The Influence of Family Social Status on Farmer Entrepreneurship: Empirical Analysis Based on Thousand Villages Survey in China

Abstract

Researchers in the fields of psychology and sociology have demonstrated the profound influence of social status on people’s behavior. Although existing studies show that social status matters, scholars have devoted little attention to how family social status affects individuals’ risk-taking entrepreneurial behavior. In Chinese rural areas, where the idea of “family” is deeply embedded, how family social status affects farmers’ entrepreneurial behavior is still an unsolved question. In this paper, we analyze the impact of family social status on farmer entrepreneurship and investigate the moderating effects of external institutional factors, such as regional policy uncertainty and local family culture. Our findings show that family social status has an important impact on farmers’ risk preferences, therefore affecting their motivation to undertake entrepreneurial behaviors. By inspecting the role of social status at the family level, our study offers important implications for social class and entrepreneurial theorizing.

1. Introduction

Farmer entrepreneurship has been regarded as a critical force of economic growth in rural areas as it improves farm productivity [1]. Compared with nonagricultural areas, the development of the agricultural economy is in a relatively slow stage restricted by many factors, such as relatively blocked information, narrow financing channels, poor infrastructure, and a lack of human capital and professional knowledge. Being different from non-farmer entrepreneurship, farmers operate in a tightly constrained, complex, and multi-faceted rural environment, which acts as a significant barrier to entrepreneurial activity [2]. Besides, farmers do not have high-level human capital; they often lack sufficient management skills [3] and entrepreneurial spirit [4], so they may not take full advantage of economic opportunities to improve their entrepreneurial performance [5]. What is more, farmer entrepreneurs always have limited access to business support [6,7], and their main support is more likely to be from family networks. Strong family orientation causes farmers to utilize family members as their trusted advisors, and family may be determinant regarding farmer decision-making, such as entrepreneurial behavior [8]. Researchers have paid significant attention to the driving forces of rural entrepreneurship. Some have focused on individual characteristics, such as personality traits [9,10], learning ability [11], age [12], education level [13], marital status [14], effort, willpower, and courage [15], while others have paid attention to family-level characteristics, such as family resource endowment [16] and family structure [17,18]. Many scholars have shown the importance of the external environment, including factors such as the roles of microcredit support [19,20], the policy environment [21,22], institutional quality [23], and infrastructure [24].

Although previous studies have helped to explore the deep-rooted motivations of farmer entrepreneurship from different perspectives, few scholars have researched the effect of social class on farmers’ entrepreneurial activities [25]. Studies in the fields of psychology and sociology have shown that social class is one of the most meaningful cultural dimensions for human life [26,27] and can have a profound impact not only on people’s judgment and decision strategies [28,29] but also on personal risk preferences [30]. Risk preference can be defined as people’s orientation and willingness to take risks [31,32], which reflects people’s psychological attitude in the face of uncertainty [33]. Generally, risk preference people tend to be more adventurous and pay more attention to the opportunities of risky choice [34,35,36]. Entrepreneurial behavior can be risky because new ventures tend to suffer a high rate of failure [37], so potential entrepreneurs have to accept different kinds of risks while starting new businesses [38]. The heterogeneity of social class reflects people’s differences in material resources, cognitive tendencies [28,39], and habitus [29]. As a high-risk behavior, farmer entrepreneurship also has an inseparable correlation with ranking in a given social group. People from different social classes have different personal characteristics, family assets, and action modes [40], and their diverse class experiences are associated with various levels of access to resources, which has significant impacts on their risk preferences and entrepreneurial behaviors [30]. In Chinese rural areas, traditional culture and elite governance are popular, and informal authority has become a dominant force in the rural social structure, playing a decisive role in farmers’ access to scarce village resources. In the current transition period, China is experiencing a shift from a centrally planned economy to a more market-oriented economy [41]. Even though this transition may be reflected in increased privatization and institutional reforms, the framework of laws, regulations, and property rights protections has been lacking, and they are not well-defined [42,43]. In particular, the transition period is characterized by high uncertainty and institutional inconsistency [44,45,46], so China’s current social and economic foundations are still weak, and the market rules are not perfect, making informal social classes quite crucial factors in farmers’ entrepreneurial activities [47]. Meanwhile, considering the deep idea of family embedded in Chinese rural areas, farmers rely heavily on family and show stronger family dependencies in their entrepreneurial processes compared with nonagricultural groups. Therefore, family social status is an important factor that restricts farmers’ entrepreneurial motivations; however, few researchers have focused on this issue.

By integrating literature on social class theory and entrepreneurship theory, we theorize that higher family social status increases the likelihood of farmer entrepreneurship occurring because of its positive effects on resource availability and psychology. Farmers owned by high social status families are perceived to not only hold a high position in terms of economic class [26] but also have wide access to resources. Additionally, farmers from families with high social status are more likely to have a sense of psychological safety, giving them more optimistic attitudes towards entrepreneurship and a high tolerance to the risks of entrepreneurship [48].

This study presents a farmer entrepreneurship model from the perspective of family social status and explores the complex role of family social status on farmers’ risky entrepreneurial behavior. Based on data from the Thousand Villages Survey in China conducted by Shanghai University of Finance and Economics (SUFE), this study shows that peasant households with a high social class are more likely to have positive resource and psychological effects, thus stimulating farmers’ entrepreneurial willingness. The positive effect of family social status on farmer entrepreneurship is moderated by the external institutional environment. To be specific, formal institutional aspects, such as regional policy uncertainty, will impact the relationship between family social status and farmer entrepreneurship. With the increase in regional policy uncertainty, the positive effect of family social status on farmer entrepreneurship will be stronger because family social status will generate a formal institutional substitution effect on this condition. Meanwhile, informal institutions, such as the local family culture, will also affect the relationship between family social status and farmer entrepreneurship. In areas with strong family culture, farmers cherish their families and family members are tightly connected with each other. That is to say, in areas with higher levels of family culture, farmers can gain more material and spiritual support from their families, and family social status may produce a more positive entrepreneurial effect. This paper expands the research scope of social class theory from the family dimension, enriches the antecedents of farmer entrepreneurship, and deepens the understanding of the impact of social class on farmers’ career choices.

The remainder of this paper is arranged as follows: Section 2 presents a theoretical analysis and hypothesis development. Section 3 portrays the methods, including the data collection, variable measurement, and empirical strategy. Section 4 reports the baseline results, endogeneity test, and robustness checks. Section 5 presents the main conclusions and discusses the implications and directions for future research.

2. Theory and Hypothesis

2.1. Literature Review

Entrepreneurship is an inherently risky activity [49], and an entrepreneur has to assume some accountability for the inherent risks [50]. Being different from other types of entrepreneurship, farmer entrepreneurship is more risky and challenging because farmers have to run businesses in a constrained and complex rural environment [2]. Due to the lack of high-level educational background, farmers do not have sufficient management skills [3] and entrepreneurial spirit [4] to deal with all the challenges and obstacles in the process of starting a new business [5]. Farmer entrepreneurs have to behave within an extremely uncertain, novel, and turbulent rural environment [51] where they may experience risks that lead to entrepreneurial failure [52].

Researchers have investigated different kinds of risks that farmers may face in entrepreneurship. For example, Rana et al. [53] indicate that disasters in developing rural regions will affect agriculture, food supply, sanitation, and availability of water, which, consequently, makes farmers face the risk of assets loss and entrepreneurial failure. Ranjan [54] points out that, as a lower group in society, farmers are fragile and often at the risk of losing common resources, which hinders their ability to accumulate financial capital and start businesses. Kangogo et al. [55] propose that climate change adds another layer of risks to agricultural production and rural development and thus threatens farmer entrepreneurial tendency. Phan et al. [56] further point out that there are five distinct risky factors in the agricultural sector: production, market, credit, personal, and environmental, all of which may impede farmers’ entrepreneurial motivation. Cao [52] discusses that farmer entrepreneurs may experience risks of market, natural, management, policy, and other man-made factors, and any one of them or a combination will hamper the success of farmer entrepreneurship.

Given these risks, a large number of farmers rush into the tide of entrepreneurship, which has attracted scholars’ extensive attention to study the driven force of farm entrepreneurship. In essence, farmer entrepreneurship is a process in which farmer entrepreneurs achieve transforming match and dynamic balance among different elements, such as opportunities and resources, under the joint influence of various internal and external factors in the heterogeneous rural environment [57]. In this process, the individual characteristics of farmer entrepreneurs dominate the entrepreneurial orientation, and cross-level factors, such as regulations and entrepreneurial environment, drive farmers’ entrepreneurial behavior [58]. Therefore, the existing literature mostly discusses farmers’ entrepreneurial motivation from the aspects of individual characteristics and environmental factors.

In terms of individual characteristics, researchers mainly pay attention to demographic characteristics, personality traits, individual cognition, and personal social network. Zhao and Seibert [59] analyze the influence of people’s demographic characteristics on their entrepreneurial intention, such as age, gender, education level, family background, and personal experience [60,61]. Zhao et al. [59] explore the relationship between personality traits, such as the “big five”, and entrepreneurial motivation and find that achievement motivation, risk-taking tendency, independence, and extroversion may have an enormous influence on entrepreneurship. Boyd and Vozikis [62] study the effect of individual cognitive characteristics on farmers’ entrepreneurial intentions based on self-efficacy. They find that people with high risk preference are more willing to undertake entrepreneurial behavior and will have a high level of opportunity identification efficacy. Researchers also propose that individual entrepreneurial decisions are derived from cognitive biases, which include overconfidence, control fantasy, belief in the law of small numbers, planning fallacy, and optimism bias [63]. Many scholars study the relationship between personal social network and entrepreneurial tendency [64,65] and find that network characteristics have an important impact on their entrepreneurial decision-making [66].

For the environmental factors, scholars have investigated the influence of macroeconomic, industrial, financial, geographical environment, political system, and social and cultural environment on the incidence of entrepreneurship [67]. The Global Entrepreneurship Monitor (GEM) systematically studies the impact of financial support, government policies, business environment, cultural and social norms, infrastructure, and nine other dimensions in the entrepreneurial environment on the rate of new business formation [68]. The important role of government is raised by many scholars; for example, Huang [69] indicates that the success of farmer entrepreneurship could not be achieved without the support of the government. Qi et al. [70] find that the heterogeneity of industrial environment and institutional environment has a significant impact on farmer entrepreneurship. In recent years, some researchers have studied the influence of cultural environment on entrepreneurial decisions and find that different cultural backgrounds at the national or regional level will also affect individuals’ entrepreneurial motivation and thus impact the entrepreneurial rate of a certain region [71]. Entrepreneurial climate and atmosphere will significantly influence people’s entrepreneurial activities too [72]. Jiang and Guo [73] find that the entrepreneurial atmosphere in rural areas will affect the entrepreneurial intention of farmers embedded in them. Successful entrepreneurs in family, relatives, or friends will enhance farmers’ entrepreneurial intentions.

Although the above research helps to explain the reasons why farmers engage in risky entrepreneurial behavior, most of it emphasizes the influence of individual characteristics and environmental elements on farmers’ entrepreneurial ability endowment and entrepreneurial motivation [74]. Few scholars systematically and comprehensively investigate this issue from the perspective of family. As a matter of fact, farmers are generally influenced by the deep-rooted “family culture” in rural areas, and the cohesion formed by unique blood ties, kinship, and family emotions helps farmers cope with various risks in the process of entrepreneurship [75]. It is worth noting that, in China’s transition period, marked by institutional inconsistency, the traditional culture is relatively preserved and intact, and the whole family endowment could be fully utilized by a single family member in China’s rural areas. Farmers’ personal development capacity and individual genius are highly dependent on their family. Therefore, farmers’ individual decisions are often optimized based on their family endowment [76], and the state of family condition will affect farmers’ entrepreneurial decision as well.

In addition, the lack of perfect formal institutions in China’s transition period makes informal social status quite crucial in determining farmers’ entrepreneurial behavior. People with high social status tend to have high social dominance orientation [77], enabling them to obtain key information and scarce resources for entrepreneurial activities [78]. This institutional substitution effect highlights the importance of social status in farmer entrepreneurship. Considering the importance of family and social status for potential entrepreneurs in the rural context, a combinatorial effect might occur through family social status, which should not be neglected in studying farmer entrepreneurship. However, few researchers have focused on this issue. Therefore, this paper aims to discuss the relationship between family social status and farmer entrepreneurship.

2.2. Family Social Status and Farmer Entrepreneurship

The gap between Chinese urban and rural dual structure results in farmers locating in a low social class position [79]. Considering that social class origin has a lasting impact on people’s risk preferences and propensity [29], farmers’ motivation to undertake entrepreneurial activities will also be subject to their class status. Due to farmers’ lack of formal financial services [80,81], material resources, social capital, habitus, and psychological implications derived from family social status are quite important [82].

According to social class theory, social class represents individuals’ perceived economic class positions [26], which reflects their differences in access to resources and social status [28]. Material conditions shape people’s psychological perception of their position in a group by influencing their endowments and experiences [39]. Complex social systems are group-based in nature, with dominant groups at the top and subordinate groups at lower levels [83]. Upper-class social groups tend to have a high social dominance orientation [77]. They maintain their superior status through the implementation of powerful means. In addition, they use philosophical, religious, psychological, or genetic explanations to legitimize their status [84], while groups in lower social classes are more likely to suffer from social and economic discrimination [85]. The differences between classes lead to strong social boundaries, and the single social network generated by social boundaries further limits lower class groups’ access to social capital, making it difficult for them to obtain resources through social relations [86].

The imparity of family social status will produce obvious “social boundaries”, leading to farmers from different classes having different entrepreneurial resources and social capital. Existing studies have shown that, when farmers are constrained by limited capital, technology, information, and other resources, social capital and network become key factors to initiate their motivation to start a new business [18,87]. Farmers from different social classes occupy unequal positions in the social network. Being born to a family of high social class gives an individual a high opportunity of occupying a central position in the social network [88], which gives them social class advantages and enables them to obtain key information and scarce resources [78]. Therefore, the higher family social status is, the closer farmers are to the top of the social pyramid and the more easily they are able to control and obtain entrepreneurial resources by using their network and social capital advantages [89,90]. In addition, high family social status is linked with a rich social network and more social capital, and, by utilizing this, farmers can access entrepreneurial information [91] and identify more entrepreneurial opportunities [47]. However, farmers from low social class families cannot access the network and resources owned by the elite group, and resource constraints make it difficult for them to become self-employed. Therefore, farmers at the bottom of society are the least likely to start a new business [85].

Cultural values vary among different social class groups [92] because most people form their unique “habitus” through social experiences. This is not a purely “internalized” and subjective psychological state but a kind of behavior system that accompanies social structure and guides individual actions. Under the influence of habitus, people from a particular class will develop a set of normative personalities, expectations, and behaviors that are unique to members of that class [93]. These characteristics can better reflect the differentiated social skills among different classes. The “habitus” engendered by class experience shapes farmers’ social class imprints and is accompanied by resource and class differences, which continue to influence their risk perception and entrepreneurial motivation. As a result, people from different social statuses will have completely different attention directions when facing the same event: “threat” or “opportunity” [29]. If people pay more attention to threats, it means that they are afraid of any possible losses in the future and are less willing to engage in risky behaviors. On the contrary, if people are more likely to focus on opportunities, they will seek positive results with a more optimistic attitude, so they are more willing to undertake risky behaviors, such as entrepreneurial activities.

Farmers from higher social status families usually grow up in environments that are rich in resources, which generates a positive psychological effect. The psychological effect mentioned in this paper refers to various psychological outcomes and perceptions produced by different family social status. Higher family social status can help an individual build a huge psychological safety net [94], putting them in a cognitive state that they do not need to worry about any negative impact to their self-image, status, or career [95]. Psychological perceptions influence people’s behavior [96]: in a psychologically safe environment, people are more comfortable to take risks [97]. As pointed out by Williams [81], people from upper-class families are often encouraged to try new things, and, if they get into trouble, their parents will find ways to help them out. Therefore, farmers from families with high social status tend to think that the world is safe and full of opportunities [98] and are more likely to have high tolerance for potential losses and the risks of entrepreneurship [81]. In addition, being from a superior social class is associated with optimism in upper-class farmers while facing risky situations [39]. Even though entrepreneurship may produce potential negative consequences or losses, farmers from high social class families are more likely to detect positive consequences or benefits from risks [99]. Finally, different family social statuses are also associated with different levels of power or control over resources [100,101]. People who act in a rewarding environment will have a higher perception of power, and an improvement in power perception increases the probability of approaching behaviors, such as entrepreneurial activities [102]. Therefore, farmers from families with higher social status are more interested in the return of entrepreneurship while downplaying its potential loss; in other words, they have more optimistic attitudes towards the success of entrepreneurship [102] and are more likely to have a proximity motivation for entrepreneurship.

In contrast, lower family social status is associated with restricted resources and necessities. Compared with those from the upper social class, farmers from families with low social status usually grow up in relatively poor and uncertain environments [103] and, therefore, have smaller and more fragile safety nets [93]. Due to the lack of adequate safety nets to cushion the negative effects of accidents [103], low-status groups are more inclined to magnify the severity of negative consequences produced by small errors [29]. Rough growth experiences make them fearful and stressed when facing risks [104]. Over time, farmers from families with lower social status develop greater levels of sensitivity and avoidance motivation to potential threats [102,105], and their tolerance to risk is further reduced. Therefore, farmers from lower social class are more inclined to allocate their attention to the uncertainty and threats of entrepreneurship, which is accompanied by greater attention to the negative effects of entrepreneurship [106]. Thus, they are more likely to avoid risky entrepreneurial behaviors. Finally, the inherent value system associated with a specific group or society shapes personality traits [107]. Underclass farmers often lack financial knowledge compared with those at the advanced level [108]. Cognitive development related to financial knowledge is important for self-employment, which might further restrict the entrepreneurial willingness of farmers with low family social status. Based on the above discussion, we propose the below hypothesis:

Hypothesis 1 (H1).

Farmers from high social status families are more inclined to conduct entrepreneurial activities.

2.3. The Moderating Effect of Regional Policy Uncertainty

External environment is an important factor that affects people’s strategic decisions [109], and institutional characteristics also restrict farmers’ entrepreneurial behaviors. Based on institutional theory, institution can be defined as “the rules of the game in a society” [110]. These rules include both formal institutions and informal institutions. Formal institutions refer to the constitutional, legal, and organizational framework for people’s actions, while informal institutions are mainly about cultures, values, and norms [110]. The institutionalist framework is widely used to study entrepreneurial behavior [111,112,113].

According to North’s study [110], stable formal institutions can help to reduce uncertainty and transaction costs connected with entrepreneurial activity. Meanwhile, during the period of China’s transition toward a market-oriented economy, the formal institution is void and the legal framework is inadequate, which will lead to a lack of property rights and stable policy. In this context, formal institutions, such as regional policy uncertainty, represent a constraining force to farmer entrepreneurship [114].

Regional policy uncertainty refers to the unpredictability caused by the unclear direction of local policies and regulations related to business operation, which makes it difficult for local entities to predict whether, when, and how the government will adjust the current policies [115,116]. Compared with other sectors, the agricultural sector is more susceptible to sudden and unpredictable external shocks, including both natural disasters and policy changes [117]. It is difficult for farmers and enterprises in rural areas to achieve long-term development by their own strength, which, in turn, leads to their high dependence on regional policy support [118]. As a result, when the regional policy becomes uncertain and unpredictable, it will affect the way family social status stimulates farmer entrepreneurship. In regions with weak policy consistency, farmers have to face environmental uncertainty and policy variability, which significantly increases their risk of entrepreneurial failure. Policy uncertainty is associated with rapid change in external conditions and major institutional deficiencies, thus resulting in additional operating costs for farmer entrepreneurs. Therefore, the stability of regional policies cannot be ignored when studying farmers’ entrepreneurial strategies [119]. With the rapid change in the economic environment, the diversification of customer demands, and the development of technology, external unpredictability will impact farmers’ entrepreneurial decisions, as well as affect the influence of family social status on farmer entrepreneurship [120].

Policy uncertainty will strengthen the class differentiation effect of family social status. In regions with frequent policy changes, the instability and dynamicity of policy will increase the risk of entrepreneurship as farmer entrepreneurs in these regions have to deal with a high level of policy-related unpredictability, so they need to invest more time and resources into buffering these negative impacts. Nevertheless, farmers generally lack experience in coping with risks, so they need to rely more on the network, information, and social capital provided by their family social status to boost the predictability of their entrepreneurial results. Additionally, while facing a high level of policy uncertainty, farmer entrepreneurs require a positive attitude when dealing with potential losses and risky entrepreneurial activities, which could stimulate the positive “safety net” effect of high social status. Under the protection of this physical and psychological safety net, farmer entrepreneurs have more courage and confidence to cope with entrepreneurial risks and policy uncertainty. Therefore, in regions with high policy uncertainty, the promoting effect of high family social status on farmer entrepreneurship will be further strengthened. Meanwhile, when policies are frequently volatile, farmer entrepreneurs need to rely more on nonmarket factors, such as family social status, to obtain business information and alleviate their fear of entrepreneurial failure. Based on the above discussion, we propose the below hypothesis:

Hypothesis 2 (H2).

Regional policy uncertainty moderates the positive effect of family social status on farmer entrepreneurship. The effect is stronger in regions with unpredictable policy than in regions with stable policy.

2.4. The Moderating Effect of Local Family Culture

Farmer entrepreneurship is restricted not only by formal institutional factors but also by informal institutional factors. North [110] noted that “although formal rules may change overnight as the result of political and judicial decisions, informal constraints embodied in customs, traditions, and codes of conduct are much more impervious to deliberate policies.” Therefore, in regions where formal institutions are weak, informal institutional constraints may play a more important role in impacting people’s behavior [111]. One primary informal institution related to culture is an extensively important constraint; for example, the local family culture acts as a widespread informal institutional factor constraining farmers’ entrepreneurial behavior.

Compared with urban residents, farmer entrepreneurship places more emphasis on family culture [121]. In China, regions differ not only in terms of economic development but also in terms of people’s attachment to their family. Farmers living in regions with tight family relationships pay significant attention to, and have strong dependence on, family and marriage, which leads to a high level of local family culture.

In regions with strong local family culture, farmers pay more attention to their traditional families, and families can offer them more material and spiritual supports for entrepreneurial behavior. The stronger family culture is, the closer family relationships will be kept by locals, and the stronger level of family cohesion will be [122]. Under these conditions, family social status will further boost farmers’ motivation to start new businesses. On the one hand, when facing external uncertainty and fierce competition, close family relationships shaped by strong family culture helps farmers to achieve strategic consistency with family members, making it possible for farmer entrepreneurs to improve their decision-making efficiency in the complicated and changeable external environments and to seize fleeting opportunities. On the other hand, the “family first” view, fostered by a strong local family culture, makes high social class members more willing to exploit entrepreneurial needs for their family members. Potential farmer entrepreneurs are more likely to obtain scarce entrepreneurial resources and opportunities through these family relations. At the same time, emphasis on the family unit helps to strengthen family cohesion, and family members will provide each other with strong material and emotional support. Thus, farmer entrepreneurs can not only obtain entrepreneurship resources from the whole family endowment [123] but also be able to deal with business risk more positively by reducing their sensitivity to business failure and increasing their tolerance of entrepreneurial failure. Therefore, in areas with strong family culture, the promoting effect of family social status on farmer entrepreneurship is further strengthened.

On the contrary, in regions with weak family culture, the concept of “family” is becoming increasingly diluted, and family cohesion is constantly declining [124]. Compared with areas with strong local family culture, “families” are more likely to be disorganized, with weak family culture conditions, where individualistic characteristics are highlighted. Therefore, in farming areas with weak family culture, people probably lack the family concept, and the altruism psychology within families may be weak. Under these conditions, it is difficult for peasant households to form family cohesion. Even in families with an eminent social status, other family members’ networks and resources cannot be fully utilized by potential rural entrepreneurs. The lack of family closeness weakens the incentive effect of family social status on farmers’ entrepreneurial motivation. Similarly, for farmers from families of low social class, willingness to start a business is also reduced by the weakening family cohesion. Based on the above analysis, we propose the following hypothesis:

Hypothesis 3 (H3).

Local family culture moderates the positive effect of family social status on farmer entrepreneurship. The effect is stronger in areas with tight family relationship than in areas with weak family culture.

We develop a conceptual framework set forth in Figure 1 to guide the design of methods. We propose that farmer entrepreneurship could be driven by the level of family social status, and resource availability and psychological support might act as a potential mechanism. This mechanism is not tested in this paper due to the unavailability of data, so we use dash lines in the framework. The relationship between family social status and farmer entrepreneurship is moderated by the external institutional environment, including both formal institutions, such as regional policy uncertainty, and informal institutions, such as local family culture.

3. Methods

3.1. Data and Sample

The data utilized in this study are from the Thousand Villages Survey in China conducted by Shanghai University of Finance and Economics (SUFE). This survey is a large-scale, nationwide social investigation with issues concerning “agriculture, countryside and farmers” as the main research object. It aims to investigate the status quo of China’s farmer entrepreneurship, dig out the key factors affecting the entrepreneurial vitality in rural areas, evaluate the entrepreneurial characteristics of farmer entrepreneurs, find out the success or failure logic of rural entrepreneurs, and, finally, determine the growth path of rural entrepreneurial enterprises. In this investigation, we use the face-to-face interview method to acquire data. This method is preferred not only for its high response rate but also because it ensures that detailed information and opinions can be extracted from the participants. In addition, this tactic can help to improve the accuracy of data from the respondents by allowing easy clarification of issues. The one-to-one discussion between researcher and respondent can help to operationalize the concepts involved and allow for deeper examination of the research questions than other methods [125].

To enhance the construct validity of the questionnaire measures, we conducted a pretest by interviewing 15 rural private business owners and 40 villagers. Each interview was approximately 60 to 75 min in length. In order to make the interviews clearer and easier to complete, and minimize the likelihood of social desirability bias, we used feedback from the interviews to revise the format of the questionnaire and the specific wording of questions. The layout of the questionnaire was very simple so as to encourage meaningful participation by the respondents.

We conducted the personal interviews by 30 teachers and 2188 university students in SUFE from June to September 2016. Five experts (including one of the authors) were involved in the process, and their main jobs included, but were not limited to, arranging meetings, coordinating efforts, and managing data collection process. Before starting the work, a detailed training session was conducted to ensure the quality of the data. Additionally, each interview was conducted in the native language of the interviewee so that they could express their thoughts, feelings, and opinions completely. The interviews were structured according to the questionnaire and interview protocol designed. The average length of the main interviews was one to two hours.

The sample frame for this study consisted of 3860 entrepreneurial farmers and 9806 non-entrepreneurial farmers from 1167 Chinese villages located in 31 different provinces. The entrepreneurial sample covers 16 industries, including, but not limited to, manufacturing, construction, forestry and fishing, mining, real estate, and information technology services. Sample selection from multiple industries rather than being limited to a specific industry contributes to increase the external validity of the research conclusions [126,127]. Similar to the study in Ref. [128], we applied stratified sampling technique to select the respondents. This method is preferred because it can minimize bias when dealing with the population. The technique helps to frame samples in relatively homogeneous groups before sampling and makes the final sample more representative in terms of the stratified groups [129]. The respondents are geographically dispersed across China (see

Table 1. Distribution of sample.

No.	Area	Frequency	Percentage	Cumulative Percentage
1	Southeast	4227	30.931	30.931
2	Bohai Rim	1646	12.044	42.975
3	Central region	2830	20.708	63.684
4	Northeast	731	5.349	69.033
5	Southwest	2425	17.745	86.777
6	Northwest	1807	13.223	100
		13,666	100

) instead of from one specific region, which further helps to increase the applicability and external validity of the conclusions. In general, the sample represents a large population of rural private business owners and villagers in China reasonably well and indicates good generalizability of the results.

3.2. Measures

Family social status. Previously, objective economic indicators were commonly used to measure social class [130], but more and more scholars are starting to pay attention to the subjective indicators of social class perceived by social members [131]. We adopt the method of Kish-Gephart and Campbell [29] in this study and use the below question to measure family social status: “What kind of social class level is your family in?” We coded answers from 1 to 5 to stand for “low level”, “low middle level”, “middle level”, “upper middle level”, and “upper level”, respectively.

Farmer entrepreneurship. Being similar to McElwee [2], we define farmer entrepreneurship as both farm and non-farm activities undertaken by local villagers in rural area for profitable gains through establishing new firms by themselves or in partnership with others. Referring to the study by Yang et al. [121], we set farmer entrepreneurship as a dummy variable. As the questionnaires were filled out by both entrepreneurial farmers and non-entrepreneurial farmers, we regard those conducting entrepreneurial behaviors as entrepreneurial farmers (=1) and others as non-entrepreneurial farmers (=0). Considering that the dependent variable is dummy variable, we test the above hypothesis through logistic model.

Regional policy uncertainty. Emphasis on unified leadership is an important feature of Chinese leadership system; that is, the power of local governments is concentrated in party committees, and the power of party committees is concentrated in party secretaries [132]. As a result, the replacement of local party secretary is taken as the proxy for regional policy uncertainty [133]. If the party secretary in a region changed, the regional policies are likely to change accordingly; on that condition, regional policy uncertainty is coded 1; otherwise, it is coded 0.

Local family culture. According to previous studies, the overall marital status in a region can reflect the attitudes of locals towards family relations [134], so scholars use the regional level marital status as a proxy for local family culture [135]. This is measured by the difference between the natural logarithm of number of registered marriages and number of divorce cases per 10,000 people in a region. The larger this value is, the stronger local family culture is, and the closer family relationship will be. Data on this variable are derived from China Civil Affairs Statistical Yearbook.

Control variables. We include several control variables that may influence farmer entrepreneurship. For individual-level variables, we control for gender (male = 1, otherwise 0), age (years), marital status (married = 1, otherwise 0), skills (having special skill = 1, otherwise 0), and network (measured by number of contacts in the mobile phone address list). For family-level variables, we control for family size, which is measured by the number of family members living together. We also include the cadre proportion and male proportion in the family as these factors might influence a farmer’s entrepreneurship motivation in different ways. The factor single-parent family (yes = 1, otherwise 0) is also controlled for because those from a single-parent family generally lack safety and, thus, are less likely to undertake risky entrepreneurial activities. We control for the area of cultivated land per capita and the per capita family income (ten thousand Yuan/person) in the model as these indicators reflect the family-level resource endowment, which is vital for entrepreneurial behavior. Considering that entrepreneurship atmosphere (measured by number of family members engaging in entrepreneurial activities) will also affect farmer entrepreneurship, we control it in the model. We also control for some village-level variables, such as the per capita income of village in the previous year (10 thousand Yuan), which may have a positive effect on farmers’ entrepreneurial behavior. The number of companies in the village is also controlled for as this illustrates the competition faced by the focal firm established by local farmers. Additionally, the number of permanent resident households, the number of migrant households, and the number of families involved in entrepreneurial activities in the village are also controlled for. Finally, we control for the marketization index [136] in the model. Marketization index is an indicator to measure the institutional differences among the provinces in China [137]. A high marketization environment is characterized by established capital market structures with well-developed legal supports for operating business, such as contract enforcement and limited government intervention in activities of economic resource allocation [137,138,139]. In such an environment, people’s motivation to start a new business might be higher. In

Table 2. Descriptive statistics and correlations.

	Variable	Mean	SD	1	2	3	4	5	6	7	8	9	10
1	Farmer entrepreneurship	0.282	0.450	1.000
2	Family social status	3.197	0.770	0.234 ***	1.000
3	Regional policy uncertainty	0.379	0.234	0.010	−0.009	1.000
4	Local family culture	1.148	0.290	0.007	0.004	0.370 ***	1.000
5	Gender	0.740	0.439	0.139 ***	0.058 ***	0.055 ***	0.031 ***	1.000
6	Age	45.020	11.956	−0.020 **	0.006	−0.011	−0.037 ***	0.135 ***	1.000
7	Marital status	0.895	0.307	0.117 ***	0.044 ***	0.031 ***	0.001	0.037 ***	0.414 ***	1.000
8	Skills	0.331	0.471	0.063 ***	0.061 ***	0.014	0.027 ***	0.091 ***	−0.009	0.039 ***	1.000
9	Network	133.890	149.910	0.318 ***	0.218 ***	0.019 **	0.025 ***	0.127 ***	−0.156 ***	0.039 ***	0.073 ***	1.000
10	Family size	3.041	0.909	−0.016 *	−0.007	0.095 ***	0.138 ***	0.007	−0.054 ***	0.014	0.009	0.012	1.000
11	Cadre proportion	0.040	0.128	0.021 **	0.072 ***	0.002	0.008	0.009	0.153 ***	0.052 ***	−0.010	0.006	−0.082 ***
12	Male proportion	0.481	0.212	0.008	−0.007	−0.018 **	0.013	0.039 ***	−0.051 ***	−0.028 ***	−0.011	0.007	−0.195 ***
13	Single-parent family	0.032	0.177	−0.022 **	−0.030 ***	0.004	0.020 **	0.016 *	0.028 ***	−0.011	0.009	−0.034 ***	−0.008
14	Cultivated land per capita	2.076	5.149	0.079 ***	0.033 ***	0.074 ***	−0.061 ***	0.060 ***	0.016 *	0.033 ***	0.008	0.048 ***	−0.057 ***
15	Per capita income of family	3.750	5.928	0.268 ***	0.280 ***	−0.083 ***	−0.136 ***	0.054 ***	−0.062 ***	−0.032 ***	0.039 ***	0.267 ***	−0.133 ***
16	Entrepreneurship atmosphere	3.145	10.038	0.119 ***	0.058 ***	0.030 ***	0.044 ***	0.035 ***	−0.037 ***	0.027 ***	0.072 ***	0.157 ***	0.021 **
17	Per capita income of village	0.092	0.180	−0.018 **	−0.009	−0.095 ***	−0.160 ***	−0.023 ***	−0.027 ***	−0.024 ***	0.007	0.015 *	−0.044 ***
18	Number of permanent resident households	7.624	0.932	0.005	−0.028 ***	−0.061 ***	−0.015 *	−0.024 ***	−0.032 ***	−0.018 **	0.012	0.047 ***	−0.017 **
19	Number of companies	8.698	16.997	0.009	0.012	−0.098 ***	−0.083 ***	−0.034 ***	−0.032 ***	−0.030 ***	0.006	0.068 ***	−0.057 ***
20	Number of migrant households	3.678	2.961	−0.018 **	0.005	−0.156 ***	−0.229 ***	−0.062 ***	−0.059 ***	−0.053 ***	0.002	0.073 ***	−0.065 ***
21	Number of families involved in entrepreneurial activities	14.440	43.643	0.010	−0.001	−0.018 **	0.038 ***	−0.004	−0.051 ***	−0.011	0.002	0.048 ***	−0.001
22	Marketization index	7.468	1.870	0.001	0.018 **	−0.239 ***	−0.440 ***	−0.039 ***	0.083 ***	0.001	−0.005	0.021 **	−0.101 ***
	Variable	11	12	13	14	15	16	17	18	19	20	21	22
11	Cadre proportion	1.000
12	Male proportion	0.061 ***	1.000
13	Single-parent family	0.022 ***	−0.006	1.000
14	Cultivated land per capita	0.044 ***	0.045 ***	0.011	1.000
15	Per capita income of family	−0.012	0.090 ***	−0.017 **	0.055 ***	1.000
16	Entrepreneurship atmosphere	−0.006	0.000	−0.002	0.005	0.061 ***	1.000
17	Per capita income of village	−0.020 **	−0.006	−0.006	0.000	0.099 ***	−0.014 *	1.000
18	Number of permanent resident households	0.003	0.011	−0.016 *	−0.076 ***	0.077 ***	0.030 ***	−0.129 ***	1.000
19	Number of companies	−0.016 *	0.006	−0.021 **	−0.069 ***	0.154 ***	0.066 ***	0.075 ***	0.303 ***	1.000
20	Number of migrant households	−0.042 ***	0.008	−0.025 ***	−0.086 ***	0.174 ***	0.019 **	0.164 ***	0.441 ***	0.379 ***	1.000
21	Number of families involved in entrepreneurial activities	0.019 **	0.006	−0.006	0.026 ***	0.023 ***	0.054 ***	−0.023 ***	0.106 ***	0.156 ***	0.085 ***	1.000
22	Marketization index	0.033 ***	0.016 *	−0.037 ***	−0.109 ***	0.210 ***	−0.015 *	0.124 ***	0.195 ***	0.270 ***	0.322 ***	−0.028 ***	1.000

Note: n = 13,666. *** p < 0.01, ** p < 0.05, * p < 0.1.

, we report the descriptive statistics and correlations for all variables.

3.3. Empirical Strategy

We used logistic regression to test our hypothesis for the dependent variable, which is a dummy variable in our case. Logistic regression is defined as a classical statistical model, which is used to analyze the relationship between a dichotomous or binary dependent and one or more nominal, ordinal, interval, or ratio-level independent variables [140]. Logistic regression is also referred to as logit model, which is widely applied in the field of entrepreneurship [141,142]. Based on the logit transformation of the dependent variable, logistic regression model could quantify the occurrence “odds” of an event and predict the outcome probabilities for dependent variable [143]. The use of logistic regression model in this study is to evaluate the probability of farmers’ entrepreneurial behavior based on a set of independent variables.

A multicollinearity test was carried before the regression and the results show that there is no serious multicollinearity problem in the model. We winsorized some continuous variables with extreme values by 1% to eliminate the deviation caused by outliers. In order to eliminate heteroscedasticity and autocorrelation of residual errors, all regressions presented in this paper adopted a robust standard error.

Considering that entrepreneurship can be regarded as an important path for farmers to pursue social class [144], there may be a reverse causal relationship between family social status and farmer entrepreneurship, and variable omission might also cause potential endogeneity. Additionally, family social status may be formed from a wide range of circumstances, so it is inevitable that common factors will lead to improvement in family social status and affect farmers’ entrepreneurial motivation. In order to correct potential endogeneity problems, we use an instrument variable regression and propensity score matching method (PSM) to modify the original model. In order to ensure the reliability of the research conclusions, we further adopt Heckman’s two-step method to correct any sample selection bias that may exist in the original model.

4. Results

4.1. Logistic Regression Results

Table 3. Logistic regression results.

	Model 1	Model 2	Model 3	Model 4	Model 5
Gender	0.575 ***	0.575 ***	0.575 ***	0.575 ***	0.574 ***
	(0.054)	(0.054)	(0.054)	(0.054)	(0.054)
Age	−0.010 ***	−0.010 ***	−0.010 ***	−0.010 ***	−0.010 ***
	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)
Marital status	1.213 ***	1.213 ***	1.215 ***	1.212 ***	1.216 ***
	(0.095)	(0.095)	(0.095)	(0.095)	(0.095)
Skills	0.077 *	0.077 *	0.076 *	0.077 *	0.078 *
	(0.045)	(0.045)	(0.045)	(0.045)	(0.045)
Network	0.003 ***	0.003 ***	0.003 ***	0.003 ***	0.003 ***
	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)
Family size	−0.014	−0.014	−0.014	−0.014	−0.014
	(0.024)	(0.024)	(0.024)	(0.024)	(0.024)
Cadre proportion	0.176	0.176	0.171	0.177	0.176
	(0.165)	(0.165)	(0.166)	(0.166)	(0.166)
Male proportion	−0.158	−0.159	−0.159	−0.158	−0.160
	(0.104)	(0.104)	(0.104)	(0.104)	(0.104)
Single-parent family	−0.178	−0.178	−0.180	−0.178	−0.179
	(0.127)	(0.127)	(0.127)	(0.127)	(0.127)
Cultivated land per capita	0.019 ***	0.019 ***	0.019 ***	0.019 ***	0.019 ***
	(0.004)	(0.004)	(0.004)	(0.004)	(0.004)
Per capita income of family	0.079 ***	0.079 ***	0.079 ***	0.079 ***	0.079 ***
	(0.004)	(0.004)	(0.004)	(0.004)	(0.004)
Entrepreneurship atmosphere	0.027 ***	0.027 ***	0.027 ***	0.027 ***	0.027 ***
	(0.003)	(0.003)	(0.003)	(0.003)	(0.003)
Per capita income of village	−0.345 ***	−0.345 ***	−0.348 ***	−0.346 ***	−0.352 ***
	(0.130)	(0.130)	(0.130)	(0.130)	(0.130)
Number of permanent resident households	0.034	0.033	0.032	0.034	0.033
	(0.027)	(0.027)	(0.027)	(0.027)	(0.027)
Number of companies	−0.003 *	−0.003 *	−0.003 *	−0.003 *	−0.002 *
	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)
Number of migrant households	−0.042 ***	−0.042 ***	−0.042 ***	−0.042 ***	−0.043 ***
	(0.009)	(0.009)	(0.009)	(0.009)	(0.009)
Number of families involved in entrepreneurial activities	−0.000	−0.000	−0.000	−0.000	−0.000
	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)
Marketization index	−0.020	−0.019	−0.019	−0.020	−0.019
	(0.013)	(0.013)	(0.013)	(0.014)	(0.014)
Family social status	0.437 ***	0.437 ***	0.437 ***	0.437 ***	0.435 ***
	(0.030)	(0.030)	(0.030)	(0.030)	(0.030)
Regional policy uncertainty		0.016	−0.003
		(0.087)	(0.089)
Family social status×Regional policy uncertainty			0.228 *
			(0.121)
Local family culture				−0.008	−0.041
				(0.084)	(0.085)
Family social status×Local family culture					0.298 ***
					(0.100)
_cons	−4.199 ***	−4.200 ***	−4.198 ***	−4.188 ***	−4.145 ***
	(0.275)	(0.275)	(0.275)	(0.296)	(0.296)
Chi2	2618.909	2618.942	2622.604	2618.919	2627.815
R2_p	0.161	0.161	0.161	0.161	0.162
n	13,666	13,666	13,666	13,666	13,666

Note: *** p < 0.01, ** p < 0.05, * p < 0.1; standard error in brackets.

lists the logistic regression results of the relationship between family social status and farmer entrepreneurship. Model 1 shows a significant positive correlation between family social status and farmer entrepreneurship (beta = 0.437, p < 0.01), which means that, the higher family social status is, the more likely farmers are to choose entrepreneurship. Hypothesis 1 is supported.

Model 3 illustrates that the coefficient of interaction between family social status and regional policy uncertainty is significantly positive (beta = 0.228, p < 0.1), which means that, in areas with greater policy uncertainty, farmer entrepreneurs’ reliance on family status becomes stronger, and the incentive effect of high family social status on a farmer’s entrepreneurial enthusiasm will be strengthened. Hypothesis 2 is supported. Model 5 shows that the coefficient of interaction between family social status and local family culture is significantly positive (beta = 0.298, p < 0.01), indicating that, in regions with strong family culture, the positive effect of family social status on farmers’ entrepreneurship will be strengthened. Hypothesis 3 is verified.

4.2. Propensity Score Matching Test

In order to correct the potential endogeneity problems, we apply the propensity score matching method (PSM) to modify the original model in this paper. Specifically, we use gender, age, marital status, craft skill, social network, and income as feature variables to match the experimental group and control group. By using the nearest neighbor matching method, we set a proportion of 1:1 for sample matching, and the results show that the mutual support hypothesis and balance assumptions are satisfied. We re-test the logistic regression model by using the samples (6875 observed values) matched by PSM method, and the results are shown in

Table 4. Propensity score matching test.

	Model 6	Model 7	Model 8	Model 9	Model 10
Gender	0.595 ***	0.595 ***	0.593 ***	0.595 ***	0.595 ***
	(0.067)	(0.067)	(0.067)	(0.067)	(0.067)
Age	−0.012 ***	−0.012 ***	−0.012 ***	−0.012 ***	−0.012 ***
	(0.003)	(0.003)	(0.003)	(0.003)	(0.003)
Marital status	1.301 ***	1.301 ***	1.306 ***	1.302 ***	1.306 ***
	(0.118)	(0.118)	(0.118)	(0.118)	(0.118)
Skills	0.059	0.059	0.057	0.059	0.058
	(0.053)	(0.053)	(0.053)	(0.053)	(0.053)
Network	0.003 ***	0.003 ***	0.003 ***	0.003 ***	0.003 ***
	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)
Family size	−0.011	−0.011	−0.013	−0.012	−0.011
	(0.029)	(0.029)	(0.029)	(0.029)	(0.030)
Cadre proportion	0.037	0.036	0.035	0.034	0.032
	(0.182)	(0.182)	(0.183)	(0.183)	(0.183)
Male proportion	−0.132	−0.131	−0.134	−0.135	−0.135
	(0.125)	(0.125)	(0.125)	(0.126)	(0.126)
Single-parent family	−0.235	−0.235	−0.231	−0.236	−0.233
	(0.151)	(0.151)	(0.151)	(0.151)	(0.151)
Cultivated land per capita	0.020 ***	0.020 ***	0.020 ***	0.020 ***	0.020 ***
	(0.005)	(0.005)	(0.005)	(0.005)	(0.005)
Per capita income of family	0.075 ***	0.075 ***	0.075 ***	0.075 ***	0.075 ***
	(0.005)	(0.005)	(0.005)	(0.005)	(0.005)
Entrepreneurship atmosphere	0.018 ***	0.018 ***	0.018 ***	0.018 ***	0.018 ***
	(0.004)	(0.004)	(0.004)	(0.004)	(0.004)
Per capita income of village	−0.358 **	−0.357 **	−0.357 **	−0.354 **	−0.357 **
	(0.156)	(0.156)	(0.156)	(0.156)	(0.156)
Number of permanent resident households	0.032	0.032	0.032	0.030	0.030
	(0.032)	(0.032)	(0.032)	(0.032)	(0.032)
Number of companies	−0.002	−0.002	−0.002	−0.002	−0.002
	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)
Number of migrant households	−0.038 ***	−0.038 ***	−0.038 ***	−0.037 ***	−0.037 ***
	(0.011)	(0.011)	(0.011)	(0.011)	(0.011)
Number of families involved in entrepreneurial activities	0.000	0.000	0.000	0.000	0.000
	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)
Marketization index	−0.023	−0.022	−0.023	−0.020	−0.019
	(0.015)	(0.015)	(0.015)	(0.016)	(0.016)
Family social status	0.445 ***	0.445 ***	0.445 ***	0.444 ***	0.443 ***
	(0.035)	(0.035)	(0.035)	(0.035)	(0.035)
Regional policy uncertainty		0.042	−0.052
		(0.113)	(0.124)
Family social status×Regional policy uncertainty			0.282 *
			(0.151)
Local family culture				0.057	−0.037
				(0.102)	(0.109)
Family social status×Local family culture					0.285 **
					(0.119)
_cons	−4.167 ***	−4.189 ***	−4.149 ***	−4.242 ***	−4.129 ***
	(0.333)	(0.338)	(0.338)	(0.358)	(0.360)
Chi2	1765.445	1765.584	1769.085	1765.764	1771.512
R2_p	0.160	0.160	0.160	0.160	0.160
n	8596	8596	8596	8596	8596

Note: *** p < 0.01, ** p < 0.05, * p < 0.1; standard error in brackets.

.

The results in Table 4 show that the conclusion of this study is still valid when the matching bias is taken into account. The improvement in family social status will significantly stimulate farmers’ enthusiasm to start new businesses (beta = 0.445, p < 0.01), and this stimulating effect is found to be more significant in regions with greater policy uncertainty (beta = 0.282, p < 0.1) and strong family culture (beta = 0.285, p < 0.05).

4.3. Instrumental Variable Regression

We also adopt two-stage instrumental variable methods to correct for the endogeneity of family social status. The provincial average of NPC (the National People’s Congress) deputies and CPPCC (The Chinese People’s Political Consultative Conference) members is used as instruments because it is theoretically correlated with family social status but unrelated to farmers’ motivation for starting a new business. The results of the instrumental variable regression are shown in

Table 5. Instrument variable regression.

	Model 11	Model 12	Model 13	Model 14	Model 15
Gender	0.296 ***	0.296 ***	0.294 ***	0.293 ***	0.293 ***
	(0.056)	(0.057)	(0.057)	(0.060)	(0.060)
Age	−0.006 ***	−0.006 ***	−0.006 ***	−0.006 ***	−0.006 ***
	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)
Marital status	0.601 ***	0.602 ***	0.604 ***	0.595 ***	0.595 ***
	(0.116)	(0.117)	(0.118)	(0.125)	(0.125)
Skills	0.018	0.018	0.016	0.016	0.017
	(0.041)	(0.041)	(0.041)	(0.042)	(0.042)
Network	0.001 ***	0.001 ***	0.001 ***	0.001 **	0.001 **
	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)
Family size	−0.016	−0.016	−0.017	−0.016	−0.016
	(0.015)	(0.015)	(0.015)	(0.015)	(0.015)
Cadre proportion	−0.105	−0.104	−0.107	−0.115	−0.115
	(0.226)	(0.227)	(0.227)	(0.233)	(0.232)
Male proportion	−0.030	−0.030	−0.030	−0.026	−0.027
	(0.082)	(0.083)	(0.083)	(0.085)	(0.085)
Single-parent family	−0.040	−0.040	−0.038	−0.036	−0.037
	(0.091)	(0.091)	(0.091)	(0.093)	(0.093)
Cultivated land per capita	0.011 ***	0.011 ***	0.011 ***	0.010 ***	0.010 ***
	(0.003)	(0.003)	(0.003)	(0.003)	(0.003)
Per capita income of family	0.027	0.027	0.027	0.026	0.026
	(0.020)	(0.020)	(0.020)	(0.021)	(0.021)
Entrepreneurship atmosphere	0.010 ***	0.010 ***	0.010 ***	0.010 ***	0.010 ***
	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)
Per capita income of village	−0.111	−0.111	−0.113	−0.108	−0.110
	(0.113)	(0.114)	(0.114)	(0.115)	(0.115)
Number of permanent resident households	0.036 *	0.036 *	0.036 *	0.038 *	0.037 *
	(0.022)	(0.022)	(0.021)	(0.022)	(0.022)
Number of companies	−0.001	−0.001	−0.001	−0.001	−0.001
	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)
Number of migrant households	−0.022 ***	−0.023 ***	−0.023 ***	−0.023 ***	−0.023 ***
	(0.006)	(0.006)	(0.006)	(0.006)	(0.006)
Number of families involved in entrepreneurial activities	−0.000	−0.000	−0.000	−0.000	−0.000
	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)
Marketization index	−0.007	−0.007	−0.007	−0.008	−0.008
	(0.009)	(0.009)	(0.009)	(0.009)	(0.009)
Family social status	0.691 *	0.690 *	0.699 *	0.714 *	0.715 *
	(0.411)	(0.414)	(0.412)	(0.426)	(0.424)
Regional policy uncertainty		−0.015	−0.029
		(0.056)	(0.057)
Family social status×Regional policy uncertainty			0.234 ***
			(0.070)
Local family culture				−0.024	−0.037
				(0.053)	(0.052)
Family social status×Local family culture					0.130
					(0.082)
_cons	−3.723 ***	−3.714 ***	−3.731 ***	−3.755 ***	−3.741 ***
	(1.143)	(1.159)	(1.152)	(1.143)	(1.138)
Chi2	2517.967	2516.551	2545.908	2577.966	2593.045
n	13,666	13,666	13,666	13,666	13,666

Note: *** p <0.01, ** p <0.05, * p < 0.1; standard error in brackets.

.

Model 11 shows that family social status is positively related to farmer entrepreneurship (beta = 0.691, p < 0.1), and Hypothesis 1 is supported. The coefficient of interaction between family social status and regional policy uncertainty presented in Model 13 is significantly positive (beta = 0.234, p < 0.01), which supports Hypothesis 2. In Model 15, the coefficient of interaction between family social status and local family culture is positive (beta = 0.130, ns), and the T value is 1.59, approaching the level of significance.

4.4. Heckman’s Two-Stage Test

Heckman’s two-step method is used to remedy model misspecification due to sample selection bias. First, we construct a probability equation to indicate whether a farmer comes from a high social status family, then calculate the inverse Mills ratio (IMR) using the estimated results. Then, a regression equation of the influence of family social status on farmer entrepreneurship is constructed. In order to obtain more accurate estimates, the IMR is added into the regression equation as the error adjustment term (see

Table 6. Heckman’s two-stage test.

	Model 16	Model 17	Model 18	Model 19	Model 20
Gender	0.450 ***	0.450 ***	0.450 ***	0.450 ***	0.450 ***
	(0.056)	(0.056)	(0.056)	(0.056)	(0.056)
Age	−0.017 ***	−0.017 ***	−0.017 ***	−0.017 ***	−0.017 ***
	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)
Marital status	1.165 ***	1.166 ***	1.167 ***	1.165 ***	1.168 ***
	(0.095)	(0.095)	(0.095)	(0.095)	(0.095)
Skills	−0.070	−0.070	−0.071	−0.070	−0.069
	(0.048)	(0.048)	(0.048)	(0.048)	(0.048)
Network	0.002 ***	0.002 ***	0.002 ***	0.002 ***	0.002 ***
	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)
Family size	−0.054 **	−0.055 **	−0.054 **	−0.054 **	−0.054 **
	(0.025)	(0.025)	(0.025)	(0.025)	(0.025)
Cadre proportion	0.142	0.141	0.136	0.142	0.143
	(0.166)	(0.166)	(0.166)	(0.166)	(0.166)
Male proportion	−0.061	−0.062	−0.063	−0.060	−0.064
	(0.105)	(0.105)	(0.105)	(0.105)	(0.105)
Single-parent family	−0.170	−0.170	−0.172	−0.170	−0.171
	(0.127)	(0.127)	(0.127)	(0.127)	(0.127)
Cultivated land per capita	0.015 ***	0.015 ***	0.015 ***	0.015 ***	0.015 ***
	(0.004)	(0.004)	(0.004)	(0.004)	(0.004)
Per capita income of family	0.027 ***	0.027 ***	0.027 ***	0.026 ***	0.027 ***
	(0.008)	(0.008)	(0.008)	(0.008)	(0.008)
Entrepreneurship atmosphere	0.026 ***	0.026 ***	0.026 ***	0.026 ***	0.026 ***
	(0.003)	(0.003)	(0.003)	(0.003)	(0.003)
Per capita income of village	−0.340 ***	−0.340 ***	−0.343 ***	−0.341 ***	−0.348 ***
	(0.130)	(0.130)	(0.130)	(0.130)	(0.130)
Number of permanent resident households	0.028	0.028	0.027	0.028	0.028
	(0.027)	(0.027)	(0.027)	(0.027)	(0.027)
Number of companies	−0.003 *	−0.003 *	−0.003 *	−0.003 *	−0.003 *
	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)
Number of migrant households	−0.043 ***	−0.043 ***	−0.043 ***	−0.044 ***	−0.044 ***
	(0.009)	(0.009)	(0.009)	(0.009)	(0.009)
Number of families involved in entrepreneurial activities	−0.000	−0.000	−0.000	−0.000	−0.000
	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)
Marketization index	−0.025 **	−0.025 **	−0.024 *	−0.026 *	−0.025 *
	(0.013)	(0.013)	(0.013)	(0.014)	(0.014)
IMR	−1.849 ***	−1.849 ***	−1.847 ***	−1.850 ***	−1.841 ***
	(0.223)	(0.223)	(0.223)	(0.223)	(0.224)
Family social status	0.407 ***	0.408 ***	0.407 ***	0.408 ***	0.406 ***
	(0.031)	(0.031)	(0.031)	(0.031)	(0.031)
Regional policy uncertainty		0.015	−0.002
		(0.088)	(0.089)
Family social status×Regional policy uncertainty			0.224 *
			(0.121)
Local family culture				−0.015	−0.045
				(0.085)	(0.085)
Family social status×Local family culture					0.303 ***
					(0.101)
_cons	−0.825 *	−0.827 *	−0.828 *	−0.805	−0.777
	(0.490)	(0.490)	(0.490)	(0.503)	(0.503)
Chi2	2675.956	2675.987	2679.518	2675.987	2685.021
R2_p	0.165	0.165	0.165	0.165	0.165
n	13,648	13,648	13,648	13,648	13,648

Note: *** p <0.01, ** p <0.05, * p <0.1; standard error in brackets.

).

Model 16 shows a significant positive relationship between family social status and farmer entrepreneurship (beta = 0.407, p < 0.01), indicating that higher family social status will boost farmers’ motivation to undertake entrepreneurial activities. Model 18 and Model 20 show that the uncertainty of regional policies and local family culture will strengthen the promoting effect of family social status on farmer entrepreneurship (beta = 0.224, p < 0.1; beta = 0.303, p < 0.01). The moderating effects of the institutional environment on the relationship between family social status and farmer entrepreneurship are further verified.

4.5. Robustness Check

We conduct several tests for a robustness check. First, because the dependent variable is binary in our study, both the probit model and logistic regression could be used to describe the effect of family social status. Thus, we apply the probit model for the robustness check, and the above conclusions still hold. Second, because the sample datasets are of large size, we use the linear probability model (LPM) to further test the reliability of our conclusions, and all the results are supported.

5. Conclusions

In a transition economy, which is also known as a state-controlled economy [145], entrepreneurs are the main driving force for economic growth. Their dynamic entrepreneurial behaviors help create jobs, supply goods, mobilize savings, and produce real welfare gains for the country [146]. Farmer entrepreneurs are particularly important for transition economies with large rural populations [147]. As an integral part of the agricultural sector, farmer entrepreneurship could not only increase farmers’ incomes but also improve their standard of living [148]. More importantly, farmer entrepreneurs are the major contributor to rural poverty alleviation and economic revitalization [149,150,151]. Despite their important role in promoting economic development, entrepreneurs have to act in insufficiently formal institutions fraught with institutional voids. One salient feature of controlled economies is that government power is predominant, accompanying a deficient market intermediary and a weak legal system [152,153,154,155]. Entrepreneurs have to face formal institutional challenges in transition economies, including insecure property rights and an inadequate legal framework. On that condition, informal family social status becomes quite crucial in farmers’ entrepreneurial activities [47].

Based on data from the Thousand Villages Survey in China, we analyzed the relationship between family social status and farmer entrepreneurship in this study and examined the moderating effects of regional policy uncertainty and local family culture from both formal and informal institutional aspects. The results show that (1) high family social status can stimulate farmers’ enthusiasm to start new businesses, i.e., with an improvement in family social status, the probability of a farmer undertaking entrepreneurial behavior becomes higher; (2) formal institutional factors, such as regional policy uncertainty, moderate the relationship between family social status and farmer entrepreneurship. That is, with the fluctuation in local policies, the resource effect and safety net effect of family social status become more prominent, thus strengthening farmers’ entrepreneurial initiative; (3) informal institutional factors, such as local family culture, play a moderating role on the entrepreneurial effect of family social status. That is, in regions with strong family culture, family cohesion and support not only increase farmers’ entrepreneurship resources but also improve their entrepreneurship enthusiasm and optimism, thus strengthening the influence of family social status on farmers’ likelihood of undertaking risky entrepreneurial behaviors.

This research has several theoretical contributions. First, by integrating social class theory and entrepreneurship theory, this study brings the sociology class construct into a business model and provides new insights into why and how family social status influences farmers’ career choices and entrepreneurial decisions. We point out the importance of family-level demographic characteristics by analyzing how family features affect farmers’ perceptions and entrepreneurial behavior, which expands the antecedents of farmer entrepreneurship. Second, this study contributes to research on social class by extending the interpretation dimensions of its influence on farmer entrepreneurship. Despite the increasing focus on this topic, researchers’ interpretation of how social class impacts farmers’ risk preferences is still limited. Our work helps to verify the influence of social class on farmer entrepreneurship, not only from a resource and network perspective but also by examining the peculiar psychological effect. Third, this study further explores the contingent factors between family social status and farmer entrepreneurship from the institutional perspective. Both formal institutional factors (regional policy uncertainty) and informal institutional factors (local family culture) are included in the research framework, which has theoretical implications for the development of the farmer entrepreneurship theoretical paradigm.

This research also has important practical contributions by offering recommendations for policymakers. Farmer entrepreneurship is testified to be an effective way to promote rural economic development, and its poverty alleviation effect has been widely recognized by previous studies. This research shows that farmers from high social status families are the main force of rural entrepreneurship, while it is difficult for farmers with low family social status to access the network and resources owned by upper-class social groups, which restricts their enthusiasm to engage in risky entrepreneurship. In order to better stimulate mass entrepreneurship, policymakers should, therefore, focus on interventions that support farmers with low family social status through advice, resources, and networks, help them expand the ways of financial support, and create better information consulting, legal, accounting, and entrepreneurial training services. From the risky assessment in the early stage to the entrepreneurial insurance and failure relief in the later period, a series of governmental supportive policies should be made to stimulate the entrepreneurial willingness and risk-bearing capacity of farmers with low family social status.

This paper also points out several directions for future research. First, there are different modes of entrepreneurship, such as opportunity entrepreneurship and survival entrepreneurship, and one interesting area for future research is the examination of whether family social status has different effects on different modes of farmer entrepreneurship. Second, our work shows the positive effect of social class on farmer entrepreneurship, but the potential mechanism is not provided. We presume that psychological safety and resource support might act as important intermediary roles, while, due to the unavailability of data, it could not be verified in this study. Future research could further explore the mechanism from the perspectives of psychological characteristics and resource support, investigating how social class promotes farmer entrepreneurship. Third, we analyzed the contingency effect of regional policy uncertainty, while agricultural policies, regulations, and regulatory documents that are specific for the agricultural sector might produce a more direct influence for farmers and enterprises in rural areas. Future research could find ways to collect the relevant data and investigate the impact of the specific agricultural policy measures on family social status and farmer entrepreneurship. Finally, external validity of this study is ensured by utilizing samples from different locations and industries, but we might not be able to generalize the research conclusions to other transition countries. Future research should continue to collect data from other transition economies to verify the effectiveness of this model and increase the external validity of the research conclusions.