Left-Wing Xenophobia in Europe

Given rising populist nationalism and multiplying meanings of “right” and “left,” this paper assesses whether Europeans who identify as extremely left-wing on the political spectrum hold anti-immigrant attitudes. In contrast to right-wing xenophobes, we further examine whether the political left, who conventionally emphasize class conflict, oppose immigrants less for cultural reasons and more for materialist reasons. We also consider whether socioeconomic status and values traditionally associated with the political left—favoring redistributive policies, egalitarianism, or social rights to benefits and services for immigrants—temper left- more than right-wing xenophobia. We find that a surprisingly large share of those who identify as far left do express extremely xenophobic attitudes, and we profile them in contrast to far right xenophobes. With logistic regression analysis of nine waves of the European Social Survey (2002–2018), we find that, all things equal, socioeconomic status influences far left xenophobia more than far right xenophobia, but inegalitarian values, less support for redistributive policies, and welfare chauvinism can only partially account for far left xenophobia and unexpectedly do not distinguish it from far right xenophobia. This implies that far left parties might adopt anti-immigrant policies to try to retain their loyal voters, even though such policies do not comport with broader left-wing values and may increase racial and ethnic inequality. Controlling for demographic and attitudinal differences reduces the probability of xenophobia among the far left by about sixty percent, but there remains some residual anti-immigrant attitudes among this group still to be explained.

Whether the respondent has a post-secondary degree (short-cycle tertiary education and above, ISCED level 5) What is the highest level of education you have achieved? (edulvla and edulvlb) Dichotomous (recoded) 0= no post-secondary degree 1= having a post-secondary degree Labor Force Status Describes the respondent's main activity. "Out of the labor force" includes respondents who are in education, unemployed and not looking for job, permanently sick or disabled, retired, or doing community or military service, or housework Which of these descriptions best describes your situation (in the last seven days)? (mnactic) Categorical ( Agreement with the statement that the government should reduce differences in income levels. Those who are coded 1 either agree or strongly agree with the statement. Please say to what extent you agree or disagree with each of the following statements "The government should take measures to reduce differences in income levels". (gincdif) Dichotomous (recoded) 0= reject income redistribution 1=support income redistribution

Importance of Egalitarianism
The respondent's attitudes towards equal treatment and opportunity. The category of respondents who find egalitarianism "somewhat important" includes those who replied "somewhat" and "a little like me," and the category "not important" those who replied "not like me" and "not like me at all."  The importance the respondent assigns to following traditions and customs. The category "somewhat important" includes those who replied "somewhat" and "a little like me," and the category "not important" those who replied "not like me" and "not like me at all." Now I will briefly describe some people. Please listen to each description and tell me how much each person is or is not like you. "Tradition is important to her/him. She/he tries to follow the customs handed down by her/his religion or her/his family." (imptrad) Categorical ( Table 2. SE = Standard Error. p-values: + p < .1, * p < .05, ** p < .01, *** p Note: The model calculates the predicted probabilities for the average person with the characteristics in Table 1, controlling for all variables included in the logistic regression on extremely anti-immigrant attitudes reported in Table 2.   Source : European Social Survey, cumulative dataset, waves 1-9. Notes: The model controls for all variables included in the logistic regression on extremely antiimmigrant attitudes reported in Table 2. SE = Standard Error. p-values: + p < .1, * p < .05, ** p < .01, *** p Note: The model calculates the predicted probabilities for the average person with the characteristics in Table 1, controlling for all variables included in the logistic regression on extremely anti-immigrant attitudes reported in Table 2

Interaction Effect of Political Orientation and Social Trust from Logistic Regression on Extremely Anti-Immigrant Attitudes
Source : European Social Survey, cumulative dataset, waves 1-9. Notes: The model controls for all variables included in the logistic regression on extremely antiimmigrant attitudes reported in Table 2. SE = Standard Error. p-values: + p < .1, * p < .05, ** p < .01, *** p Note: The model calculates the predicted probabilities for the average person with the characteristics in Table 1, controlling for all variables included in the logistic regression on extremely anti-immigrant attitudes reported in Table 2

90% Confidence Interval
Note: The model calculates the predicted probabilities for the average person with the characteristics in Table 1, controlling for all variables included in the logistic regression on extremely anti-immigrant attitudes reported in Table 2.
Source : European Social Survey, cumulative dataset, waves 1-9. Notes: The model controls for all variables included in the logistic regression on extremely antiimmigrant attitudes reported in Table 2. SE = Standard Error. p-values: + p < .1, * p < .05, ** p < .01, *** p      Interaction of Political Ideology and Importance of Traditionalism

90% Confidence Interval
Note: The model calculates the predicted probabilities for the average person of the sample used in the regression analysis reported in Table 3, controlling for all variables included in the logistic regression on extremely anti-immigrant attitudes reported in Table 3.
Note: The model calculates the predicted probabilities for the average person with the characteristics in Table 1, controlling for all variables included in the logistic regression on extremely anti-immigrant attitudes reported in Table 2.