Cohesion funds and regional patenting: inventive capacity in innovation systems under construction

Cohesion policy is the European Union's main instrument for reducing regional disparities, yet whether it builds the capacity of lagging regions to generate new knowledge remains poorly understood. The causal literature has measured income and growth rather than invention, and has treated absorptive capacity as a precondition for transfers to work rather than as something transfers might create. This paper asks whether cohesion funds build inventive capacity, and where. It applies the Callaway and Sant'Anna difference-in-differences estimator to European NUTS-3 regions from 2000 to 2022, taking European Patent Office applications by inventor region as the outcome and exploiting the 75% eligibility threshold. Eligibility raised patenting by around a fifth, but the effect became detectable only in the seventh year and accumulated thereafter; the same lag emerges independently in the following programming period. The return was concentrated among regions of the member states acceding in 2004 and 2007, where patenting roughly doubled, and was absent in the convergence regions of longer-standing member states however little they patented before. Cohesion funds build inventive capacity where the surrounding innovation system is under construction rather than where inventive output is low, so evaluations aligned with the seven-year programming cycle will systematically miss the effect.