In the EU, a sizable part of innovation is attributed to the activities other than R&D such as purchases of advanced machinery, licenses, patents and minor modifications in products or processes. These non-R&D innovation activities receive substantial funding from the European cohesion policy (ECP). In this paper we applied the dynamic spatial computable general equilibrium model RHOMOLO to evaluate the ex-ante short and long run economic impacts of 2014-2020 non-R&D innovation subsidies allocated to the EU27 NUTS2 regions. The results of computer simulations show that the most notable welfare improvements (GDP, production and household consumption) were observed in the Eastern EU regions that receive the largest share of funding. Such outcome is in line with the goals of the European Cohesion Policy of stimulating economic convergence of the least developed regions. As was expected, the magnitude of macroeconomic impacts positively correlates with the amount of non-R&D subsidies allotted to the regions.