The paper describes the further development of an agent-based multiregional input-output model of the Russian economy. We consider the idea of incorporating the government into the model and analyze the results of experimental calculations for the conditional example of spatial economy. New agents are included into the model such as the federal and regional governments, pension fund and also the state enterprises producing public goods at the federal and regional levels. The government sets four types of taxes (personal and business income taxes, VAT and payroll taxes), ensures the provision of public goods and provides social, investment and interbudgetary transfers to households, firms and budgets. Social transfers consist of social assistance and unemployment benefits. The utility functions of households are expanded by the terms associated with national and regional public goods. The budget policy is designed in accordance with the maximization of isoelastic function of social welfare that formalizes the choice between the different concepts of social justice. The Gini index is used for the monitoring the inequality of income distribution. The results of experimental calculations present the convergence of the new version of the model to the state of quasi-equilibrium. The special attention is paid an optimal level of the taxation maximizing the social welfare function. Four variants of the optimal tax rates are defined: for three major taxes at a fixed proportion of rates and for each of the tax separately at zero rates of two other taxes. The further directions of modelling are identified, they allow to investigate the spatial development of the Russian economy taking into account the decision-making by private agents in responding to government policies.

The resilience of a political, economic or trade alliances of countries or regions depends significantly on to what extent parties of a union consider their membership as beneficial. Dependent on predominant assessments, there may occur a complete disintegration of multi-regional systems, an expansion or contraction in the composition of participants, and maintenance of the status quo. The intensity of interregional exchange is used as an indicator to measure the degree of the regional integration of economic system. The paper represents the capabilities of the coalition analysis method for estimating the degree of interdependence in the multi-regional economy of the Russian Federation. The authors present an analytical review of the structural and statistical models used for estimating the effects of interregional integration as well as substantiate the choice of the approach. The basis of the coalition analysis is the input-output method and the theory of cooperative games. Moreover, the spatial input-output model of the Russian economy is developed on the methodology of national accounts statistics in terms of federal districts. We use a semi-dynamic version of the optimization multi-regional input-output model (OMIOM) on the statistical database of 2013 for developing the forecast up to 2030. On this basis, a coalition analysis of interregional interaction effects is carried out. The calculations revealed a high degree of regional integration as well as of involvement of the national economy into the world one. Depending on the structure of production and consumption of the federal districts’ economies, there are varying capacities to adapt to hypothetical changes in the volume of interregional links. The North-Western, Urals, Siberian and Far Eastern federal districts better adapt to the break of interregional links in comparison to the rest of the regions. The Central, North-Caucasian, Southern and Volga districts benefit more of their integration into the multiregional system and of the access to international markets. Prospects for using coalition analysis are associated with the introduction of an equilibrium condition, which will allow us to assess the degree of equivalence of interregional exchange.