SOCIOECONOMIC CONTEXT AND WELFARE NATIONALISM IN THE EURO AREA FOLLOWING THE GREAT RECESSION CONTEXTO SOCIOECONÓMICO Y NACIONALISMO DE BIENESTAR EN LA ZONA DEL EURO TRAS LA GRAN RECESIÓN

This study examined the association between the socioeconomic context of the Eurozone countries in the post-Great Recession era and attitudes of solidarity towards immigrants among the general population. A multilevel analysis was conducted using data from Round 8 (2016) of the European Social Survey as well as Eurostat and OECD statistics. Specifically, we assessed the link between the share of people at risk of poverty/social exclusion in 2016 and citizens ́ support for excluding immigrants from the rights to social services. Our results, in agreement with theories explaining that anxiety over-scarcity compromises out-group solidarity, confirmed the existence of such a relationship. Additional findings, however, suggested that some factors, such as holding individual-level universalism values, or the persistence of hardship during the recession recovery period (2011-2016) at country-level, may have had a protective role against chauvinistic attitudes. Recommendations for practice and research are discussed.


Introduction
The continuity of the European project depends on its ability to maintain a robust welfare protection system and appropriate levels of social cohesion in increasingly diverse and segmented societies (Ferrera & Burelli, 2019). However, the Great Recession and its socioeconomic consequences have challenged the advancement of the European social model, questioning the grounds of European multiculturalism (Ferrera & Burelli, 2019;Taylor-Gooby, Leruth, & Chung, 2017). The current study investigated whether the contextual socioeconomic characteristics in the aftermath of the Great Recessionparticularly the share of people at risk of poverty/social exclusion-eroded public support for granting immigrants the rights to social benefits and services. We focused on the countries of the Euro area, formed by the 19 member states of the European Union that have adopted the euro as their common currency. The rationale for targeting those countries is that economic and social struggles were particularly severe across the Euro area (also known as the Eurozone) during the post-recession period (Ólafsson, Daly, Kangas, & Palme, 2019;Taylor-Gooby et al., 2017). This work seeks to build a better understanding of the factors that shape preferences for redistribution and solidarity in times of economic uncertainty and sociopolitical change.

The European social model in times of uncertainty
European societies are becoming more and more diverse in terms of nationof-origin and ethnic composition as a result of, among other processes, the economic globalization and the rise of transnational migration (Castles, De Hass, & Miller, 2013). In such a context, the welfare state systems, built under predominantly homogeneous societies, have to accommodate an increasing number of newcomers excluded from the basic entitlement of social rights upon citizenship (Schierup, Hansen, & Castles, 2006). There are ongoing debates within the European population concerning the nature and scope of immigrants' access to social benefits (Svallfors et al., 2012). Moreover, these discussions are held in a context of significant transformations in the welfare state architecture. In particular, changes in demographics (e.g., population aging), culture (e.g., the transformation in gender roles), and economic (e.g., labor market segmentation), are calling for a significant reconceptualization of the European model of social protection (Castles & Schierup, 2012). New patterns of socioeconomic inclusion/exclusion are also weakening the cross-class coalition that used to constitute the primary source of support for generous redistributive policies (Esping-Andersen, 2009).
The incorporation of newcomers unveils some of the limitations and contradictions of the European system of social protection. In particular, the processes of racialized exclusion push many migrants towards the lower strata of society, which significantly impacts social cohesion by reinforcing us vs. them type narratives (Castles & Schierup, 2012;Schierup et al., 2006). Similarly, Sainsbury (2012) demonstrated that welfare policies have a lower effect on poverty reduction among immigrants than among the general population, which evidences persistent patterns of segregation and social stratification. Furthermore, the modern manifestations of racism and national-populism have increasingly labeled immigrants as a social and economic burden, which fostered prejudices even in traditional multicultural European societies such as the Netherlands (Schierup et al., 2006).
The Great Recession in the late 2000s, which was considered the greater economic crisis at a global level since the Great Depression that occurred during the 30s of the last century, has been significantly consequential for reshaping the European social model. The starting point of the recession varied among regions through 2007-2008, but evidence of its severity spread across Europe during late 2009 (Taylor-Gooby et al., 2017). The broad consequences of the Great Recession expanded throughout a long aftermath period that lasted until 2016 (Jauer, Liebig, Martin, & Puhani, 2019). As highlighted by Taylor-Gooby et al. (2017), the recession triggered a complex net of crises (social and political on top of financial and economic) that challenged the European project´s solidaristic principles.
The Euro area was particularly impacted by significant increases in poverty and social exclusion (Gómez Serrano, Molero-Simarro, & Buendía, 2016;Ólafsson et al., 2019). By the year 2016, the unemployment rates in several Eurozone countries were still above the pre-crisis level, and the European Union alerted of persistent risks for social cohesion (European Union, 2016). Artiles and Meardi (2014) explained that increasing competition for scarce resources fostered hostility against immigrants among the general population. Welfare exclusionism (also known as welfare chauvinism or welfare nationalism), defined as the belief that granting immigrants access to social benefits should be restricted, has been regarded as a genuine manifestation of those anti-immigrant sentiments (Reeskens and Van Oorschot, 2012).

Shaping solidarity towards immigrants
Public opinion is a critical element for the advancement of the European social policy (Van Oorschot, 2006). Attitudes not only reflect the views of the population on a variety of issues but also shapes the agenda-setting and the policy implementation processes (Sabatier & Weible, 2014 Previous studies assessed the intersection of individual-level factors (e.g., socioeconomic status; education) and contextual conditions (e.g., unemployment rate; share of immigrants) to explain differences on people´s preferences (Artiles & Meardi, 2014;Mewes & Mau, 2012;Reeskens & Van Oorschot, 2012). Scholars on political ethics have also suggested that holding values informed by transcending self-interest and the promotion of others´ enhancement, such as universalism and benevolence, predicted welfare attitudes and solidarity towards immigrants (Davidov & Meuleman, 2012). An economy-rooted theory for explaining preferences for intergroup redistribution is the realistic conflict approach (Böhm, Rusch, & Baron, 2018;Esses, Jackson, & Armstrong, 1998). The theory underlines that competition for resources, whether actual or perceived, predicts the extent of solidaristic attitudes. According to the model, material hardships and unequal distribution of resources develop resources stress and group-competition over scarcity (Böhm et al., 2018;Esses et al., 1998). Fiske (2002) has also underlined the role of ideology, pointing out that intergroup biases rooted in the combined effect of economic competition and perceived conflict concerning groupidentities. Furthermore, Reeskens and Van Oorschot (2012) suggested that the principle of reciprocity (i.e., benefits upon contribution) might mediate people´s support for granting immigrants access to social services.
There is a gap in the literature assessing whether the share of people suffering hardships in the post-recession period related to welfare attitudes and the inclusion of newcomers as recipients of social benefits. The current study aimed to address such a gap by assessing the case of the Euro area. We have formulated the following research question: Is the share of people at risk of poverty/social exclusion in the year 2016 associated with individual support for excluding immigrants from the rights to social services? The conclusions of this Eurozone-based study will contribute to the growing body of comparative literature concerning public attitudes on welfare policies in multicultural settings. This research aligns with the purpose of the social work discipline (specifically its commitment to promoting social cohesion and enhancing people' s wellbeing) as expressed in its global definition: Social work is a practice-based profession and an academic discipline that promotes social change and development, social cohesion, and the empowerment and liberation of people. Principles of social justice, human rights, collective responsibility and

Methodology
The current study aimed to examine the association between poverty/ social exclusion and attitudes towards immigrants´ rights to social services among the Eurozone´s citizenry in the post-Great Recession era. Under the assumption that both individual and contextual characteristics shape welfare attitudes, we tested a series of individual-level and country-level socioeconomic predictors using a multilevel analysis approach. Multilevel analysis -also known as hierarchical linear model-is particularly indicated for analyzing nested structures, such as students in schools, employees in companies, or population within countries like in the case under examination (Snijders & Bosker, 2012). As pointed out by Raudenbush and Bryk (2002), multilevel modeling increases the control over misestimating-precision and the errors of interpretation that may occur when including different units of analysis in the same statistical model.
Following the work of Artiles and Meardi (2014), Mewes and Mau (2012), and Reeskens and Van Oorschot (2012), we combined survey datasets and macro-level statistics towards addressing the following research question: Is the share of people at risk of poverty/social exclusion in the year 2016 associated with individual support for excluding immigrants from the rights to social services? The insights from the literature on attitudes about redistribution in times of scarcity allowed us to hypothesize that larger proportions of people at risk of poverty/social exclusion will relate to higher exclusionist positions. The current study used 2016 as the year of reference, which was consistent with the upper bound of the post-recession period across the Euro area (Jauer et al., 2019).

Sources of data
Our sources of data consisted of individual-level data drawn from the European Social Survey (ESS), and macro-data provided by Eurostat and OECD statistics at the country-level. Specifically, we focused on the Eurozone countries that participated in the Round-8 of the ESS (2016). These countries (N=13) were Austria, Belgium, Estonia, Finland, France, Germany, Ireland, Italy, Lithuania, Netherlands, Portugal, Slovenia, and Spain.

Individual-level data: the European Social Survey
The ESS is a cross-national survey of public attitudes and behaviors promoted by a consortium of research institutions from twenty-one European countries. The survey was conducted every other year since 2002, using cross-sectional probability samples drawn from the population aged 15 and above at each participating country (European Social Survey, 2017). Some features of the ESS enhanced the cross-national reliability and validity of its instruments of data collection. In particular, the use of the TRAPD methodology (i.e., Translation, Review, Adjudication, Pretesting, and Documentation) facilitated the cross-cultural equivalence of the translated versions of the English-language source questionnaire (Jowell, Roberts, Fitzgerald, & Eva, 2007). Furthermore, the European Social Survey subscribed to the Declaration of Professional Ethics of the International Statistical Institute.
Each round of the ESS included a set of items on core topics such as subjective wellbeing and social trust, as well as socio-demographic data and a series of questions based on Schwartz´s (2012) Human Values framework. Of interest for the current investigation is a rotating module on welfare attitudes that the survey included in two of its waves (Round-4 [2008] and Round-8 [2016]). Such a welfare component aimed to generate knowledge concerning perceptions of social services and welfare deservingness among the European population (European Social Survey, 2017).
Specifically, the current study used Round-8 of the ESS, as its fieldwork in 2016 concurred with the upper limit of the post-recession period. Twentyone countries of Europe, as well as Russia, participated in Round-8. In line with our research goal, we focused exclusively on the survey samples from the country-members of the Eurozone (Table 1). In all cases, the survey administration mode was computer-assisted personal interview (CAPI), except for two countries where paper-and-pencil interview (PAPI) applied. Round-8 response rates ranged from 30.6 to 68.4, based on a formula consistent with AAPOR´s (2016) standards: relates with welfare attitudes. Specifically, we used country-based Eurostat data for assessing (a) poverty/social exclusion, (b) income inequality, and (c) foreign-born population. Concerning the OECD, the organization provides standardized macro-data on several domains, including demographic and socioeconomic. The current study used OECD statistics on (a) economy size, (b) unemployment, and (c) social expenditures.

Variables and measures
We have conceptualized a country-level variable named poverty/social exclusion as the main predictor for the outcome welfare exclusionism. We also defined a set of variables concerning values and personal characteristics to control for at the individual-level. Similarly, a series of economic and social indicators at the country-level were specified. Alternativas. Cuadernos de Trabajo Social, (2021) (2014) defined as an unambiguous expression of exclusionist attitudes-to conceptualize welfare exclusionism as a dummy variable. The variable was measured as 1 = respondent meets the condition of expressing that immigrants should never get the same rights to social benefits and services as citizens, and 0 = respondent does not meet the condition, including all other categories. The current study used welfare exclusionism as the outcome variable (Table 2).
(ii) Universalism and benevolence. Universalism expresses the degree of preference for the value universalism. The variable was measured as the individual mean response, from 1 to 6, to three survey items (ESS, 2016; items C, H, and S) drawn from the Human Values framework (Schwartz, 2012). Universalism capsulizes disposition to out-group solidarity, and one can expect that holding the value will shape positive attitudes towards redistribution and immigrants (Davidov & Meuleman, 2012;Schwartz, 2012). Accordingly, we included universalism as a control variable in our statistical models.
Benevolence indicates the degree of preference for the value benevolence. It was built upon two items from the Schwartz´s Human Values framework (2012) included in the survey questionnaire (ESS, 2016; items L and R). The variable measure consists of the individual mean response to those items (values 1 to 6). Since benevolence represents solidarity towards those within the circles of family and friendship, holding this value will likely foster mistrust for out-group redistribution (Davidov & Meuleman, 2012;Schwartz, 2012). Hence, we have included benevolence as a control variable in our statistical models. (iii) Sociodemographic variables. We also accounted for a series of sociodemographic variables, namely age, gender, education, ideology, income, and ethnicity, as individual-level control predictors. Age was a continuous measure of a person´s years of age, from 15 to 97, based on self-reported information. Gender was measured dichotomously upon self-identification as male or female. Education represented the self-reported higher level of education attained, measured from 1 to 7. The variable ideology consisted of values from 0 to 10, based on a scale of self-identification in left-right terms included in the survey questionnaire. Income referred to family income, and it was measured upon self-reported household net income, ordered from 1st to 10th decile. Ethnicity was constructed as a dummy variable as 1 = respondent self-identified as a member of an ethnic minority.  Algan, Guriev, Papaioannou, & Passari, 2017;Kunovich, 2004), the current study has conceptualized a series of economic control-variables, namely inequality, economy size, unemployment, and social spending, as the changesamount from 2011 to 2016. Algan et al. (2017) showed that it was the change in unemployment during the Great Recession period, rather than its level at a reference year, which correlated to sociopolitical attitudes. Moreover, Kunovich (2004) pointed out the convenience of using a five-year window to assess potential country-level effects on anti-immigrant prejudices.
We measured inequality using the Gini index provided by Eurostat statistics. The index (from 0 to 1) quantifies the extent to which the distribution of income within a country deviates from a perfectly equal distribution represented by a Gini value of zero (European Union, 2016). We followed Grabka and Frick (2013) -who assessed the evolution of Gini upon the formula of relative change-to define our variable inequality as Gini´s relative change from 2011 to 2016 in a given country.
The variable economy size was measured upon OECD data on the gross domestic product (GDP). GDP consists of the monetary value of all the finished goods and services produced within a country' s borders in a specific period (Ólafsson et al., 2019). We followed the work of Mewes and Mau (2012) to define our variable economy size as the absolute change of the GDP per capita (as purchasing power parity in US dollars) from 2011 to 2016 in a given country, divided by 1,000. OECD statistics were also used for data on unemployment and social expenditures.
Concerning unemployment, its measure consists of the absolute change from 2011 to 2016 on a given country´s unemployment rate. The rate consists of the number of unemployed people as a percentage of the labor force (Mewes & Mau, 2012 a percent of GDP, in a given country. It included cash benefits, direct in-kind provision of goods and services, and tax breaks for social purposes. (iii) Social variables. Finally, we framed the variables migration flow and AROPE change based on Eurostat datasets. Following previous works, such as Gómez Serrano et al. (2016), and Mewes and Mau (2012), we included both predictors as control variables in our statistical models. Migration flow was measured as the percentage point change of the foreign-born population in a given country, using 2016 as the time-reference. Concerning AROPE change, we measured the variable as the absolute change from 2011 to 2016 on the rate AROPE for each country (non-anchored rate).

Analysis
Two-level multilevel logistic regressions were estimated to assess the association between support for excluding immigrants from the rights to social services and the share of people at risk of poverty/social exclusion in the year 2016. The multilevel logistic analysis is also known as mixed-effect logistic or hierarchical generalized linear model (Snijders & Bosker, 2012). The method allowed us to estimate the odds of the occurrence of exclusionist stands -the binary outcome in the current study-while taking into account that the survey respondents (Level-1 units) were nested within a higher-level structure conformed by countries. Our analyses were conducted using the multilevel modeling software HLM-6 (Raudenbush, 2004) with a maximum likelihood estimation procedure set to 14 interactions.
We started analyzing a null model which consisted of an initial specification with no predictors at either level (model 1; Table 3). The output of such unconditional configuration provided information on the variability between countries regarding the outcome towards confirming the convenience of using the multilevel approach. Subsequently, we examined the response of key individual-level variables by regressing a set of Level-1 predictors on welfare exclusionism (model 2). Finally, we modeled a fully conditional specification (model 3). Model 3 allowed us to estimate the odds of supporting welfare exclusionism as a function of the Level-2 predictor poverty/social exclusion while also controlling for a set of lower-level and higher-level variables, as suggested by the literature. The full-model´s specifications included Bernoulli distribution, and population size and post-stratification survey weights at Level-1. All the predictors (except for gender and ethnicity that were binary measured) were grand-mean centered to facilitate the interpretation of the results. Due to the limited number of units at the upper-level, this study did not consider robust estimation. Following the work of Papazoglou and Galariotis (2020), we applied listwise deletion to handled missing data. As

Results
The estimated results for the unconditional specification (model 1) indicated that, assuming that countries' log-odds of welfare exclusionism were normally distributed with mean -2.79 (SE 0.209) and variance 0.55, we would expect about 95% of the countries to have values between -4.24 and -1.24. If converting to probabilities for a 95% confidence-level, these results indicated that Eurozone countries' scores concerning support for excluding immigrants from the rights to social services laid between 0.014 and 0.207. We observed that the typical probability for a random effect = 0 of support for welfare exclusionism was 0.057. In summary, the unconditional model revealed the existence of considerable variation to be explained and confirmed the convenience of applying the multilevel approach. Concerning model 2, the results demonstrated the significant association (p<.001) between a series of characteristics at the personal level (i.e., embracing the values universalism and benevolence, education, ideology, and household income) and individual views on welfare exclusionism. The observation of such significant relationships supported the inclusion of those variables in our final regression model. The results for the fully conditional specification (model 3) showed that a higher rate of people at risk of poverty/social exclusion in the year 2016, holding constant the other predictors in the model, was significantly associated with higher odds (OR=1.14) of supporting the exclusion of immigrants from the rights to social services (p<.05). Such findings supported the hypothesis that a more substantial proportion of people at risk of poverty/social exclusion at the upper limit of the post-recession period will correspond to chauvinistic positions. However, our results also provided statistically significant evidence that a raise in the AROPE change from 2011 to 2016 appeared to be negatively related (OR=0.57) with the odds of endorsing welfare exclusionism (p =.05).
Regarding the individual-level variables included in the full model, the results pointed out some significant findings. We observed that respondents holding a stronger inclination for the value universalism were less likely to support the exclusion of immigrants from social services (OR=0.54, p<.001) compared with those less universalism-committed. A significant negative association was also shown by education and income. One unit increase on educational attainment (OR=0.76, p<.001) and household income (OR=0.89, p<.001) decreased by 0.274 and 0.122, respectively, the predicted log-odds of welfare exclusionism. Finally, holding an ethnic-minority status also correlated with lower odds of supporting exclusion (OR=0.42, p<.01). Conversely, the results indicated that respondents showing a greater alignment to the value benevolence (OR= 1.27,p<.001), and to the right-wing ideology (OR= 1.17, p<.001), were more likely to express a preference for denying immigrants the rights to social services. For the full specification, Wald t-statistic takes a value of 75.47 (5 degrees of freedom, and p<.001).

Discussion
This study confirmed that a larger share of people at risk of poverty/social exclusion in 2016 was significantly associated with support for excluding immigrants from the rights to social services among the general population of the Euro area. Such a finding aligned with existing literature underlining the linkage between socioeconomic adversity and exclusionist stands (e.g., Artiles & Meardi 2014;Mewes & Mau, 2012;Reeskens & Van Oorschot, 2012). More precisely, our study was consistent with works rooted in the realistic conflict theory (Böhm et al., 2018;Esses et al., 1998), which explained that perception of financial risk and scarcity over resources generate resources-stress that trigger out-group anxieties. However, our findings also indicated a puzzling effect as the AROPE change from 2011 to 2016 related to lower rather than higher chauvinistic attitudes. Examining such apparent contradiction may constitute one of the main contributions of this work.
Although both poverty/social exclusion in the year 2016 and AROPE change from 2011 to 2016 accounted for socioeconomic adversity at the country-level, they actually invoked distinctive concepts. While the former portrays the context at the upper limit of the post-recession period, the later speaks of the challenges of the recovering process. Finding that a povertystricken setting in the year 2016 correlated with a preference for withholding immigrants´ social rights does not necessarily conflict with identifying that persistent hardships over time were associated with a lower inclination for excluding immigrants from social services. In other words, although a higher number of people in a situation of poverty/social exclusion seems to have eroded out-group solidarity, an arduous recovering from the challenges of the Great Recession may have contributed to forging a sort of in/out-group solidarity in the struggle.
Additional research is needed to explore the extent to which contextual factors modulate the effect of resources-stress on solidarity towards immigrants. Perhaps, exploring the evolution of poverty/social exclusion and inequality at a regional level will help us to capture the specific responses of the socioeconomic indicators under examination. Recent studies, such as Czaika and Di Lillo (2018), have used the Nomenclature of Territorial Units for Statistics (NUTS regions) as the upper-level unit of analysis for hierarchical inquiries concerning anti-immigrant attitudes. As shown by Moretti and Whitworth (2020), the use of sub-national statistics is a promising avenue to improve comparative research on welfare attitudes throughout a deeply segmented Europe in the aftermath of the Great Recession. Specifically, since our analysis did not provide significant evidence to confirm that unemployment, inequality, or migration flow related to welfare exclusionism, it will be worthwhile to examine indicators for both predictors in a regional dimension. Assessing the within-country variability concerning the relative changes in net migration from 2011 to 2016 will provide a valuable account of the migration dynamic over the post-recession period.
The current study has also demonstrated that a series of individuallevel factors, namely values, political orientation, education, income, and ethnicity, were significantly related to the welfare exclusionism outcome. Concerning values, we found that holding universalism was associated with a greater inclination for granting social services to immigrants, which is consistent with the out-group solidaristic foundation of a pro-universalism ethic (Davidov & Meuleman, 2012;Schwartz, 2012). Conversely, we showed that an individual´s preference for the value benevolence was associated with supporting immigrants´ exclusion. This finding was consistent with Schwartz´s (2012) thesis that benevolence accounts for conditional solidarity upon family-and-friendship ties, fostering hesitation towards out-group redistribution. Future studies applying the insights from the minority threat theory (e.g., Wang, 2012) would analyze symbolic causes of the perceived threat against immigrants, such as a feeling of socio-cultural competition.
By revealing a series of factors and mechanisms that shaped people´s views regarding immigrants accessing social services, this study contributed to advance the comparative literature on welfare attitudes. Specifically, we have generated some knowledge to better understand the roots of solidarity among the general population of the Eurozone in the aftermath of the Great Recession. Hence, this paper contributed to the ongoing discussion held by policy-makers and social practitioners concerning the solidaristic and multicultural dimensions of the European project. From a social work perspective, confirming an association between poverty/social exclusion and welfare exclusionism can inform policy advocacy towards the promotion of equitable access and distribution of resources and wealth, in line with the ethical principle of promoting social justice (International Federation of Social Workers, 2018). By identifying how different human values may modulate the effect of economics on anti-immigrant attitudes, this study could also support the development of a cultural-sensitive program of education towards building solidarity and promoting inclusive communities.
We expect that the contributions of the current study will counterbalance some of its limitations. First, there was a conceptual limitation in the way the questionnaire of the ESS prompted respondents using the term immigrants without any specification. Such a lack of precision omitted the existence of various categories of immigrants, such as economic-migrants or refugees, that would likely motivate different attitudes among the general population.
Secondly, the current study could not examine all the societies of the Euro area, since six Eurozone-members did not participate in the Round-8 of the ESS. The list of Eurozone countries that were not assessed included Cyprus, Greece, Latvia, Luxemburg, Malta, and Slovakia. Finally, given the limited number of Level-2 units in our multilevel model (N=13), we could not consider robust estimation when conducting the logistic analysis.
The findings of the current study confirmed the applicability of the ESS module on welfare attitudes to analyze people' s stances about out-group redistribution. However, the complexity of the processes of the formation of attitudes suggests the need for further analysis accounting for a more comprehensive set of variables. The use of items from the survey sections on social exclusion and national and ethnic identity will allow us to examine significant dimensions, such as resources stress and cultural expectations, that have not been evaluated in the present work. Furthermore, future research applying a fixed effect approach for analyzing data from different rounds of the European Social Survey will enhance the current analysis´s scope, strengthening its conclusions.

Conclusions
The current study assessed the effect of the socioeconomic context -specifically the levels of poverty/social exclusion-on solidaristic attitudes towards granting immigrants the rights to social services. By using multilevel modeling and focusing on the post-Great Recession period in the Euro area, we have developed an original approach that reinforced some of the conclusions presented in previous investigations. In particular, this study confirmed that a higher share of people at risk of poverty/social exclusion in the year 2016 was significantly associated with exclusionist attitudes among the Eurozone´s citizenry. Such a finding was consistent with previous literature suggesting that anxiety over scarcity compromises out-group solidarity.
An additional contribution of this study was to identify that the persistence of hardships over the recession recovering period, from 2011 to 2016, related to lower rather than higher chauvinistic attitudes. Such a paradoxical phenomenon contradicted the principles of the realistic conflict theory. We have speculated that unremitting poverty/social exclusion in the aftermath of the Great Recession reinforced inter-group solidarity, which pointed out a direction for further research for social work and related fields.
The modeling of a limited set of variables, the use of exclusively one round of survey data, and the insufficient attention paid to the symbolic threats concerning immigrants were some of the limitations of the current study. Assessing additional characteristics of the context, such as the scope and quality of the systems of social services, would advance our knowledge on how welfare attitudes are conformed among the increasingly diverse societies of the Euro area. We also suggested that examining socioeconomic indicators at the regional level will enhance the country-level conclusions drawn by the current study.
Our findings underlined the need for advocacy towards the distribution of resources and wealth, as alleviating poverty and social exclusion will contribute to neutralizing anti-immigrant stands. Since the value universalism appeared associated with solidaristic attitudes, this study can also inform evidence-based educational programs towards the promotion of universalismbased ethics. Ultimately, this work constitutes a significant albeit modest contribution to the advancement of social work practice and research across the multicultural and complex European contexts in times of generalized discontent.