Explaining economic development from spatial configuration: The influence of similarity and complementarity

The present paper presents a geosimulation agent-based model of firm development that is based on firms' behavioural rules. The model describes growth, closure, spin-offs and relocation of firms in 21 industrial sectors in the Netherlands, based on a dataset of all individual firms in the Netherlands. An important aspect of the model is the potential interaction that exists between firms, which is represented by various agglomeration indicators. The indicators represent the effect of the distance to other firms, which may have implications fro market potential, agglomeration advantages and congestion/competition. By including positive and negative attraction effects a proper description of concentration and deconcentration tendencies is achieved. The interactions may take place on difference scales, depending on the type of effect (agglomeration advantage, competition for space) and economic sector. The interactions affect growth, closure and spin-off of the individual firms in the model. The model is applied to simulate the distribution of economic development in the Netherlands during a ten year period (1998-2007). The simulation results are then compared to existing 2007 data. This comparison confirms the effectiveness of the model. Simulation carried out with and without various positive and negative agglomeration factors indicate that including agglomeration factors into the model seriously adds to the quality of predictions.