Two-component structure of household income distributions in Poland

Research background: Studies of the structures of the income distributions have been performed for about 15 years. They indicate that there is no model which describes the distributions in their whole range. This effect is explained by the existence of different mechanisms yielding to low-medium and high incomes. While more than 97% of the distributions can be described by exponential or log-normal models, high incomes (about 3% or less) are in agreement with the power law. Purpose of the article: The aim of this paper is an analysis of the structure of the household income distributions in Poland. We verify the hypothesis about two-part structure of those distributions by using log-normal and Pareto models. Methods: The studies are based on the households? budgets microdata for years 2004?2012. The two-component models are used to describe the income distributions. The major parts of the distributions are described by the two parametric log-normal model. The highest incomes are described by the Pareto model. We also investigate the agreement with data of the more complex models, like Dagum, and Singh-Madalla. Findings & Value added: One has showed that two or three parametric models explain from about 95% to more than 99% of income distributions. The poorest agreement with data is for the log-normal model, while the best agreement has been obtained for the Dagum model. However, two-part model: log-normal for low-middle incomes and Pareto model for the highest incomes describes almost the whole range of income distributions very well.