Introduction

Whether individualizing rangeland access leads to positive or negative outcomes is contested for many mobile pastoral systems. Typically, rangelands in arid areas used by mobile pastoralists are characterized as common pool resources with low excludability, low resource rents, and high environmental variability (Goodhue and McCarthy 2000; Ostrom et al. 1994). These characteristics allegedly challenge the practicability of individualized rights (van den Brink et al. 1995). Field evidence supports this characterization, as most rangelands used by traditional mobile pastoralists are managed under common property regimes or regimes with less clearly defined boundaries (Fernandez-Gimenez and Le Febre 2006). In contrast, western, commercial “ranching” systems are dominated by privatized access to rangelands, even in arid regions (Galaty and Johnson 1990). Nevertheless, because ranches of a fixed size are not always able to supply enough forage as rangeland productivity varies, ranchers mostly rely on external inputs or sometimes develop agreements for reciprocal access to each other’s rangelands (McAllister et al. 2006).

Individualization policies in communally managed rangelands are criticized mostly because the formalization of titles ignores the multiple layers of use rights accounted for in customary tenure regimes (Meinzen-Dick and Mwangi 2009). Shifts to individualized rights are often associated with commercialization and changing land use, e.g., to agriculture or stationary forms of herding, whose sustainability is questionable (Behnke 2008; Beyene 2010; Lesorogol 2003; Williams 1996).

The large rangeland areas in Caucasian and Central Asian (CCA) transition countries are a major experiment in restructuring rangeland access, the starting point in this case being socialist property regimes (Verdery 2004). While livestock was privatized in the late 1990s to varying degrees in the various countries (Kerven 2003b), in most cases governments enabled individualized access to pasture plots. However, problems with insufficient implementation of and compliance with individualized rights persisted in the subsequent years (Suttie and Reynolds 2003). To solve these problems, options for the reorganization of property rights and their advantages and disadvantages have been discussed (Banks 2003; Behnke et al. 2005; Fernandez-Gimenez 2002; Fernandez-Gimenez and Batbuyan 2004). Authors reject individualized tenure options, arguing that common property and co-management solutions are better adapted to the common pool resource characteristics of rangelands (e.g., Banks 2003; Fernandez-Gimenez 2002).

This paper contributes to this debate by presenting a new case study on mobile pastoralism in post-socialist Azerbaijan. A comparatively large number of studies shed light on the transition processes of arable land in Azerbaijan (Dudwick et al. 2005; Kaneff and Yalcin-Heckmann 2003; Lerman and Sedik 2010; Yalçin-Heckmann 2005), but the widespread existence of mobile pastoralism in Azerbaijan has been largely neglected by researchers. Valuable works on pasture management and improvement were prepared during the Soviet period (Aliyev et al. 1965; Prilipko 1949), and the culturally closely related Shahsevan pastoralists of northern Iran are well researched (Tapper 1979b).

The aim of this paper is twofold: first, to systematically review the literature on the Central Asian and Caucasian pastoral transition processes regarding arguments in favor of or against individualized pasture access; and second, to analyze pasture leasing and the related institutional framework in Azerbaijan, where individualization has been widely implemented. Finally, informed by evidence from the case study, I critically revisit the arguments in favor of and against individualized tenure.

I use the framework for analyzing social-ecological systems (SES) described by Ostrom (2009). SES are understood as social systems “in which some of the interdependent relationships among humans are mediated through interactions with biophysical and non-human biological units” (Anderies et al. 2004:3), which accurately describes pastoralism in Azerbaijan. The SES framework was developed for the analysis of resource management problems and places particular emphasis on the interrelatedness of users, governance, and ecology (Ostrom and Cox 2010). The components of the framework facilitate a complete description of the case, which is useful for both the analysis and the generalization of results (Ostrom 2009).

The Debate on Individualized Rangeland Access in Caucasian and Central Asian Countries

Pastoralism in all CCA countries was significantly impacted under Communist regimes mainly through the collectivization of livestock and pastures, the organization of herding into collective or state farms, restriction and regularization of seasonal movements, and provision of supplementary fodder (Scholz 1995). A large-scale, input-intensive system of pastoralism with reduced mobility evolved up to the 1980s. In the transition period the framework for rangeland utilization was heavily restructured in all countries. Livestock was mostly privatized while the state retained authority over pastures (Kerven 2003a). Exceptions are Turkmenistan and Uzbekistan where only extensive management rights for livestock have been transferred while the state retains ownership (Behnke et al. 2005; Zanca 2000).

CCA countries differ in terms of the extent of their pasture resources, which are more widespread in Central Asia than in the Caucasus (FAOSTAT 2011), as well as mobility patterns, which vary between vertical (dominant in the Caucasus, Kyrgyzstan and Tajikistan) and horizontal (dominant in Kazakhstan, Uzbekistan or Mongolia) migration (Scholz 1995).

Individualization policies in the agricultural sector, including for pastoral land, were regarded as central in the transition process since private property rights to resources are one important precondition for the successful establishment of a market economy according to western economic theories (Csaki and Lerman 1994; Lerman et al. 2004). According to property rights theory (Demsetz 1967), exclusive, individual rights foster the conservation of resources and agricultural development.

In most countries, the restructuring of rangeland access and management took place later than the reorganization of arable land and was only partly successful. Governments generally directed greater attention to the privatization of arable land than rangeland, either providing no special legislation for rangelands (Robinson et al. 2010; Schillhorn van Veen et al. 2004) or stipulating confusing and opaque regulations (Fernandez-Gimenez and Batbuyan 2004). Lease regimes in particular were poorly implemented. A number of publications evaluate the results of the first restructuring efforts and offer recommendations for improving pasture legislation, generally favoring common property or co-management solutions on the basis of one or more of the following arguments:

  1. 1)

    Common pool characteristics, the uneven pasture quality and distribution of water points and salt licks make the subdivision of pastures and boundary supervision of individualized plots unfeasible or difficult. The claim that pastures are common pool resources characterized by the extractability of private benefits and the non-excludability of other users (Ostrom 1990) is supported with the extensiveness and low productivity of pasture, and the seasonal variability of resource availability (Banks 1997; Banks et al. 2003). Historical non-exclusive usage supports this classification (Li et al. 2007). Due to the low resource value fencing seems unfeasible, while monitoring of exclusive rights to large unfenced tracts of pasture is difficult (Banks 1997; Li et al. 2007).

  2. 2)

    Individualization inhibits mobility, which is central to the sustainability of pastoral resource use. This argument rests on empirical evidence that pasture resources under mobile management are less degraded than those under stationary management (Sneath 1998). Being able to move over large areas allows pastoralists to exploit spatially and temporally varying niches of resource availability (Fernandez-Gimenez 2002). It also allows them to track resource availability – a key requirement for good pasture management in variable environments (Robinson et al. 2010; Robinson and Whitton 2010). However, the most important constraint to mobility is lack of access to transport and infrastructure (Baibagushev 2011; Farrington 2005; Kerven et al. 2012; Robinson and Whitton 2010; Sternberg 2008). Individualized pasture access can also inhibit movement when obtaining access rights to different pasture plots causes high transaction costs (Schoch et al. 2010) or when it blocks migration routes, for example with fences (Ning and Richard 1999; Schillhorn van Veen et al. 2004; Zhaoli et al. 2005).

  3. 3)

    Individualization of pastures undermines flexibility, which is necessary to cope with low spatial and temporal predictability of forage supply. Most arid and semi-arid grasslands used by mobile pastoralists are characterized by inter-annually variable precipitation patterns, which lead to non-equilibrium ecosystem characteristics with variable forage supply (Ellis and Swift 1988; Scoones 1994). Initially described for African grassland ecosystems, this paradigm of rangeland ecology has important implications for management, which under these conditions is more successful if mobility patterns are flexible and opportunistic (Ellis 1994). In rangelands with prevailing non-equilibrium characteristics, the commonly used variability indicator, the Coefficient of Variation, exceeds 33 % (Ellis 1994), but there is a continuum between equilibrium and non-equilibrium conditions in rangelands (Fernandez-Gimenez and Allen-Diaz 1999). Regarding CCA countries, the high frequency of droughts and severe winter weather require flexible mitigation strategies (Banks 1997, 2003; Banks et al. 2003; Bedunah and Harris 2002; Li and Huntsinger 2011; Ojima and Chuluun 2008). Reciprocal access regulations to other users’ pastures in emergencies are crucial for maintaining flexibility (Johnson et al. 2006; Li and Huntsinger 2011). In addition, pastoral societies show a high degree of social flexibility as the composition of herding groups may change frequently (Fernandez-Gimenez 2002). Under individualized ownership or tenure spatial and social boundaries harden, directly inhibiting mobility or increasing transaction costs for herders who need access to other’s pastures (Fernandez-Gimenez 2002; Kerven et al. 2011; Vanselow et al. 2012). Reduced flexibility might also harm the long-term capacity of pastoralism to adapt to environmental changes, such as climate change (Li and Huntsinger 2011; Ning and Richard 1999).

  4. 4)

    Individualization of pastures leads to negative distributional consequences and exclusion of the poor. Adverse effects of distribution processes in CCA countries, such as regulations favoring rich and excluding poor pasture users, are widely reported (see, e.g., Schillhorn van Veen et al. 2004; Undeland 2005; Zhaoli et al. 2005). This must be attributed first to the greater ability of richer households to deal with formal procedures and their better personal networks (Robinson and Whitton 2010). In addition, vague regulations for the distribution process may open potential for land grabbing (Schillhorn van Veen et al. 2004). Especially regarding tenure regulations, distributional effects might be mixed up with effects of mobility and demand for pasture: because mobile pastoralism requires a large herd and complementary assets such as access to labor and transport, richer households are commonly more mobile than poorer households and therefore have a greater interest in gaining formal access to remote pastures (Alimaev and Behnke 2008; Kerven et al. 2011; Vanselow et al. 2012). However, negative distributional effects and exclusion of poorer households may also occur under common property and fuzzy access regulations (Upton 2009), for example, the exclusion of poorer herders from the distribution of campsites in Mongolia (Fernandez-Gimenez 2002; Fernandez-Gimenez and Batbuyan 2004).

  5. 5)

    Individualization is not appropriate in the cultural and social context of nomadic pastoralists. Pastoralists may have preserved a strong ethic of reciprocal access based on the conviction that pastures cannot be individually appropriated (Fernandez-Gimenez 2000, 2002) so that government regulations instituting exclusive access to rangelands are unlikely to be implemented on the ground. Norms of reciprocal pasture access among pastoralists are especially strong in Mongolia (Fernandez-Gimenez 2000), and individual appropriation of pastures has led to conflicts in Kyrgyzstan or Georgia (Tsomaia et al. 2003; Undeland 2005; see also Banks 1997; Li and Huntsinger 2011 for information on China).

  6. 6)

    Individualization of pasture rights leads to the destruction of herding groups. This has multiple adverse effects, such as loss of economies of scale, increased labor pressure for households, and breakdown of social capital. During the appropriation of pasture under the Household Responsibility System in China, households were required to herd their livestock on their private pastures, which were sometimes fenced with enormous state subsidies. The economies of scale inherent in group herding were lost, leading to a simplification of herding structures (Li and Huntsinger 2011) and increased workloads for pastoralists (Zhaoli et al. 2005). In addition, the social capital created by self-regulating pasture access was lost, causing a reduction in pastoralists’ relational assets in other spheres, for example coordinating flexible movements (Li and Huntsinger 2011). When discussing future changes in rangeland access regulations, individual ownership or tenure is directly associated with a breakup of herding structures, whereas group tenure or ownership is directly linked to collective herding structures (Banks 1997; Robinson and Whitton 2010).

  7. 7)

    Individualization of pasture access should enable investments; however, this link is weak in arid and semi-arid grasslands. Economic theory argues that individual pasture tenure facilitates investments, which lead to user benefits and to long-term preservation of pasture resources. However, this link is weak in arid and semi-arid grasslands due to the low productivity of pasture resources (Banks 1997; Fernandez-Gimenez 2002; Sternberg 2008). However, investments in specific structures, such as barns or water pumps, may increase the productivity of pastoral production (Behnke et al. 2005).

  8. 8)

    The lack of a cadaster inhibits the registration of leases, the definition of boundaries, and the eventual planning of stocking rates. The lack of a cadaster inhibited the implementation of the “old” pasture law in Kyrgyzstan (Undeland 2005) and is used as an argument against the tenure approach in Mongolia (Fernandez-Gimenez 2002). A number of authors recommend collective forms of rangeland access, mainly community-based resource management (Ning and Richard 1999; Robinson and Whitton 2010; Zhaoli et al. 2005) and rangeland co-management (Banks 1997; Li and Huntsinger 2011). Fernandez-Gimenez (2002) recommends the regulation of seasonal movements for Mongolian rangelands. However, Kerven et al. (2012) note that collective management may be heavily influenced by power relations within local communities and may thus fail to achieve socially and ecologically sustainable results.

Methodology

Field investigations were conducted in winter and summer study areas targeting two mobile pastoralist systems in Azerbaijan (Fig. 1). The eastern system links winter pastures in Gobustan (study region 1) with summer pastures in the Greater Caucasus (study region 2), while in the western system herders migrate between winter pastures in the Jeiranchel region (study region 3) and summer pastures in the Lesser Caucasus (study region 4).

Fig. 1
figure 1

Mobile Pastoralism in Azerbaijan and study regions. Map source: Aliyev et al. (1965)

Full size image

The study regions were selected for covering core regions of mobile pastoralism on the largest possible east–west gradient. Because rules of pasture use, their implementation, and farm organization were strikingly similar in both systems, I analyze the data together and note minor differences in the results.

The study follows a largely qualitative approach because it aims to explain causalities of pastoralism development in Azerbaijan with limited ex ante information (Yin 2003). The fieldwork was an iterative process distributed over five periods of one to three months in 2007 and 2008, in which literature, documents and statements of different interview partners were constantly cross-checked. I conducted semi-structured and structured interviews, analyzed official documents, and engaged in participatory observation (Table 1). Interviews were translated from Azeri to German by consecutive interpreting and were transcribed into a database. Losses through translation were kept as low as possible by recapitulating each interview together with the translator.

Table 1 Field data collection and study regions coverage
Full size table

Focus was laid on comparing information from the administration with information from herders. Officials working in pasture administration explained the de jure rights and the administrative view on individualized leases, while the herders provided information on the de facto implementation of regulations, customary rights, and their view of the practicability of the present regulations.

Semi-structured interviews were conducted with representatives state organizations, including the SLCC (State Land and Cartography Committee), its regional offices in the districts, and officials in seven district administrations who were responsible for pasture leases (icra hakimiyetti), and seven municipalities (belediye). In the study regions, municipalities leasing pastures to mobile pastoralists are found only in the summer pastures (study regions 2 and 4).

For the description of pasture users and their organization, the reference unit is the farm, which is one herding unit based on one winter pasture. In some cases several farms are managed jointly by one owner in a livestock enterprise; however, in most cases the farms are only loosely linked. A farm comprises several households having a common budget. Households according to this criterion may be composed of persons living on the pastures and in a village.

Among herders, I conducted semi-structured interviews on pasture leases with 68 farm managers. While I visited more than 100 farms in each study region for mapping purposes, I stopped at every third to fifth farm to conduct an interview. Thus, interviewed farms were distributed evenly in the study regions. However, not all farm managers could provide the required information because the responsibility for pasture leases sometimes lay with other persons, for example relatives or absentee owners. Thus, interview partners with special experiences, for example subleasers or absentee owners, were approached selectively using the snowball method (Bailey 1978). The interviewed administration officials and farm managers provided lease contracts, maps, and other documents.

In-depth information about lease contracts was gathered on four key farms in the eastern system (Table 2). The farms were selected during the first field period to represent the largest possible differences in organizational structure (see below) and on the criterion of having especially cooperative farm managers who explained their views frankly and patiently. During repeated visits in each field period an atmosphere of trust developed in which sensitive questions could be discussed and information from other interview partners could be validated. I conducted participatory observation on these farms and gained insights into their internal structure. In the western system, information on farm assets and organization was collected in structured interviews at 56 farms in study region 3, for which descriptive statistics are provided.

Table 2 Information about key farms in study regions 1 and 2
Full size table

I researched historical data collected by ethnologists in the study region (Klug 2008; Salzer 2008) and results from Tapper’s (1979b) anthropological study of the Shahsevan nomads in northern Iran, who are ethnic Azerbaijanis. Ethnic Azerbaijanis in Azerbaijan and Iran were separated by the establishment of the Iranian-Russian border in 1828 but share the same cultural background and language. While in Azerbaijan Soviet rules were implemented with crude force, deeply changing the lives of mobile pastoralists (Baberowski 2003), the Shahsevan in Iran were less affected by reforms, although they, too, experienced a decline in pasture resources and the impacts of settlement policies (Tapper 1979b). Therefore, allowing for differences in political circumstances, we can expect traditional regulations of pasture use among the Shahsevan to be similar to those of mobile pastoralists in Azerbaijan.

Results

I follow the SES framework presented by Ostrom (2009), which also frames the analysis conceptually. I describe subsequently the social, economic and political setting, the resource system and units, the resource users, the governance system, related ecosystems, and the interaction of these factors.

Social, Economic, and Political Settings:

Azerbaijan’s transition process began later than in other post-Soviet countries due to the Nagorno-Karabakh conflict with Armenia. The period between 1990 and 1996 is characterized by a lack of formal institutions, political instability, and war in Nagorno-Karabakh, leading to a decline in production and consumption and growing inflation. However, after 1995, Heydar Aliyev’s government rapidly implemented reforms, making Azerbaijan a “moderate reformer” among transition countries from the late 1990s onwards (Kray and Csaki 2005; Lerman and Sedik 2010). The country benefits from its oil reserves which between 2002 and 2008 fuelled an economic growth of over 10 % per year. In 2011 Azerbaijan generated a GDP of 51,157.5 million AZN (1 AZN = 1.27 US$, Nov 2012; SSCA 2012). Despite these encouraging developments, the economy still suffers from insufficient institutional reform of the business environment. Furthermore, Azerbaijan scores poorly on indicators for democratic freedom and control of corruption (Lerman and Sedik 2010). A market trend that directly affects the pastoral economy is the growth in consumer demand for meat products, a result of increasing average income. Azerbaijanis traditionally prefer meat from locally slaughtered sheep, the main product of the pastoral economy since the transition in the 1990s (Economist Intelligence Unit 2011).

Numbers of livestock dropped between 1990 and 1996, but have since recovered significantly (Fig. 2). In 2011, 8.6 million small ruminants were registered in Azerbaijan, compared to the pre-transition peak of 5.7 million in 1989 (SSCA 2012). Most small ruminants are kept in mobile pastoral systems, although stationary herding is also practiced. Separate statistics on the mobile pastoralist sector are not available.

Fig. 2
figure 2

Development of livestock numbers in Azerbaijan from 1935 to 2012. Source: State Statistical Committee of Azerbaijan (2012)

Full size image

Mobility in Azerbaijani pastoralism was negatively impacted by the annexation of northern Azerbaijan by the Russian Empire in 1828. The tsarist and the socialist regimes perceived mobile pastoralism as a backward, inefficient, and an obstacle to the modernization of Azerbaijani society and implemented settlement policies. Later, collective pastoral farms allowed “well-ordered” mobility including the migration between summer pastures in the Greater Caucasus and winter pastures in the foothills. In addition, the large-scale conversion of pastures into irrigated arable land for cotton production in the Kura-Arax lowlands significantly reduced the total area of the winter pastures (Baberowski 2003).

More recently, pastoral mobility is negatively affected by the conflict in Nagorno-Karabakh, due to which a significant share of the summer pastures traditionally used by Azerbaijani herders is inaccessible. In addition, Georgian herders, who used to graze their livestock on winter pastures in Azerbaijan, are not allowed to cross the Azerbaijani-Georgian border (study region 3).

The Resource System: Pastoral Ecosystems With Buildings and Access to Water

Azerbaijan’s pasture resources include 1.7 million ha of winter pastures, mainly in the foothills of the Greater and Lesser Caucasus (0–700 m.a.s.l., study regions 1 and 3, Fig. 1; Mamedov 2003). The climate of the winter pastures is characterized by low precipitation (approx. 240–350 mm/year) and a considerable inter-annual variation of rainfall and thus forage supply (Huseynov and Malikov 2009). The Coefficient of Variation for the precipitation is 28 % in the driest parts of study region 1 and around 22 % in the wetter parts (Peper 2010). The summer pastures total around 0.6 million ha (Mamedov 2003) and are located in the Greater and Lesser Caucasus at altitudes of approx. 1400–3500 m.a.s.l. (study regions 2 and 4). They consist of subalpine and alpine meadows with a huge diversity of perennial grasses and herbs (Neudert, et al. 2013). The higher precipitation in these regions leads to a balanced forage production every year.

Besides pasture, buildings and the access to water are key assets in the pastoral production system. Stables and houses are used on the winter pastures. The condition of stables is a key determinant of livestock survival during adverse weather conditions. Water access is a problem on the arid winter pastures, where wells are rare. Rain is collected and stored, but is often insufficient to provide drinking water for animals and humans over the entire winter season. Farms therefore resort to external water supplies delivered by truck at high, though acceptable costs. From the farms interviewed with a structured questionnaire for study region 3, 15 (27 %) use cisterns, while 43 (73 %) rely on ponds. Among the key farms (Table 2), A and C have cisterns for livestock, while B and D continue to use ponds.

The Resource Units: Seasonal Variability of Forage Supply

Winter pastures are dominated by steppe and semi-desert vegetation (Peper et al. 2010). Dwarf shrubs constitute the main winter forage for livestock grazing on the winter pastures from October to May. The spring peak of precipitation leads to abundant growth of herbs and grasses before the summer weather makes livestock herding undesirable in these regions. The summer pastures with constant regrowth of vegetation are grazed between June and September.

In sum, the forage supply is scarce during winter, but usually abundant in spring and summer. Variability in forage availability is greatest on winter pastures in spring, when rain can be abundant or completely absent, and in winter, when extreme cold and snowfall can threaten the survival of herds.

The Users: Pastoral Farms and Their Organization

The most economically important livestock products for Azerbaijani herders are fattened lambs, which are sold on local markets or to traders visiting the pastures. The mode of production in Azerbaijani herding can be characterized as modern and strongly market oriented (Neudert and Allahverdiyeva 2009).

At the time of fieldwork in 2007–2008, herd size in Azerbaijan normally ranged between 200 and 2000 ewes (Table 3). Qualitative investigations indicate that in study region 1 herds are somewhat larger (up to 2000 ewes per farm). In addition to sheep, nearly all farms keep some goats and cattle.

Table 3 Assets of pastoral farms in Jeiranchel region
Full size table

Large farms can capture economies of scale in herding activities and for overhead costs. Because a typical household lacks both livestock and labor to run a farm on its own, livestock owners need to hire shepherds or cooperate with other households. Most pastoral households have a home village where they own a house and maintain their social connections to the village or neighborhood community. The shepherds may bring their own livestock, which is kept with the herd of their employer. Kinship and friendship ties influence the selection of cooperating households and shepherds. If several shepherds are engaged, their animals can constitute more than half of the total livestock on a farm. Three major organizational forms were identified:

  1. 1)

    Most farms are organized hierarchically (owner enterprise e.g., key farms A and B) where one entrepreneur owns the majority of livestock and employs shepherds, who may bring animals of their own. The owner pays all overhead and fodder costs, including the costs for the shepherds’ animals, and sometimes even for the shepherds’ food and clothing. He may hire different shepherds each year. Livestock entrepreneurs are typically among the richer members of the pastoral community (see Table 2).

  2. 2)

    Absentee ownership is very common among hierarchically organized farms (absentee owner enterprise, e.g., key farm C). The entrepreneur nevertheless retains authority in central decisions such as sale of livestock, pasture leasing, or winter fodder storage, but the daily management of the herds is carried out by head shepherds. Sometimes absentee owners even employ a manager, who may supervise several farms (key farm C). Absentee owners spend most of their time on activities which often contribute a larger share to their household income than livestock keeping. Their motivation for running a livestock enterprise is mainly driven by capital investment strategy considerations.

  3. 3)

    The least common form of farm organization is the cooperation of nearly equal partners (cooperative enterprise, e.g., key farm D). In cooperative enterprises, two to four livestock owners with 150–300 ewes each, who are often linked by friendship or kinship, join their herds. Their individual herds are not large enough to set up an owner enterprise but too large to easily find employment as shepherds. The members of a cooperative enterprise share the overhead costs for the pasture and the daily work. The structured questionnaire for study region 3 showed that 52 out of 56 farms (93 %) were organized hierarchically, while only four were organized cooperatively. Out of the total sample, 34 % (19 farms) were owned by absentee entrepreneurs.

My results are regarding organization of farms strikingly similar to the results of Tapper (1979b:39) described for the Shahsevan in Iran although recorded camp sizes in their winter pastures are larger (20–40 tents or households).

The Governance System: Administration of Individualized Lease

Rangeland access for mobile herders is formally granted by the local district administration or municipality in the form of individual lease contracts. The organization of leases observed in 2007–2008 is based on Azerbaijan’s reformed regulations for pastoral land tenure (Land Code, June 25, 1999). The implementation began with the distribution of the first lease contracts from 2000 onwards and was nearly completed in 2007–2008 (Neudert and Rühs 2013).

On the national level, pastoral land is under the jurisdiction of the State Land and Cartography Committee (SLCC), which also is responsible for the cadaster and the registration of all lease contracts longer than 1 year. Direct responsibility for lease contracts is assigned to district administrations (icra hakimiyetti) or municipalities (belediye). The district administrations control rangelands that are comparatively remote from villages, which constitute the greater share of pastures used by mobile pastoralists. Municipalities are responsible for rangelands in the vicinity of villages, and in the study regions they are found in summer pastures only. Nevertheless, the majority of summer pastures is leased out by district administrations. Municipal pastures are designated as a common resource for the village residents’ stationary livestock; however, if villages have excess pasture, they may rent it to mobile pastoralists. With district administrations, contracts can be issued over one to 99 years, but in practice 15 years is the normal term. Municipalities issue most contracts between one and 15 years duration, but some municipalities refuse to issue contracts longer than 1 year.

Lease contracts specify the size of the pasture and the lease price and include precise boundary maps of the allotted land but no prescriptions regarding the sustainable use of pastures. The prices for district pastures are legally fixed (0.34 to 1 AZN/ha/yr), making them low compared to other costs of herding. Municipalities can set their fees at any level as long as they are above the prices for district pastures. In 2007–2008 the lease price of municipality pasture could be as high as 6 AZN/ha/yr. The boundary maps for lease contracts from district and municipal administrations are prepared by the SLCC using the cadastral maps from Soviet times. The buildings and campsites on winter pastures are registered in the cadaster separately as private property.

Related Ecosystems: Herd Migration Routes

Pastoral farms influence other ecosystems as they pass through settled and forested areas during their seasonal migrations between summer and winter pastures. The migration routes are designated by the administration and campsites are provided in many places, although they may be used for one night only. The migration routes also pass by village pastures, which are used based on personal agreements between herders and the local municipality. If they are on good terms with the municipality, herders may use village pastures for several days.

Interactions: Implementation of Individualized Lease and Pasture Management

Individualized leasing has been almost fully implemented in Azerbaijan. Nearly every entrepreneur has lease contracts for his summer and winter pastures. Because one contract is issued for the grazing area of one farm, the size of leased pasture equals all land used by one farm (mean size: 415 ha, Table 3). In cooperatively organized farms, only one of the livestock owners functions as the legal leaseholder. Farm owners from mountain villages have preferential access to their municipalities’ pastures (see key farms A and D). Some of them can even rely on verbal agreements, which take their credibility from the good personal relations between the municipality and the farm owner (e.g., key farm A). If the use pressure on the village pastures is low, mobile pastoralists can also access common pastures.

Boundaries between the properties of different leaseholders are clearly marked by streams, ridges, or artificial stone piles. The implementation of clear boundaries is facilitated by informal rules of reciprocity and mental models of individualized pasture use among pastoralists. It is a matter of reciprocal respect not to violate the neighbors’ boundaries, and this principle remains unquestioned even by farm owners with insufficient pasture resources. Conflicts about borders of pastures and trespassing are rarely mentioned as major problems and are mainly associated with inconsistencies in lease contracts as described for key farm D.

Villagers sometimes oppose the leasing of common pasture land to mobile pastoralists as this reduces their access to pasture for their own livestock; this as well as shows their opposition to municipal administrations exploiting monetary incentives. Nevertheless, it is clear that villagers are not opposed to leasing pasture land in general. For the seasonal migrations of livestock every herd has the right to access all pasture land.

The allocation of lease contracts happened very quickly, with the majority of lease contracts issued within 2 or 3 years, and the process was nearing completion in 2007–2008. As a result, the actual leaseholder selection process was based mainly on personal relations, patronage relationships remaining from the Soviet period, or payments to officials and disadvantaged poorer livestock owners (Neudert and Rühs 2013). For example, key farm D did not succeed in obtaining a lease contract.

A further option for obtaining rights to pastures is sublease. While technically illegal, field investigations showed that this rule was not being enforced. Sublease arrangements are found in both summer and winter pastures. A sublease for a winter pasture in study region 3 costs 1000 to 2000 AZN (1270–2540 US$) per 400–500 ha pasture, compared to up to 5000 AZN (6350 US$) per approx. 700 ha pasture in study region 1. This is 2.5 to 7 times as much as the administration lease price. Sublease arrangements come without legal documents and imply no long-term rights for the sub-lessee. Subleasers use the pasture either permanently, typically livestock owners having enough livestock and entrepreneurial skills to successfully run a farm (e.g., key farm D), or short-term, generally individuals who have various reasons for not wanting to lease a pasture permanently (e.g., key farm B with regard to summer pastures). A livestock owner explained he subleased winter pastures from season to season, depending on the quality of the forage.

Individualized rights to pastures are widely accepted in the study regions. Herders have local names for the pastures, which were fairly clearly delimited even before legal individualization. However, whether historical ownership to pastures was individual or clan-based remains unclear. During pre-Soviet times, herders originating from the lowlands rented plots in the summer pastures from the local authorities in mountain villages (Klug 2008; Salzer 2008).

Buildings have to be registered by owners of the lease contract for the surrounding pasture separately as private property; however, the initial cadastral registration of buildings had not been completed in 2007–2008. Entrepreneurs felt the need to register and thus secure their ownership of buildings only if they had made substantial investments. However, many old stables remain from the Soviet era, which herders rebuild as needed. Consequently only few newly built stables are registered. Information from the key farms indicated that the two richer entrepreneurs (key farms B and C) rebuilt some stables and houses, while the less well-off entrepreneurs of key farms A and D restricted their investments mostly to their village or town houses. Investment on key farm D is particularly limited due to its short-termed sublease arrangement.

The two annual migrations between summer and winter pastures and general pasture management are adapted to forage availability. Even though winter pastures are accessible, migration to temporarily available summer pastures allows them to rest. The most important traditional pasture management is of winter forage (xam), which needs to last the herd until March when plant growth restarts. Forage is rationed by extending the grazing area of the herd day by day into areas ungrazed during the season. By the end of March, 10 % of the pasture should be left in case the spring rains come late or fail entirely. Thus, on some pastures the reserve area may remain un-grazed in some years (see Tapper 1979b). On the summer pastures, forage grows constantly, so the herds utilize the whole pasture every day, or with short rotations of at most four days. Resting pastures over an entire summer season is not practiced.

There are several strategies for coping with variability of environmental conditions in the winter pastures. According to herders, fat and healthy animals are the best insurance. In addition, livestock owners store hay and barley to tide the herd over periods of extreme cold. But because extreme cold spells occur erratically, some livestock owners systematically store less than advised by traditional knowledge, mainly due to a lack of ready cash. But in an emergency, they may be forced either to buy fodder at inflated prices or face livestock losses of up to one half of their herd.

During local or regional spring droughts, livestock owners bring in water by truck and move their herd to find water and forage (Table 2). In spring 2008, when a drought affected the lower parts of study region 1, key farms B and C, having trucks and water tankers of their own, brought in water and fodder from external sources. Moving the herd to other pastures was facilitated by either by paying for access to unoccupied pastures (key farm B) or by relying on traditional reciprocal access to pastures (key farm D). However, the owner of key farm D reported that people in his home village denied him access to their pasture. It seems, therefore, that this traditional right is no longer generally respected. Lease contracts do not address the topic of reciprocal access in case of drought.

Discussion

This case study from Azerbaijan shows that mobile pastoralists do not necessarily require collective tenure of pasture lands for mobile herding practices. It should be noted that pastures in Azerbaijan are still not completely privatized, but remain state property. However, since plots with clear boundaries are leased out to individuals who enjoy far-reaching management and use rights the property regime is strongly individualized. Nevertheless, during seasonal migrations, every herder is allowed to move freely on the pastures.

Based on this evidence some arguments against individualized tenure (see above) can be rejected, while others remain valid. The extent and openness of rangelands, which invite classification as a common pool resource, do not generally mean that boundaries are not respected and other users excluded. In Azerbaijan exclusion is enforced without fences, based upon an informal rule of reciprocal respect among pasture users. That common-pool resource characteristics do not determine management systems even in theory is acknowledged (Bromley 1997; Ostrom et al. 1994), but is occasionally ignored by empirical researchers in post-socialist countries. Similarly, in Azerbaijan the demarcation of individual boundaries does not inhibit regular mobility. Decisive factors enabling mobility in the pasture access framework are regulations accompanying individualized leases, such as the right to access any pasture land and the official maintenance of migration routes.

The predicted demise of mobile pastoralism as a consequence of individualized leases has not occurred in Azerbaijan, although pasture access regulations do not explicitly recognize herding groups (cf. Banks 1997). However, in Azerbaijan I observed a high proportion of hierarchically organized farms, an exception in the CCA region (see Kerven et al. 2012). Individualized leases do appear to encourage investments. In Azerbaijan, this applies especially to resources such as stables, and undermines the claim that in semi-arid rangelands investment and productivity increases are only weakly linked.

In addition, in Azerbaijan restricted access rights to pasture were practiced in pre-Soviet times, and consequently individualized leasing fits the cultural and social context, and indeed is even supported by informal rules among pasture users. This prevalence of individualized rights is exceptional for the CCA region. Individual access traditions seem more likely to occur in areas with higher rainfall, higher population densities, and higher pasture productivity (Tapper 1979a). In addition, Azerbaijan’s existing pasture cadaster facilitated the rapid implementation of the lease approach, in contrast, for example, to Mongolia (Fernandez-Gimenez 2002) and Kyrgyzstan (Undeland 2005).

Some of the negative consequences of individualized leasing as observed or predicted by other authors did occur in Azerbaijan. I found distributional effects of individualized tenure, which are probably stronger than elsewhere. Access to lease contracts was in many cases facilitated by personal relations and patronage networks. Long-term, costly sublease arrangements deprive households that have long been engaged in herding of assets they urgently need. These arrangements probably developed as a reaction to the high market value of rangelands, which is not reflected by the low lease prices. The sublease practice observed in 2007–2008, with its high prices and insufficient security for permanent sub-lessees has on the whole negative effects. According to a market economic framework, unrestricted legalization of sublease would lead to the efficient allocation of resources. Yet current developments in Azerbaijan suggest rather that it would boost land grabbing by wealthy urbanites. Thus, moderate alterations in the legal framework are needed: for example, regulations for the exchange of pasture shares would contribute to a better allocation of pasture resources, at least among neighbors.

Employment as a shepherd or participation in a cooperative gives a wide range of households de facto access to pastures. Nevertheless, the stratification of pastoral society into entrepreneurs and shepherds might harden due to these regulations. Traditionally, political and economic stratification in pastoral societies is relatively low due to the unpredictability of livestock survival (Fernandez-Gimenez and Le Febre 2006). However, stratification is not primarily a result of individualized tenure but of increasing market-orientation and integration of pastoralists into state structures (Salzman 1999). In addition, individualized leases as implemented in Azerbaijan in 2007–2008 does not provide enough flexibility for herders to mitigate variable weather conditions and adjust stocking rates to actual forage supply. Flexibility could be added by complementing the lease approach with other regulations, such as a controlled legalization of subleases.

One advantage of individualized leases that should be noted is that they make it possible to exclude potential users and to regulate stocking rates and thus ensure the sustainable use of limited resources. Restricting use might also become more important for other rangelands with more pronounced non-equilibrium characteristics as the increasing market orientation of pastoralists and the availability of external resources remove some of the constraints on herd size. Under collective management, mechanisms for exclusion and regulation rely on participants’ capacity and will to cooperate. However, in post-socialist societies social capital remains scarce, especially in medium-sized groups (Mearns 1996; Rose-Ackerman 2001). Therefore, if local collective pasture management is established, we can expect developing self-governance to be challenging and to require long-term external assistance. This is especially important if pastoral societies cannot draw on a heritage of collective regulatory mechanisms. In addition, despite the proven capacity of societies to coordinate and self-regulate resource management (Baland and Platteau 1996; Ostrom 1990), the future prospects of newly established collective pasture management systems are dim under massive economic and social pressures, such as rising livestock numbers and the resulting pressure on pasture resources, increasing market orientation of pastoral production, and an increased connectedness of pastoral societies with the “outside” world in general. Reacting to these pressures requires a constant reworking of rules, which is difficult even for societies with a more recent, successful history of collective management systems (Fleischman et al. 2010). Under these circumstances it remains questionable whether pastoral societies in CCA countries have the time and capacity to build stable cooperation and self-regulation mechanisms.

Finally, there is the question of whether individualized leases, in forms similar to that found in Azerbaijan, are appropriate for other post-socialist countries. I have shown that some arguments against individualized tenure do not hold for the specific circumstances in Azerbaijan, implying that institutional possibilities may be more diverse than the binary categories of individual or collective management suggest. Nevertheless, there are historical, social, economic, and ecological differences among CCA countries beyond the scope of this paper. Given the complexity of regulating pasture use, insufficiently implemented regulations do not automatically indicate a general misfit of the local approach to tenure; rather, such problems in many cases point to the need for continuing efforts in adjusting and improving the existing regulations.