This paper develops a discrete variant of the semiparametric methodology of DiNardo, Fortin, and Lemieux (1996) (DFL) to measure the effects of socio-economic variables on the income distribution. Although the proposed method is also based on the calculation of hypothetical income distributions by reweighting the original population, it is much easier to implement. The framework is applied to examine the distributional effects of rising unemployment, decreasing female labor market participation, and widening income structure in East Germany following the reunification with West Germany in 1990. The empirical results suggest that both these tendencies contributed considerably to the recent increase in income inequality in East Germany.