Gold Price Shocks and Social Conflict

Causal evidence from the Peruvian Amazon

Causal Inference
Stata
Python
Mining
Spatial Econometrics
Spatial econometrics and DiD analysis examining how commodity price shocks affect conflict probability in resource-rich regions. SSIV and spillover modeling.
Author

Maykol Medrano

Published

October 20, 2025

Overview

This project investigates the causal relationship between international gold price fluctuations and the incidence of social conflicts in mining districts of Madre de Dios, Peru.

Research Question

How do international commodity price shocks affect the probability and intensity of social conflicts in resource-rich regions?

Methodology

Identification Strategy: - Shift-Share Instrumental Variable (SSIV) approach - Difference-in-Differences (DiD) with spatial econometrics - Two-way fixed effects accounting for spillover effects

Key equation:

\[ Z_{dt} = \text{Endowment}_d \times \log(\text{Price}_{t}) \]

Where \(Z_{dt}\) is the instrument for district \(d\) at time \(t\).

Data Sources: - ENAHO household surveys (2015-2022) - International commodity price data (World Bank) - Social conflict registry (Defensoría del Pueblo) - Spatial weights matrix for spillover modeling

Key Findings

  • A 10% increase in gold prices leads to higher conflict probability
  • Effects concentrated in districts with weak institutional presence
  • Spatial spillover effects extend to neighboring districts
  • Impact driven by competition for land rights and resource access

Implementation

* Load data and setup project structure
usecasen 2022, clear
projectinit, setup("Conflict_Analysis")

* Define Shift-Share Instrument
gen shift_share_iv = gold_endowment * log(gold_price_int)

* Main Specification (Fixed Effects)
reghdfe conflict_dummy shift_share_iv canon_transfers ///
    population_density, ///
    absorb(district_id year_month) ///
    vce(cluster district_id)

* Export Results
outreg2 using "results/main_table.tex", replace tex
import geopandas as gpd
import matplotlib.pyplot as plt

# Load district data with conflict intensity
peru_map = gpd.read_file("data/peru_districts.geojson")

# Spatial visualization
fig, ax = plt.subplots(1, 1, figsize=(10, 10))
peru_map.plot(column='conflict_intensity',
              ax=ax,
              legend=True,
              cmap='OrRd',
              scheme='quantiles')

plt.title("Spatial Distribution of Social Conflicts")
plt.show()

Policy Implications

Findings suggest need for: - Enhanced conflict prevention mechanisms in extractive zones - Strengthened local governance in resource-rich areas - Early warning systems for commodity price-driven tensions - Improved benefit-sharing mechanisms from mining activities

Citation

@unpublished{medrano2025mining,
  title={Gold Price Shocks and Social Conflict: Causal Evidence from the Peruvian Amazon},
  author={Medrano, Maykol},
  year={2025},
  note={Working Paper},
  url={https://maykolmedrano.github.io/projects/mining-conflict.html}
}

APA Format: Medrano, M. (2025). Gold Price Shocks and Social Conflict: Causal Evidence from the Peruvian Amazon [Working Paper]. https://maykolmedrano.github.io/projects/mining-conflict.html

Full working paper and replication code available upon request.