The paper attempts to quantify the effect of employment discrimination on the basis of disability status in Russia. We use data from the Russian Longitudinal Monitoring Survey – Higher School of Economics (RLMS-HSE) for 2005. This round of the RLMS-HSE included a question on the presence of limitations on usual activities. The question allows distinguishing the effect of unobservable differences in productivity from the effect of discrimination on the basis of disability status. Parametric and nonparametric methods of decomposition are used to solve a problem of non-comparability of disabled and able-bodied individuals and to control for unobserved differences in productivity. Our findings show that nonparametric methods are more applicable to disability discrimination studies due to “lack of common support” problem.

The evidence suggests that individuals with poor health face substantial discrimination on the basis of disability status in Russia. The discrimination explains up to 25 percent points of the total gap in employment probabilities. This effect should be interpreted as an upper bound of the discrimination after control for differences in observed and unobserved productivity characteristics. The effect may still include the impact of cash and non-cash disability benefits, self-selection into disability, environmental barriers, and wage discrimination. Our findings imply that current policy measures are not efficient in facilitating employment of the disabled.