“To avoid a Manchester-type bombing on American soil, integrate Muslims”,1 “We must look at Muslim immigration with clear eyes”,2 and “Accept Islamic Terror as the New Normal?”3 are just three of many recent headlines that relate the threat of terrorist attacks to Muslim migrants or immigration in general. According to the National Review (2017): “[i]n Hungary, Poland, and the Czech Republic which have strict immigration laws and few Muslims, there have been no major terror attacks. In Germany, Belgium, France, and England, which have liberal immigration laws and large Muslim populations, terror attacks have become an almost weekly occurrence.”
President Trump seems to share the widespread concern that foreigners – and in particular, immigrants from Muslim-majority countries – are more likely to commit terrorist attacks. His recent ban on travel to the US is an attempt to prevent people from six countries with predominantly Muslim population from entering the country.
The US travel ban is not a unique response. Plenty of evidence suggests that stricter immigration and visa policies are a preferred reaction to terrorist attacks (Martin and Martin 2004, Avdan 2014). On the other hand, evidence is surprisingly scarce that either liberal immigration and integration policies or having more foreigners living in a country will foster terrorism. The only systematic statistical analysis we are aware of reports a negative correlation between migration and terrorist attacks (Bove and Böhmelt 2016). Previous studies that address the link between terror and migration either examine the effect of terror on migration (Dreher et al. 2011) or employ data on terrorists with immigration status, rather than relying on systematic cross-country time-series data on migration and terrorist attacks (Kephart 2005, Leiken and Brooke 2006).
Studies focusing on terrorists with immigrant backgrounds find a close link between immigration and terrorism. Given that they do not examine overall flows of immigration, but only those cases in which immigrants have been involved in terrorist activity, these studies are not reliable indicators of how overall migration and terrorist attacks relate (Spencer 2007).
The number of terrorist attacks
In our recent work (Dreher et al. 2017) we provide a detailed analysis of how the number of foreigners living in a country has affected the number of terrorist attacks made by foreigners on citizens of their host countries. According to the raw data, in OECD countries between 1980 and 2010, for every million foreigners in the population, 0.8 terror attacks are committed per year, per country (there were 662 transnational attacks). While it is obvious that the number of attacks increases with the number of people living in a country (after all, with no foreigners in a country, no foreigners would commit any attacks), on average these numbers amount to about one attack by foreigners per year and host country, and 1.3 people die from these attacks in the average country and year.
Transnational terror is dwarfed in absolute numbers by the number of attacks made by the domestic population. In the 20 OECD countries that our sample covers, there were 2,740 attacks arising from the domestic population. In relative terms though, the picture is different – there were fewer than 0.18 terrorist attacks for every one million locally born citizens in a typical country and year. Overall, while the probability that foreigners are involved in an attack on the domestic population was much higher than the risk that citizens were involved in attacks on their own country, the risk associated with each additional foreigner was tiny.
In our statistical analysis, we investigate whether, and to what extent, an increase in the foreign population of the average OECD country would increase the risk of terrorist attacks from foreigners in a host country. We identify exogenous variation in the number of foreigners living in an OECD country using changes in migration resulting from natural disasters. These changes affected host countries differently, according to the specifics of each host- and origin-country pair.
Using data for 20 OECD host countries, and 187 countries of origin between 1980 and 2010, we find that the number of terror attacks increased with the number of foreigners living in a host country. This scale effect that relates larger numbers of foreigners to more attacks does not imply, however, that foreigners are more likely to become terrorists than the domestic population. When we calculate the effect of a larger local population on the frequency of terror attacks by locals, the effect is of a comparable size. We conclude that, in this period, migrants were not more likely to become terrorists than the locals of the country in which they were living.
To put these results in perspective, consider the expected effect of a decrease in the domestic population of 0.0002% (which is the average decrease in the domestic population of the 20 OECD countries we studied in 2015, according to the OECD). According to our model, this would have reduced the number of terrorist attacks by 0.00025 per country and year. The increase in the stock of foreigners living in these countries was 3.6% in the same year. According to our estimates, this would have created 0.04 additional attacks. We might argue that this hardly justifies a ban on foreigners as a group.
We find little evidence that terror had been systematically imported from countries with large Muslim populations. The exceptions were Algeria and Iran, where we found a statistically higher risk of being involved in terrorist attacks against the local population, compared to the average effect of foreigners from non-Muslim countries. In this light, the phrases ‘Muslim terror’ or ‘Islamist terror’ does not seem accurate or useful. Only 6% of the terrorist attacks in the US between 1980 and 2005 period were carried out by Muslims, and less than 2% of all attacks in Europe had a religious motivation between 2009 and 2013 (Alnatour 2017).
Immigration and integration
Contrary to the expectations of many politicians and pundits, introducing strict laws that regulate the integration and rights of migrants does not seem to have been effective in preventing terror attacks from foreign-born residents. We rather find that repressing migrants already living in the country with these laws has alienated a substantial share of this population, which increases the risk of terror. Stricter laws on immigration thus have the potential to increase the risk of terror, at least immediately following the ban.
Of course, our results are preliminary rather than definitive. We do not have data detailing which specific group of migrants from which country were involved in terrorist attacks, and so we rely on gross data on the number of terrorist attacks by all foreigners. This means we can estimate the risk of terror associated with a larger number of migrants, but we cannot test whether migrants from a particular country were engaged in terrorist events. This analysis would require individual-level data than currently available, for a large sample of countries and years.
Also, our sample ends in 2010, and so it does not cover recent increases in foreign populations after the refugee crisis. Our focus on numbers of migrants living in a country rather than inflows lessens this external validity concern. Focusing on total numbers, however, gives little weight to the most recent immigrants, who might be different compared to those of previous years.
Our results illustrate an important trade-off. While stricter immigration laws could reduce the inflow of (violent) foreigners and thus potentially the number of future terrorist attacks, the restrictions would also increase the probability that those foreigners already living in the country become more violent. Immigration bans, like those recently introduced in the US, would arguably increase the short-term risk of attacks, before potentially reducing risk when the number of foreigners in the population has decreased.
Costs and benefits of immigration
Driving fast on motorways leads to accidents and fatalities, planes crash and people die, and more people living in cities leads to a larger number of murder cases. Yet few people favour strict bans on motorways and planes, or cities. Any analysis focusing exclusively on (terrorist-related) costs of immigration would ignore a broader calculation of costs and benefits. A larger number of people leads to a higher risk that some of them become terrorists. This holds for local and foreign populations alike, and so hardly qualifies as reason to ban migration. Rather, the increased risk of terror has to be confronted by the many other – positive and negative – effects that come with immigration. While our results do not include these broader calculations, they clearly show that bans on Muslim immigration would be likely to increase the risk of terror, rather than increasing the safety of the domestic population.
Alnatour, O (2017), “Muslims Are Not Terrorists: A Factual Look at Terrorism and Islam”, Huffington Post.
Avdan, N (2014), “Controlling Access to Territory: Economic Interdependence, Transnational Terrorism, and Visa Policies”, Journal of Conflict Resolution 58(4): 592-624.
Bove, V and T Böhmelt (2016), “Does Immigration Induce Terrorism?”, Journal of Politics 78(2): 572-588.
Dreher, A, P Schaudt and M Gassebner (2017), “The Effect of Migration on Terror – Made at Home or Imported from Abroad?”, CEPR Discussion Paper No. 12062.
Dreher, A, D Meierrieks, and T Krieger (2011), “Hit and (They Will) Run: The Impact of Terrorism on Migration”, Economics Letters 113(1): 42-46.
Kephart, J L (2005), Immigration and Terrorism – Moving Beyond the 9/11 Staff Report on Terrorist Travel, Washington: Center for Immigration Studies.
Leiken, R S and S Brooke (2006), “The Quantitative Analysis of Terrorism and Immigration: An Initial Exploration”, Terrorism and Political Violence 18(4): 503-521.
National Review (2017), “Terrorism is not random”, 28 May.
Spencer, A (2007), “Using Immigration Policies as a Tool in the War on Terror”, Crossroads 7(1): 17-53.