Authors: Zachary Neal*, Michigan State University, Ben Derudder, Ghent University, Peter Taylor, Northumbria University
Topics: Urban Geography, Quantitative Methods, Economic Geography
Keywords: network, world city, methodology
Session Type: Paper
Start / End Time: 9:55 AM / 11:35 AM
Room: Diplomat Room, Omni, West
Presentation File: No File Uploaded
World city networks derived using the interlocking world city network model (IWCNM) have been useful for explaining a range of urban economic and transportation outcomes. This widely used model views economic linkages between cities as forged by the co-location of advanced producer services firms. To date, this research has been retrospective, collecting data about the current location of firms, then deriving a world city network. However, if the future location of advanced producer service firms can be accurately predicted, then the IWCNM can be applied to these predicted locations to forecast a future world city network. In this paper, we explore this possibility, using a recently developed firm location forecasting model. Specifically, using firm location data collected between 2000 and 2018, we examine the accuracy of predictions generated using this model over different prediction windows. We find that predictions are accurate even when made 15+ years into the future, but that prediction accuracy is impacted by whether or not a significant destabilizing event occurs during the window.