This paper examines whether big land transfers to investors in Africa cause local unrest. The author compares sites where large-scale land acquisitions (LSLAs) were actually implemented to similar sites where deals failed. Using 1,391 geocoded deals across 38 African countries from 1997 to 2025, the study finds that implemented deals produce a sustained increase in civic unrest — about a 158% rise relative to the pre-deal average.

To get at cause and effect, the researcher uses a control group of exogenously failed deals and a statistical method called a staggered difference‑in‑differences estimator. The conflict data come from the Armed Conflict Location & Event Data Project (ACLED) and the deal information from the Land Matrix. On average, implemented deals raise protest and riot events by about 1.48 events per location per year versus a pre-treatment mean of 0.94 events. The paper reports several robustness checks, including sensitivity bounds and the exclusion of failed deals that showed signs of pre-existing community resistance.

The increase in protest is not uniform. It is strongest when domestic investors acquire community or state land to grow food crops. Deals led by transnational investors and projects for biofuels or other non-food crops generate smaller and less persistent protest responses. The authors interpret this pattern as evidence of local dispossession and “elite capture” — that is, politically connected domestic actors using state authority and unclear land records to appropriate customary land without community consent.

The study brings in three additional strands of evidence that fit this mechanism. A panel built from six rounds of Afrobarometer survey data shows reduced trust in traditional leaders and less contact with them in exposed communities. Media analysis using the GDELT Global Knowledge Graph finds a rise in local coverage about property rights, corruption, and agriculture near implemented deals. Finally, constituency-level election data from the Comparative Legislative Elections Archive (CLEA) show higher opposition vote shares and greater turnout near implemented deals over the next four to five years.