Demographic Change and Commons Governance: Examining the Impacts of Rural Out-Migration on Public Open Spaces in China Through a Social–Ecological Systems Framework

Abstract

Rapid urbanization in China has driven substantial rural population out-migration, raising concerns about its implications for the governance of land commons in villages. While existing studies have acknowledged the effects of migration on rural resource management, little attention has been paid to its influence on the self-governance of rural public open spaces (POSs). This study adopts the social–ecological systems (SES) framework to examine how rural out-migration shapes POS self-governance mechanisms. Based on survey data from 594 villagers across 198 villages in Taigu District, partial least squares structural equation modeling (PLS-SEM) and a mediation model grounded in the SES framework were employed for analysis. The results indicate that rural out-migration does not exert a direct impact on POS self-governance. Instead, it negatively influences governance outcomes through full mediation by villager organizations, the left-behind population, collective investment in POSs, and self-organizing activities. Notably, the mediating roles of the left-behind population and self-organizing activities account for 67.38% of the total effect, underscoring their critical importance. Drawing on these insights, the study proposes four policy recommendations to strengthen rural POS self-governance under conditions of demographic transition. This research contributes to the literature by being the first to incorporate an external social factor—rural out-migration—within the SES framework in the context of POS governance, thereby advancing both theoretical and practical understandings of rural commons management.

1. Introduction

Public open spaces (POSs) are defined in various ways due to their diverse functions and characteristics [1]. In the context of rural China, POSs refer to open areas within villages primarily designated for residential use. These spaces are artificially constructed and not occupied by buildings or major structures. They serve as multifunctional areas for villagers—supporting recreation, social interaction, and sports—while also fulfilling roles related to ecological landscaping, agricultural production, and disaster prevention and mitigation [2]. Rural POSs take multiple forms, including public squares, green areas, sports fields, village entrances, and waterfront zones [2]. These spaces are integrally linked to rural social life, economic activities, and ecological landscapes, influencing the sustainability of rural social–ecological systems (SESs).

To address spatial resource demands across urban and rural contexts, China implemented a dual land tenure system comprising state ownership in urban areas and collective ownership in rural regions. According to the Land Administration Law of the People’s Republic of China (2019, Articles 9 and 11) and the Civil Code (2020, Articles 260 and 261), rural POSs are legally classified as collectively owned construction land. The village collective holds rights of ownership, use, operation, and management. Moreover, under the Organic Law of Villagers’ Committees (2018, Articles 8 and 28), these collectives are formally responsible for the management and maintenance of rural collective land. This property rights regime has institutionalized villagers’ self-governance as the primary management model for rural POSs. Ostrom [3] posited that such collective governance models can foster sustainable commons management under appropriate institutional, social, and ecological conditions. While this property rights system offers formal support for sustainable rural development, it has also contributed to persistent urban–rural disparities [4]. Industrialization and urbanization have triggered significant out-migration from rural areas in China [5,6]. Rural population out-migration, as an external force arising from broader economic transformations [7], has the potential to disrupt self-governance mechanisms, thereby affecting the sustainable development of rural communities. Although rural POSs are classified as club goods under the property rights system, they exhibit traits of land commons when self-governance mechanisms weaken [8,9]. In such cases, these spaces function as common-pool resources, shared land that is rivalrous in use and difficult to exclude others from, thereby facing governance challenges similar to those of traditional land commons, such as irrigation systems [10].

The impacts of rural out-migration on commons governance are often indirect. Prior studies have shown that migration can lead to diminished local leadership, erosion of social capital, and weakening of reliance on communal resources—factors detrimental to effective rural land commons governance [11,12]. However, migration may also yield positive effects by introducing advanced knowledge, enhancing financial capital through remittances, facilitating land transfers, and reducing resource consumption pressure [13,14,15,16]. These outcomes are context-dependent, varying across locations and institutional settings [13]. Given the complexity of migration’s effects on rural land commons governance, it is critical to empirically investigate how rural out-migration influences the self-governance of POSs in rural China.

To analyze the conditions under which communities engage in cooperative governance and sustain institutional arrangements, Ostrom [3] developed the social–ecological systems (SES) framework. This framework offers a multidisciplinary and integrative lens by embedding institutional, social, and ecological variables within a single analytical structure. Compared to other potential theoretical frameworks, such as the institutional analysis and design (IAD) framework [17], the complex adaptive system (CAS) framework [18], and the ecosystem services (ES) framework [19], the SES framework not only accounts for the internal components of a resource governance system but also incorporates external socio-economic and political settings, as well as broader ecological dynamics. It places particular emphasis on the interdependence among system components. Furthermore, unlike other frameworks, the SES framework offers a systematic classification of the factors affecting commons governance from a comprehensive SES perspective, and its multi-tiered structure allows for a more fine-grained diagnosis of micro-level governance issues. Importantly, the SES framework was originally developed to study the sustainability of commons governance. Given that rural POSs in China function as a form of land commons, the SES framework serves as a robust diagnostic tool for comprehensively examining related governance challenges [9]. These characteristics make the SES framework particularly suitable for studying the impact of rural out-migration as an external social driving factor on rural POS self-governance. This analytical utility has been demonstrated in studies on irrigation systems in rural China [11].

Although some scholars have applied the SES framework to explore POS governance [20,21,22], few have examined the role of external social factors such as rural out-migration. Recent work by Shi and Ling [23] has begun to address this gap by illustrating how external social forces shape POS self-governance within the SES framework. However, a lack of empirical research providing robust, data-driven insights into these dynamics remains.

In response to these research gaps, this study applies the SES framework to investigate the effects and underlying mechanisms of rural out-migration on the self-governance of POSs in rural China. Focusing on 198 villages in Taigu District, the study analyzes 594 villager responses using partial least squares structural equation modeling (PLS-SEM). This approach allows for the identification of mediating pathways across institutional, social, and ecological subsystems within rural SESs. The study aims to reveal both the direct and mediated effects of out-migration on the capacity for POS self-governance. This research contributes to the literature by providing one of the first empirical applications of the SES framework to examine the influence of an external social factor—rural out-migration—on commons governance in the context of POSs. In doing so, it not only advances theoretical insights into rural SES dynamics but also offers practical guidance for policy interventions aimed at enhancing self-governance in the face of demographic shifts.

The remainder of this paper is structured as follows: Section 2 presents a literature review and develops the conceptual framework and hypotheses. Section 3 outlines the study area, data collection strategy, and methodological design. Section 4 reports the empirical findings and offers a detailed discussion. Section 5 concludes with a summary of the study’s key contributions, limitations, and implications for policy and future research.

2. Conceptual SES Framework and Research Hypotheses

Ostrom [3] developed the SES framework based on extensive research on common-pool resources [7,24]. As shown in Figure 1, the SES framework primarily consists of three components. The outermost layer includes two systems representing the external environment: social, economic, and political settings (S) and related ecosystems (ECO). The factors in these two subsystems act as exogenous factors that directly and indirectly influence the SES components of resource management [25]. Previous studies treated population out-migration as an external force stemming from economic changes [7]. Therefore, rural out-migration in China can be categorized as a sub-variable of the social, economic, and political settings (S) within the SES framework. The second layer of the SES framework consists of internal factors that influence the management of resource SES. It includes four subsystems: resource systems (RS), resource units (RU), governance systems (GS), and actors (A). The innermost layer of the SES framework comprises focal action situations, which include subsystems representing the interaction processes of the SES subsystems’ interactions (I), as well as a subsystem representing the outcomes of the subsystems’ interaction outcomes (O). The structure of the SES framework provides a theoretical foundation for studying the impact of rural out-migration on local commons self-governance in China. Therefore, this study adopts the SES framework as its theoretical framework to identify the effects of rural out-migration on the self-governance of rural POSs in China. To develop a robust conceptual framework for this study and propose compelling research hypotheses, the study reviews five potential impact mechanisms through a literature review.

2.1. Mechanism 1: Rural Out-Migration → Villager Organizations → Quality of POS

Village organizations serve as platforms that facilitate collective action among community members [27]. Although the establishment of these organizations is not directly aimed at the management of POSs, they have been demonstrated to enhance villagers’ collective action ability and build social capital through various forms of self-governance activities. Consequently, these organizations indirectly contribute positively to POS governance. This effect has been empirically validated in the context of POS self-governance in both China and the UK [9,27].

The collective organizations in rural China primarily include farmers’ professional cooperatives and farmers’ cultural organizations. Farmers’ professional cooperatives are cooperative entities based on the household contract responsibility system in rural areas. They are typically formed voluntarily by farmers with shared economic interests, enabling members to achieve resource sharing, risk sharing, and mutual support in agricultural production, marketing, and technical training [9]. These cooperatives require that farmers constitute at least 80% of their total membership [28]. On the other hand, farmers’ cultural organizations are social entities spontaneously established by local villagers passionate about arts and culture. They serve the cultural development of rural communities, primarily functioning to preserve and promote local culture, enhance villagers’ cultural literacy, and advance rural spiritual civilization and cultural revitalization through various cultural activities [9]. These village organizations are significantly impacted by rural out-migration. For farmers’ professional cooperatives, rural out-migration reduces their scale and operational efficiency, as villagers’ dependence on agricultural production diminishes [29,30]. Similarly, for farmers’ cultural organizations, the migration of villagers with artistic and cultural talents hinders the establishment of such organizations and the organization of related activities. These village organizations serve as critical platforms for enhancing villagers’ collective action capacity and building social capital [28,31,32]. Unlike other resources, social capital can deplete if not actively engaged in collective action over time [33]. Consequently, rural out-migration may lead to a decline in villager organizations, thereby negatively affecting the self-governance of rural POSs.

2.2. Mechanism 2: Rural Out-Migration → Left-Behind Population → Quality of POS

In rural China, population out-migration can be categorized into labor migration and family migration. The former refers to the movement of working-age individuals from rural areas to cities or other regions, while the latter involves the relocation of entire families to the same destination [34]. Due to China’s urban–rural dual household registration system and the reluctance of some rural families to relinquish their rural benefits, typically, only capable labor force members migrate to cities, which has led to the phenomenon of rural left-behind populations [35,36]. These populations mainly include left-behind children, the elderly, and women [37]. The left-behind population is generally considered to have a lower collective action capacity due to poorer physical and mental health and greater pressure from household agricultural production [35,38,39,40]. However, some studies suggest that labor migrants often remit money to rural households to improve their living conditions, and these remittances may help improve rural commons [41]. Additionally, the presence of left-behind populations increases the connection between migrant laborers and rural areas, with migrants being more likely to settle in nearby cities and return more frequently, thus creating opportunities for their involvement in rural commons management [42]. In contrast, under the family migration model, where the entire rural family migrates, the connection between the migrant population and rural areas is weaker. Furthermore, rural families that have lived away for extended periods tend to have lower dependence on rural commons, which may reduce their motivation to participate in the self-governance of rural POSs. The dependence of rural families in China on rural commons has been shown to affect rural commons governance [11].

2.3. Mechanism 3: Rural Out-Migration → Educational Level of Village Leaders → Quality of POS

Leadership can be defined as a process of social influence in which an individual seeks the assistance and support of others to accomplish a common task [11]. Leadership plays a crucial role in the self-governance of rural commons in China by establishing both vertical and horizontal connections with other relevant stakeholders [9,11]. The educational level of village leaders is regarded as an indicator of their leadership capacity [43], and research has found that higher educational levels among village leaders positively impact the self-governance of rural commons in China [9].

Rural leadership in China primarily includes the party branch secretary and the village committee director, both of whom are elected from among the villagers [44]. Individuals with leadership potential often have the ability to secure employment opportunities and earn higher salaries in urban areas. As a result, the migration of rural populations to cities has led to a significant loss of talent in rural areas, including those villagers with higher education [11,45,46]. The depletion of rural leadership resources reduces the likelihood of successful collective action for the management of the commons [47].

2.4. Mechanism 4: Rural Out-Migration → Collective Investment for POS Self-Governance → Quality of POS

In rural China, the collective funding sources for investing in the self-governance of POSs are diverse, including allocations from higher-level governments, income from collective land leasing, profits from collective enterprises, and individual investments and donations [9]. Research has found that moderate rural out-migration can promote rural land transfer [48,49], and the increase in land transfer rates enables the full utilization of rural resources, thereby boosting income from collective land leasing [50]. However, rural out-migration may also lead to the loss of skilled and managerial talent capable of creating value for collective enterprises, which could negatively impact their revenue [51]. Nevertheless, this effect is not absolute, as rural out-migration may also enable returning migrants to introduce new technologies to collective enterprises, enhancing their operational efficiency [52]. Additionally, rural out-migration can lead to an increase in remittances, which have been shown to generate externalities affecting the provision of rural public goods [53]. Given these findings, there is no consensus on the overall impact of rural out-migration on collective investment in rural commons. Studies have demonstrated that rural commons in China require sustained financial support [54,55], and this is also true for rural POS self-governance [9].

2.5. Mechanism 5: Rural Out-Migration → Self-Organizing Activities for POS Self-Governance → Quality of POS

Self-organizing activities for POS self-governance include village collective meetings and collective action focused on the management and maintenance of rural POSs, such as community-led POS sanitation campaigns and collaborative maintenance of POS facilities. Rural out-migration has been shown to negatively affect the frequency of villagers’ participation in self-organizing activities for POS self-governance. This is primarily due to the reduction in the number of villagers staying in the village year-round, which decreases individuals’ time and opportunities to engage in community life and interact with other villagers. As a result, social capital among villagers diminishes. Additionally, the higher time and economic costs associated with migrants’ participation in POS management reduce the perceived benefits of engaging in self-organizing activities for POS governance [47]. Furthermore, villagers who reside elsewhere for an extended period do not depend on rural POSs for their daily living or economic income, as most outmigrants transition from rural agrarian lifestyles to urban living, with their income sources shifting from agriculture to industrial and service sectors. According to Ostrom [56] and Acheson [57], collective action for resource management is more likely to occur when the actors depend upon the resources, as dependency creates incentives for cooperation. The impact of Chinese villagers’ dependency on rural POSs on their self-governance has been empirically demonstrated [9]. Consequently, rural out-migration reduces villagers’ reliance on rural POSs for both production and daily living, which may make it more difficult for them to be mobilized to participate in the self-organizing activities related to the management and maintenance of rural POSs.

This study employs the SES framework as its theoretical framework. Building upon the aforementioned mechanisms of influence, the research has developed a conceptual SES framework, as shown in Figure 2.

Based on the conceptual SES framework, this study proposes six hypotheses regarding the relationship between rural out-migration and POS self-governance in rural China, which will be empirically tested.

Hypothesis 1.

Rural out-migration has a direct impact on the quality of POS.

Hypothesis 2.

Rural out-migration negatively affects villager organizations, and more villager organizations positively impact the quality of POS.

Hypothesis 3.

Rural out-migration results in a larger left-behind population in rural areas, and these populations affect the quality of POS.

Hypothesis 4.

Rural out-migration leads to a shortage of highly educated leaders in rural areas; highly educated leaders positively affect the quality of POS.

Hypothesis 5.

Rural out-migration influences collective investment for POS self-governance, and increased investment positively affects the quality of POS.

Hypothesis 6.

Rural out-migration negatively affects self-organizing activities for POS self-governance, while increased self-organizing activities positively influence the quality of POS.

The research hypotheses are presented in Figure 3.

3. Research Methodology

3.1. Study Area

This study covers 198 administrative villages in Taigu District, Jinzhong City, Shanxi Province. Figure 4 illustrates the geographic location of Taigu District within Shanxi Province, while Figure 5 presents the spatial distribution of the 198 villages. According to the Statistical Yearbook of Taigu District (2022), the per capita gross domestic product (GDP) of Taigu District is RMB 37,700, significantly lower than the national average of RMB 85,700. Located on the Loess Plateau, Taigu is a typical semi-arid region characterized by severe soil erosion and ecological fragility. Compared to the economically developed regions in Eastern China, the governance of rural POSs in Taigu faces more complex and pressing challenges. Furthermore, limited economic development and poor living conditions have led to large-scale rural out-migration. According to the Sixth (2010) and Seventh (2020) National Population Census of China, the proportion of the rural resident population in Taigu decreased from 61.18% (182,787 people) to 42.77% (137.752 people), reflecting a typical pattern of out-migration in underdeveloped regions of Central and Western China and the associated challenges in POS governance. Therefore, the selection of Taigu District as the study area offers valuable insights into the difficulties of rural POS self-governance under conditions of resource scarcity and demographic decline. It also provides theoretical and practical implications for developing governance strategies tailored to underdeveloped regions, thereby contributing to the broader goal of promoting balanced improvements in human settlement environments across China. It is worth noting that the villages within the study area exhibit considerable variation in natural geography and socio-economic conditions, offering a diverse set of cases for analyzing how rural out-migration influences POS governance across different social-ecological contexts. This diversity enhances the explanatory power and applicability of the study’s findings.

There are various types of POSs in rural China. This study focuses exclusively on the self-governance of central POSs within the 198 administrative villages of Taigu District. In this context, ‘central POSs’ refer to the POSs located at the center of a village, which serve the entire community for both residential and productive activities. Typically, central POSs are associated with temples, ancestral halls, and opera stages, and are used for gathering people and hosting commercial and entertainment activities during festivals. At other times, central POSs facilitate the organization of rural collective political activities, drying agricultural products, airing clothes, and daily entertainment for villagers [2]. As the largest and most prevalent POS in a village, central POSs offer the broadest range of services and facilities for both daily life and production and are present in every village across China. Consequently, this study examines the impact of rural out-migration on the quality of central POSs in the 198 administrative villages of Taigu District.

3.2. Research Variables

Based on the conceptual SES framework and research objectives, this study categorizes the examined factors into dependent variables, independent variables, mediating variables, and control variables to analyze the impact and underlying mechanisms of rural out-migration on rural POS self-governance.

3.2.1. Dependent Variable

The objective of this study is to analyze the impact of rural out-migration on the self-governance of rural POSs. Self-governance can be measured using either the process method or the output method [11]. In this study, the self-governance of rural POSs is measured using the output method. According to Shi and Ling [23], the self-governance of rural POSs can be assessed by the quality of rural POSs. Based on existing theories and local conditions, the latent variable ‘quality of POS’ is comprised of five observed variables: number of human-constructed facilities, maintenance of human-constructed facilities, hygiene, landscape, and usability.

Human-constructed facilities in the POS significantly influence users’ experiences. The level of human-constructed facilities encompasses both their quantity and quality. Adjei Mensah [58] and Makworo and Mireri [59] argued that POSs lacking adequate infrastructure are unattractive to users and detrimentally impact their experience. Similarly, the quality of these facilities is also important. Ling et al. [60] found that the maintenance of POS facilities is among users’ top concerns. Poor maintenance implies inadequate facility provision, which negatively impacts the user experience. Additionally, poor sanitation in POSs diminishes their attractiveness and poses health risks to users [58]. Liu [2] noted that one of the functions of rural POSs in China is to provide ecological landscapes. Consequently, poor ecological landscapes can adversely impact the user experience. The usability of POSs is also a critical factor in measuring their quality. POSs that do not offer adequate services negatively impact users’ experiences and decrease their usage frequency [58]. These five observed variables are measured using a three-point Likert scale.

3.2.2. Independent Variable

In this study, the independent variable is rural out-migration, measured as the proportion of the rural out-migration population to the rural registered population. According to the definition provided by the National Bureau of Statistics of China, the registered population refers to Chinese citizens who have been officially recorded in the household registration system at the public security bureau of their habitual residence, in accordance with the ‘Regulations on Household Registration of the People’s Republic of China.’ Meanwhile, based on the ‘Communiqué of the Seventh National Population Census,’ the permanent residents refer to individuals who have actually and regularly resided in a given area for more than six months. In this study, the number of the out-migration population from a village is calculated as the difference between the rural registered population and the permanent residents who hold household registration in that village. Therefore, both the rural permanent residents and the out-migration population, as defined in this study, are subsets of the rural registered population. Specifically, the rural out-migration population refers to individuals who hold household registration in the village but reside elsewhere for more than six months per year. This category includes circular migrants, seasonal migrants, and long-term migrants who spend more than half a year outside the village. Conversely, circular and seasonal migrants who reside in their household-registered village for more than six months annually are classified as rural permanent residents in this study. Villagers absent from the village for over six months predominantly rely on urban public facilities and employment opportunities in the industrial and service sectors for their daily livelihood and income. Consequently, they exhibit lower dependency on rural POSs and maintain weaker social connections with the village. This study examines the strategies implemented by long-term village residents to manage rural POSs, ensuring the fulfillment of their daily production and living needs amidst the frequent absence of resource-sharing partners.

3.2.3. Mediating Variables

Based on the potential mechanisms of rural out-migration on the quality of rural POSs, this study examines the mediating effects of five critical variables. In rural China, farmers’ professional cooperatives and farmers’ cultural organizations are the most common forms of villager organizations. The establishment of these organizations requires a sufficient number of villagers [28], and rural out-migration may lead to a decline in their numbers. Therefore, this study measures “villager organizations” using two observed variables: the number of farmers’ professional cooperatives and the number of farmers’ cultural organizations. Data for these variables were collected through the survey questions: ‘How many farmers’ professional cooperatives in the village?’ and ‘How many farmers’ cultural organizations in the village?’

In this study, the variable ‘left-behind population’ includes left-behind children, women, and elderly individuals who continue to reside in rural areas while the primary labor force members work away from home for more than three consecutive months per year. This group is considered one of the sources of social challenges in rural China [61,62,63]. Left-behind children refer to individuals under the age of 18 whose parents, either one or both, have migrated for work or other reasons, leaving them under the care of relatives or neighbors [38,64]. The left-behind elderly are those aged 60 and above who, due to their children’s migration for work or other reasons, lack direct care and companionship from their children and live alone or depend on other relatives. Left-behind women are married women aged 20 to 59 who remain in rural areas while their husbands migrate for work or other reasons [35]. In this study, the latent variable ‘left-behind population’ is measured using the observed variable ‘proportion of left-behind population,’ which is calculated as the ratio of the total number of left-behind children, women, and elderly to the number of permanent residents with household registration in the village.

In rural China, the primary local leaders are the party branch secretary and the village committee director, who are responsible for managing most village affairs and are elected by local villagers [44]. Changes in their leadership tend to have a greater impact on village resource management compared to other villagers. Therefore, in this study, the latent variable ‘educational level of village leaders’ is measured using two observed variables: ‘educational level off party branch secretary’ and ‘educational level of village committee director.’ Depending on local conditions, both variables are assessed using a five-point Likert scale, ranging from junior high school (1) to postgraduate (5), to measure the educational levels of the party branch secretary and the village committee director.

Collective investment for POS self-governance refers to the total annual financial resources allocated by the village collective toward the maintenance and development of the central POS. These funds are typically derived from multiple sources, including government grants, collective land lease revenues, profits from village-owned enterprises, individual investments, and private donations [9]. The availability of these funding sources may be influenced by rural out-migration, potentially affecting the overall investment capacity of the village collective in a rural central POS. Data for this variable are collected through the survey question: ‘How much did the village invest in the central square and its facilities this year? (Unit: Yuan).’

Self-organizing activities for POS self-governance encompass collective meetings and self-governance activities related to the management of the village’s central POS and its associated facilities. These activities include, but are not limited to, collective sanitation activities for the central POS, collective maintenance activities for public facilities within the central POS, and village collective meetings related to rural POS self-governance. Rural out-migration may hinder the organization of such self-organizing activities, leading to a decline in their frequency [56,57]. Therefore, this study measures the variable ‘self-organizing activities for POS self-governance’ by assessing the number of public meetings and activities for central POS governance held in the village over the past year. Data for this variable are collected through the survey question: ‘How many community meetings and activities about managing central public squares were held in the village this year? (Unit: Times) (Example: village collective meetings about or including central square’ management affairs, village collective cleaning activities for central square, collective maintenance action for facilities related to the central square).’

3.2.4. Control Variables

To enhance the validity of the research findings, this study incorporates a series of institutional–social–ecological factors that may influence rural POS self-governance as control variables. Among these, the resource systems and units (RSU) include two ecological factors: ‘size of POS’ and ‘proximity to urban area.’ Wade [65] suggests that smaller resource systems can be managed more effectively. Ling et al. [66] found that managing large POSs typically incurs higher costs. However, larger POSs may be more attractive, as users are often more willing to invest in them to obtain higher returns [67]. Data for this variable are collected through the survey question: ‘What is the central square in the village? (Unit: square meter).’

The proximity of rural POSs to urban areas may influence their self-governance, with both positive and negative effects. In rural areas near cities, lower rental costs and a more affordable business environment may attract an influx of external populations, increasing pressure on the use and management of rural POSs [68]. However, rural areas near urban areas also benefit from easier access to urban resources [69], which may help mitigate the negative impacts of the dual land ownership system on rural POS self-governance. The variable ‘proximity to urban area’ is measured by the distance from the village’s central square to the central business district (CBD) of Taigu District.

This study selects three institutional factors as control variables: government assistance, POS management regulations, and monitoring systems. In certain cases, government assistance can address deficiencies in the self-governance of rural POSs. Although the ownership of rural POSs belongs to the village collective, the government may intervene as a manager, investor, facilitator, coordinator, promoter, or educator to support rural POS self-governance when these spaces significantly influence regional development [70]. For example, in Daye County, China, the local government has invested substantial funds to improve rural POSs and related public facilities. Additionally, the government provides skills training for POS self-governance and serves as a mediator between villagers and private investors [71]. In this study, the variable ‘government assistance’ is coded as one if the rural POS construction and management receive government assistance and two if it does not.

The management regulations for rural POSs define the rights and responsibilities of relevant stakeholders and standardize the self-governance process. These regulations not only regulate the behavior of the involved actors but also help managers identify governance issues and propose timely solutions [72]. In rural China, POS management regulations are typically incorporated into village rules and agreements, specifying both villagers’ rights to use rural POSs and their corresponding obligations [73]. The data on this variable is collected through the survey question: “Are there any rules for managing the central square in the village?” A response indicating the presence of POS management regulations is assigned a value of one, while the absence of such regulations is assigned a value of two.

In rural China, the role of monitors is typically fulfilled by ‘grid members,’ although in some villages, this responsibility may also be assumed by village leaders. One of the key duties of these monitors is to identify violations by POS users and enforce corrective measures in accordance with relevant regulations [74]. Scholars such as Wade [65], Ostrom [75], and Baland and Platteau [76] have emphasized that the presence of monitors is a critical condition for successful commons management. In this study, the variable ‘monitoring system’ is operationalized as a binary variable, where a value of one indicates that the village has designated monitors for rural POSs, while a value of two indicates that no such monitor has been assigned.

The control variables in this study include four social factors: the number of actors, the income level of villagers, the per capita homestead area, and the number of village cadres. Generally, organizing collective action in larger groups incurs higher costs, as it is more challenging to unify individual intentions and monitor individual behaviors [22,33]. However, studies on rural POS self-governance in China suggest that a larger number of villagers may have a positive impact on POS governance [9]. As villagers are the primary participants in rural POS self-governance, the number of actors is measured by the number of permanent residents with household registration in the village.

As a social attribute, the income level of villagers may influence rural POS self-governance. Agrawal [77] suggests that a lower level of poverty is one of the conditions for facilitating successful commons management. This study measures villagers’ income level using their annual per capita income.

Homesteads, which serve as residential land for each household in a village, constitute a critical space for villagers’ daily production and living activities [78]. The functions of homesteads overlap with those of rural POSs to some extent. Consequently, when the homestead areas are larger, villagers have more private space, reducing their dependence on rural POSs, which may, in turn, weaken their motivation to participate in POS self-governance [60,79]. This study collects data on the variable ‘per capita homestead area’ through the survey question: ‘What is the average residential land area per villager? (Unit: square meter).’

In addition to the village party branch secretary and the village committee director, village collectives in China also elect and appoint other village cadres to assist in rural governance. In villages with a larger number of cadres, the allocation of responsibilities for managing rural public affairs tends to be more efficient, thereby enhancing the effectiveness of POS self-governance [9]. This study measures the variable ‘number of village cadres’ by surveying the total number of cadres in the village.

The relevant variable information for this study is presented in

Table 1. Information about variables.

Type of Variables	SES Subsystems	Name of Latent Variables	Name of Observed Variables	Measurement Level	Variables’ Assignment	Description
Dependent variable	Outcomes (O)	Quality of POS (O)	Number of human-constructed facilities	Ordinal level of measurement	Low = 1; Medium = 2; High = 3	The total count of built structures and installations within the POS, such as benches, shelters, fitness equipment, and other infrastructure.
			Maintenance of human-constructed facilities	Ordinal level of measurement	Low = 1; Medium = 2; High = 3	The upkeep and repair of man-made facilities within the POS to ensure their proper use and durability.
			Hygiene	Ordinal level of measurement	Low = 1; Medium = 2; High = 3	The cleanliness and sanitary condition of the POS.
			Landscape	Ordinal level of measurement	Low = 1; Medium = 2; High = 3	The physical features, design, and natural elements of the POS that affect its appearance and functionality.
			Usability	Ordinal level of measurement	Low = 1; Medium = 2; High = 3	The ease and convenience with which villagers can access and use the POS for its intended purposes.
Independent variable	Social, economic, and political settings (S)	Rural out-migration (S1)	Proportion of rural population out-migration	Interval-ratio level of measurement		The percentage of registered villagers who have lived outside the village for more than six months.
Mediating variable	Governance systems (GS)	Villager organizations (GS1)	Number of farmers’ professional cooperatives	Interval-ratio level of measurement		The total count of officially registered cooperatives formed by farmers to organize production, marketing, or other agricultural activities.
			Number of farmers’ cultural organizations	Interval-ratio level of measurement		The total count of farmer-led groups or associations in the village that promote cultural activities, traditions, and community cohesion.
	Actors (A)	Left-behind population (A1)	Proportion of left-behind population	Interval-ratio level of measurement		The percentage of rural residents, such as children, women, and elderly, who remain in the village while others migrate for work.
		Education level of village leaders (A2)	Education level of party branch secretary	Nominal level of measurement	Junior high school = 1; Senior high school = 2; Junior college = 3; Bachelor = 4; Postgraduate = 5	The village party branch secretary’s highest formal education.
			Education level of village committee director	Nominal level of measurement	Junior high school = 1; Senior high school = 2; Junior college = 3; Bachelor = 4; Postgraduate = 5	The village committee director’s highest formal education.
	Interactions (I)	Collective investment for POS self-governance (I1)	Collective investment for POS self-governance	Interval-ratio level of measurement		The total amount of funds contributed by the village collective to support the self-governance of the central POS.
		Self-organizing activities for POS self-governance (I2)	Number of public meetings and activities	Interval-ratio level of measurement		The total count of organized gatherings and events held in the village to engage residents in community affairs for POS management.
Control variables	Resource systems and units (RSU)	Size of POS (RSU1)	Size of POS	Interval-ratio level of measurement		The total area of the central POS, reflecting its spatial scale.
		Proximity to urban area (RSU2)	Distance between urban CBD and rural POS	Interval-ratio level of measurement		The physical distance from the central POS to the city’s CBD.
	Governance systems (GS)	Government assistance (GS2)	Government investment projects	Nominal level of measurement	POS with government investment projects = 1; POS without government investment projects = 2	Projects funded or supported by the government aimed at developing, constructing, or managing central POS.
		POS management regulations (GS3)	POS management regulations	Nominal level of measurement	POS with public management regulations = 1; POS with public management regulations = 2	Rules and guidelines established to govern the use, maintenance, and management of the central POS.
		Monitoring system (GS4)	Participation of monitors	Nominal level of measurement	POS with designated monitors = 1; POS without designated monitors = 2	The involvement of supervisors in monitoring the use of central POS.
	Actors (A)	Number of actors (A3)	Number of permanent residents	Interval-ratio level of measurement		The number of villagers who have lived in the village for more than six months.
		Income level of villagers (A4)	Per capita income of villagers	Interval-ratio level of measurement		The villagers’ per capita annual income this year.
		Per capita homestead area (A5)	Per capita homestead area	Interval-ratio level of measurement		The average homestead land area per household in the village.
		Number of village cadres (A6)	Number of village cadres	Interval-ratio level of measurement		The total count of local village officials responsible for governance and administration.

.

3.3. Data Collection

The study collected relevant data through a questionnaire survey, which consisted of two types: a village-level questionnaire for village leaders and a household-level questionnaire for rural households. The rationale for this approach is that the independent variables, mediating variables, and control variables of the study involve village-level basic information related to rural POS self-governance. Village leaders, being the most knowledgeable about these matters, are best positioned to provide accurate information for the study. Therefore, informed consent forms and questionnaires were distributed to village committee directors from 198 villages to gather information on the relevant variables. The dependent variable of the study pertains to the assessment of rural POS quality. Compared to village leaders, villagers are more likely to provide accurate assessments of rural POS quality, as they are not directly responsible for village management. To minimize potential biases in villagers’ perceptions of POS quality, the study implemented specific measures in both questionnaire design and data sampling method. In terms of questionnaire design, the study included the evaluation criteria alongside response options for questions related to rural POS quality to enhance the accuracy of the responses. For data sampling, a random sampling method was employed. Each village was considered with its central POS as the core. Families were assigned numerical identifiers based on their proximity to the central POS, with smaller numbers indicating closer proximity. Three rural households were then randomly selected from this list in each village. Informed consent forms and questionnaires were distributed to these households, with each participating household designating one adult representative most familiar with the questionnaire content to complete the survey. If a selected household was unable or unwilling to participate, the random sampling process was repeated until three valid questionnaire responses were obtained from each village. According to the seventh national census of China, there are approximately 53,392 rural households across these 198 villages. Following Krejcie and Morgan’s guidelines [80], the sample size provides a margin of error of less than 0.05 with a 95% confidence interval. Therefore, the sample size for assessing rural POS quality is considered sufficient and representative. After completing data collection, the study matched the data collected from village-level questionnaires with household-level questionnaires. Specifically, the three sets of POS quality-related data collected from each village were averaged to represent the POS quality level of the corresponding village. The averaged POS quality data were then matched with the 198 datasets collected from village leaders, ultimately resulting in 198 datasets for data analysis.

3.4. Data Analysis

The study employs partial least squares structural equation modeling (PLS-SEM) for data analysis, a second-generation structural equation modeling technique developed by Wold [81]. PLS-SEM is widely regarded by scholars as suitable for research involving latent variables and complex cause-and-effect relationships [82]. PLS-SEM offers several unique advantages that align with the data analysis requirements of this study. Specifically, PLS-SEM enables the measurement of latent variables through a single observed variable, whereas covariance-based methods typically require at least three observed variables per latent variable. This flexibility aligns well with the design requirements of the variables in this study. Additionally, PLS-SEM does not require data to follow a normal distribution, which can be challenging to achieve in sociological research. Furthermore, PLS-SEM effectively manages measurement error, allowing for more accurate estimation of the impact effects of variables. The PLS-SEM model diagram of this study is shown in Figure 6.

According to the conceptual framework of this study, the primary focus is on testing the mediation effect model. The classic method for measuring the mediating effect of variables is the causal steps approach [83]. However, this method is prone to significant errors in model estimation [84,85]. In contrast, the PLS-SEM model employs the bootstrapping approach to analyze the mediation effects, which helps mitigate such errors [86]. The bootstrapping approach constructs confidence intervals for indirect effects, allowing for a direct test of the mediation effects and offering a more convenient alternative to the causal steps method. Moreover, the bootstrapping approach is less affected by sample size issues, providing more reliable results with smaller samples [78,87]. For the bootstrapping test in this study, 5000 bootstrap subsamples were used, with a two-tailed test and a significance level set at 0.05. The data analysis was conducted using SmartPLS 4.

4. Research Results

4.1. Descriptive Statistics

Table 2. Descriptive statistics for the items (observed variables).

Name of Latent Variables	Name of Observed Variables	Mean	Median	Scale Min	Scale Max	Observed Min	Observed Max	Standard Deviation	Excess Kurtosis	Skewness
Quality of POS (O)	Number of human-constructed facilities	1.379	1	1	3	1	3	0.579	0.635	1.315
	Maintenance of human-constructed facilities	2.172	2	1	3	1	3	0.436	0.810	0.477
	Hygiene	2.217	2	1	3	1	3	0.675	−0.869	−0.320
	Landscape	1.635	1.667	1	3	1	3	0.667	−0.792	0.582
	Usability	1.847	2	1	3	1	3	0.733	−1.228	0.210
Rural out-migration (S1)	Proportion of rural population out-migration	0.205	0.054	0	0.893	0	0.893	0.268	−0.115	1.188
Size of POS (RSU1)	Size of POS	1886.753	1175	82	11,352	82	11,352	2002.674	5.472	2.184
Proximity to urban area (RSU2)	Distance between urban CBD and rural POS	17,333.939	14,300	920	57,000	920	57,000	12,135.874	0.497	0.940
Villager organizations (GS1)	Number of farmers’ professional cooperatives	2.909	2	0	21	0	21	3.940	5.434	2.191
	Number of farmers’ cultural organizations	0.778	1	0	8	0	8	1.074	13.476	2.995
Government assistance (GS2)	Government investment projects	1.929	2	1	2	1	2	0.256	9.487	−3.375
POS management regulations (GS3)	POS management regulations	1.227	1	1	2	1	2	0.419	−0.283	1.312
Monitoring system (GS4)	Participation of monitors	1.566	2	1	2	1	2	0.496	−1.949	−0.267
Left-behind population (A1)	Proportion of left-behind population	0.158	0.005	0	0.935	0	0.935	0.255	1.323	1.610
Education level of village leaders (A2)	Education level of party branch secretary	3.374	3	1	5	1	5	0.683	0.878	0.705
	Education level of village committee director	3.288	3	1	5	1	5	0.605	1.701	0.724
Number of actors (A3)	Number of permanent residents	1068.313	689	6	6456	6	6456	1203.516	3.208	1.725
Income level of villagers (A4)	Per capita income of villagers	39,444.444	40,000	20,000	60,000	20,000	60,000	12,110.370	−1.176	−0.168
Per capita homestead area (A5)	Per capita homestead area	174.167	150	30	650	30	650	93.292	3.052	1.246
Number of village cadres (A6)	Number of village cadres	5.732	6	1	16	1	16	2.658	1.274	0.693
Collective investment for POS self-governance (I1)	Collective investment for POS self-governance	17,252.525	7000	0	160,000	0	160,000	25,969.618	9.171	2.875
Self-organizing activities for POS self-governance (I2)	Number of public meetings and activities	5.389	5	1	15	1	15	2.976	−0.860	0.257

presents the descriptive statistics for all observed variables in the 594 data samples. The measurement scale and observed scale are consistent for all variables, indicating that the design of the measurement scale for the research variables is appropriate. The standard deviation reflects the dispersion of the data values: a smaller standard deviation suggests that data values are closely clustered around the mean, indicating less variability, while a larger standard deviation signifies greater variability. Among the research variables, the standard deviations for ‘size of POS,’ ‘distance between urban CBD and rural POS,’ ‘number of permanent residents,’ ‘per capita income of villagers,’ ‘per capita homestead area,’ and ‘collective investment for POS self-governance’ are relatively large, reflecting significant differences among villages but remaining within a reasonable range. The excess kurtosis and skewness values indicate that the data for all research variables do not follow a normal distribution. Examining the excess kurtosis values, most of the values of the variables do not exhibit excessively high excess kurtosis. Compared to other variables, ‘size of POS’, ‘number of farmers’ professional cooperatives’, ‘number of famers’ cultural organizations’, ‘proportion of left-behind population’, and ‘collective investment for POS self-governance’ show a relatively noticeable excess kurtosis. This indicates that, for these variables, some villages have significantly higher data values compared to others. Regarding skewness, most variables do not display extreme skewness. Nevertheless, ‘historical and cultural protection POS’ and ‘government investment projects’ exhibit noticeable negative skewness, suggesting that only a few villages possess central POSs with historical and cultural protection value or government assistance. Conversely, the ‘proportion of left-behind population’ shows a higher positive skewness compared to other variables, indicating that some villages have a markedly higher proportion of out-migrants.

4.2. Assessment of Measurement Model

When analyzing data using PLS-SEM, it is essential to first assess the measurement model to ensure the validity of the constructs. In this study, each construct (latent variable) is comprised of formative indicators (observed variables). Thus, the validity of the measurement model is evaluated by examining the indicator weights, the variance inflation factor (VIF), and inter-construct correlations.

When a construct is measured by multiple indicators, this study assigns equal weights to each indicator.

Table 3. Indicator validity of the measurement model (formative model).

Construct (Latent Variable)	Indicator (Observed Variable)	Outer Weights	p-Values	Variance Inflation Factor (VIF)
Quality of POS (O)	Number of human-constructed facilities	0.319	0.000	1.095
	Maintenance of human-constructed facilities	0.319	0.000	1.096
	Hygiene	0.319	0.000	1.173
	Landscape	0.319	0.000	1.370
	Usability	0.319	0.000	1.446
Rural out-migration (S)	Proportion of rural population out-migration	1.000	0.000	1.000
Size of POS (RSU1)	Size of POS	1.000	0.000	1.000
Proximity to urban area (RSU2)	Distance between urban CBD and rural POS	1.000	0.000	1.000
Villager organizations (GS1)	Number of farmers’ professional cooperatives	0.618	0.000	1.106
	Number of farmers’ cultural organizations	0.618	0.000	1.106
Government assistance (GS2)	Government investment projects	1.000	0.000	1.000
POS management regulations (GS3)	POS management regulations	1.000	0.000	1.000
Monitoring system (GS4)	Participation of monitors	1.000	0.000	1.000
Left-behind population (A1)	Proportion of left-behind population	1.000	0.000	1.000
Education level of village leaders (A2)	Education level of party branch secretary	0.611	0.000	1.129
	Education level of village committee director	0.611	0.000	1.129
Number of actors (A3)	Number of permanent residents	1.000	0.000	1.000
Income level of villagers (A4)	Per capita income of villagers	1.000	0.000	1.000
Per capita homestead area (A5)	Per capita homestead area	1.000	0.000	1.000
Number of village cadres (A6)	Number of village cadres	1.000	0.000	1.000
Collective investment for POS self-governance (I1)	Collective investment for POS self-governance	1.000	0.000	1.000
Self-organizing activities for POS self-governance (I2)	Number of public meetings and activities	1.000	0.000	1.000

presents the analysis results for the indicator validity of the measurement model. All outer weights of the indicators are greater than 0.2, and the p-values are less than 0.05, indicating that all constructs are valid [88]. Additionally, assessing multicollinearity among the formative indicators is crucial for evaluating the validity of the measurement model. The VIF for the formative indicators is calculated and shown in Table 3, with all VIF values being below 3.3, suggesting that there is no multicollinearity among the formative indicators [89]. Finally, the discriminant validity of the measurement model is evaluated by calculating the correlations between constructs. As shown in Figure 7, all correlations between constructs are below 0.7, indicating sufficient discriminant validity among the constructs.

4.3. Assessment of Structural Model

The assessment of the structural model includes evaluating its validity and analyzing the path coefficients. Consequently, the results of this assessment are presented in two parts.

4.3.1. Assessment of Validity of the Structural Model

The validity of the structural model in this study is evaluated using the coefficient of determination (R²) and predictive relevance (Q²).

Table 4. Assessment of the structural model validity.

Model Configuration	Target Variable	R-Square	Standard Deviation (STDEV)	T Statistics (|O/STDEV|)	p-Value	SSO	SSE	Q-Square
M1	Villager organizations (GS1)	0.156	0.025	6.160	0.000	396.000	355.995	0.101
	Left-behind population (A1)	0.479	0.090	5.296	0.000	198.000	106.327	0.463
	Education level of village leaders (A2)	0.015	0.020	0.771	0.440	396.000	394.287	0.004
	Collective investment for POS self-governance (I1)	0.095	0.018	5.293	0.000	198.000	179.980	0.091
	Self-organizing activities for POS self-governance (I2)	0.359	0.047	7.622	0.000	198.000	127.804	0.355
M2	Quality of POS (O)	0.320	0.048	6.671	0.000	990.000	865.982	0.125
M3	Quality of POS (O)	0.596	0.038	15.712	0.000	990.000	767.015	0.225
M4	Quality of POS (O)	0.728	0.033	21.964	0.000	999.000	858.211	0.133

Notes. M1: Independent variable (rural out-migration) influencing mediating variables; M2: Independent variable (rural out-migration) influencing dependent variable; M3: Independent variable (rural out-migration) and mediating variables influencing dependent variable; M4: Independent variable (rural out-migration), mediating variables, and control variables influencing dependent variable.

presents the results of the structural model validity assessment under different model configurations. The first model configuration (M1) examines the direct effects of the independent variable (rural out-migration) on five mediating variables. Specifically, the coefficient of determination (R²) for villager organizations (GS1) is 0.156, indicating that 15.6% of the variation in the number of rural nongovernment organizations can be explained by rural out-migration. The R² for the left-behind population (A1) is 0.479, meaning that 47.9% of the variation in the proportion of the left-behind population can be explained by rural out-migration. The p-value of the path coefficient for the impact of rural out-migration (S1) on the education level of village leaders (A2) is greater than 0.05, suggesting that rural out-migration does not significantly affect the education level of village leaders. The R² for collective investment for POS self-governance (I1) is 0.095, indicating that 9.5% of the variation in the amount of collective investment for POS self-governance can be explained by rural out-migration. The R² for self-organizing activities for POS self-governance (I2) is 0.359, signifying that 35.9% of the variation in self-organizing activities for POS self-governance can be explained by rural out-migration.

The second model configuration (M2) examines the direct impact of the independent variable (rural out-migration) on the dependent variable (quality of POS). The R² for POS quality is 0.320, meaning that 32.0% of its variation can be directly explained by rural out-migration.

The third model configuration (M3) evaluates both the direct effect of the independent variable on the dependent variable and its indirect effects through five mediating variables. Under this configuration, the R² for quality of POS is 0.596, indicating that 59.6% of the variation in POS quality can be explained by rural out-migration and its indirect effects through the mediating variables.

The fourth model configuration (M4) represents the full assessment of the structural model. In this configuration, the R² for quality of POS is 0.728, indicating that 72.8% of the variance in POS quality can be explained by institutional–social–ecological factors, including the independent variable, mediating variables, and control variables. This result demonstrates the strong explanatory power and the rationale of the model design in this study. The inclusion of five mediating variables and nine control variables significantly enhances the model’s capacity to analyze how rural out-migration influences the quality of rural POS.

Furthermore, as shown in Table 4, the Q² values for all mediating and dependent variables are greater than zero, confirming that the research model has predictive relevance.

By evaluating the validity of the structural model using the above indicators, the results demonstrate that the model effectively validates the impact effects and mechanisms of rural out-migration on quality of POS.

4.3.2. Analysis of Impact Path Effects

Given that the research objective is to examine the impact effects and mechanisms of rural out-migration on local rural POS self-governance, this section focuses on reporting both the direct effect of rural out-migration (S1) on the quality of POS (O) and the mediating effects of the mediating variables. The path coefficients of the PLS-SEM model are shown in Figure 8.

Table 5. The impact paths of rural out-migration (S1) on quality of POS (O).

Impact Path	Estimate	Standard Deviation (STDEV)	T Statistics (|O/STDEV|)	p-Value	Lower 2.5%	Upper 2.5%
Rural out-migration (S1) → Quality of POS (O)	0.009	0.078	0.119	0.905	−0.149	0.161
Rural out-migration (S1) → Villager organizations (GS1) → Quality of POS (O)	−0.058	0.022	2.604	0.009	−0.103	−0.016
Rural out-migration (S1) → Left-behind population (A1) → Quality of POS (O)	−0.108	0.045	2.410	0.016	−0.196	−0.021
Rural out-migration (S1) → Education level of village leaders (A2) → Quality of POS (O)	−0.001	0.006	0.155	0.876	−0.013	0.011
Rural out-migration (S1) → Collective investment for POS self-governance (I1) → Quality of POS (O)	−0.048	0.015	3.096	0.002	−0.079	−0.018
Rural out-migration (S1) → Self-organizing activities for POS self-governance (I2) → Quality of POS (O)	−0.111	0.043	2.546	0.011	−0.199	−0.028

summarizes the analysis results for all impact paths from rural out-migration (S1) to the quality of POS (O).

As shown in Table 5, the study first examined the direct effect of rural out-migration (S1) on the quality of POS (O). Based on the p-values and confidence intervals, the results indicate that rural out-migration does not exert a significant direct effect on POS quality. Therefore, research hypothesis H1 is rejected.

However, villager organizations (GS1) mediate the relationship between rural out-migration (S1) and quality of POS (O), with a path coefficient of −0.058. As illustrated in Figure 8, rural out-migration negatively affects villager organizations with a coefficient of −0.394, while villager organizations, in turn, positively influence POS quality with a coefficient of 0.147. Hence, research hypothesis H2 is accepted.

Furthermore, the mediation pathway ‘Rural out-migration (S1) → Left-behind population (A1) → Quality of POS (O)’ is statistically significant, with a path coefficient of −0.108. Specifically, as shown in Figure 8, rural out-migration leads to an increase in the left-behind population (coefficient = 0.692), which negatively impacts POS quality (coefficient = −0.157), thereby accepting hypothesis H3.

In contrast, the education level of village leaders (A2) does not significantly mediate the effect of rural out-migration (S1) on quality of POS (O), as shown in Table 5. Accordingly, research hypothesis H4 is rejected.

The results also reveal that collective investment for POS self-governance (I1) serves as a mediating variable in the pathway ‘Rural out-migration (S1) → Collective investment for POS self-governance (I1) → Quality of POS (O),’ with a path coefficient of −0.048. As depicted in Figure 8, rural out-migration negatively affects collective investment (coefficient = −0.308), whereas collective investment positively influences POS quality (coefficient = 0.156). Therefore, research hypothesis H5 is accepted.

Additionally, the mediation pathway ‘Rural out-migration (S1) → Self-organizing activities for POS self-governance (I2) → Quality of POS (O)’ is significant and negative, with a total effect of −0.111. Figure 8 shows that rural out-migration negatively impacts self-organizing activities (coefficient = −0.599), which in turn, positively affect POS quality (coefficient = 0.185), thereby accepting hypothesis H6.

Overall, rural out-migration (S1) does not directly affect the quality of POS (O), but rather influences it indirectly through the full mediating effects of villager organizations (GS1), the left-behind population (A1), collective investment for POS self-governance (I1), and self-organizing activities for POS self-governance (I2), resulting in a total indirect effect of −0.325. Among these mediating variables, the pathways through the left-behind population (A1) and self-organizing activities for POS self-governance (I2) contribute the most, accounting for 67.38% of the total effect.

5. Discussion

This section discusses the four impact pathways through which rural out-migration affects the POS quality identified in this study, aiming to reveal the underlying mechanisms of the mediating variables, namely villager organizations (GS1), left-behind population (A1), collective investment for POS self-governance (I1), and self-organizing activities for POS self-governance (I2).

The findings reveal that rural out-migration leads to a shortage of human capital and talent in villages, which hinders the establishment of villager organizations related to villagers’ daily lives and production. Although these organizations are not directly linked to rural POS self-governance, they serve as platforms for building social capital within the village collective. Their decline reduces the frequency of villagers’ participation in collective action for village governance, thereby diminishing social capital. The role of villager organizations in fostering social capital at the village collective level has been demonstrated in several studies [28,32]. Lehavi [33] argued that the prolonged absence of collective action depletes social capital. Social capital facilitates collective action because individuals who participate in such activities are more likely to trust others, invest in collective action, and exhibit lower tendencies toward free-riding behavior, thereby reducing the organizational costs of collective action [90]. Therefore, an increase in the number of villager organizations improves rural POS self-governance. The role of social capital is also reflected in the self-governance of other rural commons. Multiple studies have shown that, in villages with well-developed social capital, local villagers are more capable of efficiently managing commons, including water resources, irrigation systems, forest, and integrated pest management [11,90,91,92,93]. While many studies have found that rural out-migration reduces local social capital and adversely affects the self-governance of rural commons [47,94], this study further explores the underlying mechanisms of this effect from the perspective of villager organizations.

The findings indicate that, in rural Taigu, most out-migrants are working-age individuals, and many families are either unable or unwilling to relocate entirely to urban areas. Although some studies have found that the left-behind population may generate positive externalities for rural commons governance under the context of rural out-migration, such as through remittances and maintaining ties between migrants and maintaining ties between ties between migrants and their villages [41,42], this study reveals that such effects are not significant in the context of POS self-governance in rural Taigu. Some studies have shown that the left-behind population plays a dual role in rural resource management. On the one hand, remittances can positively influence rural resource management through financial support. On the other hand, the loss of the labor force can lead to a decline in local resource management capacity, exerting a negative impact [95,96]. The left-behind population has been identified as a vulnerable group in Chinese society, often associated with a range of social issues, including mental health problems, physical health challenges, low educational attainment, marital instability, and inadequate social security [61,62,63]. The adverse social attributes of the left-behind population result in a lack of motivation and capacity to participate in rural POS self-governance. Additionally, although left-behind individuals receive remittances from outmigrants, these funds are primarily allocated to daily household expenses rather than rural commons management, as the left-behind population faces higher poverty rates and greater pressure from agricultural production [39,97]. It is worth noting that, despite the lack of management capacity in the rural commons due to labor out-migration, the left-behind population has the potential to play a positive role in commons management. Studies have found that, when marginalized left-behind women are empowered, they can rapidly develop their capabilities and participate effectively in the governance of rural commons [98]. The negative impact of the left-behind population on rural commons governance may not solely stem from their individual capacity to engage in self-governance. Instead, transforming traditional patriarchal norms and granting them greater rights could improve rural POS self-governance.

The funding sources for rural POS self-governance are diverse. For rural households, although remittances from out-migrants are sent back to rural families, these funds do not significantly increase collective investment for rural POS self-governance. This finding contrasts with other studies that have identified remittances as contributing to public expenditures for rural commons [53]. The primary purpose of remittances is often rooted in intra-family “altruism” [99], where migrant family members, unable to provide direct care, contribute financially to alleviate the material and emotional burdens of their families. Consequently, remittances are typically prioritized for covering household expenses rather than public investments. Furthermore, due to the household registration system and household contract responsibility system, a significant portion of rural out-migrants retain their rural household registration, as well as ownership, of homestead land and farmland. This has led to the abandonment of large areas of farmland [100,101]. Since rural POSs often serve as spaces for grain drying, the abandonment of farmland reduces villagers’ reliance on rural POSs, diminishing the village collective’s motivation to invest in rural POSs. Meanwhile, rural out-migration leads to a decline in POS users, further diminishing the motivation for village collectives to improve these spaces [51]. Although China has a rural land transfer system, not all village collectives can generate revenue through land transfers to improve the rural living environment, especially in areas lacking industry and large-scale agricultural enterprises [100]. More evidently, rural out-migration leads to local labor shortages and decreased consumption potential, resulting in a less attractive rural capital environment. This unfavorable situation hampers the operation of village collective enterprises and deters the entry of external economic entities [51], leading to a decline in village collective economic income and thereby reducing the village collective’s capacity to invest in POS self-governance. Some studies suggest that rural out-migration can increase the income levels of villagers and village collectives [50,52]. However, this study finds that current rural out-migration does not significantly enhance the economic income levels of villagers and village collectives, nor their willingness to invest in POS self-governance. As the sources of collective investment for POS self-governance are diverse, the negative impact of rural out-migration on rural POS self-governance is, in fact, the result of a combination of multiple factors. These include local labor shortages, deficiencies in management systems, lagging local development, inadequate technological promotion, and villagers’ perceptions. The research findings demonstrate that rural POS self-governance in China requires sustained financial support.

Similar to other rural commons, on the one hand, the reduced participation of long-term migrants in rural community life depletes social capital among villagers, leading to a lack of confidence in self-organizing activities for POS self-governance, thereby making such activities less likely to occur [47,102]. On the other hand, migrants have lower economic and daily life dependence on rural POSs, and returning to participate in POS self-governance entails significant time and financial costs. However, they are unable to derive substantial benefits from such participation, making it difficult to mobilize them for POS self-governance activities [103,104]. These factors collectively increase the costs for village collectives to organize POS self-governance activities. Although many migrants retain housing and farmland in their villages, their primary motivation for maintaining ties with the village is not to facilitate participation in rural commons management but rather to serve as a fallback option in case of economic instability or inability to continue working in urban areas [101,105]. The research findings indicate that, like other rural commons, POS governance requires the active participation of relevant actors to maintain its quality. Self-organizing activities are even used as a direct measure of the management level of the commons [11].

6. Conclusions and Implications

This study adopts the social–ecological systems (SES) framework to examine the impact pathways through which rural out-migration influences the self-governance of rural public open spaces (POSs) in China. Utilizing partial least squares structural equation modeling (PLS-SEM), the study analyzes data from 198 villages to identify the mediating mechanisms underlying this relationship. The results indicate that research hypotheses H2, H3, H5, and H6 are accepted, while hypotheses H1 and H4 are rejected. Therefore, rural out-migration does not have a direct impact on POS self-governance but exerts a significant indirect negative effect through four mediating variables: villager organizations (GS1), left-behind population (A1), collective investment in POS governance (I1), and self-organizing activities (I2). These findings contribute to a nuanced understanding of the social–institutional dynamics affecting rural land commons under demographic transition, offering empirical evidence for rural land governance grounded in the SES framework.

The results suggest that challenges in managing rural POSs amid widespread out-migration arise from multiple interrelated factors. Accordingly, the study offers four policy recommendations to enhance rural POS self-governance:

(1)

Strengthen rural social capital:

Village committees and local leadership should promote the formation of farmer cooperatives and cultural organizations. These villager institutions help build social trust, foster cooperation, and enhance community capacity for self-governance;

(2)

Optimize rural industrial structures:

As agriculture remains the backbone of rural economies, its effective regulation directly supports POS self-governance. Policies should aim to formalize land transfers via written contracts, prevent land abandonment through active monitoring, and reclaim idle land for reallocation. Agricultural subsidies (e.g., for planting, processing, and marketing) and the promotion of modern technologies should be enhanced. These measures can improve collective income, strengthen villagers’ dependence on POSs, and attract return migration, thus reinforcing local governance capacity;

(3)

Empower the left-behind population:

Village committees should implement awareness campaigns and education programs to encourage capable women and elderly residents to participate in POS governance. As primary users of these spaces, their engagement can compensate for the absence of out-migrants and improve management outcomes;

(4)

Incentivize self-organizing activities:

Villages should consider introducing reward mechanisms (e.g., dividend bonuses from collective income) for individuals actively contributing to POS self-governance. Such incentives can catalyze broader participation in communal management.

This research holds both theoretical and practical significance. Theoretically, it is the first to apply the SES framework to empirically analyze the impact of external demographic factors, particularly rural out-migration, on the self-governance of the POS. To date, the SES framework has primarily been employed in the governance of traditional natural commons, such as irrigation systems, forests, and fisheries. This study, therefore, extends the application of the SES framework to a new domain of commons governance—the management of the rural POS—thereby providing a novel theoretical perspective and empirical foundation for the use of the SES framework in governing non-traditional commons. Practically, the study provides robust empirical insights into the mechanisms through which out-migration influences local land governance, offering implications for the design of rural collective land management policies under shifting demographic conditions. In particular, the study aligns with the objectives of Sustainable Development Goal 11.7, which aims to provide inclusive, safe, and accessible green public spaces, including in rural areas.

Nevertheless, several limitations should be noted. First, the study focuses exclusively on rural out-migrants who retain their rural household registration. Individuals who have formally canceled their registration were excluded, as their rural land rights are typically revoked, and they are no longer directly involved in rural POS management. However, future research should consider whether and how such groups may still influence local governance from a distance. Second, while the study identifies four key mediating mechanisms, other potential pathways may exist. Future studies could expand the model to include additional institutional or behavioral variables. Third, the study focuses on a single district in China. Although Taigu District offers a degree of representativeness, rural POS governance may differ across regions. Broader, multi-regional studies are needed to validate and generalize the findings.