institutional trust is neutral. Only after 2010, the relation between being employed and institutional trust becomes negative. This relation remains an enigma, however weak it is. Macro and individual aspects combined: the country profile of trust Now, to gain a full understanding of both institutional and interpersonal trust, we combine the macro and the individual level of measurement of our data. As we have argued that the effect on trust can have differentiated outcomes due to fragmentation in society, we will look further into this. Therefore, we will regress our model combining different levels of measures on the individual level of trust for different scores on socio-economic status (see Appendix 3 for the results without splitting). We do this by creating eight different socio-economic status (SES) categories, along the lines of Berzofsky, Smiley-McDonald, Moore and Krebs (2014) (see Appendix 2 for more details). This allows us to investigate the effect of SES, i.e., income and education, on trust in more detail. Country aspects on individual trust According to the results of combining the two levels of measurement shows us that macro-level components do not affect the individual level of trust as much as the individual contributors, except for institutional quality. It shows a significant effect of institutional quality for all SES categories aside from the SES category zero. The effect of institutional quality increases the level of both institutional trust and interpersonal trust as the SES level gets higher. Increasing GDP decreases interpersonal trust, where the effect starts to get significant around average SES scores and decreases in size when SES gets higher – albeit remaining significant. Globalisation does not have an effect on either form of trust, neither does expenditure, with some exceptions for the 7th SES group with expenditure in education and for the 0th SES group with total expenditure (see Appendix 4). However, the positive and significant effect of the quality of institutions shows possibilities of ways to intervene for governments, and possibly shift focus from expenditure merely to inclusive policies (as argued by Iammarino, Rodriguez-Pose & Storper (2019) and Rodrik & Stantcheva (2021a/b). We have conducted the analysis with the same model for a smaller group of countries. This sample contained only Northern- and Western European countries, whereas the sample now also includes some Southern- and Eastern European countries. In the models with the more homogeneous group of only Northern- and Western European countries (see Appendix 5), we found different results. There, globalisation had a significant negative effect on both forms of trust. Moreover, diversity, inequality and economic decline also showed significant coefficients in Table 6 Individual institutional and interpersonal trust in a pooled fixed effects regression with standardised variables. Institutional trust Interpersonal trust Education (years) 0.106*** 0.145*** (0.0191) (0.0165) Income 0.156*** 0.147*** (0.0130) (0.0151) Employement relation (1) =employed) −0.025* 0.00239 (0.0113) (0.0129) Born in country (1 =yes) −0.0362 0.0142 (0.0175) (0.0138) Age −0.0163 0.0456* (0.0176) (0.0160) Religion (subjective scale 1–10) 0.103** 0.0321 (0.0276) (0.0269) Gender (1 =male) 0.0109 0.0110 (0.00601) (0.00621) Left-right scale (0 =left, 10 =right) 0.0680** 0.00456 (0.0178) (0.0130) Constant 0.119 0.0791 (0.0704) (0.0878) N 199,914 220,477 r2 0.0685 0.0560 Standard errors in parentheses * p <0.05, ** p <0.01, *** p <0.001. L. Verhoeven and J. Ritzen Research in Globalization 7 (2023) 100142 15 Frenk, J., G´ omez-Dant´ es, O., & Moon, S. (2014). From sovereignty to solidarity: A renewedconcept of global health for an era of complex interdependence. The Lancet, 383(9911), 94–97. Glanville, J. L., & Shi, Q. (2020). The Extension of particularised trust to generalised and out-group trust: The constraining role of collectivism. Social Forces, 98(4), 1801–1828. Gray, P., & Bjorklund, D. F. (2014). Psychology ([7th ed.].). New York, United States: Worth Publishers. Grechyna, D. (2016). On the determinants of political polarization. Economics Letters, 144, 10–14. Green, A., Janmaat, G., & Cheng, H. (2011). Social cohesion: Converging and diverging trends. National Institute Economic Review, 215, R6–R22. Gruber, J. (2015). Public Finance and Public Policy [5 th edition]. New York, United States: WorthPublishers Inc. Gygli, S., Haelg, F., Potrafke, N., & Sturm, J. E. (2019). The KOF globalisation index–revisited. The Review of International Organizations, 14(3), 543–574. Habich, J., & Nowotny, V. (2017). A Closer Look at Globalization–Roots and Remedies. A Closer Look at Globalization–The Positive Facets and the Dark Faces of a Complex Notion. Background paper, Trilogue, Salzburg 2017: Arts, Economy, Politics, 3-19. De Haas, H., Castles, S. & Miller, M.J. (2020). The Age of Migration. International Population Movements in the Modern World [6 th edition]. Londen, United Kingdom: Red Globe Press. Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2006). Multivariate data analysis. Hoboken, United States: Prentice Hall Pearson Education. Harms, P., & Schwab, J. (2020). Depression of the deprived or eroding enthusiasm of the elites: What has shifted the support for international trade? European Journal of Political Economy, 64, Article 101901. Heggem, I.Ø., & Jakobsen, T. G. (2016). The challenge of globalisation: A worldwideinvestigation of public attitudes on individual vs. government responsibility. World Review of Entrepreneurship, Management and Sustainable Development, 12(4), 381–398. Heimberger, P. (2020). Does economic globalisation affect income inequality? A metaanalysis. The World Economy, 43(11), 2960–2982. Iammarino, S., Rodriguez-Pose, A., & Storper, M. (2019). Regional inequality in Europe: Evidence, theory and policy implications. Journal of economic geography, 19(2), 273–298. Inglehart, R., & Welzel, C. (2005). Modernization, Cultural Change, and Democracy: The Human Development Sequence. Cambridge: Cambridge University Press. Kim, D. G. (2019). Positional concern and low demand for redistribution of the poor. European Journal of Political Economy, 56, 27–38. Komisarof, A., & Leong, C. H. (2020). Viewing intercultural adaptation and social inclusion through constructs of national identity: An introduction. International Journal of Intercultural Relations, 78(2020), 1–9. Kriesi, H., Grande, E., Lachat, R., Dolezal, M., Bornschier, S., & Frey, T. (2008). West European politics in the age of globalization. Cambridge, United Kingdom: Cambridge University Press. Lewin, K. (1948). Resolving social conflicts; selected papers on group dynamics. Harper. Magalh˜ aes, P. C. (2014). Government effectiveness and support for democracy. European Journal of Political Research, 53(1), 77–97. Manca, A.R. (2014). Social Cohesion. In A.C. Michalos (ed.), Encyclopedia of Quality of Life and Well-Being Research (pp. 6026-6028). Dordrecht, Netherlands: Springer. Marien, S., & Hooghe, M. (2011). Does Political Trust Matter? An Empirical Investigation into the Relation between Political Trust and Support for Law Compliance. European Journal of Political Research, 50(2), 267–291. Newton, K. (1999). Social capital and democracy in modern Europe. Social capital and European democracy, 3, 24. Nowakowski, A. (2021). Do unhappy citisens vote for populism? European Journal of Political Economy, 68, Article 101985. OECD. (2022). Foreign-born population: Total, % of population, 2002 – 2018. In Society - Migration. Ottaviano, G. I., & Peri, G. (2006). The economic value of cultural diversity: Evidence from US cities. Journal of Economic geography, 6(1), 9–44. Ozili, P. K. (2020). Social inclusion and financial inclusion: International evidence. International Journal of Development Issues, 19(2), 169–186. Paskov, M., & Dewilde, C. (2012). Income inequality and solidarity in Europe. Research in Social Stratification and Mobility, 30(4), 415–432. Pervaiz, Z., Chaudhary, A. R., & van Staveren, I. P. (2013). Diversity, inclusiveness and social cohesion. International Institute of Social Studies of Erasmus University Rotterdam (ISS), 1, 1–51. Pettigrew, T. F., & Tropp, L. R. (2008). How does intergroup contact reduce prejudice? Meta-analytic tests of three mediators. European Journal of Social Psychology, 38(6), 922–934. Potrafke, N. (2015). The evidence on globalisation. The World Economy, 38(3), 509–552. Putnam, R. (1993). The prosperous community: Social capital and public life. The American Prospect, 13(4), 1–11. Putnam, R. D. (2000). Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Schuster. Putnam, R. D. (2007). E pluribus unum: Diversity and community in the twenty-first centurythe 2006 Johan Skytte Prise Lecture. Scandinavian Political Studies, 30(2), 137–174. Putnam, R. D. (2020). The Upswing: How America Came Together a Century Ago and how We Can Do it Again. New York, United States: Simon & Schuster. Ritzen, J. (2017). Higher Education and Economic Development. In J. C. Shin & P. Teixeira (Eds.), Encyclopedia of International Higher Education Systems and Institutions (pp. 1–7). Springer Netherlands. 10.1007/978-94-017- 9553-1_125-1. Ritzer, G. & Dean, P. (2015). Globalisation: a Basic Text (2e ed.). New York, United States: Wiley-Blackwell. Rodríguez-Pose, A., & Garcilazo, E. (2015). Quality of government and the returns of investment: Examining the impact of cohesion expenditure in European regions. Regional Studies, 49(8), 1274-1290. Rodrik, D., & Stantcheva, S. (2021a). A policy matrix for inclusive prosperity (No. w28736). National Bureau of Economic Research. Rodrik, D., & Stantcheva, S. (2021b). Economic Inequality and Insecurity: Policies for an Inclusive Economy. In Major Future Economic Challenges (pp. 193–328). International Commission. Saint-Sup´ ery Ceano-Vivas, M., Rivera Lirio, J. M., & Mu˜ noz-Torres, M. (2014). The role of fiscal policy in the achievement of social cohesion. Tax mix and public social spending. Department of Finance and Accounting, University Jaume I. Sangha, H., & Riegler, R. (2020). Can globalisation promote female labour force participation ? International Journal of Development Issues. Sapienza, P., Toldra-Simats, A., & Zingales, L. (2013). Understanding trust. Economic Journal, 123(573), 1313–1332. Seligman, A. B. (1997). The Problem of Trust. Princeton, United States: Princeton University Press. Simmel (1908/1971). The Stranger. In D. Levine (ed.), Georg Simmel (pp.143-149). Chicago, United States: University of Chicago Press. Snower, D. J., & Bosworth, S. J. (2021). Economic, social and political fragmentation: Linking knowledge-biased growth, identity, populism and protectionism. European Journal of Political Economy, 67, Article 101965. Soroka, S., Johnston, R., & Banting, K. (2004). Ethnicity, trust, and the welfare state. Brussels: De Boeck Cultural Diversity versus Economic Solidarity. van Staveren, I., & Pervaiz, Z. (2015). Ethnic Fractionalization and its Effect on Social Cohesion:” It’s the Social Exclusion, Stupid! ECINEQ, University of Luxembourg, 13–15. Stiglitz, J. E. (2017). The Overselling of Globalisation. Roosevelt Institute Working Paper, 1, 1–16. Stock, J. H., & Watson, M. W. (2015). Introduction to Econometrics [third edition]. London, United Kingdom: Pearson Education. Swank, D., & Betz, H. G. (2018). Globalisation, institutions of social solidarity, and radical right-wing populism in Western Europe. In presentation at the 2018 Annual Meetings of the American Political Science Association, August. Tajfel, H. (1970). Experiments in intergroup discrimination. Scientific American, 223(5), 96–103. Taylor, D. M., Moghaddam, F. M., Gamble, I., & Zellerer, E. (1987). Disadvantaged group response to perceived inequality: From passive acceptance to collective action. The Journal of Social Psychology, 127(3), 259–272. Teorell, J., Sundstr¨ om, A., Holmberg, S., Rothstein, B., Alvarado, N. & Cem Mert Dalli, P. (2021). The Quality of Government Standard Dataset [dataset]. University of Gothenburg: The Quality of Government Institute, doi:10.18157/qogstdjan21. Thompson, G., & Hirst, P. (2002). The future of globalisation. Cooperation and Conflict, 37 (3), 247–265. Uslaner, E. M. (2002). The moral foundations of trust. Cambrigde, United Kingdom: Cambridge University Press. Vergolini, L. (2011). Social cohesion in Europe: How do the different dimensions of inequality affect social cohesion? International Journal of Comparative Sociology, 52 (3), 197–214. Villaverde, J., & Maza, A. (2011). Globalisation, growth and convergence. The World Economy, 34(6), 952–971. Wilkinson, R. G. (1997). Comment: Income, inequality, and social cohesion. American Journal of Public Health, 87(9), 1504–1506. Wilkinson, R. G., & Pickett, K. E. (2017). The enemy between us: The psychological and social costs of inequality. European Journal of Social Psychology, 47(1), 11–24. Wishlade, F., Gross, F., Yuill, D., Gorzelak, G., Kozak, M., & Mendez, C. (2010). The objective of economic and social cohesion in the economic policies of the Member States. Glasgow, United Kingdom: European Policies Research Centre. Wooldridge, J. M. (2010). Econometric analysis of cross section and panel data. MIT press. World Bank. (2022). Gini index: SI.POV.GINI. In World Development Indicators. Wright, M. (2011). Diversity and the imagined community: Immigrant diversity and conceptions of national identity. Political Psychology, 32(5), 837–862. Zak, P.J. and Knack, S. (2001) Trust and Growth. The Economic Journal, 111, 295-32. L. Verhoeven and J. Ritzen