Does the rejection of immigration have objective economic bases? In other words, can changing economic conditions in a country affect attitudes towards immigrants? And do employment conditions on an individual level also affect those attitudes? To answer these questions, two types of analysis are presented. Firstly, the impact is analysed of the economic crisis of 2008-2012 on the attitudes of native workers in a score of European countries. It is confirmed that rejection of immigration grew more in countries that suffered greater falls in GDP during the first phase of the recession (2008-2010), especially if those countries had additionally experienced high growth in their immigration rates in the years prior to the crisis. Secondly, the individual conditioning factors of attitudes towards immigration are analysed and it is confirmed that both the economic vulnerability of households and the exposure of workers to labour market competition are related to a significant increase in rejection attitudes towards immigration in Europe.

Studying the effects of the economic crisis of 2008-2012 on attitudes towards immigration is of enormous interest because rejection of immigrants is at the heart of many of the populist discourses currently booming in Europe. For social researchers, the study of this crisis is also extremely interesting from a methodological viewpoint because it offers a kind of natural experiment: a strong external shock that simultaneously affects all countries in Europe, but does so with varying intensities. This variation in the recession’s impact greatly facilitates the causal identification of the relationship between economic outlook and attitudes.

In a piece of research published in Socio-Economic Review (Polavieja, 2016) the impact was analysed of the first phase of the crisis on changing attitudes towards immigration among European workers, using data from the second and fifth waves of the European Social Survey, conducted respectively in 2004-2005 and 2010-2011, i.e., before and after the recession’s first phase, which ran from 2008 to 1010 and was the most intense phase. In the study mentioned a representative sample of approximately 35,000 native workers from twenty European countries  was used.

All workers were employed and aged between 20 and 64 years old. These data enabled estimation of what the average change had been in attitudes towards immigrants in each country between the two waves studied, once differences in education and the age and gender structure of the working population were adjusted for. The estimates also took into account a long series of subjective indicators related to attitudes towards immigration (political ideology, social confidence, religiousness, egalitarian values and propensity toward happiness), therefore the estimated change between the two waves can be considered a net change.

Graph 1 shows the average scores on a scale of rejection of immigration that ranges from value -10 (minimum anti-immigration sentiment) to +10 (maximum rejection of international immigration) for the typical worker in each of the twenty countries in the study in the year 2004, as well as the net change in attitudes towards immigration taking place between 2004 and 2010. We use the term “typical worker” to mean workers with average scores in all the explanatory variables of the model (education, age, gender and subjective control indicators), in other words, the most representative worker of each country. What do we find?

Firstly, we discover that a major variation exists both in average starting attitudes (2004) and in the intensity and direction of the change. We also observe that the two parameters are not related to each other, in other words the intensity of the change during the crisis does not depend on the initial values. Looking specifically at the magnitude of the change in attitudes that took place between 2004 and 2010, we must highlight the high degree of variation observed among the 20 countries analysed. In fact, the rejection of immigration among native workers increased significantly in seven countries (Ireland, Greece, Spain, Czech Republic, Hungary, Slovakia and the United Kingdom), while it fell significantly in another seven (the Netherlands, Poland, Denmark, Germany, Sweden, Portugal and Estonia) and did not show any significant variation in the remaining six. Could these differences observed in change of attitude respond to the different intensity of the recession in the countries analysed?

A simple analysis relating attitudes towards immigration and the economic situation enables us to answer this question on macroeconomic determinants affirmatively. In effect, we find statistically significant correlation between the intensity of the recession (measured as a difference in the rate of growth of Gross Domestic Product between 2004 and 2010) and the change in attitudes. In general, the worse the economic situation (i.e., the deeper the fall in GDP), the more rejection of immigration increased among native workers (graph 2). Thus, for example, we see that, between 2004 and 2010, the Greek GDP fell by 8 percentage points and rejection of immigration increased by an average of 0.8 points on our scale; meanwhile in Germany, the GDP did not fall, rather it grew by 2.5 percentage points during the period, and the rejection of immigration fell by 0.44 points. In general, we observe a tendency that links the severity of the crisis with the increase in rejection. In other words, the rejection of immigration was greater in countries with greater falls in GDP during the crisis. This tendency is expressed by the red line in graph 2.

The experience of the recession is not, however, the only macroeconomic factor that is important when explaining the magnitude of the change in attitudes of European workers with respect to immigration. There is another factor that appears to be equally decisive: the evolution of the immigration rate prior to the crisis (graph 3). Countries such as Greece, Ireland and Spain not only experienced strong falls in GDP during the first phase of recession but they did so after experiencing a fast increase in the rate of immigration in the preceding years. Again, it is important to highlight that it does not seem to be the size of the foreign population in itself (the percentage of immigrants), but the speed of its growth before the crisis that favours (in combination with the intensity of the recession) the increase in the rejection of immigration in Europe between 2004 and 2010 (Polavieja, 2016). In short, the data suggest that the rejection of immigration increased more in those countries that experienced fast growth in their immigrant population before the crisis.

The macroeconomic analyses that we present in this article coincide with the so-called realist conflict theories (see, for example, Ceobanu and Escandell, 2010), according to which fast growth of the immigrant population combined with a strong deterioration in the economic situation would add up to increased competitive pressure on native workers. However, to demonstrate that the rejection of immigration has economic bases, it is not sufficient to establish macroeconomic correlations. We need evidence that connects the specific economic experiences of European workers with their attitudes against immigration in the micro sphere, in other words, microeconomic data. This requires individual indicators of economic vulnerability and exposure to competition in the labour markets.

To date, the majority of studies on individual conditioning factors of attitudes towards immigration had focused on the education of those surveyed as a measure of their market power (Mayda, 2006). A host of studies have demonstrated that the higher the educational level of the survey respondent, the lower their rejection of immigrants (Hainmueller and Hopkins, 2014). However, if what is wanted is proof that increased labour market competition, caused by the arrival of foreign workers, generates increased rejection of immigration, then looking at education level is problematic. This is because education can affect attitudes directly and via pathways that have nothing to do with working experience, for example, by increasing tolerance, a taste for diversity or political correctness. For this reason, to obtain direct evidence of economic conditioning factors in the rejection of immigration, objective indicators are needed that do not depend on education level.

In an older study (Polavieja, 2016), it was proposed to use three occupational indicators that are directly related with workers’ degree of replaceability and, therefore, their exposure to labour market competition. These indicators are: 1) the importance of on-the-job-training (the requirements of specific on-the-job training); 2) the difficulty for employers to measure with precision worker productivity (monitoring costs); and 3) the mixture of communication skills and manual skills required for each occupation. Specific on-the-job training protects workers from outside competition by equipping them with skills that make them less replaceable. Among the occupations with the highest on-the-job training, there are many with low formal education requirements (for example, cable assemblers, tobacco producers, clockmakers, etc.).

The costs of measuring productivity (monitoring costs) are directly related to degree of exposure to labour market competition, since the more difficult it is for employers to measure worker productivity, the less their workers are exposed to being replaced. Thus, as happens with on-the-job training, the high costs of measurement favour the closure of employment relations to external competition (Goldthorpe 2007). Among the jobs with the highest measuring costs again there are many occupations that do not require high levels of formal education such as, for example, transport workers, metal workers and office staff.

Finally, the mixture of communication skills and manual skills required for each occupation is another potentially key factor, because native workers have an advantage when performing tasks that require linguistic skills and specific cultural knowledge of their country. Thus, the greater the occupation’s communication content, the less competition faced from immigrants. Again, we find variations in the mixture of manual and communication skills in occupations with very different requirements in terms of formal education (for example between psychologists and surgeons, or between machinery salespeople and operators). Using these three objective dimensions of occupations enables us to better identify the economic bases of rejection for immigration without falling into the problems of interpretation raised by the use of education levels.

By combining the second and fifth waves of the European Social Survey, it is possible to estimate the net impact of each of these occupational dimensions for a sample of approximately 35,000 workers (graph 4). A positive value reflects the fact that, adjusting for other aspects such as country, sex, ideology and socio-political values, that variable or dimension increases rejection of immigration. In contrast, a negative value reflects a reduction in the rejection of immigration. The impact of formal education, represented in the first bar with a negative value,  is very evident: it reflects how a higher educational level among workers considerably reduces the rejection that they show towards immigration. In other words, those with a higher level of education show less rejection of immigration.

Other factors that are also relevant for explaining the rejection of immigration are the three objective occupational indicators of the degree of labour market competitiveness of workers, i.e.: specific on-the-job training, costs of measuring productivity and communication skills, all of which have a negative impact on the rejection of immigration. In effect, the results of the second, third and fourth bars show how survey respondents employed in positions that require high specific training, those employed in occupations that have a high content in communication skills tend to show lower levels of rejection toward immigration than those employed in occupations that have low specific training, low costs in measuring productivity and low costs in communication skills, respectively. In other words, exposure to labour market competition, measured through these three occupational indicators, increase the rejection of immigration.

Finally, graph 4 also includes an indicator of economic vulnerability of the households of survey respondents (fifth bar). The indicator distinguishes between survey respondents who state that they have difficulties making it to the end of the month in their households and those who do not. As observed in the graph, workers with economic difficulties in their households present significantly higher scores on the rejection of immigration scale, even once the effect of the rest of the variables estimated is neutralised. In other words, European workers who state that they have difficulties getting to the end of the month in their households show a higher rejection of immigration. Between 2004 and 2010, the number of people in households with economic difficulties increased considerably in the majority of countries most punished by the recession (in Ireland the number increased from 13% to 37%, in Greece from 38% to 59%, in Hungary from 33% to 49% and in Spain from 13% to 24%, according to own estimates with data from the European Social Survey). This increase in economic vulnerability of households directly connects the recession with the increase in the rejection of immigrants in Europe.

The results presented in this study suggest that the rejection of immigration has objective economic bases. Firstly, it has been seen how the macroeconomic scenario most favourable for increasing the rejection of immigration is one where an accelerated increase in immigration takes place followed by a severe economic crisis. This was the sequence experienced by Ireland, Greece and Spain, three economies with a growth pattern very closely linked to the construction bubble. Out of twenty countries analysed, these three showed the greatest increases in rejection of immigration during the first phase of the recession.

In second place, it was seen how specific job training, the costs of measuring productivity and the mixture of communication and manual skills required for the post are three very relevant dimensions of employment for the study of attitudes towards immigration. These three dimensions are related with the degree of exposure of workers to competition in the labour markets. The data provided indicate that workers employed in occupations more exposed to competitive market pressures tend to express greater rejection toward immigration, independently of their level of education.

Finally, it was seen how workers who claimed to be experiencing economic difficulties in their households also tended to show significantly higher scores on the rejection of immigration scale, independently of their own employment situation, sociodemographic characteristics and socio-political attitudes and values. The economic crisis of 2008-2012 increased the economic vulnerability of many European families, acting as a mechanism that directly links the macroeconomic situation with the rejection of immigration.

The rejection of immigration lies at the base of the national-populist discourses that are booming in Europe. It is also in the same base of support for Trump in the United States. The political consequences of this rejection cannot, therefore, be  minimalised. We are facing a political problem of maximum proportions. The results presented in this study question the more subjectivist interpretations of this problem, according to which the rejection of immigration would have fundamentally ideological bases. It is true that the ideological maps, identities and values of citizens play a fundamental role in the shaping of attitudes towards immigrants, but it would be an error to think that education in values can, on its own, turn the current levels of rejection around. For many European workers, economic vulnerability and exposure to labour market competition constitute very real experiences. Populism feeds off these experiences as much or more as off the values of our fellow citizens

CEOBANU, A. M., e X. ESCANDELL (2010): “Comparative analysis of public attitudes toward immigrants and immigration using multinational survey data: a review of theories and research”, Annual Review of Sociology, 36.

EUROSTAT (2015): Real GDP growth rate-volume [12.08.2015].

GOLDTHORPE, J. (2007): On sociology, vol. 2, Stanford, CA: Stanford University Press.

HAINMUELLER, J., e D. J. HOPKINS (2014): “Public attitudes toward immigration”, Annual Review of Political Science, 17.

MAYDA, A. (2006): “Who is against immigration? A cross-country investigation of individual attitudes toward immigrants”, The Review of Economics and Statistics, 88.

OCDE (2015): Foreign-born population [indicador].

O’ROURKE, K. H., e R. SINOTT (2006): “The determinants of individual attitudes towards immigration”, European Journal of Political Economy, 22.

POLAVIEJA, J. (2016): “Labour-market competition, recession and anti-immigrant sentiments in Europe: occupational and environmental drivers of competitive threat”, Socio-Economic Review, 14(3).