Introduction

The development level of rural areas in China is relatively low, with a significant disparity compared to urban areas, and rural poverty remains an issue. In 2023, the per capita disposable income of urban residents was 51,821 yuan, while that of rural residents was 21,691 yuan. The urban-to-rural per capita disposable income ratio was approximately 2.39:1. China’s total grain production reached 695.41 million tons, rural regions are rich in agricultural and labor resources, and infrastructure is continuously improving. However, these areas face challenges such as poor market information flow and weak logistical foundations. The government has implemented a series of measures to support the online sale of agricultural resources, such as enhancing rural infrastructure and promoting rural e-commerce. The scale of online shopping users in China is enormous, with a very high penetration rate. China’s online retail sales totaled 15.42 trillion yuan in 2023, marking an 11% increase and maintaining its position as the world’s largest online retail market for the 11th consecutive year. E-commerce is characterized by its large market size and rapid growth. As of December 2023, the number of rural internet users in China reached 326 million, accelerating the closure of the urban-rural digital divide. Farmers use e-commerce platforms to sell agricultural products to urban consumers, increasing the added value of these products and farmers’ incomes (Han et al. 2021). Rural e-commerce has provided new opportunities for the economic development of rural areas (Huijing et al. 2021). Government departments have fostered the creation of e-commerce cooperatives and leading rural e-commerce enterprises, building a comprehensive rural e-commerce ecosystem that has created more entrepreneurial opportunities and significantly promoted rural economic development. However, due to the lagging economic development in rural areas and the shortage of e-commerce talent, rural e-commerce entrepreneurship faces multiple challenges.

The establishment of the rural e-commerce entrepreneurial ecosystem is primarily driven by dominant rural industries, wherein e-commerce platforms serve as the carrier. This system strategically leverages diverse entrepreneurial resources to promote the high-quality development of rural industries (Huijing et al. 2021), thereby accelerating the process of achieving a distinctly Chinese form of modernization (Baozhou et al. 2020). In recent years, initiatives have been implemented to encourage young people to return to their hometowns to pursue entrepreneurial endeavors. These efforts aim to enhance the support system for rural entrepreneurship and optimize the entrepreneurial ecosystem (Cao and Shi, 2021). However, the reality reveals that rural entrepreneurship encounters challenges arising from outdated infrastructure, limited digital resources, and an environment less conducive to entrepreneurial pursuits. The evolution of the rural e-commerce entrepreneurial ecosystem is further complicated by the absence of reference cases and an unclear path for fostering its growth (Qiao et al. 2019). In this context, it is of great significance to explore the key factors influencing the successful establishment of the rural e-commerce entrepreneurial ecosystem, analyze their complex effects, and dissect the driving pathways. This is essential for achieving high-quality development of rural e-commerce and ensuring the sustainability of rural e-commerce entrepreneurship.

The entrepreneurial ecosystem is a complex adaptive system approach (Roundy, et al. 2018). It involves the aggregation of innovative elements, the formation of advantageous industry clusters, and the integration of key technologies, among other multifaceted factors that empower the rural e-commerce entrepreneurial ecosystem. Due to regional disparities and variations in natural and resource endowments among provinces, such complexities in governance issues cannot be sufficiently explained by a single factor. This necessitates the exploration of the potential effects of multiple-factor configurations. The construction of a high-level entrepreneurial ecosystem for rural e-commerce poses a crucial yet underexplored reality and scientific question. This paper addresses this gap by adopting a configuration perspective and employing a qualitative comparative analysis method to investigate the systematic, holistic, and complex issues within the rural e-commerce entrepreneurial ecosystem.

China has 34 provincial administrative regions. However, data is unavailable for three of these regions: Hong Kong Special Administrative Region, Macao Special Administrative Region, and Taiwan Province. Therefore, this study utilizes data from the remaining 31 provincial administrative regions. Focusing on these 31 provinces and adopting the technology-organization-environment framework, this research investigates the key elements and driving paths contributing to the high-level development of rural e-commerce entrepreneurial ecosystems. The analysis considers the interactions of seven factors across the technology, organization, and environment dimensions, with the ultimate aim of enhancing the quality of rural e-commerce entrepreneurship and fostering rural economic development.

Literature review and model construction

Research progress on rural e-commerce entrepreneurial ecosystem

The exploration of the connotations of entrepreneurial ecosystems can be categorized into two schools of thought: the “environmental” perspective and the “actor-environment” perspective. Since the introduction of entrepreneurial ecosystem theory, an enduring debate has persisted between the “environmental” theorists (Cohen, 2006) and the proponents of the “actor-environment” perspective (Suresh and Ramraj, 2012; Vogel, 2013), with consensus remaining unlikely. The “environment” theory emphasizes the influence of external environments on entrepreneurial activities, asserting that the formation and development of entrepreneurial ecosystems are primarily constrained and influenced by external factors. These factors include market demand, competitive conditions, and policy support (Shane and Venkataraman, 2000). The “actor-environment” theory, on the other hand, highlights the interactive relationship between entrepreneurial actors and their environment, arguing that the subjective initiative of the entrepreneurial actors is crucial for the development of the entrepreneurial ecosystem. Through interaction with the external environment, entrepreneurs continuously adjust their strategies to adapt to environmental uncertainties and complexities. In the study of the rural e-commerce entrepreneurial ecosystem in China, rural environments are characterized by resource scarcity and asymmetric market information. The “environment” theory alone cannot adequately explore the influence of internal actors within the entrepreneurial ecosystem. However, the “actor-environment” theory helps in understanding how the operations and collaborations of multiple actors within the rural e-commerce entrepreneurial ecosystem impact its development (Sarasvathy, 2001). Research on entrepreneurial ecosystems in rural e-commerce is intertwined with the examination of entrepreneurial actors and the rural environment. Furthermore, multiple studies support extending the “actor-environment” perspective to the Chinese context (Huang et al. 2023). Therefore, this study adopts the “environment-actor” perspective, as proposed by Suresh (Suresh and Ramraj, 2012), as a representative scholar, emphasizing the diverse and complex interactions between internal actors and external environmental elements, thereby constituting a dynamic and interconnected system.

Existing scholarship has explored the connotations and types of rural entrepreneurial ecosystems (Aguilar, 2021; Cai et al. 2016; Leong et al. 2016). Asmit et al. (2024) employed bibliometric methods to highlight that rural entrepreneurial ecosystems consist of both actors and non-actors. Lin et al. (2024) investigated the mechanisms by which participation in rural e-commerce influences farmers’ entrepreneurial behaviors and agricultural entrepreneurship. Zeng et al. (2019) identified that agricultural e-commerce clusters involve factors such as technology introduction, industrial bases, network infrastructure, entrepreneurial talent, and local governments. Zang et al. (2023) discussed the beneficial outcomes associated with well-developed e-commerce ecosystems in rural areas. In summary, members of rural e-commerce entrepreneurial ecosystems achieve resource-sharing and complementary advantages through collaborative efforts. This ecosystem aims to promote rural entrepreneurial development through e-commerce, encompassing multiple dimensions such as technological support, organizational structures, and policy environments. Existing research has laid the groundwork for understanding the connotations and components of rural e-commerce entrepreneurial ecosystems. However, most studies are qualitative and often focus on individual entrepreneurial entities or communities, lacking a systematic, multidimensional perspective. Additionally, research topics frequently address rural e-commerce and entrepreneurial ecosystems separately rather than as integrated systems.

This study expands upon existing research on rural e-commerce entrepreneurial ecosystems by addressing three critical areas. Firstly, it examines the construction paths of these ecosystems, which have been overlooked in current literature that primarily focuses on micro-content and immediate development. A deeper analysis is required to systematically deduce provincial-level construction paths. Secondly, it advocates for a holistic perspective and configuration approach, as previous studies have lacked comprehensive views and relied on singular perspectives. Further exploration into configuration approaches is necessary to fully understand the influencing paths. Lastly, the paper introduces qualitative comparative analysis methods to bridge the gap between micro-level descriptions or policy analyses and macro-level studies. By constructing a macro-level research framework, these methods facilitate the emergence of sustainable development paths for rural e-commerce ecosystems.

Theoretical foundation

Complex Adaptive Systems (CAS) theory involves understanding how different parts of a system interact and co-evolve to adapt to complex and dynamic environments (Miller and Page, 2009). This theory has broad applications across fields such as ecology, economics, social sciences, and management. In complex systems, individuals interact to form structures or patterns, adjusting their structures and behaviors to respond to changes in the dynamic environment. The overall properties of the system cannot be simply deduced from its individual parts; rather, they emerge through the interactions among the parts (Holland and Hidden, 1995).

CAS theory is well-suited for studying entrepreneurial ecosystems. Brett, in “Admired Disorder: A Guide to Building Innovation Ecosystems,” describes entrepreneurial ecosystems as complex dynamic systems. Levin (1998) applied CAS theory to ecological systems, discussing self-organization, adaptability, and emergent behaviors within ecosystems. Uhl-Bien et al. (2007) proposed Complexity Leadership Theory, which explores how leaders in complex adaptive systems guide organizations to adapt and evolve by facilitating interactions and fostering innovation. CAS theory not only addresses the adaptive interactions within organizations but also considers the adaptability to external environments. An entrepreneurial ecosystem, as a complex adaptive system, requires effective internal interactions among its elements and must flexibly respond to and adjust to changes in the external environment (Hazy and Goldstein, 2010). This involves analyzing the roles and behaviors of organizations within a broader ecosystem, studying how organizations develop and utilize dynamic capabilities to adapt to external changes (Teece et al. 1997), and examining how interactions between organizations and their external environments co-evolve (Lewin and Volberda, 1999).

Theoretical framework

The “actor-environment” theory of entrepreneurial ecosystems and the internal and external adaptability of CAS theory both emphasize that entrepreneurial ecosystems are based on the synergy of internal organizational elements and the resource endowments of the external environment. Given the relatively underdeveloped nature of rural areas in China, technological factors such as infrastructure and digital resources play a particularly crucial role in rural e-commerce entrepreneurship. Through a literature review, it has been found that the construction of China’s rural e-commerce entrepreneurial ecosystem is influenced by factors at the technological, organizational, and environmental levels. Synthesizing these arguments, this study introduces the TOE framework.

The TOE Framework (Technology-Organization-Environment Framework) was proposed in 1990 as a technological governance framework (Rnatzky and Fleische, 1990). It comprises three pivotal components: technological factors, organizational factors, and environmental factors. Technological factors encompass the technological capabilities associated with modernization and digitization (Chau and Tam, 1997). Organizational factors pertain to internal influences within the system. Environmental factors encompass potential external influences on the system. The TOE Framework has been extensively applied in research related to information systems and developmental stages (Han et al. 2023). The rural e-commerce entrepreneurial ecosystem shapes an intricate network structure through the collaborative creation of value by multiple stakeholders. It lends itself well to examination within the TOE theoretical framework. The TOE framework provides a structured analytical tool that helps identify and categorize the key drivers influencing the rural e-commerce entrepreneurial ecosystem. The operations of internal organizations within the ecosystem and the resource endowments of the external environment are significant influences (Guangping et al. 2020). Technological and innovation capabilities are critical factors in assessing whether the internal and external elements can drive the efficient operation of the system (Brett, 2019). The selection of conditions follows the standards of relevant theories and employs the method of “literature collection” to conduct a comprehensive review of the forefront of the research field, thereby constructing a specific research design that considers the conditions within existing studies as much as possible. Based on a literature review related to “rural e-commerce entrepreneurship,” an analytical framework for driving the construction of the rural e-commerce entrepreneurial ecosystem is proposed under the CAS theory and the TOE framework, as shown in Tables 1 and 2.

Table 1 The matching of the TOE framework with specific indicators.
Full size table

Technological factors and the construction of the rural e-commerce entrepreneurial ecosystem

Technological factors encompass two key secondary elements: digital infrastructure and technological innovation capability. Technological factors refer to the various technological conditions and capabilities that influence outcomes, including technological infrastructure and technological capability. Digital infrastructure and technological innovation capability fall under this category. Digital infrastructure encompasses the hardware and software support for technology (Wang et al. 2022), while technological innovation capability refers to the ability to use and innovate with technology (Miles and Morrison, 2020). Both factors impact the rural e-commerce entrepreneurial ecosystem through the technological dimension. For instance, robust digital infrastructure can support more diverse e-commerce applications, and strong technological innovation capability can lead to more innovative e-commerce models (Abeysinghe and Malik, 2021). Digital infrastructure provides foundational support, while technological innovation capability drives the effective utilization and innovation of technology. According to CAS theory, technological innovation capability fosters internal innovation and adaptability within the system. Through technological innovation, new paths for survival and development can be found in complex environments, enhancing the adaptability and sustainable development of the entire system.

Digital infrastructure

Digital infrastructure refers to the “services and facilities related to digital technology and organizational structure that facilitate the functioning of industries”. The development of digital infrastructure transcends the information divide between urban and rural regions, streamlines the alignment of urban and rural markets, lowers transaction costs, modernizes the distribution model for agricultural products, innovates consumer behavior, triggers the coupling of diverse resource elements, and catalyzes the growth of rural industries (Chao et al. 2021). Fundamental components of digital infrastructure, including the internet, network infrastructure, intelligent supply chains, and smart logistics, are indispensable building blocks for establishing the entrepreneurial ecosystem in rural e-commerce (Xu and Zhang, 2022; Zhou and Li, 2021).

Technological innovation capability

Entrepreneurship is inseparable from innovation. Within an entrepreneurial ecosystem, entrepreneurial resources are intrinsically linked with technological innovation capability, serving as a fundamental prerequisite for the conversion of resources. The creation of an entrepreneurial ecosystem centered around rural e-commerce demands heightened technological empowerment and innovation capability. This process propels the flow of technology, capital, and talent toward rural areas, ensuring harmonization with the rural context and adaptability to rural industries (Yan et al. 2020). Consequently, it propels the advancement of e-commerce entrepreneurship with an emphasis on high quality, endowing the rural e-commerce entrepreneurial ecosystem with both stability and sustainability.

Organizational factors and the construction of rural e-commerce entrepreneurial ecosystem

Organizational factors specifically encompass two secondary conditions: agricultural leading enterprises and the foundation of e-commerce cooperatives. Organizational factors refer to the internal components of the system, their characteristics, and resource capabilities. Leading agricultural enterprises are large companies that hold a controlling and leading position in the agricultural industry, primarily focusing on the processing or distribution of agricultural products. These enterprises connect with farmers through various benefit-sharing mechanisms, helping farmers enter the market and organically integrating production, processing, and sales of agricultural products, thus mutually promoting each other. Leading agricultural enterprises, positioned at the core of the agricultural industry, leverage their scale, technology, and management advantages to drive the development of the entire agricultural sector and advance the process of agricultural modernization. Rural e-commerce cooperatives break the traditional barriers of information isolation and fragmented operations in agriculture, providing farmers with more market information and sales channels. Utilizing Internet platforms, these cooperatives offer their members services related to the sale, processing, transportation, and storage of agricultural products, as well as technical and informational services related to agricultural production and management. This facilitates the circulation of agricultural products and plays a crucial role in promoting the development of the rural e-commerce entrepreneurial ecosystem. Organizational capability is crucial for the sustainability of innovation (Bertot and Jaeger, 2006). In relevant documents promoting the high-quality development of rural e-commerce, the Chinese government emphasizes the crucial role of leading agricultural enterprises and e-commerce cooperatives in rural e-commerce entrepreneurship. Therefore, these two factors are classified as organizational factors.

Agricultural leading enterprises

These enterprises signify the level of industry specialization and concentration. The concentration of agricultural industries within a particular geographic area mirrors the agglomeration capacity of industrial resources in the regional innovation ecosystem. Agricultural leading enterprises play a pivotal role in driving local industrial growth, streamlining commodity trade channels, enhancing industrial resource diversity, and, by promoting supply chain standardization and automation, acting as a strong catalyst for rural e-commerce entrepreneurship. Leading agricultural enterprises play a crucial role in the rural e-commerce entrepreneurial ecosystem. They typically possess strong capabilities in resource integration, brand influence, and market channels, which can significantly drive the development of rural e-commerce (Gao, 2022).

Foundation of e-commerce cooperatives

This represents the amalgamation of the starting point and entrepreneurial potential within e-commerce entrepreneurship in a designated region. It functions as a pivotal hub for facilitating online commodity transactions between urban and rural areas. A robust cooperative foundation promotes the upward flow of e-commerce agricultural products and the downward flow of industrial goods. It widens the scope for entrepreneurial collaboration and diminishes the investment overheads associated with rural e-commerce entrepreneurship (Cristobal-Fransi et al. 2020). E-commerce cooperatives play a crucial role in the rural e-commerce entrepreneurial ecosystem by aggregating resources, reducing costs, and enhancing bargaining power (Chen et al. 2023).

Environmental factors and the construction of rural e-commerce entrepreneurial ecosystem

Environmental factors specifically include government attention allocation, agricultural development level, and human capital level as three secondary conditions. These factors respectively empower the construction of the rural e-commerce entrepreneurial ecosystem as the government support environment, agricultural economic environment, and talent resource environment.

Government attention allocation

This refers to the degree of government attention and support for rural e-commerce entrepreneurship. Government attention to rural entrepreneurship enhances the willingness of returnees and rural residents to start businesses and improves the institutional environment to incentivize more entrepreneurs to start businesses in rural areas (Asmit et al. 2024). Many countries use policy and political support to promote innovation and entrepreneurship (De Rosa and McElwee, 2015), increasing the number of entrepreneurial entities in the construction of the rural e-commerce entrepreneurial ecosystem.

Level of agricultural development

Agricultural resources are the primary foundation for rural entrepreneurship, representing the attractiveness and development potential of rural areas. Entrepreneurs tend to gravitate toward regions with abundant agricultural resources (Mei et al. 2020). This factor is a critical attribute of the inherent endowment of rural areas, exerting multifaceted influences on the entrepreneurial ecosystem (Suresh and Ramraj, 2012). The availability and quality of agricultural resources, as well as the modernization and technological advancement of agriculture, play a significant role in shaping the entrepreneurial environment in rural areas. A high level of agricultural development provides a solid foundation for rural e-commerce entrepreneurship by ensuring a stable and diverse supply of agricultural products.

Level of human capital

Human capital refers to the capital embodied in laborers, including their knowledge, skills, cultural level, and technological proficiency, injecting new vitality into rural e-commerce entrepreneurship as a key element of talent resources. Human capital is a scarce resource in rural areas and serves as an important entrepreneurial support factor, driving the modernization and intelligent development of agriculture. Although not necessarily directly involved in entrepreneurial teams, high levels of human capital in rural areas can enhance entrepreneurial enthusiasm. High levels of human capital in rural areas facilitate the benign development of rural entrepreneurship and innovation, providing new momentum for the sustainable development of rural e-commerce entrepreneurship.

Drawing upon existing research findings and primary field data and adhering to the CAS theory and TOE framework, we developed a research model illustrated in Fig. 1. The theory of Complex Adaptive Systems (CAS) posits that individuals and components within a system interact and adapt continuously, leading to emergent complex behaviors of the whole. In the rural e-commerce entrepreneurial ecosystem, various elements interact, collectively determining the system’s developmental trajectory and outcomes. CAS theory aids in elucidating the intricate relationships among diverse factors within the rural e-commerce ecosystem.

Fig. 1
figure 1

Research framework: a driving model for rural e-commerce entrepreneurship ecosystem.

Full size image

This study applies the Technological-Organizational-Environmental (TOE) framework to assess the rural e-commerce entrepreneurial ecosystem, examining the specific roles of technological innovation, organizational structures, and environmental factors in system development. The research framework encompasses seven key variables across technological, organizational, and environmental dimensions. At the technological level, it involves digital infrastructure and technological innovation capability. The organizational dimension focuses on foundation of e-commerce cooperatives and agricultural leading enterprises, while the environmental aspect investigates the impacts of government attention allocation, level of agricultural development, and level of human capital. These variables are considered pivotal in shaping the formation and evolution of the rural e-commerce entrepreneurial ecosystem.

Grounded in CAS theory and the TOE framework, this study aims to uncover and analyze the driving pathways of China’s rural e-commerce entrepreneurial ecosystem. By developing a robust research framework and categorizing variables systematically, it provides theoretical support and practical guidance for promoting rural e-commerce entrepreneurship and rural economic development.

Research methodology and data sources

Method

This study employs Qualitative Comparative Analysis (QCA), a set-theoretical approach introduced by the American sociologist Ragin (Ragin, 2014). Rooted in a configurational perspective, QCA adeptly addresses complex causal relationships, integrating qualitative research techniques with formal mathematical methods. Based on CAS theory, QCA suits rural e-commerce ecosystem development for three reasons: Empirical insights reveal diverse factors and pathways contributing to high-level rural e-commerce ecosystems. It explores multiple developmental pathways through varied factor configurations. QCA is advantageous for analyzing a small number of cases, typically 10 to 50 when traditional statistical analysis methods are impractical. The method offers a holistic viewpoint, facilitating cross-case analysis, beneficial for investigating complex scenarios involving multiple cases, common in rural e-commerce ecosystems.

QCA has been applied in various academic disciplines, including management, sociology, and political science. Given that rural e-commerce entrepreneurial ecosystem development is a multifaceted endeavor involving multiple stakeholders, including farmers, agricultural enterprises, cooperatives, and entrepreneurs, and characterized by complex interrelationships among various influencing factors, we adopted the Fuzzy Set Qualitative Comparative Analysis (fsQCA) method (Fiss, 2011). Unlike the other two QCA approaches (csQCA and msQCA), fsQCA allows for the observation of degrees of membership and extent of change in addressing research questions. This methodological choice is intended to enhance the robustness and depth of our analysis in understanding the dynamics of rural e-commerce entrepreneurial ecosystems.

Case selection

Rural e-commerce entrepreneurship in China is still in its nascent stage, and there are significant regional variations in rural e-commerce-related outcomes at the provincial level. Selecting cases at the provincial level is informative and can reflect reality and research issues. Different provinces exhibit diverse regional characteristics, levels of economic development, and degrees of policy support (Han et al. 2023). Analyzing data at the provincial level allows for a comprehensive examination of how these differences affect the rural e-commerce entrepreneurial ecosystem. Provincial-level data provide a comprehensive perspective, covering various regions and aiding in a more thorough understanding of the overall situation and characteristics of the rural e-commerce entrepreneurial ecosystem. Moreover, provincial-level data are relatively easy to obtain and have higher data quality and credibility, thus providing ample data foundation for research.

China has a total of 34 provincial-level administrative regions. Since data from the Special Administrative Regions of Hong Kong and Macau, as well as Taiwan Province, are missing, this paper selects the remaining 31 provinces (autonomous regions, municipalities directly under the central government) as samples. The main reasons for this selection are as follows: First, at the current stage, the construction and development of rural e-commerce entrepreneurial ecosystems in China are still relatively dispersed, with rural e-commerce entrepreneurship data and ecosystem construction mainly distributed at the provincial level. Second, selecting sample cases ensures better representativeness. Among the 31 provinces (autonomous regions, municipalities directly under the central government), there are regions with developed rural e-commerce, as well as underdeveloped areas; there are areas with well-developed digital economies, as well as regions with lagging levels of informatization.

Data sources

In this study, all conditional variables are indicators for the year 2020, considering the lagged nature of the results, while the outcome variables are indicators for the year 2021.

Outcome variable

The primary outcome variable of interest in this study is the level of development of rural e-commerce entrepreneurial ecosystems. With the rapid development of rural e-commerce, imbalances in regional development persist. The entrepreneurial ecosystem is a multi-level, multi-agent complex system, presenting challenges in the construction and measurement of indicators. Drawing on existing research (Cheng and Lv, 2021; Feng and Zhang, 2022), this study endeavors to objectively assess the development level of the Chinese rural e-commerce ecosystem by constructing a comprehensive evaluation indicator system and applying the entropy weight method. Building on prior research on evaluating entrepreneurial ecosystem development, a sustainable entrepreneurial ecosystem emphasizes comprehensive and integrated system development, maintaining equilibrium across three dimensions: economic, organizational, and social. To gauge the developmental level of rural e-commerce entrepreneurial ecosystems in each province, three specific indicators are employed:

  1. (1)

    Provincial-level rural e-commerce cooperatives’ network retail situation: This indicator assesses the economic dimension of the ecosystem’s development. For this indicator, the datasets were acquired from the “2021 National Report on the Development of Digital Agriculture and Rural E-commerce in Counties.”

  2. (2)

    Number of provincial-level rural entrepreneurship parks (bases): This indicator measures the level of innovation and entrepreneurship within the ecosystem. The data was obtained from the “2021 National Report on Rural Entrepreneurship Parks (Bases).”

  3. (3)

    Number of Taobao villages at the provincial level: This indicator reflects the level of cluster effects within the ecosystem (Liu et al. 2020). The data for this indicator were sourced from the Alibaba Research Institute.

To assess the overall development level of rural e-commerce entrepreneurial ecosystems in the 31 provincial regions, a comprehensive evaluation was conducted using the entropy weight method. These data sources and indicators provide a multidimensional perspective on the development of rural e-commerce entrepreneurial ecosystems, capturing economic, innovative, and cluster-related aspects. The use of multiple indicators allows for a more comprehensive and nuanced assessment of these ecosystems’ development levels across different regions in China.

Conditional variable

Technological factors

a. Digital infrastructure: This variable, which indicates the level of digital infrastructure in each province, was determined using the “Digital Development Composite Index” for each province in 2020. This composite index is derived from the comprehensive evaluation of digital innovation inputs, digital infrastructure construction, digital economic development, and the development of digital society and government, as reported in the “China Regional Digital Development Index Report.” It provides an objective assessment of the digital technology development in all 31 provinces.

b. Technological innovation capability: To measure the technological innovation capability of each province, the study utilized the comprehensive indicator values of China’s regional innovation capabilities. Data for this variable were acquired from the “China Regional Innovation Capability Evaluation Report” published by the Ministry of Science and Technology (Chen et al. 2022).

Organizational factors

a. Agricultural leading enterprises: This parameter represents the number of key national leading agricultural enterprises at the provincial level. These enterprises play a crucial role in driving industry clustering and promoting the gradual maturation of the e-commerce industry. Data for this variable were obtained from the “China Agricultural Yearbook.”

b. Foundation of e-commerce cooperative: This variable was determined using the number of rural e-commerce cooperatives at the provincial level. The data were sourced from the “National Report on the Development of Digital Agriculture and Rural E-commerce in Counties.”

Environmental factors

a. Government attention allocation: To gauge the government’s attention allocation, the number of policy documents related to “rural e-commerce entrepreneurship” issued by each province or city was used as a proxy indicator. The quantity of such policy documents reflects the level of support for rural e-commerce entrepreneurship by the provincial governments. The count of relevant policy documents was obtained from the policy files disclosed on the official websites of provincial and municipal governments.

b. Level of agricultural development: Assessed by the proportion of the added value of the primary industry to the GDP, this variable reflects the level of agricultural development in each province (Chen et al. 2022). Data for each province in the year 2020 were acquired from the “China Statistical Yearbook.”

c. Level of human capital: The level of human capital in the primary industry of each province is calculated based on the average years of education for the labor force in the primary industry and the number of labor force in that sector (Bai and Yang, 2019). Data for this variable were obtained from the “China Labor Statistical Yearbook.”

As shown in Table 2, These conditional variables are essential for the study, representing the technological, organizational, and environmental factors that can influence the development of rural e-commerce entrepreneurial ecosystems in each province.

Table 2 Indicator description of variables.
Full size table

In the fsQCA framework, each case exhibits membership degrees across multiple variable sets. Prior to conducting data analysis, the calibration of both the outcome and condition variables is crucial. Calibration involves assigning membership scores to cases within sets, distinguishing between direct and indirect calibrations. This study employs direct calibration, utilizing three anchor points selected based on the original data. Drawing inspiration from previous research (Rihoux and Ragin, 2009), the 0.95 quantile, 0.5 quantile, and 0.05 quantile represented complete membership, crossover points, and complete non-membership, respectively. The calibration results are presented in Table 3.

Table 3 Variable calibration results.
Full size table

Data analysis and results

Necessary condition analysis

Using the fsQCA 3.0 software, we analyzed whether any single factor constitutes a necessary condition for the establishment of a rural e-commerce entrepreneurial ecosystem. This entailed evaluating the consistency and coverage of individual variables (refer to Table 4). A condition variable is considered necessary if its consistency surpasses 0.9, signifying its capability to independently explain the outcome variable.

Table 4 Analysis of necessary conditions.
Full size table

The tabulated findings suggest that, for both high-level and non-high-level rural entrepreneurial ecosystem development, the consistency of the prerequisite condition variables falls short of 0.9. This indicates that no single factor qualifies as a necessary condition for the formation of a rural e-commerce entrepreneurial ecosystem. The consistency value for the variable ‘Foundation of E-commerce Cooperatives’ is 0.899, which is very close to the 0.9 threshold for a necessary condition. The data indicate that while the presence of E-commerce cooperatives is highly significant, it is not an absolute requirement in all cases. In fsQCA, conditions typically do not operate in isolation but rather interact with other conditions to influence the outcome. Therefore, it is essential to consider the interplay between this condition and others, and to analyze how different combinations of conditions affect the outcome through configurational analysis. The attainment of a high-level rural e-commerce entrepreneurial ecosystem results from the combined influence of technological, organizational, and environmental factors. Consequently, subsequent research is needed to explore the configurational effects of the causal variables.

Configuration analysis

This study utilized the fsQCA software to construct a truth table and perform a configuration analysis. This was aimed at exploring the sufficiency of various configurations formed by multiple condition variables leading to the emergence of outcomes. Drawing from established research practices, we set the consistency threshold for sufficiency at 0.8 (Ordanini et al. 2014), combined with case sample and frequency thresholds set at 1, and the PRI (proportional reduction in inconsistency) threshold at 0.6 (Greckhamer et al. 2018). This analysis produced complex solutions, intermediate solutions, and parsimonious solutions. Intermediate solutions only included logical remainders that conformed with theoretical expectations, while parsimonious solutions encompassed all logical remainders.

In accordance with established research practices, we primarily report intermediate solutions, identifying core conditions and marginal conditions within the configuration. Core conditions are variables present in both intermediate and parsimonious solutions, while variables exclusive to intermediate solutions are considered marginal conditions. Based on the fsQCA analysis, the pathway configurations that drive and hinder the construction of the rural e-commerce entrepreneurial ecosystem were identified, as presented in Table 5. Table 5 shows that fsQCA effectively identified 10 pathways, of which 4 configurations can lead to a high-level rural e-commerce entrepreneurial ecosystem, while 6 configurations result in a non-high-level rural e-commerce entrepreneurial ecosystem.

Table 5 fsQCA results.
Full size table

Configuration of high-level construction

Among the configurations that form a high-level rural e-commerce entrepreneurial ecosystem, the overall solution consistency is 0.94. The coverage rate is 0.729, meaning that these 4 configurations can explain 72.9% of the cases in which the rural e-commerce entrepreneurial ecosystem is at a high level. Individual configuration conditions have a consistency exceeding 0.9, demonstrating the strong explanatory power of these conditions. The results suggest that empirical analysis is effective.

Configuration represents the arrangement and combination of different resources that yield similar outcomes. Diverse contextual configurations signify various pathways driving the construction of the rural e-commerce entrepreneurial ecosystem. From the distribution of elements within contextual configurations, “agricultural leading enterprises,” “foundation of e-commerce cooperatives,” and “government attention allocation” emerge as the core dynamic capabilities essential for a high-level rural e-commerce entrepreneurial ecosystem. Among these, the configuration coverage rates are notably high, reaching up to 0.474 for the configuration centered on “agricultural leading enterprises” and “foundation of e-commerce cooperatives.” This indicates that the construction level of the rural e-commerce entrepreneurial ecosystem relies on the local agricultural leading enterprises for resources, while the foundation of e-commerce cooperatives is critical for promoting high-quality development within the system, and government attention allocation substantially supports its construction level.

“Digital infrastructure,” “technological innovation capabilities,” and “level of human capital” typically function as supplementary dynamic capability conditions within the configuration. Their impact on the construction level of the rural e-commerce entrepreneurial ecosystem remains relatively low. Although reliant on the resource environment of digital infrastructure hardware conditions, technological innovation capabilities, and the level of human capital, these factors fail to confer a relative advantage to the ecosystem during its developmental construction process. “Level of agricultural development” often exists as a condition that is either absent or inconsequential, signifying that its influence on outcomes is not significant. The original coverage rates of the four pathways for constructing a high-level rural e-commerce entrepreneurial ecosystem all surpass the singular coverage rate, indicating instances supporting the existence of multiple compound causal pathways. Utilizing the identification method of conditions, these four pathways can be categorized into three typical patterns: technology-organization-government-driven, organization-environment-driven, and holistic synergy-driven. Table 5 provides detailed information on these pathways, outlining the driving forces for the high-level construction of rural e-commerce entrepreneurial ecosystems. A specific analysis follows:

Segment 1: Technology-Organization- Government Driven

Configuration one indicates that technological factors, organizational factors, and government attention allocation are all essential prerequisites for the development of high-level rural e-commerce entrepreneurial ecosystems. Among these, organizational factors and government attention allocation play a central role, while technological factors serve as a marginal condition, leading to its classification as “technology-organization-government-driven.” The consistency of this pathway is 0.942, with an original coverage of 0.367 and a unique coverage of 0.062. This suggests that this pathway can explain 36.7% of high-level cases in the construction of rural e-commerce entrepreneurial ecosystems, with approximately 6.2% of cases exclusively explained by this pathway. This particular pathway underscores the significance of organizational conditions in facilitating the construction of rural e-commerce entrepreneurial ecosystems. Even in provinces with lower levels of agricultural development and human capital, the support of government and organizational conditions remains instrumental. This effect is especially pronounced in provinces with good digital infrastructure and technological innovation capabilities. The results suggest that technical and organizational conditions can effectively overcome the constraints of objective environmental endowments, such as agricultural resources and human capital, in the construction of rural e-commerce entrepreneurial ecosystems.

In this driving mode, the government formulates policies, implements actions, and encourages rural entrepreneurship, thereby expanding the population of entrepreneurial entities. A robust foundation of rural cooperatives, coupled with the strength of agricultural leading enterprises, reduces the costs and risks associated with rural e-commerce entrepreneurship, facilitating a virtuous cycle and sustainable development of the rural e-commerce entrepreneurial ecosystem.

Exemplary cases of this driving mode are concentrated in the eastern coastal provinces (Fig. 2), including Guangdong, Zhejiang, Fujian, and Shandong, with Guangdong Province being particularly representative. Guangdong’s strategic emphasis on agricultural product processing and technological research and development aligns with national directives. The province has established an exemplary database of rural innovation and entrepreneurship leaders, frequently organizing events promoting technological innovation in agricultural products. As a result of these initiatives, Guangdong has emerged as a focal point for national agricultural processing enterprises and leading agricultural industrialization companies.

On one hand, Guangdong excels in establishing comprehensive service points for farmers’ cooperatives at the county, township, and village levels. This involves the creation of platforms for production and sales integration, logistics coordination, online sales, and transaction facilitation. The province adeptly integrates social resources to provide specialized and integrated services. On the other hand, Guangdong’s “Yue” branded agricultural products focus on distinctive industries and resource elements, promoting the development of Lingnan industrial clusters. These efforts align with industry clusters and brand culture, effectively shaping the rural e-commerce entrepreneurial ecosystem.

Fig. 2
figure 2

Explanation case of Segment 1.

Full size image

Segment 2: Organization-Environment Driven

The composite path analysis indicates that the organizational variables serve as core driving conditions, while the environmental variables play a supplementary role in driving outcomes. Termed the “organization-environment-driven” model, this configuration suggests that, even in the absence of high-level digital technology and innovation capabilities, a high-level rural e-commerce entrepreneurial ecosystem can be established with favorable conditions in agricultural development and human capital. In this model, abundant agricultural resources and robust human resources form the foundation for constructing the rural entrepreneurial ecosystem. Agricultural leading enterprises enhance the ecosystem by establishing reliable and stable interest connection mechanisms, thus increasing entrepreneurial entities within the system. These enterprises, along with rural cooperatives, drive the virtuous operation of supporting entities such as e-commerce platforms, entrepreneurial entities, logistics organizations, and financial institutions in the rural e-commerce entrepreneurial ecosystem. This leads to a positive cycle and sustainable development of the entrepreneurial ecosystem.

Typical cases of the “organization-environment-driven” model are found in Hebei and Heilongjiang (as shown in Fig. 3). Taking Hebei as an example, the province prioritizes the development of agricultural industrialization, exemplified by the issuance of the “Opinions on Enlarging and Strengthening Agricultural Industrialization Leading Enterprises.” Key provincial leading enterprises possess “triple excellence and one standard” certification, emphasizing the development of distinctive advantageous industries. In addition, Hebei Province emphasizes the construction of professional cooperatives, with 120,000 farmer cooperatives leading modern agricultural development. These cooperatives establish bases and cultivate multiple rural entrepreneurial and innovative zones, implementing a comprehensive “regional, enterprise, product” integrated brand strategy. Collaborative efforts involving various governmental departments, including the Ministry of Finance and the Ministry of Agriculture, are pivotal in building regional agricultural leading enterprises, promoting e-commerce industry development, and supporting brand construction. The construction of rural professional cooperatives and entrepreneurial zones creates an atmosphere for rural e-commerce entrepreneurship, facilitating the organized development of the high-level rural e-commerce entrepreneurial ecosystem.

Fig. 3
figure 3

Explanation case of Segment 2.

Full size image

Segment 3: Holistic Synergy-Driven

The combined pathways S3a and S3b illustrate that technological, organizational, and environmental factors collectively serve as prerequisite conditions for the high-level development of rural e-commerce entrepreneurial ecosystems, referred to as the “Comprehensive Synergistic-Driven” model. The S3a pathway indicates that provinces with well-established technological infrastructure, high innovation capabilities, and a high level of human capital can achieve high-level development if they also possess a strong foundation of agricultural leading enterprises and e-commerce cooperatives. On the other hand, the S3b pathway suggests that, in the context of supplementary conditions of technological factors and human capital, a rural e-commerce entrepreneurial ecosystem with a robust foundation of e-commerce cooperatives and government attention can achieve high-level development.

Both pathways require the simultaneous and synergistic interaction of technological, organizational, and environmental factors. The core condition in driving outcomes is the creation of e-commerce cooperatives, while digital infrastructure, technological innovation capabilities, and human capital levels are complementary conditions influencing the construction of this ecosystem.

Notable case studies exemplifying this model are concentrated in the central regions and adjacent provinces, including Anhui, Hubei, Henan, Jiangsu, and Shaanxi (as shown in Fig. 4). These provinces, possessing digital resources, innovation capabilities, and e-commerce development levels aligning with the national average, require a solid foundation of rural e-commerce cooperatives alongside a strong human capital base for high-level development. Analysis of the National Farmer Cooperative Demonstration Index reveals that these five provinces consistently rank among the top six nationally in the number of national demonstration cooperatives. Their robust foundational strength, high development vitality, strong entrepreneurial capabilities, numerous registered trademarks, and multiple certifications for the quality of agricultural products help support the establishment of entrepreneurial and innovative zones and promote the development of “Taobao Villages,” propelling the high-level development of rural e-commerce entrepreneurial ecosystems. Taking Anhui as an example, the province, with abundant agricultural resources, has not only played a leading role nationally in forums on smart agriculture, sustainable agricultural development, and the training of the “Three Products and One Standard,” but has also vigorously promoted the improvement and efficiency enhancement of farmer cooperatives. Anhui has established multiple farmer cooperative-operated companies and innovated business models, elevating the construction level of the rural e-commerce entrepreneurial ecosystem.

Fig. 4
figure 4

Explanation case of Segment 3.

Full size image

Configuration of non-high-level construction

Table 5 shows that the fsQCA analysis identified 6 configurations that result in a non-high-level rural e-commerce entrepreneurial ecosystem. The overall coverage rate is 0.92, meaning that these 6 configurations can explain 92% of the cases where the rural e-commerce entrepreneurial ecosystem is not at a high level. Based on the missing conditions in the configurations, the 6 configurations are divided into two types: technology-organizational constraints and organizational-environmental constraints.

In configurations K1a, K1b, K1c, and K1d, the absence of technical and organizational factors plays a core role, categorizing these four configurations as technology-organizational constraint types. Representative cases include 11 provinces such as Shanxi, Inner Mongolia, and Liaoning, which share the common characteristics of poor technical conditions and low urban development vitality. In configurations K2a and K2b, the absence of organizational and environmental factors plays either a core or peripheral role, categorizing these two configurations as organizational-environmental constraint types. Representative cases include 5 provinces such as Beijing, Tianjin, and Shanghai, where, despite good economic development, the rural e-commerce entrepreneurial environment is relatively poor. These results inversely verify the positive role of agricultural leading enterprises and the foundation of e-commerce cooperatives, while further testing the configuration paths that lead to a high-level rural e-commerce entrepreneurial ecosystem.

Robustness analysis

Firstly, the qualitative comparative analysis method, based on set theory, is susceptible to variations in data processing. Therefore, to test the robustness of the results, the consistency threshold was adjusted, and changes in the configurations were observed. The consistency threshold was adjusted from 0.8 to 0.85. As shown in Table 6 (S1’-S3b’), the analysis results were consistent with the original configurations. Secondly, the data used in this study were collected from the years 2020 and 2021, during the COVID-19 pandemic, which may have influenced the economic and technological variables. To account for this, the data for the three variables “digital infrastructure,” “technological innovation capability,” and “level of agricultural development “ were replaced with 2022 data, and a robustness check was designed. All other aspects of the design were kept unchanged, and the fsQCA software was used again to derive the configurations shown in Table 6 (S1”-S3b”). The results were largely consistent with the original configurations. In summary, the configuration analysis results are robust.

Table 6 Robustness Analysis.
Full size table

Predictive validity test

Due to the lack of more recent data, this study could not conduct a traditional predictive validity test. Using cross-validation to assess the predictive validity of the fsQCA model is an effective alternative, especially in the absence of additional samples (Schneider and Wagemann, 2012). To evaluate the predictive capability of the model, this study employed a 5-fold cross-validation method. Specifically, the 31 samples were randomly divided into 5 groups, ensuring each group contained a similar number of samples, with 6–7 samples per group. In each iteration, 4 groups of data were used as the training set to build the fsQCA model, while the remaining 1 group served as the test set to evaluate the model’s performance. This process was repeated 5 times, ensuring each sample was validated once. After each training session, the test set was used to assess the model’s predictive performance, and the consistency and coverage metrics were recorded. Finally, the evaluation results from all test sets were aggregated to calculate the average consistency and coverage, which were then compared with the results from the overall data to indirectly validate the model’s predictive validity.

As shown in Table 7, the cross-validation results showed that the model’s average consistency across different data subsets was 0.945, and the average coverage was 0.704. These results were close to the model performance on the overall data (consistency of 0.94 and coverage of 0.729), indicating that the model’s predictive performance on unseen data is relatively robust.

Table 7 Results of predictive validity test.
Full size table

Discussion

The research findings reveal that configurations of four conditions form three distinct driving pathways, indicating multiple approaches to enhance the level of rural e-commerce entrepreneurial ecosystem development. This validates the “convergence in divergence” outcome from a configurational perspective. Organizational factors were found to be core conditions within all configurations, underscoring the pivotal role of agricultural enterprises and e-commerce cooperatives in shaping the rural e-commerce entrepreneurial ecosystem. This underscores the pivotal role of organizational conditions in driving ecosystem development, aligning with existing literature. Based on these findings, we present the following theoretical and practical implications.

Theoretical implications

This research enriches our understanding of the rural e-commerce entrepreneurship ecosystem and proposes a comprehensive analytical framework for its construction. Rural e-commerce, though offering new entrepreneurial opportunities (Yang et al. 2021), also presents operational challenges (Liu et al. 2021). A well-developed rural e-commerce entrepreneurship ecosystem can mitigate costs and improve efficiency. Metrics for assessing such ecosystems in Chinese rural areas must consider unique contextual factors like regional resources, government support, and e-commerce industry clusters (Mei and Jiang, 2020). Agricultural enterprises and rural cooperatives play pivotal roles, fostering rural entrepreneurial innovation (Gaddefors et al. 2020; Dias et al. 2019).

Using the fsQCA method, this study elucidates key factors and pathways for enhancing the construction of rural e-commerce entrepreneurship ecosystems. Unlike linear approaches (Guo, 2015), our study adopts a configurational perspective, integrating three dimensions (technology, organization, environment) and seven variables. Organizational factors like “agricultural enterprises” and “e-commerce cooperatives” emerge as critical elements in ecosystem construction.

The application of the TOE framework has been expanded within entrepreneurial ecosystems. First, the TOE theoretical framework is typically used in areas such as technological innovation, government construction, and regional innovation (Chen et al. 2022). This study applies the TOE framework to rural e-commerce entrepreneurial ecosystems, based on the “ actor-environment “ perspective. Second, by introducing CAS theory, this study explores the unique manifestations and interactions of technological, organizational, and environmental factors in rural e-commerce entrepreneurship, enriching related research. By integrating the TOE framework with complex adaptive systems theory, new theoretical perspectives and models are proposed. Lastly, the “organization-environment” configurational results, frequently found in literature combining the TOE framework and the fsQCA method, are extended to the rural e-commerce entrepreneurial ecosystem.

Practical implications

Of prime importance is the coordination of the combined effects of technological, organizational, and environmental conditions to enhance the construction level of rural e-commerce entrepreneurial ecosystems. This approach mitigates the risk of overemphasizing single influencing factors and highlights the importance of synergistic interactions among multiple factors.

Additionally, there is a critical focus on the foundational roles of e-commerce cooperatives and leading agricultural enterprises. Establishing a comprehensive and robust rural e-commerce cooperative mechanism and striving to develop leading agricultural enterprises are essential to maximize the impact of these core conditions.

Furthermore, it is imperative to develop tailored construction plans and strategies based on local conditions. The driving paths of rural e-commerce entrepreneurial ecosystems vary across different regions, necessitating the adoption of region-specific development approaches.

Conclusions and future work

Conclusions

The rural e-commerce entrepreneurial ecosystem undergoes a transformation characterized by empowerment, intrinsic dynamism, and holistic coordination, ultimately leading to value co-creation. Achieving this transformation requires harnessing regional advantages and industrial characteristics, advancing the optimization and upgrading of rural industries, and establishing and refining rural entrepreneurial and commercial ecosystems. These measures are critical in propelling the rural e-commerce industry towards the progressive development of a high-value chain. In this study, which encompasses a sample of 31 Chinese provinces, according to CAS theory, we employed the TOE framework and utilized the fsQCA method to investigate the driving pathways that influence the level of development in rural e-commerce entrepreneurial ecosystem construction. Our findings are summarized as follows:

It is crucial to recognize that the advanced construction of the rural e-commerce entrepreneurial ecosystem is not driven by any single condition in isolation. No preceding condition alone can independently initiate its high-level development. The high-level construction of the rural e-commerce entrepreneurial ecosystem is propelled by three distinct configurations of conditions: technology-organization-government-driven, organization-environment-driven, and holistic synergy-driven. A consistent element across all three configurations is the pivotal role of organizational factors. The foundational components of e-commerce cooperatives and the presence of leading agricultural enterprises are of paramount importance. These factors underscore the fundamental significance of these precursor conditions in achieving the advanced construction of the rural e-commerce entrepreneurial ecosystem. The research findings guide policymakers to consider the synergistic effects of various factors and to formulate systematic policies that promote the coordinated development of these elements. Enhancing support for leading agricultural enterprises and fostering the growth of e-commerce cooperatives will strengthen their foundational roles. Additionally, the government should devise region-specific differentiated support policies based on local conditions to ensure that the construction of rural e-commerce entrepreneurial ecosystems is tailored to regional needs.

Limitations and future work

This study is not without its limitations. One potential area for improvement is the measurement of government attention. Due to the lack of standardization in the classification of documents across different provincial government websites, it was not possible to obtain a total count of policy documents, which precluded the use of the percentage method comparing the number of relevant documents to the total number of policy documents. To address this issue, future research could refine and extend the measurement methodology, potentially incorporating factors such as the relative speed at which different local governments respond to the same policy.

This study’s another significant limitation lies in its exclusive reliance on cross-sectional data, neglecting the temporal dimension. Therefore, this approach limits the study’s explanatory power over time. Future research could utilize longitudinal data to obtain more comprehensive and reliable conclusions. Additionally, the study focuses on delineating the driving pathways for the high-level development of rural e-commerce entrepreneurial ecosystems. To address this limitation and explore the concept of “asymmetric causality,” future studies should delve into the mechanisms that facilitate the establishment of low-level rural e-commerce entrepreneurial ecosystems.