Urban Network Capital: How do Chinese Cities Grow in a Networked Economy?

Department: Real Estate and Construction
Active Dates: 2019-2022
Principal Investigator: S.K. WONG (PI), S. SHI (Co-I)


After four decades’ rapid growth, China’s economy is entering a consolidation phase in which industrial restructuring activities are substantially intensified. Enormous capital flows between cities have created a complex inter-urban network that becomes intriguing and significant to understand China’s urban dynamics in the new development era. The primary question this research seeks to answer is whether urban growth is driven by a networked economy underpinned by inter-city capital flows. Urban network capital refers to the connections a city has with other cities that enable virtual, rather than physical, exchange of information, thoughts, knowledge and capital. It serves to bridge the gap between mobile capital flows and immobile local market. This research will utilize Mergers and Acquisitions (M&A) of firms in China as a proxy of capital flows to construct an inter-urban capital network. The results of this research will help urban planners and enterprises clarify their cities’ strategic positions and formulate policies to strengthen urban competitiveness.


  1. Unveil China’s underlying inter-urban capital network through tracing the M&A capital flows.
  2. Establish empirical models to examine the relationship between network capital and urban growth.
  3. Provide theoretical implications on the nexus between mobility and immobility in urban development and policy implications on improving urban competitiveness.

Initial Results:

  1. In the economic network space, outperforming cities, particularly Beijing, Shanghai and Shenzhen, hold most of the network resources, while other cities are at a disadvantage to compete, illustrating a centralised core-periphery pattern.
  2. network capital is significantly related to urban growth attuned to flow directions and network positions.
  3. network capital and local agglomeration factors are interactive in the process of economic growth, illuminating an emergent networked agglomeration economy.

Initial Discussions:

  1. Although the transportation cost has been largely reduced, building distant economic linkages requires a higher threshold and additional maintenance costs to overcome the barriers created by information asymmetry and heterogeneous institutional settings. The threshold will grant competitive edges to big enterprises equipped with powerful search abilities and business resources as well as the cities where they are spatially clustering in, which ultimately leads to further regional divergence.
  2. The results add weight to the proposition that urban economic growth is increasingly realised by the growth of capital flow networks, indicating that the network capital is an unneglectable composite of the economic growth function.
  3. Urban development patterns are reshaped by the dynamic interplay between agglomeration and network externalities. the underlying mechanism of urban development should be seen as an evolutionary and adaptive process in which cities interact and adapt to each other attuned to their localised attributes instead of the simple sum of the properties of city nodes and their direct interlinkages.
  4. policy implications should focus on improving cities’ institutional organizing capacity towards important flows in order to ‘borrow’ positive spillovers from strategic networks and counter ‘agglomeration shadows’. Policies demands a ‘place-based’ and ‘coordinative’ perspective under a networked agglomeration economy, particularly in terms of the local moderators of network capital (e.g. policies on local education and R&D development).
Figure 1 The Link Mechanism between M&As and Local Economies.Figure 2 The Spatial Distribution and Subgroup Clustering  of China’s Inter-city Capital Flow Network (Notes: top 100 links are illustrated based on the number of M&As that pair cities invest in each other (reciprocity); top 10 links are highlighted in red; city subgroups are indicated by different colours, the modularity is optimised at 0.207 through sensitivity analysis; the M&A gravity is devised to the southeast by 31 degrees in  450 km).