Exploring external urban relational processes: inter-city financial flows complementing global city-regions

ABSTRACT

External urban relations are commonly described as one of two types: hierarchical local hinterlands (central place theory) and networked non-local hinterworlds (central flow theory), referred to as town-ness and city-ness, respectively. This paper builds on and develops these generic concepts to make them specifically relevant to today’s corporate globalization. The central place process is represented by multi-nodal global city-regions, and the central flow process is represented by inter-city capital investment flows. We find that capital flows in global cities increase flows to proximate smaller cities within their regions. This empirical link between city-ness and town-ness has theoretical and policy implications.

KEYWORD:

• advanced producer service
• central flow theory
• central place theory
• foreign direct investment flows
• global cities
• regional cities

JEL:

• O18
• R11
• R12

ACKNOWLEDGEMENTS

The authors thank the editor and reviewers for their constructive comments on earlier drafts of this paper. The usual disclaimers apply.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the authors.

Notes

1. The service value of a city to a firm is a measurement of the relative importance of the office in a certain city within the firm’s overall office network. These values range from 0 (a firm having no office in a city) to 5 (a city housing the global headquarters of a firm).

2. The advanced service sector includes accounting, advertising, insurance, investment management, management consulting services, legal services, banking, scientific and technical services, auxiliary financial services, and business school services.

3. Excluding regional cities generates quantitatively robust results. If we estimate the model: $FLo{w}_{i,j}^G=\exp [a+{\beta }^{APS}\ln (r_{i,j}^{APS})+\delta \ln (Dis{t}_{i,j})+{\gamma }_1\ln (Eco{n}_i)+{\gamma }_2\ln (Eco{n}_j)+\ln (e_{i,j}^G)$using only global cities, ${\beta }^{APS}$ is significantly positive and has a value of 1.50. The complete regression results are available from the authors upon request.

4. The geographical distance is calculated based on the ‘haversine’ formula and the latitudes and longitudes of the cities. In a separate analysis, we also used driving distance and driving time as proxies (see Table A3 in Appendix A in the supplemental data online). Driving distance is measured using the Google Map API for every two cities. Driving time is measured using the Google Maps Application Programming Interface (API) for each pair of cities. When replacing geographical distance by driving time with a bandwidth of 6 hours, the significant positive influence holds. When using the driving distance with a bandwidth of 600 km, the results are also robust. A 1% increase in the flows to nearby regional cities is associated with a maximum 0.11% increase in the investment flows to the regional city. The regression results are given in Table A2 in Appendix A in the supplemental data online.

5. The correlation coefficient between independent variables is reported in Table A4 in Appendix A in the supplemental data online.

6. Our empirical results only show the average intensity of the relationship among the 247 OECD cities based on the APS global network. When multi-scale networks are included, some regional globalizing centres may influence flows to global cities in some specific sectors (e.g., Martinus et al., 2015). However, given the limited space and scope of this study, we leave in-depth individual case analyses for future research.

7. Table A1 in Appendix A in the supplemental data online lists four groups of cities: Alpha- and Beta-level GaWC global cities; Gamma, High Sufficiency, and Sufficiency level GaWC global cities, remaining GaWC cities, and other OECD regional cities that are not included in the GaWC list.

 

Additional information

Funding

Kathy Pain and Bing Zhu thank the Reading Real Estate Foundation (RREF), University of Reading, for its financial support in providing access to the Financial Times fDi Markets data.