Modelling innovative business clusters

Science and Technology Parks (STPs) are often used as tools to foster regional development. They encourage innovation amongst the constituent firms, including by networking and knowledge spillover between the inhabitants and other actors. The high failure rate of STPs led us to evaluate a case study using panel data analysis as well as simulate how STP architecture can best cope with a changing innovation environment. Data from the Ratsit database was obtained for firms in industry sector 62X (IT and related industry) in Linköping, Sweden and then divided into those on-cluster or off-cluster. Inhabitancy conferred protection for on-cluster firms against externalities. Longitudinal studies showed that micro-firms entering the STP exodus point was seen around 15-17 years when firms, grown to around 150 employees, either plateau out in growth or depart the locality. Size and age influence corporate turnover, as does the ability to innovate, but whereas size and age have a quadratic (non-linear) impact on financial growth, innovation capabilities have a positive linear impact. Employment is mainly correlated to age, previous years' innovation and shareholder investment. Innovation output is correlated to networking measured as social expenditure, which in turn exhibits a positive influence on innovation capabilities. From the point of view of the host cluster, we simulated three organizational topologies for STPs; firstly, in the star model all are connected to the cluster initiative (CI), secondly the strongly connected model, when all are connected to each other, and finally the randomly connected model, where the network follows no centralised topology. Analyses used adjacency matrixes and Monte-Carlo simulation, trading transaction (networking) costs against knowledge benefit. Results show that star topology is the most efficient form from the cost perspective. Later, when the cost of knowledge transformation is lowered, then the strongly connected model becomes the most efficient topology. iii | P a g e Then, Agency-based Monte-Carlo simulations were then applied to clusters organisation to understand the impact of managers quality on innovation distribution using both poor and good innovation. Results show that it is very beneficial to have a central Cluster Initiative (CI) controlling the decision-making process in the early stages of STP development. However, with early maturity and commitment to a high-growth trajectory, high quality of decision-making is required amongst managers and decisions are best taken by the CI with the input of individual on-cluster firms. The scenario where CI is supported by good-quality decisions from on-cluster firms - an ambidextrous situation - is superior when good innovations abound and the STP has acquired a degree of maturity.