Violence and the Perception of Risk Associated with Hosting Refugees

lex Braithwaite, Tiffany S. Chu, Justin Curtis, Faten Ghosn

Public Choice, Issue 178 (2019), Pages, 473-492


This paper examines whether individuals’ experiences of political violence affect their perceptions regarding the risk associated with hosting refugees. The authors focus on recent exposure to violence within Lebanon, which hosts more than one million Syrian refugees. The analysis is based on a nationally representative survey of 2,400 Lebanese residents, administered between June and August 2017. The authors compare attitudes of individuals interviewed before and after two key violent events: a June 30 suicide attacks by Syrian militants on the Lebanese military in Arsal town, close to the Syrian border; and a July 21 offensive Lebanese Hezbollah to remove militants from refugee camps in and around Arsal, in coordination with the regular armed forces of Lebanon. The authors match respondents on the basis of social, economic and demographic characteristics, as well as proximity to the site of violent events.

Main results:

  • Following the June 30 attack by Syrian militants, Lebanese citizens are substantially more likely to consider both Syrian refugees and militants to be a security threat for their family, community, and country. Respondents interviewed up to 15 days after the attack are more likely to believe that refugees and Syrian rebels are a security problem. The effect diminishes both substantively and statistically as the time window for treatment is extended to 30 days and then to the widest time window (to end of August 2017).
  • After the July 21 Hezbollah-led offensive, respondents are substantially less likely to consider refugees and militants to be security risks than those who responded to the survey prior to that event. The results show no meaningful differences between individuals surveyed 15 days before and after July 21. Using a 30-day time window, respondents interviewed after July 21 report feelings that refugees and Syrian rebels are less of a security problem, suggesting that it takes time for respondents to believe that the security-enhancing effect of violence perpetrated by Lebanese forces to establish that militants have successfully been rooted-out from the refugee camp.
  • Other socioeconomic and demographic factors do not appear to play consistent roles in threat perceptions. Lebanese tended to react to violence in ways not affected by religiosity or socioeconomic situations. This suggests that events on the ground, rather than characteristics such as religious identity, income, or education, may have intervened to change attitudes.

The authors conclude that, rather than seeing refugees as fellow victims of violence and considering them deserving of support, local residents find it difficult to distinguish Syrian civilians from Syrian militants. This is despite the fact that Syrian refugees in Lebanon are escaping the very same violent groups that increasingly are targeting Lebanese civilians and property. In contrast, force used against militants can have the effect of ameliorating hardened attitudes, which suggests that local residents trust armed forces to provide adequate security.