Local Deprivation, Inequality, and Resource Wealth Predict Conflict in Sub‑Saharan Africa

Why This Question Matters

Civil wars often erupt in particular regions rather than across whole countries, yet large-N cross-country studies frequently miss these within-country patterns. Gudrun Østby, Ragnhild Nordås, and Jan Ketil Rød ask whether subnational inequalities and local deprivation help explain where civil conflict begins in Sub‑Saharan Africa—and whether these local conditions interact with the presence of natural resources to raise risk.

What the Authors Did

The authors combine Demographic and Health Surveys (DHS) with GIS operations to build new, disaggregated measures of welfare and socioeconomic inequality for subnational regions in 22 Sub‑Saharan African countries. They link these regional welfare measures to geographically coded conflict onset data covering 1986–2004 to test whether local conditions predict the start of violent conflict.

How the Analysis Works

• Data sources: DHS household and education indicators, GIS-based regional aggregation, and mapped conflict locations (1986–2004).
• Unit of analysis: subnational regions within 22 countries, allowing comparison of between‑ and within‑region inequalities.
• Approach: spatially disaggregated statistical tests of whether regional education, assets, and intraregional inequality predict conflict onsets, and whether resource presence amplifies those relationships.

Key Findings

• Regions with lower average education are more likely to experience conflict onset.
• Strong relative deprivation in household assets (regions worse off relative to their country) increases conflict risk.
• High intraregional inequality is associated with greater likelihood of conflict starting in a region.
• The combination of natural resources and local relative deprivation substantially raises the probability of conflict onset in a region.

What This Means for Theory and Policy

The results highlight the value of looking below the national level: subnational patterns of education, asset deprivation, and intra‑regional inequality matter for where violence starts. These findings lend weight to theories of relative deprivation and grievance as drivers of conflict and show that local resource wealth can exacerbate tensions when paired with deprivation. For policymakers, the study implies that addressing regional disparities in education and asset access—and managing resource-related tensions—may reduce the local conditions that precede civil violence.

Research Transparency

The paper documents a novel data construction strategy (DHS + GIS) for measuring local welfare differences and links those measures to spatial conflict data, providing a replicable path for future work on subnational drivers of conflict.