Import substitution today is not so much an economic value as a strategic one. Therefore, it is fundamentally important to identify priorities and directions in which, first of all, economic and national security must be ensured through own production development. But at the same time it is necessary to assess the possibilities of replacing a particular product, objectively consider all risks, since it is not always possible to ensure a full production cycle at the expense of national resources and even the capabilities of friendly countries. In this context, it is critical to develop scientific approaches for effective import-independence processes.
In developing a comprehensive import substitution policy, the results of economic modeling should be relied upon. Mathematical models for assessing the necessity and effectiveness of import substitution are based on various economic variables and principles and are in some cases complex and multifactorial. Among the most common models are Input—Output models and dynamic stochastic general equilibrium models (DSGE).
Input—Output models are significant for the analysis and evaluation of import substitution, since it is often not enough to rely on macroeconomic indicators reflected in the System of National Accounts. In particular, gross exports do not take into account the size of intermediate consumption of goods and services that come from a unit of export revenue. If we take net exports as the difference between exports and imports of products of a certain industry, the calculation of the indicator also does not take into account intersectoral relationships. This does not allow linking exports with intermediate consumption of imported products from other industries. In this case, based on the considered indicators, the assessment of the contribution of exports to GDP will be distorted, since industries have different parameters of the share of intermediate imports in gross output.
In addition, it is advisable to use international Input—Output databases within the Union State in order to assess integration effects, as well as to consider a system of indicators calculated on their basis, for example, industry indicators of import intensity.
In the last decade, DSGE has brought major changes to the empirical practice of macroeconomic research. In particular, there has been a qualitative shift from applying vector autoregressive models (VAR) to the mass use of DSGE. These models are widely used by modern central banks in developing economic policy. The specification of the agents utility functions allows to study the effects of various types of macroeconomic interventions and macroeconomic policies on the welfare of agents, which is impossible with VAR models. DSGE are more complex than Input—Output models, but they can take into account many variables and factors affecting import substitution.