Scholarlink

<p>This study examines the relationship between income inequality and violence using a global regression discontinuity framework with data from 1990 to 2020. We identify discontinuities in violence outcomes at specific levels of income inequality where the regression discontinuity assumptions hold. Empirical estimations are conducted using a novel approach that applies local linear regressions with optimal bandwidths, employing robust and bias-corrected estimators. Additional sensitivity analyses are also performed. Our findings reveal a positive and statistically significant discontinuity at a Gini index cutoff of 41.0, indicating an effect of inequality on violence with semi-elasticities of 0.43 and 0.46. This relationship appears particularly driven by data from Latin America, suggesting a context-specific generalization of the inequality–violence nexus. The validity of this cutoff for the broader sample is supported by the absence of manipulation in income inequality, robustness to alternative bandwidth choices, and balanced covariates.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

What Do We See from Inequality and Violence? Evidence from a Global RDD Approach

  • 期刊论文

  • 发表日期：

    2026年2月24日

查看全文
  • John Michael Riveros-Gavilanes,
  • Jeisson Andres Riveros-Gavilanes,
  • Sherif M. Hassan,
  • Oluwaseun Samuel Oduniyi

摘要

This study examines the relationship between income inequality and violence using a global regression discontinuity framework with data from 1990 to 2020. We identify discontinuities in violence outcomes at specific levels of income inequality where the regression discontinuity assumptions hold. Empirical estimations are conducted using a novel approach that applies local linear regressions with optimal bandwidths, employing robust and bias-corrected estimators. Additional sensitivity analyses are also performed. Our findings reveal a positive and statistically significant discontinuity at a Gini index cutoff of 41.0, indicating an effect of inequality on violence with semi-elasticities of 0.43 and 0.46. This relationship appears particularly driven by data from Latin America, suggesting a context-specific generalization of the inequality–violence nexus. The validity of this cutoff for the broader sample is supported by the absence of manipulation in income inequality, robustness to alternative bandwidth choices, and balanced covariates.