Kinetic theory and Brazilian income distribution

TL;DR

This study analyzes Brazilian income distribution using survey data, confirming a semi-empirical model that combines Pareto and Boltzmann-Gibbs distributions, and finds decreasing inequality and persistent subgroup disparities from 2001 to 2014.

Contribution

It validates a semi-empirical kinetic model for income distribution in Brazil and provides a comprehensive inequality analysis across different social groups over time.

Findings

Income inequality decreased from 2001 to 2014.

Disadvantaged groups have more equalitarian income distribution.

Subgroup disparities remained stable over the studied period.

Abstract

We investigate the Brazilian personal income distribution using data from National Household Sample Survey (PNAD), an annual research available by the Brazilian Institute of Geography and Statistics (IBGE). It provides general characteristics of the country's population. Using PNAD data background we also confirm the effectiveness of a semi-empirical model that reconciles Pareto power-law for high-income people and Boltzmann- Gibbs distribution for the rest of population. We use three measures of income inequality: the Pareto index, the average income and the crossover income. In order to cope with many dimensions of the income inequality, we calculate these three indices and also the Gini coefficient for the general population as well as for two kinds of population dichotomies: black / indigenous / mixed race versus white / yellow; and men versus women. We also followed the time series of these indices for the period 2001-2014. The results suggest a decreasing of Brazilian income inequality over the selected period. Another important result is that historically-disadvantaged subgroups (Women and black / indigenous / mixed race),that are the majority of the population, have a more equalitarian income distribution. These groups have also a smaller monthly income than the others and this social structure remained virtually unchanged in the period of time.