This paper evaluates the rate of educational assortative mating and its impact on income inequality in Russia. We use data from the Russian Longitudinal Monitoring Survey – Higher School of Economics (RLMS-HSE) for 1995–2015. Our findings suggest that the Russian marriage market is characterized by positive assortative mating at all levels of education. It means that marriages occur among individuals with the same level of education more frequently than it would be expected under random matching with respect to education. The level of educational assortative mating has not changed in the last two decades: the observed decline is statistically insignificant. The lack of a strong trend of assortative mating can be explained by substantial heterogeneity in developments across education sub-groups. Assortative mating has been declining over time for university graduates, whereas individuals with low level of education had growing incentives to sort themselves into educationally homogeneous marriages. Changes in assortative mating had little effect on the level of household income inequality. The effect has been increasing over time but still remains below the estimates obtained for developed countries. If marriages were formed randomly across time the counterfactual Gini for couples’ incomes would be lower than the actual one on average by 1,3%. The effect has been stronger and inequality-enhancing at the top of the distribution. At the same time, educational assortative mating had neutral or even equalizing for couples at the bottom of the distribution. Economic shocks were found to amplify the effect of assortative mating on inequality in both parts of the distribution.