This paper models income distribution in four Central and Eastern European (CEE) countries (the Czech Republic, Hungary, Poland and the Slovak Republic) in 1990s and 2000s using parametric models of income distribution. In particular, we use the generalized beta distribution of the second kind (GB2), which has been found in the previous literature to give an excellent fit to income distributions across time and countries. We have found that for Poland and Hungary, the GB2 model fits the data better than its nested alternatives (the Dagum and Singh-Maddala distributions). However, for Czech Republic and Slovak Republic the Dagum model is as good as the GB2 and may be preferred due to its simpler functional form. The paper also found that the tails of parametric income distribution in the Czech Republic, Poland and the Slovak Republic have become fatter in the course of transformation to market economy, which provides evidence for growing income bi-polarization in these societies. Statistical inference on changes in income inequality based on parametric Lorenz dominance suggests that, independently of inequality index used, income inequality in the Czech Republic, Poland and the Slovak Republic has increased during transformation. For Hungary, there is no Lorenz dominance and conclusions about the direction of changes in income inequality depend on the cardinal inequality measure used.