Finance: Theory and Practice | 2026
Authors: Borisov O.I.
DOI: 10.26794/2587-5671-2026-30-1-160-173
Journal: Finance: Theory and Practice
Year: 2026
Publisher: Financial University under The Government of Russian Federation
Document Type: Article
Open Access: All Open Access; Gold Open Access
Cited by: 0
This article analyzes dividend income taxation across Russian federal subjects using statistical data from the Federal Tax Service of Russia and Rosstat for 2012–2023. The purpose of the study is to identify federal subjects exhibiting signs of personal income tax base erosion related to dividend income. This objective predetermined the following research tasks: to develop a regression model for structural and dynamic analysis of dividend income; to identify federal subjects with significant deviations from nationwide indicators; to determine potential tax optimization schemes that may cause these deviations; and to develop a risk-oriented model for identifying federal subjects where non-residents and individual entrepreneurs may potentially utilize “dividend” tax evasion schemes. The research employs econometric modeling (panel regression model with fixed effects), comparative analysis, and statistical analysis. The study establishes a statistically significant relationship between the share of dividends in total individual income and corporate profit dynamics, as well as a strong influence from the proportion of non-residents and individual entrepreneurs receiving dividends. The analysis controls for macroeconomic shocks and time-invariant regional factors such as industry structure and investor profiles. Federal subjects exhibiting signs of personal income tax base erosion related to dividend income were identified, and key factors that tax authorities should consider during tax audits were highlighted. The novelty of this research lies in constructing a regression model for structural and dynamic analysis of dividend income, as well as developing a risk-oriented model for identifying federal subjects where non-residents and entrepreneurs may potentially employ “dividend” tax evasion schemes. The practical significance stems from enabling tax authorities to use the proposed models to identify potential centers of systematic tax evasion scheme usage and facilitate their prompt suppression. The proposed research methodology allows for decomposition at to the municipal level, enabling more precise identification of sources of potential personal income tax base erosion. © Borisov О.I., 2026.
dividend taxation; fixed-effects regression; PIT; profit distribution; regional analysis; risk score; tax avoidance; tax policy