Research

INTRODUCTION

The economic and environmental importance of common-pool resources in many Global South countries has motivated governments to adopt community-based natural resource management (CBNRM) programs, wherein rural people manage collectively owned natural resources through participatory processes (Cook et al. 2023, Cook 2024). These programs are designed to protect fragile environments, conserve ecosystem services, and support rural development. However, rural livelihoods remain highly vulnerable to environmental uncertainty and resource scarcity (Steffen et al. 2018), thus prompting many households to rely on out-migration as a livelihood adaptation strategy.

CBNRM relies on voluntary participation, which makes it essential to understand the factors that influence engagement in these programs (Cook 2024). For example, in Nepal’s community forestry program, rural villagers engage in afforestation, forest monitoring, and rule enforcement to sustain local forests (Cook et al. 2023). Studies have highlighted the success of such initiatives in improving conservation outcomes (Oldekop et al. 2019), but those initiatives also play a crucial role in rural development. More than one billion people globally live in or near forests, and many depend on forest resources for daily subsistence and income (Newton et al. 2020). Community forestry can further benefit rural economies by generating revenue through the sale of forest products, which finances local public goods such as schools, roads, and other important infrastructure (Pokharel et al. 2007, Dongol et al. 2009, Ojha et al. 2009, Cook 2024).

The degree to which households participate in CBNRM is shaped by economic incentives and livelihood strategies. In Nepal, participation in community forestry often involves membership in one of the approximately 22,000 community forest user groups, which collectively manage 35% of the country’s forests (Gentle et al. 2020, Cook et al. 2023, Cook 2024). Beyond membership, participation can entail taking on leadership roles, engaging in forest management activities, or having a voice in the decision-making processes (Molinas 1998, Agarwal 2001, 2016, Cook 2024). Many rural households depend on forests for firewood, fodder, and non-timber forest products (Angelsen et al. 2014, Cook 2024), which creates strong incentives for engagement in local governance. However, as households gain access to alternative income sources, such as remittances from migrating family members, those incentives may shift, raising concerns about declining participation in CBNRM.

Out-migration has become increasingly common in rural areas, leading many households to transition away from subsistence-based economies toward remittance-dependent livelihoods. As rural households increasingly rely on remittances, global financial flows reflect this shift. In 2018, low- and middle-income countries received 529 billion U.S. dollars in remittances, a nearly 10% increase from the previous year (World Bank Group 2018). In some regions, migration has reduced reliance on local agriculture and common-pool resources because remittances provide an alternative source of income (Marquardt et al. 2016).

Despite increasing attention to migration’s role in rural economies, little is known about how remittance income affects participation in community-based resource management. While research has linked out-migration to labor shortages that reduce participation in collective resource management (e.g., Bista et al. 2023), an alternative hypothesis suggests that remittance income may weaken household members’ incentives to engage in CBNRM institutions by reducing reliance on common-pool resources for both subsistence and small-scale commercial activities (Robson and Nayak 2010, Shrestha and Fisher 2017, Poudyal et al. 2023). However, few studies have systematically tested the effect of remittance income on participation in CBNRM across different localities, or estimated the magnitude of the relationship.

We address this gap by using nationwide survey data from Nepal to examine the relationship between remittance income and two household outcomes: community forest use and participation in community forestry. We define out-migration as the relocation of one or more household members outside the village while the rest of the household remains in place, rather than full-household migration. This distinction is important in our study context, where partial household migration is the dominant pattern and it aligns with our theoretical framework, which emphasizes the role of remittance income in reshaping economic incentives for participation in CBNRM.

By leveraging nationwide panel survey data that spanned nearly a decade, we systematically evaluate how remittance income influences household engagement in community forestry, and distinguish this effect from related factors such as labor shortages due to migration. While recent research has linked out-migration to labor shortages that reduce participation in collective resource management (e.g., Bista et al. 2023), an alternative possibility is that remittance income plays a more decisive role in reducing the use of community resources and participation in community resource management. Thus, we empirically test which mechanism has a stronger influence on participation in Nepal’s community forestry program. This distinction is crucial for policy design because it suggests that governments that are seeking to sustain participation in CBNRM models in high-migration environments may need to develop financial incentives or institutional adjustments that reflect evolving economic realities.

A growing body of qualitative and case study research suggests that multiple factors, including remittance income, labor shortages, shifts in agrarian economies, and preferences for traditional resource management over government-led programs, may contribute to declining participation in community-based management (Poudel 2019, Poudyal et al. 2023). However, few studies have systematically examined the specific role of remittance income in shaping participation in CBNRM at a national scale or have sought to disentangle its effects from other migration-related factors. Our study fills that gap by providing empirical evidence on the relationship between remittances and participation in community forestry, and contributing to broader debates on migration, rural livelihoods, and common-pool resource governance.

A theoretical model of the effects of out-migration and remittances on participation in community-based natural resource management, tested on empirical data from community forest systems, is important for policymaking and practice in the forestry sector. Since community engagement is fundamental to the sustainability of these governance models, socioeconomic changes that weaken participation in CBNRM programs may threaten their long-term viability. If remittance income reduces household members’ incentives to engage in collective governance, CBNRM institutions may face declining participation, which will require policymakers to consider adaptive strategies that sustain engagement amid economic and demographic shifts. Our study contributes to these discussions by testing a theoretical model of how out-migration and remittances affect participation in community-based resource management, using empirical data from Nepal’s community forestry system. Strengthening institutional support for rural communities that are experiencing these transitions will be essential to ensuring the resilience of community-based governance models.

RESEARCH CONTEXT

We use community forestry in Nepal as a test case through which to explore the relationships between remittance income, community forest use, and community forestry participation. In Nepal, the community forestry initiative has been implemented under the Forest Act of 1993 (Kanel and Kandel 2004). The Forest Act ordered the Department of Forests to establish community forest user groups in forested rural communities, starting in the 1990s (Ministry of Forests and Soil Conservation 2013, Cook et al. 2023, Cook 2024, Cook et al. 2025). As of 2020, there were more than 22,000 community forest user groups across Nepal (Gentle et al. 2020). These groups managed approximately 35% of Nepal’s forest resources, through the participation of approximately 3 million member-households (Gentle et al. 2020). Community forest user groups exist across the three ecological regions in Nepal—the northernmost Mountain Region of the Himalaya, the southern Terai Region that borders India, and the Middle Hill Region (Cook et al. 2023, Cook 2024).

Previous research has described how community forestry is governed (Ojha et al. 2009, Cook et al. 2023, Cook 2024, Cook et al. 2025). Once a community forest user group is established, collective property rights over a plot of forested land are formally granted to the group. The community forest user group is charged with governing the communal use of forest products—mainly firewood, fodder, and other non-timber forest products—by its respective member-households. Since community forest user groups are largely self-governed, they are tasked with writing their own rules, establishing rationing systems and collecting royalties on allowable forest products, and engaging in monitoring, forest maintenance activities, and enforcement (Ojha et al. 2009,Cook et al. 2023, Cook 2024).

As in many low- and middle-income countries, rural livelihoods in Nepal depend on natural resources, but they also depend on remittances from out-migration. In 2020, the country received 7.4 billion U.S. dollars in remittances, accounting for 23% of its GDP (World Bank Group 2020). Nearly 50% of households in Nepal have at least one member working abroad (International Organization for Migration 2019). Nepal is an ideal test case for understanding the relationship between migration, remittance income, and participation in community-based natural resource management. Not only is the community forestry program mature, large, and well-institutionalized, but widespread rural out-migration during our study period—coupled with the fact that forest dependence has historically been high in rural Nepal—makes the country a most likely case for detecting these relationships. Furthermore, recent scholarship argues that community forestry participation is likely declining in some parts of Nepal, and qualitative evidence suggests that this decline might be driven by out-migration and reduced reliance on forests, among other factors (Poudel 2019, Poudyal et al. 2023). Thus, we use nationwide survey data and econometric methods to examine the role that remittance income may be playing in this changing context.

Our empirical analysis focuses on the study period of 2003–2012, due in part to the availability of large survey datasets from that period (see Methods). However, because our goal was to use Nepal as an empirical test case for exploring the general, theorized relationship between remittance income and community forestry engagement (established in Literature Review), this time period is particularly appropriate because there was a dramatic influx of remittance income as a percentage of GDP (World Bank Group 2025). Much of this was driven by out-migration from the types of rural communities that use community forest management models (Giri and Darnhofer 2010). Thus, although our data are from 2003–2012, we argue that Nepal during this study period provides a test case through which we may understand social processes that likely weaken participation in CBNRM in settings beyond Nepal that experience similar social and economic dynamics.

LITERATURE REVIEW

We explore the dynamics of rural out-migration and its implications through two key lenses. First, we review the current literature on “push” and “pull” factors in rural out-migration. This review spans the field of development studies, including development economics, to provide a comprehensive understanding of how scholars have studied the factors driving rural out-migration in the Global South context. Second, we review the relationship between rural out-migration and collective action in the context of common-pool resource management. This examination draws extensively on scholarship from the fields of common-pool resource governance and collective action, and offers insights into how rural out-migration influences and interacts with collective efforts to manage shared resources.

Push and pull factors in rural out-migration

Theory on rural out-migration has established that the motives for livelihood diversification through out-migration vary significantly across socioeconomic groups (Lambin et al. 2001, Zimmerer 2010). There is an important distinction between out-migration undertaken to manage risks and cope with environmental stressors on natural resources, characterized primarily by “push” factors, and out-migration undertaken for wealth accumulation, hence driven by “pull” factors (Reardon et al. 2007). The literature examined these push and pull factors by focusing on patterns of household adaptation through income diversification strategies in the Global South (García-Barrios et al. 2009, Hoffmann et al. 2019, Leblond 2019). While out-migration driven by push factors is usually associated with households’ adaptation to poverty and consumption and risk smoothing (de Janvry et al. 1991, Dressler et al. 2016), out-migration driven by pull factors is usually positively associated with an upward spiral of household wealth (Barrett et al. 2001a, Gray 2009).

Rural households are pushed to out-migrate to cope with environmental risks, especially where missing insurance and credit markets often lead households to pursue different coping strategies against uncertainty in resource availability (Barrett et al. 2001b). Empirical evidence confirms that a key factor that pushes households to migrate and seek nonfarm livelihoods is a decline in seasonal income from farm-related activities (Abdulai and Delgado 1999). Therefore, remittances from seasonal out-migration allow these households to smooth their income inter-seasonally (Von Braun et al. 1990, Reardon et al. 2007). This type of out-migration is in fact not a means of coping with a shock, but is a planned, ex-ante adaptation to a long-term seasonal variation in resource availability and income. A second push factor for out-migration is a transitory decline in income due to an unexpected stressor (e.g., drought) that forces households to out-migrate as an ex-post adaptation strategy (Choithani et al. 2021). Out-migration is particularly prominent in rural communities where resource-based livelihoods constitute the dominant economic activity. This is because households depend more on remittances that are not subject to environmental risks that are covariant with those of the local agricultural economy (Poapongsakorn et al. 1998, Barrett and Swallow 2006).

On the other hand, households in resource-rich areas are more likely to out-migrate to pursue attractive income diversification opportunities. For example, in the wetter and more stable agricultural zones of West Africa, households are more likely to out-migrate and diversify into nonfarm activities (Reardon et al. 1992, Haggblade et al. 2010). Indeed, empirical evidence shows that high-income households in buoyant rural economies are more likely to diversify into non-resource-based livelihoods (e.g., food processing and preparation, farm equipment repair, manufacturing) that have high returns because they have the necessary financial capital and skilled labor to pursue these profitable activities (Haggblade et al. 2010, Loison 2016). In such instances, there is evidence of a Markovian process whereby households invest remittances from out-migration into activities that enhance their resource productivity and/or human capital, such as technology upgradation, cash cropping, education, and further rounds of out-migration and income diversification (Estudillo and Otsuka 1999, Mohapatra et al. 2006, Robson and Berkes 2011, Hajjar et al. 2021).

Linking rural out-migration and collective action

Debates about how out-migration impacts collective management of natural resources have never been definitively settled, in part because there are multiple forms of migration in different resource contexts and participant groups (Connell and Conway 2000, Choithani 2017, Bhattarai 2020). We identify four key mechanisms through which out-migration may shape collective action in common-pool resource governance:

• Labor constraints: Increased out-migration reduces labor availability for managing shared resources, which increases the costs of collective action (Cárdenas et al. 2017, Shin et al. 2022).
• Resource dependence: Households that receive remittances may become less reliant on common-pool resources, which lowers their incentive to participate in community-based management (Wang et al. 2016).
• Opportunity costs: Increased access to external incomes raises the opportunity costs of engaging in collective management, thus making participation less attractive (Rudel 2011, Sapkota et al. 2020).
• Compensatory participation: In some cases, remaining household members (e.g., women) may step into leadership roles and contribute more to collective action efforts (Hecht et al. 2015, Leder et al. 2024).

These mechanisms highlight the ambiguous relationship between out-migration and collective action, with some pathways reducing participation while others may sustain or even increase engagement. Furthermore, participation in collective action may have feedback effects on migration decisions: communities with well-functioning collective institutions may provide local economic opportunities that reduce the need for out-migration, while declining participation in resource management may degrade natural resources and further incentivize migration.

To clarify these relationships, we present a conceptual figure on these mechanisms and feedback loops (Fig. 1). This figure provides a structured framework for interpreting how out-migration influences participation in CBNRM and how these relationships may reinforce or weaken each other over time.

Participation in CBNRM is shaped in part by cost-benefit trade-offs. Households weigh the excludable benefits of participation, such as access to forest products and financial capital, against the costs, including time spent in meetings, resource maintenance, and opportunity costs (Ministry of Forests and Soil Conservation 2013, Bluffstone et al. 2020, Cook 2024). Remittances influence these trade-offs by reducing dependence on community-managed resources and making alternative income sources more viable (Shrestha and Fisher 2017, Fox 2018). Consequently, households that receive remittances may opt out of participating in CBNRM, thus increasing the marginal cost of management for poorer households that are reliant on those resources (Angelsen et al. 2014, Nguyen et al. 2015, Cook 2024).

While the research identifies four key mechanisms through which out-migration may shape collective action, we focus on how remittance income influences two of them: household use of community forests, and participation in CBNRM activities. These theoretical propositions lead to the following two hypotheses:

H1: Households that receive more remittances will be less likely to use collectively managed natural resources.

H2: Households that receive more remittances will participate less in community-based natural resource management.

METHODS

To test our hypotheses, we used data from two sources: the Nepal Living Standards Survey (NLSS) and the Poverty Environment Network (PEN) survey initiative. The NLSS is a national-level, multi-topic household survey, collected during three periods: NLSS-I (1995–1996), NLSS-II (2003–2004), and NLSS-III (2010–2011) (Central Bureau of Statistics 1996, 2005, 2011a). The PEN initiative is a global comparative survey of households in 334 communities across 24 Global South countries (PEN 2016a). We used PEN data collected in Nepal, which provides socioeconomic, institutional, and environmental data at the household level from sampled locations in Mustang District and Gorkha District. We tested H1 by using data from the NLSS to fit a household-level model that predicted the collection of forest products from a local community forest as a function of the amount of remittance income received by the household, and we tested H2 by using the PEN data to fit a household-level model that predicted time allocation to community forest user group activities as a function of the amount of remittance income received. While previous analyses modeled forestry outcomes as a function of changes in migration at the aggregate level (Oldekop et al. 2018), our household-level analysis allowed us to measure household-level decisions, and to draw inferences about micro-level processes that link migration to community forest use and participation.

Testing the relationship between remittances and community forest use

The model used to test H1 was fit on an analytic sample of 3661 households in 338 rural communities, drawn from two distinct cross-sections from the 2003–2004 and 2010–2011 periods of the NLSS. Equation 1 represents this regression model:

(1)

The dependent variable (community forest use) is a dichotomous measure of whether the household reported collecting firewood or fodder from a community forest in the past 12 months, since these are two of the key products available to participants in community forestry (Agarwal 2010, Ministry of Forests and Soil Conservation 2013). Here, clogit represents the standard conditional fixed effects logistic regression model described in Chamberlain (1980), and π represents a vector of community-level conditional fixed effects. To measure remittance income, we used a variable (remittances) that represented the amount of remittance income the household received during the past year based on the household’s self-reported data. We converted these amounts to U.S. dollars using exchange rates from the year of each survey period (Central Intelligence Agency 2022) and converted them to their dollar equivalents in January 2022 (U.S. Bureau of Labor Statistics 2022). To facilitate interpretation, we rescaled this variable to be represented in hundreds of dollars. This variable includes both domestic and international remittances. While it is true that international migration tends to yield larger amounts of remittance income than does domestic migration in our study context, our theoretical model suggested that receiving a given amount of remittance income will shift the perceived costs and benefits of participating in community forestry similarly, regardless of whether the source is international or domestic. Thus, we used absolute amounts of remittance income from all sources as our independent variable of interest.

We controlled for several household-level covariates. First, we controlled for whether the household belonged to the ethnic or caste group that was most numerous in the community (ethnic majority). Because ethnicity is assigned at birth, this variable is exogenous to migration and the receipt of remittances. Previous studies have found ethnicity and caste to be highly related to community forestry participation rates and the distribution of benefits from community forestry, which makes this an important covariate to control for (Agarwal 2016, Cook 2024). We also included a dichotomous educational attainment variable (education) that measured whether the oldest living male household member received any formal schooling. We added this control because educational attainment is a widely recognized proxy for a household’s economic background and has significant implications for economic behavior and decision-making processes (Duflo 2001). Additionally, we controlled for household size (measured as the number of individuals living in the household). Controlling for household size is important because it affects the distribution of resources within the household and the overall economic burden. Larger households might have different consumption needs and income-generating capacities compared to smaller ones (Lanjouw and Ravallion 1995).

Finally, we controlled for three additional income sources: respondent households’ net income from crop sales (crop income), the sale of animal products (animal income), and enterprises owned by household members (enterprise income). We measured these three covariates in the NLSS data using households’ reported gross income from each activity type during the year preceding the survey, minus their reported expenditures for inputs related to each activity type during the same period. Like the remittance income variable, we converted these values to January 2022 U.S. dollars and scaled them so they were expressed in hundreds of dollars. Including these income sources allowed us to account for variations in household economic activities and their impact on overall income, thereby addressing potential biases from unobserved heterogeneity (Morduch 1995, Dercon 2002). While our list of controls did not capture every possible household characteristic, they were chosen based on their relevance to our research questions and their empirical support in existing literature. Our approach balanced the need for parsimony in model specification with the inclusion of key variables that were critical for our analysis.

We calculated nonparametric bootstrap-clustered standard errors using the cluster resampling method recommended by Cameron et al. (2008) to correct the confidence intervals and P values reported for our logistic regression model. In line with this method, communities were resampled with replacement, coefficient estimates were calculated for each replication, and standard errors were calculated from the resultant distribution using the procedure provided in Cameron et al. (2008:416).

This modeling approach predicted the relative odds of a household’s community forest use as a function of the amount of remittance income received by that household, while holding the household-level control variables constant. Additionally, community fixed effects held community characteristics constant in the model. Because our analytic sample from the NLSS was from two distinct, cross-sectional samples of communities across the two periods included in our study, the community fixed effects were equivalent to community–year fixed effects, and therefore also controlled for potential confounding differences between the two time periods that may have influenced the estimates. This analysis of the NLSS microdata thus allowed us to estimate the relationship between remittance income and community forest use at the level of the individual household while controlling for household characteristics, community fixed effects, and time effects. However, because remittance income was not assigned at random to households, there is still the possibility of omitted household-level variables that may have biased the estimated relationship between remittance income and community forest use. We therefore cannot assume that the causal effect of remittance income on community forest use was necessarily identified in our model, and we treat our analysis as correlational. In Appendix 1, we replicated the results while controlling for additional household characteristics.

Testing the relationship between remittances and participation

To explore the relationship between remittance income and participation in community forest user groups, we used the Nepal subsample of the PEN data. Our analytic sample consisted of 453 households across five communities in Mustang District and two communities in Gorkha District, surveyed in 2005–2008. While the PEN survey protocol involved interviewing households at multiple different time points, each household was asked about community forest user group participation during only one of those interviews. Our sample is therefore cross-sectional. In the villages surveyed, households in Mustang were asked about their participation in 2005. In Gorkha, households were asked about their participation in 2008.

To estimate the association between remittances and participation in community forestry, we used a zero-inflated negative binomial regression model. It was estimated in two stages, represented by Equation 2 and Equation 3:

(1)

(1)

The first stage of this model (Eq. 2) predicted the logged odds of whether or not the respondent household reported zero days of participation in community forest user group activities over the past 12 months (nonparticipation). Then, for the subset of respondent households that reported a non-zero number of days spent participating, the second stage (Eq. 3) predicted their degree of participation, measured by the number of person-days that household members reportedly spent on community forest user group activities over the past 12 months (days). One person-day was equivalent to one full working day for a single household member (see PEN 2016b). Here, logit refers to the standard logistic regression model, NBin refers to a negative binomial regression model, and σ is a dispersion parameter.

To measure remittances in the PEN data, we used the same technique as for the NLSS data. The independent variable used in this model represented the total annual amount of remittances received by the household from individuals living outside the village during the survey year, expressed in hundreds of dollars. In both stages, the zero-inflated negative binomial regression model included the same household-level controls as described previously. δ and λ represent vectors of unconditional community-level fixed effects included in the first and second stages of the model, respectively. Because the analytic dataset was cross-sectional, the community fixed effects were equivalent to community–year fixed effects, and thus controlled for differences between communities as well as potentially confounding temporal differences between the 2005 time period (when the Mustang households were surveyed) and 2008 (when the Gorkha households were surveyed). Although there were seasonality differences between the Mustang and Gorkha households, with the former surveyed in December and the latter surveyed in March, our community-level fixed-effects controlled for seasonality differences since the season in which households were surveyed varied across communities but not between them; thus, the community-level fixed-effects addressed the effects of seasonality. We corrected the reported confidence intervals and P values for community-level clustering by calculating cluster-robust standard errors using the sandwich estimator described by Rogers (1993).

As before, this analysis controlled for key characteristics of households, as well as community fixed effects, but we could not rule out the possibility of unmeasured confounding variables at the household level in this correlational analysis. However, in Appendix 1, we replicated the results of this analysis while controlling for additional household characteristics.

Characteristics of samples

Table 1 shows summary statistics for each analytic dataset, and Fig. 2 maps the locations of surveyed households at the district level. Our analytic sample from the NLSS spanned all three ecological regions of the country—the northernmost Mountain Region, the Terai Region that borders India, and the Middle Hill Region, which lies between. Furthermore, rural households from almost every district were included in the sample. By contrast, our analytic sample from the PEN survey covered the districts of Gorkha and Mustang. Thus, while the Middle Hill Region and Mountain Region were reflected in our test of H2, the Terai Region was not because it was not covered by the PEN survey. Readers should note this limitation. While our test of the relationship between remittance income and community forest use was based on a large analytic sample with substantial geographic and temporal coverage, our test of the relationship between remittances and community forestry participation was somewhat more limited, and it is possible that the results of our test of H2 do not generalize to the Terai Region.

Although the PEN sample did not cover the Terai region, the trends in our variables of interest were largely similar across both datasets. As shown in Table 1, the NLSS and PEN samples exhibited similar trends in household income portfolios, education, ethnic majority status, and household size. Both samples showed substantial average remittance incomes ($98 for the NLSS sample and $535 for the PEN sample). On average, households earned more from enterprise income ($321 for the NLSS sample and $104 for the PEN sample) than from livestock, and the average household in both samples experienced net negative crop incomes (-$43 for the NLSS sample and -$48 for the PEN sample). More than 50% of households in the NLSS sample and nearly 42% in the PEN sample had formal education. More than 50% of households in the NLSS sample and more than 60% in the PEN sample belonged to the local ethnic majority. The average household size was 5.2 persons for the NLSS sample and 6.2 persons for the PEN sample.

RESULTS

Remittance-receiving households are less likely to use community forest resources

Table 2 shows the conditional logistic regression results from the household-level model fit on the NLSS data. Estimate (a) shows the estimated relationship between a household’s remittance income and their relative odds of collecting forest products from a community forest. This estimate suggests that remittances have a negative and statistically significant association with the likelihood of collecting products from a community forest (P < 0.006). We calculated the average magnitude of this association on the probability scale using the average semi-elasticity method (Kitazawa 2012). On average, an increase in remittance income of one standard deviation predicted a roughly -8% change in the probability of reporting community forest use (95% CI: -14%, -2%). While the purpose of our model was to estimate the relationship between remittances and community forest use while controlling for the covariates, some readers may also find it worth noting that household size was the only covariate with a statistically significant test statistic (in the positive direction). Estimates on the ethnic majority and education variables exhibited wide confidence intervals.

Remittance-receiving households spend less time on community forestry participation

Table 3 shows the results from the zero-inflated negative binomial regression model fit on the PEN survey data. Estimate (c) in Table 3 represents the estimated relationship between a household’s remittance income and the number of person-days spent by household members on community forest user group activities (expressed as an incidence-rate ratio), and estimate (b) represents the relationship between remittance income and the odds of a household reporting participation at all. Estimate (c) indicates that remittances had a statistically significant negative association with the number of person-days spent on community forestry activities among households that reported participating.

These model results are easiest to interpret through model predictions expressed as the raw number of predicted person-days that households spent on community forestry activities (Fig. 3). These predictions suggest that on average, households that did not receive remittances spent just over four person-days per year on community forestry activities. In contrast, households that reported receiving US$2500 per year in remittances (a difference of roughly two standard deviations, relative to households that did not receive remittances) were predicted to spend three person-days per year on community forestry activities, or nearly 25% less time relative to households that did not receive remittances. In our PEN survey sample, 13% of remittance-receiving households reported receiving this amount or more.

In addition to the statistically significant relationship between remittance income and community forestry participation, three covariates had statistically significant test statistics in the first stage of the zero-inflated negative binomial regression, with negative signs: household size, education, and ethnic majority. Confidence intervals for other covariate coefficient estimates were generally wide.

Data limitations and alternative specifications

Our econometric models controlled for a range of covariates that could confound the relationship between remittance income, community forest use, and community forestry participation. In addition to the observable household-level characteristics discussed in the Methods, fixed effects controlled for unobservable and observable confounding variables at the community–year level. Nonetheless, it was not possible to rule out all potential confounding variables in this observational study, particularly if they operated at the household level (rather than the community level or community–year level). For example, the 2003–2004 period of the NLSS coincided with a Maoist insurgency that occurred across rural Nepal, and some evidence suggests that local exposure to the conflict impacted the governance of community forest user groups (Nightingale and Sharma 2014). Local conflict intensity, if it is indeed a confounding variable, is most likely to be a community-level confounding variable rather than a household-level confounding variable; therefore, it is less likely that our results were confounded by local exposure to the conflict once community-level fixed effects were controlled for. Nonetheless, if household-level conflict exposure was correlated with households’ remittance income and with their decisions to use community forests or participate in community forestry for enough households in our samples (after controlling for community–year fixed effects and our other covariates), then the variable would cause household-level confounding that we would be unable to control for.

Thus, since our analysis was based only on observable household-level covariates, it is possible that certain omitted household-level social or economic characteristics influenced our results. For example, gender, caste, and ethnicity are likely correlated with remittance income and are known to influence decisions about community forestry participation and benefits from community forestry (Agarwal 2016, Cook et al. 2023, Shrestha et al. 2023, Cook 2024). Furthermore, the number of remittance-senders, not remittance income, may reduce the likelihood that a household will use collectively managed natural resources. All of these covariates are measurable in the NLSS data, and gender is measurable in the PEN data, whereas the other covariates are not (in both datasets, gender is operationalized through the gender of the household head). In Appendix 1, we replicated our results while controlling for these characteristics; our results were stable when these additional covariates were included in our models.

Because we cannot completely rule out the possibility of unobservable household-level confounding variables in our study, future studies should use more robust, quasi-experimental research designs to further explore the relationships between out-migration, remittance income, and the use and governance of collectively managed natural resources.

DISCUSSION

Empirical insights

We provide quantitative evidence that out-migration, via remittance income, influences both the use of shared natural resources and participation in CBNRM. Our findings from the forestry sector in Nepal support two key mechanisms. First, households that receive remittances are less likely to depend on products from community forests. As rural livelihoods shift from resource-based activities to urban employment and service-sector jobs (Jaquet et al. 2019), reliance on shared natural resources diminishes. This reduced dependence weakens households’ incentives to participate in collective management.

Second, and relatedly, remittance income is associated with lower levels of participation in CBNRM activities. While research has suggested that multiple factors, such as out-migration-related labor shortages, socioeconomic transformations, and resistance to government schemes, influence participation in CBNRM (Poudel 2019, Shahi et al. 2022, Poudyal et al. 2023), our study isolates the effect of remittance income, which is distinct from the number of out-migrants. These findings contrast with those of Bista et al. (2023), who hypothesized a link between community forestry participation and remittance income but found no significant statistical evidence. Our zero-inflated negative binomial regression model suggests that while remittance income does not reduce the probability of participation, it significantly reduces the amount of time households allocate to community forestry activities. This may be because households that receive remittances maintain superficial ties to community-based institutions, such as occasionally attending community forest user group meetings, while drastically reducing their engagement in actual forest management activities. This would align with previous qualitative research that has shown that in some communities where participation has dwindled, members or even community forestry leaders maintain nominal ties to community institutions but contribute little time to resource management (Poudyal et al. 2023). The moderate magnitude of the estimate indicates that in settings where many households receive a large amount of annual remittance income, as is the case for some households that receive remittances from international locations (Central Bureau of Statistics 2011b), we would expect to see a noticeable reduction in the amount of time and effort allocated to community forestry. The estimate does not, however, suggest that there are substantial changes in participation by households that receive only a modest amount of such income (as is common for some other households, particularly those that receive remittances from domestic locations).

It is also notable that these relationships hold when controlling for the number of out-migrants, and the number of out-migrants has no apparent relationship with our forestry-sector outcome variables, while controlling for the amount of remittances received (see Appendix 1). This is in contrast to some other empirical work that has shown that decisions to implement many other conservation activities in the agricultural sector show the opposite pattern: they appear to be influenced by the number of out-migrants and not the amount of remittances received (Williams and Paudel 2020). This highlights the importance of developing sector-specific models for understanding the social and economic dimensions of conservation decision-making, rather than assuming that those social and economic dimensions are the same across sectors.

CONCLUSION

Implications for policy and future research

Although our study was based in Nepal, the findings have broader relevance for community-based resource management programs globally. The effectiveness of CBNRM depends on sustained participation for monitoring, enforcement, ecosystem restoration, and rural development. If participation declines due to socioeconomic shifts linked to out-migration, policymakers may need to reconsider how to sustain collective governance structures.

One potential response is to develop targeted programs that strengthen incentives for participation among households that receive remittances. Research suggests that women left behind by out-migration face participation constraints shaped by caste and social networks (Shrestha et al. 2023). To address these challenges, pro-poor entrepreneurship strategies in community-based resource management (Paudel 2012) could help marginalized households derive economic benefits from forestry. Effective implementation would require partnerships with government and non-governmental organizations (Cronkleton et al. 2012), which could help to ensure that such initiatives do not erode local participation but instead foster sustainable engagement. Such partnerships could help develop effective donor programs with local user buy-in by encompassing tasks such as identifying poor and marginalized households, devising production and benefit-sharing plans, implementing monitoring and evaluation procedures, and establishing conflict resolution mechanisms regarding resource use. These efforts could ultimately build the resilience of community-based natural resource systems (Nightingale and Sharma 2014). Therefore, future policy development should focus on creating an enabling an environment for commercial livelihoods in the context of community-based programs (Sapkota et al. 2020).

Our findings also raise important questions for future research. While we documented participation shifts due to remittance incomes, further studies should examine the long-term impacts on the sustainability of CBNRM institutions. Specifically, how do declining participation rates influence afforestation efforts and resource management outcomes? Additionally, exploring whether participation patterns differ across social groups, such as wealthier versus marginalized households, would provide insights into equity concerns in CBNRM governance (Sunam and McCarthy 2016, de Brauw 2019). If wealthier households withdraw from collective management, this may leave marginalized groups with disproportionate responsibility for maintaining community forests, which would reinforce existing socioeconomic inequalities (McCarthy et al. 2009, Maharajan et al. 2012, Dustmann and Okatenko 2014, Cook 2024).

Finally, while our study identified correlations between out-migration, remittance incomes, and participation in CBNRM, future research should employ causal identification strategies to strengthen policy recommendations. The findings also open avenues for cross-national comparisons using similar datasets, such as those developed through the cross-national PEN survey initiative. Expanding this research across different regions would help determine whether the observed patterns in Nepal apply to other community-based governance systems that are experiencing high out-migration. Similarly, there is room for future work to investigate whether our results from the 2003–2012 period fully capture present-day dynamics in Nepal. Although this study period is useful for studying the general relationship between remittance incomes and community forestry, and yields lessons that are relevant to countries that are currently experiencing out-migration from community forest systems, it is possible that some of these dynamics have either changed or intensified in Nepal since the time of data collection. We note this limitation so as to discourage readers from misinterpreting our results, and to encourage future data collection to study the dynamics of migration and forestry in Nepal.

Our study underscores the complex relationship between out-migration, remittance income, and community-based resource management. As rural economies transition due to migration, natural resource governance institutions must adapt to shifting participation dynamics. Understanding these evolving relationships is critical for ensuring the long-term sustainability of CBNRM models, particularly in regions where migration-driven socioeconomic transformations are accelerating.

LITERATURE CITED

Abdulai, A., and C. L. Delgado. 1999. Determinants of nonfarm earnings of farm‐based husbands and wives in northern Ghana. American Journal of Agricultural Economics 81(1):117-130. https://doi.org/10.2307/1244455

Agarwal, B. 2001. Participatory exclusions, community forestry, and gender: an analysis for South Asia and a conceptual framework. World Development 29(10):1623-1648. https://doi.org/10.1016/S0305-750X(01)00066-3

Agarwal, B. 2010. Gender and green governance: the political economy of women’s presence within and beyond community forestry. Oxford University Press.

Agarwal, B. 2016. Gender challenges. Oxford University Press.

Angelsen, A., P. Jagger, R. Babigumira, B. Belcher, N. J. Hogarth, S. Baucher, J. Börner, C. Smith-Hall, and S. Wunder. 2014. Environmental income and rural livelihoods: a global-comparative analysis. World Development 64:S12-28. https://doi.org/10.1016/j.worlddev.2014.03.006

Barrett, C. B., M. Bezuneh, and A. Aboud. 2001a. Income diversification, poverty traps and policy shocks in Côte d’Ivoire and Kenya. Food Policy 26:367-384. https://doi.org/10.1016/S0306-9192(01)00017-3

Barrett, C. B., T. Reardon, and P. Webb. 2001b. Nonfarm income diversification and household livelihood strategies in rural Africa: concepts, dynamics, and policy implications. Food Policy 26(4):315-331. https://doi.org/10.1016/S0306-9192(01)00014-8

Barrett, C. B., and B. M. Swallow. 2006. Fractal poverty traps. World Development 34(1):1-15. https://doi.org/10.1016/j.worlddev.2005.06.008

Bhattarai, B. 2020. How do gender relations shape a community’s ability to adapt to climate change? Insights from Nepal’s community forestry. Climate and Development 12(10):876-887. https://doi.org/10.1080/17565529.2019.1701971

Bista, R., S. Graybill, Q. Zhang, R. E. Bilsborrow, and C. Song. 2023. Influence of rural out-migration on household participation in community forest management? Evidence from the Middle Hills of Nepal. Sustainability 15(3):2185. https://doi.org/10.3390/su15032185

Bluffstone, R., A. Dannenberg, P. Martinsson, P. Jha, and R. Bista. 2020. Cooperative behavior and common pool resources: experimental evidence from community forest user groups in Nepal. World Development 129:104889. https://doi.org/10.1016/j.worlddev.2020.104889

Cameron, A. C., J. B. Gelbach, and D. L. Miller. 2008. Bootstrap-based improvements for inference with clustered errors. Review of Economics and Statistics 90(3):414-427.

Cárdenas, J. C., M. A. Janssen, M. Ale, R. Bastakoti, A. Bernal, J. Chalermphol, et al. 2017. Fragility of the provision of local public goods to private and collective risks. Proceedings of the National Academy of Sciences 114(5):921-925. https://doi.org/10.1073/pnas.1614892114

Central Bureau of Statistics. 1996. Nepal Living Standard Survey 1995/1996.

Central Bureau of Statistics. 2005. Nepal Living Standard Survey 2003/2004.

Central Bureau of Statistics. 2011a. Nepal Living Standard Survey 2010/2011.

Central Bureau of Statistics. 2011b. Nepal Living Standards Survey 2010/11 statistical report. Vol. 2. Kathmandu, Nepal.

Central Intelligence Agency. 2022. The world factbook – Central Intelligence Agency. https://www.cia.gov/the-world-factbook/about/archives/2022/

Chamberlain, G. 1980. Analysis of covariance with qualitative data. Review of Economic Studies 47(1):225-238. https://doi.org/10.2307/2297110

Choithani, C. 2017. Understanding the linkages between migration and household food security in India. Geographical Research 55(2):192-205. https://doi.org/10.1111/1745-5871.12223

Choithani, C., R. J. van Duijne, and J. Nijman. 2021. Changing livelihoods at India’s rural–urban transition. World Development 146:105617. https://doi.org/10.1016/j.worlddev.2021.105617

Connell, J., and D. Conway. 2000. Migration and remittances in island microstates: a comparative perspective on the South Pacific and the Caribbean. International Journal of Urban and Regional Research 24(1):52-78. https://doi.org/10.1111/1468-2427.00235

Cook, N. J. 2024. Experimental evidence on minority participation and the design of community-based natural resource management programs. Ecological Economics 218. https://doi.org/10.1016/j.ecolecon.2024.108114

Cook, N. J., M. E. Benedum, G. Gorti, and S. Thapa. 2023. Promoting women’s leadership under environmental decentralization: the roles of domestic policy, foreign aid, and population change. Environmental Science & Policy 139:240-249. https://doi.org/10.1016/j.envsci.2022.11.007

Cook, N. J., B. K. Karna, J. Steinberg, and G. Torrens. 2025. Ostromian institutions and violence: community forestry and Nepal’s civil war. World Development 192:107018. https://doi.org/10.1016/j.worlddev.2025.107018

Cronkleton, P., J. M. Pulhin, and S. Saigal. 2012. Co-management in community forestry: how the partial devolution of management rights creates challenges for forest communities. Conservation and Society 10(2):91-102. https://doi.org/10.4103/0972-4923.97481

de Brauw, A. 2019. Migration out of rural areas and implications for rural livelihoods. Annual Review of Resource Economics 11:461-481. https://doi.org/10.1146/annurev-resource-100518-093906

de Janvry, A., M. Fafchamps, and E. Sadoulet. 1991. Peasant household behaviour with missing markets: some paradoxes explained. Economic Journal 101:1400-1417. https://doi.org/10.2307/2234892

Dercon, S. 2002. Income risk, coping strategies, and safety nets. World Bank Research Observer 17(2):141-166. https://doi.org/10.1596/16419

Dongol, C., K. Hughey, and H. Bigsby. 2009. Capital formation and sustainable community forestry in Nepal. Mountain Research and Development 34:70-77.

Dressler, W., J. de Koning, M. Montefrio, and J. Firn. 2016. Land sharing not sparing in the “green economy”: the role of livelihood bricolage in conservation and development in the Philippines. Geoforum 76:75-89. https://doi.org/10.1016/j.geoforum.2016.09.003

Duflo, E. 2001. Schooling and labor market consequences of school construction in Indonesia: evidence from an unusual policy experiment. American Economic Review 91(4):795-813. https://www.aeaweb.org/articles?id=10.1257/aer.91.4.795

Dustmann, C., and A. Okatenko. 2014. Out-migration, wealth constraints, and the quality of local amenities. Journal of Development Economics 110:52-63. https://doi.org/10.1016/j.jdeveco.2014.05.008

Estudillo, J. P., and K. Otsuka. 1999. Green revolution, human capital, and off-farm employment: changing sources of income among farm households in Central Luzon, 1966–1994. Economic Development and Cultural Change 47(3):497-523. https://doi.org/10.1086/452417

Fox, J. 2018. Community forestry, labor migration and agrarian change in a Nepali village: 1980 to 2010. Journal of Peasant Studies 45(3):610-629. https://doi.org/10.1080/03066150.2016.1246436

García-Barrios, L., Y. M. Galván-Miyoshi, I. A. Valsieso-Pérez, O. R. Masera, G. Bocco, and J. Vandermeer. 2009. Neotropical forest conservation, agricultural intensification, and rural out-migration: the Mexican experience. BioScience 59(10):863-873.

Gentle, P., T. N. Maraseni, D. Paudel, G. R. Dahal, T. Kanel, and B. Pathak. 2020. Effectiveness of community forest user groups (CFUGs) in responding to the 2015 earthquakes and COVID-19 in Nepal. Research in Globalization 2:100025. https://doi.org/10.1016/j.resglo.2020.100025

Giri, K., and I. Darnhofer. 2010. Outmigrating men: a window of opportunity for women’s participation in community forestry? Scandinavian Journal of Forest Research 25(sup9):55-61. https://doi.org/10.1080/02827581.2010.506769

Gray, C. L. 2009. Rural out-migration and smallholder agriculture in the southern Ecuadorian Andes. Population and Environment 30(4):193-217. https://doi.org/10.1007/s11111-009-0081-5

Haggblade, S., P. Hazell, and T. Reardon. 2010. The rural non-farm economy: prospects for growth and poverty reduction. World Development 38(10):1429-1441. https://doi.org/10.1016/j.worlddev.2009.06.008

Hajjar, R., J. A. Oldekop, P. Cronkleton, P. Newton, A. J. Russell, and W. Zhou. 2021. A global analysis of the social and environmental outcomes of community forests. Nature Sustainability 4(3):216-224. https://doi.org/10.1038/s41893-020-00633-y

Hecht, S. B., A, L. Yang, B. Sijapati Basnett, C. Padoch, and N. L. Peluso. 2015. People in motion, forests in transition: trends in migration, urbanization, and remittances and their effects on tropical forests. Center for International Forestry Research. https://doi.org/10.17528/cifor/005762

Hoffmann, E. M., V. Konerding, S. Nautiyal, and A. Buerkert. 2019. Is the push-pull paradigm useful to explain rural-urban migration? A case study in Uttarakhand, India. PloS One 14(4):e0214511. https://doi.org/10.1371/journal.pone.0214511

International Organization for Migration. 2019. Migration in Nepal: a country profile 2019. IOM and UN Migration.

Jaquet, S., T. Kohler, and G. Schwilch. 2019. Labour migration in the Middle Hills of Nepal: consequences on land management strategies. Sustainability 11(5):1349. https://doi.org/10.3390/su11051349

Kanel, K. R., and B. R. Kandel. 2004. Community forestry in Nepal: achievements and challenges. Journal of Forest and Livelihood 4(1):55-63.

Kitazawa, Y. 2012. Hyperbolic transformation and average elasticity in the framework of the fixed effects logit model. Theoretical Economics Letters 2(2):192-199. https://doi.org/10.4236/tel.2012.22034

Lambin, E. F., B. L. Turner, H. J. Geist, S. B. Agbola, A. Angelsen, J. W. Bruce, et al. 2001. The causes of land-use and land-cover change: moving beyond the myths. Global Environmental Change 11(4):261-269. https://doi.org/10.1016/S0959-3780(01)00007-3

Lanjouw, P., and M. Ravallion. 1995. Poverty and household size. Economic Journal 105(433):1415-1434. https://doi.org/10.2307/2235108

Leblond, J.-P. 2019. Revisiting forest transition explanations: the role of “push” factors and adaptation strategies in forest expansion in northern Phetchabun, Thailand. Land Use Policy 83:195-214. https://doi.org/10.1016/j.landusepol.2019.01.035

Leder, S., R. Upadhyaya, K. van der Geest, Y. Adhikari, and M. Büttner. 2024. Rural out-migration and water governance: gender and social relations mediate and sustain irrigation systems in Nepal. World Development 177:106544. https://doi.org/10.1016/j.worlddev.2024.106544

Loison, S. A. 2016. Rural livelihood diversification in Sub-Saharan Africa: a literature review. Journal of Development Studies 51:1125-1138. https://doi.org/10.1080/00220388.2015.1046445

Maharjan, A., S. Bauer, and B. Knerr. 2012. Do rural women who stay behind benefit from male out-migration? A case study in the hills of Nepal. Gender, Technology and Development 16(1):95-123. https://doi.org/10.1177/097185241101600105

Marquardt, K., D. Khatri, and A. Pain. 2016. REDD+, forest transition, agrarian change and ecosystem services in the hills of Nepal. Human Ecology 44(2):229-244. https://doi.org/10.1007/s10745-016-9817-x

McCarthy, N., C. Carletto, T. Kilic, and B. Davis. 2009. Assessing the impact of massive out-migration on Albanian agriculture. European Journal of Development Research 21:448-470. https://doi.org/10.1057/ejdr.2009.12

Ministry of Forests and Soil Conservation. 2013. Persistence and change: review of 30 years of community forestry in Nepal. Government of Nepal, Kathmandu, Nepal.

Mohapatra, S., S. Rozelle, and J. Huang. 2006. Climbing the development ladder: economic development and the evolution of occupations in rural China. Journal of Development Studies 42(6):1023-1055. https://doi.org/10.1080/00220380600774988

Molinas, J. R. 1998. The impact of inequality, gender, external assistance and social capital on local-level cooperation. World Development 26(3):413-431. https://doi.org/10.1016/S0305-750X(97)10066-3

Morduch, J. 1995. Income smoothing and consumption smoothing. Journal of Economic Perspectives 9(3):103-114. https://doi.org/10.1257/jep.9.3.103

Newton, P., A. T. Kinzer, D. C. Miller, J. A. Oldekop, and A. Agrawal. 2020. The number and spatial distribution of forest-proximate people globally. One Earth 3(3):363-370. https://doi.org/10.1016/j.oneear.2020.08.016

Nguyen, T. T., T. L. Do, D. Bühler, R. Hartje, and U. Grote. 2015. Rural livelihoods and environmental resource dependence in Cambodia. Ecological Economics 120:282-295. https://doi.org/10.1016/j.ecolecon.2015.11.001

Nightingale, A., and J. R. Sharma. 2014. Conflict resilience among community forestry user groups: experiences in Nepal. Disasters 38(3):517-539. https://doi.org/10.1111/disa.12056

Ojha, H. R., L. Persha, and A. Chhatre. 2009. Community forestry in Nepal: a policy innovation for local livelihoods. Pages 123-160 in D. J. Spielman and R. Pandya-Borck, editors. Proven successes in agricultural development: a technical compendium to millions fed. International Food Policy Research Institute.

Oldekop, J. A., K. R. E. Sims, B. K. Karna, M. J. Whittingham, and A. Agrawal. 2019. Reductions in deforestation and poverty from decentralized forest management in Nepal. Nature Sustainability 2(5):421-428. https://doi.org/10.1038/s41893-019-0277-3

Oldekop, J. A., K. R. E. Sims, M. J. Whittingham, and A. Agrawal. 2018. An upside to globalization: international outmigration drives reforestation in Nepal. Global Environmental Change 52:66-74. https://doi.org/10.1016/j.gloenvcha.2018.06.004

Paudel, D. 2012. In search of alternatives: pro-poor entrepreneurship in community forestry. Journal of Development Studies 48(11):1649-1664. https://doi.org/10.1080/00220388.2012.716152

Poapongsakorn, N., M. Ruhs, and S. Tangjitwisuth. 1998. Problems and outlook of agriculture in Thailand. TDRI Quarterly Review 13(2):3-14.

Pokharel, B. K., P. Branney, M. Nurse, and Y. B. Malla. 2007. Community forestry: conserving forests, sustaining livelihoods and strengthening democracy. Journal of Forest and Livelihood 6(2):8-19.

Poudel, D. P. 2019. Migration, forest management and traditional institutions: acceptance of and resistance to community forestry models in Nepal. Geoforum 106:275-286. https://doi.org/10.1016/j.geoforum.2019.09.003

Poudyal, B. H., D. B. Khatri, D. Paudel, K. Marquardt, and S. Khatri. 2023. Examining forest transition and collective action in Nepal’s community forestry. Land Use Policy 134:106872. https://doi.org/10.1016/j.landusepol.2023.106872

Poverty and Environment Network (PEN). 2016a. CIFOR’s Poverty and Environment Network (PEN) global dataset. https://doi.org/10.17528/CIFOR/DATA.00021

Poverty and Environment Network (PEN). 2016b. PEN technical guidelines.

Reardon, T., J. Berdegué, C. B. Barrett, and K. Stamoulis. 2007. Household income diversification into rural nonfarm activities. Pages 115-140 in T. Reardon, S. Haggblade, and P. B. R. Hazel, editors. Transforming the rural nonfarm economy: opportunities and threats in the developing world. Johns Hopkins University Press.

Reardon, T., C. Delgado, and P. Matlon. 1992. Determinants and effects of income diversification amongst farm households in Burkina Faso. Journal of Development Studies 28(2):264-296. https://doi.org/10.1080/00220389208422232

Robson, J. P., and F. Berkes. 2011. Exploring some of the myths of land use change: Can rural to urban migration drive declines in biodiversity? Global Environmental Change 21(3):844-854. https://doi.org/10.1016/j.gloenvcha.2011.04.009

Robson, J. P., and P. K. Nayak. 2010. Rural out-migration and resource-dependent communities in Mexico and India. Population and Environment 32(2):263-284. https://doi.org/10.1007/s11111-010-0121-1

Rogers, W. H. 1993. Regression standard errors in clustered samples. Stata Technical Bulletin 13:19–23.

Rudel, T. 2011. The commons and development: unanswered sociological questions. International Journal of the Commons 5(2). https://doi.org/10.18352/ijc.248

Sapkota, L. M., H. Dhungana, B. H. Poudyal, B. Chapagain, and D. Gritten. 2020. Understanding the barriers to community forestry delivering on its potential: an illustration from two heterogeneous districts in Nepal. Environmental Management 65(4):463-477. https://doi.org/10.1007/s00267-019-01224-0

Shahi, N., P. Bhusal, G. Paudel, and J. N. Kimengsi. 2022. Forest–people nexus in changing livelihood contexts: evidence from community forests in Nepal. Trees, Forests and People 8:100223. https://doi.org/10.1016/j.tfp.2022.100223

Shin, H. C., S. Vallury, J. K. Abbott, J. M. Anderies, and D. J. Yu. 2022. Understanding the effects of institutional diversity on irrigation systems dynamics. Ecological Economics 191:107221. https://doi.org/10.1016/j.ecolecon.2021.107221

Shrestha, K., and R. Fisher. 2017. Labour migration, the remittance economy and the changing context of community forestry in Nepal. Pages 171-192 in R. Thwaites, R. Fisher, and M. Poudel, editors. Community forestry in Nepal. Routledge. https://doi.org/10.4324/9781315445168-9

Shrestha, G., E. L. Pakhtigian, and M. Jeuland. 2023. Women who do not migrate: intersectionality, social relations, and participation in Western Nepal. World Development 161:106109. https://doi.org/10.1016/j.worlddev.2022.106109

Steffen, W., J. Rockström, K. Richardson, T. M. Lenton, C. Folke, D. Liverman, et al. 2018. Trajectories of the Earth System in the Anthropocene. Proceedings of the National Academy of Sciences 115(33):8252-8259. https://doi.org/10.1073/pnas.1810141115

Sunam, R. K., and J. F. McCarthy. 2016. Reconsidering the links between poverty, international labour migration, and agrarian change: critical insights from Nepal. Journal of Peasant Studies 43(1):39-63. https://doi.org/10.1080/03066150.2015.1041520

United States Bureau of Labor Statistics. 2022.

Von Braun, J., D. Puetz, and P. Webb. 1990. Irrigation technology and commercialization of rice in the Gambia: effects on income and nutrition. Food and Nutrition Bulletin 12(2):1-2. https://doi.org/10.1177/156482659001200204

Wang, Y., C. Chen, and E. Araral. 2016. The effects of migration on collective action in the commons: evidence from rural China. World Development 88:79-93. https://doi.org/10.1016/j.worlddev.2016.07.014

Williams, D., and K. P. Paudel. 2020. Migration, remittance, and adoption of conservation practices. Environmental Management 66(6):1072-1084. https://doi.org/10.1007/s00267-020-01362-w

World Bank Group. 2018. Migration and remittances, April 2018: recent developments and outlook. World Bank, Washington, D.C., USA.

World Bank Group. 2020. Migration and development. Brief 33. World Bank, Washington, D.C., USA.

World Bank Group. 2025. Personal remittances, received (% of GDP) – Nepal. https://data.worldbank.org/indicator/BX.TRF.PWKR.DT.GD.ZS

Zimmerer, K. S. 2010. Biological diversity in agriculture and global change. Annual Review of Environment and Resources 35:137-166. https://doi.org/10.1146/annurev-environ-040309-113840