Research

INTRODUCTION

Rangelands, covering 54% of the Earth’s terrestrial surface, are home to almost a billion people (Bardgett et al. 2021, UNEP 2021). However, they are largely threatened by degradation, with 49% of global rangeland areas degraded to some extent (Bardgett et al. 2021). In China, rangeland constitutes the country’s most extensive terrestrial ecosystem, covering 41.7% of the national land area, with extensive grazing remaining the primary land use (Wang 2022). However, this vital ecosystem has experienced significant degradation, with 90% of rangelands affected since the 1950s (Liu and Diamond 2005, Wu et al. 2024). Overgrazing is a principal driver of degradation, which occurs when the actual stocking rate exceeds the rangeland’s carrying capacity, as resource users maximize economic profit from livestock production (Hu et al. 2019, Su et al. 2021, Wang et al. 2022). Heavy grazing intensity has led to a 65% decline in aboveground biomass, which exceeds the global average (Yan et al. 2013). In particular, household livestock holdings in Inner Mongolia were found to be 3.2 times greater than the local rangelands’ carrying capacity (Briske et al. 2015).

Given the global challenge of rangeland degradation, a large body of literature examines the determinants of herders’ sustainable use of rangeland, including household individual characteristics, herd size, herding strategy, and rangeland management practices (Waldron et al. 2010, Li and Bennett 2019, Feng et al. 2023). Herders’ stocking rate decisions are also socially interdependent. Peer effects, whereby high stocking by peers induces similar behavior, create self-reinforcing pressure on pastures (Shi et al. 2022). A parallel literature evaluates collective action and finds that rangeland under collective arrangements often exhibits lower degradation (Tang and Gavin 2015, Li et al. 2018, Yang et al. 2021). In China, such arrangements have deep historical roots but have been weakened since the privatization of rangeland started in the 1980s (Li and Huntsinger 2011). Subsequent grazing bans and rapid market integration further eroded cooperative traditions, diminishing environmental benefits previously associated with collective management (Li and Huntsinger 2011, Yu and Farrell 2013, Hou et al. 2023).

Collective action offers a practical bottom-up approach to realigning incentives and coordinating resource use at the community level (Ostrom 2010, Wang et al. 2024). Evidence from diverse contexts shows that collective arrangements lower transaction costs, enhance bargaining power, pool infrastructure, and often deliver environmental gains, for example, in invasive species control, river basin management, and agroforestry (Markelova et al. 2009, Zulu et al. 2018, Hazard et al. 2022). In rangelands, collective action is also linked to greater equity, managerial efficiency (Hausner et al. 2012, Cai and Li 2016), and ecological recovery in some contexts. Case studies from Uzbekistan and Mongolia report increased biomass and households’ incomes under collective arrangements (Christmann et al. 2015, Oniki et al. 2018). Evidence from China similarly indicates that collective action enhances households’ livelihoods, promotes social equity, and supports ecological sustainability (Cao et al. 2018a, Yang et al. 2020, Yu et al. 2025). Outcomes, however, are contingent on enabling conditions, especially social capital (e.g., norms, communication networks, and trustworthiness) and local informal institutions, which support monitoring and compliance (Labonne and Chase 2011, Li et al. 2021, Zhou et al. 2024). Where these conditions hold, participation in and positive impacts of collective action can be broadened, and an inclusive society can thus be realized, further reinforcing both households’ livelihoods and governance capacity (Agrawal et al. 2023).

There are two primary forms of collective action in pastoral communities of China, namely cooperatives and joint management (Wang et al. 2013, Yang et al. 2020, 2021, Li et al. 2024). Cooperatives are formal, registered organizations that provide financial and technical support to their members, enabling them to achieve economies of scale and improve welfare outcomes (Tang and Gavin 2015, Wossen et al. 2017, Ma et al. 2018). By reducing transaction costs and enhancing individual bargaining power (Mojo et al. 2017), cooperatives can foster greater environmental awareness among members, thus promoting ecological sustainability (Lise et al. 2006), with evidence of gains also in carbon efficiency and rangeland utilization (Li et al. 2018, 2024). In contrast, joint management is an informal, small-scale arrangement in which a few neighboring households or relatives pool their resources (e.g., livestock and grassland) and cooperatively coordinate day-to-day herding and other pastoral practices (e.g., selling, mobility; Bijoor et al. 2006, Tan et al. 2018). Studies show joint management can generate significant ecological benefits while enhancing livestock production outcomes (Yang et al. 2021). This approach has effectively balanced environmental goals with preserving household livelihood, offering substantial ecological and economic benefits to participating herders (Grundy et al. 2000, Cao et al. 2011, 2018b, Mazunda and Shively 2015).

Nevertheless, the benefits of collective action are unevenly distributed. Empirical studies found that collective action tends to benefit more disadvantaged groups with limited education or in poor economic conditions, by providing them with sufficient knowledge, better market access, and lower transaction costs (Feleke and Zegeye 2006, Abebaw and Haile 2013, Zulu et al. 2018). Meanwhile, collective action can advantage those with less political power by constraining dominant actors and reinforcing peer oversight. For example, individuals with political status (holding a government leadership position) often adhere to the behavioral norms to maintain their social standing and rights within the village, which involves accepting oversight and balancing livestock numbers with available forage (Agrawal et al. 2023, Feng et al. 2023).

Existing case studies provide valuable insights into the effect of collective action on rangeland sustainability, but they face limited generalizability across contexts (Christmann et al. 2015, Wang et al. 2024). More importantly, existing research mainly focuses on ecological outcomes, leaving grazing behavior (i.e., whether households exceed carrying capacity) and the underlying mechanisms linking collective action to overgrazing behavior underexplored. Heterogeneous effects across household types are insufficiently specified, yet crucial for identifying priority groups and tailoring interventions.

This study examines whether and how collective action reduces overgrazing in Chinese pastoral regions, focusing on cooperatives and joint management. Using household survey data collected from 484 herders in pastoral areas in China, each household’s stocking rate is compared to locally estimated carrying capacity to determine overgrazing. The average treatment effects of participation in collective actions on the probability of overgrazing are assessed, which accounts for self-selection using propensity score matching (PSM), and heterogeneity is examined across household characteristics such as education and income. Additionally, the underlying mechanisms that affect overgrazing through collective action are explored. Our results reveal the conditions under which collective action curbs overstocking and who benefits the most from the collective action. These findings highlight the critical role of herders’ behavior in promoting sustainable rangeland management under private property regimes, and offer implications for policy makers in setting targeted interventions for diverse groups toward promoting sustainable rangeland use and fostering a more inclusive society.

METHODS

Study area

We researched Qinghai Province and Inner Mongolia Autonomous Region from 2020 to 2022 (Fig. 1A). Qinghai Province is located on the Qinghai-Tibet Plateau in western China (31°40′-39°19′N, 89°35′-103°04′E), characterized by a plateau continental climate with long, cold seasons and low oxygen levels in the air (Zhao et al. 2020). It is one of China’s major pastoral areas and is part of its five largest grazing regions (Zhao et al. 2020). The province has a total rangeland area of 36.45 million hectares (NAHS 2017), accounting for 47% of the land area in the region (Liu et al. 2018), dominated by alpine meadows and alpine steppe. The region’s ecosystems are fragile, with grassland degradation becoming increasingly prominent, overgrazing, increasing climate disasters, and pasture fragmentation as the main reasons (Dong et al. 2020, Qi 2021, Tan et al. 2025). Qinghai Province administers eight prefecture-level regions, of which four have experienced varying degrees of overgrazing (Gao et al. 2023). By 2018, the total livestock numbers exceeded the maximum carrying capacity of the rangelands by 34.5%, with 24 counties listed in the high-risk overgrazing zone, urgently requiring measures for grassland protection and carrying capacity regulation (Wei et al. 2024). Over the last decade, the number of cooperatives has increased, though individual households are still the main actors in pastoral practices. As of 2023, Qinghai Province has developed 17,600 agricultural and pastoral cooperative organizations and established 19,700 family farms. This includes 65 national-level model cooperatives, 865 provincial-level model cooperatives, and 573 model family farms (MARAPRC 2023). Studies from the region show that cooperatives integrate labor and capital, enabling rotational grazing, higher livestock productivity, and greater resilience to climate shocks (Wang et al. 2021, Yuan and Luo 2022). They also offer training, financial services, and information platforms that improve human capital and adaptive capacity (He et al. 2024). Joint management facilitates labor sharing and collective grazing, reducing costs such as fencing and water access (Cao et al. 2011), increasing mobility, supporting grassland recovery, and livelihood diversification (Zhou et al. 2021).

Inner Mongolia Autonomous Region is located in northern China (37°24′-53°23′N, 97°12′-126°04′E), with a relatively flat terrain and a cold, dry climate. The total rangeland area is 76.54 million hectares (NAHS 2017), mainly temperate typical steppe and sandy steppe. The region is facing grazing pressure. In 2023, a total of 15.61 million hectares of rangelands were designated as overgrazing warning zones, of which 37.88% were classified as severe overgrazing warning zones (NFGA 2024). As of 2024, Inner Mongolia has actively fostered new types of business entities to encourage herders to adopt pasture rotation and moderate-scale operations. Pilot initiatives in banner- and county-level shareholding cooperatives, family ranches, and joint household ranches have increased by 3471, reaching 22,000. In addition, 247 smart ranches have been newly established, achieving 59% of the annual target (Zhang 2024). In Inner Mongolia, studies show that cooperatives have helped reduce transaction costs, enhance members’ bargaining power, and improve market access, especially in remote regions (Alho 2015, Verhofstadt and Maertens 2015, Menggendalai 2023). Joint management complements these gains in climate adaptation. By coordinating grazing routes and sharing pasture during extreme weather events, joint groups reduce household vulnerability and enhance rangeland resilience (Wang 2013, Tang and Gavin 2015). In the study area, collective action shows the potential to improve household livelihoods and resilience to climate variability and foster sustainable grazing practices (e.g., rotational grazing, mobility) in Qinghai and Inner Mongolia. These prospects motivate the empirical assessment of whether, and how, collective action could foster ecological gains by reducing overgrazing.

Data collection

The household survey was conducted in four pastoral counties in the Qinghai Autonomous Region and two pastoral counties in the Inner Mongolia Autonomous Region from 2021 to 2023. Households were selected from villages representing diverse rangeland types and herd management strategies. The survey employed a multistage sampling strategy. In each selected county, 1–2 townships were chosen, and within each township, 3–4 villages were further selected. The number of surveyed households in each village was adjusted according to village size, with a target of approximately 35 households per village. In some small-scale villages, a complete enumeration approach was adopted. However, in pastoral areas, especially in Qinghai, the wide geographic dispersion of households made reaching the target number in certain villages difficult. In total, 484 valid responses were obtained. Our survey included 308 samples from Qinghai collected in 2020 (Fig. 1B) and 176 from Inner Mongolia collected in 2021 and 2022 (Fig. 1C). The survey was conducted through face-to-face interviews, including questions on household characteristics, household grazing behavior, household rangeland management, herd ownership, composition, and transactions. The head of household ranged in age from 17 to 80 years old; the average age was 45 years old, with 25% younger than 38 years old and 25% older than 52 years old. The head of households are almost always individuals who identify themselves as male (93.3%), with females accounting for only 6.7%.

Calculation of theoretical carrying capacity

The theoretical carrying capacity refers to the maximum number of livestock a unit of rangeland area can sustainably support over a defined grazing period under ecological conditions. Equation 1 calculates the theoretical carrying capacity of rangelands, considering three primary parameters: rangeland forage yield, forage utilization rate, and daily feed intake per livestock unit.

(1)

Rangeland biomass, defined as fresh or dry grass yield, is crucial for determining carrying capacity and reflects the rangeland’s ecological status. Net primary productivity (NPP, kg C/m²/year) is widely used as an indicator for biomass because it combines vegetation growth potential under climatic and soil conditions. Annual NPP data for 2021 are obtained from the MODIS17A3 dataset, and extracted at the township level using rangeland vector boundaries. Because only the above-ground biomass is available for livestock consumption, an adjustment factor f_bnpp is applied to account for the below-ground proportion of NPP. Based on measurements at 207 field sites, Sun et al. (2021) use machine learning to create a global f_bnpp map. The utilization coefficient k represents the proportion of available forage consumed by livestock during the grazing period. According to the “Calculation of Rational Carrying Capacity of Natural Rangeland” (NY/T 635-2015) issued by the Ministry of Agriculture (MOA 2015), utilization rate (k) varies by rangeland type and grazing practice. Value of f_bnpp and k used in this study are shown in Table A.1.

The parameter M represents the daily feed intake (kg C) per sheep unit, following existing research findings and official standards. According to the Ministry of Agriculture of the People’s Republic of China (2015), the daily feed intake for sheep units is 1.8 kg of dry hay (14% moisture content), equivalent to approximately 1.548 kg dry weight. To ensure consistent measurement units, all biomass data in this study are expressed in carbon mass (kg C), applying a conversion coefficient of 0.45 from plant biomass to carbon content, as applied by Fang et al. (1996) and Yang et al. (2022). Parameter D represents the annual grazing days, determined based on local grazing practices observed during field surveys. Based on forage growth periods, the theoretical grazing days are set at 300 days annually. For herders who move livestock to communal pastures in summer (approximately 60 days), the actual grazing days are set at 240 days. In the Inner Mongolia study area, herders typically graze in the warm season from May to October and practice enclosure feeding in the cold season, yielding an annual grazing period of around 180 days.

Calculation of stocking rate

The stocking rate refers to the grazing density per unit area of rangeland, converting different livestock types to standard sheep units. Herders often buy additional forage to reduce grazing pressure on their pastures, lowering the actual stocking rate. Therefore, with reference to the standards set by the MOA (2015) and incorporating necessary adjustments, the calculation of the actual stocking rate Y is shown in equations (2) and (3).

(1)

(1)

Index i represents the type of grazing livestock, and n represents the total number of grazing livestock types.

The parameter Q_i represents the quantity of the i^th type of livestock raised by the herder, while S_i denotes the amount of the i^th type sold. The coefficient Y_i indicates the conversion factor of the i^th type of livestock into standard sheep units, as presented in Table A.2. Additionally, adjustments are made to exclude livestock sold in August or September, with only the total number of animals remaining in winter counted. This winter inventory serves as the basis for assessing the actual grazing pressure, a method also commonly adopted by the government to estimate whether overgrazing occurs. The symbol A refers to the herder’s total area of pasture grazed.

The parameter P in equations (2) and (3) presents the number of standard sheep units supported by purchased supplementary feed. As supplementary feeding reduces grazing pressure on pastures, it must be subtracted from the total livestock inventory when calculating the actual stocking rate to reflect the real grazing load. Symbol m indicates the weight of supplementary feed provided by the household (kg), which is calculated from the total cost of supplementary feeding reported in the household survey and the local forage price. In Qinghai, extensive grazing is widely practiced, and the livestock carrying capacity is adjusted by accounting for the weight of winter supplementary forage. In equation (3), the actual grazing days are expressed as D, and the daily feed intake per livestock unit is denoted as M; both align with the grazing days (D) and daily feed intake (M) in equation (1). In the Inner Mongolia study area, summer supplementary feeding is minimal, and livestock are generally kept in barns during the cold season, so no adjustment for supplementary feeding is made.

According to the National Forestry and Grassland Administration (NFGA 2021), a household is defined as “overgrazing” when its grazing intensity exceeds the theoretical carrying capacity by more than 15%. In China, rangeland carrying capacity and stocking rate are measured by the number of sheep an area can support, expressed as “sheep/mu” (1 mu = 0.0667 ha).

Propensity score matching method

Complications arise because it is impossible to observe both potential outcomes for any single herder at the same time, a situation known as the “counterfactual problem” in impact evaluation literature. To address this issue, the PSM technique reduces selection bias by matching groups based on observable characteristics that predict participation in collective action (Rosenbaum and Rubin 1983, Ji et al. 2019). The PSM method has been applied to several studies related to collective action, demonstrating that collective action significantly improves outcomes such as household income, asset accumulation, and the adoption of agricultural technologies and safe practices among households (Abebaw and Haile 2013, Mojo et al. 2017, Ji et al. 2019). The PSM method has also been applied to identify the impact of credit support on grazing intensity, effectively mitigating the problem of unobservable counterfactuals (Teng et al. 2025).

The treatment variable is defined as a binary indicator denoting whether herder i participated in collective action (CA_i = 1) or not (CA_i = 0), while the outcome variable captures whether the herder engaged in overgrazing. To address potential selection bias, the propensity score (PS_i) is introduced as the conditional probability of participating in collective action, conditional on the observed characteristics X_i. The treatment effect of collective action can then be formally expressed as shown in equation (4):

(1)

The propensity score constructs a wide range of factors (X_i), driving households’ decision making to participate in collective action. Economic and social motivations affect the decision-making process for collective action practices, including individual characteristics, household compositions, financial statuses, and land management (Yang et al. 2020, Feng et al. 2023).

The average treatment effect on the treated (ATT) represents the effect of collective action on overgrazing. It estimates the difference in overgrazing behavior between herders who participated in collective action and the behavior they would have shown had they not attended:

(1)

Here, G_i¹ denotes the observed grazing outcome for herder i under participation, and G_i⁰ represents the unobserved grazing outcome that would have occurred had the same herder not participated. Because the counterfactual outcome is not directly observable, it is approximated by identifying matched non-participants with similar propensity scores. After propensity score estimation, the study utilizes Nearest Neighbor Matching (NNM) techniques. The methods facilitate a comparative analysis between participants and non-participants of collective action. A balance test is also conducted to ensure the quality of the matching process. Following standard practice (Rosenbaum and Rubin 1983, Mojo et al. 2017, Ji et al. 2019), covariates are considered well-balanced if the absolute standardized mean differences are below 0.1, corresponding to a 10% standardized difference. In addition, the change in pseudo R² and the average difference across variables are also examined to assess overall balance. These measures help determine whether the treatment and control groups became more similar after matching.

RESULTS

Summary statistics

The summary statistics of various collective action participants are shown in Figure 2, with the number of observations labeled in a Venn diagram of overlapping circles. Among all the research samples, 19.01% participate in cooperatives and 28.51% engage in joint management. Specifically, 14.46% participate only in cooperatives, 4.96% participate only in joint management, and 14.05% participate in both forms of collective action. In total, 33.47% of households participate in at least one of the two forms, which are regarded as participation in collective action in this study.

On average, the actual stocking rate slightly exceeds carrying capacity, and 42.8% of households overgraze, indicating that overgrazing is unevenly distributed across regions. Overgrazing is especially prevalent in Qinghai’s study area compared to Inner Mongolia, with Gangcha having the highest overgrazing rate of over 80% (Fig. 3). In Chenbaerhu Banner, favorable climate and rangeland types yield higher quality forage, theoretically reducing the area required per head. Across all sample counties, the average actual stocking rate is 0.24 sheep/mu (approximately 3.60 sheep/ha), while the average theoretical carrying capacity is 0.21 sheep/mu (approximately 3.15 sheep/ha).

Variables that may influence both herders’ likelihood of overgrazing and their participation in collective action are listed in Table 1, including herder characteristics (e.g., age of the household head, education level), household compositions (e.g., labor, government leadership position), economic statuses (e.g., household income, loan), as well as rangeland management (e.g., per capita owned pastures, land transfer, rotation), as they may influence herders’ behavior of overgrazing. These factors are theoretically and empirically linked to stocking decisions and cooperative behavior (Waldron et al. 2010, Li and Bennett 2019, Feng et al. 2023). Individual characteristics (e.g., age, education level), household compositions (e.g., labor, government leadership position) and economic statuses (e.g., household income, loan) can influence herders’ perceptions of climate and ecological changes (Klein et al. 2014), which condition willingness to participate in collective action (Van Gevelt et al. 2019). Land management also affects both grazing pressure and cooperation incentives. For example, larger rangelands are more likely to join cooperative rangeland management (Yang et al. 2020). Herders who transfer rangeland are more likely to participate in collective action because it ensures more transfer benefits are retained by reducing transaction costs (Zhang et al. 2017). Additionally, herders who advocate rotational grazing are more likely to participate in collective action because it enhances land access and facilitates the implementation of rotational grazing (Fabusoro 2009).

The likelihood of herders’ overgrazing is lower in the group “participants” than in their “non-” group, indicating that herders engaged in collective action are less likely to overgraze (Table 2). The two groups, however, differ significantly on several covariates in education level, labor, household income, per capita owned pasture, and rotation. Households participating in collective action usually have low education levels and enough labor. They tend to have poor economic conditions and own fewer pastures, with rotation grazing behavior. These systematic differences indicate that households’ decision making on participating in collective action is not random but affected by their household characteristics, household composition and other factors. Consequently, directly comparing the overgrazing between participants and non-participants households may underestimate the impact of collective action on overgrazing. The matching method is applied in the empirical analysis to address this self-selection.

Logit model on determinants of participation in collective action

The results of herders’ decision making on whether to participate in collective action are shown in Table 3, indicating that factors influencing collective action participation include government leadership position, household income, and rotation. Farmers’ willingness to join cooperatives also varies from social and political backgrounds (Abebaw and Haile 2013, Mojo et al. 2017). Village leaders play an important role in organizing core community groups, encouraging them more likely to participate in and facilitate collective action (Liu et al. 2020). Also, higher expected profits in collective action encourage households to engage in cooperative management, especially for those who are in poor economic statuses (Hernández-Espallardo et al. 2013, Gezahegn et al. 2019). Also, collective action can enhance land access (Fabusoro 2009), aligning with those who want to rotate grazing.

Matching balance test

After matching, the balance test presents a well-balanced distribution between the treatment and control groups. The standardized bias (% bias) after matching for most of the variables is less than 10%, and all the t-test results after matching do not reject the null hypothesis, indicating that there is no systematic difference between the treatment and control groups (Fig. 4; Table A.3). Compared to the results before matching (Unmatched), the standardized bias for all the variables have significantly reduced. Also, balance statistics before and after propensity score matching have been estimated (Table A.4). Before matching, the Pseudo R² is high (0.171), the LR chi-square value is significant (105.57), and the mean and median biases are 31.8% and 33.4%, respectively. After matching, these values significantly decrease, with Pseudo R² dropping to 0.014, LR to 6.46 (non-significant p-value of 0.693), and mean and median biases reducing to 6.2% and 5.1%, indicating improved covariate balance.

After matching, the propensity score distribution indicates considerable overlap between treated and untreated households (Fig. 5), suggesting that the common support assumption is met. Most observations in both groups fall within a similar range of propensity scores, with only a small number of “untreated” households lying outside the support region. This overlap means that the matched control group can serve as a reasonable comparison for the treated group, helping to ensure that differences in observable characteristics do not drive the estimated treatment effects.

Overall effects and robustness test

Collective action can reduce overgrazing (Table 4). The ATT indicates that participation significantly reduces overgrazing by 29.6% for those involved. Additionally, herders can benefit from collective action participants, with cooperative membership providing a 23.9% and joint management providing a 60.0% reduction in overgrazing (Column 1 in Table 5). It is aligned with previous studies that collection action can improve rangeland management (Cao et al. 2018a, Yang et al. 2020).

Heterogeneous effects of collective action on overgrazing among household groups are also estimated (Table 4). First, households with no formal schooling benefit the most from collective action, and it holds for those who are cooperative members and adopting joint management (Column 2-3 in Table 5). Second, we split households by annual income into two groups. Herders in poor economic condition (family income < = 110,000 CNY/year) experience significant reductions in overgrazing under collective action (Column 4-5 in Table 5). Third, although families with members who hold government leadership positions see more reductions in overgrazing when participating in joint management, their involvement in cooperatives does not have a significant impact. Families without leadership positions reduce overgrazing under either cooperatives or joint management (Column 6-7 in Table 5). Taken together, collective action delivers the strongest gains for less-educated and lower-income households, with cooperatives particularly effective when local political authority is unpresented.

Three methods are applied to conduct the robustness check (Table 5). First, the dependent variable is changed to “overgrazing rate,” representing the ratio of overgrazed animals to the theoretical carrying capacity. Also, the definition of overgrazing is changed from “exceeding the theoretical value by 15%” to “exceeding the theoretical value by 0%.” Second, the matching technique for PSM is changed. Specifically, Kernel Matching is applied. As displayed in Table 5, there is no significant change in either the sign or the magnitude of the effects of collective action. All the results of ATT are negative and statistically significant for all the results, indicating that collective action contributes to overgrazing reduction and confirming the conclusion in Table 4 (Column 1).

Mechanism analysis

The potential channels through which collective action influences overgrazing behavior are explored (Table 6). Collective action is regressed by rotational grazing, livelihood diversity, and the behavior-cognition gap. First, the results indicate that collective action had statistically significant and positive effects on the promotion of rotation behavior, applicable to both cooperative members and those engaged in joint management (Column 1). Second, livelihood diversity is measured as the share of non-agricultural income in total family income. Collective action increases livelihood diversity and reduces dependence on livestock income, and the results for cooperative membership and joint management are consistent (Column 2). Third, the results show that collective action significantly reduces the perception-action deviation. The deviation between herders’ perceptions and grazing practices is recorded during the field survey. Herders were asked how much pasture is needed to raise one cow (or one sheep) to determine their perception of carrying capacity, the answer of which was converted into a sheep-unit measure. Consistency was defined as the case where the actual stocking rate was equal to or lower than the perceived stocking rate, and coded as 0; deviation, where the actual rate exceeded the perceived rate, was coded as 1. Over half of the herders showed a deviation, grazing beyond their perceived capacity. Collective action reduces the discrepancy between herders’ grazing practices and their perceived rangeland carrying capacity, with consistent results for cooperative membership and joint management (Column 3).

DISCUSSION

This research demonstrates that collective action reduces overgrazing and delivers disproportionately greater benefits to vulnerable social groups. Such findings can be attributed mainly to collective arrangements that facilitate resource sharing, mutual risk mitigation, and the diffusion of knowledge on sustainable rangeland management (Brush 2007). Through cooperation, herders, especially those with fewer assets, access critical information such as market prices, sustainable practices, and weather response strategies (Cao et al. 2018b, Weng et al. 2023). This reduces their exposure to market and climate variability and helps alleviate overgrazing. In contrast, affluent herders, who possess greater social capital and resources, tend to rely less on collective actions, and leadership roles are also essential to these processes. Herders in leadership often serve as monitors, enforcing grazing bans and verifying livestock numbers (Qiu et al. 2020). Although individuals with higher political status usually follow established norms to maintain their standing (Feng et al. 2023), their dominance in cooperatives can weaken such regulatory impact. However, villager-led joint management encourages mutual oversight, including that of those with leadership roles, thereby enhancing fairness and regulatory effectiveness. Sustaining these benefits requires participatory, rather than dominant, leadership, supported by a transparent decision-making mechanism. Additionally, fostering a knowledge network among participants is crucial to improving the effectiveness of collective action. Such networks allow herders to gain management knowledge, develop new skills, learn from peers, and adopt sustainable practices (Shi et al. 2022). Integrating local knowledge with external information, building multi-actor knowledge networks, and promoting collaboration between farmers and researchers can further enhance the effectiveness of collective action (Šūmane et al. 2018).

Three main channels through which collective action reduces overgrazing are identified. First, participating in collective action effectively promotes rotational grazing, enabling members to utilize different pastures across seasons. This reduces the constraints on mobility caused by fencing and increases the flexibility of pasture use throughout the year (Fernández-Giménez 2002, Cao et al. 2013). By pooling labor and rangeland, pastoralists can balance herd size with available labor (Næss 2021). Consequently, sufficient mobility helps prevent excessive trampling associated with livestock increases (Dlamini et al. 2014) and improves seed dispersal and plant regeneration ability (Ciftcioglu 2017). Through rotational grazing, collective action activates social capital networks. It strengthens information-sharing mechanisms, reducing livestock production costs (Cao et al. 2018a, Weng et al. 2023), which may alleviate economic pressures that drive overgrazing.

Second, collective action enables herders to diversify their income sources (Tan et al. 2023). Income diversification reduces dependence on local natural resources (Asfaw et al. 2017), contributing to the mitigation of overgrazing. At the same time, the shift to non‑pastoral activities enhances living standards by increasing non-pastoral income (Liu and Zhang 2009). Moreover, it can optimize household energy consumption transitions by reducing reliance on traditional fuels and increasing electricity use, promoting sustainable livelihoods (Li and Liu 2022).

Third, collective action helps bridge the gap between households’ ecological perceptions and grazing practices. Although some herders know the severity of rangeland degradation, ecological consequences are often insufficiently accounted for in production practices (Borges et al. 2014, Li et al. 2022). By increasing adaptive measures such as destocking, seasonal grazing, and rotational grazing, collective action could mitigate this “knowing-doing” gap and promote sustainable rangeland governance.

Although collective action is particularly effective in reducing overgrazing, the development of cooperatives remains challenging and inaccessible primarily to disadvantaged households. In practice, the role of cooperatives as voluntary organizations aimed at mutual benefit has been weakened (Shen and Shen 2018). The success of cooperatives largely depends on their leaders’ administrative and managerial skills (Mojo et al. 2017). Yet, rural communities in China are experiencing rapid out-migration (Liu et al. 2016, Wilmsen et al. 2023), which strains leadership capacity in future villages. Additionally, low-income and less-educated households often encounter barriers to participating in cooperatives, primarily because of limited financial resources and insufficient knowledge. Many cooperatives exclude pastoralists who contribute only livestock grazing but not rangeland management, leaving households with limited labor capacity unable to engage in collective action. As a result, these households often resort to renting out their rangelands (Yang et al. 2020). Compared to cooperatives, joint management offers a more flexible and inclusive approach to collective action, providing ecological benefits with fewer barriers to participation. It allows households to make their own decisions about livestock management, including when and where to graze, and how to comply with grazing restrictions (Hua and Squires 2015). This approach is particularly beneficial for vulnerable groups with low income, limited education, or no connections to the local government leadership because it requires neither formal organization nor substantial financial investment. The success of joint management depends on social capital, making it more effective in communities with stronger trust (Feng et al. 2023). Small-scale joint management can also enhance ecological outcomes by fostering greater herder engagement in rangeland management (Li et al. 2007). The increased trust resulting from these relationships can lead to more effective community outcomes. Therefore, building trust in these communities is crucial to improving the benefits of such an approach.

CONCLUSION

Overgrazing is a major driver of rangeland degradation (Li et al. 2019). This study examines the impacts of collective action in mitigating and reducing household overgrazing, using household data collected from pastoral regions in China between 2021 and 2023 with a propensity score matching approach. The results show that collective action significantly mitigates overgrazing, with a 29.6% reduction compared to the counterfactual condition. Both cooperatives and joint management contribute to this reduction, decreasing overgrazing by 23.9% and 60.0% respectively. Also, the impacts of collective action vary among different household groups, with those having lower education levels, lower income, or no family members in government leadership positions benefiting the most. This illustrates the potential of collective action to foster an inclusive society by empowering vulnerable groups and strengthening community resilience. In addition, collective action promotes rotational grazing, enhances livelihood diversity, and bridges the gap between ecological awareness and grazing practices, further reducing overgrazing.

Achieving long-term sustainable development and social inclusivity requires more than financial and technical support. It also requires effective monitoring, trust building, and transparent decision making. The effectiveness of collective action can be further improved by lowering participation barriers, promoting trust building, offering supplementary incentives, fostering a knowledge network, and ensuring fair decision making. These improvements will enable greater participation and benefit for herders, particularly those from vulnerable groups, such as low-income, less educated, and politically unconnected households. By implementing these measures, collective action can facilitate sustainable resource management and promote social welfare in a more inclusive society.

LITERATURE CITED

Abebaw, D., and M. G. Haile. 2013. The impact of cooperatives on agricultural technology adoption: empirical evidence from Ethiopia. Food Policy 38:82-91. https://doi.org/10.1016/j.foodpol.2012.10.003

Agrawal, A., J. Erbaugh, and N. Pradhan. 2023. The commons. Annual Review of Environment and Resources 48(1):531-558. https://doi.org/10.1146/annurev-environ-112320-113509

Alho, E. 2015. Farmers’ self-reported value of cooperative membership: evidence from heterogeneous business and organization structures. Agricultural and Food Economics 3(1):23. https://doi.org/10.1186/s40100-015-0041-6

Asfaw, A., B. Simane, A. Hassen, and A. Bantider. 2017. Determinants of non-farm livelihood diversification: evidence from rainfed-dependent smallholder farmers in northcentral Ethiopia (Woleka sub-basin). Development Studies Research 4(1):22-36. https://doi.org/10.1080/21665095.2017.1413411

Bardgett, R. D., J. M. Bullock, S. Lavorel, P. Manning, U. Schaffner, N. Ostle, M. Chomel, G. Durigan, E. L. Fry, D. Johnson, J. M. Lavallee, G. Le Provost, S. Luo, K. Png, M. Sankaran, X. Hou, H. Zhou, L. Ma, W. Ren, X. Li, Y. Ding, Y. Li, and H. Shi. 2021. Combatting global grassland degradation. Nature Reviews Earth & Environment 2(10):720-735. https://doi.org/10.1038/s43017-021-00207-2

Bijoor, N., W. Li, Q. Zhang, and G. Huang. 2006. Small-scale co-management for the sustainable use of Xilingol Biosphere Reserve, Inner Mongolia. AMBIO: A Journal of the Human Environment 35(1):25-29. https://doi.org/10.1579/0044-7447-35.1.25

Borges, J. A. R., A. G. J. M. Oude Lansink, C. Marques Ribeiro, and V. Lutke. 2014. Understanding farmers’ intention to adopt improved natural grassland using the theory of planned behavior. Livestock Science 169:163-174. https://doi.org/10.1016/j.livsci.2014.09.014

Briske, D. D., M. Zhao, G. Han, C. Xiu, D. R. Kemp, W. Willms, K. Havstad, L. Kang, Z. Wang, J. Wu, X. Han, and Y. Bai. 2015. Strategies to alleviate poverty and grassland degradation in Inner Mongolia: intensification vs production efficiency of livestock systems. Journal of Environmental Management 152:177-182. https://doi.org/10.1016/j.jenvman.2014.07.036

Brush, S. B. 2007. Farmers’ rights and protection of traditional agricultural knowledge. World Development 35(9):1499-1514. https://doi.org/10.1016/j.worlddev.2006.05.018

Cai, H., and W. Li. 2016. A study on efficiency and equity of rangeland management on Qinghai-Tibet plateau in different property right systems. [Translated from the Chinese.] Journal of Natural Resources 31(8):1302-309.

Cao, J. J., N. M. Holden, J. F. Adamowski, R. C. Deo, X. Y. Xu, and Q. Feng. 2018b. Can individual land ownership reduce grassland degradation and favor socioeconomic sustainability on the Qinghai-Tibetan Plateau? Environmental Science & Policy 89:192-197. https://doi.org/10.1016/j.envsci.2018.08.003

Cao, J., M. Li, R. C. Deo, J. F. Adamowski, A. Cerdà, Q. Feng, M. Liu, J. Zhang, G. Zhu, X. Zhang, X. Xu, S. Yang, and Y. Gong. 2018a. Comparison of social-ecological resilience between two grassland management patterns driven by grassland land contract policy in the Maqu, Qinghai-Tibetan Plateau. Land Use Policy 74:88-96. https://doi.org/10.1016/j.landusepol.2017.07.027

Cao, J., Y. Xiong, J. Sun, W. Xiong, and G. Du. 2011. Differential benefits of multi- and single-household grassland management patterns in the Qinghai-Tibetan Plateau of China. Human Ecology 39(2):217-227. https://doi.org/10.1007/s10745-011-9384-0

Cao, J., E. T. Yeh, N. M. Holden, Y. Yang, and G. Du. 2013. The effects of enclosures and land-use contracts on rangeland degradation on the Qinghai-Tibetan Plateau. Journal of Arid Environments 97:3-8. https://doi.org/10.1016/j.jaridenv.2013.05.002

Christmann, S., A. Aw-Hassan, T. Rajabov, and A. Rabbimov. 2015. Collective action for common rangelands improvement: a climate change adaptation strategy in Uzbekistan. Society & Natural Resources 28(3):280-295. https://doi.org/10.1080/08941920.2014.933927

Ciftcioglu, G. C. 2017. Assessment of the resilience of socio-ecological production landscapes and seascapes: a case study from Lefke Region of North Cyprus. Ecological Indicators 73:128-138. https://doi.org/10.1016/j.ecolind.2016.09.036

Dlamini, P., P. Chivenge, A. Manson, and V. Chaplot. 2014. Land degradation impact on soil organic carbon and nitrogen stocks of sub-tropical humid grasslands in South Africa. Geoderma 235-236:372-381. https://doi.org/10.1016/j.geoderma.2014.07.016

Dong, S., Z. Shang, J. Gao, and R. B. Boone. 2020. Enhancing sustainability of grassland ecosystems through ecological restoration and grazing management in an era of climate change on Qinghai-Tibetan Plateau. Agriculture, Ecosystems & Environment 287:106684. https://doi.org/10.1016/j.agee.2019.106684

Fabusoro, E. 2009. Use of collective action for land accessibility among settled Fulani agro-pastoralists in southwest Nigeria. Sustainability Science 4(2):199-213. https://doi.org/10.1007/s11625-009-0082-4

Fang, J., G. Liu, and S. Xu. 1996. Biomass and net production of forest vegetation in China (in Chinese). Acta Ecologica Sinica 16(5):497-508.

Feleke, S., and T. Zegeye. 2006. Adoption of improved maize varieties in Southern Ethiopia: factors and strategy options. Food Policy 31(5):442-457. https://doi.org/10.1016/j.foodpol.2005.12.003

Feng, X., Q. Zhao, J. Li, and H. Qiu. 2023. The impact of political status and social capital on herders’ overgrazing behaviors in the pastoral areas of China. Journal of Environmental Management 327:116861. https://doi.org/10.1016/j.jenvman.2022.116861

Fernández-Giménez, M. E. 2002. Spatial and social boundaries and the paradox of pastoral land tenure: a case study from postsocialist Mongolia. Human Ecology 30(1):49-78. https://doi.org/10.1023/A:1014562913014

Gao, J., X. Zhao, W. Liu, X. Yang, C. Zhang, Y. Yu, Q. Cao, Y. Liu, X. Zhang, and Q. Dong. 2023. Spatio-temporal analysis of alpine grassland carrying capacity in Qinghai province considering supply-consumption relationship (in Chinese). Acta Prataculturae Sinica 32(5):1-12.

Gezahegn, T. W., S. Van Passel, T. Berhanu, M. D’Haese, and M. Maertens. 2019. Big is efficient: evidence from agricultural cooperatives in Ethiopia. Agricultural Economics 50(5):555-566. https://doi.org/10.1111/agec.12509

Grundy, I., J. Turpie, P. Jagger, E. Witkowski, I. Guambe, D. Semwayo, and A. Solomon. 2000. Implications of co-management for benefits from natural resources for rural households in north-western Zimbabwe. Ecological Economics 33(3):369-381. https://doi.org/10.1016/S0921-8009(00)00146-4

Hausner, V. H., P. Fauchald, and J.-L. Jernsletten. 2012. Community-based management: under what conditions do Sámi pastoralists manage pastures sustainably? PLoS ONE 7(12):e51187. https://doi.org/10.1371/journal.pone.0051187

Hazard, L., J. Locqueville, and F. Rey. 2022. A facilitation method to foster collective action in transitions toward sustainable agriculture—a case study. Agronomy for Sustainable Development 42(6):106. https://doi.org/10.1007/s13593-022-00838-9

He, X., J. Yan, L. Yang, J. Wang, H. Zhou, and X. Lin. 2024. Linking smallholders’ livelihood resilience with their adaptation strategies to climate impacts: insights from the Tibetan Plateau. Ecology and Society 29(2):7. https://doi.org/10.5751/ES-14639-290207

Hernández-Espallardo, M., N. Arcas-Lario, and G. Marcos-Matás. 2013. Farmers’ satisfaction and intention to continue membership in agricultural marketing co-operatives: neoclassical versus transaction cost considerations. European Review of Agricultural Economics 40(2):239-260. https://doi.org/10.1093/erae/jbs024

Hou, L., P. Liu, and X. Tian. 2023. Grassland tenure reform and grassland quality in China. American Journal of Agricultural Economics 105(5):1388-1404. https://doi.org/10.1111/ajae.12357

Hu, Y., J. Huang, and L. Hou. 2019. Impacts of the grassland ecological compensation policy on household livestock production in China: an empirical study in Inner Mongolia. Ecological Economics 161:248-256. https://doi.org/10.1016/j.ecolecon.2019.03.014

Hua, L., and V. R. Squires. 2015. Managing China’s pastoral lands: current problems and future prospects. Land Use Policy 43:129-137. https://doi.org/10.1016/j.landusepol.2014.11.004

Ji, C., S. Jin, H. Wang, and C. Ye. 2019. Estimating effects of cooperative membership on farmers’ safe production behaviors: evidence from pig sector in China. Food Policy 83:231-245. https://doi.org/10.1016/j.foodpol.2019.01.007

Klein, J. A., K. A. Hopping, E. T. Yeh, Y. Nyima, R. B. Boone, and K. A. Galvin. 2014. Unexpected climate impacts on the Tibetan Plateau: local and scientific knowledge in findings of delayed summer. Global Environmental Change 28:141-152. https://doi.org/10.1016/j.gloenvcha.2014.03.007

Labonne, J., and R. S. Chase. 2011. Do community-driven development projects enhance social capital? Evidence from the Philippines. Journal of Development Economics 96(2):348-358. https://doi.org/10.1016/j.jdeveco.2010.08.016

Li, A., J. Wu, X. Zhang, J. Xue, Z. Liu, X. Han, and J. Huang. 2018. China’s new rural “separating three property rights” land reform results in grassland degradation: evidence from Inner Mongolia. Land Use Policy 71:170-182. https://doi.org/10.1016/j.landusepol.2017.11.052

Li, D., L. Hou, and A. Zuo. 2021. Informal institutions and grassland protection: empirical evidence from pastoral regions in China. Ecological Economics 188:107110. https://doi.org/10.1016/j.ecolecon.2021.107110

Li, P., and J. Bennett. 2019. Understanding herders’ stocking rate decisions in response to policy initiatives. Science of the Total Environment 672:141-149. https://doi.org/10.1016/j.scitotenv.2019.03.407

Li, S., J. Jiao, A. A. Degen, W. Wang, T. Qi, M. Huang, S. Xu, and Z. Shang. 2024. Cooperative management can mitigate trade-offs between livestock production and ecological functions to promote grassland sustainability. Agriculture, Ecosystems & Environment 370:109057. https://doi.org/10.1016/j.agee.2024.109057

Li, T., L. Cui, M. Zhuang, X. Cui, Y. Yang, Y. Jia, Z. Xu, and Y. Wang. 2022. Herdsmen’s cognitive level about sustainable utilization of grassland and its influencing factors: a case study of five counties in the headwater region of the Yellow River, China. [Translated from the Chinese.] Acta Ecologica Sinica 42(20):8193-8201.

Li, W., S. H. Ali, and Q. Zhang. 2007. Property rights and grassland degradation: a study of the Xilingol Pasture, Inner Mongolia, China. Journal of Environmental Management 85(2):461-470. https://doi.org/10.1016/j.jenvman.2006.10.010

Li, W., and L. Huntsinger. 2011. China’s grassland contract policy and its impacts on herder ability to benefit in Inner Mongolia: tragic feedbacks. Ecology and Society 16(2):1. https://doi.org/10.5751/ES-03969-160201

Li, Y., J. Li, K. S. Are, Z. Huang, H. Yu, and Q. Zhang. 2019. Livestock grazing significantly accelerates soil erosion more than climate change in Qinghai-Tibet Plateau: evidenced from ¹³⁷Cs and ²¹⁰Pbex measurements. Agriculture, Ecosystems & Environment 285:106643. https://doi.org/10.1016/j.agee.2019.106643

Li, Z., and M. Liu. 2022. Livelihood diversification helps herder households on the Mongolian Plateau reduce emissions: a case study of a typical pastoral area. Agronomy 12(2):267. https://doi.org/10.3390/agronomy12020267

Lise, W., S. Hess, and B. Purev. 2006. Pastureland degradation and poverty among herders in Mongolia: data analysis and game estimation. Ecological Economics 58(2):350-364. https://doi.org/10.1016/j.ecolecon.2005.07.035

Liu, G., H. Wang, Y. Cheng, B. Zheng, and Z. Lu. 2016. The impact of rural out-migration on arable land use intensity: evidence from mountain areas in Guangdong, China. Land Use Policy 59:569-579. https://doi.org/10.1016/j.landusepol.2016.10.005

Liu, J., and J. Diamond. 2005. China’s environment in a globalizing world. Nature 435(7046):1179-1186. https://doi.org/10.1038/4351179a

Liu, J., X. Huang, X. He, and K. Shen. 2018. Estimation of grassland yield and carrying capacity in Qinghai province based on MODIS data. [Translated from the Chinese.] Pratacultural Science 35(10):2520-2529.

Liu, J., and Y. Zhang. 2009. Research on the relation between migration of rural labor force and the development of rural economy in poverty areas—statistical analysis based on the data of ten poverty villages in Gansu province. [Translated from the Chinese.] China Rural Survey 96(3):63-74.

Liu, P., Y. Zhao, N. Ravenscroft, and M. K. Harder. 2020. Responsibility-driven collective action in the context of rapid rural depopulation. Journal of Rural Studies 75:48-56. https://doi.org/10.1016/j.jrurstud.2020.02.008

Ma, W., A. Abdulai, and R. Goetz. 2018. Agricultural cooperatives and investment in organic soil amendments and chemical fertilizer in China. American Journal of Agricultural Economics 100(2):502-520. https://doi.org/10.1093/ajae/aax079

Markelova, H., R. Meinzen-Dick, J. Hellin, and S. Dohrn. 2009. Collective action for smallholder market access. Food Policy 34(1):1-7. https://doi.org/10.1016/j.foodpol.2008.10.001

Mazunda, J., and G. Shively. 2015. Measuring the forest and income impacts of forest user group participation under Malawi’s forest co-management program. Ecological Economics 119:262-273. https://doi.org/10.1016/j.ecolecon.2015.09.016

Menggendalai. 2023. From co-enrichment to co-governance: the practical path and mechanism of the effective participation of professional cooperative of herdsmen in the governance of pastoral areas. [Translated from the Chinese.] Journal of Hui Muslim Minority Studies 2:97-104.

Ministry of Agriculture (MOA). 2015. Calculation of rangeland carrying capacity. [Translated from the Chinese.] Standard NY/T 635-2015, Ministry of Agriculture, Beijing, China.

Ministry of Agriculture and Rural Affairs of the People’s Republic of China (MARAPRC). 2023. Two new agricultural business entities in Qinghai selected as national fourth-batch exemplary cases. [Translated from the Chinese.] MARAPRC, Beijing, China. https://www.moa.gov.cn/xw/qg/202302/t20230224_6421448.htm

Mojo, D., C. Fischer, and T. Degefa. 2017. The determinants and economic impacts of membership in coffee farmer cooperatives: recent evidence from rural Ethiopia. Journal of Rural Studies 50:84-94. https://doi.org/10.1016/j.jrurstud.2016.12.010

Næss, M. W. 2021. Collaborative foundations of herding: the formation of cooperative groups among Tibetan pastoralists. Journal of Arid Environments 186:104407. https://doi.org/10.1016/j.jaridenv.2020.104407

National Animal Husbandry Services (NAHS). 2017. China forage statistics. [Translated from the Chinese.] China Agriculture Press, Beijing, China.

National Forestry and Grassland Administration (NFGA). 2021. Evaluating criterion for balance of forage supply and livestock requirement (in Chinese). NFGA, Beijing, China.

National Forestry and Grassland Administration (NFGA). 2024. Inner Mongolia solidly carries out special campaign to tackle grassland overgrazing. [Translated from the Chinese.] NFGA, Beijing, China.

Oniki, S., K. Shindo, S. Yamasaki, and K. Toriyama. 2018. Simulation of pastoral management in Mongolia: an integrated system dynamics model. Rangeland Ecology & Management 71(3):370-381. https://doi.org/10.1016/j.rama.2018.02.003

Ostrom, E. 2010. Analyzing collective action. Agricultural Economics 41(s1):155-166. https://doi.org/10.1111/j.1574-0862.2010.00497.x

Qi, Y. 2021. A comparative analysis of definition and implementation of different grassland property rights from a transaction cost perspective. Beijing Chinese Rural Economy (11):55-71.

Qiu, H., L. Su, X. Feng, and J. Tang. 2020. Role of monitoring in environmental regulation: an empirical analysis of grazing restrictions in pastoral China. Environmental Science & Policy 114:295-304. https://doi.org/10.1016/j.envsci.2020.08.023

Rosenbaum, P. R., and D. B. Rubin. 1983. The central role of the propensity score in observational studies for causal effects. Biometrika 70(1):41-55. https://doi.org/10.1093/biomet/70.1.41

Shen, M., and J. Shen. 2018. Evaluating the cooperative and family farm programs in China: a rural governance perspective. Land Use Policy 79:240-250. https://doi.org/10.1016/j.landusepol.2018.08.006

Shi, Y., G. Qin, M. Zhao, Y. Cai, and C. Li. 2022. Impact of neighborhood effect on herders’ stocking rate decision: evidence from pastoral areas in northern China. [Translated from the Chinese.] China Population, Resources and Environment 32(1):155-167.

Su, L., J. Tang, and H. Qiu. 2021. Intended and unintended environmental consequences of grassland rental in pastoral China. Journal of Environmental Management 285:112126. https://doi.org/10.1016/j.jenvman.2021.112126

Šūmane, S., I. Kunda, K. Knickel, A. Strauss, T. Tisenkopfs, I. D. I. Rios, M. Rivera, T. Chebach, and A. Ashkenazy. 2018. Local and farmers’ knowledge matters! How integrating informal and formal knowledge enhances sustainable and resilient agriculture. Journal of Rural Studies 59:232-241. https://doi.org/10.1016/j.jrurstud.2017.01.020

Sun, Y., Y. Feng, Y. Wang, X. Zhao, Y. Yang, Z. Tang, S. Wang, H. Su, J. Zhu, J. Chang, and J. Fang. 2021. Field‐based estimation of net primary productivity and its above‐ and belowground partitioning in global grasslands. Journal of Geophysical Research: Biogeosciences 126(11):e2021JG006472. https://doi.org/10.1029/2021JG006472

Tan, J., Gongbuzeren, and Zelangdongzhu. 2025. International experiences of grassland management system in the evolution of pastoral social ecosystems. [Translated from the Chinese.] Acta Agrestia Sinica 33(5):1567-1577.

Tan, S., H. Du, W. Dai, and S. Wang. 2023. Overcoming the dilemma of grassland resource governance through collective action. [Translated from the Chinese.] China Population, Resources and Environment 33(2):176-187.

Tan, S., T. Li, B. Liu, and L. Huntsinger. 2018. How can sedentarised pastoralists be more technically efficient? A case from eastern Inner Mongolia. Rangeland Journal 40(3):241-249. https://doi.org/10.1071/RJ17128

Tang, R., and M. C. Gavin. 2015. Degradation and re-emergence of the commons: the impacts of government policies on traditional resource management institutions in China. Environmental Science & Policy 52:89-98. https://doi.org/10.1016/j.envsci.2015.04.023

Teng, C., K. Lyu, F. Han, and C. Zhang. 2025. Can household borrowing curb overgrazing: empirical evidence from pastoral areas of central and western Inner Mongolia. [Translated from the Chinese.] Journal of Agrotechnical Economics (4):54-70.

United Nations Environment Programme (UNEP). 2021. New atlas reveals rangelands cover half the world’s land surface, yet often ignored despite threats. UNEP, Nairobi, Kenya. https://www.unep.org/news-and-stories/press-release/new-atlas-reveals-rangelands-cover-half-worlds-land-surface-yet

Van Gevelt, T., H. Abok, M. M. Bennett, S. D. Fam, F. George, N. Kulathuramaiyer, C. T. Low, and T. Zaman. 2019. Indigenous perceptions of climate anomalies in Malaysian Borneo. Global Environmental Change 58:101974. https://doi.org/10.1016/j.gloenvcha.2019.101974

Verhofstadt, E., and M. Maertens. 2015. Can agricultural cooperatives reduce poverty? Heterogeneous impact of cooperative membership on farmers’ welfare in Rwanda. Applied Economic Perspectives and Policy 37(1):86-106. https://doi.org/10.1093/aepp/ppu021

Waldron, S., C. Brown, and J. Longworth. 2010. Grassland degradation and livelihoods in China’s western pastoral region. China Agricultural Economic Review 2(3):298-320. https://doi.org/10.1108/17561371011078435

Wang, C. 2022. Protection and restoration: China’s grassland construction and management enter a new stage. [Translated from the Chinese.] National Forestry and Grassland Administration of China, Beijing, China.

Wang, J., L. Alita, M. Jiang, Z. Nie, Q. Tu, and M. Liu. 2024. Collective action in the pastural area of Inner Mongolia, China: evidence from a lab-in-the-field experiment. Land Use Policy 144:107255. https://doi.org/10.1016/j.landusepol.2024.107255

Wang, J., D. G. Brown, and A. Agrawal. 2013. Climate adaptation, local institutions, and rural livelihoods: a comparative study of herder communities in Mongolia and Inner Mongolia, China. Global Environmental Change 23(6):1673-1683. https://doi.org/10.1016/j.gloenvcha.2013.08.014

Wang, W., X. Zhao, H. Li, and Q. Zhang. 2021. Will social capital affect farmers’ choices of climate change adaptation strategies? Evidences from rural households in the Qinghai-Tibetan Plateau, China. Journal of Rural Studies 83:127-137. https://doi.org/10.1016/j.jrurstud.2021.02.006

Wang, X. 2013. Grassland drought in the context of institution change—impacts of pastoral sedenterization, pasture segmentation and market economy. [Translated from the Chinese.] Journal of China Agricultural University (Social Sciences) 30(1):18-30.

Wang, Y., W. Lv, K. Xue, S. Wang, L. Zhang, R. Hu, H. Zeng, X. Xu, Y. Li, L. Jiang, Y. Hao, J. Du, J. Sun, T. Dorji, S. Piao, C. Wang, C. Luo, Z. Zhang, X. Chang, M. Zhang, Y. Hu, T. Wu, J. Wang, B. Li, P. Liu, Y. Zhou, A. Wang, S. Dong, X. Zhang, Q. Gao, H. Zhou, M. Shen, A. Wilkes, G. Miehe, X. Zhao, and H. Niu. 2022. Grassland changes and adaptive management on the Qinghai-Tibetan Plateau. Nature Reviews Earth & Environment 3(10):668-683. https://doi.org/10.1038/s43017-022-00330-8

Wei, Q., X. Zhao, J. Wang, B. Zhou, R. Xiao, Y. Duan, K. Jia, and W. Wang. 2024. The dynamics and early warning of the balance of forage supply and livestock requirement in Qinghai province. Journal of Lanzhou University (Natural Sciences) 60(4):480-487.

Weng, C., Y. Bai, B. Chen, Y. Hu, J. Shu, Q. Chen, and P. Wang. 2023. Assessing the vulnerability to climate change of a semi-arid pastoral social-ecological system: a case study in Hulunbuir, China. Ecological Informatics 76:102139. https://doi.org/10.1016/j.ecoinf.2023.102139

Wilmsen, B., S. Rogers, Y. Duan, and J.-H. Z. Wang. 2023. Farmer cooperatives and the limits of agricultural reform in rural Hubei. China Journal 89:1-23. https://doi.org/10.1086/722258

Wossen, T., T. Abdoulaye, A. Alene, M. G. Haile, S. Feleke, A. Olanrewaju, and V. Manyong. 2017. Impacts of extension access and cooperative membership on technology adoption and household welfare. Journal of Rural Studies 54:223-233. https://doi.org/10.1016/j.jrurstud.2017.06.022

Wu, S., J. Hao, H. Diao, Y. Su, K. Dong, C. Wang, and X. Zhao. 2024. Short-term grazing diminished ecosystem multifunctionality of grassland in northern China. Plant and Soil 505(1-2):829-844. https://doi.org/10.1007/s11104-024-06711-w

Yan, L., G. Zhou, and F. Zhang. 2013. Effects of different grazing intensities on grassland production in China: a meta-analysis. PLoS ONE 8(12):e81466. https://doi.org/10.1371/journal.pone.0081466

Yang, M., S. Dong, Q. Dong, P. Wang, W. Liu, and X. Zhao. 2020. Cooperative grassland management practices promoted by land tenure system transformation benefit social-ecological systems of pastoralism on the Qinghai-Tibetan Plateau, China. Journal of Environmental Management 261:110215. https://doi.org/10.1016/j.jenvman.2020.110215

Yang, M., S. Dong, Q. Dong, Y. Xu, Y. Zhi, W. Liu, and X. Zhao. 2021. Trade-offs in ecological, productivity and livelihood dimensions inform sustainable grassland management: case study from the Qinghai-Tibetan Plateau. Agriculture, Ecosystems & Environment 313:107377. https://doi.org/10.1016/j.agee.2021.107377

Yang, Y., Y. Shi, W. Sun, J. Chang, J. Zhu, L. Chen, X. Wang, Y. Guo, H. Zhang, L. Yu, S. Zhao, K. Xu, J. Zhu, H. Shen, Y. Wang, Y. Peng, X. Zhao, X. Wang, H. Hu, S. Chen, M. Huang, X. Wen, S. Wang, B. Zhu, S. Niu, Z. Tang, L. Liu, and J. Fang. 2022. Terrestrial carbon sinks in China and around the world and their contribution to carbon neutrality. Science China Life Sciences 65(5):861-895. https://doi.org/10.1007/s11427-021-2045-5

Yu, L., and K. N. Farrell. 2013. Individualized pastureland use: responses of herders to institutional arrangements in pastoral China. Human Ecology 41(5):759-771. https://doi.org/10.1007/s10745-013-9580-1

Yu, L., S. Qiu, Q. Chen, and L. Hou. 2025. Beyond property rights: all roads lead to sustainable grassland management. Global Environmental Change 94:103029. https://doi.org/10.1016/j.gloenvcha.2025.103029

Yuan, P., and Q. Luo. 2022. Can ecological livestock cooperatives become effective organizations to promote the transformation of herders from “natural persons” to “professional persons”? A case study based on analysis of Qinghai Lageri Ecological Livestock Cooperative. [Translated from the Chinese.] Chinese Rural Economy (6):45-64.

Zhang, M., L. Zhang, Y. Zhang, Y. Xu, and J. Chen. 2017. Pastureland transfer as a livelihood adaptation strategy for herdsmen: a case study of Xilingol, Inner Mongolia. Rangeland Journal 39(2):179-187. https://doi.org/10.1071/RJ15121

Zhang, X. 2024. Multimodal solution for overgrazing problems in grasslands in Inner Mongolia. [Translated from the Chinese.] Inner Mongolia Economic Network, 27 August. https://www.nmgsb.com.cn/system/yaowen/2024/0RQEV2024.html

Zhao, H., X. Li, D. Zhang, and R. Xiao. 2020. Aboveground biomass in grasslands in Qinghai Province estimated from MODIS data and its influencing factors. Acta Prataculturae Sinica 29(12):5-16.

Zhou, L., Y. Zhou, W. T. De Vries, Z. Liu, and H. Sun. 2024. Collective action dilemmas of sustainable natural resource management: a case study on land marketization in rural China. Journal of Cleaner Production 439:140872. https://doi.org/10.1016/j.jclepro.2024.140872

Zhou, X., R. Dong, C. Qin, Z. Tang, L. Wang, and Y. Zhang. 2021. The impact of multi-household rangeland operation on non-agricultural employment and household income in the Qinghai-Tibet Plateau: an empirical analysis based on Qinghai and Gansu survey data. [Translated from the Chinese.] Pratacultural Science 38(12):2524-2535.

Zulu, L. C., E. A. Adams, R. Chikowo, and S. Snapp. 2018. The role of community-based livestock management institutions in the adoption and scaling up of pigeon peas in Malawi. Food Policy 79:141-155. https://doi.org/10.1016/j.foodpol.2018.06.007