This paper decomposes the gender gap in pay in the Russian Federation along the earnings distribution for the period 1996–2011. The analysis uses a reweighted, recentered influence function decomposition that allows estimating the contribution of each covariate on the wage structure and composition effects along the earnings distribution. The paper finds that women are in flat career paths compared with men; the importance of observable characteristics that proxy human capital in the gender pay gap decrease along the earnings distribution; and if women's pay took into account their educational degrees as much as men's, the gender pay gap would disappear or even reverse at the top of the earnings distribution. The results suggest that women at the bottom of the earnings distribution should be helped to increase their labor market skills, and women at the top of the distribution should be helped to break the glass ceiling and be remunerated for their skills to the same extent as men.