Considerable research on developed countries shows the economic factors matter for voters. But for transitional democracies there has been uncertain evidences of the economics and elections connection using microdata. This view fits the Russian case well, where research on aggregate data supports economic voting but the conclusion can be limited due to institutional restrictions on access to elections and self-selection of voters. We consider modelling economic voting as a two step process, where the voters first decide to turn out and then choose the party at the polls. Given the subjective question this analyse uses, particularly in the context of likely endogeneity, an instrumental variable method is called for. We observed that negative egotropic perceptions does not decrease the probability to vote for the incumbent.