Measuring inequality beyond the Gini coefficient may clarify conflicting findings

Area of Science:

• Socioeconomic studies
• Econometrics
• Data science

Background:

• Previous research on economic inequality shows inconsistent findings.
• The Gini coefficient, a common measure, offers a limited view of inequality.
• A narrow focus on single-parameter measures may explain contradictory results.

Purpose of the Study:

• To conceptualize inequality measurement as a data reduction task for income distributions.
• To evaluate the performance of various inequality models using fine-grained data.
• To identify superior models for analyzing the relationship between inequality and policy outcomes.

Main Methods:

• Utilized a dataset of 3,056 US county-level income distributions.
• Estimated the fit of 17 proposed inequality models.
• Conducted simulations to validate model performance and interpret parameters.

Main Results:

• Multi-parameter models consistently outperformed single-parameter models like the Gini coefficient.
• The two-parameter Ortega model demonstrated superior fit, distinguishing between lower- and top-income inequality.
• The Ortega parameters showed different correlations with 100 policy outcomes compared to the Gini coefficient.

Conclusions:

• Multi-parameter models provide a more comprehensive understanding of economic inequality.
• Data-driven approaches and advanced models are crucial for accurate inequality research.
• The findings suggest a need to reconsider traditional inequality metrics in policy analysis.