Abstract
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.
Highlights
The studies of structures of the income distributions have been performed for about 15 years
We assume the structure is of the form: log-normal with power law tail and we study its characteristics vs. time
The smallest values of R2 are observed for the lognormal model, which describes the smallest part of the incomes distributions: from 94.9% to 98.3%, depending on year
Summary
The studies of structures of the income distributions have been performed for about 15 years They indicate that there is no one model which describes the whole ranges the distributions. In the majority of studies incomes are best described by lognormal model with power law tail. 463–470) studied Japanese income distributions for years 1887−1998 He showed that two-part model, lognormal with power law tail is the universal structure describing distributions of personal incomes in Japan. 213– 221) studied the income distributions in United Kingdom (1994–1999) and in individual US states (1998) They showed that income distributions had two-part structure. 440–459) studied income distributions in Japan and the US based on the individual income tax returns data from 1960 to 1999 They confirmed the hypothesis about the two-part structure of income distributions. The authors described the left-central part of the distributions by the exponential model and the top 1% of incomes — by the power law model. Clementi and Gallegati (2005, pp. 3–14) investigated income distributions of households in the US
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
