The Structural Features and Centrality Optimization of a Firm Interlocking Network of the Nodal Cities on the South Route of the 21st-Century Maritime Silk Road: The Case of Fujian Province

Abstract

The 21st-Century Maritime Silk Road is an important part of the “Belt and Road Initiative”. It is intended to carry out broader and deeper regional cooperation and to provide a Chinese solution for reforming the world economic governance model. Fujian Province, as the starting point of the historical Maritime Silk Road, its development status, and the development strategy is of vital importance for the 21st-Century Maritime Silk Road’s sustainable and balanced regional development. Based on the data on the spatial distribution of the headquarters and branches of enterprises in information-related industries, this study, from the perspective of enterprise connection, constructed a firm interlocking network along the south route of the 21st-Century Maritime Silk Road covering nodal cities along the southeast coast of China and cities in Southeast Asian countries. In addition, this paper analyzed the structural features of the firm interlocking network in the region from the perspectives of centrality and connectivity and considered the goal of constructing the core area of the 21st-Century Maritime Silk Road to extract the influencing factors of the firm interlocking network. The core factors influencing the centrality of cities were analyzed and identified by taking Fujian Province as an example. The study found that cities in Fujian Province generally have problems such as unremarkable centrality and weak connectivity in the firm interlocking network in the region and that the three dimensions of policy coordination, financial integration, and technology exchange in the three cities of Fujian Province need to be strengthened despite the achievements made in the three dimensions of facility connectivity, unimpeded trade, and closer people-to-people bonds. Policy suggestions for promoting the centrality of nodal cities in Fujian Province and accelerating the integration into regional city networks were made based on the above research findings and Fujian Province’s development goal of constructing a core area of the 21st-Century Maritime Silk Road.

1. Introduction

The 21st-Century Maritime Silk Road has gradually attracted the attention of researchers since China adopted the “Belt and Road Initiative”, where Belt is short for the “Silk Road Economic Belt”, and Road refers to the “21st-Century Maritime Silk Road”. The “Belt and Road Initiative” is a global sustainable developmental strategy initiated by China and represents innovation in the model of world economic governance and international regional cooperation [1]. Since the issuance of the Vision and Action to Promote the Construction of Silk Road Economic Belt and 21st Century Maritime Silk Road [2] (hereafter referred to as the Vision and Action) by China in March 2015, the Belt and Road has rapidly become a research hotspot. Therefore, researchers have begun to focus on the connotation of the “Belt and Road” construction, the impact of opening up policies, infrastructure cooperation, international trade, and investment cooperation [3,4,5,6,7], but there has been a lack of discussion on the regional impact of cities. According to the evolution pattern of the ancient Silk Road, the Maritime Silk Road is generally divided into three major routes, namely the east route, the south route, and the west route [8], as shown in Figure 1. The south and west routes start from the South China Sea and extend to Southeast Asia, South Asia, West Asia, East Africa, and European countries. Some scholars have noted that as the Belt and Road stretches thousands of kilometers, a few geographic pivots are needed to perform the functions of connecting, linking, and radiating out to surrounding spaces for the efficient construction and smooth operation of the Belt and Road [9]. Therefore, the South China Sea and its surrounding areas are planned as the strategic basis of the Maritime Silk Road due to their important geostrategic position [10]. Southeast Asian countries, as the closest part within the spatial range of the 21st-Century Maritime Silk Road to China, are the most important foothold for the construction of the 21st-Century Maritime Silk Road beyond the Chinese border. Fujian Province and Southeast Asian countries have interlinked geography, close folks, and kindred cultures. As the gateway into both the south route and west route of the 21st-Century Maritime Silk Road, the province is an incomparably important location on the 21st-Century Maritime Silk Road that faces the South China Sea and the whole world. Meanwhile, the province, as the starting point of the historical Maritime Silk Road, has continued its pivotal role in different ways and forms for a millennium. In addition, the Vision and Action advocated for the construction of a core area of the 21st-Century Maritime Silk Road in Fujian Province [2]. Therefore, promoting the construction of a core area of the 21st-Century Maritime Silk Road in Fujian Province, and studying the effects on regional sustainable development is of vital importance for the further establishment of gateways from the South China Sea into the surrounding Southeast Asian countries for cooperation, which would have good effects on China to deepen geostrategic cooperation with Southeast Asian countries and to ensure more sustainable and balanced regional development. However, few existing studies have quantitatively measured the regional influence of Fujian Province on the 21st-Century Maritime Silk Road, and further research on the centrality and connectivity of Fujian Province in the region is required.

In the context of globalization, cities are no longer isolated as they interact more closely with cities and regions beyond national boundaries [11,12]. Globalization has given rise to the theory of space of flows [13] and has promoted research on world city networks [14]. City networks focus on inter-city relationships and underscore inter-city connections and cooperation [15]. According to a literature review, existing studies on city networks both in China and in other countries are centered on population, economy, transportation, information, and innovation [16,17,18,19] and characterize urban connections through empirical paths such as transportation infrastructure networks, business organizations and connections, economic connections, and innovation cooperation, in order to reveal the levels of cities and the features of association networks in the region [20,21,22]. Taylor and some other scholars proposed a theory and method of characterizing the urban network by using the internal connections of advanced producer services enterprises, which has been widely recognized by scholars [23]. They pointed out that enterprises are the “actors” that drive the flow of factors in city networks and that the spatial layout of enterprises and inter-enterprise connection networks are important research perspectives of city networks [24]. City network research based on the perspective of enterprise networks is now the hot spot and frontier of urban geography research [25]. To date, studies on city networks based on enterprise connections are mainly focused on whole industries and relatively mature industries as the sample of research [26,27,28]. City networks based on emerging industries (such as information-related industries and technology hardware and equipment industries) are still relatively rare. Some scholars have studied the network among strategic emerging industries, such as the photovoltaic industry [29,30]. Some scholars have analyzed the corporate networks in manufacturing subsectors, including in the technology hardware and equipment subsectors [31,32]. Additionally, there is room to develop research on the features of city networks from the perspective of emerging industries. Therefore, this study selects information-related industries as representatives of emerging industries to explore the new patterns of regional city networks in the context of the development of information-related industries.

Since the 1990s, the rapid development of digital technology and computer networking has brought the global economy into a new information age [33,34], in which the new manufacturing sector represented by informatization and artificial intelligence and emerging industries such as e-commerce and other similar platforms have become important directions for the future economic development of cities. According to the Provisional Regulations on Statistical Classification of Information-related Industries issued by the National Bureau of Statistics of China, information-related industries mainly refer to the collection of various activities associated with electronic information. These industries, as part of the emerging sectors in the new information age, have gradually become a crucial means for cities to optimize the industrial structure and to promote competitiveness. The research focus of city networks varies at different times. Compared with traditional producer services, information-related industries are becoming the main driving forces to reshape regional spatial patterns due to the better role that they play in guiding the direction of the future economic development of cities. The strategic and emerging nature of these industries, coupled with other characteristics such as the high liquidity of the market structure, will continuously affect the levels of cities and the evolution of the functional structure in regional networks, thus resulting in the reshaping of the city networks [35]. Therefore, the construction of a firm interlocking network of information-related industries at a regional scale and the analysis and identification of the hierarchical structure and regional connections of Fujian Province in the network can, to a large extent, provide a reference for the analysis of the potential of Fujian Province’s nodal cities on the 21st-Century Maritime Silk Road in regional competition.

Based on the above analysis, a firm interlocking network covering 31 cities (including nodal cities along the southeastern coast of China and cities in Southeast Asian countries) on the south route of the 21st-Century Maritime Silk Road was constructed to explore the status quo of the centrality and connectivity of the nodal cities in Fujian Province and the core influencing factors of the firm interlocking network. Furthermore, policy suggestions on promoting the centrality and hierarchical structure of nodal cities in Fujian Province in the regional city network were made based on the core influencing factors. The aim was to promote the construction of a core area of the 21st-Century Maritime Silk Road in Fujian Province and to establish gateways from the South China Sea into the surrounding Southeast Asian countries for cooperation.

2. Materials and Methods

2.1. Research Area

After the proposal of the 21st-Century Maritime Silk Road, China’s Maritime Silk Road diplomacy began to involve Southeast Asian countries, South Asian countries such as India and Pakistan, and Arab League countries in West Asia and North Africa. This study selected 31 cities on the south route of the 21st-Century Maritime Silk Road (including Chinese cities and Southeast Asian cities) as the research areas (

Table 1. Nodal cities on the south route of the 21st-Century Maritime Silk Road.

Geographical Area	Cities
Nodal cities along the southeast coast of China	Shanghai, Tianjin, Ningbo, Guangzhou, Shenzhen, Zhanjiang, Shantou, Qingdao, Yantai, Dalian, Fuzhou, Xiamen, Quanzhou, Haikou, Sanya
Nodal cities in Southeast Asian countries	Manila, Cebu, Hanoi, Ho Chi Minh, Vientiane, Phnom Penh, Yangon, Mandalay, Bangkok, Chonburi, Kuala Lumpur, Selangor, Singapore, Jakarta, Dili, Bandar Seri Begawan

). Specifically, 15 nodal cities along the southeast coast of China were selected according to the Vision and Action, and the 16 Southeast Asian cities selected were national capitals, major cities along the route, and cities with a relatively high level of economic development (Figure 2).

2.2. Data Source

This study selected enterprises above a certain size in information-related industries in the world as the samples [36]. On the Fortune Global 500 2020 list [37], 67 enterprises are in information-related industries. Data on these enterprises were collected from the Internet. From publicly accessible data that were accurate and updated in a timely manner, such as the information on official websites and annual reports, the business scopes of the enterprises, the names and addresses of the headquarters and branches of each enterprise, and the number of branches were obtained. On this basis, we established the criterion that an enterprise must have at least two cross-prefecture-level city-affiliated parts and at least one of them must be located in the research area. A total of 54 eligible enterprises were selected. According to the data on the headquarters and branches, an undirected city network matrix based on the enterprise connection between 31 cities was constructed.

2.3. Research Framework

As shown in Figure 3, the study constructed a regional firm interlocking network based on the data on the headquarters and branches of enterprises in information-related industries. Secondly, the social network analysis method was adopted to analyze network centrality. The centrality of the cities in the region was measured according to their degree centrality, and the cities were classified into various levels by centrality. The interlocking network model was used to analyze the connectivity between cities to measure the connections between cities in the region. Thirdly, the factors influencing the association network were selected to analyze the main factors affecting the centrality of the cities. Lastly, suggestions for optimizing and improving centrality were made based on the structural features and influencing factors of the firm interlocking network.

2.4. Approaches to Network Centrality, Network Connectivity, and the Influencing Factors

2.4.1. Social Network Analysis

Social network analysis is an important research method to investigate the formation of multiple relationships between members. Social networks are mainly divided into two analytical frameworks, ego-centered networks and whole networks, which can reflect both the individual positions in the network structures and the structural features of the overall networks [38]. Network centrality is an important indicator to measure network centrality, from the following three perspectives, degree centrality, betweenness centrality, and closeness centrality. The degree of centrality is the most direct index used to characterize urban centrality [16]. This study used the degree of centrality to measure the power and position of each nodal city in the network, and a larger degree centrality means greater importance and stronger centrality [39]. The formula is as follows [40]:

$$CD\left(i\right)=\sum _{j=1}^n{X}_{ij}$$

(1)

where $CD\left(i\right)$ refers to city $i$’s degree of centrality in the undirected network, and $X_{ij}$ represents the undirected connection value between city $i$ and city $j$ in the undirected network.

2.4.2. The Interlocking Network Model

Proposed by the Globalization and World Cities Research Network (GaWC), the interlocking network model is a research model used to depict city networks from the perspective of enterprise connection. According to the theory of the interlocking network model, enterprises produce city networks, and the network relationships between cities around the world are obtained by summarizing the connections between enterprises in various cities and the quantized data [41]. At present, studies on world city networks using the interlocking network model have achieved remarkable results [42].

Based on the interlocking network model, the connectivity in a city network can be obtained through an inter-city association network constructed based on the connections between enterprises. This study analyzed the connectivity of each city in a network to measure the city’s linking function in the regional space. A city with a higher connectivity can better integrate itself into the firm interlocking network and play a certain role.

The interlocking network model is used to represent a city network based on enterprise connection. Assuming that there are $m$ enterprises in $n$ cities in the region, the connection between city $a$ and city $b$ established by enterprise $j$ is:

$$r_{abj}=v_{aj}\times v_{bj}$$

(2)

where $v_{aj}$ is the service value of enterprise $j$ in city $a$, $v_{bj}$ is the service value of enterprise $j$ in city $b$, and $r_{abj}$ is the connection between city $a$ and city $b$ denoted by enterprise $j$.

Assuming that there are $m$ enterprises in city $a$ and city $b$, the network connectivity between city $a$ and city $b$ is:

$$r_{ab}={{\displaystyle \sum }}_{j=1}^m{r}_{abj}$$

(3)

Assuming that there are $n$ cities in the region, each city has at most $n-1$ such connections, and the overall level of connection of city $a$ in the regional city network can be defined as:

$$r_a={{\displaystyle \sum }}_{i=1}^n{r}_{ai}\left(a \ne i\right)$$

(4)

where $r_{ai}$ is the connection between city $a$ and city $i$, and $r_a$ refers to city $a$’s overall level of connection with all of the other cities in the regional network (i.e., the network connectivity of city $a$).

The importance of each city in the enterprise network is reflected by the service value so that different types of enterprises can be compared. According to GaWC’s value assignment, the headquarters and affiliated parts of enterprises can be divided into six levels, and each level is assigned a value (

Table 2. Service values of different parts of an enterprise.

Enterprise Size	Service Value
no branch	0
agency or office	1
branch or larger agency	2
larger branch	3
regional headquarters	4
enterprise headquarters	5

).

2.4.3. Influencing Factors

The existing studies show that the formation and development of a city network can be affected by enterprises’ multi-location choices [43]. A firm interlocking network that can effectively characterize the city network contains the circulation of factors such as capital, information, and personnel. These factors, arranged and guided by enterprises’ multi-location choices, circulate, gather, and spread in various cities through the association network between enterprises, which drives the evolution of the city networks. Therefore, this study started from the driving factors of the changes in the firm interlocking network from the perspective of the influencing factors of enterprises’ multi-location choices, in order to better grasp the development direction of the regional city network and to optimize the centrality of nodes in the city network.

Regarding the relationship between enterprises and cities, factors such as the policy advantages, infrastructure level, economic scale, resource endowment, and geographical location of cities exert an important influence on enterprises’ multi-location choices. Keeping in mind the many factors that influence enterprises’ multi-location choices, this study adopted the principal component analysis method to determine the core factors. On the one hand, the impact of overlapping information between original indicators is determined through dimension reduction. On the other hand, as the method can solve the limitation that weights for indicators are subjectively determined in the comprehensive evaluation, it has some advantages [44].

3. Results

3.1. Analysis of the Structure of the Firm Interlocking Network

3.1.1. Network Centrality of the Cities

The degree centrality of the 31 cities in the city network was obtained by calculating the service values of different levels of headquarters and affiliated parts of enterprises in the cities in the region, as shown in

Table 3. Degrees of centrality of the nodal cities on the south route of the 21st-Century Maritime Silk Road.

Rank	City	Degree Centrality
1	Shanghai	7665
2	Singapore	6510
3	Shenzhen	4419
4	Guangzhou	4254
5	Manila	3849
6	Kuala Lumpur	3721
7	Bangkok	3459
8	Hanoi	3329
9	Dalian	3278
10	Jakarta	2825
11	Tianjin	2546
12	Qingdao	2340
13	Fuzhou	2277
14	Ho Chi Minh	2063
15	Xiamen	1697
16	Ningbo	1488
17	Selangor	1280
18	Yangon	1118
19	Chonburi	1013
20	Quanzhou	777
21	Phnom Penh	773
22	Haikou	719
23	Yantai	458
24	Shantou	402
25	Sanya	347
26	Zhanjiang	269
27	Vientiane	186
28	Cebu	177
29	Mandalay	168
30	Bandar Seri Begawan	93
31	Deli	70

. In the firm interlocking network of information-related industries, a higher degree centrality means that a city has strong centrality in the network. This study adopted the natural breaks classification method to classify the 31 cities into four levels according to their degrees of centrality: core nodal cities, secondary core nodal cities, backbone cities, and ordinary nodal cities.

The results show that Shanghai and Singapore belong to the first level, and they have the strongest centrality and absolute dominance in the regional city network. As the core nodal cities in the region, they serve as hubs in the city network. Shenzhen, Guangzhou, Manila, Kuala Lumpur, and three other cities belong to the second level and are the secondary core nodal cities in the region. The development gap on this level promotes the circulation of factors and improves the secondary cycle in the network. Jakarta, Tianjin, Fuzhou, Qingdao, and three other cities belong to the third level. There is a small development gap among the cities in this level. According to Table 3, the degree of centrality of the third-level cities is only half that of the first-level cities. However, as the third-level cities can connect with 16 cities, they are the backbone cities that establish enterprise connections in the network. The fourth-level cities include Quanzhou, Haikou, Selangor, and other 12 cities that are still influential in the network despite their low degree centrality. These cities, the ordinary nodal cities in the network, develop based on the development of the core cities.

Fuzhou, Xiamen, and Quanzhou do not have a high centrality in the research areas and are ranked 9th, 16th, and 17th, respectively. Fuzhou’s degree of centrality is less than half that of Shanghai, a core nodal city. Similarly, other cities in Fujian Province do not have a high level of centrality in the firm interlocking network of information-related industries in the region. Therefore, it is urgent that the province has a core nodal city that plays a cohesive and motivating role. This type of city performs the functions of connecting, linking, and radiating out to other areas in the network and undertakes information exchange and transmission while guiding and driving the development of other cities in the province. Therefore, it is imperative for Fujian Province to build a city that can serve as an information network hub in order to consolidate its centrality in the regional network. In addition, the province should make this a key point to consider in the formulation of urban and regional policies in order to take positive measures to bring itself into a larger regional network.

3.1.2. Network Connectivity of the Cities

The service value of a city, as a city-to-city undirected connection value, is used to calculate a city’s degree of centrality and that reflects the city’s importance in the network. The connectivity of a city network focuses on the strength of the connections between cities, which more specifically reflect the connectivity capability of each core nodal city, secondary core nodal city, backbone city, and ordinary nodal city in the network.

Equations (2)–(4) of the interlocking network model were used to calculate the connectivity strength of the city network covering 31 cities based on the connections between enterprises in information-related industries. A 31 × 31 city service value matrix V was obtained by calculating the pairwise connections among the 31 cities. A total of 465 city pairs in the city network form connection flows. The city pairs whose $r_{ab}$ values are in the top 2%, 5%, and 10% in the matrix were spatially visualized, and three levels of connectivity strength were classified, which were plotted into a city connection network diagram (Figure 4).

The $r_{ab}$ of a city pair reflects the strength of the connection between two cities. By analyzing the city pairs ranked in the top 2%, 5%, and 10% in terms of the strength of connection, the following results were found. Firstly, on the south route of the 21st-Century Maritime Silk Road, the transnational connections between the nodal cities along the southeast coast of China and the cities in Southeast Asian countries are stronger than the connections between Chinese cities. Specifically, among the top 10% of city pairs in terms of strength of connection, 52% are between a Chinese city and a Southeast Asian city, 26% are between two Chinese cities, and 22% are between two Southeast Asian cities. From the spatial structure of the city connection network in Figure 4, Shanghai, Singapore, Shenzhen, Guangzhou, and Manila have strong connections with each other, and the connections among the five cities have formed a cyclic connection structure within the region. Secondly, Shanghai dominates in the first level of connection networks. Specifically, in the city network, all of the top three city pairs in terms of strength of connection include Shanghai. Among the top 2% of city pairs in terms of strength of connection, 66% include Shanghai, and among the top 10%, 29% include Shanghai, indicating that Shanghai has a great influence and many partners in the network. Thirdly, Kuala Lumpur is the preferred choice for Chinese cities. Specifically, among the top 10% of city pairs in terms of strength of connection, 25% include Kuala Lumpur, and only 21% include Singapore.

Figure 5 is the city connection network diagram showing the connection flows between all 31 cities. From Figure 5, the structure of the inter-city connection network in the region has a “circular” shape in the outer part and “triangles” in the inner part, and all of the connection flows also show that the transnational connections between the nodal cities along the southeast coast of China and the cities in Southeast Asian countries are stronger than the connections between Chinese cities. Although the connections between the nodal cities along the southeast coast of China and the cities in Southeast Asian counties have become dominant in the regional city network, there are problems of uneven distribution regarding the space for production and cooperation and insufficient cooperation in some areas.

Regarding the connection networks between Chinese cities and Southeast Asian cities, the connection flows between Shanghai (a core nodal city in China) and cities in Southeast Asian countries account for 25.68% of the total flows, and those of Shenzhen and Guangzhou (secondary core nodal cities in China) account for 13.84% and 11.33%, respectively. The total connection flows between secondary core nodal cities in China and Southeast Asian cities equal the total connection flows between Shanghai and Southeast Asian cities. The connection flows between the backbone cities and ordinary nodal cities in the third and fourth levels and cities in Southeast Asian countries account for 49.16% of the total. The connection flows between the three cities (Fuzhou, Xiamen, and Quanzhou) and Southeast Asian cities make up 7.95%, 4.17%, and 2.74% of the total, showing a large gap. It can be seen from Figure 5 that the core nodal cities and secondary core nodal cities (Shanghai, Shenzhen, and Guangzhou) have connections, both strong and weak, with 94% of the nodal cities being in Southeast Asian countries. Fuzhou, Xiamen, and Quanzhou have connections with 88%, 70%, and 52% of the Southeast Asian nodal cities, respectively. In summary, in the firm interlocking network of information-related industries in the region, cities in Fujian Province generally have a weak connectivity, and the situation in Fuzhou is different from that in Xiamen and Quanzhou. Although Fuzhou is more widely connected with the nodal cities in Southeast Asia than Xiamen and Quanzhou, it has weaker connections with the core cities in China. Xiamen and Quanzhou have fewer and weaker connections than other similar cities in China.

3.2. Driving Factors of the Structure of the Firm Interlocking Network

3.2.1. Principal Component Analysis

According to the Vision and Action, countries along the Belt and Road should promote policy coordination, facility connectivity, unimpeded trade, financial integration, and people-to-people bonds (i.e., the five major goals). Based on existing studies [45,46,47,48,49,50,51], the present study added the goal of technology exchange, which is vital for information-related industries and its influencing factors. In total, 6 first-level indicators and 10 s-level indicators were established in the six dimensions, and 12 indicators, including per capita GDP and the amount of foreign investment actually used, were selected as the influencing factors of the firm interlocking network (

Table 4. Influencing factors and explanatory variables of the association network of enterprises.

Item	First-Level Indicators	Second-Level Indicators	Code	Definition	Reference
Five major goals	policy coordination	the openness of the city	X1	the amount of foreign investment actually used (unit: CNY 100 million)	[45]
	facilities connectivity	transport facilities	X2	Port cargo throughput (unit: 10,000 tonnes)	[46]
			X3	the number of civil aviation passengers (unit: 10,000 person–time)	[46]
		communication facilities	X4	total telecommunication business (unit: 100 million yuan)	[46]
	unimpeded trade	economic scale	X5	per capita GDP (CNY)	[47]
		the foundation of industrial development	X6	the number of enterprises in information-related industries	[48]
			X7	the average wage of city employees (CNY)	[48]
		the level of trade and investment	X8	total imports from and exports to Southeast Asian countries (unit: CNY 100 million)	[46]
	financial integration	external financial environment	X9	the total amount of overseas deposits and loans by financial institutions (unit: CNY 100 million)	[46]
	people-to-people bonds	attention index	X10	Mean value of Baidu Search Index	[47]
		Spatial distance	X11	Average straight-line distance to Southeast Asian cities (km)	[47]
New One goal	technology exchange	investment in technology resources	X12	local fiscal expenditure on science and technology (unit: CNY 10,000)	[45]

).

The results of the KMO and Bartlett’s test on the aforementioned influencing factors showed that the value of KMO was 0.798, Bartlett’s sphericity test value was 213.675, and sig = 0.000 < 0.05, indicating that the variables selected in this study were suitable for principal component analysis. The variance contribution rate and the cumulative variance contribution rate of each component were calculated, and the first three components with eigenvalues greater than one were selected, with the cumulative variance contribution rate of the principal components reaching 89.623% (

Table 5. Factors explaining the total variance of original variables.

Component	Initial Eigenvalues
	Total	Percent Variance	Cumulative Percent Variance
1	8.153	67.938	67.938
2	1.450	12.079	80.017
3	1.153	9.605	89.623

). Therefore, the first three principal components obtained could retain the information contained in all the original factors to a great extent.

The first principal component had high loadings on factors X1, X2, X3, X4, X5, X6, X7, X9, X10, and X12, reflecting the comprehensive strength of a city in the six dimensions. Therefore, the first principal component was defined as the “factor of comprehensive strength”. The second principal component had a high loading on factor X8, reflecting the level of trade and investment between a city and Southeast Asian countries. Therefore, the second principal component was defined as the “factor of trade and investment”. The third principal component had a high loading on factor X11, reflecting the spatial distance between a city and Southeast Asian countries. Therefore, the third principal component was defined as the “factor of spatial distance”.

In this study, the proportion of the variance contribution rate of each principal component in the maximum cumulative variance contribution rate was taken as the weight to construct the equation for calculating the centrality development strength in the regional city network. The comprehensive centrality scores of the 15 Chinese nodal cities and the rankings were obtained (

Table 6. The comprehensive centrality scores and the rankings of the Chinese nodal cities.

Rank	City	Score
1	Shanghai	5.03
2	Shenzhen	2.79
3	Guangzhou	1.91
4	Tianjin	0.56
5	Qingdao	0.41
6	Ningbo	0.15
7	Dalian	−0.37
8	Xiamen	−0.39
9	Fuzhou	−0.74
10	Quanzhou	−1.13
11	Yantai	−1.16
12	Haikou	−1.58
13	Zhanjiang	−1.78
14	Sanya	−1.83
15	Shantou	−1.86

). The results showed that the centrality development strength of all the cities, except for Dalian and Ningbo, was roughly consistent with their centrality in the regional firm interlocking network, indicating that cities with stronger comprehensive strength for development generally occupy a more central position in the regional firm interlocking network, while cities with weaker and uneven development are less important in the association network.

3.2.2. Driving Factors of the Network Structure in Fujian Province

The results of the principal component analysis showed that the calculated rankings in terms of the centrality development strengths of the three cities in Fujian Province were roughly consistent with their actual rankings in the firm interlocking network. The actual ranking of Fuzhou was higher than its calculated ranking, and the two rankings of the other two cities were the same. All three cities did not have high centrality in the regional city network. From the perspectives of the three principal component factors (i.e., the factor of comprehensive strength, the factor of trade and investment, and the factor of spatial distance), the factors of comprehensive strength and the factors of trade and investment of the three cities did not score high, which was the main reason for the low centrality of the cities in Fujian Province.

In terms of the factor of comprehensive strength, Fuzhou scored low on influencing factors X1, X6, X9, and X12, which denote the amount of foreign investment actually used, the number of enterprises in information-related industries, the total amount of overseas deposits and loans by financial institutions, and local fiscal expenditure on science and technology, respectively, and Xiamen scored low on influencing factors X1, X9, and X12. Quanzhou scored low on influencing factors X1, X3, X6, X7, X9, and X12, which denote the amount of foreign investment actually used, the number of civil aviation passengers, the number of enterprises in information-related industries, the average wage of city employees, the total amount of overseas deposits and loans by financial institutions, and local fiscal expenditure on science and technology, respectively. In terms of factors influencing X8 (i.e., total imports from and exports to Southeast Asian countries) corresponding to the factor of trade and investment, Xiamen scored relatively high, and Fuzhou and Quanzhou scored low. In terms of the factor of spatial distance, the three cities scored relatively high.

According to this study conducted based on the construction of the Belt and Road and the five major goals proposed in the Vision and Action, the three cities in Fujian Province have not performed well on all the five major goals despite their efforts to address them. The three cities still have some shortcomings. In terms of policy coordination, the countries and regions along the Maritime Silk Road need to develop comprehensive cooperation, strengthen cooperation and exchanges between governments, and pioneer new policies. In terms of facility connectivity, remarkable achievements have been made in the construction of ports and ship routes. On this basis, Quanzhou can further develop its passenger and cargo transportation to help Fujian Province fully play into its unique geographical advantages and to become a hub of transport connectivity on the Maritime Silk Road. In terms of unimpeded trade, based on their good trade and cooperative relations, Fujian Province and Southeast Asian countries can further complement each other’s economies and explore the huge potential for economic and trade cooperation. In terms of financial integration, the cities in Fujian Province lack foreign exchange reserves despite their abundant private capital. In terms of people-to-people bonds, the three cities in Fujian Province have significant advantages. As the starting point of the Maritime Silk Road, Fujian Province has a solid foundation in terms of social relations, business networks, as well as language and culture, which can help to further promote the friendly relationships between the three cities and Southeast Asian countries in terms of cultural exchange, academic exchange, talent exchange, and media cooperation. Lastly, although technology exchange is not one of the five major goals, interdependence between cities and countries is gradually shifting from trade and investment to technological cooperation in the context of scientific and technological globalization. Therefore, the cities in Fujian Province need to further enhance their technological innovation capabilities and boost inter-city technological cooperation and exchange.

4. Discussion and Conclusions

With the constant development of information technology in recent years, the development of cities has developed beyond their own boundaries and has had closer connections and interaction with cities and regions in other countries, with a diversity of global or regional city networks being formed. The effective identification of the city network structure is related to the sustainable development of the region. Therefore, we based our data on the headquarters and branches of enterprises in information-related industries, and this study analyzed the structural features of the firm interlocking network in the nodal cities along the southeast coast of China and in Southeast Asian countries on the 21st-Century Maritime Silk Road, identifying the core influencing factors of the network’s centrality. The following conclusions have been drawn:

• According to the spatial structures of the inter-city connection network already formed in the spatial range of the south route of the 21st-Century Maritime Silk Road, a circular connection structure has been formed within the region, and it gradually drives other cities in the region to further integrate into the development of the city network. Different levels of cities play their role in the connections in the network. The network is led by the core nodal cities, is driven by the secondary core nodal cities, and is supported by the backbone cities and the ordinary core nodal cities;
• From the perspective of the centrality of the city network, Shanghai, Guangzhou, and Shenzhen, China, are the basis for China “going global” in relation to the construction of the Silk Road in Southeast Asia. Cities such as Singapore, Manila, and Kuala Lumpur represent the foothold beyond the Chinese borders this study has focused on. From the perspective of the connectivity of the city network, a regional city network has been formed based on inter-regionally interconnected cities in China and Southeast Asian countries. Meanwhile, the connectivity between the nodal cities in the Chinese section and the ones in Southeast Asian countries is better than that between Chinese nodal cities;
• The cities in Fujian Provinces are not playing an important role in the regional city network. Fujian and Xiamen belong to the third level of cities, namely the backbone cities, and Quanzhou belongs to the fourth level and is an ordinary nodal city. None of these cities are playing a dominating or driving role. Therefore, it is urgent to improve their centrality in the regional city network;
• The connectivity of the cities in Fujian Province is relatively low in the regional city network. Although Fuzhou and Xiamen are widely connected with other cities, the strength of the connections is much weaker than that of the core nodal cities and secondary core nodal cities. Quanzhou does not have connections with most of the cities in the region, resulting in it having a low connectivity in the regional network;
• The three dimensions of policy coordination, financial integration, and technology exchange in the three cities of Fujian Province need to be strengthened despite the achievements made in the three dimensions of facility connectivity, unimpeded trade, and closer people-to-people bonds. The low level of investment in technology resources in all three cities reflects, to a certain extent, the neglect of science and technology development in Fujian Province. Therefore, Fujian Province should try to realize rapid development and catch up with other cities in the regional city network by investing more in science and technology.

Based on the above conclusions, the following suggestions have been made regarding the optimization of the centrality and connectivity of the cities in Fujian Province in the regional city network that further support Fujian’s construction of a core area of the 21st-Century Maritime Silk Road.

In terms of policy coordination, Fujian Province needs to actively enhance exchange and communication with the governments of Southeast Asian countries, establish a regular exchange mechanism, and promote pragmatic cooperation between governments. A favorable institutional environment should be provided to attract high-quality enterprises and to strengthen the exchange of factors with international cities.

In terms of financial integration, Fujian Province should promote the construction of economic and trade cooperation platforms and should explore an open economic system. The China (Fujian) Pilot Free Trade Zone can serve as an engine to drive foreign investment into sectors such as high-tech industries. In addition, the province can support enterprises to go public and raise capital in countries and to regions along the Maritime Silk Road, guiding and supporting qualified enterprises to build economic and trade cooperation zones outside of China.

In terms of technology exchange, Fujian Province should focus on strengthening technological cooperation with countries and regions along the route, enhancing its strength, and accelerating the construction of influential science and technology innovation centers in the region. Thus, Fujian Province should further promote the gathering, transfer, and diffusion of scientific and technological innovations among countries and regions along the Maritime Silk Road and should strengthen the connectivity of Fujian Province in the future regional spatial pattern reshaped by information-related industries.

In terms of facility connectivity, unimpeded trade, and closer people-to-people bonds, Fujian Province should maintain and consolidate its advantages. In addition, the province needs to continuously deepen the core area’s role in leading, gathering, and affecting other areas in various fields such as connectivity, economic and trade cooperation, and people-to-people exchanges.