Inference Tests for Gini-Based Tax Progressivity Indexes

Abstract

Distribution-free statistical inference procedures for changes in Lorenz- and Gini-based indexes of tax progressivity are developed and applied. Related but distinct tests for the Reynolds–Smolensky index of residual progression and the Kakwani index of liability progression are provided. The inference procedures are illustrated by applying them to Luxembourg Income Study microdata for Sweden, the United Kingdom, and the United States before and after periods of tax reform. In each country a finding of significant change depends on the choice among progressivity indexes. No single index exhibits a consistent pattern of significant change in all countries across time.

KEY WORDS:

• Gini index
• Kakwani index
• Luxembourg Income Study
• Reynolds–Smolensky index
• Tax reforms