An Investigation into the Mechanism of Government Embedment and Organizational Environment Influencing Farmers’ Credible Commitment in Regard to the Collective Governance of Rural Residential Land

Abstract

The purpose of this study is to investigate the influence of Government Embedment and Organizational Environment on farmers’ Credible Commitment in regard to the Collective Governance of Rural Residential Land and to inform decision-making processes that promote farmers’ compliance with Rural Residential Land management regulations and enhance the effectiveness of the Collective Governance of Rural Residential Land. The research employs various models, including Oprobit, Ologit, IV-2SLS, and moderated mediation. The research results show that (1) both Government Embedment and the Cultural Environment within the Organizational Context significantly enhance Farmers’ Credible Commitment in regard to the Collective Governance of Rural Residential Land. Notably, the quality of the Cultural Environment, when improved by Government Embedment, boosts this commitment further. (2) Endogeneity issues can lead to an overestimation of the effects of these factors. (3) Punishment intensity only positively moderates the impact of the Cultural Environment on Farmers’ Credible Commitment. (4) The moderating effect of punishment intensity correlates positively with its strength, exhibiting a threshold value. In conclusion, our research underscores the need to establish a robust regulatory framework for Rural Residential Land use, foster a conducive village cultural atmosphere, and consistently enhance the punishment intensity for illegal Rural Residential Land use. These measures will encourage farmers to adhere to Credible Commitment in regard to Rural Residential Land use, thereby facilitating the effective implementation of Collective Governance over Rural Residential Land.

1. Introduction

With the implementation of the rural revitalization strategy, the level of rural development has been significantly improved, and Rural Residential Land has become an important resource in the rural revitalization strategy, which provides important support for the in-depth promotion of the rural revitalization strategy. However, the ensuing problems, such as over-occupation of Rural Residential Land utilization, cannot be limited, should be approved but cannot be approved, cannot be given for free, cannot be stopped or transferred, cannot be managed well in violation of the law, and cannot be controlled by the number of problems highlighted in [1]. In addition, the villagers illegal occupation of arable land to build houses and conduct private transfer transactions on Rural Residential Land is not conducive to the fair, efficient, and sustainable use of Rural Residential Land. For this reason, China has carried out a series of Rural Residential Land system reform explorations, and, in 2015 and 2020, the country launched two rounds of Rural Residential Land reform pilots, allowing localities to take the approach of “crossing the river by touching the stones” while exploring and innovating a more efficient Rural Residential Land governance and fully solving the problem of Rural Residential Land utilization. Jiangxi Yujiang and other places with Rural Residential Land reform experience show that the governance of Rural Residential Land is essentially a multi-level governance of public pond resources with strong closed characteristics, and the key lies in strengthening the capacity of grass-roots self-organization and its sustainability [2], fully activating the villagers’ self-governance [3], and realizing the collective governance of Rural Residential Land. The core of Rural Residential Land collective governance is to improve the collective governance capacity of Rural Residential Land in order to realize collective action, the most critical element of which is to solve the problem of “people” in rural society [4], i.e., how to carry out the collective governance of Rural Residential Land by leaders and the public. Rural Residential Land collective governance relies on villagers to trust each other and reach a consensus on the harmonious use of Rural Residential Land, which in essence is to solve the problem of Credible Commitment among villagers in rural society.

Credible Commitment is one of the three main challenges to achieving collective governance of public pond resources proposed by Ostrom, referring to how a group of interdependent people can organize themselves to govern themselves so as to achieve lasting mutual gains in the face of the temptation of free-riding, avoiding responsibility, and other opportunistic behaviors [5], i.e., the members of the organization agree with each other that “if you keep the promise, I will keep the promise”. Credible Commitment in regard to collective governance of Rural Residential Land refers to the villagers utilizing cooperation and common compliance to maintain the order of Rural Residential Land, and the violation of Credible Commitment by members of the organization is regarded as “free-riding” in the collective action of public goods provision [6]. In short, Credible Commitment means that villager A complies with the regulations on the management of Rural Residential Land and villager B also complies with the regulations on the management of Rural Residential Land. However, Credible Commitment is a concept in game theory which involves both the first and the second actors, and the game is about whether the first actor should believe in the behavior of the second actor and, based on the behavior of the first actor, what kind of behavior the second actor takes. Therefore, Credible Commitment in regard to Rural Residential Land collective governance includes three processes, i.e., affective commitment, normative commitment, and continuous commitment, namely, whether villager X trusts other villagers, whether villager X is willing to comply with Rural Residential Land management regulations, and whether villager X is still willing to comply with Rural Residential Land management regulations in the event of other villagers’ violation of Rural Residential Land management regulations. At present, there are fewer studies on the factors influencing the Credible Commitment of collective governance of Rural Residential Land, but some progress has been made on the research related to the factors influencing the Credible Commitment of public goods governance. It is found that individual non-compliance with Credible Commitment will lead to a decline in the overall level of Credible Commitment of organizational members, which in turn undermines the organizational health structure and leads to a decline in the ability to supply collective public goods [6]. A perfect monitoring mechanism and incentive mechanism can effectively improve the level of Credible Commitment of organizational members [7] and enhance the collective action capacity. An Organizational Environment involves the rules of the environment in which economic activities take place, governing the production, exchange, and distribution of resources [8]. The promise-keeping behavior of organizational individuals is not only limited by material conditions in the Organizational Environment [9] but also influenced by specific social relationship networks such as interpersonal trust, culture, and reputation [10]. Specifically, the pattern of social connections constructed through the relational networks of social members determines the degree of interdependence and the way people interact with each other [11]. In addition, the government is embedded in rural governance in the form of authority, institutions, and resources [12] and ideas [13,14], which helps to enhance villagers’ trust in each other and build a mechanism for collective governance action on Rural Residential Land. Some of the literature has also analyzed the impact of contractual rules and group norms on Credible Commitments. Du Yanqiang et al. found that resource mobilization, relative deprivation, and group sentiment can promote farmers from action dilemmas to proactive participation [15], while Zhang Ping et al. argued that the combination of group norms, such as “entrepreneurial” leadership, governance rules, and the spirit of joint participation, can help promote individuals’ compliance with Credible Commitments [16] and realize collective action in shared resource governance.

The above studies provide useful references for this paper, but there are still some shortcomings: (1) Existing studies affirm the influence of Government Embedment and Organizational Environment on the Credible Commitment of organizational members but lack a systematic analysis of them, which makes it difficult to reveal the relationship between Government Embedment, organizational environment, and Credible Commitment under the multi-level governance of public pond resources. (2) Existing studies on the analysis of the Credible Commitment of public pond resource governance are mainly theoretical studies, lacking empirical level analysis and supporting evidence. Therefore, this paper draws on the institutional analysis and development framework to construct the analytical framework of “Government Embedment-Organizational·Environment-Credible Commitment to collective governance of Rural Residential Land” to explore the role of Government Embedment and Organizational Environment in the Credible Commitment to collective governance of Rural Residential Land and to test the role of penalty and punishment. We investigate the mechanism of Government Embedment and Organizational Environment on Credible Commitment regarding collective governance of Rural Residential Land and test the moderating role of punishment, which is of great theoretical and practical significance, in promoting the effective implementation of collective governance of Rural Residential Land.

2. Theoretical Analysis and Research Hypotheses

2.1. The Impact of Government Embedment on Credible Commitment to Collective Governance of Rural Residential Land

External coercion is effective in solving the problem of Credible Commitment to the use of public ponds [17], and an externally imposed regulation is indispensable for the governance of public pond resources [18]. The state’s involvement in rural governance is essential and provides an important external regulation and elemental support for the collective governance of Rural Residential Land, and the long-term embeddedness of the state’s administrative power in the villages since the founding of the PRC has formed the rural order pattern of “rural politics” [19]. From a detailed point of view, the system supply, resource input, supervision, and management of Government Embedment is an important guarantee for the effective operation of Rural Residential Land collective governance [12]. In essence, Government Embedment involves establishing the authority of the village Rural Residential Land management system and strengthening the supervision of Rural Residential Land utilization in order to promote the administrative power of the Rural Residential Land collective governance of the socialization of the generation of capital to play an important role in the capital [20], and to promote the villagers to perform the Rural Residential Land utilization provision commitments. In addition, the government’s institutional embedment can also promote external governance resources, rules and subjects to be embedded in the grassroots governance system by setting up normative procedures and recreating self-governance institutions, etc., to enhance the grassroots governance capacity [21] and to strengthen villagers’ promise-keeping behavior. Based on this, this paper proposes Hypothesis 1:

Hypothesis 1.

Government Embedment can increase the level of farmers’ Credible Commitment to the collective governance of Rural Residential Land.

2.2. The Impact of Organizational Environment on the Credible Commitment to Collective Governance of Rural Residential Land

Organizational Environment can be accepted and perceived by organizational members and described through characteristic values [22], which can be disassembled into the four dimensions of ecology, social system, culture, and context from the perspective of organizational climate [23]. This paper is based on Rural Residential Land collective governance practice and the Organizational Environment has been classified. The Organizational Environment in Rural Residential Land collective governance is divided into the institutional environment, resource environment, technological environment and Cultural Environment, and a further analysis of its impact on Credible Commitment to the collective governance of Rural Residential Land has been conducted. Combined with the practice of Rural Residential Land governance, Rural Residential Land resource conditions, rural community characteristics, and rule design are important institutional safeguards for the collective governance of Rural Residential Land, and the combination of selection rules, scope rules, and rural social networks can motivate farmers to actively participate in Rural Residential Land governance, constrain the opportunistic behaviors such as “nail households” and unauthorized structures [2], and promote villagers’ compliance with collective Rural Residential Land management regulations. The influence of rural social capital on the Credible Commitment of Rural Residential Land governance should not be ignored. On the one hand, the tradition of reciprocity and cooperation rooted in the acquaintance society and the reputation effect prevent and dissolve all kinds of opportunistic behaviors that may hinder the collective governance of Rural Residential Land [24,25]; on the other hand, the accumulation of social capital in rural areas and the members of the organization with the role of leadership can effectively reduce the cost of organizing, coordinating, and executing the Rural Residential Land sorting and solve the collective action dilemmas such as “hitchhiking”, “free-riding”, and so on [26,27]. Based on this, this paper proposes Hypothesis 2:

Hypothesis 2.

Organizational Environment can enhance the level of farmers’ Credible Commitment to the collective governance of Rural Residential Land.

2.3. The Mediating Role of Organizational Environment

Under the background of rural revitalization strategy, the rural governance model of “administrative embeddedness + rural autonomy” has become a highly efficient method of governance, creating a new pattern of grassroots social governance [28]. The government embedded in the governance of Rural Residential Land is actually the government and the village collectives cooperating in the governance of Rural Residential Land, the effective interaction between the embedded government and the rural autonomy can give full play to the positive driving role of the government, stimulate the endogenous motivation of the main body of self-government, and enhance the effectiveness of rural governance [13]. Embedded autonomy activates grassroots organizations through the delegation of supervisory power and authority, centralization of governance objectives, and decentralization of governance responsibilities [29], i.e., Government Embedment improves the supervision and management of the order and utilization of resources in villages by giving the village collectives the corresponding authority and implementing the corresponding management responsibilities, encouraging the villagers to comply with the regulations of Rural Residential Land utilization. In detail, Government Embedment can significantly improve the governance capacity of grassroots organizations, effectively solve the second-order dilemma of institutional provision, and leverage the potential of “self-governance” [30] to realize the provision of public affairs in villages. Based on the above analysis, this paper puts forward Hypothesis 3:

Hypothesis 3.

Organizational Environment mediates the effect of Government Embedment on the Credible Commitment to collective governance of Rural Residential Land.

2.4. The Moderating Effect of Punishment Intensity

The effects of Government Embedment and Organizational Environment on the Credible Commitment to the collective governance of Rural Residential Land also depend on the strength of punishment for non-compliant Rural Residential Land utilization. In Rural Residential Land governance, the punishment mechanism is a positive incentive for collective action and a constraint on “free-riding” behavior. When the sanctions for violations are more severe, collective members are more fearful of the consequences of violating Rural Residential Land regulations and may be more willing to comply with the Credible Commitment and contribute a certain amount of time, money, and psychological commitment [31]. Meanwhile, in public goods provision, the punishment mechanism can lead to the maximization of collective interests, increase the average benefit of the collective, and enhance the level of individual cooperation [32], and whether the punishment mechanism can play an effective role in promoting cooperation mainly depends on the adequacy of punishment [33]. Generally speaking, the stronger the punishment is, the higher the price paid by villagers for violating Rural Residential Land management regulations, which in turn enhances the restraining effect on Rural Residential Land utilization. Based on this, research Hypothesis 4 and research Hypothesis 5 are proposed:

Hypothesis 4.

Penalty intensity significantly moderates the relationship between Government Embedment and Credible Commitment in regard to the collective governance of Rural Residential Land.

Hypothesis 5.

Penalty intensity significantly moderates the relationship between Organizational Environment and Credible Commitment in regard to the collective governance of Rural Residential Land.

The five theoretical hypotheses above constitute the analytical framework of this paper, as shown in Figure 1. Government Embedment can have both a direct positive effect on the level of farmers’ Credible Commitment to collective governance of Rural Residential Land and an indirect positive effect on the collective governance of Rural Residential Land through the sub-dimensions of the Organizational Environment; the intensity of punishment has a positive moderating effect on the relationship between both Government Embedment and the Organizational Environment in regard to the level of farmers’ Credible Commitment to the collective governance of Rural Residential Land.

3. Data and Methods

3.1. Data Sources

The data used in this paper come from a field survey conducted by the authors and their research team in July–August 2023 in Sichuan Province. The team selected Lu County, PI District, and Chongzhou City as the research areas in accordance with the Rural Residential Land Reform pilot, the Operational Construction Land in the Market Pilot and the Control Sample, respectively, in order to ensure the representativeness of the selected samples, as shown in Figure 2. In the three sample counties (cities, districts), according to the distance from the center of the county, from near to far, they randomly selected three townships (towns), and in the selected townships (towns), each township randomly selected two sample villages (notably, in the township of Lu County, the number of surveyed households in the selected village of Fangdong was insufficient, and therefore three sample villages were selected). The research covered 3 counties (cities and districts), 9 townships (towns), and 19 villages (communities), and it collected data from 473 households, of which 450 were valid samples, with a questionnaire validity rate of 95.14%.

3.2. Methods

Selection of Model Variables

Referring to the above studies related to public pond resource governance and China’s Rural Residential Land governance practice, this paper selects the level of farmers’ Credible Commitment as the explained variable, Government Embedment and Organizational Environment as the core explanatory variables, and the intensity of punishment as the moderating variable. In order to control other factors affecting the level of farmers’ Credible Commitment, this paper adds head of household characteristics, head of household family characteristics, characteristics of Rural Residential Land use, satisfaction with the living environment, labor mobility, and collective size as control variables in the model.

(1)

Explained variables

The explained variable in this paper is the level of farmers’ Credible Commitment in regard to Rural Residential Land governance. Credible Commitment is an important concept in the theory of autonomous governance which involves the behavioral game between organizational members, including whether they believe other organizational members, their established behavioral decisions, and the behavioral decisions taken with they are aware of the behavioral decisions of other organizational members, which can be measured from three aspects: affective, normative and sustained commitment [34]. First, in regard to affective commitment, the question is set as “Do you believe that other villagers will comply with the regulations on the utilization of Rural Residential Land”? The second aspect is normative commitment, which mainly involves understanding whether the farmers behavior in the utilizing of Rural Residential Land is appropriate and involves elements such as multi-family residences, exceeding the allotted size, and illegal construction (without directly asking the villagers, and judging through the basic information survey of the questionnaire). The third element is continuance commitment, which is set as “Are you willing to continue to abide by the regulations on the management of the utilization of the Rural Residential Land if other people violate the regulations?”. The three answer options are set as “0 = no; 1 = yes”, and the results of the three questions are summed up to obtain the level of the farmers’ Credible Commitment, which may take the values of 0, 1, 2, and 3, and belongs to the ordered categorical variables.

(2)

Core explanatory variables

The core independent variables in this paper are Government Embedment and Organizational Environment. Referring to the interpretation of Organizational Environment from the perspective of organizational atmosphere [23], this paper decomposes the Organizational Environment into Institutional Environment, Technological Environment, Resource Environment, and Cultural Environment, and the above variables are evaluated by the entropy weight method to obtain the corresponding values.

Referring to the Government Embedment research, this paper analyzes the system coverage, management scope [35], villagers’ participation, system awareness and system effect, the number of management personnel [36], the rationality of management structure, the number of consultations, the quality of management work and community relations [37], the number of introduced projects, the amount of invested funds, the applicability of platforms, the effect of introduced projects and effect of invested funds [13,38], publicity and training [39], the number of participating villagers, acceptance of publicity and training, the participation rates of villagers, and the quality of publicity. The number of times, the villagers’ participation rates, the propaganda acceptance, and the villagers’ recognition [39,40,41,42] are used to evaluate the Government Embedment.

Referring to the related studies on Organizational Environment, this paper evaluates the Institutional Environment in terms of institutional clarity, institutional authority, institutional rationality, institutional fairness, and institutional design openness [42,43] and evaluates the Technological Environment in terms of collective leadership capacity, number of village sages, governance methods, and governmental and non-governmental helping capacity [44,45,46]. The paper also evaluates the Resource Environment in terms of resource utilization order, Rural Residential Land reserves, Rural Residential Land value, and the collective economic level to evaluate the resource environment, while also evaluating the Cultural Environment in terms of clan atmosphere, villager relationships, and social network strength [47,48].

(3)

Moderating variable

The moderating variable in this paper is the intensity of punishment. The question is set as “How severe do you think the government punishes illegal Rural Residential Land use in the village?” The options available to farmers are “1 = very not severe; 2 = not severe; 3 = average; 4 = quite severe; 5 = very severe”, with a positive integer value between [1,5].

(4)

Control variables

Referring to the theory of autonomous governance and the research related to collective action in the governance of rural public things, the characteristics of the head of the household, the characteristics of the family, the characteristics of the utilization of the Rural Residential Land, the satisfaction of the living environment, the mobility of labor, and the size of the village are the important factors affecting the level of the Credible Commitment of the farmers in regard to the collective governance of the Rural Residential Land. The assignment of model variables and descriptive statistical analysis are detailed in

Table 1. Variable definition assignment and descriptive statistics.

Variable	Definition and Measure	Mean	SD
Explained variable
Level of Credible Commitment	Summed from three sub-indicators	2.049	0.785
Emotional commitment	Do you believe that other villagers will comply with the regulations on the utilization of residential land? 0 = No; 1 = Yes	0.651	0.477
Normative commitments	Does your family’s utilization of residential land include multiple dwellings, over-sized areas and illegal structures? 0 = No; 1 = Yes	0.756	0.430
Continuing commitment	Would you still be willing to continue to comply with the regulations governing the use of Rural Residential Lands if they were violated by others? 0 = No; 1 = Yes	0.642	0.480
Core explanatory variables
Government Embedment	Obtained by entropy weighting method for comprehensive evaluation of relevant indicators	0.736	0.051
Institutional Environment	Obtained by entropy weighting method for comprehensive evaluation of relevant indicators	16.318	2.033
Technical Environment	Obtained by entropy weighting method for comprehensive evaluation of relevant indicators	12.845	1.402
Resource Environment	Obtained by entropy weighting method for comprehensive evaluation of relevant indicators	30.586	4.113
Cultural Environment	Obtained by entropy weighting method for comprehensive evaluation of relevant indicators	6.148	0.751
Characteristics of the head of household
Sex	0 = female; 1 = male	0.776	0.418
Age	year	57.687	11.060
Educational attainment	1 = Elementary school and below; 2 = Middle school; 3 = High school and secondary school; 4 = College and above	1.802	0.830
Family characteristics
Annual household income	Yuan (take logarithm)	10.844	0.953
Number of laborers	people	3.102	1.194
Presence of village officials at home	0 = No; 1 = Yes	0.184	0.388
Availability of party members	0 = No; 1 = Yes	0.316	0.465
Whether or not arable land is transferred	0 = No; 1 = Yes	0.487	0.500
Characteristics of Rural Residential Land use
Number of Rural Residential Land plots	How many Rural Residential Lands does your family own? 0 = 1; 1 = 2 or more	0.073	0.261
Situation of unused Rural Residential Land	Is your Rural Residential Land unused? 0 = No; 1 = Yes	0.060	0.238
Construction costs	ten thousand dollars	17.078	17.776
Per capita of Rural Residential Land	square meter	47.519	29.764
Whether or not the right is established	Is your family’s Rural Residential Land titled? 0 = No; 1 = Yes	0.804	0.397
Satisfaction with the living environment	How satisfied are you with your current living environment? 1 = Very dissatisfied; 2 = Unsatisfied; 3 = Average; 4 = Quite satisfied; 5 = Very satisfied	3.916	0.788
Labor mobility	Village labor mobility is ? 0 = inflow; 1 = outflow?	0.884	0.320
Village size	How many households reside in the village? 1 = Up to 500; 2 = [500, 1000); 3 = [1000, 1500); 4 = 1500 and above	1.904	1.009
Moderator variable
Intensity of punishment	How strong is the government’s punishment for non-compliance with the use of Rural Residential Lands? 1 = very low severity; 2 = low severity; 3 = average; 4 = relatively severe; 5 = very severe	3.733	0.823

.

3.3. The Models

(1)

Oprobit model and OL model

The level of Credible Commitment may take the values of 0, 1, 2, 3, corresponding to no Credible Commitment, low Credible Commitment, medium Credible Commitment, and high Credible Commitment, respectively, and the larger the value, the higher the level of Credible Commitment of the farmers, which is an ordered categorical variable. Green gave a comprehensive summary of the thinking framework for regression response analysis of this type of data as a class of discrete choice model, which is called the “Oprobit model” [49]. Therefore, the Oprobit model is selected to empirically analyze the effects of Government Embedment and Organizational Environment on the level of farmers’ Credible Commitment, and the Ordinal Logistic (OL) regression model is adopted to conduct the robustness test. The Oprobit model has the same settings as the Ordinal Logistic model, and the specific model settings are as follows:

$$P_j=a_j+\alpha Go{v}_{\mathrm{j}}+\beta Or{g}_j+{\displaystyle {\sum }_{i=1}^n{\lambda }_{ij}Co{n}_{ij}+{\mu }_j}$$

(1)

In Equation (1), $P_j$ is the level of Credible Commitment in regrd to the collective governance of Rural Residential Land, $Go{v}_j$ is the core explanatory variable Government Embedment, $Or{g}_j$ is the core explanatory variable Organizational Environment, $Co{n}_{ij}$ is a control variable, $j$ and $i$ denote the $i$ control variable for the $j$ sample, ${\mu }_j$ is a random error term obeying a standard normal distribution, and $a_j,\alpha ,\beta ,{\lambda }_{\mathrm{ij}}$ is the coefficient to be estimated.

(2)

Moderating effects model

Before conducting tests of moderating effects, the level of the independent and moderating variables should be clarified. The core independent variables in this chapter are Government Embedment and Organizational Environment, which are continuous variables. The moderating variable is the intensity of punishment, the value range is an integer between [1,5], and the value interval is uniform. According to Wen Zhonglin’s research [50], the value of the punishment strength can be approximated as a continuous variable, and the regression model with a product term is selected to do a hierarchical regression analysis in order to test the interaction between the governmental embedment, the Organizational Environment, and the intensity of punishment, and then to judge the intensity of punishment in the governmental embedment and the Organizational Environment influencing the level of Credible Commitment in regard to the collective governance of Rural Residential Land. The specific model settings are as follows:

$$P_j=a_j+\alpha Go{v}_j+\beta Or{g}_j+\gamma Pu{l}_j+{\mu }_j$$

(2)

$$P_j=b_j+\overline{\alpha }Go{v}_j+\overline{\beta }Or{g}_j+\overline{\gamma }Pu{l}_j+\varphi Go{v}_{j}Pu{l}_j+\phi Or{g}_{j}Pu{l}_j+{\sigma }_j$$

(3)

The judgment criteria for the moderating effect are (1) the moderating effect is significant if the $R_2^2$ measured by the regression of Equation (3) is greater than the $R_1^2$ measured by the regression of Equation (2); (2) if the coefficients of the direct effects and moderating effects are significant, the moderating effect is significant.

(3)

Mediating effects model

Government Embedment not only has a direct impact on the level of Credible Commitment regarding the collective governance of Rural Residential Land but may also have an indirect impact on it through the Organizational Environment. Referring to Wen Zhonglin’s testing process of mediating effect, the mediating effect testing model set in this study is as follows:

$$P=a_0+{\alpha }_0{X}_{gov}+{\beta }_0{K}_{con}+{\mu }_1$$

(4)

$$M_{org}=a_1+{\alpha }_1{X}_{gov}+{\beta }_1{K}_{con}+{\mu }_2$$

(5)

$$P=a_2+{\alpha }_2{X}_{gov}+\lambda M_{org}+{\beta }_2{K}_{con}+{\mu }_3$$

(6)

In Equations (4)–(6), the coefficient ${\alpha }_0$ represents the total effect of the independent variable $X_{gov}$ on the dependent variable $P$, the coefficient ${\alpha }_1$ represents the effect of the independent variable $X_{gov}$ on the mediator variable $M_{org}$, the coefficient $\lambda$ is the effect of the mediator variable $M_{org}$ on the dependent variable $P$ with the addition of control variable $K_{con}$ and independent variable $X_{gov}$, the coefficient ${\alpha }_2$ is the direct effect of the independent variable $X_{gov}$ on the dependent variable $P$ with the addition of control variable $K_{con}$ and mediator $M_{org}$, and ${\mu }_1,{\mu }_2,{\mu }_3$ are the regression residuals. In the classical linear regression model, the mediating effect is equal to the indirect effect, i.e., the product of ${\alpha }_1$ and $\lambda$, and the total effect ${\alpha }_0$ is equal to the direct effect ${\alpha }_2$ + indirect effect $\lambda {\alpha }_1$:

$${\alpha }_0={\alpha }_2+\lambda {\alpha }_1$$

(7)

4. Results

4.1. Baseline Regression

Before the regression analysis of the data, the multicollinearity test is conducted first, and the test results show that the variance inflation factor (VIF) of each variable is significantly less than 10, which can exclude the multicollinearity between the variables. The results of the empirical analysis of the impact of Government Embedment and Organizational Environment on the level of farmers’ Credible Commitment regarding the collective governance of Rural Residential Land are shown in

Table 2. The Impact of Government Embedment and Organizational Environment on Credible Commitment in Regard to Collective Governance of Rural Residential Land.

Variable	Oprobit		Ologit
	(1)	(2)	(3)	(4)
Government Embedment	2.791 ** (1.186)		4.715 ** (2.017)
Institutional Environment		0.000 (0.031)		0.008 (0.053)
Technical Environment		−0.078 (0.049)		−0.138 * (0.083)
Resource Environment		−0.012 (0.015)		−0.022 (0.025)
Cultural Environment		0.281 *** (0.089)		0.463 *** (0.152)
Sex	−0.186 (0.132)	−0.194 (0.133)	−0.304 (0.222)	−0.334 (0.224)
Age	−0.004 (0.006)	−0.002 (0.006)	−0.008 (0.010)	−0.003 (0.010)
Educational attainment	−0.177 ** (0.079)	−0.154 * (0.080)	−0.298 ** (0.134)	−0.251 * (0.135)
Annual household income	−0.106 * (0.059)	−0.126 ** (0.060)	−0.203 * (0.101)	−0.231 ** (0.102)
Number of laborers	0.086 * (0.047)	0.112 ** (0.048)	0.165 * (0.081)	0.208 ** (0.082)
Presence of village officials at home	−0.217 (0.153)	−0.181 (0.155)	−0.404 (0.259)	−0.363 (0.261)
Availability of party members	0.160 (0.129)	0.102 (0.130)	0.260 (0.218)	0.177 (0.219)
Whether or not arable land is transferred	0.000 (0.109)	−0.017 (0.109)	−0.018 (0.184)	−0.037 (0.184)
Number of Rural Residential Land plots	0.033 (0.206)	0.127 (0.207)	0.098 (0.345)	0.237 (0.345)
Situation of unused Rural Residential Land	0.128 (0.227)	0.116 (0.228)	0.240 (0.376)	0.221 (0.378)
Construction costs	0.008 ** (0.003)	0.007 ** (0.003)	0.013 ** (0.005)	0.013 ** (0.005)
Per capita of Rural Residential Land	0.000 (0.002)	0.001 (0.002)	0.001 (0.003)	0.002 (0.003)
Whether or not the right is established	−0.121 (0.140)	−0.089 (0.141)	−0.167 (0.232)	−0.118 (0.234)
Satisfaction with the living environment	0.318 *** (0.070)	0.260 *** (0.073)	0.543 *** (0.119)	0.434 *** (0.124)
Labor mobility	0.932 *** (0.190)	0.671 *** (0.178)	1.521 *** (0.330)	1.057 *** (0.309)
Village size	−0.006 (0.055)	0.013 (0.060)	−0.021 (0.093)	0.019 (0.102)
sample size	450	450	450	450
Pseudo R2	0.067	0.075	0.066	0.073
Chi-square	68.19	75.54	66.53	73.81
Prob > chi2	0.000	0.000	0.000	0.000

Note: Robust standard errors in parentheses, ***, **, and * represent significance at the 1%, 5%, and 10% statistical levels, respectively; same table below.

. Among them, the impact of Government Embedment on the level of farmers’ Credible Commitment is significant within the 5% confidence interval, and its impact coefficient is positive, indicating that Government Embedment significantly enhances the level of farmers’ Credible Commitment, and the more the government supports and intervenes in the collective governance of Rural Residential Land, the more the farmers comply with the regulations of Rural Residential Land utilization commitment. Among the Organizational Environment sub-dimensions, only the Cultural Environment has a significant effect on the level of farmers’ Credible Commitment at the 1% confidence interval, and its impact coefficient is positive, implying that the cultural Environment significantly enhances the level of farmers’ Credible Commitment, and farmers believe that the better the cultural Environment of the village, the more willing they are to comply with the Rural Residential Land utilization regulations commitment. Columns (3) and (4) use the OL model to observe whether the results change or not, and the results are found to be basically the same as the regression results of the Oprobit model, indicating that the above conclusions have good robustness.

4.2. Mediated Effects Test

As can be seen from

Table 3. Test results of mediating effect of Organizational Environment.

	(1)	(2)	(3)	(4)
Variable	Technical Environment	Cultural Environment	Credible Commitments	Credible Commitments
Government Embedment	5.734 *** (1.187)	3.420 *** (0.734)	3.082 ** (1.216)	2.151 * (1.219)
Technical Environment			−0.050 (0.045)
Cultural Environment				0.194 ** (0.077)
Control variable	Controlled	Controlled	Controlled	Controlled
Sample size	450	450	450	450
Pseudo R2	0.351	0.174	0.069	0.073
Prob > chi2	0.000	0.000	0.000	0.000
F	14.15	5.40	69.43	74.47

Note: Robust standard errors in parentheses, ***, **, and * represent significance at the 1%, 5%, and 10% statistical levels, respectively.

, the coefficient of the impact of Government Embedment on the Technological Environment in the estimation results of Model 1 is significantly positive, which verifies that Government Embedment is conducive to the improvement of the Technological Environment, and the impact of the Technological Environment on the Credible Commitment in the estimation results of Model 3 is insignificant, indicating that its mediating effect is not significant, which is probably due to the fact that the level of compliance with the Credible Commitment of the farmers’ Rural Residential Land use is more dependent on their own rights and interests in the use of Rural Residential Land and on the surrounding neighbors. The reason may be that the level of compliance with the Credible Commitment to Rural Residential Land utilization is more dependent on their own Rural Residential Land utilization rights and interests, as well as the neighboring neighbors’ Rural Residential Land utilization behaviors, and has nothing to do with the village Rural Residential Land governance capacity. The estimation results of model 2 show that the coefficient of Government Embedment on Cultural Environment is significantly positive, which verifies that Government Embedment is conducive to the improvement of Cultural Environment. The estimation results of model 4 show that the impact of Cultural Environment on Credible Commitment is significantly positive, and the impact of Government Embedment on Credible Commitment is still significantly positive, which indicates that Cultural Environment plays a part of the mediating effect.

4.3. Endogeneity Test

Farmers’ perception of Government Embedment may be endogenous due to bidirectional causality or omitted variables with the level of Credible Commitment regarding the collective governance of Rural Residential Land, and this paper adopts the instrumental variable approach to deal with the endogeneity problem. In regard to the current method of applying instrumental variables for multivalued ordinal variables, the IV-2SLS model is relatively more mature and characterized by a higher estimation accuracy, higher estimation efficiency, and a wider range of application in the presence of endogenous explanatory variables. More importantly, when the sample size is large enough, IV-2SLS is guaranteed to provide a parameter of heterogeneity and provide largely accurate and unbiased estimates. Referring to the research idea of Nyberg et al. [51], the mean value of the perception of Government Embedment among other farm households in the same village is selected as the instrumental variable. The reason is that the perception of Government Embedment is closer to that of other farmers in the same village because of the similarity in the utilization of Rural Residential Land, production and living habits, and economic development conditions, which satisfies the correlation requirement; at the same time, the mean value of the perception of Government Embedment of other farmers will not be affected by the level of farmers’ Credible Commitment, and there is no reverse causality.

The results of the endogeneity test are shown in

Table 4. Endogeneity test results.

Variable	First Stage	Second Phase
	Government Embedment	Credible Commitments	Credible Commitments
	(1)	(2)	(3)
Government Embedment	1.351 (0.925)	1.406 * (0.939)	1.454 * (0.900)
Technical Environment		−0.027 (0.027)
Cultural Environment			0.128 ** (0.050)
Constant term	0.773 (0.906)	−0.103 *** (0.032)	0.015 (0.879)
Control variable	Controlled	Controlled	Controlled
Sample size	450	450	450
R2	0.141	0.142	0.154
Wald chi2	77.95	78.51	91.06
F	106.43	116.14	139.09
p	0.000	0.000	0.000

Note: Robust standard errors in parentheses, ***, **, and * represent significance at the 1%, 5%, and 10% statistical levels, respectively.

, which tested the mediating effect of the Organizational Environment on Government Embedment affecting the level of Credible Commitment in regard to farmers and the collective governance of Rural Residential Land. Regarding the mediating effect of the Organizational Environment in general, Government Embedment on the level of Credible Commitment regarding the collective governance of Rural Residential Land is significant at the 1% level, while the coefficient of influence is positive and the mediating effect of the Organizational Environment is insignificant, which is consistent with the results of the estimation before the endogeneity treatment was carried out. Regarding the mediating effect of each sub-dimension of the Organizational Environment, the mediating effect of the Cultural Environment on the impact of Government Embedment on the level of Credible Commitment regarding the collective governance of Rural Residential Land is still significant, which is consistent with the results obtained before the endogeneity treatment. This suggests that the above estimation results are relatively robust.

4.4. Moderating Effects Test

This paper not only focuses on the effects of Government Embedment and Organizational Environment on the level of farmers’ Credible Commitment regarding the collective governance of Rural Residential Land, but also meticulously discusses the moderating role of punishment intensity in the effects of Government Embedment and Organizational Environment on the level of farmers’ Credible Commitment. The interaction term between the intensity of punishment and Government Embedment in column (1) of

Table 5. The results of moderation effect testing.

Variable	(1)	(2)
Government Embedment	2.942 ** (1.207)
Institutional Environment		−0.013 (0.032)
Technical Environment		−0.096 * (0.050)
Resource Environment		−0.005 (0.015)
Cultural Environment		0.186 * (0.099)
Intensity of punishment	0.294 *** (0.070)	0.258 *** (0.082)
Government Embedment * Punishment intensity	1.302 (1.231)
Technical Environment * Punishment intensity		−0.079 (0.054)
Cultural Environment * Punishment intensity		0.216 * (0.093)
Control variable	Controlled	Controlled
Sample size	450	450
Pseudo R2	0.086	0.091
Chi-square	87.48	91.91
Prob > chi2	0.000	0.000

Note: Robust standard errors in parentheses, ***, **, and * represent significance at the 1%, 5%, and 10% statistical levels, respectively.

is not significant, while both Government Embedment and the intensity of punishment significantly increase the level of farmers’ Credible Commitment, indicating that the intensity of punishment directly affects the level of farmers’ Credible Commitment and does not have a moderating effect on the effect of Government Embedment on the level of farmers’ Credible Commitment. The interaction term between the intensity of punishment and Cultural Environment in column (2) of Table 4 is significant at a 10% confidence interval, indicating that the intensity of punishment has a moderating effect on the effect of Cultural Environment on farmers’ Credible Commitment levels; at the same time, it is found that both Cultural Environment and the intensity of punishment can significantly enhance farmers’ Credible Commitment levels, indicating that the intensity of punishment not only directly enhances the level of farmers’ Credible Commitment but also enhances the role of the Cultural Environment. In this paper, we find that increasing the intensity of punishment can increase farmers’ fear of violations of Rural Residential Land laws and motivate them to comply with the Credible Commitment in regard to the collective of governance of Rural Residential Land, but it is difficult to compensate for the lack of government supervision in the utilization of Rural Residential Land and to weaken the calling power of collective leaders and township sages; therefore, the reinforcing effect of the intensity of punishment on the Credible Commitment of the Government Embedment and Technological Environment is not obvious.

4.5. Robustness Tests

In order to test the robustness of the estimation results of the mediating and moderating effects, the self-help method test (Bootstrap test) was conducted using SPSS 27 software, the significant level was set at 95% confidence interval, and the model used was the moderated mediating effect model. The test results are shown in

Table 6. The results of Moderated Mediation analysis.

Self-Service Testing Method	Effect	se/BootSE	t	p	LLCI/BootLLCI	ULCI/BootULCI
Direct effect	1.286	0.797	0.113	0.108	−0.281	2.853
Indirect effect	0.447	0.207	/	/	0.107	0.909

, and they indicate that the direct effect of Government Embedment on Credible Commitment in regard to the collective governance of Rural Residential Land is not significant and that the indirect effect of Cultural Environment on Credible Commitment in regard to the collective governance of Rural Residential Land is significant, illustrating that Cultural Environment has a fully mediated effect and that there is, indeed, a transmission mechanism of Government Embedment to improve the Cultural Environment of villages to improve the level of Credible Commitment of farm households in regard to collective governance of Rural Residential Land. Although the direct effect of Government Embedment is not significant, it is close to being significant at the 10% level of significance, which is basically consistent with the estimation results above, indicating that the estimation of the mediating effect by the stepwise regression method is more robust.

Table 7. The moderation effects at different levels of punishment intensity.

Variable	Severity of Punishment	Effect	BootSE	BootLLCI	BootULCI
Cultural Environment	−0.823	−0.067	0.299	−0.638	0.572
	0.000	0.277	0.241	−0.153	0.804
	0.823	0.620	0.316	0.058	1.297

shows the estimation results of the moderating effect of Punishment Intensity in the Cultural Environment on Credible Commitment in regard to the collective governance of Rural Residential Land; its moderating effect is not significant when the punishment intensity is at the mean and negative one standard deviation, and its moderating effect is significant when the punishment intensity is at the positive one standard deviation. This shows that, in the collective governance of Rural Residential Land, the positive moderating effect of punishment intensity on the level of Credible Commitment of farmers increases with the increase in the intensity of the punishment; additionally, the moderating effect of low punishment intensity is not obvious, and its moderating effect starts to be significant only when the punishment intensity reaches a certain threshold. Figure 3 provides a more intuitive display of the changes in the regulatory effect of punishment intensity and also shows the regulatory effect of punishment intensity at different levels of Cultural Environment. When the punishment intensity is low, the Cultural Environment does not have a large impact on Credible Commitment in regard to the collective governance of Rural Residential Land; when both the Cultural Environment and the punishment intensity are high, the level of Credible Commitment in farmers in regard to the collective governance of Rural Residential Land is the highest and the regulatory effect is the best.

5. Conclusions and Policy Implications

Using field research data on the collective governance of Rural Residential Land from July to August 2023, this paper tests the influence of Government Embedment and Organizational Environment on farmers’ Credible Commitment in regard to the collective governance of Rural Residential Land, the mechanism of action, and the moderating effect of punishment intensity based on the Oprobit model, using the instrumental variables method to adopt the IV- 2SLS model to verify the endogeneity of the core explanatory variables and the validity of instrumental variables before finally adopting the self-help test method to select the mediated effect model with moderation to verify the robustness of the main conclusions. The conclusions are as follows: (1) both Government Embedment and Cultural Environment can significantly increase the level of farmers’ Credible Commitment in regard to the collective governance of Rural Residential Land, and Government Embedment can improve the level of farmers’ Credible Commitment by improving the Cultural Environment. (2) Punishment intensity can both directly increase the level of farmers’ Credible Commitment and increase the positive effect of the Cultural Environment on it. (3) Low levels of punishment intensity make it difficult to increase the level of farmers’ Credible Commitment, and the level of farmers’ Credible Commitment is increased only when the punishment reaches a certain threshold value; reaching this threshold can significantly improve the level of Credible Commitment in farmers.

Under the background of rural revitalization strategy, Rural Residential Land has become an important development element of rural revitalization, and the government has undertaken many interventions. The findings of this paper are of scientific significance for understanding the logic and connotation of farmers’ Credible Commitment behavior in the collective governance of Rural Residential Land under the influence of Government Embedment and village Organizational Environments, and they are also of reference significance for the effective implementation of the collective governance of Rural Residential Land and for deepening the reform of the Rural Residential Land systems. Based on the public pond resource attribute of Rural Residential Land, the implementation of collective governance of Rural Residential Land is an inevitable choice. Farmers’ compliance with the trustworthy commitment of collective governance of Rural Residential Lands is not only closely related to their own interests in Rural Residential Lands but also a precondition for the realization of the collective governance of Rural Residential Lands, which is closely related to the governmental embeddedness and the village Organizational Environment. Based on the research findings and policy implications, the policy suggestions of this paper are as follows: first, establish a sound system of supervision and assistance for the utilization of Rural Residential Land. Strengthen the supervision and management of Rural Residential Land utilization and implement the responsibility of Rural Residential Land supervision; provide support for villages with governance conditions, such as funds, projects, and other development elements, and optimize the pattern of Rural Residential Land utilization. Second, create a good cultural atmosphere in the village. Carry out Rural Residential Land utilization related publicity, education and training, and village cultural activities to enhance villagers’ understanding of Rural Residential Land governance, promote mutual understanding and trust among villagers, and reach a consensus on the harmonious utilization of Rural Residential Land. Third, continue to strengthen the punishment of illegal Rural Residential Land utilization. Strictly examine the use of illegal Rural Residential Land and strengthen the punishment of illegal behavior to ensure that the punishment can reach the threshold value before giving full play to its deterrent effect. Fourth, strengthen the synergistic interaction between the government and village collectives. Strengthen the institutional synergy, organizational synergy, resource synergy, and conceptual synergy between the government and rural collectives in the governance of rural public affairs, promote multi-dimensional co-governance, and give full play to the positive role of the government’s administrative power in the governance of rural public affairs.