Regional Knowledge Creation and R&D collaboration in Europe

Strong dynamics of innovation in regions are crucial for achieving national and regional economic growth objectives. In this respect, policy recommendations are being sought by European, national as well as regional policymakers. Regional innovation results from the innovative performance of firms, which increasingly have to face global competition, tightened by accelerated technological development and specialization. The necessary new knowledge can no more be created within organizations only, but has increasingly to be sourced through R&D collaborations in flexible networks with partners nearby as well as from anywhere in the world. In this regard, inter-regional networks for the creation of knowledge are gaining importance for regional policy. However, there is little systematic evidence so far in Regional Science on the relationship between localised knowledge creation processes in firms and their embedding in inter-regional R&D collaboration networks. With this proposal we aim at closing this research gap by modelling knowledge creation in regions, taking into account firm-level R&D collaborations, inside and outside of the region. We adopt an empirically driven agent-based modelling (ABM) approach to simulate knowledge creation in European regions. In particular we focus on three main objectives: (i) characterising the influence of inter-regional R&D collaboration on the regional knowledge creation, (ii) disentangling local buzz and global pipelines as triggers of the regional knowledge creation, and (iii) characterising technological specialization and geographical concentration tendencies (as well as future prospects) among and within European regions. The project will demonstrate the power of the ABM approach up to now rarely used in the context of Regional Science to analyze the interdependencies between R&D networks and the technological performance as well as the collaborative behaviour of firms. Moreover, the novelty of the project is in its extensive use of empirical data for the initialization and the calibration of the ABM, which will enable us to apply the model to real world contexts: In policy experiments the effect of political framework conditions on the regional knowledge creation can be simulated, in order to support the ex- ante evaluation of different public policy measures. Both the theoretical foundation of the model and its empirical foothold will equip the simulation model with a high level of relevance for policymakers at regional, national and European levels.

Regional economic development is closely related to innovative performance of firms who are increasingly facing severe global competition. Continuous product differentiation, accelerating technological change and breakthroughs are challenging even large corporations, forcing them to complement their own R&D activities with external as well as distant knowledge, in order to keep pace with their global competitors. From a regional innovation policy perspective, the question arises how knowledge creation in firms and returns from regional specialization might synergize locally for emergent and promising technological strengths. R&D networks within the region but also outward-oriented ones are in general attributed an enabling and sustaining role for regional innovation. How micro- processes like collaboration and knowledge exchange activities interact for nurturing aggregate regional innovation dynamics remains one of the urging questions in regional innovation research. The project Regional Knowledge Creation and R&D Collaboration in Europe has been focusing on this issue and has analysed the complex dynamics of regional knowledge creation using computer simulations. To this end, an agent-based model (ABM) was developed representing firms as model agents who evolve over time with respect to their technological profiles by performing R&D and by mutually exchanging knowledge. In this way, depending on the intra-regional and inter-regional links at the micro-level, also the region-level technological specialisation patterns co-evolve. The model is highly relevant for regional innovation research and policy because it discerns effects of local exchange processes and global knowledge channels and provides a means to identify technological specialization tendencies and geographical concentration processes by producing quantitative future scenarios. The value of the model is in its strong empirical foundation, given by scientifically established model processes on the one hand and the use of extensive data on European regions (data on patents, collaborative R&D projects, regional economic statistics and company data) on the other.