Revisiting the Resource Curse based on Geocoded Events Data

• Research unit: COST
• Project number: C12.0087

Key words (English)

Civil War; Natural Resources; Events-Data; GIS; Iraq; Africa

Research programs (English)

COST-Action IS1107 - European Network for Conflict Research (ENCoRe)

Short description (English)

There is mounting evidence for the claim that natural resource abundance increases the risk of civil war. Yet, we still know little about how the presence of high-value natural resources in contested territory affects the actual behavior and strategies of warring parties. The proposed research addresses this gap by analyzing newly available disaggregated, high-volume geo-referenced events data to uncover the particular strategies of violence in resource-abundant regions. This project will build on our network of COST partners to supply data on conflict events and natural resources, and provide a key contribution to the COST Action by integrating these various data sources into our in-house, open-access data platform GROWup. Building on these data integration efforts, we will rely on Geographic Information Systems (GIS), statistical analysis, and qualitatively informed reasoning to analyze patterns of violence in two resource-rich regions, Iraq and Africa. This project will benefit the scientific community by offering existing and new data on conflict events and natural resources through a consolidated platform, close an important gap in the research on the effects of natural resources on conflict, and provide valuable policy advice for sustainable resource extraction.

Further information (English)

Full name of research-institution/enterprise: ETH Zürich Professur für Internationale Konfliktforschung IFW D 49.2

Partners and International Organizations (English)

BE; CY; FI; DE; ELM IE; IL; IT; NL; NO; PT; ES; SE; CH; MK; UK; AT; F:Y:R of Macedonia; TR; BA; PL; RO

Abstract (English)

This project’s goal is to understand how natural resource extraction affects the outbreak and conduct of civil war. To this end, we combine disaggregated data on violent conflict with newly geo-coded information on hydrocarbon extraction sites, thus allowing the statistical analysis of the resource-conflict nexus with unprecedented spatial precision. Equipped with these data, we propose a novel instrumental variable design based on the geological determinants of oil and gas deposits to reevaluate the causal relationship between petroleum production and the outbreak of conflict. In contrast to a number of recent high-profile publications, we find that oil and gas production exert a substantially large and statistically highly significant causal effect on the outbreak of armed secessionist conflict. Moreover, we find that the oil-conflict link is limited to settlement areas of small and politically underrepresented ethnic communities, thus indicating that ethno-nationalist mobilization is a key driver of political violence in resource-rich regions. Beyond these substantive insights, the project has also led to considerable methodological advancements: The collection of the new oil-field data was performed using newly developed Python software that permits extracting arbitrary cartographic features from scanned map data with minimal human input. Further, to facilitate the analysis of the various geographic data sources required for the project, we developed a new R package for rapid, large-scale raster data processing.

References in databases (English)

Swiss Database: COST-DB of the State Secretariat for Education and Research Hallwylstrasse 4 CH-3003 Berne, Switzerland Tel. [phone] Swiss Project-Number: C12.0087