What makes people welcome or reject refugees? What research in Germany reveals

Across the EU, immigration is one of the most divisive topics in politics today. Germany, a country once known for its “Willkommenskultur” (welcome culture), is a case in point.

The German chancellor, Friedrich Merz, has recently said that Syrians no longer have “grounds for asylum in Germany”, and that they will be encouraged to return voluntarily. Some could also be deported in the near future.

Polling suggests that tough approaches to immigration resonate well with the public, reflecting a broader shift toward more negative immigration attitudes.

What determines whether people in a host country like Germany welcome or reject refugees? This is what my colleagues Bruno Martorano, Laura Metzger and Melissa Siegel and I sought to better understand in a recent paper.

Through a survey experiment, we tested how different factors would affect whether participants express concern for refugees’ wellbeing, or consider them a threat.

The study was designed to assess the effects of different factors on people’s attitudes towards refugees. For example, whether a participant held humanitarian values (is committed to help fellow humans in need). We measured this based on their responses to a frequently-used set of four questions. Humanitarian considerations have received little attention in earlier studies in this area.

We also measured if people’s views changed depending on the amount of adversity refugees faced (such as whether they were fleeing war), and the personal characteristics of the refugees – their age and gender, and whether they were part of families.

We surveyed more than 2,000 participants in 2023, using short, professionally-produced videos about Syrian refugees in Turkish refugee camps.

Some participants watched a control video, which provided only some background information. Others watched one of four videos: two emphasised the humanitarian situation of Syrian refugees in refugee camps in Turkey, the other two stressed challenges that the immigration of these refugees may imply for German society.

One of the “humanitarian message” videos and one of the “threat message” videos focused on families with small children among the refugees. The other two highlighted young refugee men.

After they watched the videos, we surveyed respondents about their views and concerns about the refugees.

Humanitarian compassion

On average, respondents overall showed a moderate level of concern for the wellbeing of Syrian refugees. They were somewhat more worried about the impact on Germany’s welfare system, security and cultural life. Fears that refugees might take away jobs were less common.

We identified a strong correlation between how humanitarian someone generally is, and the compassion that respondent expressed toward the refugees. We also found that exposure to short videos highlighting the humanitarian plight of refugees made participants care significantly more for most aspects of refugees’ wellbeing, compared to those who only saw the control video.

Additionally, we gave participants the possibility to sign pro-refugee policy petitions within the survey. Only a quarter of respondents overall signed a petition calling for increased funding for Syrian refugees abroad. An even smaller share supported a petition for more admissions of Syrians to Germany. But highlighting the humanitarian plight of refugees largely increased the share of respondents advocating for more support for refugees abroad.

The limits of this kind of messaging were also apparent. Watching the humanitarian videos did nothing to reduce immigration-related fears, nor did it increase acceptance for allowing refugees into the country.

Scepticism of (some) refugees

Those who watched videos of young male refugees were significantly less likely to support allowing more refugees into the country. Our data suggests that this is likely due to heightened concerns about negative cultural effects among those who viewed a video featuring young refugee men, rather than economic concerns or participants feeling more physically threatened.

Those who watched the videos highlighting families were more concerned about the refugees’ safety. Yet, they also expressed greater concerns that refugees may represent a burden for the welfare state.

The videos did not impact all respondents equally. For example, among respondents who identified as politically leftwing, seeing a video with a humanitarian message was associated with fewer cultural concerns about immigration, compared to the control group. For right-leaning respondents, we observed the opposite: seeing one of the humanitarian videos was associated with more concerns.

In addition, it was remarkable how differently east and west Germans reacted to our experiment. The political legacy of eastern Germany – the region which used to be the socialist authoritarian German Democratic Republic (GDR) until 1990 – is relevant in explaining persistent differences between the eastern and western German populations. It has been well established that east and west Germans differ in their values, preferences and voting behaviour, including support for the anti-immigration party AfD.

While similar at baseline, we found that exposure to the four videos affected the views of east Germans more negatively than those of west Germans, regardless of the exact message or the group of refugees the video highlighted. For example, focusing on refugee families largely boosted the share of west Germans who supported increasing support for refugees abroad. Among east Germans though, it had if anything the reverse effect.

It was remarkable how different these populations reacted to the very same message. Their reactions diverge more strongly than across any other divide, such as age or education.

Taken together, our results suggest that people’s opinions on immigration are more complex than a simple pro- v anti-immigration split. Whether a political message is effective or not – that is, whether it changes minds – depends on the framing of the message itself, as well as the views and values held by the people receiving that message.

Tobias Hillenbrand does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.