That’s one conclusion of a methodologically innovative study of European immigrants suggesting that, other qualifications being equal, migrants from certain countries may be roughly 40 percent less likely than others to gain citizenship.
Even if an immigrant has obtained a good education and job, and solid language skills, simply being from the wrong country can be a “massive disadvantage,” says Jens Hainmueller, an associate professor of political science at MIT and a co-author of the study, along with Dominik Hangartner of the London School of Economics and Political Science.
“By far the most decisive factor in the applicants of immigrants is their country of origin,” Hainmueller says.
However, there is good news for immigrants: The study also suggests that such biases against people from particular nations tend to shift over time, and often occur due to temporary social circumstances, such as the number of immigrants arriving from one country at a particular moment.
“The results suggest that interactions with immigrant groups can really remove prejudice over time,” Hainmueller says.
Citizens speaking at the ballot box
The study’s findings are based on voting from 1970 to 2003 in Switzerland, where many municipalities used direct referendums in which local citizens voted on citizenship applications — an unusual approach among European countries. In particular, immigrants from Turkey and the former Yugoslavia fared worse in the voting than applicants from other countries. The results are in a paper — “Who Gets a Swiss Passport? A Natural Experiment in Immigrant Discrimination” — published in the latest issue of the American Political Science Review.
The use of voting data allowed Hainmueller and Hangartner to circumvent problems that may occur when studying attitudes toward immigration in other ways. While public opinion surveys about immigration may be useful, not all respondents reveal their true views to pollsters, especially if they think those opinions might seem controversial. But the Swiss data reflect “people who are just voting their true preferences, what they really think about this,” Hainmueller says. “And the outcome really mattered to people.”
Moreover, the detailed descriptions of the immigrants on their applications allowed Hainmueller and Hangartner to construct close matches between applicants, finding cases in which, for instance, characteristics such as the level of education or type of job were equal, but the country of origin differed.
All told, the researchers studied 2,429 naturalization cases in 44 municipalities. The overall rejection rate for applicants was 37 percent — but for Turkish and Yugoslavian immigrants, that increased to more than 50 percent, an increase of about 40 percent compared to the overall rate. By contrast, for immigrants applying for citizenship from central or eastern Europe, Asia, or other non-European countries, the rejection rate never topped 45 percent, while immigrants from southern Europe fared better than the average applicant.
To an extent, the researchers believe, this bias reflects the presence of a straightforward anti-immigrant sentiment among many voters; in municipalities where Switzerland’s leading anti-immigration party was popular, voters were three times as likely to reject Turkish or Yugoslav naturalization applications, compared to the municipalities where the anti-immigration party was least popular. “It has much to do with stereotypes and prejudice,” Hainmueller says.
However, Hainmueller says, the voting results do not reflect an immutable Swiss bias against Turks or Yugoslavs. The source of the prejudice, the researchers believe, is partly due to the volume of immigration from the countries in question. The dynamic, Hainmueller says, represents a kind of “threat mechanism” in which some Swiss citizens become concerned that having too many immigrants from one particular country at one time will disrupt the social order.
“As the immigrant groups got larger, particularly in the 1990s when there was a lot of immigration from Turkey and Yugoslavia, the country of origin-based discrimination increased dramatically,” Hainmueller says. “Turks were doing all right [in naturalization votes] in the 1970s and 1980s when there were not as many around.” In the 1960s and 1970s, a larger proportion of immigrants to Switzerland were Italian, and in turn, Italians fared worse in naturalization votes in the 1970s.
But over time, Hainmueller suggests, immigrants from any given country become more accepted in their new land, because “people get used to them, and recognize their contribution to society.”
Power to the people?
While the study focused on Switzerland, Hainmueller suggests that the results, with proper caution, may apply to other countries as well.
As Hainmueller notes, the precise constellation of political forces and social conditions that affects sentiment about immigrants varies in every country. However, he adds, survey evidence — whatever its limitations — does show similar levels of anti-immigrant sentiment across many European countries.
“It’s a big topic that, in particular, right-wing parties are using to drum up support now,” Hainmueller says. “We see this in France, Denmark, Austria and other countries. If in these countries, people were given the opportunity to vote on these applications, it does strike me as plausible that the results could look similar.”
Rafaela Dancygier, a political scientist at Princeton University, calls the study “very important in helping us understand native hostility toward immigrants,” particularly in its use of data gathered over three decades, which she terms “a big improvement over most work, which tends to only take a snapshot of a given point in time.” This allows researchers to analyze the interplay between changing economic and political circumstances and attitudes toward immigrant groups.
Stanford University political scientist David Laitin also praises the paper’s use of the data, noting, “It is as if the Swiss cantons held up a generous gift to social science and Hainmueller and Hangartner ingeniously grabbed it.” Still, he adds, scholars should “keep looking for replication opportunities” to see how generally the current findings might apply across Europe.
Hainmueller thinks the immigration issue also bears on the question of when direct democracy should be used, as opposed to indirect democratic mechanisms.
“I think the results suggest that the [impact] of direct democracy can be very substantial for these immigrants,” Hainmueller says. “On the one hand, this seems like a very legitimate way of deciding policies: Let the people vote. On the other hand, there is a serious cost for minorities if the majority decides based on preferences which might be discriminatory.”
To pursue the issue further, Hainmueller and Hangartner have conducted a follow-up study of immigration applications in Switzerland after 2003, when power to approve naturalization petitions was handed back to local legislators; their initial findings suggest that elected representatives are significantly more likely to approve citizenship applications.