...this study analyses whether climatic changes between 1963 and 2014 impacted the risk of conflict and displacement of people in East Africa... found that climate variations as recorded by the Palmer Drought Severity Index (PDSI) and the global temperature record did not significantly impact the level of regional conflict or the number of total displaced people (TDP). The major driving forces on the level of conflict were population growth, economic growth and the relative stability of the political regimes... Instead, we suggest rapid population growth, low or falling economic growth and political instability during the post-colonial transition were the more important controls...I've elided a bit there that I'll come back to, but I'm showing the important bit. This via an article on it at the Conversation. I find this unsurprising, because I said much the same in Infertile Crescent: these people's major problem is that their government is crap. As O+M point out, Halvard Buhaug reached much the same conclusions in 2015, and nicely notes that journals and journalists alike appear especially attracted to sensational findings and tabloid conclusions, with the result that researchers are pushed to oversell their findings. Oh, and also it does not follow that a long‐term shift in normal conditions (e.g., a 2°C warmer world) will have the same impact on social systems as a short‐term anomaly of a similar magnitude (2°C above monthly mean). Unfortunately, researchers are often not clear on the distinction between climate variability and climate change, and findings of behavior related to the former are often used as foundation for projecting impacts of the latter. And if you're interested in climate change, and people's vulnerability to it, then In unstable corners of the world, ending violent conflict may be the most efficient and cost‐effective way to improve social resilience to climate change. I seem to have got a bit distracted from O+M, sorry.
Back at O+M: I've just skimmed it. It isn't desperately exciting, I'd say. Their method - take a bunch of things and correlate them against another bunch of things - doesn't seem very powerful, and since there are piles of correlations floating around, some are likely to be significant by chance. The one thing they find is The displacement of people, particularly across international boundaries as refugees, in this region is also shown to be in part driven and exacerbated by intense droughts which can be in part linked to the long-term drying trend that has been ascribed to climate change, which is what I elided from the blockquote, because it's rather minor.
So, there we go. What I thought I already knew is reinforced. If you'd like to be told otherwise, then maybe try Jesse Anttila-Hughes, in particular Different perspectives on climate and conflict.
And to avoid the obvious misunderstanding, neither I nor O+M are suggesting no role for climate change. Just that other things are significantly more important. And also, I'd argue, more fixable. At least in principle. In practice, less so.
* FirstEnergy Seeks Bankruptcy Protection for Ailing Coal and Nuclear Subsidiaries.