Detecting the existence of space–time clustering of firms

The use of the K-functions (Ripley, 1977) has recently become popular in the analysis of the spatial pattern of firms. It was first introduced in the economic literature by Arbia and Espa (1996) and then popularized by Marcon and Puech (2003), Quah and Simpson (2003), Duranton and Overman (2005), and Arbia et al. (2008). In particular in Arbia et al. (2008) we used Ripley's K-functions as instruments to study the inter-sectoral co-agglomeration pattern of firms in a single moment of time. All this research has followed a static approach by disregarding the time dimension. Temporal dynamics, on the other hand, play a crucial role in understanding the economic and social phenomena particularly when referring to the analysis of the individual choices leading to the observed clusters of economic activities. With respect to previous contributions to the literature, this paper uncovers the process of firm demography by studying the dynamics of localization through space–time K-functions. The empirical part of the paper will focus on the study of the long run localization of firms in the area of Rome (Italy), by concentrating on the Information and Communication Technology (ICT) sector data collected by the Italian Industrial Union in the period 1920–2005.