Country level

Our country-month (cm) forecasts are presented in the maps above. Our models yield results in line with mainstream studies of conflict at the country level. For instance, we forecast a higher risk of state-based conflict in countries with large populations, in non-democracies and countries with recent regime change, with low or negative growth rates, and with low education levels or other indicators of low socio-economic development.

We continue to forecast a high probability of state-based conflict in countries that have a recent history of conflict or protest events. Particularly in Egypt, Mali, Burkina Faso, Nigeria, Cameroon, Sudan, DR Congo, Somalia and Mozambique, the risk of at least one state-based conflict event is high and over 50%.

Compared to last month's forecast, the greatest increase of the risk of a state-based conflict is in South Africa, where in the early weeks of September organized crowds committed xenophobic violence in attacks on migrant-owned businesses in and around Johannesburg. The unrest led to around 20 dead, including one killed in clashes with security forces trying to disperse a mob on 8 September. A relevant increase of risk furthermore occurred in South Sudan, where 16 government soldiers where killed in an ambush perpetrated by the National Salvation Front (NAS) in Yei River State on 6 September. Sudan also shows a slight increase in the risk of state-based violence, given an apparent SLM/A attack on the government in Gereida on 18 September, in which one army lieutenant, two civilians, and five assailants were killed. Conversely, the risk of state-based violence significantly decreased for Niger, where in September no cross-border events near Mali and Burkina Faso occurred.

The forecast maps for non-state conflict follow partly the same patterns as state-based conflict, but the patterns of past events do differ across conflict types. Libya, Mali, Burkina Faso, Nigeria, DR Congo, Kenya, and Somalia remain at particularly high risk of non-state violence this month.

Compared to last month's forecast, the risk of non-state conflict has increased in South Africa in particular, given the events of xenophobic violence described above. We also a find a relevant rise in the risk of non-state violence in Burkina Faso this month, where violence between JNIM and the Dozo community led to two deaths in September. In Central African Republic, clashes in Birao between Patriotic Front for the Renaissance in the Central African Republic (FPRC) and Movement of Central African Liberators for Justice (MLCJ) led to 24 dead on the first of September, and then again 39 dead on 14 September. This is despite the rival groups being part of a peace deal signed earlier in February. The risk of non-state violence appears to have reduced this month for Cameroon, Chad, and Sudan particularly, given that no (clear) non-state conflict events were recorded in the UCDP Candidate Events Data for September 2019.

The probability of one-sided violence events remains especially pronounced this month in Mali and Burkina Faso, Nigeria (predominantly given Boko Haram/IS), DR Congo, Sudan, and Somalia (predominantly given Al-Shabaab). Mozambique continues to be at especially high risk of one-sided violence as well, given persistent civilian killings by Islamist militants in the country's Cabo Delgado Province.

Compared to our October forecast the model ensemble responds strongly to the case of Namibia in particular, where on 5 September a civilian was killed during a joint Police and Defense anti-crime operation in Windhoek. South Africa again shows an elevated risk of one-sided violence, too, given the xenophobic violence in early September. Algeria shows a slightly elevated probability of one-sided violence, given the killing of two civilians in Relizane by police responding to a protest that was spiked by the police having run over and killed a teenager a day before on 18 September. Conversely, Kenya and Niger show the strongest reduction, given that no (clear) one-sided violence event has been recorded by the UCDP there.

For more information see, see the monthly forecasts report for November 2019

Download maps

Please cite:  Hegre, Håvard, Marie Allansson, Matthias Basedau, Michael Colaresi, Mihai Croicu, Hanne Fjelde, Frederick Hoyles, Lisa Hultman, Stina Högbladh, Naima Mouhleb, Sayeed Auwn Muhammad, Desiree Nilsson, Håvard Mokleiv Nygård, Gudlaug Olafsdottir, Kristina Petrova, David Randahl, Espen Geelmuyden Rød, Nina von Uexkull, Jonas Vestby (2019) ‘ViEWS: A political violence early-warning system’, Journal of Peace Research, 56(2), pp. 155–174. doi: 10.1177/0022343319823860.