Killing in the Amazon: Local Political Power and Violence Over Land and the Environment in Brazil

Value of the Information & Strength of the Argument

The value of this research lies in its causal identification strategy, which elevates a largely correlational literature to a higher standard of evidence. The argumentation is well-structured: the presenter clearly motivates the question, describes the context, details the data sources, and presents findings incrementally. He also addresses potential alternative explanations, such as donor capture, and demonstrates that effects are strongest when the mayor’s own economic livelihood is tied to extractive enterprises. The Q&A section further tests the robustness of the results, showing receptiveness to suggestions for additional analyses.

Scientific Rigor, Source Quality, Title Accuracy

The scientific rigor is high: the study relies on official administrative records (IBAMA, mortality registry), satellite data, and hand-coded NGO reports (CIMI and CPT), with explicit attention to data cleaning and reconciliation. The regression discontinuity design is a well-established method for causal inference, and the survey complements it with resident perspectives. The title is apt, capturing the core focus on violence driven by local political power in the Amazon. No comments were provided, so public reception cannot be assessed.

The title accurately reflects the content, focusing on violence over land and environment in the Amazon linked to local political power.

8/10

The talk presents a rigorous regression discontinuity design and original survey data, conducted by a team at Brown University, with transparent methodology and clearly stated findings. The data sources are reputable Brazilian institutions and NGOs, and the analysis includes multiple robustness checks.

Markers derived by PSI from the transcript: the creator did not define chapters.

• 00:00Introduction and announcements by Dennis Gonzalez; welcoming Professor Robert Blair.
• 03:00Blair sets the empirical backdrop: extractive industries expanding into the Amazon, conflict over land, and statistics on violence.
• 10:00Presentation of the study design: regression discontinuity with mayoral elections and a survey of 2,000 residents.
• 20:00Main findings: municipalities electing agricultural/livestock producers see ~11.5 more attacks per 100k, roughly tripling the control group; no effect on general homicides.
• 30:00Discussion of federal environmental enforcement: significant increases in violations, fines, and blacklisting, but no change in deforestation.
• 40:00Survey evidence: respondents in affected municipalities fear expropriation, formalize land claims, but report less violence, suggesting defensive adaptation.

• CIMI (Conselho Indigenista Missionário) — Used as a source for coding conflicts involving indigenous peoples.
• CPT (Comissão Pastoral da Terra) — Primary source for land conflict reports and violence data.
• IBAMA (Instituto Brasileiro do Meio Ambiente) — Federal environmental agency providing data on violations and fines.