Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Competition among cities for export trade brings diversification: The experience of China’s urban export trade development

  • Enkang Li,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Visualization, Writing – original draft, Writing – review & editing

    Affiliation School of Architectural Engineering, Jinling Institute of Technology, Nanjing, Jiangsu, China

  • Yingyi Ma ,

    Roles Funding acquisition, Project administration, Resources, Supervision

    Affiliation School of Architectural Engineering, Jinling Institute of Technology, Nanjing, Jiangsu, China

  • Yi Wang,

    Roles Supervision, Validation, Visualization

    Affiliation School of Economics and Management, Nanjing University of Science and Technology, Nanjing, Jiangsu, China

  • Yu Chen,

    Roles Project administration, Resources, Visualization

    Affiliation School of Geography, Nanjing Normal University, Nanjing, Jiangsu, China

  • Bo Niu

    Roles Investigation, Resources, Software

    Affiliation School of Geography, Nanjing Normal University, Nanjing, Jiangsu, China


Market competition is considered to have a significant impact on product diversification, but related issues are rarely discussed on a city scale. To analyze the diversification of export commodities and export market of 270 Chinese cities, this study uses data from 2000 to 2017 based on the commodity concentration index, market concentration index, export similarity index, spatial stratified heterogeneity, and the Almon lag model. The study’s findings are: (1) The diversification of exports in most Chinese cities increased, which was higher in more developed cities in the southeast than in less developed ones in the northwest. With time, the degree of commodity and market diversification in some developed southeast Chinese cities (e.g., Shanghai) declined. This indicates the difference in and complexity of the evolution of export development in hundreds of Chinese cities between 2000 and 2017. (2) The export competition between cities became increasingly fierce, which effectively urged most of them to improve their export diversification levels. Facing increasing export competition pressure, 81.11% of the total number of cities will enhance the diversification of export commodities to cope with challenges posed by other cities. But only 56.67% will further expand the export market when the pressure of export competition increases. The biggest contribution of this study is the finding that for most Chinese cities, when export competition from other domestic cities increases, increasing diversification of products becomes a wise choice. However, the influence of competition on the diversification degree of the city’s export market is relatively weak. This study provides not only a new perspective for existing research on urban export trade, but also valuable information for cities to form a more profitable and robust trading system.


Since joining the World Trade Organization (WTO), China’s trade has developed at an amazing speed [1, 2]. It has become the world’s factory [3], providing rich raw materials and commodities to other countries, and being a strong link in the global value chain (GVC) [4]. This development is closely related to the dynamic growth of export trade in hundreds of Chinese cities, especially in eastern coastal areas. Because cities have varying natural resources and geographical locations, their export commodities have different structures and flow to various markets. For example, a high proportion of exports from Dalian City and Qingdao City in North China flows to Japan and South Korea [5], and from Shenzhen City and Guangzhou City in South China to Hong Kong [6, 7].

Existing studies on the export trade of Chinese cities have mainly focused on two aspects. One is the impact of exports on urban development, including economic growth [8, 9], urbanization [10], issues around left-behind children [11], urban residents’ happiness [12], urban pollution, and environmental protection [13, 14]. The other aspect is related to factors affecting a city’s export trade, which includes environmental regulations [15], financial development and constraints [16], and air pollution [17].

A few studies have examined the diversification of urban export products and commodity, and the destination market structures of the export trade of Chinese cities. The continuous adjustments in the pace of China’s economic transformation should advance the optimization of urban export trade that eliminates dependence on labor-intensive industries. Improving the diversification of commodities and actively exploring overseas markets are important measures for Chinese cities to enhance their export levels. China has many cities, among which export products overlap and compete widely. Studies have shown that the increase of export competitive pressure will affect the change of export product types and combinations [18, 19]. However, whether this kind of competition affects the commodity and market diversification of urban exports has not been addressed by relevant studies. Therefore, this study focuses on the following scientific question: as an “economic man” with a natural desire for economic growth, how does a city react in terms of commodity diversification and market diversification when it competes with other cities in export trade?

The framework of this study is divided into two parts (Fig 1). First, we calculated the degree of commodity and market diversification of Chinese cities’ exports to the world. Second, we calculated the pressure of export competition faced by each Chinese city and then analyzed how it affects export diversification. From the urban scale, this paper analyzes how different regions will change their export structure and destination when facing the pressure of export competition. The contribution of this research lies in the quantitative measurement of the different responses of Chinese cities to export competition pressure using the econometric model.

Literature review

Cities are prominent worldwide because of the concentration of social and economic activities within them [20]. Research on export trade based on the analysis of countries has begun to turn its focus to cities. As Kalafsky argued, export dynamics should be viewed increasingly from an urban frame [21]. One of the pillars of the framework of trade research at the urban scale is the in-depth analysis of the factors affecting urban export.

The improvement of transportation infrastructure will undoubtedly bring opportunities for rapid development of a city’s export [22]. The construction of expressways can effectively promote the export volume of a city. For border cities, there is a closer relationship between export-led growth and transportation accessibility [23]. Contrastingly, the level of innovation [24], the stability of the financial system [25], the environmental policies made by the government [15], and the tax incentives provided to enterprises [26] will have a great impact on the export development of a city. There are also occasional events that can affect a city’s exports, such as a financial crisis [27] and health-related emergencies such as the COVID-19 pandemic [28].

Despite the remarkable findings of the above studies, when discussing the different factors affecting a city’s exports, researchers have mostly focused on how the total value of the city’s exports was affected. Total value is indeed one of the most important indicators in export trade research, and it is very easy to obtain such data. However, there is another very important aspect of export trade analysis that has received little attention: the discussion of the structure of exports and the spatial distribution of the destination countries receiving a city’s exports. In fact, most studies on export product structure and destination distribution are based on the country. Many scholars have discussed the level of export diversification [2933] and what factors lead to diversification [3437]. However, export diversification at the urban scale is still scarce, and only a few scholars such as He et al. have discussed relevant issues [3840].

Another point that must be considered is that research on urban export trade is still lacking from the perspective of mutual competition. The reason why different cities compete in export trade is that the growth of export trade can effectively promote urbanization [41, 42] and urban economic development [43, 44]. Although over-reliance on exports can also lead to environmental damage [45, 46], the benefits of an export-oriented economic model have been encouraging an increasing number of cities to join the race. In China, some neighboring coastal cities are dotted with factories making identical or similar products to compete in the world market. Given the above, does this competition between domestic cities influence each city’s decisions on product mix and destination market selection? To the best of our knowledge, no previous studies have provided an answer to this question, and this is the research gap that this study tries to fill.



Commodity and market concentration indices.

Typically, the Gini-Hirschman coefficient [47] is applied to calculate the commodity concentration index (CCI) and market concentration index (MCI) as follows [4850]: (1) where expic is the commodity c value in city i‘s global exports, and expi is city i‘s total export value in the world market. The CCI value is between 0 and 100. The larger the value, the lower the degree of commodity diversification; the smaller the value, the higher the degree of commodity diversification. (2) where expim is city i‘s export value to country m, and expi is city i‘s total export value in the world market. The MCI value is between 0 and 100. The larger the value, the lower the degree of market diversification; the smaller the value, the higher the degree of market diversification.

Export similarity index.

As a reflection of the similarity in exports between regions, the export similarity index (ESI) has been widely adopted by scholars since its inception, especially those focusing on economic geography. The ESI has expounded many academic achievements involving international trade, industrial economies, and other multidisciplinary fields [5153]. It can indicate the degree of convergence of commodities in two regions in a certain market. The higher the ESI, the more intense the competitive relationship is [54, 55]. The formula is as follows: (3) where EZIijk is the similarity between cities i and j regarding exports in the market k; represents the ratio of commodity l that is exported by city i to market k to its total exports to market k; represents the ratio of commodity l that is exported by city j to market k to its total exports to market k. The value of ESIijk is between 0 and 100. The closer it is to 100, the more similar the export products of cities i and j in market k. In short, a high ESIijk indicates high export competition between cities i and j; a low ESIijk indicates less intense export competition between cities i and j. In this study, the market k is the global market.

Based on the ESI, this study creates the average export similarity index (AESI) as an indicator that reflects the competitive pressure that each city faces from other cities. The formula is as follows: (4)

This study’s research subject includes 270 cities, where each city has one ESI with every other city. This means that each city has 269 ESIs. The higher the AESIi, the greater the competitive pressure city i faces from other cities in the global market.

Geographical detector

Geographical Detector (Geodetector), is a statistical tool that measures Spatial Stratized Heteroeity (SSH). In this study, we use this method to analyze the extent to which the independent variable X (AESI) explains the spatial differentiation of the dependent variable Y (CCI and MCI). The calculation formula is as follows [56]: (5) q represents the explanatory power of X (AESI) on Y (CCI and MCI), and the larger the q, the stronger the explanatory power; q∈[0,1]. h represents the stratification of independent variables (the independent variable AESI is divided into 5 levels using quantile classification according to the study of Ding et al. [57]). Nh and N represent the number of units in layer h and whole region, respectively. and σ2 are the variances of Y in layer h and whole region, respectively.

The Almon lag model (ALM)

There is a lag in the influence of competitive pressure (AESI) on city export diversification (CCI and MCI). This is because the export enterprises in a city experience a change in competitive pressure first and then adjust the structure of export products. This process requires time. In this study, the distributed lag model is used to analyze the influence of AESI on CCI and MCI in the current and lag periods. The finite distributed lag model can be presented as follows [58]: (6) where Yt is the variable to be studied (i.e., CCI and MCI); a is the free term; bv is the sum of the values of the pulse response function; X is the explanatory factor (i.e., AESI); μt is white noise; t is time; v is the lag period; and s is the number of lag periods. This model describes how the lagged effects of changes in X are distributed over time [59].

To reduce the impact of multicollinearity, the Almon lag model (ALM) is used in this research [60, 61]. The Almon polynomial transformation is expressed as follows: (7) where φ is low order polynomial; and m represents lag weights.


The export data for each city are obtained from the customs database of China which is a database that many economists use [6265] (service trade data are not included). Based on the enterprise codes, the export situation of enterprises in the same cities is summarized and used as the original export data of those cities. In addition, the customs database includes a large span of data collection years, which has resulted in multiple versions: HS1996, HS2002, HS2007, HS2012, and HS2017. Although the differences between the versions are not significant from a macro viewpoint, each version has a certain degree of fine-tuning when compared with its previous version. Hence, they must be recoded systematically. According to the comparison table between the different commodity codes issued by the United Nations Trade Database (, this study uniformly adjusts the eight-digit commodity codes of every year of the customs data to the HS1996 version with six-digit codes, so that the CCI, MCI, and AESI between the different years can be measured comparatively. Based on data availability, 270 cities in mainland China are selected.

Study area

As shown in Fig 2, the cities analyzed are mainly located in the central and eastern regions of China. Moreover, the total export value of the 270 cities from 2000 to 2017 is close to that of the country. Therefore, the selection of these cities for analysis can comprehensively reflect the development characteristics of China’s urban exports.


The spatiotemporal pattern of export diversification

Commodity diversification.

From 2000 to 2017, an increasing number of Chinese cities diversified their export commodities. In 2000, only 73 cities had a CCI of less than 20, but the number increased to 117 in 2017 (Fig 3). The decline in CCI was significant in some cities. For example, Changchun City (the capital of Jilin Province) and Urumqi City (the capital of Xinjiang Province) both saw their CCI drop by more than 70% between 2000 and 2017.

Fig 3. Commodity concentration index (CCI) of the 270 Chinese cities analyzed in the study in the years 2000, 2005, 2010 and 2017.

However, some cities, especially those with higher economic levels, saw a rise in CCI. This means that exports from these cities became less diversified. The CCI of more developed cities, including Zhengzhou City (the capital of Henan Province), Chengdu City (the capital of Sichuan Province), and Shanghai City (a municipality directly under the central government), increased by more than 100% between 2000 and 2017.

Market diversification

As can be seen in Fig 4, the MCI of most cities in China was between 20 and 40 in 144 (year 2000), 182 (2005), 192 (2010), and 204 (2017) cities. The increase in the number of cities in the MCI range of 20–40 also reflects the increasing diversification of China’s urban export markets. In 2017, the average MCI for 270 cities was 33.46, which was 23.40% lower than the average in 2000.

Fig 4. Market concentration index (MCI) of the 270 Chinese cities analyzed in the study in the years 2000, 2005, 2010 and 2017.

Unlike the CCI, the number of cities with MCI below 20 was relatively small. For example, in 2017, only nine cities, including Jinan City (the capital of Shandong Province) and Jinhua (in Zhejiang Province), had an MCI less than 20. In 2010, although the MCI of many cities was below 20, the distribution was dispersed, and there was no spatial feature of clustering distribution like the CCI.

Most cities’ export markets became more diversified, but some did not. Of the 270 cities, 51, including Xi’an City (the capital of Shanxi Province) and Wuhan City (the capital of Hubei Province), showed a trend of growth in MCI from 2000 to 2017. In other words, their export markets were less diverse. For example, 14.49% and 8.64% of Xi’an City’s total export value in 2000 were exported to the United States and Hong Kong, respectively. In 2017, these figures increased to 17.35% and 26.99%, respectively.

As can be seen, the spatial difference of MCI is smaller than that of CCI. This is because a city can choose from thousands of commodities to export (the mix of exports varies greatly from city to city) but will have only a hundred or so markets to choose from. Taking 2017 as an example, 45.56% of cities chose the United States as their largest export market, while 14.81% and 8.52% chose Hong Kong and South Korea, respectively, as their largest export destinations (see Appendix A in S1 Appendix).

The impact of competition on export diversification

The export competition that each city faces.

As shown in Fig 5, cities in eastern and central China had significantly higher AESI values than those in the west. For example, in 2000, seven provincial capitals in the west and three in the east had an AESI of less than 5 and greater than 10, respectively. In 2017, the pattern changed slightly as many cities in the Jiangxi Province had significantly higher AESI values than their surrounding cities. In 2017, 11 of the 270 cities had an AESI greater than 16, of which 7 were in the Jiangxi Province (Fig 5). The higher AESI in Jiangxi is due to the intense competition among cities in the province (see Appendix B in S1 Appendix). The time variation of the AESI was also significant. In 2000, most cities had an AESI of less than 8. However, in 2017, 49.63% of the cities had an AESI greater than 8 (Fig 5). This change reflects the growing export competition among Chinese cities.

Fig 5. Average export similarity index (AESI) of the 270 Chinese cities analyzed in this study.

Spatial stratified heterogeneity: Spatial interaction of competition and diversification

We calculated the annual spatial interaction, which is the q value in the figure below (Fig 6). It is clear that AESI has a stronger explanatory power to the spatial pattern of CCI, and its q value remains in the range of 0.6–0.8 for a long time. In contrast, AESI is relatively weak in explaining the spatial pattern of MCI. q of AESI-MCI remained in the range of 0.1–0.2 for a long time.

Fig 6. q of AESI-CCI and AESI-MCI (p-value is less than 0.01 every year).

However, this is only a preliminary discussion of the relationship between AESI and two diversity indicators from a spatial analysis perspective. To what extent do changes in AESI affect CCI and MCI? The effect of AESI on the spatial pattern of MCI is weak from the perspective of spatial interaction, but does this mean that changes in the AESI of a city do not affect its MCI? These will be of the greatest concern to us.

The effect: Competition brings diversification

Both the AESI and CCI were panel data (18 years, 270 cities). Using Stata/MP 16.0, we found that AESI and CCI passed the panel unit root test (PURT) [66] and panel co-integration test (PCIT) [67] (Table 1). In the panel Granger causality test (PGCT) [68], we found that AESI was the Granger cause of CCI. Because under all three criteria (AIC: Akaike’s Information Criterion; BIC: Bayesian Information Criterion; HQIC: Hannan and Quinn’s Information Criterion) [6971], the null hypothesis Ha (AESI is not the Granger reason for CCI) can be significantly rejected (the p-value of both Z-bar and Z-bar tilde is below 0.01) (Table 2). Similarly, we found that CCI is also the Granger cause of AESI. In other words, AESI and CCI are bi-directional causality. Thus, AESI and CCI are endogenous. In the face of endogeneity, one of the common ways to solve the problem is by using lag variables. Therefore, the ALM was used to analyze how a change in AESI affects CCI.

Table 1. Results of panel unit root test (PURT) and panel co-integration test (PCIT) of CCI and AESI.

Table 2. Results of panel Granger causality test (PGCT) between average export similarity index (AESI) and commodity concentration index (CCI).

As shown in Fig 7A, the optimal lag period obtained by most cities is greater than 1 when analyzing the effect of AESI on CCI. For most Chinese cities, an increase in AESI led to a decrease in CCI (Fig 7B). In other words, the more a city faces competitive pressures in export, the more it diversifies its export commodities. This shows that the city is also a rational economic man to some extent. Increasing commodity diversification can prevent trade risks caused by commodity concentration and builds a healthier and stronger export system. China’s export trade has developed rapidly in the more than ten years since China’s accession to the WTO. In the face of the vast world market, although Chinese cities are competitive with each other, the huge scale of the destination market still makes it one of the important development paths for foreign trade enterprises in most cities to increase product types that will help obtain profits and cope with competition [72].

Fig 7. The optimal lag period and effect of average export similarity index (AESI) on commodity concentration index (CCI).

The optimal lag period is the lag period with the minimum mean standard error. Meaning of effect: If AESI is increased by one unit, CCI will increase/decrease by the corresponding amount. CCI’s increases are expressed as positive values and decreases as negative values.

However, 18.89% of the cities registered a positive effect of AESI on CCI (orange and red in Fig 7B). An increase in AESI would result in an increase in CCI. For example, the impact of AESI on CCI in Chongqing City, Xi’an City, Fuzhou City was 2.04, 1.37, and 1.14, respectively. Let us take Chongqing as an example to discuss why competition leads to reduced product diversification in some cities. Since 2010, in the face of fierce export competition from domestic cities, Chongqing’s foreign trade policy and economic development path have undergone major adjustments. In the same year, Chongqing has brought electronics giants to the city, including HEWLETT-PACKARD, Acer, and Foxconn. The original relatively dispersed export commodity structure has rapidly changed to electronic products as the leading export structure. In 2017, 45.79% of the total value of Chongqing’s exports were electronic products. Market subjects respond to increased competition by dropping their worst performing products [73]. For each city, although it faces competition from other cities in China, the path chosen is different. This largely depends on the city’s geographical location, policy advantages, and economic basis. Chongqing is one of the four municipalities directly under the central Government in China, and also a window of western China’s opening to the outside world, to which the Chinese central government attaches great importance. Therefore, for Chongqing, there are certain advantages to introduce the world’s high-tech companies to set up factories there and produce leading products for export.

Both AESI and MCI were panel data. Using Stata/MP 16.0, we found that AESI and MCI passed the PURT and PCIT (Table 3). In the PGCT, we found that AESI was the Granger cause of MCI (Table 4, column 4). Similarly, we found that MCI is also the Granger cause of AESI. In other words, AESI and MCI are bi-directional causality. Thus, AESI and MCI are endogenous. As in the previous analysis, ALM was used to calculate the effect of AESI on MCI.

Table 3. Results of panel unit root test (PURT) and panel co-integration test (PCIT) of MCI and AESI.

Table 4. Results of the panel Granger causality test (PGCT) between average export similarity index (AESI) and market concentration index (MCI).

The optimal lag periods of 90.37% cities are all greater than 1 (Fig 8A). For more than half of the cities (56.67%), an increase in their AESI led to a decrease in MCI (the light yellow and green areas in Fig 8B). For less than half of the cities (43.33%), an increase in their AESI led to an increase in MCI (the orange and red areas in Fig 8B). These results suggest that there does not appear to be a dominant effect of the impact of AESI on MCI. According to our analysis, this is probably related to the economic level of the city (Appendix C in S1 Appendix). Many cities with negative effects were in the southwest and northeast regions of China, including Kunming City (the capital of Yunnan Province) and Changchun City (the capital of Jilin Province). Many cities with positive effects were in the eastern and central regions of China, including more developed cities, such as Shanghai City and Nanjing City (the capital of Jiangsu Province). This phenomenon shows that more developed cities increase their export share to existing target markets when facing increased competitive pressures.

Fig 8. The optimal lag period and effect of average export similarity index (AESI) on market competitive index (MCI).

The optimal lag period is the lag period with the minimum mean standard error. Meaning of effect: If AESI is increased by one unit, MCI will increase/decrease by the corresponding amount. MCI’s increases are expressed as positive values and decreases as negative values.


Since the beginning of this century, the types of commodities that Chinese cities export to the world have become increasingly diversified, and the number of their target markets has also increased. The development characteristics of these two aspects show that the export trade strength of Chinese cities has increased significantly. Export trade has promoted China’s economic growth and transformation [74, 75]. However, because of the differences in resource endowments, location conditions, and policy treatments of cities in different regions, cities in western, northwestern, and northeastern China did not perform as well as those in southeastern China in terms of diversification of export commodities and export markets.

Based on the ESI, for the AESI that we constructed allows us to analyze the competitive pressure that each city faces. Moreover, AESI has become an important starting point for studying the factors influencing export diversification. From 2000 to 2017, export similarity between Chinese cities increased, and the competition for the global market became fierce. This process was accompanied by a continuous improvement in export diversification. We used Geodetector to discuss the spatial interaction between AESI and CCI/MCI. Notably, AESI explains the spatial pattern of CCI well, but fails to give the same ideal explanation for the spatial pattern of MCI. This stems from the calculation of spatial stratification heterogeneity, which uses cross-sectional data and does not fully incorporate the time element. Fortunately, the PGCT and ALM provide more comprehensive results. The results of the PGCT confirmed the existence of a causal relation between competition and diversification, thus laying a credible foundation for our final effect analysis using ALM.

Through ALM, we have found that competition brings diversification of export commodities. However, the impact of competition on the diversification of export markets has not shown a relatively uniform trend. Moreover, the more backward Chinese cities are, the greater the pressure of competition urges them to increase the variety of goods. This feature is more prominent in the impact of competition on market diversification. Generally, relatively backward cities have a fairly single type of export commodity and a more concentrated target market. Therefore, when the pressure of competition increases, they can use the industrial transfer and development experience of developed cities to broaden their sales channels. By contrast, developed cities have more mature and stable export industrial systems. Hence, when competitive pressure increases, they focus on superior products and key markets, thereby realizing larger-scale economic benefits. Furthermore, the product allocation behavior of an enterprise is an important mechanism for an enterprise to cope with the competition in the export market and thereby improve the productivity of the enterprise [76].

This article reveals an important discovery in the development of China’s urban trade that is, competition is not bad. In the long run, the existence of competition increases the diversification of export commodities of most cities. This is closely related to the anti-risk ability of a city’s trading system, especially for those with more fragile economic structures. COVID-19 has severely affected the export levels of cities in all countries, including China. Resistance to similar crises cannot lack a strong export structure, including the commodity and geographic structures. However, care must also be taken to avoid over-expansion caused by competition. If a city blindly expands its product range or develops a trade market with a low profit margin in order to meet the challenge of competitors, it is highly likely to backfire. On the other hand, the impact of competition on diversification has different results in different cities, and the central government should fully consider the actual situation in different regions when formulating national trade policies in order to achieve improved export performance. Specifically, for those relatively backward cities in the central and western regions that face competitive pressures and choose to expand commodity types and markets, the government should use more policy tools to enable their enterprises to expand overseas markets as far as possible. In fact, with the increasingly smooth trade routes between China, Central Asia, and Europe, the development of foreign trade in inland cities in the central and western regions offers new opportunities.

This study does not include other factors that may affect the diversification of urban exports. If more data can be gathered and sorted in the future, some other economic factors can be taken as control variables. Then, the influence of competition on diversification can be discussed more comprehensively.


We would like to thank Professor Yuqi Lu (Nanjing Normal University) and doctoral candidate Qi Cui (Nanjing University of Information Science &technology) for giving us valuable suggestions for this study.


  1. 1. Yong P. Research on international trade practice teaching reform. 2nd International Conference on Education Technology, Management and Humanities Science. 2019; 183–186. Available from:
  2. 2. Yao X, Zhou M. China’s economic and trade development: imbalance to equilibrium. World Economy. 2011; 34(12): 2081–2096.
  3. 3. Zheng L, Yue L, Zhou M, Yu H. Dental laboratory technology education in China: current situation and challenges. Journal of Dental Education. 2013; 77(3):345–347. pmid:23486899
  4. 4. Zhu Z, Zheng H, Zhu Z. Analysis on the economic effect of Sino-US trade friction from the perspective of added value. Environment Development and Sustainability. 2022; 24(1): 180–203. pmid:34093072
  5. 5. Li H, Zhao H. The influence and countermeasure of entry to RCEP on Dalian’s export. Foreign Economic Relations & Trade. 2019; (07):32–34+37.
  6. 6. Portyakov V. processes in the Pearl River Delta and the role of Shenzhen. Problemy Dal’nego Vostoka. 2016; (3):82–86.
  7. 7. Shen J. World city and its hinterland: the transformation of Hong Kong as trade & logistics-hub. Journal of Korea Trade. 2004; 8(1):25–54.
  8. 8. Hong LM. Import, export and economic growth: a GMM analysis based on dynamic panel data from Chinese prefecture city. Advanced Materials Research. 2014; 676–682. Available from:
  9. 9. Qin X, Li S, Xu W, Li X. Which export variety matters for urban economic growth, related or unrelated variety? Sustainability. 2019; 11(16): 4325.
  10. 10. Chen C, Wu Y. Impact of foreign direct investment and export on urbanization: evidence from China. China & World Economy. 2017; 25(1): 71–89.
  11. 11. Chen Z, Li B, Li T. Exports and left-behind children: empirical evidence from the China migrants dynamic survey. Review of International Economics. 2019; 27(4): 1081–1107.
  12. 12. Xin W, Smyth R. Economic openness and subjective well-being in China. China & World Economy. 2010; 18(2): 22–40.
  13. 13. Fang Z, Huang B, Yang Z. Trade openness and the environmental Kuznets curve: evidence from Chinese cities. World Economy. 2020; 43(10): 2622–2649.
  14. 14. Mao X, He C. Export upgrading and environmental performance: evidence from China. Geoforum. 2017; 86: 150–159.
  15. 15. Hering L, Poncet S. Environmental policy and exports: Evidence from Chinese cities. Journal of Environmental Economics and Management. 2014;68(2):296–318.
  16. 16. Chen Z, Poncet S, Xiong R. Local financial development and constraints on domestic private-firm exports: Evidence from city commercial banks in China. Journal of Comparative Economics. 2020;48(1):56–75.
  17. 17. Yu L, Ying R, Zhang B. How air pollution lowers the domestic value-added ratio in exports: an empirical study of China. Environmental Science and Pollution Research. 2021; 28(35): 48123–48140. pmid:33900557
  18. 18. Gaglio C. Trade liberalization and export performance: A literature review. Revue d’économie politique. 2017;127(1):25–46.
  19. 19. Kim TH. Export competitiveness of Japanese automobile industry to Germany. The Journal of Asian Studies. 2020; 23(2): 273–302.
  20. 20. Erguzel OS, Tunahan H, Esen S. Measuring the globalization of cities from the new regionalism perspective. SpringerPlus. 2016; 5 (1): 1–16. pmid:27652109
  21. 21. Kalafsky RV, Graves W. Exports and growth: learning from the case of Southern US Metropolitan Areas. Professional Geographer. 2018; 70(3): 383–394.
  22. 22. Duranton G, Morrow PM, Turner MA. Roads and trade: evidence from the US. Review of Economic Studies. 2014; 81(2): 681–724.
  23. 23. Taisarinova A, Loprencipe G, Junussova M. The evolution of the Kazakhstani Silk Road Section from a transport into a logistics corridor and the economic sustainability of regional development in Central Asia. Sustainability. 2020; 12(15): 6291.
  24. 24. Simmie J. Trading places: competitive cities in the global economy. Eur Plan Stud. 2002; 10(2): 201–214.
  25. 25. Pan Y, Ma L, Wang YZ. How and what kind of cities benefit from the development of digital inclusive finance? evidence from the upgrading of export in Chinese cities. Economic Research-Ekonomska Istrazivanja. 2021; 1–29.
  26. 26. Gourdon J, Hering L, Monjon S, Poncet S. Estimating the repercussions from China’s export value-added tax rebate policy. Scandinavian Journal of Economics. 2022; 124(1): 243–277.
  27. 27. Silva TC, Muniz FJ, Tabak BM. Indirect and direct effects of the subprime crisis on the real sector: labor market migration. Empirical Economics. 2022; 62(3): 1407–1438. pmid:33897095
  28. 28. Pei JS, de Vries G, Zhang M. International trade and Covid-19: City-level evidence from China’s lockdown policy. Journal of Regional Science. 2021; 1–26. pmid:34548696
  29. 29. Misztal P. Commodity export concentration and economic growth in the eumember countries. Ekonomista. 2011; (5): 691–708.
  30. 30. Al-Marhubi F. Export diversification and growth: an empirical investigation. Applied economics letters. 2000;7(9):559–562.
  31. 31. Hesse H. Export diversification and economic growth. Breaking into new markets: emerging lessons for export diversification. 2009;2009:55–80.
  32. 32. Del Rosal I. Export diversification and export performance by destination country. Bulletin of Economic Research. 2019; 71(1): 58–74.
  33. 33. Dikova D, Jaklic A, Burger A, Kuncic A. What is beneficial for first-time SME-exporters from a transition economy: A diversified or a focused export-strategy? Journal of World Business. 2016;51(2):185–199.
  34. 34. Parteka A, Tamberi M. Determinants of export diversification: an empirical investigation. Universita Politecnica Delle Marche, Dipartimento Di Economia Quaderno Di Ricerca. 2008; (327).
  35. 35. Dennis A, Shepherd B. Trade facilitation and export diversification. World Economy. 2011; 34(1): 101–122.
  36. 36. Regolo J. Export diversification: How much does the choice of the trading partner matter? Journal of International Economics. 2013; 91(2): 329–342.
  37. 37. Cieslik A, Parteka A. Export variety, productivity, and country size in a multi-good Ricardian model of export diversification. GUT FME Working Paper Series A. 2018; (6): 52.
  38. 38. He C, Ren Y, Li Y. The mechanism of cross-boundary product evolution in China: an empirical analysis based on export product of prefecture-level cities. Scientia Geographica Sinica. 2016; 36(11): 1605–1613.
  39. 39. Liu J, He C. Innovating independently or following the frontier: Geographical evolution path of China’s high-tech export industry. Arid Land Geography. 2020; 43(4): 1077–1087.
  40. 40. Hu X, He C. The spatial difference of market dynamics of Chinese exporters. Geographical Research. 2019; 38(9): 2129–2147.
  41. 41. Ahmed K. The sheer scale of China’s urban renewal and CO2 emissions: multiple structural breaks, long-run relationship, and short-run dynamics. Environmental Science and Pollution Research. 2016; 23(16): 16115–16126. pmid:27146684
  42. 42. Chhabra M, Giri AK, Kumar A. The impact of trade openness on urbanization: Empirical evidence from BRICS economies. Journal of Public Affairs. 2021; e2637.
  43. 43. Gao ZX, Peng J, Ma SB. international trade and its influence on economic growth. 7th International Conference on Education, Management, Computer and Medicine (EMCM). 2016; 29: 31. Available from:
  44. 44. Ma YH, Zhang Z, Huo M, et al. Multi-drivers and multi-mechanism analysis for city-level energy consumption in Suzhou based on the extended STIRPAT model. Iop Publishing Ltd; 2020; 467(1): 012197.
  45. 45. Wang J, Li J. Exploring the impact of international trade on carbon emissions: new evidence from China’s 282 cities. Sustainability. 2021; 13(16): 12.
  46. 46. Meng J, Liu JF, Guo S, Huang Y, Tao S. The impact of domestic and foreign trade on energy-related PM emissions in Beijing. Applied Energy. 2016; 184: 853–862.
  47. 47. Hirschman AO. National power and the structure of foreign trade: Univ of California Press; 1980.
  48. 48. Massell BF. Export concentration and fluctuations in export earnings: A cross-section analysis. The American Economic Review. 1964; 54(2): 47–63.
  49. 49. Love J. Export instability in less developed countries: consequences and causes. Journal of Economic Studies. 1987.
  50. 50. Silva JLMd, Montalván DBV. Exportações do Rio Grande do Norte: estrutura, vantagens comparativas e comércio intra-indústrial. Revista de Economia e Sociologia Rural. 2008; 46(2): 547–568.
  51. 51. Finger JM, Kreinin ME. A measure of export similarity and its possible uses. The Economic Journal. 1979; 89(356): 905–912.
  52. 52. Lu S, Dickerson K. The United States-Korea Free Trade Agreement (KORUS) and its impacts on China’s textile and apparel exports to the United States. Clothing and Textiles Research Journal. 2012; 30(4): 300–314.
  53. 53. Rondinella S, Agostino M, Demaria F, Drogue S. Similarity and competition in the agri-food trade among European mediterranean countries. The International Trade Journal. 2019; 33(5): 444–468.
  54. 54. Moreira MM. Fear of China: Is there a future for manufacturing in Latin America? World Development. 2007; 35(3): 355–376.
  55. 55. Caselli FG. China’s rise, asymmetric trade shocks and exchange rate regimes. Rev Int Econ. 2019;27(1):1–35.
  56. 56. Wang JF, Li XH, Christakos G, Liao YL, Zhang T, Gu X, et al. Geographical detectors-based health risk assessment and its application in the neural tube defects study of the heshun region, China. International Journal of Geographical Information Science. 2010; 24(1): 107–127.
  57. 57. Ding L, Niu X, Song X. Validating gravity model in multi-centre city: A study based on individual mobile trajectory. Acta Geographica Sinica. 2020; 75(2): 268–285.
  58. 58. Morozova E, Akulov A, Logunov T. Impact of environmental investments and air protection costs on the sustainable development of coal mining regions. EDP Sciences. 2019; 105: 02003.
  59. 59. Pitard A, Viel JF. Some methods to address collinearity among pollutants in epidemiological time series. Statistics in medicine. 1997; 16(5): 527–544. pmid:9089961
  60. 60. Almon S. The distributed lag between capital appropriations and expenditures. Econometrica: Journal of the Econometric Society. 1965: 178–196.
  61. 61. Siddiqui MA, Ieee. An application of VAR and Almon polynomial distributed lag models to insurance stocks: evidence from KSE2009. 2009 International Conference on Emerging Technologies. IEEE, 2009; 206–212.
  62. 62. Li R, Liu Y, Bustinza OF. FDI, service intensity, and international marketing agility: The case of export quality of Chinese enterprises. International Marketing Review. 2018.
  63. 63. Jarreau J, Poncet S. Export sophistication and economic growth: Evidence from China. Journal of Development Economics. 2012; 97(2): 281–292.
  64. 64. Huang Y, Lin C, Liu S, Tang H. Trade linkages and firm value: Evidence from the 2018 US-China trade war. Graduate Institute of International and Development Studies Working Paper. 2018.
  65. 65. Fan H, Lai EL-C, Li YA. Credit constraints, quality, and export prices: Theory and evidence from China. Journal of Comparative Economics. 2015; 43(2): 390–416.
  66. 66. Levin A, Lin C-F, Chu C-SJ. Unit root tests in panel data: asymptotic and finite-sample properties. Journal of Econometrics. 2002; 108(1): 1–24.
  67. 67. Westerlund J. Testing for error correction in panel data. Working Papers. 2005.
  68. 68. Dumitrescu E-I, Hurlin C. Testing for Granger non-causality in heterogeneous panels. Econ Model. 2012; 29(4): 1450–1460.
  69. 69. Akaike H. A new look at the statistical model identification. IEEE Transactions on Automatic Control. 1974; 19(6): 716–723.
  70. 70. Schwarz G. The bayesian information criterion. Ann Statist. 1978; 6: 461–464.
  71. 71. Hannan EJ, Quinn BG. The determination of the order of an autoregression. Journal of the Royal Statistical Society: Series B (Methodological). 1979; 41(2): 190–195.
  72. 72. Peng J, Lilin Y, Ronglin L. Market size, competition and export product mix—micro-evidence of Chinese multi-product export companies. Journal of International Trade. 2021; (10): 73–89.
  73. 73. Mayer T, Melitz M J, Ottaviano G I P. Market size, competition, and the product mix of exporters. American Economic Review. 2014; 104(2): 495–536.
  74. 74. Abeysinghe T, Lu D. China as an economic powerhouse: Implications on its neighbors. China Economic Review. 2003; 14(2): 164–185.
  75. 75. Gu K, Wu H. Shift to foreign-organizatin and maket:the choice of models and the arrangements for system in company’s control structure. Management World. 2003; 0(4): 9–16.
  76. 76. Sang R, Han C, Xiuzhen LI. How export market competition affects firm productivity:from the perspective of product allocation. Industrial Economics Research. 2018.