Beyond Property: Co-Management and Pastoral Resource Access in Mongolia

World Development Vol. 77, pp. 367–379, 2016 0305-750X/Ó 2015 Elsevier Ltd. All rights reserved. www.elsevier.com/locate/worlddev

http://dx.doi.org/10.1016/j.worlddev.2015.08.012

Beyond Property: Co-Management and Pastoral Resource Access in Mongolia SANDAGSUREN UNDARGAA and JOHN F. MCCARTHY* The Australian National University, Canberra, Australia Summary. — A critique of property theory points to the limitations of policies that seek to specify property rights, to strengthen or to reestablish common property institutions. Drawing on property theory and its critique, this paper presents a detailed case study of two waves of reform that attempted to reorganize property relations in Mongolia. Despite their analytical sophistication, property theories face particular challenges when translated into policy prescriptions. Reforms need to build on a broader understanding of the practices and mechanisms involved in governing resources, thereby providing a means to improve resource management. Ó 2015 Elsevier Ltd. All rights reserved. Key words — property-rights, resource access, pastoralism, Mongolia, Inner Asia

1. INTRODUCTION

continue to impede natural resource management. This is principally because, in applying these theories, policy makers misread how pastoralist institutions had in the past applied integrated approaches to manage the three critical components of pastoral production: livestock, labor, and land (hereafter ‘components’). In the process, reformers overlooked how this system had allowed for flexibility of movement, enabling herders to cope with unpredictable weather conditions. This paper identifies five limitations in these approaches for implementing pastoral land management policy. First, these theories provide universalizing concepts of property rights and thus tend to generate an overly prescriptive approach. Second, based on assumptions embedded in collective action theory, analysts misread the Mongolian landscape—misconstruing the natural resource problem as one of ‘open access’ or the absence of property rights (Fernandez-Gimenez & Batbuyan, 2004; Griffin, 2003; Ickowitz, 2003; Mearns, 2004b). Thus, they promoted approaches involving either modifying existing state territorial structures or crafting new forms of property institutions. However, this has proved inadequate to the task of managing dynamic social and resource boundaries. Third the interpretation and application of these approaches raises questions of equity and legitimacy. The reforms broke down a system that in the past had provided legitimate and secure access to all. Implementation of such policies in Mongolia increased the complexity of property institutions, leading to ambiguity regarding who has what sort of rights and authority over which property and resources. Fourth, as the reforms built on an inadequate understanding of customary resource governance, they narrowly focused on land, neglecting the management of the other components

In the mid 20th century, development policies encouraged the use of new technologies to intensify the traditional production of livestock in Africa and elsewhere. After this proved unsuccessful, development organizations attempted to reform the management of pasturelands by granting private property rights to pastoralist communities, particularly in Africa (Khazanov, 2013, p. 897). Later in the 1990s, governments combined market-based land reforms with community-based natural resource management (CBNRM), to develop policies designed to strengthen local pastoral institutions. However, these approaches are yet to provide a solution to the challenges of managing pastoral production. To a significant degree the obstacles they encounter reflect theoretical problems in the collective action 1 theory that have yet to be addressed in practice. Although the insights provided by critics of the collective action theory remain highly relevant, there is clearly a need to assess their applicability to particular cases, such as that of Mongolia, the subject of this study. This paper examines the impact of two land policy reforms that occurred during or after the implementation of a structural adjustment package (SAP) in Mongolia. In 1991, the Mongolian government privatized collective assets, effectively dismantling pastoral institutions inherited from the past. Then, drawing on collective action theory, the state applied a sophisticated approach to pastureland management prioritizing community-based natural resource management (CBNRM)) and co-management of pastoral lands involving centralized government and self-governing community institutions (Fernandez-Gimenez & Batbuyan, 2004; Mearns, 2004b; Schmidt, 2004). From 1994, the government reformed land tenure introducing exclusive individual rights to pastureland. By encouraging industry intensification and the eventual sedentarization of herders, these reforms sought to adjust the livestock industry to a market economy (Bazargur, 1998, 2009). Through an analysis of changes to historical pastoralism in one particular case, and drawing on the tools provided by collective action theory and the access approach, this paper considers how these reforms changed herders’ access to resources in Mongolia. We argue that attempts to specify property rights in land and a narrow application of collective action theory

* We owe a deep debt of gratitude to the research participants for sharing their experiences and invaluable information and insights. We thank the Ministry of Food, Agriculture and Light Industry of Mongolia for information. The Endeavour Postgraduate Scholarship of Australia and the Crawford School of Public Policy, College of Asia Pacific School and CartoGIS at the Australian National University who funded and supported this study. We are also grateful for the comments from anonymous referees on an earlier draft of the paper. Final revision accepted: August 21, 2015. 367

368

WORLD DEVELOPMENT

of production, such as labor and livestock, which are also critical to resource governance. Fifth, these approaches neglect the way actors make use of legal and extra-legal mechanisms, 2 beyond property rights, to derive benefits from pastoral production. This is largely because these theories present a ‘fixed’ menu of property-based mechanisms for defining access and use of natural resources for improving management. However, these concepts of property inadequately map onto the field. In essence, both land reform and CBNRM approaches, lack the conceptual acuity required to understand the dynamic contexts found in Mongolian pastoralism, which cannot be distilled into the fixed categories of property theory. Consequently, we conclude that the application of institutionalist collective action theory, as an alternative to privatization, has yet to improve livelihoods or environmental management. In large part, this is due to a misunderstanding of the dual control institutions that support customary pastoral governance. By moving flexibly across space, and by coordinating the components of production, herders regulate access to migratory pastoral resources. Jurisdictional control, rather than exclusive rights to land per se, facilitates this freedom of movement. Thus, rather than ‘open access’, the underlying problem remains the dismantling of this dual control over the integrated management of pastoral production components. Therefore, future reforms need to consider the possibility of reviving the historical practices of dual control. 2. CPR MANAGEMENT, PROPERTY RIGHTS AND ACCESS For over four decades, the idea of using property policy (either through state or community institutions) to manage landed resources has provided a foundational tenet of resource management. In response to Hardin’s (1968) simplistic analysis, an intuitionalist reading—referred to as the ‘collective action approach’ or ‘CPR theory’—emerged focusing on common pool resources (CPRs) that are readily overused (Johnson, 2004; Saunders, 2014). CPRs are resources from which it is difficult to ‘exclude’ others, where one person’s use subtracts from what others can use (Ostrom, Gardner, & Walker, 1994). This framework opened up a new way of thinking about, and managing the problem of complex and overlapping, bundles of rights (Feeny, Berkes, McCay, & Acheson, 1990; Robbins, 2004). In this revisionist view, the answer to CPR management lay in recognizing and re-installing or creating self-governance by community institutions. Alternatively, co-management approaches might provide a way to get co-existing management authorities to work together to govern resources held under complex bundles of rights (Agrawal & Gibson, 1999; Berkes, Feeny, McCay, & Acheson, 1989; Feeny et al., 1990). Thus, the challenge became one of clarifying the circumstances that might support improved CPR management, and either mending deficiencies in the old institutional arrangements, or crafting newer and better rules using a nested approach (Agrawal, 2001; Agrawal & Ostrom, 2001; Ostrom, Burger, Field, Norgaard, & Policansky, 1999). Recognizing complexity, Ostrom (2009) acknowledged the importance of going beyond a universal design principle and incorporating flexibility when addressing the specific social and historical aspects in various CPR governance arrangements in different contexts and tenure regimes. However, a body of literature has emerged that demonstrates that many of the projects applying these approaches have achieved ‘disappointing outcomes’ (Saunders, 2014, p. 637).

A critique of the collective action approaches has emerged that suggests that it may be difficult to engineer property institutions along the lines suggested by those actually applying collective action theory (Hall et al., 2014; Johnson, 2004; Saunders, 2014). First, policy approaches derived from property theory can be prescriptive in nature, focusing ‘more on how property regimes should be instead of how they are’ in order to promote a socio-economic agenda of equity, efficiency, and sustainability (Benda-Beckmann, Benda-Beckmann, & Wiber, 2006, p. 2). In reality, approaches that try to formalize individual property rights are limited in their effectiveness, given that processes of formalization are contingent on so many factors (Cousins, 2009). Concepts of property have emerged through unique historical processes embedded in specific socio-political contexts (Benda-Beckmann et al., 2006). Understanding these contexts is crucial to learning about how various circumstances affect actors’ decisions over the use of natural resources (Agrawal, 2003). All too often this complexity eludes simple property categories, leading to interventions that struggle to accommodate socially embedded norms, values, and interests (Saunders, 2014). Second, some users may be unable to exercise property rights due to their lack of access to the other means of production including labor, information, capital, or assets (Berry, 1993; Cellarius, 2004). The social stratification which generates inequities of access creates a problem of legitimacy. As actors work around the state law, this can lead, in the eye of those advocating exclusive property rights, to ‘fuzzy’ property rights, which ‘lack clarity of borders, owners and exclusion’ (Verdery, 1999, cited in Sturgeon & Sikor, 2004, p. 3). Third, actors’ ability to benefit from the exploitation of natural resources may have little relation to their formal property rights (Ribot & Peluso, 2003). Thus, it may be difficult to apply property rights approaches in cases where actors employ other mechanisms and strategies to benefit from natural resources, that bear little relation to prescriptive norms, or enforceable claims, supported by property institutions (Ribot & Peluso, 2003, p. 155). Indeed, Ribot and Peluso’s (2003) access approach has provided the means of rethinking CPR management problems, opening up new ways of reading changes in how actors control production and marketing. 3. STUDY SITES AND RESEARCH METHODS To date, there has been limited research on how and why transition policies of privatization and land reform have altered local pastureland management in Mongolia. This paper examines why herders changed the means they use to access seasonal pasture and how this affects management of disputes and overuse. Using a qualitative case study method, this study applies an access approach to understanding the specific mechanisms that different actors apply in accessing pastoral resources and how this is linked to underlying property relations in pastureland management in Mongolia. Here we explain why some CPR management approaches succeed, while others fail in regulating overuse and subtractibility. In particular, we investigate why these policies face difficulties in improving pastureland management, herding risk management and livestock population control. We explain why each approach faces problems when regulating local pastoral resources. Based on a case study undertaken in Herlen Bayan-Ulaan, Mongolia’s oldest and largest State Reserve Pasture Area, we argue that reinstating the historically integrated management of pastoral production is the key to

BEYOND PROPERTY: CO-MANAGEMENT AND PASTORAL RESOURCE ACCESS IN MONGOLIA

improving pastureland management in Mongolia. This goes well beyond the application of western approaches to property rights and conservation-oriented policies. Revealing the limitations of approaches which focus exclusively on property rights, this case study shows that characteristics of historical property relations in pastoralism are highly resilient. This qualitative study of the Herlen Bayan-Ulaan (HBU) Reserve Pasture Area (RPA) provides the means to explore the complexity of property relations in a reallife, contemporary context. This study chose the HBU RPA as a single case study for the following reasons. First, RPAs have provided an important strategy for herder’s risk management under unstable weather conditions (Fernandez-Gimenez, Batkhishig, & Batbuyan, 2012; SDC, 2010). Secondly, the HBU Mountain is historically renowned for its rich foraging grounds and pasturelands suitable for herders to pursue distance movement (otor) in the winter months. It has also experienced a succession of management patterns under pre-collective and socialist collective livestock husbandry systems, hosting an influx of visiting herders. Third, in response to the high demand for reserve pasture, the state re-established the HBU RPA in 2007 after 16 years of closure. This increased the value of HBU RPA for herders from all over Mongolia as a risk management destination as well as for providing access to the public services and livestock markets of the nearby cities of Ulaanbaatar and Baganuur. Fourth, the HBU RPA provides the opportunity to understand how herders developed diverse arrangements to access the area during the severe winter of 2009–10. This study explores the interactions between local and visiting herders, and their negotiated access to pastoral resources, in three neighboring areas of HBU RPA. The three microdistricts (bags) are Herlen Bayan-Ulaan (HBU) bag, Dolood (DD) bag, and Ulaan-Uhaa (UU) bag. The first two bags are situated within Delgerkhaan, a larger district (soum) within Khentii province (aimag). Ulaan-Uhaa bag is situated within Bayanjargalan soum within Tuv aimag. As shown in Map 1, (a) the HBU RPA was established in HBU bag by annexing territory from the surrounding bags and (b) DD and UU bags are the main corridor routes for visiting herders to gain access to HBU RPA. Herders from these latter two bags also use HBU RPA extensively for otor. Given the differences in management of the territorial administration and resource governance across these areas, the research could examine the impact of competition on the HBU RPA and on pasture management of the three bags. During 12 months of field work, 97 participants from three different bags were surveyed (40 from HBU, 23 from DD and 34 from UU), collecting data regarding the legal status of their rights over campsites and their pursuit of otor in the RPA zone. The survey recorded participants’ livestock ownership, herding experience, campsite possession status, sequence and pattern of seasonal and otor movements, relationship with local governments, and RPA administration regarding pasture management. 57 follow-up interviews were conducted with officials from local and national governmental and nongovernmental organizations, herding households from each bag represented by locals and migrants with different backgrounds, e.g., herding experience, wealth, and marital status. The interviews focused on how, when, where, and why herders access campsites and pastoral resources and considered the different strategies used to gain, maintain, and control access to those resources. Field observation was also conducted in all four seasons to explore the reasons for herders’ camping

369

arrangements in each season and the distance between their campsites. 4. HISTORICAL FUNDAMENTALS OF PROPERTY RELATIONS IN MONGOLIAN PASTORALISM Here we wish to explore why management of disputes and overgrazing in Mongolia proved to be more successful under some arrangements than others. We examine changes that occurred in access to pastoral resources by analyzing the history of pastoralism and changes in the control over the components of production. The latter refers to how institutions work to control 3 access to production or the resources crucial to production. The search for seasonal, free-range, grazing land suited to the five species of livestock (sheep goat, cattle, horse and camel) shapes mobile pastoralism in Inner Asia. Under the system inherited from the pre-collective period, management of customary pastoral production (ulamjlalt mal aj ahui) amounted to a system for regulating movement for grazing, providing flexibility for herders to move their stock when needed (Bazargur, 1998; Erdenetsogt, 1998). In Mongolia, this system of managing pastoral production appears to depend upon two historical fundamentals. First, it involves balancing the components of production of five species of livestock (hereafter ‘livestock’), labor and land resources (Bazargur, 1998; Bjorklund, 2003; Erdenetsogt, 1998). ‘Ecologically preferable land’ (EPL) (ecologiin zohistoi nutag) is the most important criteria for land use, shaping herders’ production strategies (Bazargur, 1998). Land identified as EPL includes a diversity of habitat types, providing a means to accommodate the need to move herds in response to season, climate and food abundance (Bazargur, 1998; Fernandez-Gimenez, 1999; Sneath, 2007). Secondly, this was a dual control system which involved interdependent forms of formal (state) and informal (herders) control over pastoral production, a system controlling the different components of production by balancing or limiting the ratio of livestock to pastoral resources within a jurisdiction. The ‘Formal’ aspect encompasses the state’s use of regulations set forth in statutory law. Rather than focusing just on land use, the state formally employed jurisdictional mechanisms to control components of pastoral production. This provided the means of obtaining political and economic benefits (Sneath, 2007). Under the Mongol Empire (1206–1691), jurisdictional lords benefited through controlling their subjects under a state territorial administration (Natsagdorj, 1972). Then, the Manchu Qing Dynasty (1691–1911) used banners (hoshuu) to delineate their territorial-administrative units and formalized their territorial boundaries, which were marked by geographical features and landmarks. At the same time, it continued to allow the ulamjlalt pastoral production system for regulating pasture use (Vreeland, 1954). Some monastery territories (Shavi Gazar) also obtained jurisdictional status in parallel with the hoshuu within an aimag (Natsagdorj, 1972; Vreeland, 1954). In this state territorial structure, each jurisdictional authority controlled production components using formal mechanisms— such as civil registration (jurisdictional membership for labor), taxes (taxable property was mostly the production of livestock) and civil services (postal, military, household work, wage herding, or rental livestock herding). The system also maintained territorial boundaries to accommodate functional EPL (Bazargur, 1998; Natsagdorj, 1967, 1972; Sneath, 2007; Vreeland, 1954). Because labor and livestock can be counted, in contrast to migratory pastoral resources, the authorities

370

WORLD DEVELOPMENT

Map 1. Herlen Bayan-Ulaan case study area in central Mongolia, 2010.

favored using such mechanisms to maximize tax revenues derived from rangeland. Smaller administrative groups 4 within the hoshuu only handled registration of residency and taxation, but did not allocate property rights to land (Natsagdorj, 1967; Vreeland, 1954). Parceling out land with formal property rights remained impractical as herders would move to other jurisdictions during times of scarcity (Natsagdorj, 1967). Rather than providing exclusive use of plots of land, the jurisdictional authorities granted specific pasture use in a territorial manner. These groups or individual herding camps had locally recognized winter territory, but shared summer pastures within their jurisdiction based on customary use (Natsagdorj, 1967; Vreeland, 1954). The state territorial administration functioned as a mechanism for defining community social and resource boundaries and for exclusion and inclusion of outsiders under various climatic conditions, by offering EPL for herding different types of livestock. This

type of public ownership needs to be distinguished from modern ‘state ownership’ (Lattimore, 1932, p. 48). In this form of public ownership, the jurisdictional territory accommodated the use of a jurisdictional community (Lattimore, 1932; Natsagdorj, 1967). This system of dual control amounted to a form of co-management over pastoral production, whereby the centralized state allowed local herders a certain degree of responsibility to manage production as well as pastoral resources (Ballet, Koffi, & Komena, 2009). Herders used an extremely complex set of ulamjlalt rules and norms to govern livestock production (Erdenetsogt, 1998). They decided when, where and how often to move depending on fluctuating herd size, type of livestock, available labor and pasture (Bazargur, 1998; Erdenetsogt, 1998; FernandezGimenez, 1999). In other words, they needed flexible arrangements and freedom of movement (Lattimore, 1932). Rather than relying on exclusive property rights to land, this

BEYOND PROPERTY: CO-MANAGEMENT AND PASTORAL RESOURCE ACCESS IN MONGOLIA

requirement shaped herders’ access to pasture. Under this principle, herders could take up different forms of movement including daily herding, four-season mobility within jurisdictional boundaries, and cross territorial movements (Bazargur, 1998; Erdenetsogt, 1998; Sampildendev, 1985). Although jurisdictional authorities controlled herders’ residency status, herder seasonal movement occurred across boundaries (Vreeland, 1954). In order to maintain the flexibility of movement required to effectively manage the land across jurisdictional boundaries and maximize stock yield, herders themselves, not the state, could most effectively enforce the rules and norms of ulamjlalt production. Thus, the state recognized herders’ practice of ulamjlal, granting them informal control (Fernandez-Gimenez, 1999). This aimed to micromanage production and pasture use. Several formal and informal rules and norms worked together to underpin the arrangements that allowed herders the right to pastoral resources. The state offered herders equal opportunity to exercise use rights to campsites and pasture, and to produce livestock consequent to their jurisdictional residency and the registration of their ownership of private herds. In other words, the state territorial administration allowed herders access to landed-resources through membership and ownership of livestock rather than through exclusive property rights to land. Within this system, the arrangements that determine who uses how much of the pastoral resource were not egalitarian; rather the way in which herders inherit and own livestock determined differences in access (Natsagdorj, 1967; Sneath, 2007). Authorities were not directly involved in pasture management, as pasture was not a taxable property. Instead, they protected taxable property (livestock) by arranging households to assist each other in producing livestock (Vreeland, 1954, p. 26). Informally, the practices of pooling household labor balanced this inequality, allowing everyone access to pastoral resources (Natsagdorj, 1972; Sneath, 2007; Vreeland, 1954). As we discussed already, pasture management in Mongolian pastoral communities requires maintaining freedom of movement. Herders utilized several informal and essential rules and norms to exercise use rights. These rules included ‘first to camp claims the pasture use’ rule (FernandezGimenez, 1999, 2006; Natsagdorj, 1972; Vreeland, 1954). This rule allowed those herders, who could afford to be mobile, the time and resources to be the first to camp while others followed and established themselves around the initial camp (Undargaa, 2013). Further use rights to campsites emerged when a herder left a mark on a particular campsite, indicating that a particular campsite was occupied (Vreeland, 1954). This locally recognized use right permitted flexible use: if a user moved their herds elsewhere to graze, they would remove the mark from the campsite, and the campsite and the surrounding pasture would be available for others to use (Vreeland, 1954). In this way, herders became familiar with all possible campsites for accessing alternate pasture and some herders maintained several locally recognized campsites (Vreeland, 1954, p. 42). Herders would only ‘camp and graze to the extent the pasture allows and would share pasture during inclement weather or at border pastures between campsites, seasonal pastures or jurisdictional boundaries (Fernandez-Gimenez, 1999, 2000; Vreeland, 1954). Herders employed these rules to negotiate camping. They let herds pass by each other, under supervision, without mixing (Undargaa, 2013). Herders and administrative leaders would mediate occasional disputes that resulted from such flexible arrangements of land use (Fernandez-Gimenez, 1999; Natsagdorj, 1967, 1972; Vreeland, 1954).

371

Over time significant changes occurred in pasture management which led to the reduction of EPL. During the Manchu era, hoshuu was used to reduce jurisdictional boundaries. During the collectivization period (1958–91), the state retained the historical emphasis of public ownership, 5 in accordance with the ideology of socialist state property (Verdery, 2004). The state divided territorial units into bag, soum and aimag. It also assigned Meetings of Working People’s Deputies as the organs of state power 6 along with appointing chairmen at each level in order to abolish the aristocratic order and eliminate inequality (Regsuren & Baljinnyam, 1973). The state assigned the main responsibility for managing production management to nested enterprises—a much smaller rural district (soum) or urban micro-district (horoo). Pastoral collectives (negdel), operated within the soum while state farms (aj ahui) operated within the horoo, providing for parallel production enterprises. The Chairmen of soums also governed the negdel and provided the negdel with state and public services (Humphrey, 1978). Negdel’s also formalized brigade groups based on the bag, which functioned as a basic unit to integrate the management of the components of production for developing intensive livestock husbandry. Thus, herders’ seasonal movements were limited within the territorial boundary of the brigade (Bazargur, 1998; Bruun, 2006; Fernandez-Gimenez, 1999; Humphrey, 1978; Mearns, 1996; Sneath, 2003). In response to reduced territory, negdel and aj ahui formal enterprises opened up available pasture by using mechanized transport, engineering wells and providing extra forage to herders to compensate for the loss of EPL. These enterprises also formalized a few ulamjlalt production and pasture use rules regarding when pastoralists might pursue seasonal movement based on the type, age and gender of their livestock. These arrangements were established rules rather than collective-period ulamjlal. They strictly oversaw the components of production, planning, and implementation. However, they continued to allow herders a certain degree of use of ulamjlalt production rules. They also allowed herders to maintain discretionary authority over selecting the location of campsites and seasonal movements according to their herding expertise (Humphrey, 1978; Mearns, 2004b; Undargaa, 2013). Thus, historically up to the end of the collective period, the state retained the fundamentals of dual control over the integrated management of production components when adjusting pastoralism to a changing socio-political and economic system. 5. DISMANTLING THE HISTORICAL FUNDAMENTALS OF PASTORAL PRODUCTION MANAGEMENT (a) Privatization in Mongolia (1991) From the 1980s, a shift in development models and policy narratives occurred. An assessment of ‘state failure’ was used to justify market liberalization in the environmental sector and to promote privatization. In the 1990s the Mongolian government liberalized its pastoral economy, privatizing the collectives (negdel) and state farms (aj ahui) and selling off livestock to members of these pastoral institutions (Korsun & Murrel, 1995). A corrupt and chaotic process of privatization heightened the disparity in living conditions among rural households. In this process, the collective and state assets used in pastoral production (trucks, infrastructure, etc.) were also destroyed from looting (Mearns, 2004a). In 1992, an amendment 7 to the Constitution of Mongolia liberalized the labor force. It led to a subsequent increase in the number of unemployed households migrating from settled communities to take

372

WORLD DEVELOPMENT

up herding in rural areas (Fernandez-Gimenez, 2001; Sneath, 2006). With the number of herding households increasing as new herders took up pastoralism, this led to a differentiation of wealth, social status and herding strategies (FernandezGimenez, 2001; Griffin, 2003). During this time, the land management system altered in several ways. Some forms of public ownership persisted, as the state retained responsibility for land and land-based resources, such as flora and fauna under the public’ or people’s power or authority (ard tumnii medel) 8. The state also developed a de-centralized form of co-management, distributing the responsibility of resource governance to a combination of central government bodies (bag, soum, and aimag administrations) and a self-governing institution (pre-existing public deputies meeting (hural) 9). In practice, the soum government retained residual power to arbitrarily interpret and enforce the land law (Mearns, 2004b). However, the soums were ineffective or rather remained passive in managing local herders’ ulamjlalt pasture use (Fernandez-Gimenez, 2001). The absence of formal control over the management of different components of production led to several changes in the collective management of pastoral production. Herders remained responsible for managing their private production. They decided when, where and how to pursue seasonal and distant migratory movements (Mearns, 2004b). Many herders concentrated their activities in smaller areas near larger settlements, where they could obtain ready access to markets, public services and where desirable pastoral resources were available. This substituted for the absence of negdel support in providing markets, mobility and public services (Fernandez-Gimenez, 2001; Mearns, 2004a). Herders adapted in this way to reduce transaction costs for marketing livestock products and other goods (Griffin, 2003; Sneath, 2006). New herders, in particular, took up such strategies because they had less ecological awareness, little social support, faced inequities in asset holdings; and lacked the customary knowledge of pre-collective ulamjlalt norms that regulated the use of pasture (Fernandez-Gimenez & Batbuyan, 2004; Mearns, 2004a; Nixson & Walters, 2006). This pattern of pasture use and semi-settled camps around jurisdictional centers led, to a certain extent, to localized overgrazing of pastures in those areas (Fernandez-Gimenez, 2001; Mearns, 2004a). We can analyze these changes from the perspective of institutionalist collective action theory. This framework suggests that increasing heterogeneity in wealth, social status, and membership within a community ‘should lead to increasingly different interests among herders and a breakdown in the ability to self-regulate pasture use’ (Ostrom, 1990, cited in Fernandez-Gimenez, 2001, p. 53). According to the property regime approach, open access prevails in the absence of property rights or during the failure of a property regime (Feeny et al., 1990). This perspective suggested that property rights should be specified as clearly as possible. As this body of thinking was applied to Mongolia, ‘open access’ emerged as a key concern. As Griffin (2003) noted: ‘No one in Mongolia owns the vast grassland of the steppe: no one regulates the use of the land. Anyone may graze their livestock on this common land and everyone is free to graze as many animals as they wish’ (Griffin, 2003, p. 67). Many analysts worried that, if individual herding households independently decided on pasture use without an effective formal pastoral institution, this would lead to overgrazing (Griffin, 2003; Ickowitz, 2003; Mearns, 2004b). However, others argued to the contrary. For instance, Upton (2005) maintained:

An ideal of rights to pasture for all. . .may convey the impression of an open-access regime as defined in CPR literature. . .However, this is to overlook or misunderstand the complex system of rights, norms, and rules that inform practice on Mongolia's herding commons (p. 589).

In a similar vein, analysts maintained that herders’ pasture use patterns remained homogenous, or changed less, given that they adhered to seasonal mobility practices (Addison, Friedel, Brown, Davies, & Waldron, 2012; FernandezGimenez, 2001; Upton, 2010). Rather than asserting that herders’ mobility and flexibility were necessarily obstructed by an emerging heterogeneity, this view held that the lack of mobility of resources such as labor, transportation, petrol and livestock had obstructed such practices (Fernandez-Gimenez, 2001; Sneath, 2003). Thus, understanding problems in pastureland management would involve more than simply reallocating property rights. (b) Changes in production management in the three bags after privatization The Herlen Bayan-Ulaan (HBU) Reserve Pasture Area (RPA) case studied here indicates that, by dismantling dual control over production, privatization disembedded the pastoral economy. Although privatization of livestock led to the re-emergence of pre-collective patterns of livestock production, local administrations and the hurals now lost their capacity to apply formal mechanisms to control labor and livestock, as historically practiced in the pre-collective and collective period. The amended constitution entitled herders to migrate freely, and the state, who retained national territorial administration, introduced benign tax policies on livestock (Tumenbayar, 2000). Unlike the dual control system of past times, the co-management institution had formal authority only over land through its management planning and enforcement of some ulamjlalt pasture use rules and norms 10. In other words, officials lost the capacity to enforce precollective ulamjlalt rules and norms regulating herders’ private production. Informal controls also changed. The collapse of formal institutions resulted in problems of open access to privatized assets. This in turn led to the neglect of state and collective infrastructure and assets- the tractors, trucks, and wells central to accessing pastoral resources. In all three bags, migration created wider disparities in wealth, social status, and herding experience. Since individual herding households now independently decided on how to organize their own production, they chose where to move or build livestock shelters and establish camps. 11 However this did not lead to open access. Herders continued to comply with pre-collective ulamjlalt rules and norms, in which the availability of production components shaped herders’ access to pastoral resources. In particular, herders selected the location of their camps and pasture according to ancestral herding knowledge and previous experience with herding certain types of livestock in specific ecological conditions. They also made decisions based on the types and size of the herd they received during privatization, and the availability of labor and transport. Thus, as in the pre-collective period, these factors were essential to production and influenced the way herders practiced ulamjlalt pasture use rules and norms, shaping access to pastoral resources. For several reasons, herders continued to pursue seasonal rotations. First, to increase their production, migrant herders followed the practices of local herders. An increasing number of new herders moved into DD and UU bags, many of them returning to their ancestral or place of birth and relying on

BEYOND PROPERTY: CO-MANAGEMENT AND PASTORAL RESOURCE ACCESS IN MONGOLIA

relatives. These new herders chose grazing areas near relatives, people with specific knowledge and control over suitable pasture areas. Using the authority of both local herders (via informal social relations) and the local administration (via formal rules) migrant herders legitimized their residency and use rights to pastoral resources. In contrast, after the collapse of the state owned HBU enterprise, only a small number of local herders remained to use the abundant pastureland in this bag. This led to an influx of herders from other areas. In the absence of local herders, incoming herders could legitimize their access to pasture through local authorities. Migration patterns in HBU bag shows that the shift in state land policies changed how the components of production were controlled and how herders and the state legitimized access in the past. Second, participants in all three bags could move up to four times per year, but within a greatly reduced area. The destruction of infrastructure and machinery combined with the lack of labor, transportation and petrol as well as the increasing number of migrant herders, reduced the availability of EPL that herders had accessed during the collective period. Herders moved their livestock over shorter distances within the boundaries of their jurisdiction. Third, due to the small number of livestock that each household owned, herders also pursued short seasonal rotations. Therefore, the HBU case area witnessed little change in livestock numbers and herders’ mobility as observed elsewhere (Addison et al., 2012; Fernandez-Gimenez, 2002; Upton, 2010). Fourth, according to the participants, in the period following privatization, optimal weather conditions offered abundant pastoral resources. Under such conditions, herders only needed to travel short distances for seasonal rotation. Fifth, with the increased number of households, and reduced access to EPL, herders continued to impose informal sanctions over those who disobeyed ulamjlalt rules and norms, regardless of their migration status. For instance, in Dolood bag, a former negdel herder was forced to comply with pasture use rules, because ‘‘. . .everybody complained that we were grazing over their winter pasture in the off-season”. Consequently, following privatization, pasture resources did not become entirely ‘open-access’. In accordance with the historical-fundamentals, those who controlled the components of production also controlled access to pastoral resources (Undargaa, 2013). Although land tenure changed in law, as we have seen, the de facto practices of the past continued to affect both formal and informal control. Herders independently decided on how to arrange livestock and labor, replicating pre-collective patterns. However, when compared to the ulamjlalt rules and norms imposed by jurisdictional authorities (government and hural), informal arrangements remained more influential. Although the state legally allowed for comanagement over pastureland, in practice jurisdictional authorities lacked the power to regulate pasture use in the absence of the formal framework used previously to control the various components of production (civil registration, taxation to balance the number of households and livestock, and EPL). Nonetheless, formal institutions still allowed for migrant herders’ residency and shaped their access to pastoral resources. In HBU bag in particular, the shift in the standing of the local authority affected the future for pastureland management in all three bags. Herders reduced the distance they traveled and used the same seasonal pastures repeatedly, and this set the pre-conditions for future overgrazing during intemperate weather conditions. In the next section we will consider the link between these outcomes and other policy reforms.

373

(c) The enforcement of land reform (1998) During the next decade a debate arose over introducing property rights to the pastoral context. Some policy advocates promoted the formalization of property rights in campsites, aiming to support the eventual intensification of livestock husbandry, transforming it into a sedentary production system (Bazargur, 1998). Others argued that it would affect customary flexibility in access and use of campsites and pastureland (Sneath, 2003). Nevertheless, in 1994 the government of Mongolia passed a land law with two provisions stipulating specific property rights over pastureland. The law aimed to provide ‘incentives to herders, farmers, and others to maximize production and to protect land from damage or degradation’ (Sneath, 2003, p. 443). In particular, these changes aimed to resolve overgrazing and disputed use of resources among herders (Fernandez-Gimenez & Batbuyan, 2004). The first provision refers to winter and spring campsites 12 (uvuljuu havarjaa), with a certificate of possession (ezemshliin gerchilgee) (Fernandez-Gimenez, 2002; Sneath, 2003). In 2002 the land law was amended 13 to provide that citizens may possess winter and spring campsites as khot ail. The 1994 law had allowed possession of small residential plots by individual households, 14 thus the local administrations allowed individual households to possess campsites. In other words, depending on household camping structure (khot ail) local authorities could issue a certificate to an individual household and a certificate to several households, each with their separate livestock shelters (Fernandez-Gimenez & Batbuyan, 2004; Undargaa, 2006). This amendment assigned exclusive management rights to campsites including the right to mortgage the certificate, 15 the right to transfer use through inheritance and to another except through sale (Fernandez-Gimenez & Batbuyan, 2004). Although a soum is supposed to approve campsite possession based on soum land management planning, 16 herders now applied for a certificate of possession after establishing their campsites (Undargaa, 2013; Upton, 2005). The 2002 amendments also specifically forbid legal possession of the pastoral resources that surrounded campsites (Fernandez-Gimenez, Kamimura, & Batbuyan, 2008). The state assigned formal co-management actors to enforce both allocation of campsite possession and ulamjlalt pasture use rules and norms. Although the state formalized ulamjlalt seasonal mobility and rotation, 17 local administrations subsequently found it difficult to enforce this reform (FernandezGimenez & Batbuyan, 2004; Upton, 2009). In other words, the state overestimated its ability to regulate access to natural resources (Berry, 1993). When the law came to be implemented, it became clear that these reforms could not resolve the problems leading to disputes over access and overgrazing. Those who owned sizable livestock herds and shelters could use the law to gain possession of campsites, thereby claiming the de facto right to use surrounding pasture (Fernandez-Gimenez & Batbuyan, 2004; Sneath, 2003). In contrast, those lacking livestock shelters, wealth and campsite rights could make few claims over pasture. This process contributed to the growing inequality among herders (Fernandez-Gimenez & Batbuyan, 2004). Wealthy herders could set up several campsites near poor households with less livestock. Further, to gain control over more campsites they could split their family unit, with members of the same family claiming legal possession of campsites in several different places (Mearns, 2004a). This diminished the informal rights of many herders and their ability to access resources. Attracted by the opportunity to possess fixed campsites, wealthier herders had an incentive to become sedentary,

374

WORLD DEVELOPMENT

thereby protecting the pasture around their campsites (Fernandez-Gimenez & Batbuyan, 2004; Upton, 2005). This process encouraged off-seasonal pasture use and the overgrazing that ensued. (d) Changes in the three bags As the system shifted away from using the mechanisms to access pasture in the past, the introduction of property rights under the land law presented several problems. In the HBU case, land reform discouraged dual control over the components of production. Following the loss of formal labor and livestock controls, herders made decisions regarding stock levels after taking into account their value in the market and their ability to access markets. As the number of households and livestock increased, this further strained bag EPL. The area could hardly contain the expanding number of legal campsites. After severe loss of livestock during the dzud (severe winter storms) of 2000 in UU and Dolood bags, many migrant households left. Over time, households who held on saw natural growth in herd numbers, even as their households grew. By contrast, in HBU bagß movement into the area added to livestock and household numbers. As the constitution provided herders with specific rights, the local administration could not forbid jurisdictional residency. As a result, these migrants obtained a legal right to campsites, irrespective of the scarcity of seasonal pastures in the bag. A local HBU herder explained ‘Households with thousands of head of livestock have come from other areas and have become residents here’. This type of access led to overgrazing in the HBU bag. The inability of authorities to apply jurisdictional mechanisms to control the number of herding households impaired resource management. Although in the past bag herders had use rights to both campsites and pasture, land reform de-coupled this system. Rather than accessing pastoral resources through informal, locally recognized, use rights, herders could use legal rights to campsite possession to access pasture. In all three bags, the authorities provided certificates of possession to individual households as a khot ail, which currently is composed of a single household in the winter and spring seasons. A survey conducted in each of the three bags found that 95% of the total participants (n = 97) legally possessed either, or both, winter and spring campsites in all three bags (Undargaa, 2013). As elsewhere, prior to applying for a legal certificate, many herders preferred to establish their permanent campsites in areas with access to pasture and water and in close proximity to markets. The state is unable to bring the movements and livestock of herders into alignment with pasture conditions, and the local administration allowed herders to select winter and spring campsites. However, this is not open access. As in the pre-collective and collective period, coordination of the components of production regulated herders’ access. For the most part herders cannot simply access whatever campsites or pasture they desire. Herd size, type, gender, herding strategy for each type of livestock and the availability of labor and pastoral resources within a particular jurisdiction determine access to pasture. Also, herders’ practices that take into account particular ecological conditions affects access outcomes just as much as differences in wealth, social status and herding experience. For instance, wealthy migrant herders from western aimags responded to a shortage of seasonal pasture in HBU bag by choosing campsites in the HBU Mountain. They were accustomed to a mountain environment and they only made legal claims over winter campsites. While they moved across boundaries to find grazing areas in other seasons, they overused the mountain pasture for both spring and winter seasons.

After the land reforms became law, the following processes shaped outcomes. First, new opportunities for campsite ownership affected the mobility of herders. Rather than make use of flexible, but informal, locally recognized use rights to campsites, herders sought legal rights to campsites. Second, the changes increased the number of campsites that were allocated and subsequently led to overgrazing. Traditionally, rather than investing in exclusive property rights to campsites, herders accessed pastoral resources, moving across the land in a flexible manner to coordinate the components of production. Now, increasing livestock densities have reduced EPL, and herders responded by attempting to possess more campsites across all three bags. To mitigate against the risk of unstable weather conditions, herders sought access to larger pastures. In the HBU case, herders sought to acquire more campsites, spreading out into other seasonal pastures, in turn generating even more overcrowding. Third, herders re-invented the norms of using mechanisms and strategies to access additional campsites from the precollective past. As mentioned earlier, using a splitting strategy, a family could gain access to campsites in several places. They could also use their membership of a jurisdiction and their ownership of livestock to further their claims. For example, the law provided that any ‘citizen who owned livestock’ had the right to possess a campsite. An ‘A list’ provided a formal census of livestock owned by individual households both at the soum center and at the herding site. Herders could use this A-list to support the claim of a family member to extra campsites, irrespective of their occupation. After the state lost control of the components of production, herders returned to these mechanisms from the past. In other words, without having property rights to land, people could use their rights of residency and ownership of livestock to benefit from pastoral resources. This strategy led to an imbalance in production components and decreased the availability of pastoral resources. As a result, herders now found it more difficult to balance the components of production. As in other post-socialist countries, land reform increased the ‘fuzziness’, leading to more indistinct, ambiguous and partial forms of property rights (Verdery et al., 1999). Following the reforms, herders could use a variety of co-existing mechanisms to access pastoral resources. Under land reform legislation, herders could not (a) exclude others from using the pasture around their campsite; (b) secure access to additional pastureland during unstable weather; and (c) protect private pastoral assets on their campsites from looting. As a result, many herders returned to pre-collective ways such as using mobile fences, practicing locally recognized use rights and using alternative campsites to ensure freedom of movement. They no longer restricted their seasonal grazing to specific land within their jurisdiction. Instead, they turned to offseason (otor) movements, even crossing jurisdictional boundaries. However, those households following such patterns came into conflict with others seeking to use their legal ownership of campsites to exclude herders from surrounding pasture. Local co-management authorities (local government and its hural) also face a dilemma. On the one hand the government legitimizes bag herders’ residency and use rights to pasture and campsite possession. On the other hand, local herders wish to balance the components of production. Bag and soum governments lack the authority to forbid migrants from taking up residency. At the same time, to manage the bag pasture, the bag self-governing hural claims the authority to limit the number of migrants. As each authority used to gain benefits from governing production, each now seeks to employ their

BEYOND PROPERTY: CO-MANAGEMENT AND PASTORAL RESOURCE ACCESS IN MONGOLIA

authority for their own advantage. Consequently, herders and state-based actors can make use of various competing formal and informal mechanisms to gain, maintain and control benefits from pastoral resources. 6. DISCUSSION AND CONCLUSION 18 Property relations emerge from the historical process of a society (Benda-Beckmann et al., 2006). They reflect sociological and political relationships that are based on long historical experiences of environmental dependencies and vulnerabilities. In the Mongolian case, property relations are deeply embedded in the historical foundations of pastoralism. Land tenure evolved as an element of pastoralism that worked as a sociotechnical system: here, the historical socio-political structure of pastoralists was integrated with the natural and material resources available in this environment (Sneath, 2004). Mongolian pastoralists developed the ulamjlalt system of pastoral production as a means of allocating political and economic benefits from natural resources in the prevailing climatic and geographical conditions (Sneath, 2007). The review of the history of Mongolian land tenure presented above demonstrated that those who controlled production also controlled access to pastoral resources. Historically, until privatization, the state regulated access to pastoral resources based on a dual (formal and informal) system of controlling the benefits from pastoral production within a jurisdiction. The formal control (jurisdictional residency and taxation of production under state territorial administration) worked to balance the number of households and livestock to match the availability of ecologically preferable land (EPL). These functioned as exclusionary, as well as inclusionary, mechanisms for facilitating the freedom of movement based on the principle of reciprocity in pastoralism. Herders employed these formal control mechanisms to claim their use rights to pastoral resources and to practice ulamjlalt pastoral production. The state’s use of jurisdictional mechanisms to extract benefits from pastoral production also depended upon herders’ informal control over production and pastureland management. The state allowed herders to micromanage daily pasture use informally based on ulamjlalt production principles and enabled freedom and flexibility of movement, in a territorial manner, within a jurisdictional boundary. Herders pursued ulamjlalt seasonal rotations and movement based on a production strategy, which depended on coordinating access to, and the availability of, livestock, labor and pastoral resources. Thus, the property relations bequeathed by historical practices were embedded in the system of pastoral production. Combining field research with a review of existing studies, this paper has explored the ways in which privatization and land reform affected herders’ access to resources. We have sought to elucidate changes to the integrated management system of dual control that previously governed pastoral production (See Table 1). In conclusion we present several challenges that emerged from this transition. Issue 1: Pursuing an overly prescriptive approach based on a misreading. During privatization, analysts interpreted the system in accordance with a highly prescriptive reading of collective action theory that argued for addressing the problems of exclusion and subtractibility inherent in CPR management. This policy narrative supported state recognition of historical forms of self-governance (Feeny et al., 1990). Accordingly, the state formalized co-management of rangeland involving the local government and its hurals in order to enforce ulamjlalt

375

pasture use rules and norms. However the formalization process overlooked the historical form of self-governance which had previously balanced the components of production. Also, land reform and co-management policy reflected a particular interpretation of property theory. Land reform, in 1994, derived from a policy narrative that called for clear property rights to land and pastoral resources. Applying this principle, the state formalized herders’ right to campsites with statutory exclusive possession rights. However, this failed to allow for freedom of movement—an essential element of the herding system. As in other cases (Cousins, 2009), this reform sought to disembed rights from their social functions in order to support the development of a market economy. In the process, it broke down pre-existing arrangements that had defended herders from losing rights to pastureland. As in other cases (Meinzen-Dick & Mwangi, 2009, cited in Cousins, 2009), these reforms attempted to transform Mongolian property rights in a way that fitted more closely with western ideals of property rights. The collective action approach, with its emphasis on the ‘bundle of rights’ concept and its preference for comanagement of land resources, provides a sophisticated framework for analysis. However, those employing this approach inadequately interpreted the realities of Mongolian pastoralism. Hence, a puzzle emerges: to what extent can the collective action approach be used to explain the characteristics of Mongolian pastureland management? Can this framework provide a prescription for reforming pastoralism in Mongolia? In other words, is the co-management approach able to remedy problems in Mongolia’s CPR resource management? As one analyst observed, ‘while pastureland [in Mongolia] remains under state ownership, it is de facto managed as common property’ (Upton, 2005, p. 586). In attempting to find a new nested relationship between state and customary institutions, co-management re-invented the wheel. The state and customary forms had already been integrated into a single, historically tested and socially embedded form of pastoral production and resource management. Issue 2: Misconstructing the natural resource challenge as ‘open access’. The application of the property rights approach rested on a narrow understanding of property in this pastoral context. This policy narrative promoted ‘a confused and confusing set of ideas that obfuscates the real problems and opportunities of rights-based approaches’ (Cousins, 2009, p. 6). Pastureland is not open access; access is regulated by coexisting formal and informal rules and norms (Klein et al., 2012; Upton, 2005). Employing the concept of ‘institutional bricolage’, Upton (2009) has emphasized that where contests emerge over notions of customary use, rules and norms, as actors seek legitimacy for their uses, they will re-invent custom. As we have argued, in coordinating the components of production, in the past the state territorial administration had met a critical need of herders, in the process shaping how herders practice pasture use rules and norms. This understanding is critical to our analysis of institutional change. We saw this ‘institutional bricolage’ in the HBU case, where the current pastoral production system combines elements derived from different periods: the re-emergence of a pre-collective pattern of private livestock production; a socialist collective pattern of smaller jurisdictional boundaries; and, a postsocialist market economy that prioritizes commoditization of land-based resources. Herders now needed to operate without the arrangements that had governed production in the past, and this created difficulties for the practice of pre-collective ulamjtlalt pasture use rules and norms. As in the past herders turned to using residency requirements and livestock

376

WORLD DEVELOPMENT Table 1. Timeline of policy changes to the management of production and land tenure

Policy period

Trends shaping pastoral institutions

Specific events in the case study

Mongol Empire (1206)

Under state territorial administration, the aristocratic Dual control involving both state and jurisdictional order controls the components of production subjects (herders) in managing livestock, labor, and (livestock, labor and land) land formalized under the state territorial administration Manchu Era (1691–1911) Jurisdictional boundaries are formalized and then Dual control continues under the state territorial reduced; jurisdictional administrative unit controls the administration components of production under the banner (khoshuu) system Collectivization (1958–91) The state re-arranged the territorial units. Brigade The state formalized some pre-collective pasture use (bag) now control the components of production. First customs (e.g., seasonal movements), but allowed initiatives to intensify livestock husbandry herders to practice many informal rules and norms under the supervision of the collective or state enterprises Privatization (1991) Constitution Privatization of livestock, livestock shelters, and other Corrupt and ambiguous privatization process led to (1992) collective and state assets to collective members; unequal distribution of assets, the collapse of pastoral Collapse of formal pastoral institutions such as state infrastructure, and the disembedding of the pastoral enterprises and collectives; Liberalized labor controls economy. Migrants move from cities to former allowed free migration collective areas (DD and UU bags), and rural populations from former state enterprise area (HBU bag) move to cities New Law of Mongolia on To intensify production and support conservation, the Without control over the components of production, Administrative and Territorial Units State shifts from controlling all the components of the state failed to enforce customary pasture use rules and their Governance and new tax and norms. Herders gain control over the production production to only land; local administration now provisions for herders (1993); First obtains authority to regulate land use based on components; Herders now select where to build land law land (1994) customary pasture use rules and norms. The new land livestock shelters, returning to ancestral herding law provides for lease of residential plots and public strategy regarding herd size and type etc; not open use rights to pastoral resources access Implementation of land law (1998); Enforcement of land law; Herders embark on long Local administration allows herders to hold campsites Nationwide severe winter 2000–01 distance cross-boundary movement (otor) as residential plots. Migration from Western and Gobi aimags begins as herders search for accessible and better pasture and markets in HBU bag Amendments to law on land (2002); The amendments to the land law allow lease of winter Local administration allows herders to choose Several local and nationwide severe and spring campsites; exemption of pastoral resources campsite locations; the number of migrant households winters 2000–10 from legal possession. Herders pursued long distance with large herd sizes increases in HBU bag; access to resources depended on herding strategy, herd size, cross-boundary movement (otor) type and number, and availability of labor and pasture; not open access. With the growth of herding households, campsites and livestock in HBU bag, herders from other bags, soums, aimags struggled to access HBU Reserve Pasture Area

ownership to facilitate their freedom of movement. As the precollective ulamjlalt patterns of production are rooted in the logic of pastureland management, patterns of using these mechanisms persist. The attempt to apply community-based natural resource management (CBNRM) through design principles faced critical challenges elsewhere in terms of strengthening collective action (Nelson & Agrawal, 2008). In Mongolia, this CBNRM approach needed to engineer co-managed institutions and to specify property rights. This involved formalizing property rights to campsites and to pasture under state territorial administration, but without making use of the well proven mechanisms discussed above. This initiative conflicted with the herders’ insistent pursuit of flexibility or freedom of movement. This model provided for smaller nested jurisdictions that did not contain sufficient EPL, thereby exacerbating this problem. The experimental CBNRM projects struggled to define social and resource boundaries and to provide for the reciprocity and the flexibility of movement required by herders in all living conditions (Addison, Davies, Friedel, & Brown, 2013; Fernandez-Gimenez et al., 2008; Upton, Moore, Nyamsuren, & Batjargal, 2013). Consequently, state and

community-led interventions need to improve how they incorporate local patterns of resource use (Sikor & Muller, 2009; Upton, 2009). Issue 3: Increasing the complexity of property institutions. In several other post-socialist countries such as Romania, Bulgaria, China and Vietnam, the process of specifying exclusive property rights exacerbated the ‘fuzzy’ quality of property relations (Verdery, 2004; Verdery et al., 1999). In other words, it intensified the gap between ‘actual property arrangements with their complex interrelationships and multiple actors’, and an ‘idealized image of exclusive private rights and obligations held by normative individuals’ set out in the new land law (Yeh, 2004, p. 108). As in these other cases ‘within a single historical period’ a government issued mutually contradictory laws, ‘allocating rights to the same resources to different parties’ (Ribot & Peluso, 2003, p. 162-163). In Mongolia the transformation of socialist state/public territorial ownership to landed-property rights reproduced this ‘fuzziness’ (Sturgeon & Sikor, 2004). Mis-diagnosing the problem as one of open access, reformers introduced property rights instead of addressing the underlying problems generating disputes and overgrazing. This in turn broke down a historically proven

BEYOND PROPERTY: CO-MANAGEMENT AND PASTORAL RESOURCE ACCESS IN MONGOLIA

system that provided legitimate secure access for herders in all living conditions in the context of non-egalitarian pastoralism. Then it created ambiguity with respect to who really controls access to what resource. As we saw in the HBU case, herders and local authorities employed competing legal and extra-legal mechanisms to extract benefits from pastoral resources. Issue 4: Production components. Privatization and land reform policies in Mongolia ignored the fundamental logic underlying the integrated management of pastoral production. As in other cases, after collectivization, the state shifted its control over the components of production to control of land only as has occurred elsewhere (Peluso, 1992). The comanagement approach focused on conservation rather than production (Mearns, 2004b). Indeed, many studies of extensive pastoralism are yet to provide a means of integrating all the components of production, insufficiently taking into account the dynamic relations between these components under unstable conditions (Bazargur, 1998). By dismantling the management of the components of production, in a similar fashion to that described by Polanyi (1944), the state disembedded the pastoral economy from its social, political, and ecological foundations. A long historical experience underlay private livestock ownership and production in Mongolia. However, by developing a system that lacked the proven, customary mechanisms to control labor and livestock, the reforms discarded a long-established understanding of what worked in the local environment. As collective action theory has emphasized, particular arrangements need to ‘fit’ ecologi-

377

cal and social-structures (Ostrom & Nagendra, 2006). In the past these mechanisms fitted with the fundamental logic of pastureland management. The healthy functioning of historically developed and resilient institutions remains critical to property relations: indeed the credibility of such institutions is more important than the form (state, communal, or individual) they take (Ho, 2013). This study has argued that jurisdictional authorities provide several resource governance mechanisms for all the actors who draw benefits from taxable pastoral production. This is why herders support state involvement in the co-management of the various components of production. Issue 5: Access mechanisms. Developing a more sophisticated analysis of production. The analysis of the political economy of resources—principally exploring why some benefit irrespective of the property rights they hold—enables us to understand pastoralism in new ways (Ribot and Peluso, 2003, p. 155). It led us to shift our focus from the management of pastureland to the management of production and its marketing, exposing the whole range of mechanisms that pastoralists employ to access pastoral resources and to draw benefits from production. Further, this critique of the application of property rights and CBNRM management in this case raises theoretical and policy questions that need to be answered in order to improve understandings of production and resource practices. This requires a broader understanding of context and the abandonment of overly prescriptive institutional approaches.

NOTES 1. See Johnson (2004), Saunders (2014) and Hall, Cleaver, Franks, and Maganga (2014) for critical reflections on CPR theory. 2. This article follows Ribot and Peluso (2003) in using the term mechanism as shorthand for the means, processes and relations used to access resources. 3. ‘‘Control” . . .refers to the checking and direction of action, the function or power of directing and regulating free action (Rangan, 1997, cited in Ribot & Peluso, 2003, p 158). 4. Aravt (unit of 10 households), bag (more than 10 households led by noble families), a soum (150 households for military service). These groups were responsible for civil registration of a group of households, its members, their taxable properties (only livestock, occasionally silver), group members’ civil offences and services and occasionally organized social events like sacred rituals. Comprehensive tax collection system from livestock production existed within each jurisdiction (Vreeland, 1954). For territorial administrative purposes, jurisdictional authorities issued decrees allocating specific areas for public services such as urtuu (postal) or haruul (border guard) households or grazing areas for livestock owned by the state or a few high-ranking religious elites (Natsagdorj, 1967). 5. Article 5 & 8, Constitution 1940. 6. Article 37, Constitution 1940. 7. Article 6.3 Constitution, 1992. 8. Article 6.1 & 6.2, Constitution, 1992. 9. Bag residents’ meeting, and soum, aimag residents’ representative meeting. Article 59.1, Constitution, 1992.

10. Law of Mongolia on Administrative and Territorial Units and their Governance 1993; Law of Mongolia on Environmental Protection 1995; Law of Mongolia on land 1994: Article 51.3; when using and managing pastureland, refer to ulamjlalt land use divisions such as winter, spring, summer and autumn pasture. Summer and autumn pasture and otor reserve pasture will be communally used within bag hot ails. 11. This does not refer to legal campsite possession, because the 1994 land reform was not enforced until 1998. 12. It contains a ger (mobile accommodation), livestock shelter and horse posts (Fernandez-Gimenez & Batbuyan, 2004). 13. Article 52.7, Law of Mongolia on Land, 2002.

14. Article 29.1 the Law on land, 1994. 15. Article 35.1.7 of the Land law, 2002 (added in July 9th, 2009). 16. Article 33.4 of the Land law, 2002. 17. It allows summer, autumn, and otor pastures to be shared among bag residents and prohibits grazing of winter and spring pasture in the off season. Article 52.1 and 52.2, the Law on land, 2002. 18. Detailed findings on historical section, changes in production and land tenure during the revolutionary period (1921–58), privatization, land reform, community-led CBNRM and state territorial strategy for NRM, and relevant discussions are available in a forthcoming book in progress: Undargaa, S (2015–2016), Pastoralism and Common Pool Resources: Rangeland Co-management, Property Rights and Access in Mongolia. Routledge.

378

WORLD DEVELOPMENT

REFERENCES Addison, J., Davies, J., Friedel, M., & Brown, C. (2013). Do pasture user groups lead to improved rangeland condition in the Mongolian. Gobi Desert?. Journal of Arid Environments, 94, 37–46. Addison, J., Friedel, M., Brown, C., Davies, J., & Waldron, S. (2012). A critical review of degradation assumptions applied to Mongolia’s Gobi Desert. The Rangeland Journal, 34, 125–137. Agrawal, A. (2001). Common property institutions and sustainable governance and resources. World Development, 29(10), 1649–1672. Agrawal, A. (2003). Sustainable governance of common-pool resources: context, methods and politics. Annual Revenue of Anthropology, 32, 243–262. Agrawal, A., & Gibson, C. (1999). Enchantment and disenchantment: The role of community in natural resource conservation. World Development, 27(4), 629–649. Agrawal, A., & Ostrom, E. (2001). Collective action, property rights, and decentralization in resource use in India and Nepal. Politics and Society, 29(4), 485–514. Ballet, J., Koffi, K. J. M., & Komena, K. B. (2009). Co-management of natural resources in developing countries: The importance of context. E´conomie Internationale, 120, 53–76. Bazargur, D. (Ed.) (1998). Geography of pastoral animal husbandry. Ulaanbaatar: Institute of Geoecology, Mongolian Academy of Science. Bazargur, D. (2009). Ecologiin Gazarzui. Mongol Uls, Ulaanbaatar: Gazarzuin Hureelen, Shinjleh Uhaanii Academy. Benda-Beckmann, F. V., Benda-Beckmann, K. V., & Wiber, M. G. (2006). The properties of property. In F. V. Benda-Beckmann, K. V. BendaBeckmann, & M. G. Wiber (Eds.), Changing properties of property (pp. 1–40). New York: Berghahn Books. Berkes, F., Feeny, D., McCay, B. J., & Acheson, J. M. (1989). The benefits of the common. Nature, 340, 91–93. Berry, S. (1993). No condition is permanent: The social dynamics of Agrarian change in Sub-Saharan Africa. Madison: University of Wisconsin Press. Bjorklund, I. (2003). Sami pastoral society in Northern Norway-National Integration of an Indigenous Management System. In D. Anderson, & M. Nuttal (Eds.), Cultivating Arctic landscapes (pp. 124–135). New York: Berghahn Press. Bruun, O. (2006). Precious steppe. Oxford: Lexington Books. Cellarius, B. (2004). Without co-ops there would be no forests’: historical memory and the restitution of forests in post-socialist Bulgaria. Conservation & Society, 2(1), 51–73. Cousins, B. (2009). Capitalism obscured: The limits of law and rightsbased approaches to poverty reduction and development. Journal of Peasant Studies, 36(4), 893–908. Erdenetsogt, N. (Ed.) (1998). Mongolian Nomadic Livestock. Ulaanbaatar: ’MMM’ Association. Feeny, D., Berkes, F., McCay, B. J., & Acheson, J. M. (1990). The tragedy of the commons: Twenty-two years later. Human Ecology, 18, 1. Fernandez-Gimenez, M. (1999). Sustaining the steppes: A geographical history of pastoral land use in Mongolia. The Geographical Review, 89 (3), 315–336. Fernandez-Gimenez, M. (2000). The role of Mongolian nomadic pastoralists’ ecological knowledge in rangeland management. Ecological Applications, 10(5), 1318–1326. Fernandez-Gimenez, M. (2001). The effects of livestock privatization on pastoral land use and land tenure in post-socialist Mongolia. Nomadic Peoples, 5, 2. Fernandez-Gimenez, M. (2002). Spatial and social boundaries and the paradox of pastoral land tenure: a case study from post-socialist Mongolia. Human Ecology, 30(1), 49–78. Fernandez-Gimenez, M. (2006). Land use and land tenure in Mongolia: A brief history and current issues. In D. J. Bedunah, E. McArthur, & M. Fernandez-Gimenez (Eds.), Proceedings of the Conference on Transformations, Issues, and Future Challenges, Salt Lake City, UT, Proceedings RMRS-P-39. Fort Collins, CO: US Department of Agriculture, Forest Service, Rocky Mountain Research Station. Fernandez-Gimenez, M., & Batbuyan, B. (2004). Law and disorder: Local implementation of Mongolia’s Land Law. Development and Change, 35 (1), 141–165.

Fernandez-Gimenez, M., Batkhishig, B., & Batbuyan, B. (2012). Crossboundary and cross-level dynamics increase vulnerability to severe winter disaster (dzud) in Mongolia. Global Environmental Change, 22, 836–851. Fernandez-Gimenez, M., Kamimura, A., & Batbuyan, B. (2008). ’Implementing Mongolia’s Land Law: Progress and Issues’. Japan: The Central for Asian Legal Exchange (CALE), Nagoya University. Griffin, K. (2003). Urban–rural migration and involution in the livestock sector. In K. Griffin (Ed.), Poverty reduction in Mongolia (pp. 56–71). Canberra: Asia Pacific Press. Hall, K., Cleaver, F., Franks, T., & Maganga, F. (2014). Capturing critical institutionalism: A synthesis of key themes and debates. European Journal of Development Research, 26, 71–86. Hardin, G. (1968). The tragedy of the commons. Science, 162, 1243–1248. Ho, P. (2013). In defense of endogenous, spontaneously ordered development: Institutional functionalism and Chinese property rights. The Journal of Peasant Studies, 40(6), 1087–1118. Humphrey, C. (1978). Pastoral nomadism in Mongolia: The role of herdsmen’s cooperatives in the national economy. Development and Change, 9, 133–160. Ickowitz, A. (2003). Poverty and the environment. In K. Griffin (Ed.), Poverty reduction in Mongolia (pp. 95–112). Canberra: Asia Pacific Press. Johnson, G. (2004). Uncommon ground: The ’Poverty of History’ in common property discourse. Development and Change, 35(3), 407–433. Khazanov, A. (2013). Contemporary pastoralism: Old problems, new challenges. In T. Sternberg, & D. Chatty (Eds.), Modern pastoralism and conservation: Old problems, new challenges. Cambridge: The White Horse Press. Klein, J. A., Fernandez-Gimenez, M., Wei, H., Changqing, Y., Ling, D., Dorligsuren, D., & Reid, R. S. (2012). A participatory framework for building resilient social–ecological pastoral systems. In M. FernandezGimenez, X. Wang, B. Batkhishig, J. A. Klein, & R. S. Reid (Eds.), Restoring community connections to the land: Building resilience through community-based rangeland management in China and Mongolia. Oxfordshire: CABI International. Korsun, G., & Murrel, P. (1995). Politics and economics of Mongolia’s privatization program. Asian Survey, XXXV:5, 472–486. Lattimore, O. (1932). Manchuria cradle of conflict. Toronto: The Macmillan Company of Canada. Mearns, R. (1996). Community, collective action and common grazing: The case of post-socialist Mongolia. The Journal of Development Studies, 32, 297–339. Mearns, R. (2004a). Sustaining livelihoods on Mongolia’s pastoral commons: Insights from a participatory poverty assessment. Development and Change, 35(1), 107–139. Mearns, R. (2004b). Decentralisation, rural livelihoods and pasture-land management in post socialist Mongolia. European Journal of Development Research, 16(1), 133–152. Natsagdorj, Sh. (1967). The economic basis of feudalism in Mongolia. Modern Asian Studies, 1(3), 265–281. Natsagdorj, Sh. (1972). Soum, khamjlaga, shavi, ard. Ulaanbaatar: Academy of Science Publishing. Nelson, F., & Agrawal, A. (2008). Patronage or Participation? Community-based natural resource management reform in Sub-Saharan Africa. Development and Change, 39(4), 557–585. Nixson, F., & Walters, B. (2006). Privatization, income distribution, and poverty: The Mongolian experience. World Development, 34(9), 1557–1579. Ostrom, E. (2009). Beyond markets and states: Polycentric governance of complex economic systems. Workshop in Political Theory and Policy Analysis, Indiana University, Bloomington, IN 47408, and Center for the Study of Institutional Diversity, Arizona State University, Tempe, AZ, U.S.A. Ostrom, E., Burger, J., Field, C. B., Norgaard, R. B., & Policansky, D. (1999). Revisiting the commons: Local lessons, global challenges. Science, 284, 278–282. Ostrom, E., Gardner, R., & Walker, J. (1994). Rules, games, and commonpool resources. Michigan: The University of Michigan Press.

BEYOND PROPERTY: CO-MANAGEMENT AND PASTORAL RESOURCE ACCESS IN MONGOLIA Ostrom, E., & Nagendra, H. (2006). Insights on linking forests, trees, and people from the air, on the ground, and in the laboratory. PNAS, 103 (51), 19224–19231. Peluso, N. L. (1992). Rich forests, poor people: resource control and resistance in Java. Berkeley: University of California Press. Polanyi, K. (1944). The great transformation. Boston, MA: Beacon Press. Regsuren, D., & Baljinnyam, A. (1973). BNMAU: Arkhangai Aimagiin Tuuh. Tsetserleg: ADKh Executive Administration, The Party Committee of Arkhangai Aimag. Ribot, J. C., & Peluso, N. L. (2003). A theory of access. Rural Sociology, 68(2), 153–181. Robbins, P. (2004). Political ecology: A critical introduction. Malden, MA: Blackwell. Sampildendev, H. (1985). Malchin ardiin zal uiliin ulamjlal. Ulaanbaatar: State Printing Place. Saunders, F. P. (2014). The promise of common pool resource theory and the reality of commons projects. International Journal of the Commons, 8(2), 636–656. Schmidt, S. M. (2004). Pastoral Community Organization, Livelihoods and Biodiversity Conservation in Mongolia’s Southern Gobi Region. In Annual Meeting of Society for Range Management, Salt Lake City, USA. SDC (2010). Swiss Agency for Development and Cooperation SDC, Ulaanbaatar. Sikor, Th., & Muller, D. (2009). The limits of state-led reform: An introduction. World Development, 37, 1307–1316. http://dx.doi.org/ 10.1016/j.worlddev.2008.08.010. Sneath, D. (2003). Land use, the environment and development in postsocialist Mongolia. Oxford Development Studies, 31(4), 441–457. Sneath, D. (2004). Property regimes and sociotechnical systems: Rights over land in Mongolia’s ‘‘Age of the Market”. In K. Verdery, & C. Humphrey (Eds.), Property in question: Value transformation in the global economy (pp. 161–182). Oxford, New York: Berg. Sneath, D. (2006). The rural and the urban in pastoral Mongolia. In O Bruun, & L Narangoa (Eds.), Mongols from country to city: Floating boundaries, pastoralism and city life in the Mongol lands (pp. 14–161). Nordic Institute of Asian Studies. Sneath, D. (2007). The headless state. New York: Columbia University Press.

379

Sturgeon, J. C., & Sikor, T. (2004). Post-socialist property in Asia and Europe: Variations on ‘‘fuzziness”. Conservation & Society, 2(1), 2–17. Tumenbayar, N. (2000). Land privatization option for Mongolia. In Constituting the commons: Crafting sustainable commons in the new millennium, the Eighth Conference of the International Association for the Study of the Common Property. Bloomington, Indiana, May 31– June 4. Undargaa, S. (2006). Gender and pastoral land use in Mongolia: Dilemmas of pastoral land tenure. Auckland: Centre for Development Studies, The University of Auckland. Undargaa, S. (2013). Property ’owners’ without rights? Exploring property relations and access in the Herlen Bayan-Ulaan Reserve Pasture Area of Mongolia. Canberra: Crawford School of Public Policy, Australian National University. Upton, C. (2005). Institutions in a pastoral society: Processes of formation and transformation in post-socialist Mongolia. Comparative Studies of South Asia, Africa and the Middle East, 25(3), 584–599. Upton, C. (2009). ‘‘Custom” and contestation: Land reform in postsocialist Mongolia. World Development, 37, 1400–1410. http://dx.doi. org/10.1016/j.worlddev.2008.08.014. Upton, C. (2010). Living off the land: Nature and nomadism in Mongolia. Geoforum, 41, 865–874. Upton, C., Moore, K., Nyamsuren, N., & Batjargal, E. (2013). Community, place and pastoralism: Nature and society in post-soviet Central Asia. University of Leicester, Mongolian State University of Agriculture. The Leverhulme Trust. Verdery, K. (1999). Fuzzy property: Rights, power, and identity in Transylvania’s decollectivization. In M. Burawoy, & K. Verdery (Eds.), Uncertain transition: Ethnographies of change in the postsocialist world (pp. 53–82). Lanham: Rowman and Littlefield Publishers. Verdery, K. (2004). The property regime of socialism. Conservation & Society, 2(1), 190–198. Vreeland, H. H. (1954). ’Mongol community and kinship structure’ Walter Hines Page School of International Relations. New Haven: The John Hopkins University. Yeh, E. T. (2004). Property relations in Tibet since decollectivisation and the question of Fuzziness. Conservation and Society, 2(1), 107–131.

Available online at www.sciencedirect.com

ScienceDirect