Multidimensional Measurement and Temporal and Spatial Interaction Characteristics of Rural E-Commerce Development Capacity in the Context of Rural Revitalization

Abstract

With the implementation of the rural revitalization strategy, rural e-commerce has become an essential means of promoting rural economic development and increasing farmers’ income. However, the development of rural e-commerce varies significantly among different regions. Based on the perspective of “three rural areas”, this study constructs a rural e-commerce development capability measurement system centered on readiness, utilization, and influence. It adopts a panel vector autoregressive model to identify key influencing factors. Through the exploratory spatiotemporal data analysis (ESTDA) method, the spatiotemporal dynamic characteristics of rural e-commerce development capacity and the interaction relationship between provinces and regions are revealed. The study shows that (1) China’s rural e-commerce development capacity gained significant improvement from 2011 to 2022, but provincial polarization is evident, with eastern and central provinces leading and western and marginal provinces lagging; the rural e-commerce development capacity shows a decreasing dynamic pattern from the east to the central and western to the northeastern regions. (2) The eastern region has active rural e-commerce development, stable spatial structure, and provincial solid correlation, which creates a significant linkage effect. The western region shows strong internal spatial dependence, the district cross-regional interaction and linkage effect are beginning to emerge, and the northeastern low-development provinces are challenging to leap to a higher level in the short term; (3) the spatiotemporal interaction network of rural e-commerce development among several provinces and regions shows a positive synergistic relationship, and it is an essential consideration for the high-quality development of rural e-commerce to strengthen regional cooperation and realize complementary advantages. The study results provide a theoretical basis for formulating differentiated regional e-commerce development policies, which can help enhance regional synergy and narrow the regional development gap.

1. Introduction

The issue of agriculture, rural areas, and farmers holds paramount importance for the national economy and the well-being of the populace [1]. It has consistently garnered considerable attention from both the Party and the government. The implementation of the rural revitalization strategy, aimed at tackling the “three rural” issues, involves comprehensively enhancing rural society, economy, and ecological environment. This strategy also seeks to promote integrated urban–rural development, marking a significant decision stemming from the 19th National Congress of the Communist Party of China [2,3]. Since the advent of the 21st century, amid China’s rapid industrialization and urbanization, the regional development gap between urban and rural areas has persistently widened, presenting “rural decline” as a global challenge [4]. Examining the global experiences of representative countries in rural development reveals a common trend: the pursuit of government-led rural development “from top to bottom” and the empowerment of rural residents “from bottom to top” [5,6,7,8,9]. In the context of China’s substantial population and limited per capita land area, the evolution of rural development should transition gradually from external to internal mechanisms. It is imperative to delineate the contemporary requisites of the rural revitalization strategy, which aims to comprehensively rejuvenate rural areas with a central focus on people-centric initiatives [10].

Advancing the comprehensive development of rural revitalization, rural e-commerce plays a vital role in addressing the “three rural issues” (agriculture, rural areas, and farmers). By utilizing the “Internet+” platform and integrating advanced technologies such as big data, cloud computing, and artificial intelligence, rural e-commerce can accurately assess market demand, optimize agricultural supply chain management, and enhance production efficiency, thereby injecting new vitality into agricultural economic development. Additionally, through data analysis, farmers are empowered to understand consumer preferences in real time, allowing for timely adjustments to planting structures and production methods. This continuous improvement of product competitiveness enables the market to be optimally satisfied. Moreover, it accelerates the achievement of “prosperous farmers, strong agriculture, and beautiful rural areas [11]”, thus serving as a key driver in the implementation of the rural revitalization strategy [12].

In recent years, rural e-commerce in China has experienced rapid development, marked by significant growth in retail sales. It has emerged as a crucial driver of rural industrial development. In 2023, national rural online retail sales surged to CNY 2.49 trillion, reflecting a 12.9% increase, while rural physical online retail sales totaled CNY 2.27 trillion, marking a 12.1% increase. The proliferation of Taobao villages surged significantly from 3 in 2009 to 7780 in 2022. In 2014, the Ministry of Commerce, alongside other pertinent departments, vigorously spearheaded comprehensive demonstration projects on rural e-commerce, striving to achieve full e-commerce coverage in national poverty-stricken counties. By 2019, online retail sales in these counties reached CNY 239.2 billion, marking a remarkable 33% year-on-year increase, contributing to the employment and income augmentation of 5 million farmers in impoverished areas. Rural e-commerce has proven instrumental in facilitating targeted poverty alleviation initiatives and has charted new pathways for advancing rural revitalization [13,14]. Rural e-commerce emerges as a pivotal component in realizing the seamless integration of digital technologies and the tangible economy. Beyond connecting production with markets, it serves to bridge urban–rural disparities, playing a significant role in stimulating consumption, fostering job opportunities, and driving economic prosperity. Within the context of promoting the rural revitalization strategy, rural e-commerce stands as a primary driving force, markedly enhancing agricultural production efficiency, fostering rural economic development, and elevating farmers’ living standards [15]. However, its development status cannot yet meet the needs of cultivating new quality productivity in the “three rural areas” and the strategy of a strong trading nation. In order to further accelerate the momentum of stable and rapid rural e-commerce development, it is urgent to identify the development capacity and spatial and temporal evolution characteristics of rural e-commerce in China. This is of great practical significance for the formulation of high-quality development policies for rural e-commerce in a targeted manner.

In summary, although current research extensively acknowledges the contribution of rural e-commerce to rural economic development and the augmentation of farmers’ incomes, the majority of studies predominantly concentrate on metrics such as online sales, market scale, and policy support, alongside the mechanisms through which rural e-commerce elevates income and fosters economic growth [16,17,18,19]. Nevertheless, there exists a paucity of comprehensive assessment and evaluation of the developmental ability of rural e-commerce, especially concerning readiness, utilization, and impact, which remain inadequately scrutinized. To address this gap, it is essential to construct a scientific measurement system that facilitates a more nuanced analysis of the developmental capability of rural e-commerce and identifies its fundamental driving variables. Furthermore, the development of rural e-commerce in China’s provinces demonstrates significant imbalance, yet limited research examines how inter-provincial dynamics interact temporally and spatially to affect each other’s developing capacity. The majority of current literature evaluates e-commerce development from a static viewpoint [20]. It fails to completely encapsulate its spatiotemporal dynamics. Consequently, it is essential to examine the developmental trajectories and association patterns of rural e-commerce across areas from a spatiotemporal interaction viewpoint to investigate collaborative or competing regional dynamics. Ultimately, rural e-commerce development policies within provinces are frequently homogenized and devoid of tailored assistance. Considering substantial regional variances in economic development levels, legislative influences, and market demands [21], methods for rural e-commerce development must be more focused and adaptable. This study examines the spatiotemporal variability and geographical dependencies of rural e-commerce development capability, offering empirical evidence and theoretical support for the formulation of localized and differentiated support policies for rural e-commerce.

This study aims to address existing research deficiencies by creating a scientifically rigorous assessment index system to quantitatively measure rural e-commerce development capacity within both single and multidimensional frameworks. Furthermore, it seeks to elucidate the interacting features of rural e-commerce across temporal trajectories, spatiotemporal shifts, and developing spatiotemporal networks, offering extensive theoretical insights and empirical data to guide policy formulation and practical implementation.

2. Literature Review

2.1. Rural Revitalization

Rural revitalization is a critical strategy for China’s recent economic and social development, aiming to achieve comprehensive progress in the rural economy, society, culture, and ecology through multidimensional reforms [22]. This initiative must adhere to the overall objectives of prosperous industries, livable environments, civilized rural customs, effective governance, and improved living standards. The rural revitalization plan tackles the “three rural issues”—agriculture, rural regions, and farmers—by advancing agricultural growth, enhancing rural prosperity, and augmenting farmers’ earnings. This technique is crucial for promoting equitable prosperity across urban and rural regions and guaranteeing the widespread allocation of developmental advantages [23]. Existing research indicates that rural revitalization requires not only policy support but also the active participation of farmers and effective governance by local governments [24]. Concurrently, scholars have diversified their discussions on pathways to rural revitalization, including the development of rural e-commerce, enhancement of agricultural technology, and promotion of farmers’ cooperatives [25,26,27]. However, some literature highlights several challenges facing rural revitalization, such as uneven resource allocation, rural population outflow, and the gradual fading of traditional culture [28,29]. To effectively advance rural revitalization, policy formulation must prioritize local realities and strengthen the synergy between ecological protection and economic development to achieve sustainable outcomes [30,31]. Our research should further investigate the experiences and lessons of rural revitalization across different regions, with an emphasis on the fact that current studies rarely examine the specific role of rural e-commerce in facilitating intercity interactions. Such an exploration would yield more targeted policy recommendations and implementation strategies.

2.2. Development of Rural E-Commerce

Rural e-commerce involves agricultural enterprises engaging in the online exchange of goods and services through modern information technology, playing a crucial role in fostering rural economic development [32]. International research has focused on the developmental stages of rural e-commerce and its influencing components. You and Vavekanand examined both the positive and negative factors in e-commerce operations, addressing issues related to distribution channels [33,34]. Lyu developed an evaluation index system to analyze the impact of ICT on rural e-commerce, highlighting the inefficiencies in information system integration [35]. However, current research reveals several contradictions and limitations, with some studies emphasizing the positive effects of rural e-commerce on poverty alleviation, while others argue that the lack of information technology in rural areas hinders e-commerce development. Additionally, rural e-commerce faces numerous possible obstacles in its operations, such as insufficient infrastructure, the digital divide, and escalating market competitiveness, which may impede its further advancement [36,37]. These characteristics pose dangers and adverse effects as rural e-commerce endeavors to realize its economic growth potential. Since 2014, domestic researchers have increasingly intensified their investigations into rural e-commerce, focusing on its development models, influencing factors, and impacts on poverty alleviation [38,39,40]. Nevertheless, they face challenges, including the need for a cohesive evaluation index system and consideration of regional disparities. “Taobao villages”, as a notable exemplar of rural e-commerce advancement, have garnered significant attention in China. These villages have significantly contributed to regional economic restructuring by utilizing Alibaba’s e-commerce platform, with local entrepreneurs and governments working together to promote industrial innovation and technical progress [41]. Taobao villages, through agglomeration, provide substantial benefits in invigorating local economies and creating employment, acting as a crucial method for numerous farmers to enhance their income [42]. Nonetheless, Taobao villages encounter other obstacles, such as excessive dependence on e-commerce platforms, which can result in industry homogenization, income volatility stemming from alterations in platform policies, and strain on local logistics and infrastructure [41,43,44]. Although current research has primarily concentrated on the impact of rural e-commerce on enhancing farmers’ lives and facilitating economic transformation, there is a necessity for a more systematic examination of its evolution across various regions and a thorough investigation of regional differences.

2.3. Research on Measuring the Development Level of E-Commerce

Research on measuring the level of e-commerce development has yielded extensive findings. Developed Western countries and international organizations have established relatively comprehensive standards for evaluating e-commerce. For instance, in 2000, the European Council initiated the eEurope Action Plan 2002, aiming to expedite online activities within the European Union [45]. Two years later, the eEurope Action Plan 2005 reinforced the measurement and evaluation of Internet effectiveness [46]. The Organisation for Economic Co-operation and Development (OECD) contends that assessing the level of e-commerce development should encompass three dimensions: readiness, usage, and impact, drawing from the S-curve model of innovation diffusion [47]. During the European Commission’s five-year strategy spanning from 2006 to 2010, comprehensive consideration was given to the readiness, usage, and economic impact of e-commerce from the perspectives of enterprises, citizens, and governments [48]. The UK’s National Statistics Office has devised a measurement framework encompassing the readiness, usage, and impact of e-commerce across individuals, enterprises, governments, and markets [49]. Similarly, in China, the National Bureau of Statistics, the National Information Center, and the China Internet Network Information Center (CINIC) gauge the level of e-commerce development utilizing the e-commerce composite index (Tian, 2023) [50]. In assessing the developmental prowess of rural e-commerce, Yu Xinyu (2019) employed the abrupt change series method to gauge its progression [51]. Guo et al. (2022) and Janom et al. (2014) scrutinized the regional rural e-commerce development through the prism of readiness, usage, and impact [52,53]. Meanwhile, Liu et al. (2022) utilized the grey prediction model to assess the coupling coordination status between regional rural e-commerce logistics and agricultural modernization, along with predicting obstacles to coordinated development and future trends [54]. Additionally, Li (2022) devised a comprehensive evaluation index system for rural e-commerce development and conducted a systematic clustering analysis to evaluate each province’s developmental status [55]. The current literature predominantly utilizes the S-shaped curve to develop the indicator system for e-commerce development levels, emphasizing the temporal changes in rural e-commerce development levels. The existing literature offers limited analysis of the spatial patterns and differentiation characteristics of rural e-commerce development capacity, which is crucial for the informed formulation of rural e-commerce policy.

In summary, existing literature predominantly establishes e-commerce development level indicator systems grounded in the S-curve paradigm, emphasizing temporal shifts in rural e-commerce development levels yet offering scant analysis of the spatial patterns and distinctive characteristics of rural e-commerce development capabilities. Consequently, this study endeavors to construct a multidimensional measurement indicator system for rural e-commerce development capabilities, viewed through the lenses of agriculture, rural areas, and farmers, leveraging the OECD’s “S-curve” framework. Utilizing panel vector autoregressive models, the study aims to discern the influential factors impacting the development capabilities of rural e-commerce. Exploratory spatiotemporal data analysis methods are subsequently employed to scrutinize the spatial–temporal differentiation patterns and interactions inherent in rural e-commerce development capabilities. This endeavor not only fosters a deeper comprehension of the intricacies and variances within rural e-commerce development but also unveils the dynamic spatiotemporal characteristics of such development from a novel perspective. It serves to enrich and augment existing research on rural e-commerce development capabilities. The research outcomes furnish a theoretical foundation for regions to devise tailored development trajectories and strategies for rural e-commerce, thereby facilitating the robust and sustainable advancement of rural revitalization efforts in China.

3. Research Methods and Data Sources

3.1. Research Methods

Researchers will face several anticipated challenges when developing a multidimensional measurement index system for enhancing rural e-commerce capabilities. The subjectivity involved in selecting indicators may compromise the objectivity of the outcomes, necessitating stringent oversight of the selection process. Furthermore, the quality and availability of data present significant obstacles; difficulties in acquiring relevant rural e-commerce data may lead to deficiencies or inaccuracies, thereby undermining the reliability of the analytical results. Various factors limit the developmental capacity of rural e-commerce, and these elements dynamically change over time in response to policy adjustments, which can threaten the stability of the indicators. This study will employ multi-source data verification by integrating diverse data sources from government, industry associations, and academic research to ensure accuracy and representativeness, including data from the “China Statistical Yearbook” and the Alibaba Research Institute. The indicator system will be dynamically modified based on literature reviews and expert assessments to accurately reflect rural e-commerce development capacity. In the data processing phase, statistical techniques will be utilized to address missing values (e.g., linear interpolation) and to rigorously assess data distribution, ensuring that data quality meets analytical standards. Through these measures, the study aims to mitigate the impact of potential challenges in identifying and evaluating the developmental capacity of rural e-commerce.

3.1.1. Preliminary Construction and Selection Method of Rural E-Commerce Development Capability Index System

• Construction of characterization and measurement indicators

This study follows the approach of the OECD, integrating the existing indicator system of e-commerce research with research findings and statistical data on rural e-commerce [35,56,57,58,59,60,61]. Adopting a perspective centered on agriculture, rural areas, and farmers, a multidimensional measurement indicator system for rural e-commerce development capabilities is devised. This system encompasses both high-frequency indicators and comprehensive features of rural revitalization, with the objective of pinpointing sensitive indicators that comprehensively depict the characteristics of rural e-commerce development capabilities and facilitate dynamic monitoring. Among them, the characterizing variables are A0, B0, and C0.

The readiness dimension (A) evaluates the infrastructure and initial investments required for rural e-commerce development, comprising five essential indicators. Yu Xin Yi (2019) posits that an advanced logistics system is crucial for the growth of e-commerce, with rural delivery routes acting as a significant measure of the enhancement and expansion of rural logistics [51]. Logistics is essential in rural e-commerce operations, enabling the efficient movement of agricultural products and industrial goods and significantly contributing to the “e-commerce helping farmers” initiative [62]. This research employs rural delivery routes (A0) to assess the robustness of e-commerce infrastructure development. Broadband access constitutes a critical infrastructure for e-commerce, as extensive internet availability facilitates farmers’ participation in e-commerce markets, significantly contributing to the advancement of rural e-commerce [63]. This study utilizes the number of rural broadband users (A1) as an indicator of network penetration in rural areas. Government financial assistance for agriculture and rural regions facilitates essential funding for the expansion of e-commerce [64]. In rural areas, where funding is notably constrained, this support significantly influences the advancement of e-commerce infrastructure and operational settings. This research utilizes government agricultural expenditure (A2) as an indicator of the extent of government investment in rural development. The effectiveness of rural e-commerce is contingent upon a robust agricultural production support system. Agricultural mechanization is essential for improving agricultural production and service systems. Regions with advanced levels of mechanization are more effective in facilitating the upward flow of agricultural products [52]. This research quantifies agricultural mechanization by assessing the total power of agricultural machinery (A3). The availability of electricity significantly influences the operational efficiency of rural e-commerce [52]. Electricity serves as a critical infrastructure that underpins contemporary agricultural production and e-commerce logistics systems. This study employs total annual rural electricity consumption (A4) to evaluate the condition of rural electricity infrastructure [65].

The usage dimension (B) indicates the activity level and market transaction scale in rural e-commerce, consisting of five primary indicators. Sales revenue, an essential indicator of e-commerce progress, has been widely utilized in numerous studies [66]. The levels of e-commerce sales serve as a direct indicator of the market penetration of rural e-commerce. This study employs e-commerce sales revenue (B0) as a metric to assess the development level of rural e-commerce. “Taobao villages” exemplify the growth of rural e-commerce, with their development and proliferation intricately connected to the evolution of the rural e-commerce ecosystem [67]. This study utilizes the number of Taobao villages (B1) as an indicator of farmers’ adoption of e-commerce. The volume of e-commerce purchases (B3) indicates the extent of rural e-commerce transactions and serves as a vital metric for evaluating the development of e-commerce [66]. Express logistics serve as a crucial link between e-commerce enterprises and consumers [68], with logistics capacity being fundamental for the advancement of e-commerce, facilitating the realization of virtual transactions. The volume of express delivery serves as a direct indicator of the logistics and distribution capabilities within rural e-commerce [69]. The proportion of e-commerce businesses with transaction volume (B5) reflects the level of commercial activity in the e-commerce sector. This criterion is fundamental for assessing the commercialization level of rural e-commerce [70].

The impact dimension (C) illustrates the significant effects of rural e-commerce on both the economy and society, comprising five essential indicators. Income level significantly influences e-commerce consumption. Studies indicate that higher income levels can improve farmers’ engagement in e-commerce [71]. This research employs the per capita disposable income of villagers (C0) as an indicator of farmers’ economic and social well-being. Consumption expenditure is significantly associated with e-commerce transactions, with regions exhibiting higher consumption expenditures generally displaying more dynamic e-commerce markets [72]. This study employs villagers’ per capita consumption expenditure (C1) to evaluate farmers’ purchasing power and online shopping behaviors. The growth of rural e-commerce has contributed to the development of local emerging industries, stimulated economic activity, and directly enhanced local tax revenue, serving as a significant indicator of e-commerce’s influence on local economies [52]. This research employs local fiscal tax revenue (C2) to assess the impact of rural e-commerce on local financial conditions. E-commerce removes intermediaries and diminishes the importance of distance, improving consumer experiences and fostering trade expansion, which in turn stimulates significant regional economic development [73]. Research indicates that rural retail sales of consumer goods (C3) may serve as a measure of the impact of rural e-commerce on local economic growth [74]. The expansion of rural e-commerce addresses public demand for enhanced quality of life, progressively diversifying and upgrading the consumption patterns of rural residents. This development contributes to the improvement of rural economic growth and the overall quality of life and production [75]. This research employs Engel’s coefficient of rural residents (C4) to assess the influence of rural e-commerce on consumption patterns.

Most of the economic indicators in this paper have a skewed distribution, and the raw values of the indicators in

Table 1. Indicators of rural e-commerce development capacity.

Dimension	Index	Meaning	Symbol	References
readinessA	A0 Rural delivery routes (km)	Characterize the investment in rural e-commerce infrastructure	lnRDR	Jennex [63]
	A1 Rural broadband access users (10,000 households)	Reflects the availability of Internet infrastructure in rural areas	lnRBAU	Lu, P [64]
	A2 Amount of Fiscal Expenditure on Agriculture (CNY 100 million)	It reflects the degree of government investment in rural development	lnFERA	Guo, N [52]
	A3 Total power of agricultural machinery (10,000 kWh)	Reflects the degree of mechanization of agricultural production	lnPAM	Guo, N [52]
	A4 Annual rural electricity consumption (100 million kWh)	It reflects the state of rural power infrastructure	lnREC	Zhang, D [65]
usageB	B0 E-commerce sales (CNY 100 million)	Characterize the development level of rural e-commerce	lnECS	Fan, P [66]
	B1 Number of Taobao villages	It reflects the acceptance of rural e-commerce by farmers	TBV	Zang, Y [67]
	B2 E-commerce Purchases (CNY 100 million)	Reflect the scale of rural e-commerce transactions	lnECP	Fan, P [66]
	B3 Express business volume (10,000 pieces)	Reflect the logistics and distribution capabilities of rural e-commerce	lnCBV	Zhang, H [69]
	B4 Proportion of enterprises with e-commerce transactions (%)	It reflects the commercial activity of rural e-commerce	lnECTA	Zhang, J [70]
influenceC	C0 Per capita disposable income of villagers (RMB)	Characterize the social living standards of farmers	lnPDI	Tang, H [72]
	C1 Per capita consumption expenditure of villagers (RMB)	Reflect the purchasing power and consumption habits of villagers	lnPCE	Guo, N [52]
	C2 Local Government Tax Revenue (CNY 100 million)	It reflects the impact of rural e-commerce on local fiscal revenue	lnlfr	Guo, M [73]
	C3 Total Retail Sales of Rural Consumer Goods (CNY 100 million)	It reflects the contribution of rural e-commerce to rural economic growth	lnRCGR	Feng, X [74]
	C4 Rural Engel coefficient (%)	It reflects the influence of rural e-commerce on the consumption structure of villagers	lnEC	Zhu, D [75]

have been logarithmized to attenuate heteroskedasticity.

2.

Panel Vector Autoregression

PVAR, a model integrating panel data models and vector autoregressive methods, is utilized to investigate dynamic interactions among multiple variables [76]. This approach amalgamates the benefits of traditional VAR models and panel data analysis, proficiently addressing heterogeneity issues among panel individuals without necessitating adherence to the long-term sequence requirement of VAR models [77]. Moreover, it treats all variables as endogenous variables and incorporates the influence of lagged variables on other variables within the model [78]. The econometric model is outlined as follows:

$${\delta }_{\mathrm{i},\mathrm{t}}={\lambda }_0+{\displaystyle \sum _{\mathrm{j}=1}^k{\lambda }_{\mathrm{j}}}{\delta }_{\mathrm{i},\mathrm{t}-\mathrm{j}}+u_{\mathrm{i}}+o_{\mathrm{i}}+{\epsilon }_{\mathrm{i},\mathrm{t}}$$

(1)

In Equation (1), ${\delta }_{\mathrm{i},\mathrm{t}}$ denotes the endogenous variable matrix, ${\delta }_{\mathrm{i},\mathrm{t}-\mathrm{j}}$ denotes the explanatory variable matrix, ${\lambda }_0,{\lambda }_{\mathrm{i}}$ represents the coefficient matrix of lagged terms, $u_{\mathrm{i}}$ denotes individual effects, accommodating individual differences, $o_{\mathrm{i}}$ represents time-fixed effects, and ${\epsilon }_{\mathrm{i},\mathrm{t}}$ signifies the random disturbance term.

3.1.2. Methods for Measuring the Development Capacity of Rural E-Commerce

To compare, analyze, and evaluate the rural e-commerce development capabilities of different provinces, this study utilizes both the vector summation method and the polyhedral method to assess both single-dimensional and multidimensional aspects of rural e-commerce development capabilities.

(1) To compare and analyze the level and differential attributes of rural e-commerce development capabilities within a single dimension, the vector summation method is utilized. This method is relatively straightforward, treating each indicator as a vector, with the vector norm representing the value of the respective indicator. The sum of these vectors for each indicator corresponds to the Rural E-commerce Development Capability Index (REDCI-A) of that dimension [79]. When computing the Rural E-commerce Development Capability Index, indicator weights are considered, and the calculation formula is as follows:

$$REDCI{\_}_A=\sqrt{{(a_A\times w_A)}^2+{(a_B\times w_B)}^2+{(a_C\times w_C)}^2}$$

(2)

$a_A-a_C$ represents the standardized indicator value, and $w_A-w_C$ represents the indicator weight.

(2) Methods for measuring multidimensional rural e-commerce development capabilities.

To comprehensively comprehend the multidimensional levels and differential characteristics of rural e-commerce development capabilities, the polyhedron method is employed for measurement. This method represents each indicator with a line segment connecting a fixed point (also referred to as the origin) inside the polyhedron to each vertex, with the length of the line segment representing the magnitude of the corresponding indicator value. The volume of the polyhedron serves as a comprehensive measure of the Rural E-commerce Development Capability Index (REDCI-V) for multidimensional development of rural e-commerce [80].

$$REDC{\_}_V={\displaystyle \frac{1}{6}}\sin \alpha (a_A\times w_A\times a_B\times w_B+\cdots +a_G\times w_G\times a_A\times w_A)\times (a_H\times w_H+a_I\times w_I)$$

(3)

$a_A-a_I$ represents the standardized indicator value; $w_A-w_I$ represents the indicator weight; $\alpha$ represents the angle between indicators.

3.1.3. Exploratory Spatiotemporal Data Analysis

Temporal and spatial effect analysis can unveil regional development imbalances and convergence. This study employs the ESTDA method to unveil the spatiotemporal relationships, patterns, and changes in the multidimensional capabilities of rural e-commerce development across Chinese provinces, addressing the limitations of traditional spatial analysis methods in the temporal dimension and accomplishing spatiotemporal interaction analysis [81]. Numerous challenges may arise during the implementation process. The limitations of LISA time path analysis stem from its reliance on data quality and the selection of spatial units, which can potentially lead to biased outcomes. The matrix classification of LISA spatiotemporal transitions may also be influenced by subjective assessments, compromising the objectivity of the classifications. To mitigate the impact of these potential issues, this study employs the following procedures: First, the accuracy and reliability of the data utilized are verified through multi-source data validation and quality control, thereby enhancing the trustworthiness of the analytical results. Second, the LISA time path analysis incorporates standardized processing and rigorous statistical testing methods to minimize data selection subjectivity and ensure the findings’ reliability. Additionally, spatiotemporal transition analysis will integrate literature reviews and expert insights to ensure the rationality and scientific validity of the classification matrix. The study aims to maintain the objectivity and consistency of results when assessing the spatiotemporal patterns of rural e-commerce development capacity, ensuring that these factors do not significantly influence the interpretation of future research findings. In the subsequent subregional analysis, China is categorized into East, Central, West, and Northeast regions according to geographic distinctions. The eastern region comprises 11 provinces: Beijing, Tianjin, Hebei, Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong, and Hainan. The central region includes six provinces: Shanxi, Anhui, Jiangxi, Henan, Hubei, and Hunan. The western region consists of 12 provinces: Inner Mongolia, Guangxi, Chongqing, Sichuan, Guizhou, Yunnan, Tibet, Shaanxi, Gansu, Qinghai, Ningxia, and Xinjiang. The northeast region comprises three provinces: Liaoning, Jilin, and Heilongjiang.

(1)

LISA Time Path

The LISA time path is continuously unveiled using the LISA Markov transition matrix, which can unveil the coordinate transitions of different spatial units in the Moran’s I scatter plot [82]. Analyzing the migration of standardized values and spatial lag values of rural e-commerce development across provinces reveals the spatiotemporal synchronous changes and dynamic characteristics of multidimensional development intensity of regional e-commerce capabilities [83]. This involves two indicators: relative length and curvature [84]. The calculation formula is as follows:

$${\Gamma }_{\mathrm{i}}={\displaystyle \frac{\mathrm{n}{\displaystyle \sum _{\mathrm{t}=1}^{T-1}\mathrm{d}({\mathrm{L}}_{\mathrm{i},\mathrm{t}},{\mathrm{L}}_{\mathrm{i},\mathrm{t}+1})}}{{\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{n}}{\displaystyle \sum _{\mathrm{t}=1}^{T-1}\mathrm{d}({\mathrm{L}}_{\mathrm{i},\mathrm{t}},{\mathrm{L}}_{\mathrm{i},\mathrm{t}+1})}}}},\mathrm{n}=31$$

(4)

$$D_{\mathrm{i}}={\displaystyle \frac{{\displaystyle \sum _{\mathrm{t}=1}^{T-1}\mathrm{d}({\mathrm{L}}_{\mathrm{i},\mathrm{t}},{\mathrm{L}}_{\mathrm{i},\mathrm{t}+1})}}{\mathrm{d}({\mathrm{L}}_{\mathrm{i},1},{\mathrm{L}}_{\mathrm{i},T})}}$$

(5)

In Equations (4) and (5), ${\Gamma }_{\mathrm{i}}$ represents the relative length of the time path, $D_{\mathrm{i}}$ represents the curvature degree, n denotes the number of provinces, T represents the time interval, $\mathrm{d}({\mathrm{L}}_{\mathrm{i},\mathrm{t}},{\mathrm{L}}_{\mathrm{i},\mathrm{t}+1})$ represents the movement distance of province i between years t and t + 1, ${\mathrm{L}}_{\mathrm{i},\mathrm{t}}$ is the position of province i in the Moran scatter plot in year t. A larger ${\mathrm{L}}_{\mathrm{i},\mathrm{t}}$ indicates a stronger dynamicity in the local spatial structure, whereas a smaller value suggests relative stability. A larger $D_{\mathrm{i}}$ indicates a more curved time path, implying greater volatility in the local spatial structure.

(2)

LISA Temporal Transition

To further explore the transition states and dynamic paths of local spatial correlation types in measuring rural e-commerce development capabilities, the spatial–temporal leap theory is referenced to construct four types of matrices (

Table 2. Temporal transition type.

Type	Meaning	Expression Symbols
Type I	The self remains the same, the field does not change	HHt→HHt + 1, HLt→HLt + 1, LLt→LLt + 1, LHt→LHt + 1
Type II	Self-transition, the neighborhood remains the same	HHt→LHt + 1, HLt→LLt + 1, LLt→HLt + 1, LHt→HHt + 1
Type III	It does not change itself, and the neighborhood jumps	HHt→HLt + 1, HLt→HHt + 1, LLt→LHt + 1, LHt→LLt + 1
Type IV (1)	Self Transition, Neighborhood Transition (Unanimous)	HHt→LLt + 1, LLt→HHt + 1
Type IV (2)	Self Jump, Neighborhood Jump (Opposite)	HLt→LHt + 1, LHt→HLt + 1

) [85].

(3)

Visualization of spatiotemporal interactions

To address the shortcomings of traditional spatial analysis in losing correlation information, graph theory is employed to depict the spatiotemporal interaction characteristics of rural e-commerce development transitions [86]. This study computes the covariance coefficients of the distances between neighboring provinces and their spatiotemporal movement trajectories. It utilizes network graphs and chord diagrams for visualization to unveil the dynamic evolution process of the competitive and cooperative state of rural e-commerce development among provinces and the similarity of development mechanisms [87].

3.2. Data Source

Micro-scale research on rural e-commerce is vital for uncovering its development trends and spatiotemporal interaction characteristics, but smaller scales also entail greater challenges in data acquisition [88]. By comprehensively considering the degree of indicator reflection and data acquisition difficulty, this study selects the 31 provincial-level administrative regions of China as the research units and the period from 2011 to 2022 as the research timeframe. The main data sources comprise the “China Statistical Yearbook” spanning from 2011 to 2022, rural broadband access users data from the China Internet Network Information Center (CNNIC), fiscal expenditure on agriculture retrieved from the “China Financial Yearbook”, the number of Taobao villages provided by the Alibaba Research Institute, rural retail sales of consumer goods obtained from the statistical bulletins of various provinces’ national economic and social development, and rural Engel coefficients extracted from the “China Urban Statistical Yearbook”, “China Rural Statistical Yearbook”, “China Social Statistical Yearbook”, among others. Due to the lack of data on e-commerce procurement and sales for 2011 and 2012, the Python 3.12.4 libraries pandas, NumPy, and scikit-learn were used to address the missing values. Linear interpolation was employed to ensure data continuity and integrity for a small amount of missing data.

4. Analysis of Results

4.1. Identification of Sensitive Factors and Construction of Measurement Index System

4.1.1. Identification of Sensitive Factors in the Development Capacity of Rural E-Commerce

The Hausman test results indicate that the usage level, along with the first-order differences of the preparedness and influence levels, all pass the test. This suggests that the three-dimensional indicators adhere to the characteristics of the fixed effects model, thus warranting the construction of separate PVAR models. To ensure the accuracy of model estimation and mitigate the risk of spurious regression, three methods—LLC, IPS, and ADF-Fisher—are employed to test the stability of variables individually. Remarkably, all indicators meet the stability requirements. When conducting PVAR model analysis, it is imperative to determine the optimal lag order of each dimensional influencing factor based on the AIC, BIC, and HQIC criteria. The test results reveal that the optimal lag order of each dimensional influencing factor is 1. To eliminate time effects, the within-group mean difference method is employed, while the forward mean difference (Helmert transformation method) is utilized to eliminate individual effects. This approach aims to prevent biased results arising from both effects. Subsequently, the GMM estimation method is employed to identify the static interactive influence relationship between variables, as demonstrated in

Table 3. Estimate the relationship between the characterization metric and the dimension metric.

Variable	h_dlnrdr	Variable	h_lnecs	Variable	h_dlnpdi
L·h_drbau	0.001 ***	L·h_tbv	0.000	L·h_dlnpdi	0.791 ***
	[0.00]		[0.00]		[0.20]
L·h_dlnrdr	0.062	L·h_lnecp	−0.000	L·h_dlnpce	−0.092 **
	[0.11]		[0.04]		[0.04]
L·h_dlnfera	0.145 **	L·h_lnecs	0.634 ***	L·h_dlnlfr	−0.031
	[0.07]		[0.08]		[0.03]
L·h_dlnpam	0.176 *	L·h_lncbv	0.394 **	L·h_dlnrcgr	−0.020 **
	[0.09]		[0.20]		[0.01]
L·h_dlnrec	−0.003	L·h_lnecta	−0.261 *	L·h_dlnec	0.0923 ***
	[0.04]		[0.15]		[0.03]

Note: “h_” indicates that the variable has been Hermite transformed; “L” represents the first-order lag of the variable; the numbers in the table denote the correlation coefficients estimated by GMM; the numbers in parentheses represent t-test values; ***, **, * denote significance at the 1%, 5%, and 10% levels, respectively.

. Variables passing the Granger causality test undergo impulse response analysis to elucidate the dynamic response path of representing indicators to various dimensional indicators, as illustrated in Figure 1.

From the perspective of rural e-commerce development readiness, at a 1% significance level, the lagged one-period rural broadband access users (RBAUs) and fiscal expenditure related to agriculture (FERA) have a significant positive impact on current rural delivery routes. The increased investment by the government in rural e-commerce infrastructure and financial support amid rural revitalization efforts has facilitated the proliferation of rural broadband networks and amplified fiscal investment, thereby bolstering the readiness of rural e-commerce. Annual rural electricity consumption (REC) also positively influences the current rural delivery routes, albeit to a lesser extent. The increase in rural electricity consumption has contributed to the improvement and development of rural infrastructure. However, its role in the overall readiness for rural e-commerce is relatively minor. In comparison to the impact of widespread broadband network coverage and financial support, its effect is rather limited.

From the perspective of rural e-commerce development usage, the lagged one-period express delivery volume (CBV) exhibits a significant positive impact on e-commerce sales, albeit gradually diminishing over time. This trend could be attributed to the stabilization of the growth rate of e-commerce sales as rural e-commerce matures, leading to a gradual weakening of the influence of express delivery volume on sales. Conversely, the lagged one-period number of Taobao villages (TBVs) and e-commerce purchase volume (ECP) demonstrate a negative impact on e-commerce sales. The escalation in the number of Taobao villages and e-commerce purchase volume may be associated with factors such as heightened competition among e-commerce platforms, product quality variations, and after-sales service disparities. Failure to effectively manage and control these factors could undermine consumer trust and satisfaction with e-commerce, consequently dampening sales.

From the perspective of rural e-commerce development impact, rural residents’ per capita consumption (PCE), local fiscal tax revenue (LFR), and rural consumer goods retail sales (RCGR) exhibit a significant negative impact on rural residents’ per capita disposable income. This phenomenon may stem from governmental initiatives aimed at enhancing the economic development level of rural areas and tackling structural challenges, such as income inequality and the singular structure of rural industries. These issues could potentially impede the income growth rate and consumption capacity of rural residents.

4.1.2. Determination of Indicators and Weights of Rural E-Commerce Development Capacity

To ascertain the definitive measurement indicators of rural e-commerce development capability, it is crucial to elucidate the varying degrees of influence exerted by different dimensional indicators on rural e-commerce development. This study conducted an equal-weight variance analysis to assess the contribution of each indicator, as shown in Figure 2.

Figure 2 illustrates that the influences on the three dimensions of readiness, usage, and impact primarily originate from within themselves. However, the contribution levels of other indicators depict a pattern of initial increase, subsequent decline, and eventual stabilization within a certain range. From a readiness perspective, the contributions of rural broadband access users, fiscal expenditure related to agriculture, and the total power index of agricultural machinery to the characterization indicators stabilize after increasing from the second to the third period. In terms of usage, the impacts of express delivery volume, e-commerce procurement amount, and the number of Taobao villages gradually increase from the second to the seventh period before stabilizing. Similarly, the contribution levels of rural per capita consumption expenditure, local fiscal tax revenue, rural Engel’s coefficient, and rural consumer goods retail sales indicators also increase from the second to the third period and gradually approach stability.

The contribution levels of the three-dimensional indicators are sorted from highest to lowest. Subsequently, the last indicator affected by the variance of each dimension is removed to determine the measurement indicators of rural e-commerce development capability. These indicators include rural broadband access users, fiscal expenditures on agriculture, rural electricity consumption, the number of Taobao villages, e-commerce procurement amount, express delivery volume, rural per capita consumption expenditure, local fiscal tax revenue, and rural consumer goods retail sales, totaling nine indicators.

The CRITIC weighting method is a relatively effective objective evaluation method, but it does not consider the discreteness of data. Conversely, the entropy method can effectively reflect the degree of data discreteness. By combining these two weighting methods, it is possible to comprehensively reflect the data’s discreteness, conflict level, and comparative strength, thereby compensating for each other’s shortcomings [89]. Therefore, this study utilizes a combination of the entropy method and CRITIC method to obtain the composite weights of each indicator, as shown in

Table 4. Metric portfolio weights.

Target Layer	Dimension	Weight (%)	index	CRITIC Method	Entropy Method	Combination Method
Rural e-commerce development capacity index	Readiness	23.75	Rural broadband access users (in 10,000 households)	7.75	8.24	7.99
			Agricultural-related fiscal expenditure (in RMB 100 million)	4.74	2.93	3.83
			Rural electricity consumption (in 100 million kilowatt-hours)	13.38	10.46	11.92
	Usage	60.38	Number of Taobao Villages (in units)	26.44	31.15	28.80
			E-commerce procurement volume (in RMB 100 million)	13.81	12.29	13.05
			Express delivery volume (in 10,000 units)	17.49	19.59	18.54
	Influence	15.87	Per capita rural resident consumption expenditure (in RMB)	3.72	2.35	3.04
			Local fiscal tax revenue (in RMB 100 million)	5.53	5.72	5.62
			Rural consumer goods retail sales (in RMB 100 million)	7.15	7.27	7.21

.

4.2. Analysis of the Results of Rural E-Commerce Development Capacity

4.2.1. Measurement Results of One-Dimensional Rural E-Commerce Development Capacity

The vector sum method was employed to evaluate the development capacity index of rural e-commerce villages in 31 provincial-level administrative regions in China from 2011 to 2022. Only the measurement results of 2011, 2016, and 2022 were depicted, as shown in Figure 3. Additionally, the spatiotemporal characteristics of rural e-commerce development capacity in different dimensions were analyzed.

Over the period from 2011 to 2022, the average indices of rural e-commerce readiness, usage, and impact in China exhibited a general upward trend. Factors such as the rural revitalization strategy, policy support, technological innovation, and market demand collectively propelled the development of rural e-commerce in China. In terms of readiness, although showing an upward trend, it was not particularly significant. The provinces of Henan, Hebei, and Sichuan experienced relatively rapid increases. Analysis of average readiness values from the graph revealed a gradual reduction in regional disparities, possibly attributed to proactive policy measures implemented by provinces, such as subsidies for the construction of e-commerce industrial parks and support for the innovative development of e-commerce enterprises, thus enhancing the readiness for rural e-commerce development.

The usage of rural e-commerce has exhibited notable patterns over the years. Between 2011 and 2016, approximately one-third of the provinces and cities demonstrated relatively high usage levels, while others lagged behind, indicating an early developmental stage. This disparity primarily stemmed from unequal resource distribution and inadequate infrastructure. However, from 2016 to 2022, with exceptions such as Zhejiang, Jiangsu, and Guangdong leading the usage trend, most provinces and cities approached the average usage level. During this period, intensified policy advocacy, technological dissemination, and infrastructure development contributed to the enhancement of rural e-commerce usage, fostering a more balanced landscape across regions.

Regarding impact, Sichuan and Anhui exhibited relatively high levels from 2011 to 2022, while other provinces and cities lagged behind. In summary, during the period from 2011 to 2022, China’s rural e-commerce witnessed varying degrees of improvement in infrastructure investment and development levels. However, there is still ample room to enhance the social living standards of rural residents.

From a spatial distribution perspective, high levels of rural e-commerce development readiness are predominantly concentrated in provinces such as Jiangsu, Zhejiang, Shandong, Guangdong, and Henan. These regions have implemented a series of policies to bolster rural e-commerce development and have made substantial investments in building infrastructure, including broadband network coverage and logistics distribution systems, thus augmenting the readiness for rural e-commerce. Conversely, low-value areas are primarily clustered in provinces such as Qinghai, Tibet, and Tianjin, which suffer from inadequate infrastructure and lack policy support, thereby constraining the readiness of rural e-commerce.

Regions with higher usage levels, such as Guangdong, Zhejiang, Jiangsu, Shandong, Fujian, Jiangxi, and Hebei in the eastern regions, exhibit a greater adoption of e-commerce among residents. This heightened usage is propelled by robust market demand, fostering the thriving development of rural e-commerce. Conversely, areas with lower usage levels, including Tibet, Qinghai, and Ningxia, face inadequate market demand, impeding the advancement of rural e-commerce.

High-impact values are concentrated in regions such as Jiangsu, Zhejiang, Guangdong, and Shandong, which boast larger market shares and wield significant influence in the e-commerce industry. Conversely, provinces like Qinghai, Tibet, Ningxia, and Jilin exhibit smaller market shares and limited influence, categorizing them as low-value areas.

4.2.2. Measurement Results of Multidimensional Rural E-Commerce Development Capacity

Using the Formula (3) of the polyhedron method, the rural e-commerce development index of the 31 provincial units in China was calculated for the years 2011 to 2022. Representative samples of the rural e-commerce development index for the years 2011, 2016, and 2022 were selected. Based on the development index and its degree of dispersion calculated for these representative years, the Jenks natural break classification method was employed to categorize rural e-commerce development capability into five dimensions: low development capability (0, 0.005865), relatively low development capability (0.005865, 0.022661), moderate development capability (0.022661, 1.259414), relatively high development capability (1.259414, 8.215478), and high development capability (8.215478, 36.80). This classification aimed to analyze the spatiotemporal evolution characteristics of rural e-commerce nationwide. See Figure 4 for reference.

From 2011 to 2022, China’s rural e-commerce development capability has notably advanced, yet regional disparities are evident, with the highest development index observed in the eastern region, followed by the central region, and the lowest in the western region. The eastern region, boasting a developed economy, robust policy backing, and vibrant market demand, demonstrates the highest rural e-commerce development capability. The central region benefits from supportive policies and infrastructure enhancement, resulting in comparatively robust rural e-commerce development capability. Conversely, the western region, hampered by factors like slower economic growth, geographical constraints, and inadequate policy support, exhibits relatively lower rural e-commerce development capability. These regions promote the development of adjacent areas by disseminating e-commerce operational frameworks, technologies, and skills. Zhejiang’s “Village Taobao” model exemplifies the effective utilization of technology and community-oriented strategies to enhance agricultural goods sales. Conversely, Shandong’s “e-commerce + logistics” model demonstrates that the integration of courier services enhances the accessibility of rural e-commerce. Regions, including Hebei, Shaanxi, and Sichuan, have progressively enhanced their e-commerce capabilities by adopting these effective approaches. Nevertheless, certain western provinces, including Qinghai and Gansu, persist in their underdevelopment of e-commerce owing to obstacles, including insufficient transportation and internet infrastructure, along with a deficiency of skilled e-commerce professionals. This indicates that forthcoming policies ought to emphasize enhanced infrastructure assistance for these areas and contemplate the creation of regional e-commerce training centers to cultivate local expertise, thus successfully narrowing the development disparity.

4.3. Analysis of Spatiotemporal Interaction Characteristics of Rural E-Commerce Development Capacity

4.3.1. LISA Timepath Analysis

(1)

Analysis of the geometric features of the time path

From Figure 5a, it is evident that the relative length of the LISA time path for rural e-commerce development is less than the mean (both are 1) in 25 provinces, indicating relatively stable local spatial patterns. The average relative lengths are as follows: eastern region (2.849) > central region (0.1979) > western region (0.088) > northeast region (0.0872). The highest and lowest values are in Zhejiang (7.582) and Hubei (0.0426), respectively, indicating more frequent local spatial association changes in the eastern and central regions, while the local spatial structure in the northeast and western regions is relatively stable. Over time, the trend of the time path changes from the central and western regions to the eastern region gradually increases. Zhejiang, Shanghai, and Jiangsu have relative lengths exceeding 5.5, indicating strong dynamic characteristics of local space. The time paths in Hubei, Chongqing, and Sichuan are relatively short, with relative lengths all less than 0.06. The eastern region, due to its higher level of economic development, strong policy support, and technological innovation, exhibits strong dynamic characteristics and changes in the time path of rural e-commerce development. Zhejiang’s “Digital Countryside” initiative and Jiangsu’s “Rural E-commerce Ecosystem” have successfully attracted resources and stimulated consumer demand, thereby enhancing local rural e-commerce development capabilities and offering a replicable model for other regions in the country. This not only bolsters local capabilities for rural e-commerce advancement but also serves as a paradigm for other places nationwide. On the other hand, the central and western regions, due to relatively weak economic foundations and remote geographical locations, experience relatively stable rural e-commerce development, but efforts are still being made to catch up. Chongqing’s rural e-commerce development, although nascent, is progressively enhancing its market competitiveness by upgrading the logistics network and promoting specialized agricultural products. This regional disparity underscores the influence of economic foundations and policy environments on rural e-commerce development while simultaneously revealing the spatial interdependence among regions, with e-commerce advancement in the eastern regions acting as a catalyst for the central and western areas of the country.

From Figure 5b, it can be observed that the curvature of the LISA time path for rural e-commerce development capacity is lower than 1 in four provinces, accounting for 12.9% of the total. This indicates a strong spatial dependence among provinces, with close economic ties and industrial layouts between different regions, especially between the eastern and western regions. This connection leads to a strong spatial dependence on rural e-commerce development capacity, where the development level in certain areas directly affects surrounding regions. During the study period, the average curvature values for the four major regions are as follows: eastern region (10.251) > western region (9.640) > central region (6.4076) > northeastern region (3.124). The highest and lowest curvature values are in Chongqing (55.723) and Fujian (0.450), respectively. Overall, the spatial pattern shows a decreasing trend from the eastern, northwestern, and central regions towards surrounding areas. The advanced economic development and resource agglomeration in the eastern area have profoundly impacted rural e-commerce growth in the western region. The successful experiences of rural e-commerce in the East have fostered the development of rural e-commerce in the West through collaboration in industrial chains, logistics and distribution networks, and the transfer of technical expertise. This is especially apparent in Shaanxi and Xinjiang, where the eastern region’s allocation of e-commerce resources has propelled the swift advancement of regional e-commerce. This phenomenon indicates that the advancement of rural e-commerce would necessitate intra-regional initiatives and inter-regional collaboration to improve the rural e-commerce environment via resource sharing and logistical interoperability.

4.3.2. LISA Spatiotemporal Transition Analysis

The length, curvature, and direction of the LISA time path can track the evolution trajectory of spatial association types but cannot deeply reveal the characteristics of mutual transfer between spatial association types. Therefore, temporal–spatial transition analysis is adopted to further reveal the transfer characteristics of local spatial association types of rural e-commerce development capacity among Chinese provinces. Please refer to

Table 5. Spatiotemporal transition matrix of multidimensional rural e-commerce development capacity.

Time	Type	HH	LH	LL	HL
2011–2016	HH	Type I (SH, JS, ZJ)	Type II	Type IV (1)	Type II (SD)
	LH	Type II	Type I (AH, FJ, JX, GX, HI)	Type III	Type IV (2)
	LL	Type IV (1)	Type III (BJ)	Type I (TJ, HE, SX, NM, LN, JL, HL, HA, HB, CQ, SC, GZ, YN, XZ, SN, GS, QH, NX, XJ)	Type II
	HL	Type III	Type IV (2)	Type II	Type I (GD)
2016–2022	HH	Type I (SH, JS, ZJ)	Type II	Type IV (1)	Type III
	LH	Type II	Type I (AH, FJ, JX, HN, GX, HI)	Type III (BJ)	Type IV (2)
	LL	Type IV (1)	Type III	Type I (TJ, HE, SX, NM, LN, JL, HL, HA, HB, CQ, SC, GZ, YN, XZ, SN, GS, QH, NX, XJ)	Type II
	HL	Type III (SD)	Type IV (2)	Type II	Type I (GD)
2011–2022	HH	Type I (SH, JS, ZJ, SD)	Type II	Type IV (1)	Type III
	LH	Type II	Type I (AH, FJ, JX, HN, GX, HI)	Type III	Type IV (2)
	LL	Type IV (1)	Type III	Type I (BJ, TJ, HE, NM, LN, JL, HL, HA, HB, CQ, SC, GZ, YN, XZ, SN, GS, QH, NX, XJ)	Type II
	HL	Type III	Type IV (2)	Type II	Type I (GD)

.

According to Table 5, the transition matrices of Type I from 2011 to 2016, from 2016 to 2022, and from 2011 to 2022 account for 90.3%, 90.3%, and 96.7%, respectively. This indicates the relatively stable spatial distribution structure of multidimensional rural e-commerce development capabilities, with a clear spatial aggregation phenomenon. In this structure, the speed of spatial correlation type transition is slow, and the development level of some regions remains fixed, potentially limiting the implementation of the rural revitalization strategy, particularly in areas with lower rural e-commerce development capabilities.

In the periods 2011–2016 and 2016–2022, only Shandong and Beijing belonged to Type III, indicating their strong overall strength in rural e-commerce development and limited influence from other provinces or involvement in spatial correlation transfer processes. No provinces belonged to Type II and Type IV during these periods, suggesting the absence of provinces in relatively low spatial correlation types (low–high or high–low). Shandong, situated in the economically developed eastern coastline region, possesses a solid economic basis and substantial policy backing, which greatly impacts its e-commerce advancement. Beijing, as the political and economic hub of the nation, propels e-commerce expansion through its distinctive resource distribution and market demand, resulting in a more autonomous development trajectory. This may be attributed to the increasing importance attached to rural e-commerce development and the general increase in investment and support for rural e-commerce under the backdrop of rural revitalization. Additionally, the local Moran’s index transfer of the four types mentioned above is not very active, and the local spatial correlation structure is relatively stable, indicating challenges in changing the relative position of each provincial unit. Future policies should facilitate the transfer of technology and market integration from the eastern area to the center and western regions, promoting resource sharing and industrial synergy. An e-commerce incubation center could be established in provinces with significant rural e-commerce development capacity to offer resource coordination and market analysis services to adjacent provinces, thereby enhancing e-commerce development in underdeveloped regions.

4.3.3. Spatiotemporal Network Analysis

Utilizing the covariance coefficient of the mobile trajectory of rural e-commerce development multidimensional capabilities, the geographic network and topological network results of provincial rural e-commerce development capabilities were obtained using ArcMap 10.8.1 and Origin 2022 software, as shown in Figure 6.

From Figure 6, it is evident that while competition exists among provinces, the majority of them maintain a cooperative relationship. Inter-provincial relationships are generally positive, although the degree of association varies. Specifically, there are weak positive associations between six pairs of provinces, constituting 8.6% of the total, primarily concentrated at the intersection of North China, Central China, and East China. For instance, tenuous connections exist among Henan, Anhui, and Shanxi, which are potentially influenced by geographic positioning and economic development inequalities. These reasons result in a comparatively low level of synergy in rural e-commerce development throughout these provinces. This diminished connection suggests that, although these areas demonstrate beneficial contacts in rural e-commerce, they have not yet developed substantial synergies.

Additionally, there are weak positive associations between five pairs of provinces, accounting for 7.1% of the total, mainly distributed between Hubei, Hunan, and Jiangxi, as well as between Hebei, Beijing, Tianjin, and Inner Mongolia. The potential for rural e-commerce advancement in these areas suggests preliminary collaboration; nevertheless, the connections are tenuous owing to disparities in industry composition and economic circumstances. The minimal correlation may arise from inadequate regulatory support and a fragile market basis, underscoring the necessity for enhanced inter-regional collaboration, especially regarding synergy and interoperability within the e-commerce supply chain.

There are relatively stronger positive associations between eight pairs of provinces, accounting for 11.4% of the total, primarily among Anhui, Guangdong, Zhejiang, and Jiangsu, as well as between Liaoning, Hebei, Henan, and Inner Mongolia. The potential for rural e-commerce advancement in these areas suggests preliminary collaboration; nevertheless, the connections are tenuous owing to disparities in industry composition and economic circumstances. The minimal correlation may arise from inadequate regulatory support and a fragile market basis, underscoring the necessity for enhanced inter-regional collaboration, especially regarding synergy and interoperability within the e-commerce supply chain.

Strong positive associations are observed among 51 pairs of provinces, accounting for 72.9% of the total, with the majority distributed in western and northeastern provinces. Provinces like Xinjiang, Gansu, Ningxia, and Qinghai, although relatively autonomous from the more economically advanced eastern regions, have established enhanced internal connections via rural e-commerce development initiatives and sustained inter-regional collaboration. The economic relationships among these provinces are predominantly shaped by internal collaboration and mutual assistance, with minimal effect from external forces. For instance, by regulatory backing and infrastructural investment, these regions have effectively grouped in e-commerce, creating more cohesive networks.

Over time, through long-term interaction, these regions have established close cooperative relationships and economic ties, promoting strong positive associations between their rural e-commerce development capabilities. In terms of spatial changes, regions with strong positive associations may exhibit relative independence from developed regions, with comparatively less communication and cooperation among them, thus fostering stronger positive associations. Relatively stronger positive associations are primarily observed in the eastern regions, where inter-regional cooperation and communication are more frequent, facilitating robust positive associations between rural e-commerce development capabilities. These variations in the degree of positive associations are closely linked to the historical development trajectories, policy initiatives, geographic locations, and economic development levels of provinces.

5. Conclusions and Suggestions

5.1. Conclusions

This study establishes a multidimensional measuring index method for evaluating rural e-commerce development capacity, focusing on the “Three Rural Issues” (Agriculture, Rural Areas, and Farmers). The system comprises three dimensions: readiness, utilization, and impact. This study quantitatively assesses the rural e-commerce development capacity at the province level in China using both multidimensional and unidimensional methodologies. Furthermore, exploratory spatial–temporal analysis is utilized to uncover the spatial–temporal inequalities and evolutionary traits of rural e-commerce development capacity. The results are as follows:

(1) The execution of the Rural Revitalization Strategy has rendered rural e-commerce an essential instrument for advancing rural economic development and augmenting farmers’ incomes. Enhancing the capacity for rural e-commerce growth can promote the advancement of rural industrial frameworks, accommodate excess labor, rejuvenate “Three Rural” resources, and support the objectives of enriching farmers, bolstering agriculture, and developing rural regions;

(2) From 2011 to 2022, the capability for rural e-commerce development in China has significantly enhanced, demonstrating distinct spatial–temporal dynamics. The aspects of readiness, utilization, and impact have all seen increases over time. Nonetheless, substantial regional disparities persist, characterized by a marked two-tier distinction. The capacity for rural e-commerce development exhibits significant agglomeration and spatial differentiation. The eastern region exhibits the greatest development potential, succeeded by the center region, whilst the western and northeastern regions are significantly disadvantaged, reflecting a pronounced spatial disparity;

(3) The eastern region has the most vigorous rural e-commerce advancement, characterized by frequent local spatial correlations. Conversely, rural e-commerce advancement in the central and western regions is comparatively constant, whereas the northeastern region exhibits the least developmental variation. The economic connections between the eastern and western regions are robust, with considerable spatial dependence. The middle and northeastern regions demonstrate a somewhat stable spatial distribution structure, characterized by slower transitions in spatial association types, hence complicating changes in the relative positions among provinces;

(4) Despite the competitive dynamics in rural e-commerce development among provinces, the majority exhibit positive synergistic interactions, with certain regions demonstrating significant and stable spatial interdependence. Collaboration is more prevalent in the eastern region, although inter-provincial relations are more intimate in the western and northeastern regions. The geographical linkages and regional collaborations amplify the interaction impacts of rural e-commerce development among provinces.

5.2. Suggestions

Initially, advanced provinces in rural e-commerce development should prioritize the enhancement of the rural e-commerce industrial chain, the cultivation of new sub-sectors and growth sources, and the promotion of creative models within rural e-commerce. These regions must broaden their geographical scope, harness developmental potential, and facilitate the advancement of rural e-commerce in all dimensions, transitioning from conventional models to innovative formats such as digital agricultural product e-commerce and novel rural e-commerce. Provinces exhibiting moderate rural e-commerce development should persist in optimizing logistics systems to diminish operating expenses. By providing legislative incentives and financial assistance, they can entice prominent firms and superior e-commerce platforms to set up branches, thereby improving the sophistication of the rural e-commerce industry chain. Advocating for the development of specialized industrial and e-commerce supply chains is essential to enhance the overall supply chain.

In provinces with underdeveloped rural e-commerce, initiatives should prioritize the enhancement of logistics systems in remote regions, primarily by upgrading road infrastructure and guaranteeing effective “last-mile” delivery of agricultural goods. Policy and financial assistance should be allocated for the construction of agricultural product warehouses at production locations, supporting nascent e-commerce ventures, and promoting the integration of fresh agricultural products into urban markets. Rural e-commerce firms ought to be assisted in leveraging new media platforms, such as Douyin (TikTok) and Kuaishou, to enhance consumer experience and broaden market share.

Secondly, inter-provincial collaboration and exchanges should be utilized to maximize the beneficial effects of regional connections on rural e-commerce advancement. Encouraging collaboration between Zhejiang and Jiangsu with Gansu and Qinghai can enhance e-commerce support and establish regular “E-commerce + Poverty Alleviation” projects to promote the integration of information technology and e-commerce in the central and western regions. This will propel the intelligent, premium, and sustainable advancement of rural e-commerce. Major e-commerce enterprises ought to be incentivized to disseminate their experience to central and western regions via compensated technical knowledge exchange, collaborative training initiatives, and resource allocation, thereby facilitating development in these areas and enhancing cooperative rural e-commerce expansion.

Third, rural e-commerce companies in adjacent regions should prioritize diverse consumer demands and implement differentiated business strategies. This strategy will allow them to enhance each other’s abilities and prevent harmful uniform competition. Provinces such as Hebei and Shandong can mitigate direct competition with adjacent regions by concentrating on specialized agricultural products and precise product placement, thereby augmenting regional e-commerce competitiveness. Additionally, collaborative efforts like e-commerce industry summits and the establishment of rural e-commerce incubators should be encouraged to create unified platforms that enable the exchange of information, resources, and markets among provinces, thereby ensuring the sustainable and robust development of rural e-commerce.

5.3. Limitations and Future Research

This research possesses multiple limitations. The research is limited to the period from 2011 to 2022 due to the lack of data for 2023 and 2024 from statistical departments. This temporal constraint may not adequately reflect the dynamic evolution of rural e-commerce in light of recent legislative and technology advancements. Therefore, subsequent studies have to integrate more current data to enhance timeliness. The development of the index system was impeded by the absence of publicly accessible data regarding rural e-commerce sales and procurement volumes. Consequently, the research employed aggregate e-commerce data as a surrogate. This approach offers insight into rural e-commerce development but may lack the detail necessary for effectively representing rural locations, thereby constraining the interpretability of the findings. Subsequent studies may enhance data sources or directly gather data to ensure the index system more accurately reflects the attributes of rural e-commerce. This study concentrates on province units, which, while beneficial for comprehending the general spatial distribution of rural e-commerce development, fail to account for differences at more granular geographical levels, such as counties or townships. This limitation may mask intra-provincial urban–rural or regional differences, so hindering the comprehension of rural e-commerce development at a micro level. Subsequent research should examine smaller geographical units to more accurately reflect regional disparities and improve the spatial analysis of rural e-commerce development capability.