Introduction

The development of e-commerce mirrors how data, networks, and technology synergistically influence urban economic activities, driving enhanced transaction efficiency, refined market structures, and the comprehensive digital transformation of the economy1. As a novel and vibrant economic form in the context of globalization and informatization, it profoundly reshapes China’s urban development models and industrial structure2. Currently, e-commerce has become a crucial driver propelling significant leaps in China’s economic strength, technological prowess, and comprehensive national power3,4. In 2023, China’s e-commerce retail sales exceeded 15.42 trillion yuan, accounting for 32.7% of the country’s total retail sales (Fig. 1). This value has increased by nearly four times compared to 2013, which indicates that the “engine” role of e-commerce in driving economic and social development is increasingly strengthening. The concept of green development originated from the United Nations Conference on the Human Environment in 1972, calling on all countries to maintain and improve the human environment for the benefit of people and future generations5. In 2015, green development emerged as a pivotal component of the five core development principles, underscoring the vibrant essence of building a “beautiful China” in this new era. Subsequently, in 2024, China further underscored the urgency of transitioning to a green development model. The central focus of this initiative is on enhancing Green Total Factor Productivity (GTFP). This clear objective not only signifies the strategic direction for China’s socio-economic development model in its new stage but more decisively underscores the central mission of deeply integrating environmental protection with high-quality development. Currently, urban e-commerce and GTFP have shifted from independent trajectories to a phase of interactive symbiosis and progressive coupling. Their collaborative development has increasingly become a key approach to addressing the issues of unbalanced and inadequate regional development in China. As the strategic cores of national development6,7, China’s urban clusters, particularly the eight major urban agglomerations, embody the pinnacle of economic sophistication, technological advancement, and cultural vibrancy, functioning as indispensable cornerstones underpinning the national economy. In the process of high-quality development, these urban clusters confront more complex and severe challenges in advancing the development of urban e-commerce and fostering GTFP than other cities8. Recently, China has proposed the concept of “integrated urban agglomeration development”. Therefore, it is of great significance to scientifically evaluate the coupling interaction between the development level of urban e-commerce and GTFP in the eight major urban agglomerations, and to analyze their temporal and spatial evolution characteristics, which is very helpful to promote the construction of “Digital China” and “a beautiful China”.

Fig. 1
figure 1

E-commerce retail sales and total retail sales in China from 2013 to 2023.

Full size image

The coupling and coordination between the development of e-commerce and GTFP play a key role in shaping the new urban development pattern in the era of the digital economy9. On the one hand, as a key component of the digital economy, the rapid development of urban e-commerce has significantly improved the efficiency of urban resource allocation and effectively improved the city’s GTFP10,11. With the help of advanced technologies such as the Internet and big data, the rapid development of urban e-commerce not only meets the diversified needs of consumers. At the same time, it also realizes the disintermediation and transparent development of the circulation of goods and services12,13. This greatly reduces the carbon dioxide emissions and energy consumption generated by the circulation of goods. In addition, the development of online sales platforms has promoted the innovation of urban business models and service systems. This has stimulated the vitality of various industries and fields within urban clusters, promoted the vigorous development of new business forms and models such as online education and telemedicine, and effectively promoted the green transformation and development of cities. On the other hand, the improvement of urban GTFP has laid an important foundation for the rapid development of e-commerce14. The improvement of urban GTFP is usually accompanied by the modernization of urban infrastructure, such as efficient public transportation systems, smart grids, and high-speed Internet, which provide the necessary physical and technical support for e-commerce15. Meanwhile, the significant symbol of the improvement of GTFP is the popularization of green production methods, which is significant in driving innovation in e-commerce products and services. For example, the emergence of second-hand trading platforms, sharing economy models, and eco-friendly products16,17,18. Therefore, the development of urban e-commerce and green development form a benign interactive development model that promotes each other and rises in a cycle.

However, a pivotal issue amidst urban development transitions centers on skillfully blending and maintaining a harmonious balance between urban e-commerce progression and green development, thereby promoting equitable and sustainable growth within and among urban clusters. To address this issue, we propose a model that measures the coupling and coordination between urban GTFP growth and e-commerce development based on relevant data from eight major urban agglomerations. The empirical analysis was conducted using the Dagum Gini coefficient and Kernal density estimation methods to examine the spatial pattern and temporal evolution trend of GTFP growth, e-commerce development, and their coupling coordination in eight major Chinese cities from 2011 to 2022. Succinctly put, our research provides an important decision-making reference for promoting the effective coordinated development of urban e-commerce development and the improvement of GTFP, and realizing the innovation-driven development of urban agglomeration and the synergy of regional economy.

The remaining sections of this study are structured as follows: “Methods and materials” presents the relevant methods employed and explains the data sources. “Results” introduces and analyzes the findings. Finally, “Conclusions” outlines the conclusions drawn from the study and offers policy implications.

Methods and materials

Coupling coordination degree model

The Coupling coordination degree Model serves as a sophisticated analytical instrument designed to assess the dynamics of interaction and the degree of alignment between multiple systems19. Extensively applied across a spectrum of fields such as regional economic and environmental studies, this model quantifies the interconnectivity, mutual influence, and collaborative potential within complex systemic frameworks20,21,22. Thus, we employ the model to delve into the intricacies of the relationship between urban e-commerce development and the enhancement of GTFP. Specifically, entropy weight TOPSIS model and meta-frontier data envelopment analysis with undesirable outputs is used to calculate the urban e-commerce development level and GTFP respectively. Then, the coupling coordination degree model is utilized to quantify the extent of coupling coordination between the two systems. The calculation formulas are as follow:

$$C=2\sqrt {\frac{{{U_1}{U_2}}}{{{{({U_1}+{U_2})}^2}}}} $$
(1)
$$T=\alpha {U_1}+\beta {U_2}$$
(2)
$$D=\sqrt {C \times T} $$
(3)

In the equations, C denotes the coupling degree between the development level of urban e-commerce and GTFP; T represents the overall coordinated index resulting from the combined interaction of the urban e-commerce development system and GTFP system; D represents the coupling coordination degree between two systems. Its values are within the interval [0, 1]; U1 signifies the aggregate evaluation index for the urban e-commerce development system; U2 represents the comprehensive evaluation index for the GTFP system. α and β are undetermined coefficients representing the weights of the municipal e-commerce development system and the GTFP system, respectively. Given the equal emphasis on both systems, their weights are both set at 0.5. Based on this, we introduce the concept of relative development degree which denoted as U1/U2, to clarify the development focus of optimizing the coupling coordination mode. This metric serves to measure the extent to which the urban e-commerce development level system leads or lags behind the GTFP system.

Based on the measurement results of the coupling coordination degree of the two systems, we refer to the research of Cheng et al. (2023)23, and divide the coupling coordination degree D into ten levels (Table 1). As D increases, the coupling coordination level transitions progressively from extreme imbalance to high quality coordination.

Table 1 The division of coupling coordination level.
Full size table

Dagum Gini coefficient

Compared to traditional approaches such as the Gini coefficient and Theil index, the Dagum Gini coefficient decomposition method uniquely addresses the issue of overlapping sub-samples and decomposes spatial disparities into inter-level difference, intra-level difference and hypervariable density, which is useful to yield more rigorous and accurate estimates24. Therefore, the Dagum Gini coefficient decomposition method is employed to investigate the spatial variations and their underlying sources in the coupling coordination degree between the development levels of urban e-commerce and GTFP. The calculation formulas are as follow:

$$G=\sum\limits_{{j=1}}^{h} {\sum\limits_{{d=1}}^{h} {\sum\limits_{{i=1}}^{{{n_j}}} {\sum\limits_{{r=1}}^{{{n_d}}} {\left. {\left| {{D_{ji}} - {D_{dr}}} \right.} \right|} } } } /2{n^2}\overline {u} $$
(4)
$${G_w}=\sum\limits_{{j=1}}^{m} {{G_{jj}}{p_j}{s_j}} $$
(5)
$${G_{nb}}=\sum\limits_{{j=2}}^{m} {\sum\limits_{{d=1}}^{{j-1}} {{G_{jd}}} } ({p_j}{s_d}+{p_d}{s_j}){H_{jd}}$$
(6)
$${G_t}=\sum\limits_{{j=2}}^{m} {\sum\limits_{{d=1}}^{{j-1}} {{G_{jd}}} } ({p_j}{s_d}+{p_d}{s_j})(1-{H_{jd}})$$
(7)

In the formulas, G, \({G_W}\), \({G_{nb}}\)and\({G_{\text{t}}}\)represent the overall Gini coefficient, intra-level difference, inter-level difference and hypervariable density, respectively; h represents the number of urban agglomerations; \({n_j}({n_d})\) indicates the number of cities within the urban agglomeration \(j(d)\); \({D_{ji}}({D_{{\text{dr}}}})\) is the coupling coordination degree between the e-commerce development level and GTFP of city \(i(r)\)in the urban agglomeration\(j(d)\); n signifies the total number of cities; \(\overline {u} \) is the mean of the two-systems’ coupling coordination degrees across all cities; \({G_{jj}}\) represents the Gini coefficient for the urban agglomeration j; \({G_{jd}}\) indicates the Gini coefficient between the urban agglomeration j and d; \({p_j}={n_j}/n\), \({s_j}={n_j}{u_j}/n\overline {u} \); \({H_{jd}}\) measures the interactive influence on the coupling coordination degrees between regions j and d.

Kernel density estimation

Kernel density estimation is a significant method for studying spatial non-uniform distributions, as it delineates the spatial characteristics of random variables through the use of continuous and smooth kernel density curves25. This method reveals the information about the location, shape, extension, and polarization phenomena of the research subjects, thereby shedding light on the dynamic evolution trends in their spatial patterns26. Consequently, this paper employs the Kernel density estimation method to investigate the spatial distribution features and evolving trends of the coupling coordination degree between two systems within China’s eight major urban agglomerations. In terms of the selection of the kernel function, this study adopts the Gaussian kernel function27. The calculation formula presented as follows:

$$f(x)=\frac{1}{{nh}}{\text{ }}\sum\limits_{{i=1}}^{n} {K(\frac{{{X_i} - x}}{h})} $$
(8)
$$K(x)=\frac{1}{{\sqrt {2\pi } }}\exp \left( { - \frac{{{x^2}}}{2}} \right)$$
(9)

Where \(f(x)\) represents the estimated kernel density value; \(K({\text{x}})\) denotes the kernel function; n stands for the number of cities within the urban agglomeration; \({X_i}\) refers to the coupling coordination degree observation value for two independently and identically distributed systems; x signifies the mean coupling coordination degree between the two systems; h represents the bandwidth parameter.

Construction of evaluation system

Based on the principles of scientificness, systematicness and operationality of an indicator system, we separately construct the evaluation indicator systems for both the urban e-commerce development and the GTFP.

Evaluation index system for e-commerce development system

E-commerce is a new form of economy, characterized by the rapid development of online transactions, the widespread coverage of modern information networks, and the deep integration of information and communication technologies. Thus, referring to the research of Zeng et al. (2023)3, Han et al.(2023)28 and other researchers25,26,29,30, we built an evaluation index system for the urban electronic commerce development system from development scale, resource conditions, and application level. More details are seen in Table 2.

Table 2 Index system for the development level of e-commerce.
Full size table

Evaluation index system for GTFP system

GTFP is defined as the maximum economic output (desirable output) and minimum pollutant emissions (undesirable output) that a decision-making unit (DUM) can achieve under constant labor and capital inputs. In terms of input indicators, referring to the research of Feng et al. (2023)31, the number of employed people at the end of the year, fixed assets investment and capital stock are selected as labor input and capital input, respectively. As for output indicators, we refer to the research of Zhang and Guo (2024)19 and takes GDP as the expected output. Referring to the practice of Xu et al. (2023)32 and Gilani and Hoseinzadeh (2021)33, the comprehensive index of environmental pollution using the entropy method is used to represent the undesirable output in the process of urban economic activities. The specific description of input-output variables is shown in Table 3.

Table 3 Input-output index system.
Full size table

Data sources

As a critical region for China’s economic development, scientific innovation, and cultural flourishing, the eight major urban agglomerations play a significant role in supporting the national economy and thus possess significant research value. Therefore, drawing on the research of Yang et al. (2024)34, this paper chooses these eight urban agglomerations as the sample for analysis (see Fig. 2). Given the continuity and accessibility of data, this study focuses on the panel data from 140 cities between 2010 and 2022. The data mainly came from the China Urban Statistical Yearbook, China Statistical Yearbook, China Environmental Statistical Yearbook, and various city-national economic and social development statistical bulletins. A few missing data were filled in using the linear interpolation method.

Fig. 2
figure 2

Geographical spatial distribution map of the eight urban agglomerations in China. Produced based on the Ministry of Natural Resources Standard Map Service website, Standard Map No. GS(2019)1822, with no modifications to the boundaries of the base map. The map was created using ArcGIS, version 10.7 (https://www.esri.com/en-us/arcgis/about-arcgis/overview).

Full size image

Results

Temporal characteristics of overall coupling coordination degree

Table 4 showcases the coupling coordination status between urban e-commerce development levels and GTFP across China’s eight urban clusters from 2011 to 2022. Analysis discloses that the average coupling coordination degree demonstrated an upward trend with fluctuations over this period, effectively transitioning from a tenuous imbalance to a more cohesive state. The relative development index consistently ascended towards unity, indicating that a balanced developmental equilibrium between these two domains is steadily materializing.

Specifically, between 2011 and 2015, the coupling coordination degree fluctuated between 0.487 and 0.682, denoting a phase transitioning from near dysfunction to primary coordination. Despite this, the relative development index remained relatively low, suggesting that during this period, there was a dearth of harmony between urban e-commerce growth and the burgeoning GTFP within these urban clusters. There existed a stark misalignment and incongruity, with urban e-commerce development lagging behind the strides made in the realm of GTFP. This shortfall hindered the optimal exploitation of the transformative potential unleashed by the latter, limiting the full-scale e-commerce boom across these urban areas. The constraint was largely due to the fact that, despite the pervasive use of information technology and internet industries, the inadequate infrastructure backbone restricted the scalability of e-commerce. Thereby failing to capitalize fully on the opportunities presented by the rapid growth of GTFP. Additionally, the economy faced structural adjustments as it endeavored to pivot from traditional to emerging growth drivers, with conventional manufacturing and service sectors encountering difficulties in transforming into digital, network-centric models. These barriers stifled the synchronized progress of both digital and offline economies, thereby inhibiting the powerful synergies that could arise from the seamless blend of e-commerce platforms with the physical economy.

From 2016 to 2020, there was a pronounced improvement in the coupling coordination degree among the eight urban clusters, progressing towards a state of high quality coordination. The relative development index climbed from 0.692 to the highest value of 0.983 in 2019. However, due to the impact of the global new crown epidemic in 2020, the coupling coordination degree and relative development degree of the two systems in China’s eight major urban agglomerations declined to a certain extent. This signifies that during the 13th Five-Year Plan period, urban e-commerce technologies and their applications within these urban clusters witnessed a heightened convergence with the maturation of GTFP within the evolving economic environment. This led to a profound interdependence between the digital commerce sphere and the material economy, resulting in a mutually beneficial coexistence based on deep integration. This synergy not only powered the robust growth of e-commerce but also triggered collaborative growth across related industries, enhancing the overall competitiveness and resilience of the regional economy. It is noteworthy that during 2021–2022, amidst the COVID-19 pandemic, the coupling coordination between the two systems in China’s eight major urban agglomerations underwent significant fluctuations. Concurrently, there was a noticeable decline in the relative development degree. This phenomenon underscores the impact of the pandemic on the interplay between the urban economic and ecological systems, highlighting the need to enhance the resilience of these systems against future uncertainties.

Table 4 The coupling coordination characteristics of two systems in the eight urban agglomerations of China.
Full size table

Temporal characteristics of regional coupling coordination degrees

To mitigate the potential impact of the COVID-19 pandemic on the coupling coordination of the eight urban agglomerations, we utilize data from 2011, 2013, 2017, and 2019 to create geographical distribution maps. This enables an understanding of the inherent changing characteristics of the coupling coordination degree between the two major systems of China’s eight urban agglomerations.

According to the Fig. 3, it is can be observed that the overall coupling coordination degrees have increased among these urban clusters. However, significant regional disparities emerge, characterized by a pattern of higher coordination in the south and east versus lower coordination in the north and west. In 2011, the coupling coordination degrees of the two systems in YRD and PRD were in the stage of primary coordination. However, other six urban clusters’ coupling coordination degrees ranged from 0.200 to 0.600, mostly hovering near dysfunction or barely coordination. This indicates a pronounced imbalance issue between urban e-commerce development and GTFP across the eight clusters. The YRD and PRD urban agglomerations, due to their advanced infrastructure and robust manufacturing bases, facilitated better integration of e-commerce sectors and GTFP. Conversely, urban agglomerations such as HC and GP faced substantial challenges due to lagging e-commerce infrastructure and slow economic growth.

In 2013, there was a marked improvement in the coupling coordination degree across all eight urban clusters. Many cities in BTH, MRYR and CY entered the stage of primary coordination, with Beijing, Hangzhou, and Guangzhou advancing to the intermediate coordination stage. HC, GP, and CP urban agglomerations transitioned from moderate imbalance towards barely coordinate. This shift can be attributed to China’s ongoing economic structural optimization and transformation of green development mode. Every urban cluster actively seeking new engines to achieve green and innovative growth. E-commerce, as a nascent industry form, effectively propelled traditional industrial upgrading and enhancement of overall urban economic vitality and sustainable development level. Consequently, metropolises such as Beijing, Hangzhou, Guangzhou, and Shanghai redoubled their policy backing for the e-commerce sector’s development. They invested substantially in the establishment of e-commerce hubs, nurtured front-runner enterprises, and encouraged relevant technological innovations. These efforts collectively expedited the exponential growth of e-commerce industries within these urban clusters. Cities like Harbin, Changchun, Zhengzhou, and Xi’an leveraged their unique geographical positions and rich manufacturing resources to fortify their logistics networks and vigorously promote a profound integration between traditional manufacturing sectors and e-commerce, thereby effectively catalyzing industrial transformations and upgrades.

In 2017, YRD, PRD, and BTH urban clusters witnessed rapid increases in their coupling coordination degrees, with most cities entering a good coordination phase. Notably, cities including Beijing, Tianjin, Shanghai, Hangzhou, Yangzhou, Guangzhou, Shenzhen, and Dongguan achieved high-quality coordination. CY and HC urban clusters showed steady improvements, with cities such as Chengdu, Chongqing, Changchun, and Harbin reaching intermediate and good coordination stages. However, it is noteworthy that the MRYR urban cluster did not show significant progress, remaining in the primary and intermediate coordination phases. In contrast, CP and GP clusters experienced a decline in their coupling coordination degrees, with cities like Xinxiang and Luoyang slipping into states nearing dysfunction or mild dysregulation, indicating severe polarization among the eight urban clusters. While YRD, PRD, and BTH had progressed to a stage of enhanced quality and efficiency in their coupled development, CP and GP seemed to regress.

This suggests that the underdeveloped logistics, warehousing, and internet infrastructure in these regions hindered the rapid growth of e-commerce and the development of GTFP. Additionally, slow industrial transformation and the incomplete formation of emerging e-commerce industries and supporting service systems led to insufficient momentum for e-commerce growth, which failed to keep pace with GTFP advancements. In contrast, the MRYR urban cluster saw substantial improvements in information infrastructure, enabling traditional industries to transform marketing models and production methods through e-commerce, thereby promoting innovative and green transformation of urban development. Moreover, the rapid development of emerging industries and high-tech sectors within the e-commerce realm injected fresh impetus into e-commerce growth, allowing many cities within the MRYR cluster to advance into good or high-quality coordination stages.

In summary, while certain urban clusters made significant strides in harmonizing e-commerce development with GTFP over time, others grappled with setbacks, often influenced by external factors and internal developmental constraints. The temporal evolution of coupling coordination degrees highlights both the achievements and challenges faced by these eight Chinese urban clusters in their pursuit of synchronized e-commerce and productivity growth.

Fig. 3
figure 3

Spatial distribution of coupling coordination degree of two systems in eight urban agglomerations. Produced based on the Ministry of Natural Resources Standard Map Service website, Standard Map No. GS(2019)1822, with no modifications to the boundaries of the base map. The map was created using ArcGIS, version 10.7 (https://www.esri.com/en-us/arcgis/about-arcgis/overview).

Full size image

Study on spatial heterogeneity

Table 5; Figs. 4 and 5 present the decomposed Gini coefficient results and respective contribution rates for the coupling coordination degree between two systems across China’s eight major urban clusters during the period from 2011 to 2022.

Fig. 4
figure 4

Gini coefficient of the coupling coordination degree of the two systems from 2011 to 2022.

Full size image

As shown in Table 5, the overall Gini coefficient for the coupling coordination between these two systems within the eight urban clusters exhibits a declining trend characterized by a “down-up” V-shaped pattern. This suggests that there has been a gradual narrowing of spatial disparities in terms of their coupling coordination across different regions in China, indicating improved synchronization and balanced development between urban clusters. However, this decline is marked by distinct temporal phases. Specifically, from 2011 to 2014, the overall Gini coefficient pertaining to the coupling coordination between the two systems in China’s eight city clusters displayed a downward trajectory with an evident reduction in spatial differences. According to Fig. 4, there was a substantial decrease in the overall Gini coefficient for the coupling coordination between the two systems among these city clusters, plummeting from 0.163 in 2011 to 0.088 in 2014, representing a notable drop of 46.08%. The average contributions from inter-group disparities (10.49%) and intra-group disparities (58.90%) account for a combined total of 69.38% of the overall Gini coefficient, surpassing the contribution from super-variable density. This finding underscores regional heterogeneity as the primary factor influencing the overall Gini coefficient, with disparities between the eight city clusters exerting the greatest impact. The reason is that during this period, various urban agglomerations implemented differentiated development strategies based on their unique resource endowments and industrial foundations. The BTH focuses on collaborative innovation, the YRD enhances industrial chain cooperation, the PRD deepens open collaboration, and the CH region emphasizes ecological construction and industrial upgrading. These measures have effectively narrowed the development gap between regions and enhanced overall coupling and coordination. Additionally, local governments in urban agglomerations such as the BTH and YRD regions have also guided the rational layout of industries through adjustments to fiscal and tax policies and land use systems, thereby avoiding disorderly competition and promoting balanced regional economic development. Consequently, during this phase, the spatial gap in the coupling coordination degree expanded gradually among the city clusters. Additionally, Fig. 4 reveals that the contribution from super-variable density fluctuated mildly downwards, signifying a relatively smaller influence on the overall disparity due to overlapping and cross-regional issues among the eight city clusters.

Fig. 5
figure 5

Contribution rates for the coupling coordination degree between two systems from 2011 to 2022.

Full size image

From 2015 to 2022, however, there was a fluctuating increase in the overall Gini coefficient for the coupling coordination degree of the two systems in China’s eight major urban clusters, accompanied by a relative expansion of spatial disparities. As seen in Figs. 4 and 5, the overall Gini coefficient rose from 0.104 in 2015 to 0.116 in 2022, registering an increment of 11.01%. The combined average contribution rate of inter-group differences (7.74%) and intra-group differences (76.27%) reached 84.01%, again underscoring that regional heterogeneity remained the primary determinant of the overall Gini coefficient. The reasons may be: On the one hand, each region has tailored distinct industrial development paths and policy orientations based on its unique resource endowments and developmental phase, exemplified by strategies like the Beijing-Tianjin-Hebei coordinated development and the Yangtze River Delta integrated development planning. However, in practice, variations in local government interests, industrial foundations, and innovation capabilities have led to regional disparities in policy outcomes, resulting in fluctuations in the degree of coupling and coordination. On the other hand, external shocks like international trade frictions and supply chain restructuring have disproportionately impacted export-oriented urban clusters like the Pearl River Delta and Yangtze River Delta, while inland clusters such as Chengdu-Chongqing and the Central Plains, with their lower reliance on external factors, have been less affected. This asymmetric impact from the external environment has exacerbated developmental imbalances among the regions. Additionally, the domestic push for economic restructuring has set varying transformational demands on different urban agglomerations. Urban clusters like MRYR and HC, with their singular industrial structures and limited innovation capacity, face significant transformational pressures, contributing to the fluctuations in their coupling and coordination degrees. Thus, future urban cluster planning and development endeavors should increasingly prioritize fostering collaborative innovation among constituent cities, implementing cross-regional cooperation projects, and facilitating industrial transfers. These actions are aimed at systematically reducing disparities in the harmonious coordination of e-commerce growth and the deployment of GTFP across the eight city clusters.

Table 5 Sources and contributions of the overall differences in the coupling coordination degree of the two systems in China’s eight urban agglomerations from 2011 to 2022.
Full size table

Research on dynamic evolution trend

In this study, Kernel Density Estimation (KDE) methodology has been employed to rigorously analyze the distributional location, morphology, extension, and polarization phenomena of the coupling coordination degree between two systems across China’s eight major urban agglomerations, as illustrated in Fig. 6.

Regarding the distribution location, the central tendency and variation range of the overall distribution curve exhibit an evolutionary trajectory characterized by a “rightward shift followed by a leftward shift,” which substantiates the finding from previous analyses that the coupling coordination degree among these eight urban clusters demonstrates a consistent upward trend over time. With respect to distribution morphology, the peak height and width of the curve indicate a pattern of diminishing peak intensity and increasing breadth in stages. This signifies that there has been a fluctuating change feature in the spatial disparities of the coupling coordination degree among the eight urban agglomerations, initially expanding and then contracting before widening again. Regarding distribution extensibility, the gradual attenuation of the right-skewed tail in the distribution curve suggests a discernible trend towards contraction in its spread, implying that cities within these urban agglomerations with lower levels of coupling coordination have, through adjustments, moved closer to those with higher degrees of coupling. As for the polarization phenomenon within the distribution, the curve transitions from an initial bimodal structure with one dominant peak accompanied by a smaller side peak, evolving into a unimodal distribution over time. This evidences a process where the dichotomy in the coupling coordination degree among the eight Chinese urban agglomerations first intensified before experiencing some degree of mitigation; thus, the polarization effect initially strengthened but eventually weakened.

The reason may be that, at first, due to the concentration of capital, technology and policy in the BTH, YRD and PRD. These urban agglomerations have robust technological innovation capacity and substantial policy support. This has promoted the synergistic development of e-commerce and GTFP. As a result, obvious polarization has been formed. With the implementation of the national regional development strategy, capital and technology gradually flow to the GP, HC, CY and other urban agglomerations. Therefore, the infrastructure construction of urban e-commerce development has been gradually improved, and the upgrading of green industry has been steadily carried out. As a result, the coupling and coordination degrees of the two systems have steadily increased, narrowing the gap with the BTH, YRD and PRD. This is reflected in the gradual improvement of polarization and prominent convergence characteristics.

Fig. 6
figure 6

Kernel estimation of coupling coordination degree of two systems in eight urban agglomerations from 2011 to 2022.

Full size image

Conclusions

Based on samples of 140 prefecture-level cities within China’s eight urban agglomerations from 2011 to 2022, this study employs the entropy weight method and the Meta-frontier Data Envelopment Analysis (DEA) to measure the development level of e-commerce in cities and GTFP. Subsequently, employing the coupling coordination degree model, Dagum distribution, and kernel density estimation techniques, it investigates the spatiotemporal characteristics and evolution trends of the coupling coordination between urban e-commerce development levels and GTFP in these eight urban agglomerations.

Findings reveal that:

  1. (1)

    From 2011 to 2022, the coupling coordination degree of urban e-commerce development level and GTFP in China’s eight urban agglomerations shows a fluctuating and increasing trend, realizing the transformation from being on the brink of imbalance to good coordination, and the balanced development mode of the two systems is gradually forming. By region, the coupling and coordination degrees of the two systems in the eight urban agglomerations have been improved on the whole, but the regional differentiation characteristics are significant. In terms of time periods, there are obvious stage characteristics in the growth of the coupling coordination degree of the two systems in China’s eight urban agglomerations. In the period of improving quality and efficiency from 2016 to 2020, the coupling coordination degree of the two systems is significantly improved, and gradually transitions to the stage of good coordination. 2021–2022 is the period of fluctuation adjustment, and the coupling coordination degree of the two systems shows an obvious fluctuation trend.

  2. (2)

    From 2011 to 2022, the spatial heterogeneity of the coupling coordination degree between the development level of urban e-commerce and GTFP in China’s eight urban agglomerations is obvious, and the overall spatial difference shows a downward trend and shows a “descending - rising” V-shaped change feature. From 2011 to 2014, the overall Gini coefficient of the coupling and coordination degrees of the two systems in China’s eight urban agglomerations shows a downward trend, and the spatial difference is constantly narrowing. From 2015 to 2022, the overall Gini coefficient of the coupling and coordination degree of the two systems in China’s eight urban agglomerations shows an increasing trend of fluctuation, and the spatial difference has expanded.

  3. (3)

    From 2011 to 2022, the coupling and coordination degrees of the two systems in China’s eight urban agglomerations show a “rising - rising” change feature. The fluctuation of the spatial gap decreases, showing the characteristics of the trend of “expanding-narrowing - expanding.” In China’s eight urban agglomerations, the cities with lower coupling and coordination degrees of the two systems gradually approach the cities with higher levels after adjustment.

Drawing upon these research conclusions, the following policy recommendations are proposed for the “15th Five-Year Plan” period to promote the coordinated development of e-commerce development levels and GTFP within China’s eight urban agglomerations:

  1. (1)

    Formulate differentiated development strategies. Given the differences in the coupling coordination degrees of the two systems in different urban agglomerations, the government should formulate differentiated development policies to promote the balanced development of the region. Each urban agglomeration should formulate different development orientations and development strategies according to its own resource endowment and development potential. For example, BTH can build intelligent and green e-commerce demonstration highlands. The YRD should focus on deepening the integration of e-commerce and manufacturing and building a smart supply chain. MRYR can focus on building a green e-commerce logistics system. By tapping the potential of cultural e-commerce, GP should promote the integration of culture and tourism, and promote the improvement of urban GTFP.

  2. (2)

    Using the convergence effect to promote the balanced development of urban agglomerations. On the one hand, administrative barriers should be broken to build a collaborative development network. The administrative boundaries of urban agglomerations can be effectively weakened by establishing a helping mechanism. Promote the flow of e-commerce resources and green development resources from developed urban agglomerations to less developed areas. Finally, we will promote the balanced development of urban agglomerations. On the other hand, we should improve the institutional framework and lay a solid foundation for cooperation. A combination of legal, policy, economic, and social means should be comprehensively used to balance the internal and external interests of various urban clusters, and provide effective institutional guarantees and constraints for cross-city e-commerce cooperation and urban green transformation and upgrading at the institutional level.

  3. (3)

    Build a sound incentive mechanism. Set up an urban e-commerce development fund, focusing on supporting the modernization and upgrading of e-commerce infrastructure in urban agglomerations with low coupling and coordination degrees, such as HC and GP, to create a more efficient and convenient operation environment for e-commerce enterprises. Further, improve the intellectual property rights system in the field of green development, encourage and protect the achievements of urban green scientific and technological innovation, and promote the green transformation and development of cities; Build a diversified sharing platform for government, industry, education, and research, and encourage cities to share excellent experience and advanced technologies in e-commerce development and GTFP improvement across urban clusters.