ISSN electrónico: 2448-7147                                           doi: /10.22185/24487147.2025.120.15

Artículos

A socioeconomic perspective on attitudes towards immigrants and political affiliation in Europe

Una perspectiva socioeconómica sobre las actitudes hacia los inmigrantes y la afiliación política en Europa

Víctor Bazán-Monasterio*

Ana Isabel Gil-Lacruz*

Marta Gil-Lacruz*

*Universidad de Zaragoza, Zaragoza, Spain

Abstract

Political parties on the far-right commonly hold very strong nationalist, anti-immigration views and this seems to be one of the main reasons why people vote for them. However, there is very little empirical research on the role that attitudes towards immigrants (ATIs) play in elections. This paper analyses how negative ATIs are associated with political affiliations using data from the European Social Survey (ESS) and the OECD. The determinants of ATIs and their association with political ideologies is analysed from a European perspective. Employment rates among the native populations have a significant association with the perception that immigration has a negative impact on the economy. The employment rates of immigrant have a negative association with the perception that immigrants take jobs away from local people. The perception that immigration is negative for the economies of the host countries is linked to right-wing political beliefs although this varies among countries with differing state welfare systems.

Keywords: Attitudes Towards Immigrants (ATIs), political affiliation, public budgets, employment rates.

Resumen

Los partidos políticos de extrema derecha suelen tener fuertes posturas nacionalistas y antiinmigratorias, y esta parece ser una de las principales razones por las que la gente vota por ellos. Sin embargo, hay muy poca investigación empírica sobre el papel que las actitudes hacia los inmigrantes (ATI) desempeñan en las elecciones. Este artículo analiza cómo las ATI negativas se asocian con las afiliaciones políticas utilizando datos de la Encuesta Social Europea (ESS) y la OCDE. Los determinantes de las ATI y su asociación con las ideologías políticas se analizan desde una perspectiva europea. Las tasas de empleo entre las poblaciones nativas tienen una asociación significativa con la percepción de que la inmigración tiene un impacto negativo en la economía. Las tasas de empleo de los inmigrantes tienen una asociación negativa con la percepción de que los inmigrantes quitan los trabajos a la población local. La percepción de que la inmigración es negativa para las economías de los países de acogida está vinculada a las creencias políticas de derecha, aunque esto varía entre países con diferentes sistemas de bienestar estatal.

Palabras clave: Actitudes hacia los inmigrantes (ATI), afiliación política, presupuestos públicos, tasas de empleo.

Received: 25/01/2023
Accepted: 19/12/2024

Introduction

This paper contemplates attitudes towards immigrants (ATIs); it is based on two frameworks that deal with the phenomena: Group Conflict Theory and Political Affiliation Theory. Group Conflict Theory suggests that people living in a given region can perceive intergroup competition for scarce goods (for example, employment, affordable housing, wages, state welfare benefits) and this can induce negative attitudes towards immigrants (Meuleman, Davidov and Billiet, 2009). Political Affiliation Theory links a political position on the left or right with negative attitudes towards immigrants (Rustenbach, 2010). Cutts, Ford and Goodwin (2011) list motivations for voting for an extreme right- wing party as racism, hostility towards the political establishment and anti-immigrant attitudes. While Group Competition Theory refers to a collective viewpoint, Political Affiliation Theory encompasses both individual and collective perspectives (Meuleman, Davidov and Billiet, 2009; Rustenbach, 2010).

This paper analyses how Europeans’ attitudes towards immigrants are associated with political opinions, with a not-used before technic to the best of our knowledge. Binary variables are utilized in a three-level mixed effects logistic regression (with STATA XTMELOGIT). Results are controlled by individual characteristics and contextual factors. The analysis comprises two parts: i) The consideration of how two different negative attitudes towards immigration are shaped by socioeconomic and contextual factors; ii) The association of these factors and negative attitudes towards immigrants with left-right political positioning (self-assessed on a scale of ‘left’, ‘centre’ or ‘right’).

Both economic and ideological factors are implicated in the research; the analysis could therefore provide useful information for other studies, lawmakers, business people or human resources management. Bauer, Lofstrom and Zimmermann (2001) emphasised the fact that immigration policy can have both an impact on the employment opportunities of the immigrant and the local population’s attitudes towards immigration. Social programmes which work with immigrants have become political priorities in host countries, irrespective of their popularity among the general population (Ceobanu and Escandell, 2010).

Literature review

It is often to find out an interaction between negative ATIs, unemployment rates and the existence of far-right political parties (Cochrane and Nevitte, 2014). However, other variables also explain the negative attitudes and it is needed to understand the principal theories that explain the interaction ATIs, economic rates and ideology.

Demographic and socioeconomic factors on attitudes towards immigrants

Generally speaking, attitudes towards immigrants depend on factors such as socioeconomic characteristics, economic stability and place of residence (Card, Dustmann and Preston, 2005, Meuleman, Davidov and Billiet, 2009; Butkus, Maciulyte-Sniukiene, Davidaviciene and Matuzeviciute, 2016). For example, negative ATIs tend to increase with age (Kunovich, 2004; Becchetti, Rossetti and Castriota, 2010; Gorodzeisky and Semyonov, 2018). Education is also a significant factor (Scheepers, Gijsberts and Coenders, 2002; Alba, Rumbaut and Marotz, 2005; Fertig and Schmidt, 2011; Easterbrook, Kuppens and Manstead, 2016): people with higher levels of education have more positive attitudes towards immigration (Haubert and Fussel, 2006; Dandy and Pe-Pua, 2010; Pichler, 2010). However, there is some evidence that when entering the labour market, higher educated individuals can become more opposed to immigration (Lancee and Sarrasin, 2015).

Ideology and attitudes towards immigrants

The most conservative sector of a society is associated with negative ATIs (Bierbrauer and Klinger, 2002; Semyonov, Raijman and Gorodzeisky, 2006; Skenderovic, 2007; Wilkes, Guppy and Farris, 2007; Andreescu, 2011; Van Prooijen, Krouwel, Boiten and Eendebak, 2015). Competition theory argues that the link between ideology and immigration is rooted in the economic threat (Cohrs and Stelzl, 2010). Sides and Citrin (2007) have shown that opposition to immigration and support for anti-immigrant political parties increase with increased unemployment. Broad ideological structures in terms of left-right self-positioning are important determinants of attitudes towards immigrants when the socioeconomic vulnerability of citizens is low (Pardos-Prados, 2011).

Higher unemployment rates appear to lead to more negative ATIs and demands for limits on immigration (Wilkes, Guppy and Farris, 2008; Markaki and Longui, 2013). Unemployment is associated with negative attitudes towards immigrants with similar skills and training as local workers (Kemnitz, 2003; Hainmueller and Hiscox, 2010). On the other hand, Rustenbach (2010) states that immigrant workers often accept jobs that the native population do not want and they can therefore improve the economy. Some studies conclude that immigration does not have a detrimental effect on unemployment rates (Pischke and Velling, 1997) and can even create employment (Peri, 2012): immigrants generate higher demand for goods and services and this leads to economic growth (Bauer, Lofstrom and Zimmermann, 2001; Betz and Simpson, 2013; Alesina, Harnoss and Rapoport, 2016). Despite institutional differences, migration flows lead to higher levels of employment, and this can generate more positive attitudes to immigrants (Fromentin, 2013). Among the better educated, the question of employment is a significant factor in tolerant attitudes towards immigration (Dustmann and Preston, 2007), but there is no solid evidence that concerns about the labour market (believed to be more prevalent among unskilled workers) are reflected in opposition to increased immigration. Although there are many authors who have found that changes in unemployment rates lead to changes in ATIs (Lancee and Sarrasin, 2015), given the diversity of results, high levels of unemployment should not necessarily be expected to lead to more negative attitudes; it is possible that the perception of the increase in unemployment is more important than the real increase in unemployment (Kehrberg, 2007). Opposition to immigration is slightly less significant in countries with higher levels of unemployment (Sides and Citrin, 2007).

Public budget, attitudes towards immigrants and political affiliation

ATIs are influenced by perceptions of how it affects public expenditure (Dustmann and Preston, 2007). European residents believe that impact of immigration on welfare state benefits and financing is bad for their countries (Bridges and Mateut, 2014). Higher social expenditure reduces anti-immigrant sentiment in the long term, but has the opposite effect in the short term (Jaime-Castillo, Marqués-Perales and Álvarez-Gálvez, 2016).

Right-wing governments are more prone to reduce expenditures and deficits after the elections than left-wing ones (Castro and Martins, 2019). Indeed, social expenditure (active labor market programs, unemployment) increased under left-wing governments when de facto trade globalization was pronounced (Florian, Sturm and Potrafke, 2020). However, governments with a deeper leftist orientation do not enlarge active labor market policies spending and reduce job creation programs (Tepe and Vanhuysse, 2013).

Materials and Methods

Sample and analysis

The goal of this research is to analyse how two negative ATIs, linked to socioeconomic issues, are related to Self-Positioning Political Affiliation (SPPA or political affiliation), due to economic and contextual factors. Data was taken from the 2002 and 2014 European Social Survey (ESS) and the 2001 and 2013 reports from the Organisation for Economic Co-operation and Development (OECD). The surveys involved 44,721 individuals. Dummy variables for dependent and explanatory variables were created in order to generate dichotomous variables (1 ‘Yes’; 0 otherwise). This means that negative ATIs could be obtained as outcome variables and predictor variables when compared to their neutral references.

The dependent variables were the ATIs and SPPA. The ATIs were related to the perception of labour market competition and the perception of an economic threat (LMPerception and ETPerception). The ESS scores ATIs on an eleven item scale (from 0=immigrants take job away/immigrants are bad for economy to 10=immigrants create new jobs/immigrants are good for economy). As our purposes of this study were analyse how negative ATIs are associated with placement on left right scale, ESS values from 0 to 3 were taken as 1 (‘Yes’), and 0 otherwise; the variables reflect the opinion that immigrants take jobs away or are bad for economy. The ESS reported SPPA on an eleven item scale. Because this variable is analysed by a multinomial logit model, SPPA results were given three values: 1 = politically left-wing (recoded SPPA values from 0 to 3); 2 = the political centre (recoded SPPA values from 4 to 6); 3 = right-wing (recoded SPPA values from 7 to 10). This means that Note that SPPA values followed a normal distribution. Further information was explained in Annex 1.

Explanatory variables concerned sociodemographic and contextual factors. Sociodemographic factors were: age (the only variable non-dummy, it is categorical); gender (female and male), marital status (married, divorced, single and widowed) education level (primary or less, secondary and tertiary), occupation (employed, student, unemployed, retired and home-maker); and, self-perceived quality of life (low quality, low to middle, middle to high and high quality). Contextual factors were: public expenditure (employment incentives, unemployment protection, and ‘others’ - health, disability, retirement and home-maker support); employment and unemployment rates for immigrants and natives; welfare systems (Mediterranean countries vs. Nordic countries; Anglo-Saxon vs. Nordic; Continental vs. Nordic; Eastern vs. Nordic), and years (2002 and 2014). ‘Welfare systems’ included the following countries:

· Nordic countries: Denmark, Norway, Finland and Sweden.

· Continental countries: Austria, Belgium, Switzerland, Germany, the Netherlands and France.

· Anglo-Saxon countries: United Kingdom and Ireland.

· East European countries: Czech Republic, Slovenia, Poland and Hungary.

· Mediterranean countries: Portugal and Spain.

Econometric specifications

A multilevel mixed-effects model[1] (STATA XTMELOGIT) is used for modelling categorical outcome variables where the categories have no natural ordering: ATIs are the dependent variables. This technique was selected to show how socioeconomic and contextual factors are associated with ATIs in a hierarchical random-effects model with binary data; it contemplates three analysis levels of negative perceptions: socioeconomic characteristics, macroeconomic factors, and European welfare systems. The XTMELOGIT formula is a three-level mixed effects logistic regression, with the predict option for the i^th observation within the j^th level-two cluster within the k^th level-three cluster, and where, z(^p) refers to the design variables of level p, and û(^p) refers to the random effects of level p (Stata Corp, 2023).

  ,

A multinomial logistic regression (mlogit STATA command) fits maximum-likelihood multinomial logistic models in which there is one dependent variable, and there is an equation corresponding to each of the outcomes (values taken on) recorded in that variable, except for the one that is taken to be the base outcome (Stata Corp, 2023). In this study, it was used to analyse the three SPPA categories: left-wing, centre (base outcome), and right-wing. MLOGIT estimates by maximum likelihood of the multinomial logit model will reveal any differences with respect to the political centre (reference). In the multinomial logit model, it estimates a set of coefficients, β (¹), β (²) and β (³), corresponding to each result. Setting β (¹) = 0, the equation would be (Stata Corp, 2023):

Where, γ is the possible outcomes (left, centre, right ideology in this study), X is the explanatory variables;  is the coefficient, and e is the exponent.

The syntaxes to develop the models were:

· For the analyses of attitudes towards immigrants (depentend variables LMPerception and ETPerception) with xtmelogit:

xtmelogit depvar [indepvars] [if] [in] [,fe_options] || re_equation [|| re_equation ...] [,options]

· For the analyses of the ideology self-positioning (dependent Political Affiliation):

mlogit depvar [indepvars] [if] [in] [weight] [options]

Statistical description

Table 1 presents the means of the dependent variables related to the ATIs, in accordance with welfare systems and years of survey. People from Nordic countries are the least likely to think that immigrants take jobs away from natives (five per cent in 2014) and that immigration is bad for the economy (19 per cent in 2014). In contrast, People from Eastern countries are the most likely to have negative ATIs (27 per cent for LMPerception and 41 per cent in ETPerception in 2014). With the exception of Anglo-Saxon countries, there was a generalised increase in the trends between 2002 and 2014.

Table 1: Attitudes Towards Immigrants

Welfare system	Years	Labour market competition perception			Economic threat perception
		Mean (%)	Standard deviation (%)	Mean (%)		Standard deviation (%)
Mediterranean	2002	19.51	39.64	21.33		40.98
	2014	19.56	39.68	25.01		43.31
Nordic	2002	5.39	22.59	18.84		39.11
	2014	5.61	22.97	19.01		39.24
Continental	2002	12.14	32.66	18.32		38.68
	2014	12.87	33.48	23.17		42.19
Eastern Europe	2002	27.46	44.64	30.26		45.94
	2014	27.52	44.67	40.68		49.13
Anglo-Saxon	2002	20.24	40.19	27.96		44.89
	2014	19.19	39.39	27.27		44.54

Source: Authors’ estimations based on the ESS 1^st and 7^th round data

Table 2 displays the means of the dependent variables related to political affiliation. People from Nordic countries are more prone to right-wing beliefs than those from Mediterranean countries.

Table 2: Political affiliation

Welfare System	Years	Left		Centre		Right
		Mean (%)	Standard deviation (%)	Mean (%)	Standard deviation (%)	Mean (%)	Standard deviation (%)
Mediterranean	2002	31.41	46.43	53.02	49.92	15.57	36.27
	2014	33.64	47.26	50.83	50.00	15.53	36.22
Nordic	2002	19.92	39.95	48.58	49.98	31.49	46.45
	2014	21.47	41.06	44.86	49.74	33.67	47.26
Continental	2002	25.70	43.70	57.96	49.36	16.33	36.97
	2014	25.10	43.36	56.31	49.60	18.59	38.90
Eastern Europe	2002	20.93	40.69	54.44	49.81	24.63	43.01
	2014	19.15	39.36	53.91	49.85	26.93	44.37
Anglo-Saxon	2002	15.79	36.47	66.44	47.23	17.77	38.23
	2014	19.81	39.86	62.19	48.50	18.00	38.43

Source: Authors’ estimations based on the ESS 1^st and 7^th round data.

Table 3 includes the statistical description of the explanatory variables. The mean age of the respondents is about 39 years (inclusion criterion was to be between 18 and 65). 48 per cent of the respondents were men, 52 per cent were married and 36 per cent were single. ten per cent were divorced and two per cent were widowed. 30 per cent of the respondents had tertiary education, 65 per cent secondary and five per cent had primary education or less. 70 per cent were employed, nine per cent were students, seven per cent unemployed, four per cent retired and ten per cent were home-makers. The most common standard of living was ‘medium-high’ (46 per cent); only four per cent of respondents declared a low standard of living.

Table 3: Independent variables

	TOTAL (100%)	N (18%)	M (10%)	A (14%)	C (37%)	E (21%)
Age	39.79	39.01	39.63	39.99	40.25	39.59
Male^α	48%	52%	48%	45%	48%	47%
Female	52%	48%	52%	55%	52%	53%
Married^α	52%	45%	55%	51%	53%	56%
Divorced	10%	10%	8%	9%	11%	9%
Single	36%	44%	35%	38%	35%	32%
Widowed	2%	1%	2%	2%	1%	3%
Education (Primary or less)	5%	2%	23%	10%	3%	1%
Education (Secondary)	65%	61%	54%	50%	66%	81%
Education (Tertiary)^α	30%	37%	23%	40%	31%	19%
Occupation (Employed)^α	70%	76%	66%	67%	72%	66%
Occupation (Student)	9%	12%	8%	6%	8%	9%
Occupation (Unemployed)	7%	5%	12%	8%	6%	8%
Occupation (Retired)	4%	3%	3%	4%	4%	8%
Occupation (Home-maker)	10%	4%	11%	15%	10%	9%
High Quality of Life^α	34%	44%	24%	35%	41%	15%
Middle to high Quality of Life	46%	45%	49%	46%	45%	54%
Low to middle Quality on Life	15%	8%	20%	15%	11%	25%
Low Quality of Life	4%	2%	7%	5%	3%	6%
Employment public expenditure	262.4	474.94	181.08	236.38	332.55	79.76
Unemployment protection public expenditure	434.8	327.92	589.35	483.26	560.95	127.71
Other expenditure	7,590.6	1,0101.85	5,843.62	6,698.56	9,001.95	4,382.48
Immigrant unemployment rate	12%	13%	22%	10%	10%	11%
Native unemployment rate	6%	6%	15%	8%	5%	10%
Immigrant employment rate	63%	64%	60%	64%	63%	62%
Native employment rate	68%	74%	60%	67%	71%	60%
Year (2002)	44%	49%	44%	43%	44%	43%
Year (2014)	56%	51%	56%	57%	56%	57%

N - Nordic countries; M - Mediterranean countries; A -Anglo-Saxon countries; C - Continental countries; E - Eastern European countries.

Source: Authors’ estimations based on the ESS 1^st and 7^th round data and OECD, 2001 and 2013, minus Poland (2012 budget variables and 2014 immigrant unemployment rates).

The data on public expenditure is given per capita, in constant 2001 value and adjusted to purchase parity. On average, budgets for active employment policies decreased from 2001 to 2013 (from 289.48€/person[2] to 281.88€/person, with a standard deviation from 188.70€/person to 163.11€/person) and unemployment protection budgets increased (from 347.05€/person to 476.68€/person, with a standard deviation of 233.31€/person and 374.30€/person, respectively). The item associated with other budgets also increased, from 6,794.47€/person to 8,488.17€/person (standard deviations of 2,279.47€/person and 2,360.06€/person, respectively).

The average budgeted for active employment policies for all the countries in this study was 285.33€/person, with a standard deviation of 175.23€/person. Among the regions that spend the most per person on active employment policies are the Nordic countries (average expenditure of 474.94€/person and a standard deviation of 171.47€/person) and the continental countries (average of 332.55€/person, standard deviation of 94.91€/person). The regions that spend less per person on active policies are the Eastern countries (average 79.76 €/person, deviation of 48.12€/person) and Mediterranean countries (average 181.08€/person, deviation of 32.81€/person).

The average expenditure on unemployment protection is 417.86 €/person. The largest budgets were in Mediterranean countries (589.00€/person, standard deviation of 302.40€/person) and Continental countries (560.95€/person, standard deviation 273.04€/person). The lowest budgets were in Eastern countries (127.71€/person, standard deviation of 44.54€/person) and Nordic countries (327.92€/person, standard deviation 261.88€/person).

The average expenditure for ‘other’ public finance categories was 7,590.60€/person. The highest was in the Nordic countries (10,101.85€/person, standard deviation of 1,640.89€/person), the lowest was in Eastern countries (4,382.48€/person, standard deviation 965.42€/person).

The unemployment rate for immigrants was higher than the native populations in all countries (12 per cent compared to six per cent). Conversely, and as might be expected, in most countries, the employment rate was lower for immigrants (63 per cent) than for the native population (68 per cent). However, this is not the case for Eastern European countries, where employment rates of immigrants (63 per cent) are lower than those of natives (68 per cent), and, in the Mediterranean countries rates of employment are 60 per cent for both groups. As employment rates decreased between 2001 and 2013, the unemployment rates increased for both natives and immigrants: the unemployment rate of immigrants increased from ten per cent to 14 per cent, while the employment rate decreased from seven per cent to six per cent; among the native populations, unemployment rose from six per cent to nine per cent and employment rates fell from eight per cent to seven per cent (OECD, 2018a; 2018b).

Analyses of variance

The results for the variable LMPerception were statistically significant. The analysis of variance provided two main results: i) it corroborated the finding that the weight of the random effects in the LMPerception disappears when explanatory variables are incorporated into the models; and, ii) it showed that differences in the unobserved characteristics are greater between countries with different state welfare systems than between countries with similar systems. The state welfare system is therefore an important reference when modelling LMPerception.

In the case of ETPerception we confirm the results obtained for ETPerception. The main difference is that the introduction of macro variables does not have the desired impact on reducing the variance of the random effects. Nevertheless, model 3 again presents the best analysis of variance, and it is confirmed that the welfare systems provide a correct criterion for classifying countries.

Table 4: Analysis of variance

			SS	P > F
Labour Market CompetitionPerception	Empty model	Between groups	18,819.41	0.00
		Within groups	3,692.27
		Total	22,511.69
	Model 1	Between groups	13,991.77	0.00
		Within groups	2,945.87
		Total	16,937.65
	Model 2	Between groups	9,145.90	0.00
		Within groups	4,197.61
		Total	13,343.52
Economic Threat perception	Empty model	Between groups	4,759.25	0.00
		Within groups	3,254.80
		Total	8,014.06
	Model 1	Between groups	2,873.44	0.00
		Within groups	3,345.62
		Total	6,219.06
	Model 2	Between groups	4,120.50	0.00
		Within groups	5,374.99
		Total	9,495.49

Prob > chi² = 0.00

Source: Authors’ estimations based on the ESS 1^st and 7^th round data.

Results

Attitudes towards immigrants

Table 5 shows the estimates of the four multilevel logistic regression models for LMPerception.

Table 5: Labour market competition perception (xtmelogit).

Explanatory variables	ATI: Labour market competition perception
	Empty model	Model 1	Model 2	Model 3
Fixed Effects
Age	--	-0.0027	-0.0013	-0.0005
Male^α	--	--	--	--
Female	--	-0.0382	-0.0307	-0.0322
Married^α	--	---	---	---
Divorced	--	0.1336**	0.1302**	0.1309**
Single	--	0.0690*	0.0718*	0.0707*
Widowed	--	0.0591	0.0419	0.0403
Education (Primary or less)	--	1.1808***	1.1932***	1.1919***
Education (Secondary)	--	0.8651***	0.8741***	0.8715***
Education (Tertiary)^α	--	---	---	---
Self-interest theory
Occupation (Employed)^α	--	---	---	---
Occupation (Student)	--	-0.5877***	-0.5942***	-0.5927***
Occupation (Unemployed)	--	0.2917***	0.2876***	0.2862***
Occupation (Retired)	--	-0.1189*	-0.1370*	-0.1391*
Occupation (home-maker)	--	0.0549	0.0459	0.0446
High Quality of Life^α	--	---	---	---
Middle to high Quality of Life	--	0.2909***	0.2908***	0.2919***
Low to middle Quality of Life	--	0.5503***	0.5540***	0.5551***
Low Quality of Life	--	1.0621***	1.0567***	1.0590***
Employment public expenditure	--	--	-0.0052	0.0122
Unemployment protection public expenditure	--	--	-0.1878*	-0.0559
Other public expenditure	--	--	-0.2010	0.2404
Immigrant unemployment rate	--	--	-0.0112	0.0203
Native unemployment rate	--	--	0.0255	-0.0262
Immigrant employment rate	--	--	-0.0247**	-0.0160*
Native employment rate	--	--	0.0156	0.0034
Year (2002)^α	--	---	---	---
Year (2014)	--	-0.0005	0.0791	-0.0808
Countries (Nordic)^α	--	--	--	---
Countries (Mediterranean)	--	--	--	1.4776***
Countries (Anglo-Saxon)	--	--	--	1.6874***
Countries (Continental)	--	--	--	0.9868***
Countries (Eastern Europe)	--	--	--	2.0800***
Intercept	-1.8385***	-2.8210***	0.4910	5.1004
Random effects
σ^2(var_cons)	0.5416	0.4036	0.3299	0.0606
LR Test (Pro>chi^2)	0.00	0.00	0.00	0.00

^aReference variable

*** . ** and * explanatory variables are statistically significant to 99.99 per cent, 99.95 per cent and 90 per cent.

Source: Authors’ estimations based on the ESS 1^st and 7^th round data and OECD, 2001 and 2013, minus Poland (2012 budget variables and 2014 immigrant unemployment rates).

According to the empty model, the random effect associated with the country of residence (σ²) is relevant, representing 54 per cent of the explanation of LMPerception. In the following three models, the categories ‘Education’ and ‘Quality of Life’ are statistically significant, as are some categories related to civil status and employment. The coefficients of the individual variables are constant, so the estimated results are robust. The belief that immigrants take away jobs is positively correlated with a low level of education, low quality of life, being divorced or single (versus married) and being a student, unemployed or retired (versus employed).

Model 2 adds a set of macro explanatory variables as fixed effects: public policy budgets (active employment, unemployment protection and other concepts), and employment and unemployment rates for both immigrants and natives. Immigrant employment rates and unemployment protection policies are positively correlated with the employment perception that natives have about immigrants: higher unemployment protection (public expenditure) and immigrant employment rates are linked to less fear of employment competition.

Random effects decreased as the variables were incorporated; this means that, in comparison with random effects, the fixed effects gain weight in the explanation of the dependent variable. This can be seen as positive, since the part of the explanation associated with the random effects decreased, so the incorporated determinants are relevant.

Model 3 incorporates the variables related to the countries of residence. The model confirms the results obtained for the employment rate. All the coefficients associated with the geographical areas were significant to 99 per cent. The Nordic and Continental countries reported the least negative employment attitudes towards immigrants; residents of Eastern European countries have the most negative attitudes.

Table 6 shows the estimates of the four multilevel logistic regression models for ETPerception. The dependent variable is dichotomous (1 if the respondent believes that the immigrant is bad for the economy, 0 otherwise). The answers range from zero to ten (from good to bad for the economy); if the respondent gave a score of 8 or higher, the value 1 was assigned, 0 otherwise.

Table 6: Economic threat perception (xtmelogit)

Explanatory variables	ATI: Economic threat perception
	Empty model	Model 1	Model 2	Model 3
Fixed Effects
Age	--	-0.0306	-0.0390	-0.0392
Male^α	--	---	---	---
Female	--	0.1093***	0.1100***	0.1093***
Married^α	--	---	---	---
Divorced	--	0.1300**	0.1498***	0.1498***
Single	--	0.0592*	0.0567*	0.0565*
Widowed	--	0.1435	0.1712*	0.1679*
Education (Primary or less)	--	1.0851***	1.0785***	1.0825***
Education (Secondary)	--	0.9110***	0.9026***	0.9004***
Education (Tertiary)^α	--	---	---	---
Occupation (Employed)^α	--	---	---	---
Occupation (Student)	--	-0.6313***	-0.6322***	-0.6325***
Occupation (Unemployed)	--	0.0741	0.0835*	0.0827*
Occupation (Retired)	--	0.0321	-0.01144	-0.0131
Occupation (home-maker)	--	0.0520	0.0532	0.0528
High Quality of Life^α	--	--	--	--
Middle to high Quality of Life	--	0.1857***	0.1923***	0.1896***
Low to middle Quality of Life	--	0.4239***	0.4275***	0.4250***
Low Quality of Life	--	0.7578***	0.7738***	0.7706***
Employment public expenditure	--	--	0.0610***	0.0670***
Unemployment protection public expenditure	--	--	0.1188	0.1745**
Other expenditure	--	--	-0.6269**	0.0018
Immigrant unemployment rate	--	--	-0.0074	0.0152
Native unemployment rate	--	--	0.0327	-0.0063
Immigrant employment rate	--	--	0.0045	0.0062
Native employment rate	--	--	0.0354**	0.0277*
Year (2002)^α	--	--	--	--
Year (2014)	--	0.2246***	0.2684**	0.0930
Countries (Nordic)^α	--	--	--	---
Countries (Mediterranean)	--	--	--	0.5131
Countries (Anglo-Saxon)	--	--	--	0.9043**
Countries (Continental)	--	--	--	0.2671
Countries (Eastern Europe)	--	--	--	1.5985***
Intercept	-1.1378	-2.0940***	-0.3991	-6.4110*
Random effects
σ^2 (var_cons)	0.1922	0.1510	0.2457	0.0880
LR Test (Pro>chi^2)	0.00	0.00	0.00	0.00

^aReference variable

*** . ** and * explanatory variables are statistically significant to 99.99 per cent, 99.95 per cent and 90 per cent.

Source: Authors’ estimations based on the ESS 1^st and 7^th round data and OECD, 2001 and 2013, minus Poland (2012 budget variables and 2014 immigrant unemployment rates)

According to the empty model, the random effect associated with the country of residence (σ²) was less relevant than in the previous case, but it still represents 19 per cent of the explanation for Perception-Economy. In model 1, Gender, Marital status, Educational level, being a Student, Quality of Life and the Year of the survey were statistically significant. Women showed a positive association with perceiving immigrants are bad for economy taking male as base-outcome. The belief that immigrants are bad for the economy is positively correlated with a low educational level and low quality of life. Opinions on the contribution of immigrants to the economy were more negative in 2014 than 2002.

Model 2 includes the macro variables: public policy budgets (active employment, unemployment protection and other concepts), and employment and unemployment rates for both immigrants and natives. Results show that active employment policies and the employment rate reinforce negative perceptions of immigrants and the economy, but the reverse is true for the importance of public expenditure.

Model 3 includes country of residence as the explanatory variable. The results obtained for the native employment rate were confirmed. Residents in Eastern and Anglo-Saxon countries showed higher coefficients of this attitude towards immigrants than residents of Nordic countries. Also, this model had the lowest estimated coefficient for random effects, so it can be concluded that models that include micro, macro and reference country variables have fixed effects with greater explanatory power.

Attitudes towards immigrants and political affiliation

Table 7 shows the estimates for the determinants of SPPA. Political affiliation is categorised by ‘left’, ‘centre’ (reference) and ‘right’. People who perceive immigrants as an economic threat are less likely to associate themselves with left-wing politics and more likely to support right-wing ideology. Among those who believe that immigrants take jobs away from the native populations, only those who consider themselves as right-wing gave statistically significant results. Mediterranean, Anglo-Saxon and Continental countries were more right-wing than Nordic countries. 2014 is associated with higher levels of auto positioning in the right-wing views than in 2002.

Table 7: Political affiliation (mlogit)

Explanatory variables	Political affiliation
	Left	Centre	Right
Fixed Effects
Labour market competition perception	0.0690	---	0.1395***
Economic threat perception	-0.1865***	---	0.3094***
Age	0.5225***	---	0.0850
Male^α	---	---	---
Female	0.0373	---	-0.3262***
Married^α	---	---	---
Divorced	0.1974***	---	-0.0514
Single	0.3536***	---	-0.0689*
Widow	0.1332	---	-0.0117
Education (Primary or less)	-0.2665***	---	-0.0716
Education (Secondary)	-0.3408***	---	-0.2147***
Education (Tertiary)^α	---	---	---
Occupation (Employed)^α	---	---	---
Occupation (Student)	0.3220***	---	-0.0183
Occupation (Unemployed)	0.1587***	---	-0.1361*
Occupation (Retired)	0.1305*	---	0.0095
Occupation (Home-maker)	-0.1033**	---	-0.0124
High Quality of Life^α	---	---	---
Middle to high Quality of Life	-0.0175	---	-0.2828***
Low to middle Quality of Life	0.1179**	---	-0.3751***
Low Quality of Life	0.3190***	---	-0.3530***
Employment public expenditure	0.0145	---	0.0641**
Unemployment protection public expenditure	-0.0751*	---	0.0500
Other expenditure	1.1727***	---	-1.2624***
Immigrant unemployment rate	-0.0314**	---	-0.0192*
Native unemployment rate	0.0403***	---	0.0001
Immigrant employment rate	-0.0277***	---	-0.0139***
Native employment rate	0.0275***	---	0.0133**
Countries (Nordic)^α	---	---	---
Countries (Mediterranean)	1.264***	---	-1.1829***
Countries (Anglo-Saxon)	0.0605***	---	-1.3818***
Countries (Continental)	0.1965	---	-1.0217***
Countries (Eastern Europe)	0.9836***	---	-1.1063***
Year (2002)^α	---	---	---
Year (2014)	-0.3114***	---	0.4618***
Intercept	-13.1241***	---	10.5872***
Random effects
LR test Prob >= chibar2	0.00

^aReference variable

*** . ** and * explanatory variables are statistically significant to 99.99 per cent, 99.95 per cent and 90 per cent.

Source: Authors’ estimations based on the ESS 1^st and 7^th round data and OECD, 2001 and 2013, minus Poland (2012 budget variables and 2014 immigrant unemployment rates).

Discussion

The main goal of this paper was identify the determinants of two ATIs (LMPerception and ETPerception) and their association with political ideologies in a European context. Although immigration does not appear have a detrimental effect on unemployment rates (Pischke and Velling, 1997) and it might even create employment in the host country (Peri, 2012), expenditure on active employment policies, unemployment protection and employment rates can influence ATIs. The results obtained in this present study are in line with those reported by Bridges and Mateut (2014) - higher immigrant employment rates dissuade people from believing that immigrants take away local jobs (taking as reference create new jobs), but active employment expenditure, unemployment protection expenditure, and a higher native employment rate encourage people to think that immigrants are bad for the economy.

It is necessary to remind that immigrant employment rates were 63 per cent, lower than native employment rates 68 per cent and it is likely people take into account this fact in the formation of attitude towards immigrants. Indeed, a change in the native employment rates is associated positively with perceiving immigrants bad for the economy. Interestingly, meanwhile native employment rates hearten self-positioning in the left wing of the political affiliation, economic threat perception discourage self positioning in this wing.

Governments with a deeper leftist orientation do not enlarge active labor market policies spending and reduce job creation programs (Tepe and Vanhuysse, 2013). This is in line with our results, a change in the employment public expenditure is associated positively with autopositioning in the right wing.

Some authors have reported empirical evidence that unemployment rates influence the formation of ATIs (Wilkes, Guppy and Farris, 2008; Markaki and Longhi 2013), as with Sides and Citrín (2007), and Jaime-Castillo, Marqués-Perales and Álvarez-Gálvez (2016), this present study found no such evidence. Higher levels of unemployment should not be expected to result in more negative ATI’s, as it is possible that the perception of rising unemployment is more important than the increase in the real unemployment rate (Kehrberg, 2007). The independent variable ETPerception was statistically significant but there was no statistical significance for LMPerception, public expenditure on active employment policies, unemployment protection and other concepts, and these results support the suggestion of Rustenbach (2010) that native populations may see immigrants as consumers and low-skilled workers who are good for the development of the economy and the welfare state. The arguments of Rustenbach (2010) are compatible with those of Jaime-Castillo, Marqués-Perales and Álvarez-Gálvez (2016); residents of host countries expect immigrants to integrate better and suffer less discrimination if the state is more generous. However, natives may also fear that increases in social spending will encourage more immigration, leading to higher levels of negative attitudes towards immigrants.

The belief that immigrants are bad for the economy was negatively correlated with left-wing views. Conversely, those who thought that immigrants were bad for the economy consider themselves to be politically right-wing. Our results support the association between right-wing views and negative attitudes towards immigrants (Semyonov, Raijman and Gorodzeisky, 2006; Skenderovic, 2007; Wilkes, Guppy and Farris, 2007; Andreescu, 2011; Dahlström and Esaiasson, 2013; Van Prooijen, Krouwel, Boiten and Eendebak, 2015).

With the exception of the welfare state system, this work did not control for individual characteristics. It is likely that micro factors like the immigration background of some respondents or the type of employment (private or public sector; business, self-employed, etc) can influence ATIs and political affiliations. Results could also vary in line with the type of question, for example, asking about the political party voted for at the last election (instead of political affiliation). Further research could consider these issues.

References

Alba, R., Rumbaut, R. G. and Marotz, K. (2005). “A distorted nation: Perceptions of racial/ethnic group sizes and attitudes toward immigrants and other minorities”. In Social forces, 84(2), 901-919. https://doi.org/10.1353/sof.2006.0002

Alesina, A., Harnoss, J. and Rapoport, H. (2016). “Birthplace diversity and economic prosperity”. In Journal of Economic Growth, 21, 101-138. https://doi.org/10.1007/s10887-016-9127-6

Andreescu, V. (2011). “Attitudes toward immigrants and immigration policy in United Kingdom”. In Journal of Identity and Migration Studies, 5(2), 61-85. Available at: http://jims.e-migration.ro/2011-JIMS%2C-volume-5%2C-number-2.php

Bauer, T. K., Lofstrom, M. and Zimmermann, K. F. (2001). Immigration policy, assimilation of immigrants, and natives' sentiments towards immigrants: Evidence from 12 OECD countries. IZA Discussion Paper No. 187.Bonn: Institute for the Study of Labor

Becchetti, L., Rossetti, F. and Castriota, S. (2010). “Real household income and attitude toward immigrants: an empirical analysis”. In The Journal of Socio-Economics, 39(1), 81-88. https://doi.org/10.1016/j.socec.2009.07.012

Betz, W. and Simpson, N. B. (2013). “The effects of international migration on the well-being of native populations in Europe”. In IZA Journal of Migration, 2, 1-21. https://doi.org/10.1186/2193-9039-2-12

Bierbrauer, G. and Klinger, E. W. (2002). “Political ideology, perceived threat, and justice towards immigrants”. In Social Justice Research, 15, 41-52. https://doi.org/10.1023/A:1016045731732

Bridges, S. and Mateut, S. (2014). “Should they stay or should they go? Attitudes towards immigration in Europe”. In Scottish Journal of Political Economy, 61(4), 397-429. https://doi.org/10.1111/sjpe.12051

Butkus, M., Maciulyte-Sniukiene, A., Davidaviciene, V. and Matuzeviciute, K. (2016). “Factors influencing society's attitudes towards internal and external EU immigrants”. In Filosifija sociologija., 27(4), 292-303. Available at: https://www.ceeol.com/search/article-detail?id=525173

Card, D., Dustmann, C. and Preston, I. (2005). Understanding attitudes to immigration: The migration and minority module of the first European Social Survey. CReAM Discussion Paper Número 3, London: Centre for Research and Analysis of Migration.

Castro, V. and Martins, R. (2019). “Budgets, expenditure composition and political manipulation”. In International Review of Applied Economics, 33(2), 172-187.

Ceobanu, A. M. and Escandell, X. (2010). “Comparative analyses of public attitudes toward immigrants and immigration using multinational survey data: A review of theories and research”. In Annual review of sociology, 36(1), 309-328. https://doi.org/10.1146/annurev.soc.012809.102651

Cochrane, C. and Nevitte, N. (2014). “Scapegoating: Unemployment, far-right parties and anti-immigrant sentiment”. In Comparative European Politics, 12, 1-32. https://doi.org/10.1057/cep.2012.28

Cohrs, J. C. and Stelzl, M. (2010). “How ideological attitudes predict host society members’ attitudes toward immigrants: Exploring cross‐national differences”. In Journal of Social Issues, 66(4), 673-694. https://doi.org/10.1111/j.1540-4560.2010.01670.x

Cutts, D., Ford, R. and Goodwin, M. J. (2011). “Anti‐immigrant, politically disaffected or still racist after all? Examining the attitudinal drivers of extreme right support in Britain in the 2009 European elections”. In European Journal of Political Research, 50(3), 418-440. https://doi.org/10.1111/j.1475-6765.2010.01936.x

Dahlström, C. and Esaiasson, P. (2013). “The immigration issue and anti-immigrant party success in Sweden 1970-2006: A deviant case analysis”. In Party Politics, 19(2), 343-364. https://doi.org/10.1177/1354068811407600

Dandy, J. and Pe-Pua, R. (2010). “Attitudes to multiculturalism, immigration and cultural diversity: Comparison of dominant and non-dominant groups in three Australian states”. In International Journal of Intercultural Relations, 34(1), 34-46. https://doi.org/10.1016/j.ijintrel.2009.10.003

Dustmann, C. and Preston, I. P. (2007). “Racial and economic factors in attitudes to immigration”. In The BE Journal of Economic Analysis & Policy, 7(1), Article Number 12. https://doi.org/10.2202/1935-1682.1655

Easterbrook, M. J., Kuppens, T. and Manstead, A. S. (2016). “The education effect: Higher educational qualifications are robustly associated with beneficial personal and socio-political outcomes”. In Social indicators research, 126, 1261-1298. https://doi.org/10.1007/s11205-015-0946-1

Fertig, M. and Schmidt, C. M. (2011). “Attitudes towards foreigners and Jews in Germany: identifying the determinants of xenophobia in a large opinion survey”. In Review of Economics of the Household, 9, 99-128. https://doi.org/10.1007/s11150-009-9084-3

Florian, H. Ã., Sturm, J. E. and Potrafke, N. (2020). Determinants of social expenditure in OECD countries (No. 20-475). KOF Swiss Economic Institute, ETH Zurich.

Fromentin, V. (2013). “The relationship between immigration and unemployment: The case of France”. In Economic Analysis and Policy, 43(1), 51-66. https://doi.org/10.1016/S0313-5926 13 50003-8

Gorodzeisky, A. and Semyonov, M. (2018). “Competitive threat and temporal change in anti-immigrant sentiment: Insights from a hierarchical age-period-cohort model”. In Social Science Research, 73, 31-44. https://doi.org/10.1016/j.ssresearch.2018.03.013

Hainmueller, J. and Hiscox, M. J. (2010). “Attitudes toward highly skilled and low-skilled immigration: Evidence from a survey experiment”. In American political science review, 104(1), 61-84. Available at: https://www.jstor.org/stable/27798540

Haubert, J. and Fussell, E. (2006). “Explaining pro-immigrant sentiment in the US: Social class, cosmopolitanism, and perceptions of immigrants”. In International Migration Review, 40(3), 489-507. https://doi.org/10.1111/j.1747-7379.2006.00033.x

Jaime-Castillo, A. M., Marqués-Perales, I. and Álvarez-Gálvez, J. (2016). “The impact of social expenditure on attitudes towards immigration in Europe”. In Social Indicators Research, 126, 1089-1108. https://doi.org/10.1007/s11205-015-0939-0

Kehrberg, J. E. (2007). “Public opinion on immigration in Western Europe: Economics, tolerance, and exposure”. In Comparative European Politics, 5, 264-281. https://doi.org/10.1057/palgrave.cep.6110099

Kemnitz, A. (2003). “Immigration, unemployment and pensions”. In Scandinavian Journal of Economics, 105(1), 31-48. https://doi.org/10.1111/1467-9442.00003

Kunovich, R. M. (2004). “Social structural position and prejudice: An exploration of cross-national differences in regression slopes”. In Social Science Research, 33(1), 20-44. https://doi.org/10.1016/S0049-089X 03 00037-1

Lancee, B. and Sarrasin, O. (2015). “Educated preferences or selection effects? A longitudinal analysis of the impact of educational attainment on attitudes towards immigrants”. In European Sociological Review, 31(4), 490-501. https://doi.org/10.1093/esr/jcv008

Markaki, Y. and Longhi, S. (2013). “What determines attitudes to immigration in European countries? An analysis at the regional level”. In Migration Studies, 1(3), 311-337. https://doi.org/10.1093/migration/mnt015

Meuleman, B., Davidov, E. and Billiet, J. (2009). “Changing attitudes toward immigration in Europe, 2002–2007: A dynamic group conflict theory approach”. In Social science research, 38(2), 352-365. https://doi.org/10.1016/j.ssresearch.2008.09.006

OECD (2018a), Foreign born-unemployment. https://doi.org/10.1787/ba5d2ce0-en Accessed on 22 August 2020.

OECD (2018b). Native-born unemployment. https://doi.org/10.1787/0f9d8842-en  Accessed on 22 August 2020.

OECD (2018c). Public spending on labour markets. https://data.oecd.org/socialexp/public-spending-on-labour-markets.htm Accessed on 4 January 2020.

OECD (2018d). Public unemployment spending. https://data.oecd.org/socialexp/public-unemployment-spending.htm Accessed on 4 January 2020.

OECD (2018e). Foreign born employment. https://data.oecd.org/migration/foreign-born-employment.htm Accessed on 4 January 2020.

OECD (2018f). Native-born employment. https://data.oecd.org/migration/native-born-employment.htm Accessed on 4 January 2020.

Paas, T. and Halapuu, V. (2011). “Competition for people as drivers of future economic growth: European people’s attitudes towards migration”. In Discussions on Estonian Economic Policy, 19(2), 58-71. http://dx.doi.org/10.2139/ssrn.1998058

Pardos-Prado, S. (2011). “Framing attitudes towards immigrants in Europe: When competition does not matter”. In Journal of Ethnic and Migration Studies, 37(7), 999-1015. https://doi.org/10.1080/1369183X.2011.572421

Peri, G. (2012). “The effect of immigration on productivity: Evidence from US states”. Review of Economics and Statistics, In 94(1), 348-358. https://doi.org/10.1162/REST_a_00137

Pichler, F. (2010). “Foundations of anti-immigrant sentiment: The variable nature of perceived group threat across changing European societies, 2002-2006”. In International Journal of Comparative Sociology, 51(6), 445-469. https://doi.org/10.1177/0020715210379456

Pischke, J. S. and Velling, J. (1997). “Employment effects of immigration to Germany: an analysis based on local labor markets”. In Review of Economics and Statistics, 79(4), 594-604. https://doi.org/10.1162/003465397557178

Rustenbach, E. (2010). “Sources of Negative Attitudes Toward Immigrants in Europe: A Multi‐Level Analysis 1”. In International migration review, 44(1), 53-77. https://doi.org/10.1111/j.1747-7379.2009.00798.x

Scheepers, P., Gijsberts, M. and Coenders, M. (2002). “Ethnic exclusionism in European countries. Public opposition to civil rights for legal migrants as a response to perceived ethnic threat”. In European sociological review, 18(1), 17-34. https://doi.org/10.1093/esr/18.1.17

Semyonov, M., Raijman, R. and Gorodzeisky, A. (2006). “The rise of anti-foreigner sentiment in European societies, 1988-2000”. In American sociological review, 71(3), 426-449. https://doi.org/10.1177/000312240607100304

Sides, J. and Citrin, J. (2007). “European opinion about immigration: The role of identities, interests and information”. In British journal of political science, 37(3), 477-504. https://doi.org/10.1017/S0007123407000257

Skenderovic, D. (2007). “Immigration and the radical right in Switzerland: ideology, discourse and opportunities”. In Patterns of Prejudice, 41(2), 155-176. https://doi.org/10.1080/00313220701265528

StataCorp. 2023. Stata 18. Statistical software. StataCorp LLC.

Tepe, M. and Vanhuysse, P. (2013). “Parties, unions and activation strategies: The context-dependent politics of active labour market policy spending”. In Political Studies, 61(3), 480-504.

Van Prooijen, J. W., Krouwel, A. P., Boiten, M. and Eendebak, L. (2015). “Fear among the extremes: How political ideology predicts negative emotions and outgroup derogation”. In Personality and social psychology bulletin, 41(4), 485-497. https://doi.org/10.1177/0146167215569706

Wilkes, R., Guppy, N. and Farris, L. (2007). “Right-wing parties and anti-foreigner sentiment in Europe”. In American Sociological Review, 72(5), 831-840. https://doi.org/10.1177/000312240707200509

Wilkes, R., Guppy, N. and “Farris, L. (2008). “`No Thanks, We’re Full: Individual Characteristics, National Context, and Changing Attitudes Toward Immigration”. In International Migration Review, 42(2), 302-329. https://doi.org/10.1111/j.1747-7379.2008.00126.x

Authors' curriculum vitae

Víctor Bazán Monasterio

Has a PhD in Economics and Management of Organizations. His doctoral thesis studied attitudes towards immigrant people in Europe. While managing projects, he has published articles related to human resource management and business, as well as researched attitudes towards science, technology, and well-being in Europe.

Email: victorbazan@unizar.es

Ana Isabel Gil Lacruz

Is a Full Professor in the Department of Business Management and Organisation at the University of Zaragoza. She received the Extraordinary Award in Social Sciences from the University of Zaragoza for her European PhD in Economics (Excellent Cum Laude). Recognized for her extensive academic and research career, she has played key roles in coordinating and developing national and international research projects focusing on corporate social responsibility, workplacewell-being, and sustainable development. Her work addresses interdisciplinary issues integrating economics, sociology, and applied psychology. Dr. Gil Lacruz is a research member of the University Institute for Research in Employment, Digital Society, and Sustainability (IEDIS). She is also a member and co-director of the Consolidated Research Group of the Government of Aragon and the University of Zaragoza, which focuses on 'Well-being and Social Capital' (https://bycs.unizar.es/). Additionally, she is part of the Campus Iberus Action Group “HealthyOrg” (https://www.campusiberus.es/grupo-de-accion-healthyorg/) and the Group of “Leadership and Health Management Studies of Aragon” (https://www.iisaragon.es/grupos-de-investigacion/). As a member and co-director of these research groups, She has promoted the intersection of technological and social innovation for worker health. This interdisciplinary approach reflects her commitment to generating applied knowledge and advancing the United Nations Sustainable Development Goals. Dr. Gil Lacruz has led multiple competitively funded projects, including recent initiatives aimed at promoting occupational health as a component of CSR and its impact on worker well-being. Notable among these is her leadership in the project "Healthy Companies: CSR and Worker Well-being," developed under the Applied Research modality and positively evaluated for its multidisciplinary approach and social applicability. These projects have yielded tangible results in the form of indexed scientific publications, management tools, and recommendations for public policy. She has had the opportunity to work in different national and international institutions (Universidad de Zaragoza, Cornell University-USA, Universidad Autónoma de Madrid and Universidad de Alcalá). She has also carried out numerous research stays in universities, research centers and international organizations (United Nations, London School of Economics, FEDEA, Deutsches Institut für Wirtschaftsforschung, The University of Sydney or Universidad del Pacífico, among others). She has recently collaborated as an external expert for Cornell University, the World Bank and the European Commission. She has actively participated in the European Health Gastein Forum since she was invited as a young academic in 2009. Undoubtedly, the knowledge of foreign languages (English, German and French) has promoted the development of this activity. Her scientific output includes over 100 articles in indexed journals, nearly 70 of which are listed in the Journal Citation Report (JCR), with a significant representation in Q1 and Q2 quartiles. Her recent publications include key contributions on the interplay between CSR, workplace health promotion, and sustainable development goals. These publications have been cited as references in impact evaluations and knowledge transfer activities in academic and business contexts. In the educational context, she has supervised more than 15 doctoral theses, many of which address the practical application of CSR and well-being concepts in organizational settings. Her teaching activities include modules in international master's programs and doctoral courses focusing on experimental economics, human resource management, and organizational psychology. This experience is supported by innovative education projects, such as implementing chatbots for personalized learning and gamification as a pedagogical tool.

Email: anagil@unizar.es

Marta Gil Lacruz

Graduate and doctorate in psychology and sociology; her Ph.D. thesis: “Health system and community” was awarded an extraordinary doctorate in Social Sciences from the University of Zaragoza. Thanks to a grant from the Ministry of Education, she completed a two-year postdoctoral fellowship at the University of California, Riverside (United States). Currently, she is a tenured professor specialized in health psycho sociology at the Faculty of Health Sciences. In the third cycle, she teaches in the following Official University Masters: Gerontology, Teacher Education of Compulsory Secondary Education, Gender and Equal Opportunities, Public Health, Sociology of Public Policies Master. She has supervised 1doctoral theses, six of them with international mention and one, with an University extraordinary award. From 2017 to 2021 (June) she directed the official master's degree in Sociology of Public Policies (University of Zaragoza). She participated in the design, development and evaluation of the European Union Program: Education and Culture DG. Lifelong Learning Program led by Mikkeli University (Finland). This network integrates Universities from Germany, Finland, Wales, Poland and Spain. Regarding research topics, Marta Gil Lacruz is director of the consolidated group: Wellbeing and Social Capital, recognized since 2005 by the University of Zaragoza and the Government of Aragon as a reference research group (BYCS). She has been the director of 6 consecutive research projects in public I + D + I national research programs. In recent years, her teacher's work could be characterized by its international projection. Highlighting the research work carried out at the Universities of Kyoto, Sophia, Hygio and Wakayama in Japan, University of Chile, the Autonomous University of Mexico, the University of Lima, the University of Guatemala, the Women's Chair of the University of Havana, the Faculty of Political Sciences of the University of Bucharest (Romania) and the University of Stendhal in Grenoble, Perpignan, Lille and Toulouse (France) Sapienza University (Italy), Athena University (Greek). One of her training stays has been in the Department of Health and Environment at the headquarters of the World Health Organization (Geneva, Switzerland). Her book publications include the monograph edited by the Sociological Research Center: “Health and sources of social support” and “Social Psychology didactic guide”, published by Promolibro, “Social Psychology and Wellbeing” by Press University; “Bodies that speak: genders, identities and social representations” ed. Montesinos Essay; “Health and Gender: Neighborhood Research for Community Development”, ed. Sight; “The labyrinth of public health” ed. Erasmus; “Ibero-American immigration processes in the province of Teruel. Social capital and community development “. One of her knowledge transfer publications is a multidisciplinary and bilingual didactic guide for community development and has been edited by the Education for Health Foundation (Madrid).

Email: mglacruz@unizar.es

Annex 1. Process of getting the data

Data was collected from European Social Survey (ESS) and Organisation for Economic Co-operation and Development (OECD). Individual observations, which included perceptions about immigrants, political affiliation and sociodemographic variables, was collected from ESS. Attitudes towards people from other country was questioned as follow:

Labour market competition perception

The question related to Labour market competition perception was gathered asking: Using this card, would you say that people who come to live here generally take jobs away from workers in (country), or generally help to create new jobs?

The possible options were 11, from 00 “Take jobs away” to 10 “Create new jobs”. It was also available the option “Don’t know”. Observations “Don´t know” was missing.

Economic threat perception

The question related to Economic threat perception was gathered asking: Using this card, Would you say it is generally bad or good for (country)’s economy that people come to live here from other countries? Please use this card.

The possible options were 11, from 00 “Bad for the Economy” to 10 “Good for the Economy”. It was also available the option “Don’t know”. Observations “Don´t know” was missing.

These two variables related to attitudes towards immigrants were changed to dummy variables in order to be analysed with xtmelogit. In doing so, ESS values from 0 to 3 were taken as 1 (‘Yes’), and 0 otherwise; the variables reflect the opinion that immigrants take jobs away or are bad for economy.

Because of the creation of dummies variables, it was also possible to include them as independent (explanatory) variables of political affiliation, analysed with the multinomial model mlogit of STATA.

Data from OECD was gathered from its database. Regarding public spending, in constant prices (2001) per person and purchasing power parity adjusted:

Foreign born-unemployment. https://doi.org/10.1787/ba5d2ce0-en

Native-born unemployment. https://doi.org/10.1787/0f9d8842-en

Public spending on labour markets. https://data.oecd.org/socialexp/public-spending-on-labour-markets.htm

Public unemployment spending. https://data.oecd.org/socialexp/public-unemployment-spending.htm

Foreign born employment. https://data.oecd.org/migration/foreign-born-employment.htm

Native-born employment. https://data.oecd.org/migration/native-born-employment.htm

---

[1] This study was also repeated with mixed model analyses. The technique is a linear model which calculates both fixed and random effects; the MLOGIT estimates maximum-likelihood multinomial logit models. Similar results to the Multinomial logistic model were obtained which shows results from a reference variable, the results from the mixed models are not reported here.

[2] Only for comparing, according to the Official European Commission Exchange rate, 1 EUR = USD 0.8887 in 2001 and 1 EUR = USD 1.3592 in 2013, both on month 12.