A new study by Dominic Rohner, Professor at HEC Lausanne (University of Lausanne), and co-authors*, published in Science, shows that top-US-aid-receiving African regions experienced a substantial increase in violence compared to non-recipients following the USAID shutdown last year.
When Donald Trump moved to shut down the United States Agency for International Development (USAID), ending one of the world’s largest aid programmes, experts warned that the move would harm fragile recipient nations, while supporters said the impact would be limited. While debate continues over whether foreign aid reduces or fuels conflict, the real-world impact of a sudden, large-scale withdrawal has remained largely untested. Until now.
Over a year after the USAID shutdown, a new study by a team of international academics*, published in Science on May 14, addresses this issue. Dominic Rohner and his co-authors conducted an empirical analysis of the withdrawal’s impact on subnational conflict across Africa—one of the main recipient regions—testing whether the abrupt dismantlement of USAID has led to increased conflict in those areas.
What the data shows
Using geocoded data on US aid disbursements merged with detailed records of violent events from the Armed Conflict Location and Event Data (ACLED), the researchers compared conflict outcomes in a sample of 870 African subnational regions from March 2024 to November 2025 (10 months before and after the shutdown), as well as across regions with varying levels of historical US aid exposure (disbursements in 2017-2020).
They found that regions that had received more US aid per capita experienced relatively more conflict following the USAID withdrawal. In regions among the top 25% US aid recipients (above $251 per 1,000 people), the shutdown was associated with a higher probability of any conflict events: the probability of protests and riots surged by 10% higher relative to regions with no US aid, the number of conflict events increased by 10.6%, battle counts by 6.9%, and battle-related fatalities by 9.3%, compared to areas not receiving US aid.
The event-study analysis showed no pre-existing differences in conflict trends between high- and low-US-aid-exposure regions before the shutdown. Whereas protests and riots responded immediately to aid cuts, the effects on battles and violence against citizens intensified over time. A key reason is that the cost of violence falls faster than the value of what groups might fight over, so people have less to lose by resorting to violence.
“We found that the effects of the aid withdrawal were substantially mitigated in countries with stronger, more open political systems. This highlights the vulnerability of regions with weak governance to humanitarian and economic shocks. Our findings stress the need to carefully design and time changes to international aid,” comments Professor Rohner.
*The authors :
Dominic Rohner, HEC Lausanne (University of Lausanne), Geneva Graduate Institute and CEPR
Uwe Sunde, University of Munich, CEPR
Oliver Vanden Eynde, CEPR, Paris School of Economics, CNRS
Austin L. Wright, The University of Chicago
Jing-Rong Zeng, Paris School of Economics
Reference : Rohner, D., Sunde, U., Vanden Eynde, O., Wright, A.L. and Zeng, J.-R., Aiding Peace or Conflict? The impact of USAID cuts on violence
