Abstract
Drawing on data from the China Family Panel Studies (CFPS), this study examines the impact of Internet usage on migrant workers’ decisions to return home and engage in entrepreneurial activities. The findings indicate that the Internet significantly facilitates entrepreneurial ventures among migrant workers in their hometowns, and the results remain robust after a variety of tests. Mechanism tests reveal that the Internet influences migrant entrepreneurship through the social interaction effect, the financing channel effect, and the human capital effect. Moreover, under the framework of rural revitalization, heterogeneity analysis shows that the Internet’s impact is more pronounced among migrant workers returning home for entrepreneurship with lower agricultural incomes, more severe ecological challenges, lower levels of education, higher government performance evaluations, and wider local prosperity gaps.
Introduction
Migrant workers play a role in rural economies worldwide, as their mobility supports not only urban labor markets but also rural development when they return to their hometowns (De Haan & Rogaly, 2002). In many countries, particularly in developing economies, rural-to-urban labor migration is a mechanism through which migrant workers acquire new skills, accumulate savings, and eventually return home with the potential to start businesses and enhance local economic conditions (Wong & Song, 2008). These movements can impact rural areas by fostering employment creation, knowledge transfer, and entrepreneurial activity, all of which collectively drive local development and reduce rural poverty (Nedelcu & Soysüren, 2022). In nations like India, Mexico, and Nigeria, returning migrant workers are viewed as agents of rural economic transformation, bringing back capital, skills, and innovation to their communities (Dahiya et al., 2023).
In China, the return of migrant workers has a particularly significant effect on rural economic development. Historically characterized by high emigration, rural areas are now witnessing a growing trend of individuals returning home to start businesses. This shift is being driven by rapid urbanization, evolving labor market conditions, and the Chinese government’s strategic focus on rural revitalization. According to the “2022 Human Resources and Social Security Development Statistical Communique” released by China’s Ministry of Human Resources and Social Security, there were 290 million migrant workers in China in 2022, representing a 1.1% increase from the previous year. Notably, among these migrant workers, 12.372 million were local migrants—those employed in the towns where their household registration is located—reflecting a significant growth rate of 2.4%. Meanwhile, 171.9 million were migrant workers employed outside their hometowns, with a growth rate of only 0.1%. This suggests that a growing number of migrant workers are returning to their hometowns to explore new entrepreneurial opportunities.
According to a press conference held by China’s State Council Information Office in Beijing in 2022, over 11 million people had returned to their hometowns to start businesses by the first quarter of that year, with approximately 70% being migrant workers. During the conference, Zeng Yande from the Ministry of Agriculture and Rural Affairs stated that comprehensive calculations by the ministry indicated that, on average, each entrepreneurial migrant worker could create stable employment for 6 to 7 farmers and flexible employment for an additional 17 farmers (http://www.moa.gov.cn/hd/zbft_news/yjdnyncjjyx/). As a result, many migrant workers have not only re-entered the labor market through self-employment, thereby boosting household incomes, but have also contributed significantly to job creation. These efforts align with China’s broader national strategy to reduce the urban-rural divide and promote high-quality economic growth.
In China, many returning migrant workers start businesses in the breeding industry or agriculture-related sectors. This trend is shaped by several factors, including traditional mindsets, the specificity of their skill sets, local resource endowments, and a strong emotional connection to their hometowns. As a result, many—particularly the first generation of migrant workers—tend to pursue entrepreneurship in agriculture. The agricultural ventures initiated by returning migrant workers play a crucial role in stimulating rural consumption, advancing agricultural supply-side structural reforms, and supporting the rural revitalization strategy. Globally, agricultural entrepreneurship by returning migrant workers is also significant. For example, it can contribute to a more diverse and higher-quality global food supply (De Brauw & Rozelle, 2008; Ellis, 2000).
One of the forces behind this phenomenon is the Internet. The widespread adoption of digital technologies has reshaped the landscape for migrant workers, providing them with unprecedented access to resources, markets, and networks. According to the 50th Statistical Report on China’s Internet Development, released by the China Internet Network Information Center (CNNIC), as of June 2022, China had over 1 billion Internet users, 293 million of whom were in rural areas, reflecting significant digital penetration in these regions. This vast number of Internet users has created opportunities for migrant workers to leverage the Internet as they return home to start their businesses.
The Internet has greatly enhanced the ability of returning migrant workers to engage in entrepreneurial activities. First, it enables them to overcome regional constraints while maintaining and strengthening existing social networks as entrepreneurs. Second, the Internet expands access to entrepreneurial information, accelerates the dissemination and diffusion of various types of knowledge, and helps mitigate challenges caused by information asymmetry. In doing so, it can narrow the “digital divide” and “information gap” between urban and rural areas in China. This technological shift aligns closely with China’s rural revitalization strategy, which aims to promote rural economic prosperity through innovation, digital connectivity, and sustainable development.
Although existing literature has examined the impact of the Internet on entrepreneurship (Dahiya et al., 2023; Janson & Wrycza, 1999), studies focusing specifically on returning migrant workers remain relatively scarce, and the findings are often contentious. Some scholars argue that the Internet’s vast reach and “long tail effect” offer significant development opportunities for enterprises, fostering innovation and enabling market expansion (Falk & Hagsten, 2015; T.K. Liu et al., 2013). Conversely, others contend that due to the limited scale and lack of endowment factors, enterprises may struggle to fully utilize the synergistic potential of the Internet (Bai et al., 2020). Nonetheless, the Internet’s potential as a tool for economic empowerment in rural areas is undeniable, particularly within the framework of China’s rural revitalization strategy.
The marginal contribution of this paper is reflected in three main aspects. First, it extends the research boundaries of the Internet’s economic effects by focusing on returning migrant workers as a key research subject. Most current studies have explored the economic implications of the Internet at both macro and micro levels. At the macro level, the Internet promotes inclusive innovation (Paunov & Rollo, 2016), enhances the technological environment (Zhang & Guan, 2017), optimizes agricultural technology (Zheng et al., 2022), facilitates low-carbon transitions (Wang, Dong, et al., 2022; Pan et al., 2023), boosts export trade (F. Zhou et al., 2022), and drives economic growth (Y. Liu et al., 2021; Luo et al., 2023; Wang, Wang, et al., 2022; B. Yu, 2022). At the micro level, it increases employment rates (Hjort & Poulsen, 2019), facilitates knowledge flow within firms (Forman & van Zeebroeck, 2019), raises rural farmers’ incomes (Khan et al., 2022; Leng, 2022), and improves residents’ health (H. Wu et al., 2022). While existing literature has examined the Internet’s impact on entrepreneurship (Dabbous et al., 2023; Dahiya et al., 2023; Janson & Wrycza, 1999), specific attention to returning migrant workers remains limited and underexplored. This paper employs panel data to examine the “long tail effect” of the Internet on migrant workers returning home to start businesses, revealing insights into the microeconomic effects of the Internet.
Second, it offers a digital economic perspective on the entrepreneurship of returning migrant workers. As a significant demographic group, migrant workers have long been the focus of researchers and policymakers (S. Chen & Liu, 2016; H. Fan et al., 2023). Existing literature has discussed factors such as immigrant work experience (Xi et al., 2024), comprehensive medical insurance (Huang et al., 2024), and capital markets (Bossavie et al., 2024) concerning migrant workers or entrepreneurship. There is also literature examining the digital economy’s role in rural entrepreneurship development, such as Dahiya et al. (2023), who explore multimedia technology’s impact on rural entrepreneurship development. Unlike these studies, our paper emphasizes the group of returning migrant workers. Return migration for entrepreneurship is more related to leveraging personal experiences and skills to contribute positively to their hometown. The difference between our paper and this prior research is that the research subjects have different entrepreneurial backgrounds. H. Fan et al. (2023) conducted a study using data from the 2016 China Labor Dynamics Survey (CLDS). In contrast to their only single-year dataset, we construct a multi-year panel dataset, offering a more comprehensive and nationally representative analysis of the entrepreneurial activities of returning migrant workers. For example, our year-grouped regressions reveal that the newer the year, the greater the regression coefficient of the core explanatory variables, indicating that the benefits of the development of the Internet are becoming more and more obvious over time. These findings highlight the Internet’s evolving role in different stages of entrepreneurship for returning migrant workers, thereby broadening the research scope.
Third, this study incorporates the unique context of China’s rural revitalization strategy, an important yet underexplored factor in existing literature. As a key component of China’s national development strategy, rural revitalization encompasses building rural areas with thriving businesses, pleasant living environments, social etiquette and civility, effective governance, and prosperity. However, the literature on promoting these goals through return migration for entrepreneurship is still limited. By analyzing the Internet’s role in returnee entrepreneurship, this paper demonstrates how their activities align with the objectives of rural revitalization. For instance, the Internet’s impact on entrepreneurship can drive rural industry upgrading, enhance ecological conditions, and foster the modernization of rural cultural practices and governance through knowledge dissemination and information sharing, ultimately contributing to common prosperity. This study offers a new perspective on how migrant workers can leverage the Internet to start businesses under the rural revitalization strategy, highlighting the Internet’s internal driving force in this process and providing insights for future policy formulation.
Literature Review and Research Hypothesis
Literature Review
Migrant Workers Returning Home for Entrepreneurship
According to the Monitoring Survey Report of China’s Migrant Workers, China’s migrant workers are defined as individuals registered in rural areas who engage in non-agricultural industries or work outside their hometowns for 6 months or more each year. This social group emerged after China’s reform and opening-up policies, with its roots primarily in the country’s distinctive household registration and land systems. As such, migrant workers are fundamentally an institutional phenomenon (Chan & Buckingham, 2008). The widespread societal concern about this group stems from their large population size and unique socio-economic status. Migrant workers often face significant challenges integrating into urban environments after leaving rural areas, as they are frequently unable to access the same public services as urban residents. This places them in a vulnerable position during China’s economic and social transformation.
In the current complex economic environment, migrant workers face increasingly prominent employment challenges, partly due to a lack of formal education and professional skills. A study by Xie and Zhou (2017) found that 13.26% of migrant workers were unemployed after returning home, underscoring the urgent need for the Chinese government to address the re-employment of migrant workers. One important solution is entrepreneurship. Upon returning to their hometowns, many migrant workers do not solely return to traditional agriculture but also become key participants in the “mass entrepreneurship and innovation” initiative.
At present, scholars have identified four main subjective motivations driving migrant workers returning home for entrepreneurship: (1) the desire to achieve higher economic income; (2) the experience of discrimination or exclusion in urban areas, leading some migrant workers to return to their hometowns to pursue self-worth; (3) a deep sense of connection and belonging to their hometowns, motivating migrant workers to establish businesses locally (Conway & Cohen, 1998); (4) the impact of long-term family separation, which reduces the happiness and security of migrant workers. Returning home to start businesses allows them to reunite with their families, thereby addressing social issues such as “left-behind children” and “empty nesters” as far as possible (X. Yu et al., 2018).
Internet
With the further advancement of the Internet, increasingly detailed data—such as the number of websites, the proportion of Internet users, and the number of broadband ports—have been generated, reflecting the level of Internet development. This has led to a growing body of research examining the relationship between the Internet and socio-economic development. These studies can be categorized into macro and micro levels based on their research subjects and data. At the macro level, research has explored the Internet’s influence on key issues like economic growth (Y. Liu et al., 2021), international trade (Choi et al., 2002), price competition (Chevalier & Goolsbee, 2003), and financial development (Goldfarb & Tucker, 2011). At the micro level, studies focus on the Internet’s impact on individual economic decision-making and welfare, covering topics such as employment choices (Feldman & Klaas, 2002), consumption preferences, and subjective well-being (Lohmann, 2015).
Internet and Entrepreneurship
The Internet and its impact on entrepreneurial activities have received attention from scholars worldwide. Researchers from various countries, including Canada (Cumming & Johan, 2010), the United States (Mack et al., 2011), Germany (Heger et al., 2011), as well as from Africa and Portugal (Branstetter et al., 2014), and China (Kotnik & Stritar, 2015; Yuan & Shi, 2019), have examined the positive influence of the Internet on entrepreneurial activities using different datasets and contextual settings.
The existing literature on the relationship between the Internet and entrepreneurial activities can be broadly categorized into two types based on research paradigm: theoretical analysis and empirical research. In terms of theoretical analysis, Dajani & Marlow (2013) deconstructed entrepreneurial theories and established a research framework, providing a theoretical basis for subsequent studies on the topic. Empirical research has utilized various databases such as the Global Entrepreneurship Monitor (GEM) data (P. Liu et al., 2013), Chinese General Social Survey (CGSS) data (G. Chen, 2015), China Family Panel Studies (CFPS) data (G. Zhou & Fan, 2018), and China Labor Dynamics Survey (CLDS) data (H. Fan et al., 2023).
The existing literature on the relationship between the Internet and entrepreneurship primarily focuses on two key aspects based on research contents: direct effect and indirect effect. Firstly, the direct impact of the Internet’s popularity and application on entrepreneurship is extensively studied. The Internet plays a pivotal role in providing abundant information resources for entrepreneurial and innovative endeavors. It facilitates the identification of entrepreneurial opportunities (X. Yu et al., 2019), expands market reach (Littunen & Virtanen, 2006), mitigates entrepreneurial risks, enhances search and matching efficiency (Stevenson, 2008), reduces startup costs and improves entrepreneurial performance (T. Wu & Wang, 2016), as well as other factors. Secondly, the indirect effect of the Internet’s popularity and application on entrepreneurship is also examined. This effect is manifested in various ways, including the enhancement of human capital among migrant workers (Davidsson & Honig, 2003), the expansion of social capital (Zhang & Zhao, 2015), and the accumulation of physical capital, which ultimately impacts entrepreneurial activities.
In conclusion, although numerous studies have discussed the positive impact of the Internet on entrepreneurship from various perspectives, there remain many unresolved issues in this field that require further exploration.
First, while existing studies have paid attention to the importance of the Internet in entrepreneurship, there is still a gap in the literature focusing on the specific group of returning migrant workers. For instance, although Dahiya et al. (2023) explored the role of multimedia technology in rural entrepreneurship, their research did not focus on the unique entrepreneurial background of returning migrant workers and lacked an in-depth analysis of how the Internet affects these migrant workers. Additionally, prior studies have predominantly remained at the theoretical and conceptual levels (Dajani & Marlow, 2013; Nedelcu & Soysüren, 2022), rarely utilizing objective data to answer how and to what extent the Internet can empower migrant workers returning home for entrepreneurship.
Second, regarding variable measurement, much of the existing literature relies on relatively limited data, making it difficult to assess the specific impact of the Internet on migrant workers returning home for entrepreneurship. For example, H. Fan et al. (2023) based their study on a single year of CLDS data from 2016, which limits the ability to capture long-term trends. In contrast, by leveraging CFPS data, our study uses multi-year panel data to offer a more detailed investigation of the channels, pathways, and effects of the Internet in enabling migrant workers to return home and pursue entrepreneurial activities. This extended data enables us to capture dynamic changes in the process of returning migrant workers to start a business and to reflect the varying roles of the Internet during different periods of this entrepreneurial process.
Third, the implementation of the rural revitalization strategy will profoundly impact rural economic and social development. Entrepreneurial activities, which are deeply influenced by these economic and social shifts, will naturally benefit from this strategy. However, existing literature has not directly examined how to promote the rural revitalization strategy when discussing the relationship between the Internet and migrant workers returning home for entrepreneurship. From the goals of the rural revitalization strategy, we identify five dimensions (industrial development, ecological environment, education level, governance measures, and prosperity gap) for our heterogeneity analysis. This paper also analyzes how the Internet’s role in enabling migrant workers to return home for entrepreneurship can actively contribute to promoting the rural revitalization strategy. This represents a new attempt and an active exploration in the field, adding depth to the existing body of research.
Research Hypothesis
Drawing upon the existing literature, our paper posits that the Internet empowers migrant workers to return home for entrepreneurship. By accelerating the dissemination of information, the Internet brings forth abundant resources that benefit migrant workers in their entrepreneurial endeavors. Firstly, the development of the Internet diminishes the constraints posed by inter-regional spatial and temporal distances, thereby broadening the information channels available to migrant workers returning home for entrepreneurship. Consequently, this reduces the costs associated with information search and aids in overcoming the adverse effects of information friction on entrepreneurial activities. Moreover, Internet technologies’ real-time and interactive nature provides timely and effective information for migrant workers pursuing entrepreneurial ventures. This enables them to stay updated on market conditions and user demands and to identify entrepreneurial opportunities, thereby informing them in making informed and strategic decisions.
It is worth noting that Internet-based information dissemination not only facilitates the identification of entrepreneurial opportunities (Shane, 2000) but also positively impacts the establishment and operation of entrepreneurial enterprises (Coate & Tennyson, 1992). In the past, limited transportation accessibility and information availability in rural areas constrained the sales areas of traditional rural industries. However, with the rapid development and widespread adoption of Internet technology, returning migrant workers now have access to a convenient and cost-effective tool to expand their business operations beyond local boundaries and reach broader markets. Consequently, this paper presents the following testable hypothesis:
Drawing upon prior studies (Fairlie & Krashinsky, 2012; Zhang & Zhao, 2015), we posit that migrant workers returning home for entrepreneurship benefit from Internet empowerment through three key mechanisms: the social interaction effect, the financing channel effect, and the human capital effect.
Firstly, the social interaction effect emphasizes the mutual influence and various outcomes arising from social interactions, which significantly impact entrepreneurial development. This mechanism manifests in two primary aspects. On the one hand, enhanced social interactions foster the accumulation of social capital, a vital prerequisite for entrepreneurs, particularly within China’s rural “relational” communities (Bian, 1997; Chai et al., 2019). On the other hand, social interactions strengthen the demonstration effect of returning migrant workers’ entrepreneurial endeavors, akin to the “role model effect” (Gan & Li, 2022). In this context, the success of these entrepreneurs encourages potential entrepreneurs to pursue similar paths. Social interaction amplifies this effect, making it more likely that others will follow suit. Consequently, the Internet, as a platform facilitating robust social interactions, not only enhances the accumulation of social capital for migrant workers returning home for entrepreneurship but also reinforces the role model effect of entrepreneurial achievements, influencing the decision-making process of returning migrant workers.
Secondly, the financing channel effect mechanism underscores that returning migrant workers need access to resources such as capital, information, and policies during the entrepreneurial process, among which capital is a critical internal factor for entrepreneurial success. Whether obtaining funds from family and friends or institutions like banks, easing financing constraints plays a key role in stimulating entrepreneurship. With the advancement of Internet technology, the Chinese government has implemented the “mass entrepreneurship and innovation” policy, providing financing channels specifically tailored to Internet entrepreneurship. This policy orientation has fueled enthusiasm for Internet entrepreneurship. The rapid development of Internet finance has mitigated the challenges of “difficult financing,”“expensive financing,” and “slow financing” faced by returning migrant workers starting their businesses. Consequently, the Internet has expanded financing channels, ensuring a more substantial capital reserve for migrant workers returning home for entrepreneurship and facilitating the sustainable development of their ventures.
Lastly, the human capital effect mechanism highlights that the Internet accelerates information dissemination, enabling returning migrant workers to acquire new entrepreneurial skills and optimize their existing knowledge structure. This process contributes to enhancing their human capital, which plays a role in entrepreneurial success. Simultaneously, the Internet enhances returning migrant workers’ understanding of new technologies and knowledge, stimulating their subjective initiative and bolstering their overall comprehensive capabilities. Higher levels of human capital among returning migrant workers increase the likelihood of making sound entrepreneurial decisions, thereby improving enterprise performance and innovation capacity during the entrepreneurial process (Davidsson & Honig, 2003). Consequently, this paper presents a second hypothesis for testing:
According to the Rural Revitalization Strategic Plan (2018–2022) issued by China’s Office of the Central Rural Work Leading Group, the general objectives of “to build rural areas with thriving businesses, pleasant living environment, social etiquette and civility, effective governance, and prosperity” may influence the effect of the Internet empowering migrant workers to return to home for entrepreneurship (J. Han, 2020).
Firstly, if returning migrant workers have a lower agricultural income, the Internet’s effect on enabling them to return home for entrepreneurship is greater. This is because returning migrant workers with higher agricultural income typically enjoy a relatively higher standard of living and may lack the motivation and pressure to engage in entrepreneurial endeavors. It is often the returning migrant workers with lower agricultural income who are more affected by the influence of the Internet as they seek to improve their livelihoods and production conditions. This subgroup of migrant workers engaging in entrepreneurship to overcome economic challenges can be characterized as survival entrepreneurship (Mead & Liedholm, 1998). Therefore, the Internet may promote entrepreneurship among returning migrant workers with low agricultural income, facilitate the upgrading of the rural industrial chain, and contribute to the integration of the three rural industries, thereby advancing the goal of “thriving businesses” in rural areas.
Secondly, if serious ecological problems exist in the hometowns of returning migrant workers, the effect of the Internet on their entrepreneurial activities is more significant. Returning migrant workers often possess an attachment and sense of belonging to their hometowns, desiring a beautiful and ecologically sustainable environment in their hometowns. This is because the preservation and enhancement of the natural environment and ecological resources in their hometown are directly linked to their production, livelihoods, and future development (Tan, 2023). The Internet itself has minimal ecological impact, and it can infiltrate traditional industries and mitigate the development pressures on environmentally taxing sectors. Therefore, if returning migrant workers aspire to improve the ecological environment of their hometowns, they are more likely to leverage the Internet to initiate entrepreneurial endeavors aimed at reducing ecological challenges. This highlights how the Internet can facilitate the engagement of returning migrant workers, who are aware of the environmental issues in their hometowns, in digital, green, and low-carbon entrepreneurship, ultimately contributing to the realization of the goal of a “pleasant living environment” in rural areas.
Thirdly, the Internet has a more pronounced effect on enabling migrant workers with lower levels of education to return home for entrepreneurship. According to the diminishing marginal law of human capital accumulation, the Internet exerts a greater impact on the accumulation of human capital among returning migrant workers with limited knowledge and skills (G. Zhou et al., 2017). Compared to individuals with higher education levels, migrant workers with lower levels of education often face challenges in accessing high-quality employment opportunities and entrepreneurial resources. Additionally, their understanding and grasp of new technologies and business models may be limited, necessitating information and support provided by Internet platforms. In this regard, the Internet guides them in understanding market demand and competition, acquiring entrepreneurial skills and knowledge, and accessing resources such as financial support and marketing promotion. The Internet enhances the success rate of their entrepreneurial endeavors. Therefore, the Internet has a more significant impact on returning migrant workers with lower levels of education, contributing to cultural advancement and social development in their hometowns and advancing the goal of “social etiquette and civility.”
Fourthly, if returning migrant workers hold a positive evaluation of the local government, the positive effect of the Internet on their entrepreneurial activities is more pronounced. With the implementation of the rural revitalization strategy, the government’s image plays a pivotal role in the effectiveness of rural governance (S. Fan et al., 2004). On the one hand, the government’s image directly influences the satisfaction and trust levels of returning migrant workers. Higher evaluations of the government by returning migrant workers indicate a greater willingness to engage with the various services and policies provided by the government to encourage return entrepreneurial activities. On the other hand, as the Internet evolves, the government can enhance information transparency and openness through online government services and information disclosure, thereby enhancing its credibility and governance capacity. This, in turn, strengthens the positive evaluation of the government by migrant workers, encouraging their active participation in rural construction and governance and promoting the improvement of rural governance. Additionally, migrant workers who have internalized concepts of health, legal systems, and regulations during their urban employment can serve as role models for other villagers upon their return. This contributes to providing more talent and intellectual support for rural governance, fostering democratization, the rule of law, and standardization in rural governance. Consequently, to some extent, the Internet can enhance the willingness and behavior of migrant workers who hold a positive evaluation of the government to return home and start businesses, further advancing the goal of “effective governance” in rural areas.
Lastly, if migrant workers perceive a significant prosperity gap in their rural areas, the Internet has a more apparent effect on empowering them to start businesses in their hometowns. In rural areas with a pronounced prosperity gap, despite the higher risks associated with entrepreneurship, returning migrant workers face lower opportunity costs. Due to the development lag in these rural areas, there is ample room for expansion and untapped potential. Moreover, influenced by the urban-rural binary structure and their low-end skill levels, some migrant workers find it difficult to secure stable income and development opportunities in urban and town settings, and it is challenging for them to find suitable employment upon returning home. Consequently, they are compelled to pursue entrepreneurial opportunities. Although they encounter various difficulties and challenges in the process, rural market opportunities and development potential are more enticing compared to the intense competition prevalent in urban environments. By starting businesses in their hometowns, returning migrant workers not only benefit themselves in terms of personal development and improved social status but also have the opportunity to improve the new look of their communities, create employment opportunities, foster innovation, and make social contributions. Overall, the Internet promotes the entrepreneurial pursuits of migrant workers who recognize the significant gap between rich and poor in their rural areas, thus advancing the goal of “prosperity” for farmers. As such, this paper proposes a third hypothesis for testing:
Research Design
Data
The microdata utilized in this study are derived from three waves of the China Family Panel Studies (CFPS) conducted in 2014, 2016, and 2018. The CFPS, administered by the China Social Science Survey Center at Peking University, is a biennial survey that examines various aspects of family and individual dynamics. The data collection began in 2010. Owing to the absence of Internet-related data in the 2012 survey and the global repercussions of the COVID-19 epidemic in 2020, the uncertainty and dynamic nature of the epidemic render the data from 2020 potentially incomparable and unstable. To ensure comparability and stability, we opt for panel data derived from household and individual questionnaires across the three survey rounds conducted in 2014, 2016, and 2018.
The CFPS design produced a nationally representative sample (Kong & Wu, 2019). This questionnaire follows a multi-stage stratified sampling approach, with counties/districts selected as the primary sampling units, followed by communities nested within these counties/districts and families nested within the communities (Kong & Wu, 2019). In the initial two stages, sample selection was guided by official administrative divisions. Regional variations in China were accounted for by applying the implicit stratified probability-proportional-to-size (PPS) method, with administrative units and socioeconomic status (SES) serving as key stratification factors. Housing units, in the third stage, were systematically selected, employing random starting points and equal probability, while trained personnel mapped the sampled communities to create frames.
Given that this paper focuses on the impact of the Internet on migrant workers returning home for entrepreneurship and recognizing that entrepreneurial decisions often involve collaboration and risk-sharing among family members, the study examines entrepreneurship at the household level. Specifically, the household’s Internet usage is measured by the Internet usage of the household head, as captured in the personal questionnaire. In the CFPS, the household head is defined as either the “head of household” or the “respondent who is most familiar with the family’s economic conditions.” Additionally, the sample is restricted to households where the head is between 18 and 85 years old. The final dataset comprises 4,686 observations across three survey rounds, covering 27 provinces and 231 counties (including districts and county-level cities). To address missing values when taking the natural logarithm, the paper adds 1 to the logarithm of continuous variables before taking their logarithm.
Empirical Model and Variables
Considering that the entrepreneurial decision of returning migrant workers is represented by a binary dummy variable, we adopt the Probit model, referring to previous research on entrepreneurial choice (Hu et al., 2022; Minns & Rizov, 2005; Radicic & Petković, 2023), to examine the influence of the Internet on migrant workers who return home for entrepreneurship.
Where Entrepreneurshipijt represents whether i family in province j started a business in year t. Internetijt is the measurement variable of the Internet. Xijt is a characteristic variable used to describe the household and head level. To control the influence of the entrepreneurial environment and time trend on the model, the province dummy variable θj and year dummy variable δt are introduced. In the empirical study, the clustering standard error at the household level is used to estimate.
Variable Description
The explained variable in this study is a binary dummy variable indicating whether migrant workers return home for entrepreneurship. To identify “migrant workers” and “returning migrant workers” in the sample, we consider rural households (with agricultural hukou) that have at least 1 year of work experience outside their township during the survey period (2014–2018) as “migrant workers.” Among these migrant workers, those who returned home during a given year within the survey period are defined as “returning migrant workers.” In the questionnaire, the question that defines entrepreneurship is as follows: “Did any member of your family engage in self-employment or start a private enterprise in the past year?” A response of “Yes” is assigned a value of 1; otherwise, it is assigned a value of 0. Table 1 shows that 5.2% of the families in the sample had returning migrant workers who started businesses.
Descriptive Statistics for the Variables.
The core explanatory variable in this study is also a binary dummy variable indicating Internet usage. The questionnaire includes questions on “Whether to access the Internet.” If a respondent has access to the Internet, the value for the Internet variable is defined as 1; otherwise, it is assigned a value of 0.
Drawing on the study of G. Zhou and Fan (2018) and Yuan and Shi (2019), we incorporate relevant control variables at the individual, household, and regional levels. At the individual level, the control variables encompass age, age2, gender, marriage, health condition, religion, education, and appearance. Additionally, variable such as intelligence level is further included in the robustness tests. At the household level, this study controls for variables such as household size, net household income, total household property, major event impact, and medical risk impact. The questionnaire includes a question on major events, such as weddings, funerals, births, and children’s school admissions, which occurred within the past 12 months. Medical risk impact is identified when medical expenses exceed 10% of the household’s net income. Furthermore, in the robustness tests, additional variables such as the logarithm of household total cash and deposits and financial product holdings are included. At the regional level, the control variables encompass provincial-level factors such as business climate, GDP, and urbanization rate. The variables are defined in Table A1 in the Appendix.
The descriptive statistics are presented in Table 1.
Empirical Results
Baseline Results
Table 2 presents the regression results using Equation (1) to examine the impact of the Internet on migrant workers returning home for entrepreneurship. Columns (1) to (4) gradually add Control variables at the individual level, household level, and regional level. Upon comparing the estimation results, it is observed that as control variables are gradually added, the direction and significance level of the core explanatory variables remain consistent, indicating the robustness of the estimation results. To provide a clearer explanation, we interpret the economic implications based on column (4). The results in this column indicate that holding other variables constant, the probability of returning migrant workers starting a business increases by 3.61 percentage points compared to rural residents who do not use the Internet. This finding has economic significance and supports Hypothesis 1. The Internet facilitates entrepreneurship among returning migrant workers by accelerating information dissemination, reducing search costs, and aiding in the identification of entrepreneurial opportunities and strategic planning. Through the Internet, returning entrepreneurs can overcome time and space limitations, access broader market information, and support the establishment and operation of their entrepreneurial ventures.
Baseline Results.
Note. *, **, and *** denote significance at the 10%, 5%, and 1% statistical levels, respectively. Same as below. The Probit model reports average marginal effects and Pseudo R2.
Regarding individual-level control variables, the relationship between age and migrant workers returning home for entrepreneurship follows an inverted U-shaped pattern. This means that as age increases, the probability of migrant workers returning home for entrepreneurship initially rises and then gradually declines. Gender is found to be insignificant in this context, indicating that being male does not have a positive impact on the probability of migrant workers returning home for entrepreneurship. While previous literature has suggested that men are more likely than women to engage in entrepreneurship upon returning home (H. Fan et al., 2023), our study comes to a different conclusion. A possible explanation is that H. Fan et al.’s (2023) research is based on 2016 CLDS data, while our analysis covers data from 2014 to 2018. Over time, women’s autonomy consciousness and entrepreneurial aspirations have likely strengthened, and the current Internet-enabled business environment no longer favors men excessively, which diminishes the gender gap. Therefore, the gender variable is not significant in this paper. Additionally, factors such as marriage, education, health condition, religious beliefs, and physical appearance all have a significant positive impact on the probability of migrant workers returning home for entrepreneurship. Overall, the estimated results for these control variables align with the main findings of existing literature.
Regarding household-level control variables, household size has a positive impact on the probability of migrant workers returning home for entrepreneurship. The larger the household, the higher the probability of entrepreneurship. Conversely, net household income has a slightly negative impact on entrepreneurial activities. This is because household income can have a dual effect: on the one hand, higher income provides the financial capital necessary for entrepreneurship; on the other hand, it raises the opportunity cost of starting a business. In this case, the opportunity cost appears to be the more dominant factor (G. Zhou & Fan, 2018). Furthermore, major event impact and medical risks impact did not significantly encourage entrepreneurship among returning migrant workers. This suggests that, in the face of such risks, they are less willing to undertake the uncertainties of starting a business and show little enthusiasm for entrepreneurial ventures.
Robustness Tests
Add Control Variable
The decision to use the Internet and to start a business are independent choices made by migrant workers based on their personal preferences. These decisions are often influenced by individual factors such as intelligence level and the amount of capital available. However, in previous literature (H. Fan et al., 2023), variables like intelligence level and individual perception are challenging to measure directly, leading to potential issues with omitted variables. To strengthen the robustness of the estimation results and mitigate the impact of possible omitted variable bias in this study, we introduced additional control variables. These include individual-level factors such as intelligence and household-level factors like total cash and deposits holding and financial product holdings. The results presented in columns (1) and (2) of Table 3 indicate that the coefficients of core explanatory variables remain significantly positive, further confirming the robustness of the findings.
Robustness Test Results I: Add Control Variable.
The coefficient of the newly added individual-level variable is negative, suggesting that a higher intelligence level among returning migrant workers is associated with a lower probability of engaging in entrepreneurship. A plausible explanation is that, compared to individuals with lower intelligence levels, those with higher intelligence are more likely to receive better education and training, have access to greater resources, and encounter more favorable employment opportunities in urban areas. As a result, they are more inclined to secure stable, well-paying jobs and are less willing or motivated to pursue uncertain entrepreneurial activities.
The coefficient of the newly added household-level variable is significantly positive, indicating that greater total household cash and deposits and financial product holding are associated with a higher probability of returning migrant workers starting a business, which aligns with practical expectations. Compared to cash-constrained migrant workers, those with more household cash, deposits, or financial products are better positioned to secure start-up capital. Additionally, these individuals tend to possess greater skills in capital planning and management, enabling them to allocate resources more effectively, thus increasing the likelihood of entrepreneurial success.
To ensure the stability and accuracy of the regression results, this paper continues to include the control variables used in the basic regression model, consistent with previous literature.
Replace the Core Explanatory Variable and Grouping Regression
The study conducted a robustness test by substituting the core explanatory variable, replacing Internet usage with the number of rural broadband access users. The regression results, presented in column (1) of Table 4, show that the estimated coefficients and significance levels remain consistent with the previous findings. This indicates that the research results are robust, as altering the core explanatory variable does not change the conclusion that Internet access facilitates migrant workers’ return to their hometowns for entrepreneurship.
Robustness Test Results II: Replace Internet Variable and Grouping Regression.
To mitigate the influence of possible discontinuities between different survey years, this paper conducted group regression by year, and the results were displayed in columns (2) to (4) of Table 4. The findings demonstrate that the estimates for each year closely align with the baseline results, confirming the robustness of the conclusions. In addition, it should be noted that the newer the year, the greater the regression coefficient of the core explanatory variables, indicating that the benefits of the development of the Internet are becoming more and more obvious over time.
Reestimation of the PSM Model
Due to the limitations imposed by data availability and variables and the potential non-randomness of migrant workers’ Internet usage, concerns regarding sample selection bias may arise during the research process. The propensity score matching (PSM) method allows for the selection of multiple variables to calculate the propensity score and mitigate the influence of confounding factors through targeted matching, thus reducing the issue of sample selection bias. Therefore, we employ PSM to construct a counterfactual framework to further examine the robustness of the Internet’s role in enabling migrant workers to return home for entrepreneurship.
Table 5 presents the PSM results obtained using different matching methods, such as nearest neighbor matching, radius matching, kernel matching, local linear regression matching, and Mahalanobis metric matching. The average treatment effect on the treated (ATT) ranges from 0.0347 to 0.0431, indicating that regardless of the PSM method employed, Internet use increases the probability of returning migrant workers starting businesses by 3.47% to 4.31%. Overall, the findings in Table 5 align with those in Table 2, demonstrating that even after considering the issue of sample selection bias through PSM, the Internet still exerts a significant positive impact on migrant workers’ entrepreneurship, affirming the robustness of the regression results.
Robustness Test Results III: Reestimation of the PSM Model.
Reestimation of Instrumental Variable Methods
The preceding research results provide evidence that the Internet enables migrant workers to return home for entrepreneurship. However, potential reverse causality should be considered. First, migrant workers who intend to start a business or who are already engaged in entrepreneurial activities may be more likely to use the Internet due to business needs rather than being motivated to start a business because of Internet access. Second, although China promotes network reforms such as “speed acceleration and fee reduction,” the cost of network services and Internet-related equipment like mobile phones and computers remains high. Consequently, farmers who have attained a certain economic capability, such as migrant workers returning home to start businesses, may have the financial capacity and willingness to purchase Internet equipment.
To improve the reliability and effectiveness of the study, we apply the instrumental variable method to conduct a robustness test. Drawing on the approach utilized by Zhao (2019), we employ three instrumental variables to represent the usage of the Internet by returning migrant workers. The instrumental variables used in this study include “mobile phone usage,”“provincial Internet penetration rate,” and “average Internet usage among interviewees.” These instrumental variables are carefully examined in terms of their correlation and exogeneity and are found to effectively represent the utilization of the Internet by returning migrant workers.
Specifically, at the correlation level, mobile phone usage is considered a precondition for Internet usage, exhibiting a strong correlation between the two variables. Additionally, the provincial Internet penetration rate and the average Internet usage among respondents reflect the overall development level of the Internet in a given region. Higher levels of Internet development in a region are associated with a greater likelihood of returning migrant workers in that area utilizing the Internet.
At the exogeneity level, there is no direct correlation between the entrepreneurial decisions of returning migrant workers and whether they use mobile phones, nor is it likely that returning migrant workers’ entrepreneurial activities would directly influence the development level of regional Internet infrastructure.
Accordingly, Table 6 presents regression results using the instrumental variables of “mobile phone usage,”“provincial Internet penetration rate,” and “average Internet usage among interviewees.” To verify the validity of the instrumental variables, we conducted a weak instrument test based on the Stock and Yogo (2005) threshold table. The F-statistic value obtained is 237.26, which exceeds the critical threshold, indicating that the selected instrumental variables do not suffer from weak instrument issues. Additionally, the p-value of the Hansen J statistic is 0.4918, suggesting no evidence of over-identification for the selected instrumental variables. Thus, the chosen instruments for Internet usage are appropriate.
Robustness Test Results IV: Reestimation of Instrumental Variable Method.
Instrumental variables (IVs) are crucial in estimation methods such as two-stage least squares (2SLS), limited information maximum likelihood (LIML), and the generalized method of moments (GMM). In this paper, we apply several IV estimation techniques, including 2SLS, LIML—known for its robustness against weak instruments—and GMM, which is efficient in the presence of heteroscedasticity, as well as iterative GMM estimation. Across all methods, the results consistently indicate that Internet access has a significant positive effect on the probability of migrant workers returning home to start businesses, with regression coefficients remaining stable at the 10% significance level. This further supports the conclusion that Internet access facilitates entrepreneurship among returning migrant workers.
Mechanism Teats
Social Interaction Effect
This paper aims to explore the impact of the Internet on migrant workers returning home for entrepreneurship, with an emphasis on the mechanism of the social interaction effect. The development of social interactions not only accelerates the accumulation of social capital but also amplifies the demonstration effect of migrant workers’ entrepreneurial activities. Therefore, the Internet, with its strong social interaction capabilities, not only facilitates the accumulation of social capital for returning migrant workers but also reinforces the influence of entrepreneurial role models, thus exerting a significant impact on their decision to start businesses.
To measure the influence of the Internet on the social interaction effect, this study refers to the work of Chai et al. (2019), which utilizes the logarithmic value of “expenditure on favors” as a proxy for the social interaction effect. The regression results are presented in Table 7. Control variables are gradually incorporated in columns (1) to (3). The results in column (3) indicate that migrant workers who use the Internet have a 46.6% higher “expenditure on favors,” suggesting that the Internet enhances the social interaction effect within the families of returning migrant workers, thereby promoting their entrepreneurial decision-making process. Hypothesis 2 is partially supported. In China’s relational society, the Internet plays a key role in the social interaction mechanism, significantly influencing the entrepreneurial decisions of returning migrant workers by enhancing the aggregation of social capital and the demonstration effect of entrepreneurship.
Mechanism Analysis: Social Interaction Effect.
Note. In this table, since the dependent variable “expenditure on favors” is a continuous variable, OLS regression was applied for the analysis.
Financing Channel Effect
Capital is recognized as a crucial factor for entrepreneurial success (Fairlie & Krashinsky, 2012). With the rapid development of Internet finance in recent years, it plays a significant role in addressing the challenges of “difficult financing,”“expensive financing,” and “slow financing” faced by migrant workers returning home for entrepreneurship. Internet finance provides affordable and convenient financial support for their entrepreneurial endeavors.
To examine whether the Internet facilitates entrepreneurship among returning migrant workers by alleviating financing difficulties, the Probit model is used for further analysis in Table 8. The dependent variable in columns (1) to (3) is whether the household of a returning migrant worker has outstanding bank loans. The result in column (3) indicates that the Internet increases the probability of returning migrant workers having outstanding bank loans by 2.51 percentage points. This supports part of Hypothesis 2, demonstrating that the Internet enhances the likelihood of returning migrant workers obtaining loans from financial institutions, thereby increasing their probability of starting a business.
Mechanism Analysis: Financing Channel Effect.
Human Capital Effect
The human capital effect of the Internet on returning migrant workers can be attributed to its facilitation of access to new knowledge and technologies, as well as its enhancement of interest and motivation in learning, leading to an improvement in their human capital. Taking into account data availability, this paper employs “whether to read books” and “expenditure on culture, education and entertainment” as indicators to gauge the human capital effect to verify whether the Internet will affect the entrepreneurial decisions of returning migrant workers through the human capital effect. This is because the reading habits and expenditure on cultural, educational, and entertainment activities among returning migrant workers can serve as rough indicators of their educational level. The process of receiving education can enhance their knowledge, skills, and cognitive abilities, thereby enhancing their human capital and increasing the competitiveness of rural migrant workers in starting businesses back home.
In Table 9, the dependent variable in columns (1) to (3) is whether the individual reads books. In columns (4) to (6), the dependent variable is the logarithm of expenditure on culture, education, and entertainment. The results in column (3) indicate that Internet use increases the likelihood of reading behavior in returning migrant workers by 13.47%, while column (6) shows that Internet use increases their expenditure on culture, education, and entertainment by 69.78%. The finding confirms that the Internet enhances the human capital effect in returnee migrant workers’ families, thereby promoting entrepreneurial decision-making. So far, Hypothesis 2 is fully supported.
Mechanism Analysis: Human Capital Effect.
Note. The dependent variable “expenditure on culture, education, and entertainment” in columns (4) to (6) is a continuous variable, and we utilized OLS for regression analysis.
Heterogeneity Analysis Based on the Background of Rural Revitalization
Thriving Businesses
In general, a high agricultural income for the households of returning migrant workers indicates a substantial endowment of agricultural resources, enabling them to maintain a high standard of living through agriculture without the need to engage in uncertain entrepreneurial activities. Conversely, it is often the returning migrant workers with low agricultural incomes who are more susceptible to the influence of the Internet and aspire to change their living circumstances through entrepreneurship. To investigate the heterogeneity among rural migrant worker families resulting from different agricultural incomes in the context of Internet application, this study groups the migrant worker household sample based on their responses to the question: “What is the total value of your farm and sideline products (in yuan per year)?” If the total value of agricultural and sideline products reported by returning migrant workers is lower than the sample mean, their families are categorized into the lower agricultural income group; otherwise, they are classified into the higher agricultural income group.
Column (1) of Table 10 presents the estimated results for the lower agricultural income group, while column (2) displays the estimated results for the higher agricultural income group. By comparing the results of the two groups, it is observed that the Internet has a more pronounced impact on returning migrant workers with lower agricultural incomes, while its impact on returning migrant workers with higher agricultural incomes is not significant. This finding suggests that the Internet has a stronger positive effect on returning migrant workers with lower agricultural incomes, thereby increasing their likelihood of engaging in entrepreneurship. Consequently, this provides support for the partial verification of Hypothesis 3. This result substantiates that the Internet can indeed promote the entrepreneurship of rural migrant workers with lower agricultural incomes, facilitating the optimization of the industrial structure and fostering the integrated development of the three industries, thus contributing to the realization of the goal of rural “thriving businesses.”
Heterogeneity Analysis: Agricultural Incomes.
Pleasant Living Environment
Given the advancements in information technology and the widespread adoption of Internet technologies, the significance that returning migrant workers place on the ecological environment in their hometown is not only an emotional identification but also a responsibility and contemporary mission. Recognizing the value that returning migrant workers attribute to the ecological environment in their hometowns. This study posits that they are more likely to access the Internet and endeavor to drive industrial transformation and upgrading through entrepreneurship, thereby enhancing the ecological environment of their hometowns. To examine the heterogeneous effects resulting from returning migrant workers’ divergent perspectives on ecological issues when utilizing the Internet, this study groups the samples based on their level of concern for the ecological environment in their hometowns.
Column (1) in Table 11 presents the regression results for the group of returning migrant workers from areas with severe ecological problems (scores > 5), while column (2) reports the regression results for the group with less severe ecological problems. The regression findings indicate that the impact of the Internet on returning migrant workers within the group deeply concerned about ecological problems in their hometowns is significantly positive, whereas the impact on the other group is not statistically significant. These results suggest that the Internet can indeed facilitate entrepreneurship among migrant workers from areas with severe ecological problems. Thus, this partially supports Hypothesis 3. The findings demonstrate that the Internet can contribute to improving the rural ecological environment through digital, green, and low-carbon entrepreneurial models, thereby promoting the goal of a “pleasant living environment” in rural areas.
Heterogeneity Analysis: Ecological Problems.
Social Etiquette and Civility
Social etiquette and civility constitute a vital component of the requirements outlined in the rural revitalization strategy. If rural residents possess higher levels of education overall, the establishment of social etiquette and civility will progress more smoothly. However, it is important to note that returning migrant workers with lower levels of education often face greater employment pressures and encounter more challenges when starting their businesses compared to their more educated counterparts. Consequently, they need to leverage the Internet to access a broader range of entrepreneurial opportunities and strategies, enabling them to adapt more flexibly to consumer demands and market fluctuations. To investigate the heterogeneity of the Internet’s impact based on different education levels, we group the samples according to years of education and examine the effect of the Internet on the entrepreneurship of returning migrant workers with varying educational backgrounds.
Column (1) of Table 12 reports the estimates for the lower education group (with years of education less than or equal to 9 years), while column (2) presents the estimates for the higher education group (with years of education greater than 9 years). The comparison between the two groups reveals that the Internet has a greater impact on returning migrant workers with lower levels of education, whereas its effect on those with higher levels of education is not significant. This indicates that the Internet plays a more substantial role in empowering returning migrant workers with limited educational backgrounds, ultimately enhancing their entrepreneurship. Therefore, the Internet indeed promotes entrepreneurship among returning migrant workers with lower educational attainment, partially supporting Hypothesis 3. The Internet contributes to achieving “social etiquette and civility” by improving the overall quality of rural residents and by preserving and promoting local culture.
Heterogeneity Analysis: Education.
Effective Governance
Both government image and the Internet play indispensable roles in achieving the objective of “effective governance” within the rural revitalization strategy. The Internet can contribute to enhancing rural governance by encouraging returning migrant workers who hold favorable opinions of the government to embark on entrepreneurial ventures in their hometowns. To explore the heterogeneity resulting from the combined effect of the government’s image and the application of the Internet, the sample is categorized based on the evaluations provided by returning migrant workers regarding the government (scores: 1–5).
Column (1) of Table 13 presents the estimates for the group with higher government evaluations (scores greater than or equal to 3), while column (2) reports the estimates for the group with lower government evaluations (scores less than 3). Comparing the two groups, it is evident that the Internet has a significantly positive effect on returning migrant workers who give higher evaluations to the government. This indicates that the Internet can indeed promote entrepreneurship among those returning migrant workers who hold the government in high regard, thereby partially supporting Hypothesis 3. The results suggest that the Internet helps maintain harmony and stability in rural society by enhancing the government’s positive image and credibility. Additionally, it contributes to improving health awareness and the rule of law consciousness among returning migrant workers, facilitating the realization of the goal of “effective governance.”
Heterogeneity Analysis: Government Evaluation.
Prosperity
“Prosperity” constitutes the foundational objective of implementing the rural revitalization strategy. Nonetheless, socioeconomic disparities persist in China, with noticeable gaps between the rich and the poor. Migrant workers who perceive these disparities as severe are particularly motivated to capitalize on entrepreneurial opportunities by the Internet. By returning to their hometowns to establish businesses, they aim to expand their income sources and uplift their household earnings, thereby coming closer to the goal of “prosperity.” To explore the heterogeneous effects of the Internet on the entrepreneurship of returning migrant workers with divergent perceptions of the prosperity gap, the sample is stratified based on the distinct perceptions of the prosperity gap among returning migrant workers.
Column (1) of Table 14 presents the results for those who perceive the prosperity gap to be severe (scores greater than 5), while column (2) reports the results for those who believe the prosperity gap is not severe. By comparing the two groups, it becomes clear that the Internet has a more substantial impact on returning migrant workers who perceive the prosperity gap as severe, indicating that the Internet significantly enhances their entrepreneurial opportunities. These findings confirm that the Internet encourages migrant workers, especially those concerned about income inequality, to return home for entrepreneurship. Thus, Hypothesis 3 is further supported. The Internet serves as a tool for migrant workers to increase their income, improve the quality of life for rural residents, and promote the integration of urban and rural development, thereby supporting the achievement of the goal of improving “prosperity” for farmers. At this stage, Hypothesis 3 is fully supported.
Heterogeneity Analysis: Prosperity Gap.
Conclusions
This paper utilizes data from the China Family Panel Studies (CFPS) to examine the influence of the Internet on the entrepreneurial decisions of returning migrant workers. The findings can be summarized as follows: (1) The Internet has beneficial effects on migrant workers returning home for entrepreneurship. This conclusion remains robust after a series of tests. (2) The mechanisms through which the Internet impacts the decision of migrant workers to return home for entrepreneurship include the social interaction effect, financing channel effect, and human capital effect. (3) Based on the background of rural revitalization, heterogeneity analysis reveals that the influence of the Internet is more pronounced for migrant workers returning home for entrepreneurship with lower agricultural income, more severe ecological problems in their hometowns, lower education levels, higher government evaluations, and a significant prosperity gap in their hometowns. Based on these findings, the following policy recommendations could be drawn from the study.
Technology Extension and Infrastructure Development
Government departments should continue to strengthen and expand Internet infrastructure, particularly in rural areas, to lower the costs of Internet use during the entrepreneurial activities of returning migrant workers. Enhancing their ability to utilize Internet technology is crucial. In tandem with the social interaction effect, the government should encourage migrant workers to use Internet platforms to build social capital, promoting resource sharing and information exchange during their entrepreneurial journeys.
Industry and Ecological Environment
The government should foster the diversification of local industries, especially in regions facing serious ecological challenges, by integrating environmental protection technologies with industry development. Returning migrant workers should receive policy support to develop sustainable industrial models that align with ecological protection, meeting environmental standards while enhancing product competitiveness and corporate image.
Education and Skills Training
As the Internet has a more pronounced effect on supporting entrepreneurship among groups with lower education levels, the government should offer more targeted skills training tailored to local economic and industrial needs. Training programs should include Internet applications, entrepreneurial management, and industry insights. Inviting mentors, experts, and successful entrepreneurs to share their experiences may boost the entrepreneurial skills of migrant workers.
Financial Support and Policy Guidance
The Internet plays a vital role in expanding financing channels for migrant workers. To further support this, the government should bolster financial policies, establish dedicated funds, optimize loan schemes, and reduce interest rates for start-up loans to ensure that returning migrant workers receive the necessary financial backing. Moreover, the government should encourage private investment and social capital to support returnee business ventures, leveraging Internet financing channels to help migrant workers achieve their entrepreneurial ambitions.
Leveraging Social Interaction and Demonstration Effects
Given the Internet’s significant role in fostering social interaction, the government should highlight successful cases of migrant workers returning home to start businesses to create a demonstration effect, encouraging others to follow suit. Additionally, through policy guidance, fostering entrepreneurial interaction among migrant workers will enable resource sharing via Internet platforms, thereby driving the high-quality development of rural economies.
This study provides a theoretical analysis and quantitative evaluation of the effect of Internet use on migrant workers returning home for entrepreneurship, offering valuable insights for promoting Internet development and fostering rural revitalization through returnee entrepreneurship. Although the China Family Panel Studies (CFPS) data used in this paper are highly representative, they offer limited detail regarding the entrepreneurial behaviors of migrant workers. This limitation hinders a deeper investigation into the Internet’s influence on specific entrepreneurial types or the overall quality of entrepreneurship among returning migrant workers. Additionally, this study focuses on whether or not the Internet is used but overlooks the diversity of different Internet platforms and applications—such as e-commerce platforms, social media, and financial technology—that may have varying effects on returning migrant workers engaging in entrepreneurship. Future research may aim to explore how different Internet applications (e.g., e-commerce, social media, online payments) affect the entrepreneurial quality of returning migrant workers by tracking aspects like business survival, growth, and performance. This would enable a more comprehensive understanding of how diverse digital tools shape returnee entrepreneurship.
Footnotes
Appendix
Variable Definitions.
| Variable | Definition |
|---|---|
| Entrepreneurship | Whether migrant workers return home for entrepreneurship: yes = 1, no = 0 |
| Internet | Internet access: yes = 1, no = 0 |
| Age | Age of household head (years) |
| Age2 | Age squared divided by 100 |
| Gender | Male = 1, female = 0 |
| Marriage | Have a partner: yes = 1, no = 0 |
| Health condition | Unhealthy = 1, average = 2, relatively healthy = 3, good healthy = 4, very good healthy = 5 |
| Religion | Religious beliefs: yes = 1, no = 0 |
| Education | Years of education (years) |
| Appearance | There are seven points: one is very bad, and seven is very good. |
| Household size | Number of people eating at the same household |
| Net household income | Total annual household net income (yuan) |
| Total household property | Total household property (yuan) |
| Major event impact | Whether have a major event impact: yes = 1, no = 0 |
| Medical risk impact | Whether have a medical risk impact: yes = 1, no = 0 |
| Business climate | The proportion of workers employed by self-employed and private Enterprises in the total employed population of district and county (%) |
| GDP | Gross domestic product of provinces (100 million yuan) |
| Urbanization rate | Proportion of permanent urban population in total population by province at year-end (%) |
Note. This table describes the measurement of the variables, including the explained, core explanatory, and control variables.
Author Contributions
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the General Project of Humanities and Social Sciences Research, Ministry of Education of China (21YJA710043).
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
The original contributions presented in the study are included in the article, further inquiries can be directed to the corresponding author.