Occupational Income Inequality of Thailand: A Case Study of Exploratory Data Analysis beyond Gini Coefficient

TL;DR

This study analyzes income inequality in Thailand using both Gini coefficients and income domination networks, revealing occupational disparities overlooked by traditional measures and providing detailed insights into inequality patterns across regions.

Contribution

It introduces a combined approach of Gini coefficient and network analysis to uncover occupational income inequality, offering a more comprehensive perspective than traditional methods.

Findings

Agricultural provinces show low inequality in both measures.

Non-agricultural provinces exhibit occupational inequality without general inequality.

Estimation statistics reveal the magnitude of income gaps among occupations.

Abstract

Income inequality is an important issue that has to be solved in order to make progress in our society. The study of income inequality is well received through the Gini coefficient, which is used to measure degrees of inequality in general. While this method is effective in several aspects, the Gini coefficient alone inevitably overlooks minority subpopulations (e.g. occupations) which results in missing undetected patterns of inequality in minority. In this study, the surveys of incomes and occupations from more than 12 millions households across Thailand have been analyzed by using both Gini coefficient and network densities of income domination networks to get insight regarding the degrees of general and occupational income inequality issues. The results show that, in agricultural provinces, there are less issues in both types of inequality (low Gini coefficients and network densities), while some non-agricultural provinces face an issue of occupational income inequality (high network densities) without any symptom of general income inequality (low Gini coefficients). Moreover, the results also illustrate the gaps of income inequality using estimation statistics, which not only support whether income inequality exists, but that we are also able to tell the magnitudes of income gaps among occupations. These results cannot be obtained via Gini coefficients alone. This work serves as a use case of analyzing income inequality from both general population and subpopulations perspectives that can be utilized in studies of other countries.