This paper investigates the impact of the 2015/16 large-scale refugee arrivals on social cohesion in Germany. During this period, over one million asylum seekers from Syria, Afghanistan, and the Western Balkans arrived in the country, doubling the population of persons seeking protection from one to two million. Initially, the German public broadly supported the influx; however, public opinion shifted dramatically following incidents such as the New Year’s Eve 2015/16 assaults in Cologne, where some perpetrators were identified as recently arrived asylum seekers.
To estimate the causal effect of refugees on social cohesion, the authors exploit the assignment mechanism of asylum seekers in Germany, which was based on the federal states’ population and tax revenues from two years prior. This assignment was unrelated to the contemporary economic, political, or social conditions. The study employs a difference-in-differences approach, comparing changes in attitudes over time between areas with high and low inflows of asylum seekers. The key assumption is that, in the absence of the inflow, the outcomes in both high and low inflow areas would have followed parallel trends.
The primary data source for social attitudes is the German Socio-Economic Panel (SOEP), which provides individual-level panel data with geo-identifiers for respondents’ counties of residence. This data is linked to county-level asylum seeker inflow data from the Central Register of Foreigners, detailing the numbers of asylum seekers in each county. Additionally, the study utilizes a geo-referenced event dataset on anti-immigrant violence from the Amadeu Antonio Foundation, documenting incidents such as assaults, attacks on refugee housing, and arson. Social cohesion is measured across several dimensions, including generalized trust, perceived fairness, attitudes towards foreigners, and anti-immigrant violence.
Main findings:
- There is no evidence that refugee inflows affected self-reported measures of social cohesion in the short run. There is no evidence that inflows of refugees affected social attitudes and perceptions such as trust, perceived fairness, and perceived helpfulness of other people. The only statistically significant effect was a small decrease (around 4 percent of the mean) in charitable donations. These null findings are notable given the substantial shift in public opinion observed during this period.
- There is some evidence that the inflow had negative effects on social cohesion in areas where people are generally more concerned about immigration. In areas with above-median employment rates, the inflow had a small positive effect on trust and perceived fairness. In areas with low employment rates, all effects were small and statistically insignificant. In areas with high vote shares for the far-right AfD party, the inflow led to stronger anti-immigrant sentiment, greater concern about crime, and a lower likelihood of donations to charity.
- The local presence of refugees increased the incidence of anti-immigrant violence. Counties receiving larger numbers of refugees saw a disproportionate increase in anti-immigrant violence. A doubling of the local asylum seeker population increased the incidence of anti-immigrant violence by 0.05 incidents per 100,000 inhabitants per month, a 166 percent increase relative to the pre-2015 average. This increase was mainly driven by attacks on asylum accommodation and was concentrated among the top 10 percent of municipalities with the largest inflows of asylum seekers. The effect was stronger in areas with high unemployment and a higher share of AfD voters. This effect on violence was temporary, subsiding after about two years.
Overall, the findings suggest that while the general population does not significantly react to the presence of asylum seekers, a small segment of the population exhibits extreme reactions, manifesting as anti-immigrant violence. The authors conclude that high-income countries can absorb large numbers of refugees without major impacts on social cohesion. However, to mitigate the negative reactions observed in certain areas, policies that foster inclusiveness and empathy towards immigrants should be implemented, particularly in regions prone to anti-immigrant violence. Based on the relevant literature, the authors recommend perspective-taking interventions, such as priming family immigration stories among natives, to increase support for refugees. Additionally, promoting narratives that highlight the hardships faced by refugees can positively influence intergroup relations and policy attitudes.