The Research on the Impact of Regional Trade Network Relationships on Value Chain Resilience in China’s Service Industry

3.1 Method for Environmental Analysis of Regional Trade Networks

3.2 Individual Characterization Methods Under Regional Trade Networks

Centrality is a quantitative analysis of an individual’s power, which can clarify what kind of power an individual has or what kind of central position it occupies in the network. Therefore, this article combines degree centrality, betweenness centrality, and closeness centrality proposed by (Freeman, 1979) to quantify the individual characteristics of the Chinese service industry in the network.

3.2.3 Closeness Centrality

Closeness centrality indicator represents the extent to which an actor is not controlled by other actors. That is, the closer a point is to other points, the easier it is for that point to transmit information, the more likely it is to be at the center of the network, and the less dependent it is on others. A lower degree of closeness centrality means that the sum of the shortcut distances for an industry in a country to trade with other economies is smaller; i.e., the closer the trade “distance” is, the lower the trade costs are, the more reliable the trade information is, and the easier it is to establish a dominant position in the value chain. The formula is as follows:

ccen it = ∑ j = 1 n d ij .

In this formula, d ij represents the shortcut distance between points i and j.

To make the overall methods and ideas clearer, a flowchart (Figure 1) has been used.

Figure 1

Article application methods and ideas.

4.1 North American Regional Trade Network Relationships

Since 2007, the density of the North American regional trade network has shown a trapezoidal fluctuation, and there were obvious inflection points in 2011, 2014, and 2019 (the details are shown in Figure 2). There was a substantial increase in the density of the North American regional trade network relationship in 2011, from 0.583 to 0.667, and the reason for that was the gradual recovery of the economies of the countries after the U.S. subprime crisis in 2007, and the trade relationships between the countries became closer; density fell from 0.667 to 0.583 in 2014 owing to the trade disputes of the countries in the region and the impact of political factors; after the signing of the USMCA, trade between the United States, Mexico and Canada became closer so that the density rebounded to 0.667 in 2019.

Figure 2

Density and connectedness. Source: Calculated from data in the input–output tables of the ADB database.

The North American regional network connectedness remains perfectly correlated between 2007 and 2021; i.e., the network correlations are all 1 (Figure 2). The main reason for this is that the North American regional trade network contains relatively few countries and maintains a stable trade in services between all countries during the years examined, resulting in a relatively high degree of stability. Under such connectedness, China, the United States, Mexico, and Canada are always the core countries in the whole trade network (Figure 3a), and the strength of countries in the North American regional trade network is relatively balanced.

Figure 3

Core-Periphery structure. Source: Calculated from data in the input–output tables of the ADB database: (a) North America, (b) the Belt and Road, and (c) EU.

4.2 European Union (EU) Regional Trade Network Relationships

Both EU regional trade density and connectedness have shown fluctuating upward trends since 2007, and both measures peaked in 2019, but COVID-19 has hampered trade associations between economies to some extent, resulting in a decline in density and connectedness in 2020 (Figure 2).

In terms of the structure of the trade network, the number of countries in the core and semi-edge regions of the EU regional trade network increases between 2007 and 2021 (Figure 3c). In 2007, Germany, France, Spain, Italy, the Netherlands, Belgium, Luxembourg, Ireland, China, and Austria were in the core tier, while Poland and Portugal were in the semi-edge tier. Subsequently, there has been a continuous influx of countries into the core and semi-edge layers. Until 2021, based on the previous network structure, Slovakia, Greece, and Finland were in the semi-edge layer for the first time. It can be seen that not only the more stable developed countries have always been in the core of the trade network, but in recent years developing countries have also gradually entered the core regions, with the number of countries increasing from one to three. Meanwhile, the number of countries in the edge tier has been decreasing year by year, from 16 in 2007 to 8 in 2021.

In terms of the subregional structure of the trade network, the EU’s regional trade network not only has the obvious characteristics of a “small world,” but also consists of subgroups of developed countries with similar levels of economic development, and has strong geographical proximity (Table 1). Germany and France were the most cohesive core countries in 2007, which led to the formation of six cohesive subgroups.

The number of cohesive subgroups decreased and then increased as a result of the fine-tuning of the subregional structure between 2011 and 2015 due to adjustments in the trade strategies of some countries. However, there is a structural change in the cohesive subgroups’ structure in 2021, where not only did the number grow to eight, but also the Netherlands jumped up to become the third cohesive core country, on top of Germany and France. In addition, developing countries gradually entered the cohesive subgroup of developed countries, where the cohesive subgroup formed by China and EU member states has accounted for almost half of the EU cohesive subgroups; i.e., China’s cohesion in the EU’s regional trade network has become stronger.

4.3 “The Belt and Road” Regional Trade Network Relationships

Although the density of trade networks in “the Belt and Road” region climbed from 0.011 in 2007 to 0.020 in 2021, the overall change was small (Figure 2). The same applied to connectedness, which rose from 0.022 in 2007 to 0.072 in 2021 (Figure 2). Although the stability of the trade network is gradually increasing, the indicator values show that the regional trade network of “the Belt and Road” is still mainly dependent on individual core countries, and when the trade of the core countries fluctuates, the entire network structure will face a huge shock. Therein, the countries at the core of the “the Belt and Road” regional trade network are mainly Singapore, South Korea, Malaysia, and China. With the further implementation of “the Belt and Road” Initiative, countries such as Russia, the Philippines, Cyprus, Italy, Luxembourg, and Vietnam have gradually moved from the edge to the core (as shown in Figure 3b).

The development trend of cohesive subgroups shows that there was only one cohesive subgroup in “the Belt and Road” regional trade network between 2007 and 2015, namely the subregional trade network comprising China, South Korea, and Singapore (Table 1). By 2021, the number of cohesive subgroups increased to three, with China and Singapore as the core cohesive countries. The newly formed cohesive subgroups may be influenced by the signing and deepening implementation of RCEP. Moreover, China and Southeast Asian countries have certain advantages of geographical distance. Frequent exchanges and cooperation in various aspects, such as politics, culture, and economy, also contribute to the formation of subregional networks.

By comparing the forms of the three regional trade networks, it can be seen that the North American regional trade network, both in terms of density and connectedness, outperforms the other two regional trade networks and it has the closest associated countries and is highly stable. The EU regional trade network is the second. Although the advantages of “the Belt and Road” regional trade network are not obvious in terms of density and connectedness, China is a central node country in the core region of the network. Compared with North America and the EU regional trade network, China’s dominant position in “the Belt and Road” regional trade network is more prominent and its cohesive capacity is superior.

5.1 Degree Centrality

In the North American regional trade network, the degree centrality of China’s service industry has remained stable, ranking second with 0.667 (as shown in Table 2). Only when China and the United States faced a fierce trade conflict in 2018, as well as the North American region renewed the free trade agreement (FTA), signed the USMCA to modify and improve the rules of trade in services, increased restrictions and protective measures for non-member states, suppressed the degree centrality of China’s service industry in that year to a certain extent, down to 0.333. But in 2019, it quickly recovered to the previous level. China’s service industry’s degree centrality has a similar development trend in the EU and “the Belt and Road” regional trade network, both of which show fluctuating growth trends (as shown in Table 2). Herein, the fluctuation is more obvious in the EU regional trade network.

5.2 Betweenness Centrality

During the study period, the degree of betweenness centrality of China’s service industry was relatively low in North America and the EU regional trade network. This indicates that in the above regional trade networks, the trading capacity of developed countries is significantly better than that of China, and the role of China’s service industry as a “bridge” in these networks is exiguous. On the contrary, in “the Belt and Road” regional trade network, the betweenness centrality of China’s service industry has obvious advantages, ranking first in 2019 and reaching a peak of 37.833 in 2021. China’s service industry plays an important “intermediary” role in “the Belt and Road” regional trade network and is an important hub for other economies to trade in services.

5.3 Closeness Centrality

In the North American regional trade network, the closeness centrality of China’s service industry is relatively stable, remaining at 0.833. In the EU regional trade network, the closeness centrality of China’s service industry is on a downward trend. This indicates that the trade “distance” where China’s service industry trades with EU economies is gradually decreasing. This is partly due to the China–Europe Railway Express, which has shortened the trade distance between China and the EU, creating a new pattern of international transport. On the contrary, China’s service industry not only shows an increasing trend in closeness centrality in “the Belt and Road” regional trade network but also has a value over 100, much higher than the other two trade networks. On the one hand, this is due to the influence of infrastructure construction. On the other hand, in the context of the rapid development of the digital economy, the digital gap between China and the countries along “the Belt and Road” has also extended the “trade distance.”

6.1 Spillover Effects of Degree Centrality

Degree centrality reflects a country’s ability to establish close trade relations with other economies in terms of trade breadth. According to the theory of risk diversification put forward by R.E. Caves and G.V. Stevens in the middle of the twentieth century, when investment and trade in the host country do not necessarily yield favorable profit gains, the risk should be diversified according to the geographic location, i.e., the diversified development of trade is the result of risk diversification. A high degree centrality on the one hand increases a country’s responsiveness to shocks and risks. A higher degree centrality means that a country expands its scope of trading partners, at which point the spread of uncertainty shocks will advance sequentially among trading partners, giving a country sufficient time to anticipate the effects of shocks and actively respond to them. Thus, it improves the overall value chain resilience from the perspective of improving operational value chain resilience. On the other hand, degree centrality reflects a country’s ability to establish direct trade relations with other economies. Therefore, when a country’s industrial value chain encounters risky shocks, it can flexibly adjust its trade strategy through direct trade associations under a high degree centrality, and the effect of the change in trade strategy can be quickly realized. That is, a high level of degree centrality enhances the structural value chain resilience through positive response to risky shocks and rapid recovery, thus strengthening the overall industrial value chain resilience.

Hypothesis 1: Degree centrality is positively related to the resilience of China’s service industry value chain.

6.2 Spillover Effects of Betweenness Centrality

Rauch (2001) states that firms can utilize trade relationship networks for information sharing, which in turn reduces the cost of searching for market information and improves the efficiency of information dissemination. The higher the degree of betweenness centrality, the more accurate and timely the crucial country in the value chain will be in grasping the information on the direction of the shock, the strength of the shock, and the effect of the response to the shock. Hence, the country can shorten its response time to the shock and even do sufficient early warning and response before the shock arrives. Moreover, according to the theory of value chain governance (Gereffi et al., 2005), global production networks correspond to different power relations, with dominant subjects having the ability to exercise direct power over subordinate subjects. Concurrently, the theory of social network centrality posits that the centrality of the value-added trade network, as a configuration of trade relations and flows, reflects the host country’s role as a “hub” at a deeper level. It is manifested in the ability to coordinate and control network relations, factor flows, greater access to resources such as knowledge and technology, and more efficient allocation, as well as improvements in institutional quality and the business environment. A high degree of betweenness centrality means that core node countries can dominantly lead the value chain by coordinating upstream and downstream trade activities and allocating resource flows, thus achieving active control over the negative impacts of shocks and enhancing the stability of the value chain.

Finally, according to the dynamic optimization framework of “three circulations” constructed by Wang et al. (2021), China is in the global “upper circulation” led by developed economies, the regional “middle circulation” composed of developing economies, and the domestic value “lower circulation” composed of eastern, middle and western regions. In the “upper circulation,” developed countries export high-quality essential factors of production to China, while China exports intermediate and final consumption products to developed countries. With the implementation of “the Belt and Road” initiative, the signing and implementation of RCEP, etc., relying on the regional trade network relationship in the “upper circulation,” China has the nature of a “chain master,” gradually diverted from the “middle circulation” formed by developing countries. So it can realize the two-way export of resource factors of the countries in the region based on the differences in endowment advantages. The “dual circulation” of value provides useful insights for our country to get rid of the low-end lock of the global value chain. However, only the “upper circulation” and “lower circulation” are not enough to support the upgrading and development of China’s industries. The construction of the “lower circulation” of value division of labor between East, Middle, West, and inland-coastal regions of China is the only way to further improve the value chain system and enhance the endogenous growth momentum of China’s industries. The regional value of the “middle circulation” is the hub that connects the “upper circulation” and the “lower circulation.” Regional cooperation is used as a means of connecting domestic and global value chains in terms of taxation and trade facilitation, to help industries climb up the value and innovation chains based on the efficient operation of the value “three circulations” system.

Therefore, countries with a high degree of betweenness centrality can efficiently connect global and domestic value chains. The smooth flow of the dual circulation can accelerate the developmental process of domestic and foreign trade integration. It is conducive to the formation of a strong domestic market and at the same time helps foreign trade enterprises to improve their ability to resist risks.

Hypothesis 2: Betweenness centrality is positively related to the resilience of China’s service industry value chain.

6.3 Spillover Effects of Closeness Centrality

According to the speculation of the iceberg trade theory, the lower the cost of inter-firm trade, the lower the cost of obtaining intermediate goods. It is conducive to enterprises improving the quality of export products and climbing up the position of the global value chain. From the perspective of technological innovation, closeness centrality reduces trade costs by shortening trade “distance,” and reserves funds for breaking through technological bottlenecks and achieving technological innovation. Technological innovation is the core element that enables industrial value chains to recover quickly from shocks, and low closeness centrality enhances the overall service value chain resilience by strengthening the technological value chain resilience. From the perspective of influence, low closeness centrality makes it easier for products to enter other countries’ markets. It establishes a good reputation among other economies through the diffusion effect of large volume and high quality, forming and strengthening the influence and competitiveness of independent brands. Conversely, the branding effect will once again promote the occupancy rate and competitiveness of products in the international market, thus forming a virtuous circle. Under the guidance of consumer preference theory, the reduction of product substitution rate can fundamentally reduce or eliminate the negative effect of shocks on the value chain resilience.

Hypothesis 3: Closeness centrality is negatively related to the resilience of China’s service industry value chain.

6.4 Stack Effects of Centrality and Density

According to the trade network theory, external factor shocks can have indirect transmission effects on other points through the trade network channel. Therefore, the above spillover effects of centrality on value chain resilience will further produce a scaling effect when superimposed density. On the one hand, network density has a scaling effect. If the trade relationship in the trade network is a virtuous interaction, the higher the density, the bigger the amplification effect. That is, if the countries in the trade network are each other’s core trading partners, and the establishment of trade relations is centered on a “win-win,” then the countries will adjust their trade strategies from a global perspective when uncertainties hit the value chain. Then the negative impacts of the shocks will not only not be spread dramatically because of the high density, but will also be weakened rapidly because of the coordinated and unified countermeasures taken by the countries. Meanwhile, if the relationships between countries in the trade network are established with a long-term interest and have strong inclusiveness, the network of trade relations will be more resilient, and the possibility of external shocks interrupting the trade relations will be sharply reduced. The weakening of the volatility of the trade network relationship will make it possible for it to steadily and consistently transmit the positive spillover effects of the centrality of the network, which will feedback doubly and positively to the value chain resilience. Conversely, density will have a contractionary effect, weakening the spillover of the positive effects of centrality on increasing value chain resilience.

7.1 Econometric Modeling and Variable Descriptions

According to the theory of value chain governance (Fernandez-Stark & Gereffi, 2019; Schrage & Gilbert, 2019), we extend the model by incorporating relational factors under regional service trade agreement constraints. To verify the hypotheses presented above, the following econometric model was constructed:

In the model, t means time, and Ln GVC PL t represents the value chain resilience of China’s service industry in year t, which is logarithmic. Ln DE t , Ln IN t , and Ln CL t respectively show the degree centrality, betweenness centrality, and closeness centrality of the Chinese service industry in year t after the logarithm transformation. Ln Ctrol t is the set of all control variables, done logarithmically. This article is based on the value chain theory, referring to the traditional influencing factors generally recognized by scholars, and combining the development context of the times, incorporating emerging elements into the model together as control variables (Geng et al., 2022; Ma et al., 2016; Xu et al., 2018). Therefore, the control variables specifically include value added in the service industry, employment in the service industry with a bachelor’s degree or higher, wages in the service industry, fixed asset investment, and foreign investment in the service industry. εt is the random error term, dt signifies Time Fixed Effects, It means Industry Fixed Effects. Particularly, all subsectors of the service industry are included in the Industry Fixed Effects. Considering the continuity of the data, the period of the data is from 2007 to 2021. At the same time, China’s National Economic Industry Classification (GB/T 4754-2017) was integrated with the industry classification of the latest World Input–Output Table published by the ADB database, and eight service industry segments were defined, which are as follows: trade services; hotels and restaurants; transport services; telecommunications; financial intermediation; real estate, renting, and business activities; public administration and defense; education, health, and social work.

Global value chain resilience has multiple dimensions. From a life-cycle perspective of value chain resilience, on the one hand, resilience centers on rebounding from shocks. On the other hand, after a shock, the quality of the value chain itself is also an important aspect of resilience as external influences become more normalized. Therefore, this article uses stability and security as entry points, indicating value chain resilience. This article refers to the article drawn by John et al. (2008) to measure the stability of value chains in terms of the magnitude of fluctuations in the length of the value chain, and the formulas are as follows:

In the above formulas, plv _ gv c ijt is the GVC length of industry j in country i in year t, plv _ gv c t ¯ is the average value chain length of all industries, which is defined as system risk, and pl ijt is the GVC length of industry j in country i in year t after removing risk. T is the observation window of opportunity, and Vol _ pl ijt is used to measure the stability of the value chain of industry j of service industry in country i in year t, and the smaller the value of this indicator, the lower the volatility of the value chain and the greater its stability.

The security of the value chain is then measured by the gap in the length of the value chain with reference to Yang and Zheng (2023), and the following is the formula:

In the formula, max ⁡ ( plv _ gv c jt ) is the maximum value chain length for industry j in year t. gap _ pl ijt is the gap in GVC length for industry j in country i in year t, and the larger the value, the wider the gap between countries’ GVC and the trading powers, the fewer the segments in which countries participate in GVC, and the less secure the GVC.

As evidenced by the calculations (Table 3), the stability of China’s service industry has been relatively stable over the years. When security was hit by the epidemic, the indicator experienced a slight decline. However, it rebounded steadily in the following year. In terms of development trends (Figure 4), there was no significant change or improvement in China’s service industry value chain resilience during the period. Specifically, the less secure industries are “transport services” and “financial intermediation,” which, in contrast, are relatively more stable, and the less stable industries are “hotels and restaurants” and “real estate, renting, and business activities,” respectively.

Figure 4

China’s Service Industry Value Chain Resilience over the years. Source: Calculated from data in the input–output tables of the ADB database.

Given that the indicators selected in this article all measure value chain resilience inversely, i.e., the larger the value of the indicator, the less resilient the value chain. Therefore, the hypotheses above are adjusted as follows:

Hypothesis 1: Degree centrality is negatively related to the resilience of China’s service industry value chain.

Hypothesis 2: Betweenness centrality is negatively related to the resilience of China’s service industry value chain.

Hypothesis 3: Closeness centrality is positively related to the resilience of China’s service industry value chain.

Table 4 below shows the selection of variables in the regression equation:

7.2 Analysis of Empirical Results

8.1 Conclusions

To clarify the development of regional trade networks in the service industry, this article conducts detailed research on the three major regional trade networks of North America, the EU, and “the Belt and Road,” as well as individual characteristics of China’s service industry therein, based on RTAs. The findings indicate that the cohesion of China’s service industry within the regional trade networks has been increasingly intensifying on an annual basis, and regional trade subgroups with China as the core have emerged increasingly. This underscores the centrality of China’s service industry in the regional trade network. Thereby, this article further explores the impact of industry centrality on value chain resilience. The results indicate that an increase in the degree and betweenness of China’s service industry significantly enhances value chain resilience. Conversely, the positive impact of closeness on value chain resilience is only evident in “the Belt and Road” regional trade network.

This article builds on existing literature in several ways. It expands the research perspective, combining the development background of the times and establishing the influence framework between the regional trade network relationship and the value chain resilience. It also accurately measures China’s service industry resilience from the meso perspective. The study finds that although external shocks negatively affect the service industry value chain resilience, the impact is relatively small. The outstanding challenge of China’s service industry value chain resilience is the low degree, which has no significant improvement over the years. On this basis, from the perspective of optimizing trade network relations, we conclude that the construction of high-quality centrality in China’s service industry will further strengthen the optimization effect on value chain resilience through balanced regional trade network relationships. On the contrary, the degree of polarization in trade network relationships will exacerbate the weakening motivation of centrality for value chain resilience.

8.2 Countermeasures Suggestions

To improve the resilience of China’s service industry value chain, it is essential to optimize the regional trade network relationships based on RTAs. China’s service industry should, on the one hand, continue to optimize its trade relations with developed economies, guiding and encouraging Chinese service industry enterprises to actively participate in the regional trade networks formed by developed countries. Cultivate their core competitiveness and narrow the gap with developed countries while fully enjoying the technology spillover effect. China’s service industry needs to take the initiative to form influence, cohesion, and voice in high-level regional trade networks, dispersing the impact of uncertainties on the value chain. At the same time, it should actively cultivate the ability to stabilize the value chain and control the negative impacts on the value chain.

On the other hand, “the Belt and Road” initiative, RCEP and other carriers, the signing of FTAs, and the continuous deepening of development offer the potential for China’s service industry to further improve the trade “distance” between China’s service industry and other economies. It can achieved by lowering trade barriers and building a mutually beneficial and “win-win” trade environment, to intensify the closeness of China’s service industry to the value chain, and to build high-quality trade platforms for shaping the resilience of the value chain. Meanwhile, the government should continue to promote the opening up of the system. As a core node country in “the Belt and Road” regional trade network, China’s service industry should make full use of its core position to coordinate the trans-regional cooperation mechanism, and actively explore docking of international high-standard economic and trade rules, such as CPTPP, DEPA, etc. In this way, it can strengthen the diversified intra-regional trade associations, enhance the degree centrality of China’s service industry in the value chain, and create a stable trade environment for cultivating the value chain resilience from the perspective of strengthening the overall intensity of trade network relations. In addition, implementing policies to promote the cultivation of leading enterprises, and then strengthen the betweenness centrality of China’s services industry in the value chain through leading enterprises, thereby properly guiding capital into medium and high technology industries, optimizing trade network relationships, making full use of the effects of technological spillovers and industrial transfers, activating research and innovation activities, and fostering core technological capabilities for optimizing the value chain resilience.

8.3 Future Studies

While the strengthening of regional trade network relations serves to reinforce China’s service industry value chain resilience, it is also important to consider the potential negative effects that may arise from the deepening development of the digital economy. In the absence of harmonized multilateral trading system regulations to address the unique challenges posed by digital trade, there is a risk of adverse consequences, including market fragmentation, increased trade costs, and a reduction in trade efficiency. These factors could impede the further upgrading of the industry’s value chain. Furthermore, from a long-term perspective, core countries that maintain a high centrality sustainably tend to occupy the middle and high end of the value chain, where they are in a dominant position, thereby creating long-term trade surpluses. The persistence of trade gains will inevitably trigger tensions in trade relations, which in turn will precipitate significant trade frictions and increased protectionist momentum. This will ultimately result in the weakening of the relatively resilient value chains that have been established. Once a negative effect is observed in trade network relationships, it will be further diffused through the transmission effect of trade network relationships, which will exacerbate the extent of undermining the value chain resilience. Therefore, future research on the impact of trade network relationships on value chain resilience should focus on the impact relationship curve between the two. To this end, it is crucial to find and clarify the inflection point of the curve and provide countermeasure suggestions for the sustainable upgrading of China’s service industry value chain resilience based on the inflection point, i.e., how to sustainably release the positive effects and delay or reduce the negative impacts.