Bottomline: Abrupt aid withdrawals affect conflict not only by reducing resources but also by disrupting commitments that previously constrained violent behavior. This has implications for understanding vulnerability to donor-side shocks in aid-dependent settings.
[7/7]

Institutional capacity matters: stronger states experience weaker or no conflict response, while political corruption amplifies these effects.
[6/7]

Two mechanisms:
1. The announced aid withdrawal created a finite-horizon environment, undermining existing arrangements between state and non-state actors.
2. The termination of aid-funded programs reduced local employment, lowering the opportunity cost of conflict.
[5/7]

Main finding: Armed conflict between organized actors increased by 15% in high-aid countries during the first 3 months after the shutdown. Violence against civilians perpetrated by militias increased by up to 10% a bit later. No substantial changes in protests or riots.
[4/7]

Countries differed substantially in their pre-shutdown dependence on USAID, generating differential exposure to a common and plausibly exogenous aid shock. We apply event-study and difference-in-differences designs to monthly data from July 2024 to June 2025.
[3/7]

In February 2025, USAID disbursements fell by roughly 90% across Sub-Saharan Africa relative to the same month a year earlier. The shock was driven by U.S. domestic politics rather than recipient-country conditions, creating a rare quasi-experimental setting.
[2/7]

Overall, our results offer new evidence on how local maize prices affect social unrest and explain why the same price change can have diverging effects. [9/9]

Using the same Afrobarometer survey data, we show that these effects are tied to changes in perceived well-being—especially relative well-being—in the wake of price shocks. [8/9]

We check suggested mechanisms with geo-located data from Afrobarometer surveys.
– Higher maize prices amplify concerns about political instability and ethnic tensions only where there is little maize agriculture or high production inequality.
– In maize-producing regions, the effect reverses. [7/9]

We then examine mechanisms.
– At or immediately after harvests, protests weaken—plausibly due to higher opportunity costs.
– Following (presumably) poor harvests, riots subside—likely due to less resentment and, possibly also, less to fight over. [6/9]

Some numbers:
– In locations with no maize agriculture, a 10% increase in prices reduces unrest (protests) by 7.5%;
– In areas with average maize cropland (relative to none), riots fall by 5.5%;
– In areas with high ethnic inequality in maize production, riots rise by 6.6%. [5/9]

To test this, we develop an index capturing maize production inequality across historical ethnic boundaries. Price increases raise unrest—especially riots—in areas with high inequality, but reduce it in more homogeneous areas. [4/9]

Grievances can be understood through the lens of relative deprivation: people feel worse when others around them seem better off. Commodity price changes create precisely such distributional effects—benefiting net producers and hurting consumers. [3/9]

Our baseline finding may appear to contradict the common perception but is hardly counterintuitive. Farmers’ grievances are typically linked to plummeting prices: e.g., “The government has let us down. The market price ... is very low …” (shorturl.at/NFNpG). [2/9]

In that order? That's bad editorship :)