Newly displaced Syrian children arrive at a refugee camp in the village of Atimah, in Syria’s Idlib province, on Sept. 11. (Khalil Ashawi/Reuters)

Violent attacks by local populations against refugee populations are depressingly common across the globe. In August, demonstrators in the Brazilian border town of Pacaraima attacked refugees from neighboring Venezuela. The rioters destroyed a provisional refugee camp and forced an estimated 1,200 Venezuelans to flee back across the border.

Meanwhile, more than 1.5 million Venezuelans have been displaced to neighboring countries by the worsening crisis, and similar reports of violence by locals against the refugees abound.

And Syrian refugees have suffered extensive harassment in Lebanon, with deliberate efforts to clear refugees from specific areas, while many Rohingya from Myanmar have been targeted in India.

Such attacks do not constitute armed civil conflicts in a conventional sense, as local nonstate actors target other nonstate actors rather than mobilizing against the state, and the state is not a direct actor in or target of the violence. In many cases, host country authorities have also actively tried, but failed, to prevent violence against refugees.

Many have emphasized how refugee flows can exacerbate the risk of civil war in hosting states, but our forthcoming article draws attention to how refugees are more likely to give rise to other forms of nonstate violence. The ability and willingness of states to exercise authority and control is key to preventing attacks against refugees.

How refugees spark conflict

The literature linking refugees and war can be limiting because it neglects the many forms of violence other than war that can afflict refugee communities. An influx of refugees often generates controversy in hosting countries, often at a hyperlocal level. Existing research points to a number of ways in which refugees can exacerbate the risk of domestic conflict. In particular, we may have increased real or perceived competition between refugees and local populations. Existing survey evidence also confirms that resentment often is widespread among local populations in hosting countries. This creates a fertile political atmosphere for violence.

Perceived competition and resentment are particularly likely to induce violence by locals directed against refugees rather than challenges against the government. We find support for this in an analysis of data on nonstate conflict from the Uppsala Conflict Data Program and the number of refugees hosted from the United Nations High Commissioner for Refugees. More specifically, we find that there is a higher risk of nonstate conflict in countries hosting large refugee populations, and the effect is particularly pronounced in countries that have the least capacity to host large numbers of refugees and mitigate the tension that may arise with local populations.

Violence in the other direction is less common. In the absence of clear ties to actors and ongoing conflicts, we are unlikely to see refugees contribute to violent mobilization against host governments. There certainly have been instances in which refugee communities were active in civil wars in hosting states such as the Great Lakes Region in Africa in the 1990s. But few would expect Venezuelan refugees to make common cause with existing Marxist rebel groups such as the National Liberation Army (ELN) in Colombia, or for Syrian refugees to turn their guns on the Turkish government.

The unequal refugee burden of weaker states

The state can play an important role in modifying the risk of violence against refugee populations. On one hand, states can help to better accommodate the needs of refugees and address the impact of refugee flows on local communities to decrease perceived conflict. On the other hand, states can also provide security, both in the sense of monitoring refugee camps and taking measures to protect refugee populations from violence by locals.

However, whether states can accommodate or protect refugees will be limited by the resources at their disposal. For example, although the Syrian refugee crisis has led to expressions of xenophobia and extremism in Germany, the authorities have enacted strong measures to monitor people involved in hate speech on social media and prevent physical attacks. By contrast, in India and Lebanon, the authorities have generally failed to prevent attacks against refugees despite often actively trying to do so.

Where the risk is greatest

Nonstate local violence against refugees is a serious security issue and cause for concern. However, the implications run completely counter to the state security rationale frequently cited in efforts to restrict access in the United States and many European countries after the refugee crisis from the civil war in Syria.

The risk of nonstate-actor conflict is actually low in hosting states that are highly capable. The real risk is to lower-capacity countries, which have been left to host 84 percent of the world’s refugees. If fleeing and fighting are potential substitutes, then efforts to crack down on refugee entry or asylum institutions in more capable states risks exacerbating conflicts elsewhere, both in their devastating impact in the countries of origin and in potentially worse consequences for low-capacity neighboring states.

Although many states hosting refugees lack the capacity to implement measures that can mitigate the risk of violence, external resources could help to compensate. The scale of the current refugee crisis may create many additional demands and strains, but failing to implement comprehensive policies to accommodate refugees can be far costlier in the long run.

Tobias Böhmelt is a professor of government at the University of Essex.

Vincenzo Bove is a reader of politics and international studies at the University of Warwick.

Kristian Skrede Gleditsch is regius professor of political science at the University of Essex and a research associate at the Peace Research Institute Oslo.