The Spatio-Temporal Differentiation and Convergence Characteristics of the Coordinated Development of Digitalization and Greening in China

Abstract

The synergistic development of digitalization and greening is an important lever for China to accelerate the formation of new quality productive forces. This study adopted the global entropy method and coupling coordination degree model to measure the level of coordinated development between digitalization and greening with the panel data of Chinese cities from 2011 to 2022. Spatio-temporal evolution characteristics were explored through kernel density estimation, the Dagum Gini coefficient, and spatial autocorrelation methods. This study further tested the convergence characteristics of coordinated development through a two-way fixed effect model and spatial econometric model. The results show the following: (1) The overall level of coordinated development of digitalization and greening in China is on the rise, with the development level in the eastern region being significantly higher than that in the central and western regions. The degree of differentiation in coordinated development shows a trend of decreasing first and then increasing, mainly due to regional differences. (2) The level of coordinated development between digitalization and greening in China shows a significant positive spatial autocorrelation feature, with a clustering pattern dominated by “low–low” clustering. (3) It is found that the coordinated development of digitalization and greening in China has significant characteristics of σ convergence, spatial β convergence and club convergence.

1. Introduction

With the continuous promotion of the “Beautiful China” and “Digital China” strategies, the synergistic development of digitization and greening has become a key path to achieve a comprehensive green transformation of economic and social development. To this end, in reports such as “The 14th Five-Year Plan for National Informatization” and the “Implementation Guidelines for the Coordinated Transformation and Development of Digitalization and Greening,” the Chinese government has, from a top-down design perspective, put forward new requirements for enhancing the level of synergistic development between digitalization and greening. However, it is important to note that while China has achieved significant results in digitalization and greening, efforts to promote their synergistic development are still in the exploratory phase. Issues such as the high energy consumption of digital infrastructure and insufficient green application of digital technologies pose obstacles to their deep integration. Additionally, the synergistic development of digitalization and greening faces numerous challenges, including regional development imbalances and transition pains. Therefore, to better achieve the digital–green synergistic transformation, it is necessary to scientifically measure the level of their synergistic development and conduct in-depth research into their spatio-temporal evolution characteristics and convergence trends.

In recent years, scholars have conducted extensive research on digitalization, greening, and their interrelationships. (1) Research on digitalization. Domestic and international scholars have primarily focused on the measurement, influencing factors, and economic effects of digitalization. For example, Jin Xingye et al. [1] identified existing issues in digitalization measurement and employed machine learning methods to scientifically measure corporate digital transformation. Wu et al. [2] investigated the impact of corporate scale and product type characteristics on digital transformation. Ling Chen et al. [3] conducted empirical analyses on the effects of digitalization on global value chains and manufacturing firms’ total factor productivity (TFP). Recent international review studies further suggest that digital transformation is increasingly intertwined with sustainable marketing, green supply chains, and broader sustainability transitions, indicating that the implications of digital technologies extend beyond productivity gains to environmentally oriented value creation [4]. (2) Research related to greening. Under the backdrop of green development, existing studies primarily measure the level of green development using indicators such as carbon dioxide emissions [5], greening composite indices [6], and green total factor productivity [7]. Additionally, scholars have also examined the driving factors of greening, such as environmental policies [8], technological innovation [9], and resource dependence [10]. International studies also show that environmental regulation may stimulate innovation and even competitiveness under certain conditions, while product greenness itself plays a pivotal role in green innovation and green new product development [11,12]. (3) Research on the interrelationship between digitalization and greening. As the national emphasis on digitalization and greening has grown, scholars have explored the relationship between the two from various dimensions. On one hand, many studies have focused on the positive role of digitalization in promoting greening. For example, Cao Yu et al. [13] used resource orchestration theory to analyze the patterns and logic of digitalization driving green transformation in manufacturing enterprises. Liu et al. [14] also confirmed that the digital economy can significantly enhance the greening level of manufacturing. He Xiaogang and Zhong Xiangfei [15] pointed out that enterprises can enhance their green technology levels and energy utilization rates through digital transformation, thereby achieving green upgrading. On the other hand, many scholars have conducted theoretical and empirical research on the logic and level of synergistic development between the two. For example, Zhou Mi and Qiao Yurong [16], Li Xuhui et al. [17], and Li Qilun [18] analyzed the mechanisms of their synergistic development and employed a coupling coordination degree model to measure and analyze the level of synergistic development between digitalization and green transformation. Li Qiang and Tang Youming [19] and Sun Bowen et al. [20] respectively analyzed the impact of information consumption and the construction of national big data comprehensive pilot zones on the synergistic development of the two. Recent studies published in international journals further show that digital inclusive finance, digital economy agglomeration, and the broader digital economy can promote green technological innovation, improve green economic efficiency, and reduce carbon emissions, often with significant spatial spillover or threshold effects [21,22,23]. Compared with the existing literature, these studies highlight that the digital–green relationship is reflected not only in direct technological empowerment but also in spatial transmission, regional heterogeneity, and nonlinear transformation mechanisms.

In summary, research findings on digitalization and greening are increasingly abundant, providing important references for this study. However, existing research primarily focuses on the provincial level, with relatively few studies exploring the spatio-temporal differentiation and evolutionary characteristics of China’s digitalization and greening synergy at the city level. Additionally, while a few studies have examined the convergence characteristics of China’s digitalization and greening integration [17,18], these analyses are primarily based on the club convergence characteristics of China’s two-in-one integration within a geographical zoning framework. This approach fails to accurately capture the influence of economic, policy, and technological factors on the convergence of two-in-one integration, making it difficult to describe the true state of China’s inter-city digitalization and greening integration clustering characteristics. Based on the above analysis, this paper attempts to achieve the following extensions: First, using panel data from 287 Chinese cities, this paper measures and compares the level of synergistic development between digitalization and greening and regional development disparities, which holds significant practical implications for further enhancing the quality of their integration. Second, this paper employs methods such as σ convergence, β convergence, club convergence, and random convergence to validate the convergence characteristics of China’s digitalization and greening from both traditional convergence and spatial convergence dimensions. This helps identify the driving factors behind the level of synergy between the two and provides reference for the government to formulate more targeted policy recommendations.

2. Theoretical Analysis of the Synergistic Development of Digitalization and Greening

Digitalization refers to the transformation process relying on the digital hard and digital soft environments, with the goal of increasing digital innovation output, deepening the application of digital technology, and increasing the income of the digital economy. Greening, on the other hand, is a development process based on green production, green governance, and green living. Digitalization and greening are an important means to promote China’s economic and social development and comprehensive green transformation. From the perspective of system theory, the two complement each other, and there is bound to be a certain interaction.

On the one hand, digitalization empowers green development. The construction of digital infrastructure and the cultivation of digital talents can broaden the information exchange channels between regions, which is conducive to the formation of collaborative innovation networks; enhance the level of technological innovation of enterprises; and provide strong support for greening development [15,24,25]. In addition, digital technology itself has strong green attributes. It can effectively improve regional energy efficiency and environmental governance efficiency and achieve green production and green governance [25,26]. For example, the use of digital media technology can improve the transparency of information, strengthen the public’s supervision of the environmental governance behavior of enterprises and governments [27,28], and promote the development of regional greening. And the application of big data, the Internet of Things, and other technologies can achieve the online monitoring of energy consumption and emissions as well as the visual management of the whole production process, which helps to improve the greening of production operations [29]. Data as a core element in the digitalization process can improve the allocation capacity and utilization efficiency of resources, reduce unnecessary energy consumption, and provide a guarantee for green development. Digital inclusive finance can also ease financing constraints for the green innovation and sustainable development of enterprises and then promote the level of regional greening [25,30,31,32]. In addition, the application of digitalization in social life, such as bike sharing and online offices, can also promote green living and improve the public’s concept and awareness of green development.

On the other hand, greening leads to digital transformation. The region is working towards multiple goals in the process of green development, such as improving the level of green production, improving the efficiency of green governance, and transforming the green way of life, which cannot be separated from the use of digital technologies such as big data and cloud computing, and also puts forward higher requirements for the hard and soft environments for the development of digitalization, which will reverse pull the development of digitalization [29,33]. In addition, with the improvement in China’s digital infrastructure, the growth of arithmetic power, and the acceleration of the replacement rate of digital equipment, the energy consumption and pollution emissions of the digital industry have also increased [34,35]. In the context of green development, this will catalyze the transformation of the digital economy toward high-quality development [36,37]. For example, the government has implemented the “East Data, West Computing” project to further optimize the layout of data centers, integrate computing power resources, achieve the scale and intensive development of national computing power, and promote the green transformation of the national economy.

3. Research Design

3.1. Indicator System Construction

Focusing on the connotation of digitalization and greening, based on the consideration of scientific, comparable, and accessible data, this paper drew on scholars such as Zhou Mi and Qiao Yurong [16], Zhao Jiali and Zhang Xiaoya [34] and Wang Yujin et al. [38] to construct urban digital development from three dimensions, the digital hard environment, digital soft environment, and digital economy index system, and constructed an urban greening development measurement index system from the three dimensions of greening production, greening governance, and greening life, as shown in

Table 1. Evaluation index system of China’s digitalization and greening development level.

Objective Layer	Tier-1 Indicators	Tier-2 Indicators	Tier-3 Indicators
Digitalization	Digital Hardware Environment	Digital Infrastructure	Mobile Phone Penetration Rate
			Internet Penetration Rate
	Digital Software Environment	Digital Talent Support	Percentage Of Employment in Information Transmission, Computer Services and Software Industry
		Digital Policy Support	Term Frequency Proportion of Digital Economy Policies
		Digital Public Attention	Baidu Search Index for Digital Economy
		Digital Innovation Environment	Proportion of Science Expenditure
	Digital Economy	Digital Economy Revenue	Proportion of Telecommunications Business Revenue
			Proportion of Postal Business Revenue
		Digital Financial Inclusion	Digital Inclusive Finance Index
		Digital Innovation Output	Number of Digital Patent Applications
Greening	Green Production	Energy Consumption	Energy Consumption per Unit of GDP
		Pollutant Emission	Sulphur Dioxide (SO2) Emission per Unit of GDP
			Carbon Dioxide (CO2) Emissions per Unit of GDP
			PM2.5 Emissions per Unit of GDP
	Green Governance	Pollution Control	Wastewater Treatment Rate
			Harmless Disposal Rate of Municipal Solid Waste
	Green Life	Greening Area	Green Space Rate of Built-Up Area
			Green Space Area per Capita
		Green Living	Number of Buses per 10,000 People
			Daily Domestic Water Consumption per Capita

.

To further illustrate the reasonableness of the weighting scheme, Appendix A 

Table A1. (a). Entropy weights aggregated by Tier-2 indicators: digitalization subsystem. (b). Entropy weights aggregated by Tier-2 indicators: greening subsystem.

Tier-2 Indicator	2011	2017	2022
(a)
Digital Infrastructure	0.1504	0.1265	0.0729
Digital Talent Support	0.0370	0.0869	0.0723
Digital Policy Support	0.0740	0.0349	0.0316
Digital Public Attention	0.1785	0.2452	0.1768
Digital Innovation Environment	0.0797	0.0825	0.1089
Digital Economy Revenue	0.0910	0.1028	0.1031
Digital Financial Inclusion	0.0198	0.0282	0.0355
Digital Innovation Output	0.3697	0.2932	0.3989
(b)
Energy Consumption	0.0267	0.0167	0.0235
Pollutant Emission	0.0549	0.0783	0.0751
Pollution Control	0.0806	0.0280	0.0671
Greening Area	0.0885	0.1099	0.1025
Green Living	0.7493	0.7671	0.7318

reports the entropy weights aggregated at the Tier-2 level for selected years. Appendix A 

Table A2. Robustness check using equal weights.

Year	Weighting Scheme	Whole Nation	East	Central	West	Regional Ranking
2011	Baseline (entropy weights)	0.2287	0.2555	0.2173	0.2111	East > Central > West
	Equal weights	0.5318	0.5706	0.5164	0.5050	East > Central > West
2017	Baseline (entropy weights)	0.2829	0.3111	0.2684	0.2671	East > Central > West
	Equal weights	0.5342	0.5745	0.5167	0.5082	East > Central > West
2022	Baseline (entropy weights)	0.3173	0.3486	0.3040	0.2966	East > Central > West
	Equal weights	0.5720	0.6090	0.5593	0.5441	East > Central > West

provides an equal-weight robustness check.

3.2. Research Objects and Data Sources

In view of data availability, the research object of this paper covered 287 prefecture-level and above cities in China, with a time span of 2011–2022. The relevant raw data were collected from China Urban Statistical Yearbook, China Urban and Rural Construction Statistical Yearbook, the statistical yearbooks of prefectural cities, Digital Financial Inclusion Index, and the government work reports of each city. For individual missing values, this paper adopted the linear interpolation method to make up for them.

3.3. Research Methods

3.3.1. Global Entropy Value Method

The entropy value method is based on the characteristics of “entropy” in information theory, and it objectively assigns and comprehensively evaluates multiple indicators. Considering that the traditional entropy value method only focuses on cross-sectional data, the evaluation results are not time-comparable. In this paper, we referred to the research of Li Jixia et al. [39] and measured the digitalization and greening level of China from 2011 to 2022 by constructing a three-dimensional time-series data table of “indicator–time–space” and adopting the global entropy value method.

3.3.2. Coupling Coordination Model

The coupling coordination model objectively reflects the degree of mutual promotion and constraints between multiple systems [40], which can more reasonably describe the synergistic development of digitalization and greening in Chinese cities.

3.3.3. Kernel Density Estimation

The kernel density estimation method can estimate the probability density distribution of the sample based on the indicator data, which is widely used to estimate the distribution characteristics of variables [34]. In this paper, the kernel density estimation method is used to conduct an in-depth analysis of the distributional characteristics of and evolutionary trends in the level of synergistic development of digitalization and greening in China.

3.3.4. Dagum Gini Coefficient

The Dagum Gini coefficient is a decomposition method based on subgroups, which can further decompose the level of digitalization and greening synergistic development of Chinese cities into intra-regional differences, inter-regional differences, and the transvariation effect [17] so as to better analyze the spatial differentiation of the level of digitalization and greening synergistic development in China. The specific calculation formula is as follows.

$$G={\displaystyle \frac{{\displaystyle \sum _{j=1}^k{\displaystyle \sum _{h=1}^k{\displaystyle \sum _{i=1}^{n_j}{\displaystyle \sum _{r=1}^{n_h}\left|D_{ji}-D_{hr}\right|}}}}}{2n^2\overline{D}}}$$

(1)

$$G_w={\displaystyle \sum _{i=1}^{n_j}{G}_{jj}}{P}_j{S}_j$$

(2)

$$G_{nb}={\displaystyle \sum _{j=2}^k{\displaystyle \sum _{h=1}^{j-1}{G}_{jh}(}{P}_j{S}_h+}{P}_h{S}_j)D_{jh}$$

(3)

$$G_t={\displaystyle \sum _{j=1}^k{\displaystyle \sum _h^{j-1}{G}_{jh}}}(P_j{S}_h+P_h{S}_j)(1-D_{jh})$$

(4)

where G represents the Gini coefficient, k is the number of sub-regions, n is the number of cities, and $n_j$ and $n_h$ denote the number of cities in sub-regions j and h, respectively. $D_{ji}$ is the synergistic development level of digitalization and greening in city i of sub-region j, and $\overline{D}$ is the mean value of the synergistic development level of digitalization and greening in China. $G_w$, $G_{nb}$ and $G_t$ represent the contributions of intra-regional differences, inter-regional differences, and the transvariation effect, respectively. $P_j={\displaystyle \frac{n_j}{n}}$, and $S_j={\displaystyle \frac{n_j\overline{D_j}}{n\overline{D}}}$. $G_{jj}=\left(\left(1/2\overline{D_j}\right){\displaystyle \sum _{i=1}^{n_j}{\displaystyle \sum _{r=1}^{n_j}\left|D_{ji}-D_{jr}\right|}}\right)/n_j^2$ is the Gini coefficient of the synergistic development level of digitalization and greening in sub-region j. $G_{jh}=\left({\displaystyle \sum _{i=1}^{n_j}{\displaystyle \sum _{r=1}^{n_h}\left|D_{ji}-D_{hr}\right|}}\right)/n_j{n}_h\left(\overline{D_j}+\overline{D_h}\right)$ is the Gini coefficient between sub-regions j and h, representing the disparity in the synergistic development level of digitalization and greening across different sub-regions. $D_{jh}={\displaystyle \frac{d_{jh}-p_{jh}}{d_{jh}+p_{jh}}}$ denotes the mutual influence of the synergistic development level of digitalization and greening between sub-regions j and h.

3.3.5. Spatial Moran’s Index

Moran’s index assumes that cities are homogeneous and can be used to analyze the spatial agglomeration characteristics of the synergistic development of digitalization and greening in China. Drawing on Li et al.’s study [41], this paper analyzed the spatial correlation of the synergistic development of the two in terms of both global and local dimensions.

Global Moran’s I index is calculated as follows:

$$I={\displaystyle \frac{{\displaystyle \sum _{i=1}^n{\displaystyle \sum _{j=1}^n{w}_{ij}(D_i-\overline{D})(D_j-\overline{D})}}}{{\displaystyle \sum _{i=1}^n{\displaystyle \sum _{j=1}^n{w}_{ij}}}{(D_i-\overline{D})}^2}}={\displaystyle \frac{{\displaystyle \sum _{i=1}^n{\displaystyle \sum _{j=1}^n{w}_{ij}(D_i-\overline{D})(D_j-\overline{D})}}}{s^2{\displaystyle \sum _{i=1}^n{\displaystyle \sum _{j=1}^n{w}_{ij}}}}}$$

(5)

where $D_i$ denotes the level of coordinated development of digitalization and greening in city i; $s^2$ represents the variance in the coordinated development level; and $W=(w_{ij})$ denotes the first-order Queen contiguity matrix, defined in Equation (6):

$$w_{ij}=\left\{\begin{array}{ll}1,\hfill & ifregionsiandjareadjacent,\hfill \\ 0,\hfill & otherwise\hfill \end{array}\right.$$

(6)

Local Moran’s I index is calculated as follows:

$$I_i={\displaystyle \frac{(D_i-\overline{D})}{s^2}}{\displaystyle \sum _{j=1}^n{w}_{ij}(D_j-\overline{D})}$$

(7)

A significantly positive Moran’s I indicates that cities with similar levels tend to cluster in space, whereas a significantly negative value indicates spatial dispersion. To further identify local clustering types, this paper also uses a local Moran scatterplot, in which the horizontal axis is the standardized local value z, and the vertical axis is its spatial lag Wz.

3.3.6. Convergence Models

• σ convergence

σ convergence means that the deviation of the level of synergistic development of digitalization and greening in different cities decreases over time, which is commonly measured by the coefficient of variation [42]. The specific calculation equation was constructed as follows.

$$\sigma ={\displaystyle \frac{\sqrt{{\displaystyle \sum _{i=1}^N{(D_i-\overline{D})}^2/N}}}{\overline{D}}}$$

(8)

where σ is the coefficient of variation of the level of synergistic development of digitalization and greening in China, N is the number of sample cities, D_i is the level of synergistic development of digitalization and greening of i city, and $\overline{D}$ is the mean value of the synergistic development level.

2.

β convergence

β convergence originated from economic convergence theory, and it can be divided into absolute β convergence and conditional β convergence. Based on the classical β convergence, further considering the spatial correlation and adding the spatial weight matrix, it can be used to detect whether cities with weak digitalization and greening synergistic development levels in China have higher growth rates and, ultimately, whether the synergistic development levels of all cities in the country achieve convergence [17]. The absolute β convergence model can be expressed as follows.

$$g_{it}=D_{it}-D_{i(t-1)}=\alpha +\beta D_{i(t-1)}+{\mu }_i+v_t+u_{it}$$

(9)

$$g_{it}=\rho W{g}_{it}+\beta D_{i(t-1)}+{\gamma }_1\times W{D}_{i(t-1)}+{\mu }_i+v_t+{\epsilon }_{it}$$

(10)

where $g_{i,t+1}$ denotes the growth rate of the level of synergistic development in city i from t to t + 1; α is a constant term; β is the convergence coefficient; ${\mu }_i$ is the city fixed effect; $v_t$ is the time fixed effect; $u_{it}$ is the random disturbance term; ${\epsilon }_{it}$ is the composite disturbance term in the spatial specification; ρ is the spatial autoregressive coefficient; and γ is the coefficient of the spatially lagged initial term. When $\lambda =0$, it is a spatial Durbin model; when $\lambda =0$ and ${\gamma }_1=0$, it is a spatial autoregressive model; when ${\gamma }_1=0,\delta =0$ and $\rho =0$, it is a spatial error model. Meanwhile, W is the spatial first-order Queen matrix and $\rho$ the spatial lag coefficient. Absolute β convergence exists when the convergence coefficient β is significantly negative, i.e., the growth rate of the level of synergistic development of cities is negatively correlated with the initial level.

Conditional β convergence assumes that the level of synergistic development of cities may be affected by socio-economic factors, and after controlling for the influencing factors, a state of convergence is eventually reached. The conditional beta convergence equation is shown below.

$$\begin{array}{l}{g}_{it}=\alpha +\beta D_{i(t-1)}+{\theta }_{1}PGD{P}_{it}+{\theta }_{2}GO{V}_{it}+{\theta }_{3}FI{N}_{it}+{\theta }_4\ln PA{T}_{it}+{\theta }_{5}L{P}_{it}\\ +{\mu }_i+{\nu }_t+u_{it}\end{array}$$

(11)

$$\begin{array}{l}{g}_{it}=\rho W{g}_{it}+\beta D_{i(t-1)}+{\theta }_{1}PGD{P}_{it}+{\theta }_{2}GO{V}_{it}+{\theta }_{3}FI{N}_{it}+{\theta }_4\ln PA{T}_{it}+{\theta }_{5}L{P}_{it}\\ +{\gamma }_1\times W{D}_{i(t-1)}+{\vartheta }_{1}WPGD{P}_{it}+{\vartheta }_{2}WGO{V}_{it}+{\vartheta }_{3}WFI{N}_{it}+{\vartheta }_{4}W\ln PA{T}_{it}+{\vartheta }_{5}WL{P}_{it}\\ +{\mu }_i+v_t+{\epsilon }_{it}\end{array}$$

(12)

where PGDP_it, COV_it, FIN_it, lnPAT_it, and LP_it are control variables in the conditional β convergence model, which respectively measure the level of the economy, the intensity of government intervention, the efficiency of financial development, the level of technological development, and the productivity of labor in city i.

3.

Club convergence

Club convergence mainly analyzes the convergence of the level of synergistic development of digitalization and greening in cities with similar characteristics, and there was steady-state convergence within different clubs. The test method of club convergence was mostly based on the log t regression test [43,44,45], and the specific equation is as follows.

$$\log \left({\displaystyle \frac{H_1}{H_t}}\right)-2\log \left(\log (t+1)\right)=\widehat{c}+\widehat{\gamma }\log t+{\widehat{u}}_{it}$$

(13)

where H_t is a cross-sectional variation statistic at time t, which measures the dispersion of all sample units relative to the common growth path; h_it denotes the relative transition parameter of city i in period t, which measures the relative position of city i compared with the panel average; $\widehat{c}$ is the estimated intercept term; $\widehat{\gamma }$ is the estimated convergence coefficient; and ${\widehat{u}}_{it}$ is the regression residual. In general, H_t can be expressed as follows:

$$H_t={\displaystyle \frac{1}{N}}{\displaystyle \sum _{i=1}^N{\left(h_{it}-1\right)}^2}$$

(14)

and the relative transition parameter is defined as

$$h_{it}={\displaystyle \frac{D_{it}}{\frac{1}{N}{\displaystyle \sum _{i=1}^N{D}_{it}}}}$$

(15)

When t < −1.65, it indicates that there is club convergence in the synergistic development of digitalization and greening in China.

4.

Stochastic convergence

The use of the stochastic convergence method can avoid the transitional state of convergence and non-convergence that exists in the level of synergistic development of digitalization and greening in China, and stochastic convergence analysis can generally be carried out by the method of the unit root test [46]. If the level of synergistic development does not have a unit root, it indicates that it exhibits convergence in the long term. Considering the ambiguity of the assumptions and interpretations of the panel unit root test results, this paper adopted the three unit root test methods of Harris–Tzavalis, Levine–Lin–Chu, and Im–Pesaran–Shin to test the stochastic convergence phenomenon of China’s digitalization and greening synergistic development level in order to obtain more robust validation results.

In this paper, convergence refers to whether the inter-city gap in digital–green coordinated development tends to narrow over time or whether cities with similar initial conditions move toward a common or club-specific steady-state path. It is examined from four complementary perspectives: σ convergence tests whether dispersion decreases over time; β convergence tests whether cities with lower initial levels grow faster; club convergence tests whether different groups of cities converge toward different steady states; and stochastic convergence tests whether the deviation from the long-run path is stationary.

4. Typical Characteristics of Synergistic Development of Digitalization and Greening in Chinese Cities

4.1. Characteristic Analysis of Time-Series Evolution

Based on the measurement results, Figure 1 provides the trend in the digitalization and greening synergistic development level in China and the three major regions from 2011 to 2022. It can be seen that the synergistic development level of digitalization and greening in China mainly exhibits an evolutionary trend from low synergistic development to medium synergistic development. The proportion of cities with a low level of synergistic development decreases from 91.99% in 2011 to 53.31% in 2022. In addition, there is an increase in the number of cities with higher levels of synergistic development; six cities, namely, Hangzhou, Dongguan, Guangzhou, Shanghai, Beijing, and Shenzhen, are at higher levels of synergistic development in 2020–2022. As for the cities with a high level of synergistic development, no city in the country has yet exceeded 0.8 on the coupling coordination index.

In Figure 2, the overall levels of synergistic development of digitalization and greening in China both show a steady upward trend. During the sample period, the national synergistic development level ranged from 0.229 to 0.317, with an average annual growth rate of 3.02%, but there is still much room for improvement. In addition, the level of synergistic development between digitalization and greening in China is spatially uneven. From 2011 to 2022, the average value of the level of synergistic development in the eastern region is 0.307, followed by the central region (0.263) and the western region (0.260). In terms of growth rate, the level of synergistic development in the western region increases the most significantly, with an average annual growth rate of 3.14%, a total increase of 0.085. The growth rates of synergistic development in the central and eastern regions are 3.10% and 2.86%, which show convergence characteristics to a certain extent.

The upward movement from 0.229 in 2011 to 0.317 in 2022 indicates that China’s digital–green coordination entered a sustained improvement stage, but the absolute level still remained in the low-to-medium synergy range. In substantive terms, the improvement was driven more by gradual diffusion than by a full transition to high coordination, because no city exceeded 0.8 on the coupling coordination index and the share of low-level synergy cities still remained above one half in 2022. The east maintained a clear first-mover advantage, while the west recorded the fastest growth rate and the largest absolute increase, implying that the national pattern was characterized by “overall improvement with partial catch-up” rather than by complete regional equalization.

From the kernel density estimation of the level of synergistic development (see Figure 3), the kernel density peak moves rightward from 2011 to 2022, indicating that the overall level of synergistic development shows a dynamic evolutionary trend of increasing year by year. In addition, the kernel density curves for each year show a right-skewed curve, with the peaks occurring in places with low coupling coordination, indicating that the level of synergistic development of digitalization and greening is still low in most cities across the country. Compared with the kernel density curve in 2011, the wave width in 2022 is significantly narrower, indicating that the absolute difference in the level of digital–green synergistic development in China is gradually decreasing.

4.2. Characteristic Analysis of Spatial Differentiation

According to

Table 2. Gini coefficient of synergistic development of digitalization and greening in China.

Year	Overall Gini Coefficient	Intra-Regional Gini Coefficient			Inter-Regional Gini Coefficient
		East	Central	West	East–Central	East–West	Central–West
2011	0.102	0.116	0.066	0.087	0.102	0.116	0.077
2012	0.098	0.114	0.060	0.085	0.099	0.112	0.072
2013	0.095	0.113	0.057	0.079	0.096	0.109	0.068
2014	0.094	0.113	0.058	0.076	0.096	0.108	0.067
2015	0.092	0.112	0.056	0.077	0.095	0.107	0.067
2016	0.087	0.111	0.052	0.070	0.092	0.102	0.061
2017	0.092	0.117	0.056	0.073	0.096	0.106	0.065
2018	0.095	0.120	0.059	0.076	0.100	0.109	0.067
2019	0.097	0.120	0.064	0.078	0.102	0.110	0.071
2020	0.099	0.118	0.065	0.076	0.101	0.113	0.073
2021	0.098	0.116	0.065	0.077	0.101	0.112	0.073
2022	0.094	0.111	0.065	0.074	0.097	0.105	0.071
Mean	0.095	0.115	0.060	0.077	0.098	0.109	0.069

, the overall Gini coefficient of China’s digitalization and greening synergistic development from 2011 to 2022 shows an “inverted N-shaped” trend of first declining, then increasing, and then declining. During the sample period, the overall Gini coefficient of the level of synergistic development of digitalization and greening in China ranges from 0.087 to 0.102, with a mean value of 0.095. From 2016 to 2020, the Gini coefficient shows a small increase. In terms of intra-regional differences, the eastern region has the largest difference in the level of digital–green synergistic development, with a mean value of 0.115, followed by the west (0.077) and the center (0.060). This indicates that the eastern region is not only the most advanced area but also the most internally differentiated one: frontier cities such as Beijing, Shanghai, and Shenzhen have moved ahead more quickly, while some eastern cities still remain at a relatively low integration level. In terms of inter-regional differences, the average values of the Gini coefficients of east–west and east–central in 2011–2022 are 0.109 and 0.098, respectively, which are higher than the national average values, which suggests that regional inequality is still primarily shaped by the leading role of the eastern region. However, their gradual decline also shows that the relative gap between the east and the rest of the country has been narrowing, even though the process is slow and nonlinear.

Figure 4 illustrates the contribution rate of regional differences in digital–green synergistic development in China. It can be found that the contribution rate of inter-regional differences has been at a leading level throughout the sample period but shows an overall decreasing trend. The mean values of the contribution rate of hypervariable density and intra-regional differences are 29.07% and 30.52%, respectively, showing fluctuating and rising change characteristics, which indicates that while focusing on the inter-regional differences in the level of digital–green synergistic development, it is also necessary to be vigilant about the status quo of urban development where the level of synergistic development is relatively lagging behind that of the different intra-regional areas.

4.3. Analysis of Spatial Agglomeration Characteristics

Table 3. Spatial autocorrelation analysis of level of synergistic development of digitalization and greening in China.

Year	Global Moran’s I	Z	p
2011	0.276 ***	7.363	0.000
2012	0.279 ***	7.420	0.000
2013	0.261 ***	6.956	0.000
2014	0.263 ***	6.970	0.000
2015	0.308 ***	8.164	0.000
2016	0.308 ***	8.184	0.000
2017	0.295 ***	7.801	0.000
2018	0.272 ***	7.213	0.000
2019	0.250 ***	6.593	0.000
2020	0.273 ***	7.171	0.000
2021	0.279 ***	7.291	0.000
2022	0.276 ***	7.181	0.000

*** p < 0.01.

reports the estimation results of the spatial Moran index for the synergistic development of greening and digitalization in China. Overall, the Global Moran’s I values calculated based on the spatial adjacency matrix are statistically significant and positive throughout the study period. This demonstrates the presence of positive spatial autocorrelation in China’s digital–green synergistic development. Specifically, cities with high (low) synergy development levels tend to be adjacent to others exhibiting similarly high (low) levels.

During the period 2011–2022, Global Moran’s I fluctuated centered around a mean value of 0.278, indicating that the spatial dependence pattern remained relatively stable over the period.

This paper further plotted a Moran scatterplot of China’s digital–green synergistic development between 2011 and 2022 to observe the evolution of agglomeration in each city (see Figure 5). The dominance of “low–low” clustering in the third quadrant indicates that the lagging status of some cities is not isolated but spatially embedded. From a policy perspective, this means that the problem is not simply one of insufficient digital infrastructure or solely one of inefficient environmental policies; rather, it reflects the co-location of weak digital foundations, limited innovation capacity, relatively low labor productivity, and insufficiently effective green governance. In such areas, neighboring cities may reinforce one another’s low-level equilibrium, making it difficult for a single city to break through by relying only on isolated local measures. By contrast, the increase in the number of “high–high” cities from 51 in 2011 to 54 in 2022 suggests that developed areas are gradually forming stronger positive spillover networks. Therefore, the prevalence of low–low clustering has an important policy implication: reducing spatial lock-in in lagging areas requires coordinated regional intervention, including digital infrastructure sharing, cross-city environmental governance collaboration, and differentiated support policies targeted at contiguous low-synergy areas.

5. Convergence Analysis of Synergistic Development of Digitalization and Greening in Chinese Cities

5.1. σ Convergence Analysis

Figure 6 depicts the trend in variation in the σ convergence of China’s digital–green synergistic development index from 2011 to 2022. It can be seen that the coefficients of variation of the digital–green synergistic development index in the national and eastern regions show a decreasing trend during the study period, with obvious σ convergence characteristics. The σ convergence coefficient of the digital–green synergistic development level in the central and western regions shows a decreasing characteristic in 2011–2016 but shows yearly growth in 2016–2021, without an obvious σ convergence characteristic. This may be due to the fact that the differences in the economic and technological basis of cities in the central and western regions and the degree of response to policies at the early stage of the construction of “Digital China” exacerbated the phenomenon of unbalanced synergistic development of digitalization and greening [16], which led to the growth rate of each city being different, resulting in an increase in intra-regional disparity.

5.2. β Convergence Analysis

5.2.1. Absolute β Convergence Analysis

In

Table 4. National absolute β convergence test.

	Traditional Absolute β Convergence	Spatial Absolute β Convergence
	Two-Way Fixed Effects	SAR	SEM	SDM
DGi(t−1)	−0.415 ***	−0.413 ***	−0.423 ***	−0.358 ***
	(0.015)	(0.014)	(0.014)	(0.013)
WDGi(t−1)				0.309 ***
				(0.015)
ρ		0.0836 ***		0.289 ***
		(0.024)		(0.022)
Constant	0.122 ***
	(0.004)
Entity FE	Y	Y	Y	Y
Year FE	Y	Y	Y	Y
N	3157	3157	3157	3157
R2	0.311	0.085	0.091	0.099
Log-likelihood		10,285.362	10,294.151	10,138.680
LM-Lag test				13.739 ***
Robust LM-Lag test				2.706 *
LM-Error test				11.061 ***
Robust LM-Error test				0.028
Wald test: SDM degenerates to SAR				15.180 ***
LR test: SDM degenerates to SAR				15.050 ***
Wald test: SDM degenerates to SEM				3.660 *
LR test: SDM degenerates to SEM				−2.530

*** p < 0.01; * p < 0.1. The SEs are in parentheses.

, the original hypothesis of no spatial lag and no spatial error effect was rejected by the results of the LM-Lag and LM-Error tests, respectively. The statistics of the Wald and LR tests indicated that the results estimated by the spatial Durbin model are superior. In addition, regardless of whether the spatial effect is considered or not, the convergence coefficient β of the development level of digital–green synergy is statistically significant at the 1% level, indicating a robust catch-up process in China’s digital–green synergistic development. In economic terms, this means that cities with a weaker initial coordination level tended to grow faster, so the development gap narrowed over time. In addition, the spatial autocorrelation coefficient is significantly positive, indicating that there may be a spatial spillover effect of the increase in the level of digital–green synergy. Therefore, convergence in this study is not purely an internal adjustment process within each city but also a spatially interconnected process shaped by inter-city diffusion.

Table 5. Sub-regional absolute β convergence test.

	Traditional Absolute β Convergence			Spatial Absolute β Convergence
	East	Central	West	East	Central	West
DGi(t−1)	−0.314 ***	−0.382 ***	−0.606 ***	−0.326 ***	−0.394 ***	−0.607 ***
	(0.023)	(0.024)	(0.031)	(0.022)	(0.023)	(0.029)
WDGi(t−1)				0.098 ***	0.145 ***	−0.045
				(0.034)	(0.047)	(0.040)
ρ				0.102 **	0.108 **	−0.015
				(0.040)	(0.043)	(0.041)
Constant	0.103 ***	0.107 ***	0.163 ***
	(0.007)	(0.006)	(0.008)
Entity FE	Y	Y	Y	Y	Y	Y
Year FE	Y	Y	Y	Y	Y	Y
N	1100	1100	957	1100	1100	957
R2	0.339	0.334	0.358	0.097	0.091	0.126
Log-likelihood				3642.037	3706.100	3043.017

*** p < 0.01; ** p < 0.05. The SEs are in parentheses.

reports the estimation results of the absolute β convergence test for the sub-regions. It can be seen that the convergence coefficients of the eastern, central, and western regions are significantly negative in both the traditional absolute β convergence model and the spatial absolute β convergence model, presenting absolute β convergence characteristics. Combined with the convergence speed, it is found that the western region has the fastest convergence speed, and the corresponding convergence period is smaller. The convergence speeds of the central and eastern regions are smaller than the national level, which may be due to the fact that the development level of urban digitalization and greening in the eastern and central regions differs greatly, resulting in a slower convergence speed than that of the west and the national average.

5.2.2. Conditional β Convergence Analysis

Considering that the convergence of the synergistic development of digitalization and greening may change when it is affected by control variables, this paper further tested the conditional β convergence characteristics of the whole country and each region (see

Table 6. National conditional β convergence test.

	Traditional Conditional β Convergence	Spatial Conditional β Convergence
	Two-Way Fixed Effects	SAR	SEM	SDM
				Main Effect	Spatial Effect
DGi(t−1)	−0.430 ***	−0.429 ***	−0.436 ***	−0.439 ***	0.075 ***
	(0.015)	(0.014)	(0.014)	(0.014)	(0.026)
PGDPit	0.044 ***	0.044 ***	0.047 ***	0.066 ***	−0.063 *
	(0.017)	(0.016)	(0.016)	(0.020)	(0.033)
GOVit	0.049 ***	0.047 ***	0.043 ***	0.024	0.080 ***
	(0.015)	(0.014)	(0.015)	(0.015)	(0.025)
FINit	0.003 ***	0.004 ***	0.003 ***	0.004 ***	−0.001
	(0.001)	(0.001)	(0.001)	(0.001)	(0.002)
lnPATit	0.001 *	0.001 *	0.001 *	0.001 *	−0.000
	(0.000)	(0.000)	(0.000)	(0.000)	(0.001)
LPit	0.009 ***	0.009 ***	0.009 ***	0.008 ***	0.008 **
	(0.002)	(0.002)	(0.002)	(0.002)	(0.003)
ρ		0.078 ***			0.096 ***
		(0.023)			(0.025)
Constant	0.106 ***
	(0.005)
Entity FE	Y	Y	Y	Y
Year FE	Y	Y	Y	Y
N	3157	3157	3157	3157
R2	0.325	0.101	0.107	0.119
Log-likelihood		10,317.500	10,322.670	10,330.180
LM-Lag test				21.634 ***
Robust LM-Lag test				11.449 ***
LM-Error test				12.255 ***
Robust LM-Error test				2.069
Wald test: SDM degenerates to SAR				25.500 ***
LR test: SDM degenerates to SAR				25.360 ***
Wald test: SDM degenerates to SEM				19.420 ***
LR test: SDM degenerates to SEM				15.030 **

*** p < 0.01; ** p < 0.05; * p < 0.1. The SEs are in parentheses.

and

Table 7. Sub-regional conditional β convergence tests.

	Traditional Conditional β Convergence			Spatial Conditional β Convergence
	East	Central	West	East		Central		West
				Main Effect	Spatial Effect	Main Effect	Spatial Effect	Main Effect	Spatial Effect
DGi(t−1)	−0.320 ***	−0.428 ***	−0.647 ***	−0.336 ***	0.144 ***	−0.437 ***	0.080	−0.648 ***	−0.032
	(0.024)	(0.025)	(0.031)	(0.022)	(0.038)	(0.023)	(0.053)	(0.029)	(0.048)
PGDPit	0.022	0.136 ***	0.040	0.040	−0.049	0.186 ***	−0.188 **	0.024	0.066
	(0.019)	(0.042)	(0.048)	(0.026)	(0.043)	(0.046)	(0.074)	(0.047)	(0.069)
GOVit	0.075 ***	0.042 *	−0.006	0.003	0.112 **	0.036	0.025	−0.000	−0.026
	(0.027)	(0.024)	(0.028)	(0.033)	(0.048)	(0.025)	(0.038)	(0.027)	(0.044)
FINit	0.004	0.001	0.016 ***	0.012 ***	−0.033 ***	0.001	0.006 **	0.020 ***	−0.011 **
	(0.004)	(0.001)	(0.004)	(0.004)	(0.007)	(0.001)	(0.003)	(0.004)	(0.006)
lnPATit	−0.001	0.002 ***	0.001	0.000	−0.002	0.002 **	0.0004	0.000	−0.000
	(0.001)	(0.001)	(0.001)	(0.001)	(0.002)	(0.001)	(0.001)	(0.001)	(0.001)
LPit	0.004	0.010 **	0.013 ***	0.004	0.006	0.007 *	0.011	0.014 ***	−0.006
	(0.003)	(0.004)	(0.004)	(0.003)	(0.005)	(0.004)	(0.007)	(0.004)	(0.007)
ρ					0.090 **		0.069		−0.006
					(0.041)		(0.044)		(0.041)
Constant	0.0981 ***	0.0913 ***	0.151 ***
	(0.010)	(0.007)	(0.010)
Entity FE	Y	Y	Y	Y		Y		Y
Year FE	Y	Y	Y	Y		Y		Y
N	1100	1100	957	1100		1100		957
R2	0.347	0.366	0.384	0.112		0.121		0.15
Log-likelihood				3665.0277		3734.4418		3065.2759

*** p < 0.01; ** p < 0.05; * p < 0.1. The SEs are in parentheses.

). It can be seen that the convergence coefficients β for the whole country and the three major regions of the east, middle, and west are consistent with Table 4 and Table 5 in terms of sign and significance, indicating that there are significant conditional β characteristics of the level of digital–green synergistic development in the whole country and the three major regions of the east, middle, and west and that the gap of the city’s development will be gradually narrowed after taking into account the control variables. Combined with the convergence speed, it is found that the conditional β convergence speed is higher than the absolute β convergence for the whole country and regions, indicating that the introduction of control variables has a facilitating effect on the realization of development convergence.

In terms of control variables, without considering the spatial effect, the impact coefficients of GDP per capita, government intervention, financial development efficiency, technological innovation, and labor productivity are significantly positive, indicating that there is a positive relationship between these factors and the level of synergistic development of digitalization and greening, which is conducive to promoting the synergistic development of digitalization and greening in cities to converge to a steady-state level. After considering the spatial effects, it can be seen that the spatial effects of government intervention and labor productivity are significantly positive, while the spatial effect of GDP per capita is significantly negative at the 10% level. Specifically for each region, the growth rate of the synergistic development of the two regions in the eastern region is mainly affected by government intervention and financial development efficiency. For the central region, the increase in GDP per capita, patent applications, and labor productivity in the region significantly contributes to the increase in the level of digital–green synergistic development, while the economic and financial development levels of the surrounding cities have positive and negative effects, respectively. The estimated coefficients of the main effect and spatial effect of financial development efficiency in the western region are 0.020 and −0.011, respectively, statistically significant at the 5% level. One possible reason for this is that cities with higher levels of financial development in the west have a siphoning effect, which affects the digitalization and greening of the surrounding areas, causing changes in the growth rate of synergistic development. In addition, improving local labor productivity is an important guideline for the convergence of digital–green synergistic development in the western region.

5.3. Club Convergence Analysis

From the test results of β convergence, there is a significant convergence trend in the level of synergistic development of digitalization and greening in the east, central, and west regions, i.e., it shows club convergence characteristics [44]. However, artificially dividing cities into three major regions and conducting the club convergence test are somewhat subjective and ignore the strategy of synergistic development among cities. Therefore, this paper further analyzed the possible club convergence characteristics of China’s digital–green synergistic development level with the help of the log t regression test, and the results are shown in

Table 8. Club convergence tests.

Clubs	Number of Cities	Cities	Ratio	t	Average Level of Synergy
Club 1	10	Shanghai, Hangzhou, Changsha, Guangzhou, Dongguan, Nanjing, Suzhou, Hefei, Wuhan, Chengdu	−0.090	−1.300	0.427
Club 2	18	Tianjin, Shijiazhuang, Wuxi, Ningbo, Fuzhou, Xiamen, Nanchang, Jinan, Qingdao, Zhengzhou, Zhuhai, Foshan, Zhongshan, Sanya, Chongqing, Guiyang, Kunming, Xi’an	−0.046	−0.954	0.371
Club 3	6	Hohhot, Shenyang, Dalian, Changchun, Lanzhou, Urumqi	0.695	5.261	0.344
Club 4	41	Taiyuan, Harbin, Changzhou, Nantong, Wenzhou, Jiaxing, Huzhou, Shaoxing, Jinhua, Wuhu, Huizhou, Jieyang, Haikou, Xining, Baoding, Langfang, Xuzhou, Yancheng, Yangzhou, Zhenjiang, Taizhou, Zhoushan, Taizhou, Lu’an, Quanzhou, Yingtan, Ganzhou, Zibo, Yantai, Weifang, Weihai, Luoyang, Hebi, Nanyang, Shantou, Nanning, Lhasa, Jiayuguan, Jinchang, Yinchuan, Keramay	−0.072	−1.314	0.300
Club 5	102	Tangshan, Qinhuangdao, Handan, Xingtai, Zhangjiakou, Chengde, Cangzhou, Hengshui, Yangquan, Jincheng, Linfen, Lvliang, Baotou, Wuhai, Ordos, Anshan, Jinzhou, Yingkou, Panjin, Qiqihar, Daqing, Lianyungang, Huai’an, Suqian, Quzhou, Lishui, Bengbu, Ma’anshan, Huaibei, Tongling, Anqing, Huangshan, Chuzhou, Fuyang, Suzhou, Bozhou, Xuancheng, Putian, Zhangzhou, Longyan, Ningde Pingxiang, Jiujiang, Xinyu, Ji’an, Yichun, Fuzhou, Shangrao, Dongying, Jining, Tai’an, Rizhao, Linyi, Dezhou, Liaocheng, Binzhou, Heze, Kaifeng, Pingdingshan, Xinxiang, Jiaozuo, Xuchang, Luohe, Sanmenxia, Shangqiu, Xinyang, Zhoukou, Zhumadian, Huangshi, Shiyan, Yichang, Xiangyang, Ezhou, Jingmen, Xianning, Zhuzhou, Xiangtan, Hengyang, Yueyang, Yiyang, Chenzhou, Huaihua, Jiangmen, Meizhou, Shanwei, Liuzhou, Guilin, Panzhihua, Mianyang, Neijiang, Zunyi, Anshun, Tongren, Yuxi, Lijiang, Tongchuan, Baoji, Xianyang, Pingliang, Longnan, Shizuishan, Zhongwei	−0.017	−0.605	0.262
Club 6	101	Datong, Changzhi, Jinzhong, Yuncheng, Xinzhou, Chifeng, Tongliao, Bayannur, Ulanqab, Fushun, Benxi, Dandong, Fuxin, Liaoyang, Tieling, Chaoyang, Huludao, Jilin, Siping, Liaoyuan, Tonghua, Baishan, Songyuan, Jixi, Hegang, Shuangyashan, Yichun, Jiamusi, Qitaihe, Mudanjiang, Suihua, Huainan, Chizhou, Sanming, Nanping, Jingdezhen, Zaozhuang, Anyang, Puyang, Xiaogan, Jingzhou, Huanggang, Suizhou, Shaoyang, Changde, Zhangjiajie, Yongzhou, Loudi, Shaoguan, Zhanjiang, Maoming, Zhaoqing, Heyuan, Yangjiang, Qingyuan, Chaozhou, Yunfu, Wuzhou, Beihai, Fangchenggang, Qinzhou, Guigang, Yulin, Baise, Hezhou, Laibin, Chongzuo, Zigong, Luzhou, Deyang, Guangyuan, Suining, Leshan, Nanchong, Meishan, Yibin, Dazhou, Ya’an, Ziyang, Liupanshui, Bijie, Qujing, Baoshan, Zhaotong, Pu’er, Lincang, Weinan, Yan’an, Hanzhong, Yulin, Ankang, Shangluo, Baiyin, Tianshui, Wuwei, Zhangye, Jiuquan, Qingyang, Dingxi, Wuzhong, Guyuan	−0.041	−1.385	0.244
Club 7	6	Hulunbuir, Baicheng, Heihe, Hechi, Guang’an, Bazhong	0.476	6.501	0.234
Diffusion	3	Beijing, Shuozhou, Shenzhen	−0.881	−27.968	0.499

. It can be seen that, after the initial identification and merger test, China’s digital–green synergistic development level can constitute seven different convergence clubs and one divergence club. The first club contains 10 first-tier and new first-tier cities such as Shanghai, Hangzhou, and Changsha, and the development level of digitalization and greening in the cities in this club is high, and the mean value of the coupled coordination degree of amalgamation is 0.427, which is in the significant leading level. The second club contains 18 cities such as Tianjin, Shijiazhuang, Wuxi, and Ningbo, and the coefficients of the estimated values are negative, which is not significantly different from 0, and there is weak conditional convergence. The third club contains fewer cities, with a log t statistic significantly higher than −1.65, and the average value of water for digital–green synergistic development for the included cities is 0.344. The fourth club contains a total of 41 cities, with a level of digital–green synergistic development slightly higher than the national average. Club 5 and Club 6 contain 35.54% and 35.19% of the cities in the study sample, respectively, and most of the cities are located in the central and western regions of China, with the level of synergistic development of amalgamation being significantly lower than the national average. In addition, the growth rate of the integration level of Club 5 is greater than that of the national average and that of Club 6. The number of cities included in Club 7 and the dispersed club is 6 and 3, respectively. Among them, Beijing and Shenzhen are both in the dispersed club, which may be due to the fact that the level of synergistic integration of Beijing and Shenzhen is much higher than that of other regions during the sample period, compared to which the level of development of Shuozhou has been in the lagging region for a long period of time. The club structure reflects differentiated development trajectories rather than simple nationwide convergence.

5.4. Stochastic Convergence Analysis

Table 9. Stochastic convergence test.

	Harris–Tzavalis		Levin–Lin–Chu		Im–Pesaran–Shin
Whole Nation	z	−12.396 ***	t*	−17.182 ***	$\overline{t}$	−12.482 ***
	p	0.000	p	0.000	p	0.000
East	z	−1.998 ***	t*	−4.894 ***	$\overline{t}$	−3.856 ***
	p	0.000	p	0.000	p	0.000
Central	z	−8.613 ***	t*	−11.689 ***	$\overline{t}$	−8.755 ***
	p	0.000	p	0.000	p	0.000
West	z	−10.549 ***	t*	−11.697 ***	$\overline{t}$	−8.888 ***
	p	0.000	p	0.000	p	0.000

*** p < 0.01.

reports the estimation results of the three unit root tests of Harris–Tzavalis, Levine–Lin–Chu, and Im–Pesaran–Shin to discern the stochastic convergence characteristics of China and the three regions. It can be seen that the p-values of all three tests are significantly less than 0.01, indicating that stochastic convergence exists for the synergistic development of digitalization and greening in the whole country and in the east, central, and west regions. This result is consistent with the previous conclusion of β convergence.

6. Conclusions and Policy Implications

In order to promote the synergistic development of digitalization and greening in China, this paper measured and analyzed the level of digital–green synergistic development of China and the trend in spatio-temporal evolution by adopting methods employing the coupling coordination degree, kernel density estimation, and Dagum Gini coefficient based on the construction of a theoretical framework and tested the convergence characteristics by means of a convergence model. The main conclusions are as follows: (1) The level of synergistic development of digitalization and greening in China increased steadily during the sample period, with the national coupling coordination index rising from 0.229 in 2011 to 0.317 in 2022, an increase of 0.088. At the same time, the proportion of cities at the low-synergy stage declined from 91.99% to 53.31%, a decrease of 38.68 percentage points. Nevertheless, no city exceeded 0.8 on the coupling coordination index, which means that the overall pattern is still characterized by low-to-medium synergy rather than high-quality coordination. Regionally, the average coordination level was 0.307 in the east, 0.263 in the center, and 0.260 in the west. The western region recorded the fastest average annual growth rate (3.14%) and the largest absolute increase (0.085), indicating a certain catch-up tendency. In terms of disparity, the overall Gini coefficient changed from 0.102 in 2011 to 0.094 in 2022, while the east–west and east–central inter-regional Gini coefficients declined from 0.116 to 0.105 and from 0.102 to 0.097, respectively, which suggests that regional gaps persisted but narrowed overall. (2) The level of synergistic development of digitalization and greening in China exhibited significant positive spatial autocorrelation throughout the sample period, with Global Moran’s I ranging from 0.250 to 0.308 and averaging about 0.278. The agglomeration pattern was dominated by “low–low” clustering, which indicates a spatial lock-in effect in lagging areas, while the number of “high–high” cities increased from 51 to 54, suggesting a strengthening spillover network among advanced cities. (3) China’s digital–green synergistic development shows significant σ convergence, β convergence, club convergence, and stochastic convergence. Financial development, technological innovation, government intervention, and labor productivity all play important roles in shaping the speed and pathway of convergence.

Based on the above conclusions, this paper puts forward policy recommendations as follows:

First, attach great importance to the regional differences in the synergistic development of urban digitalization and greening, and promote a steady improvement in the synergy between the two. At this stage, the level of China’s synergistic development of digitalization and greening is still relatively low, and the differences between cities are obvious. The government should further increase the support and cultivation of cities such as Beijing, Shanghai, and Shenzhen and fully explore their advanced development experience so as to give play to the “leader” effect and provide demonstration and radiation for cities with weaker levels of synergistic development. Regions with a weak initial level should endeavor to improve their own synergy level and seize the opportunity to overtake the others.

Second, strengthen government guidance and scientific and technological innovation to provide effective support for digital–green synergistic development. The government should increase policy support and provide guidance and support to enterprises and regions for the integration of digitalization and greening through subsidies and tax reductions. In addition, scientific and technological innovation and labor productivity improvement are key to promoting the synergistic development of digitalization and greening. Therefore, it is necessary to strengthen the research and development and transformation of new technologies such as intelligent manufacturing and new energy and pay attention to the cultivation and introduction of complex talents so as to provide technical support and talent guarantee for the synergistic development of digitalization and greening.

Thirdly, focus on regional heterogeneity and implement policies to promote digital–green synergistic development according to local conditions. The central government should strengthen the top-level design, combine the level of synergistic development of digitalization and greening and the convergence characteristics of each city, and explore the path of digital–green synergistic enhancement with regional characteristics, such as the need to pay attention to the spatial effect of government investment in the eastern region, and pay more attention to improving the labor productivity of local employees in the central and western regions.