The paper examines the quality of matching between fields of study and actual occupations in Russia. The analysis based on microdata from the 2006 Labour Force Survey data deals separately with university (tertiary A level) and college (tertiary B level) types of education. Applying multinomial logit regression technique, we estimate expected probabilities of occupational mobility conditional upon demographic characteristics and upon field of education. The most general conclusion is that the scale of mismatch is very large hinting on massive underutilization of human capital. We rely on very conservative estimates allowing that actual mismatch is likely to be even larger.