Assessment of the impact of selected science and technology policy instruments on innovative output in Russian regions

In the context of sanctions pressure and a course towards technological sovereignty, the issue of the effectiveness of Russia’s science and technology policy is becoming critically important. However, direct financing of R&D, the introduction of special preferential regimes, and personnel training often do not lead to a steady increase in the output of high-tech products, which indicates a systemic gap between investments in scientific and technological potential and final innovative production. The purpose of this study is to assess the impact of key instruments of scientific and technological policy on innovative output in the regions of Russia, taking into account the problems of endogeneity and time lags. Based on data from 82 regions for 2014–2023 using the dynamic panel model (System GMM), the following key gaps were identified. It is shown that the costs of technological innovations in the basic specification demonstrate a significant negative effect with an annual lag and a positive return with only a two-year lag. Taking into account the regional typology, this effect becomes insignificant, which indicates the heterogeneity of this effect in the context of regional groups. The presence of special economic zones (as one of the instruments of special preferential regimes) has a short-term positive effect, which is replaced by a significant negative effect at the second lag, indicating unstable returns from the SEZ institution; over the period under review, no long-term positive effect is observed. The steady negative impact of the number of researchers with academic degrees on innovative output points to a deep gap between the system of scientific personnel training and the needs of the real sector, where researchers are concentrated in basic research that is not focused on commercialization. Significant regional heterogeneity has been revealed: industrial regions demonstrate the most stable returns from R&D, while in knowledge-intensive and resource-rich regions the positive effect is replaced by a deep recession. The reconstructed individual effects confirm significant structural variation between regions. The results of diagnostic tests (AR(2), Hansen) confirm the validity of the instruments and the absence of second-order autocorrelation. The findings indicate the need to shift the policy focus from financial support measures to institutional reforms aimed at creating mechanisms for cooperation in the state-science-business relationship and building differentiated commercialization management models, taking into account regional specialization.