Well connected compared to what? Rethinking frames of reference in world city network research

Claims about the strength of cities' global connections have become commonplace in the world cities literature. Although such claims are inherently comparative, authors often do not specify the reference. London is well connected compared to what? In this paper, I adapt the stochastic degree sequence model from network analysis as a tool to derive a frame of reference that can be used to inform and substantiate such claims. Beyond providing a formal statistical method for deciding when the claim that X is well connected is justified, it also addresses a number of other challenges in this literature, including more explicitly casting firms as key agents in world city formation, providing insight into when and where global firms might be expected to locate their branch offices, and helping identify cases that warrant more detailed investigation. To illustrate, I apply the method to data on cities and firms from 2013, examining the results at five different scales, from the individual city and firm to the entire world city network. I conclude by considering how this approach allows researchers to ask different kinds of questions about the nature of world city status.