Crowding in or crowding out? Evidence from discontinuity in the assignment of business R&D subsidies

We employ a regression discontinuity design to study the effects of a flagship business R&D subsidy programme in the Czech Republic on R&D investment, patenting and economic performance of the supported firms. The R&D subsidies stimulated R&D expenditure in small and medium-sized enterprises (SMEs) but not in large firms. In SMEs, public funding succeeded in crowding in private R&D investment, and 1 unit of public subsidy was associated with about 2.3 units of additional R&D expenditure. The positive effects on R&D expenditure of SMEs were sustained after the original projects ended, possibly thanks to subsequent subsidies from the same funding provider. Supported SMEs also saw their sales increase in the short term, but we do not observe any positive effects of the support on patenting, employment or longer-term sales and productivity. We find evidence that the subsidies crowded out private R&D expenditure in large firms and financing constraints played an important role in explaining the effect heterogeneity.