Since the Lisbon Agenda (2000), European policies have been increasingly oriented toward R&D and innovation. In this context, we analyze the effects of a centralized R&D subsidy policy upon a steady state and welfare situation using an agglomeration and growth model composed of two asymmetric countries. Whereas the policy leads to a steady state characterized by a higher growth rate and lower inequalities, the welfare analysis provides a more contrasted vision on the effects of the policy. Even though such a policy can eliminate some distortions and improve global welfare, it implies a zero-sum game at the national level as the policy increases welfare in the periphery country and decreases it in the core country once knowledge spillovers are sufficiently localized. Consequently, a centralized R&D subsidy policy implemented in an economy composed of two asymmetric countries will be a positive-sum game only if international knowledge spillovers are sufficiently large.