The paper estimates the heterogeneous returns of informality in Russia. The em-pirical strategy uses a Marginal Treatment Effect model (MTE). The data comes fromthe Russian Longitudinal Monitoring Survey (RLMS) and a combination of regionaldata for the period 2009-2016. I also constructed a measure of enforcement of laborregulation at the regional level that allows me to identify the parameters of the MTEmodel. I find that formal workers have, on average, 20% higher wage rates than in-formal workers. Additionally, the MTE has a negative slope, which reflects the factthat individuals with a higher propensity score of being formal are those who are morelikely to work formally -they have the largest positive gains of this type of jobs on theirearnings. My estimates also show that those individuals with a very low propensityof working formally can be negatively affected by having these types of jobs. Thesefindings suggest that the decision of working formally versus informally responds toa comparative advantage story and not segmentation of the labor market/ability hy-pothesis.

---