1In many destination countries in the global land rush, the state lays legal claim to large swathes of land, including land occupied and used by smallholders and managed under a variety of customary governance systems. This disjuncture between formal and de facto property is exacerbated by many states not even knowing the extent of their landholdings (FAO et al., 2010), and has in recent years been used to put large amounts of land into play through the blurring of both legal and cartographic boundaries (Cotula et al., 2009; Deininger and Byerlee, 2011; HLPE, 2011; Borras Jr. and Franco, 2012). This imprecise legal geography has major implications for investors who have been lured into so-called ‘frontier’ markets by promises of cheap and abundant state land (Adler and So, 2012; Borras Jr. and Franco, 2012; de Leon et al., 2013). If state land is actually state-owned in the sense of being demarcated and uncontested, it can give investors attractive incentives: ‘one-stop’ acquisition, efficient regulation and, most important, low cost of access. But when land is state-‘owned’ in only the formal sense––that is, when it remains occupied, used, or even locally held under soft forms of title––investors are vulnerable to a range of delays, additional costs, and reputational risks (Munden Project, 2012; de Leon et al., 2013). Whether state ownership of land is actual, merely formal, or somewhere in between is thus a subject of great interest.
2The role of the private sector in helping bridge the gap between formal and de facto property remains a key point of contention in debates about land grabbing, responsible agricultural investment, and the gulf in between (Cotula and Leonard, 2010; FAO et al., 2010; UNCFS, 2012). In many ways Cambodia is exemplary of the global land grab problem, in which local elites and foreign investors collaborate with each other and with state officials to acquire and develop large concessions of putatively state-owned (but previously un-demarcated) land. State land concessions in Cambodia, various estimates of which range between 1 and 3 million hectares (Titthara and Boyle 2012a; ADHOC, 2013; 2014), have reportedly flouted legal provisions on landholdings and community protection, and many have reportedly involved substantial violence and displacement yet delivered minimal or no benefit to locals (LICADHO, 2009; Chak, 2011; Müller, 2012; Kuch and Zsombor, 2013; Neef et al., 2013). But while this may be the standard story within the Cambodian land sector (see Gironde and Senties Portilla; Cismas and Paramita, both in this volume), this pattern has nonetheless become a growing liability for the actors involved––not just companies and their investors, but also state officials at multiple levels, and even foreign donors. This article focuses on two cases where companies have sought to develop more socially benign––and, they believe, more profitable and sustainable––plantation concessions in a context that is still marred by extensive land conflict. The first is the Mong Reththy Investment Cambodia Oil Palm (MRICOP) Company (Preah Sihanouk province); the second is the Grandis Timber Company (Kampong Speu province). We examine each case empirically, investigating how two well-intentioned yet strategically oriented companies navigate the complex landscape of multiple entitlements and competing claims that lurk beneath the surface of state ownership in Cambodia’s rural hinterland. In doing so, we interrogate the private sector’s role in helping address the state land problem, both in Cambodia and elsewhere.
3The two cases sit within a national regulatory context that is increasingly linked to what government officials, private sector actors and the media call the government’s Leopard-Skin (Sbek Khla) policy. The term, apparently coined by state officials around 2010 and circulated increasingly since (CDC, 2010; RGC, 2012a; 2012c), remains vague for some, and contested for others in its status as well as its intended operational meaning (Milne, 2013; Beban, 2014). For some, it is a clear policy, carrying significant government weight (if not quite the force of law), while for others it is more like a concept, approach, or formula that is still being tested, but that holds significant appeal because of its contrast with the status quo (ADHOC, 2014). The basic idea behind it boils down to concession development through regularisation rather than eviction: treating illegal smallholders––occupants whom official discourse often terms ‘encroachers’––as legitimate parts of the economic landscape, and developing concessions around them rather than by evicting them first (CDC, 2010; RGC, 2012c; ADHOC, 2014). One foreign advisor to the Cambodian land sector described the policy as ‘leaving the people where they are and just using the rest’ (Müller, 2012, 10). Similarly, the Phnom Penh Post (Becker 2012) called the approach a ‘workaround strategy’ in the literal sense of leaving small-scale farmers in place and making companies work around them.
4Regularisation is hardly a new idea in Cambodian policy circles. A preference for avoiding evictions has been stated in various policy documents and discussions for over a decade (see e.g. CLP, 2002, 27; RGC, 2007; MAFF, no date). Yet, as has often been noted, the concept has been minimally implemented (Müller, 2012). What is notable about the last few years is the political currency that regularisation has gained, as land conflicts, evictions, and land-related arrests mushroomed, especially in the run-up to Cambodia’s national elections in July 2013 (ADHOC, 2013; Un, 2013). The case studies presented below are notable for providing a closer look at what regularisation efforts in Cambodia have actually looked like. The cases were chosen for their ability to show Leopard-Skin development in practice––both companies have publicly committed to avoiding eviction and made the business case for leaving local land users in place (Chakrya and Sherrel, 2011; Becker 2012)––and because each has been identified by outside experts, including third party certifiers (Intertek-Moody International, 2012; GFA Certification, 2013), as being a positive example in a concession landscape defined largely by its governance failures (for MRICOP, see UNCOHCHR, 2007, 20; Chakrya and Sherrel, 2011. For Grandis Timber, see IWA, 2011; Müller, 2012, 11; and Becker, 2012).
5Looking at how private sector actors negotiate the landscape of existing property claims and entitlements that confront their efforts to develop concessions of so-called state land is useful because it provides a window into the social, political-economic and legal dynamics that exist in pluralist contexts where statutory, use-based, and patronage-based ownership norms all vie for supremacy––contexts where, in short, ‘the law is not the law’ (Adler and So, 2012). Studying private sector efforts to take advantage of state-legal land claims while also acknowledging local, use-based understandings of ownership––at least in part––provides insight into the political economy of concession development, framing regularisation as a strategic move for legitimising state land concessions at a moment of crisis, and showing how the private sector performs for the state the difficult task of making the legal abstraction of state land operational on the ground. In examining this reciprocity, we show how concession development risks blurring the lines between public authority and private interest, but also suggest that at least some in the private sector are thinking more pragmatically than the government about local land entitlements. Whether this pragmatism should be equated with good practice, however, is a difficult question; the cases thus show the need to look closely at how Leopard-Skin development deals with entrenched relations of power and marginality in rural landscapes. Finally, by gesturing to both the needs and challenges associated with getting accurate information about how land acquisition occurs, the cases provide opportunities to connect with wider debates on transnational regulation, including third-party certification and soft law (see Cismas and Paramita in this volume).
6This paper is based on findings from desk research and a field study involving key informant interviews (including company directors, project staff, local officials, and residents of project-area communities) conducted in November and December 2011, as well as additional investigation during 2012 and early 2013, the periods during which both projects were being evaluated for third-party certification. The study is based on best available information. However, as is typical in the Cambodian agribusiness sector, most project documents such as impact assessments and land inventories remained unavailable to us. While rough project timelines could be constructed from fieldwork and online data, access to project documents would have increased our ability to analyse and explain the cases more substantially. Opacity is typical of the Cambodian agribusiness sector (and of investment in Cambodia more generally), but this lack of access to documents presented a limitation nonetheless. Second, our focus, both conceptually and methodologically, is on questions of land access; while these are inevitably connected to larger issues of livelihood, sustainable development, and so on, such wider issues as such are beyond the scope of this chapter. (For linkages to livelihood, human rights, and a governance analysis of agricultural investments in Cambodia, see Messerli et al., Gironde and Senties Portilla, and Cismas and Paramita, all in this volume).
7While Cambodia has featured centrally in the ‘global land grab’ debate of recent years (GRAIN, 2008; Deininger and Byerlee, 2011; Neef et al., 2013; Baird, 2014), land grabbing and related problems of unmapped state land are hardly new. Dating back to the social and economic dislocations of Khmer Rouge rule (1975-1978) and its aftermath (during the 1980s and 1990s), the distribution of land and associated questions of access and ownership have long been at the heart of contemporary Cambodian politics (Chandler, 1993; Gottesman, 2003; Hughes, 2007; Un and So, 2009; Cock, 2010; Heder, 2011). One of the central motifs through which this history has unfolded is ‘anarchic encroachment’, a term that state officials often use to describe the activities of smallholders on lands that are claimed as state property. Anarchic encroachment is far more complicated than just smallholders, however, harkening back not just to the legal disarray that followed the Khmer Rouge’s removal in 1979, but also––and, we argue, especially––to contemporary Cambodian state policy predicated on the allocation of state land for development purposes.
- 1 Cambodia is hardly unique in this regard, although a wider comparison is beyond the scope of this (...)
8Despite having an official policy of collectivised production (as in Vietnam and Laos at the time), the Cambodian government that came to power in 1979 had little choice, given the memory of Khmer Rouge rule, but to allow the traditional model of smallholder farming to return on a wide scale (Chandler, 1993; Gottesman, 2003). This occurred during the 1980s, and in preparation for the Vietnamese military withdrawal from Cambodia at the end of the decade (and presaging the formal UN-mediated transition period of 1992‒1993) the government offered farmers the chance to formalise these entitlements; by the end of 1991, roughly 4 million applications had been filed (So, 2009, 106). Only a fraction of these applications were eventually converted into actual certificates of ownership, however, in part because the Land Law of 1992 excluded agricultural land from eligibility for private ownership (Van Acker, 1999, 37; Cooper, 2002, 17). Instead, in a move that favoured elite patronage over smallholder populism, state officials focused on handing out concession rights to powerful individuals, most (in)famously for logging. Thus neoliberalism, in the form of ‘public enterprise privatization’ (RGC, 1994; MEF, 1995), articulated with the politics of patronage; timber concessions, for example, increased from roughly 2.2 million hectares in 1994 (World Bank et al., 1996, 4; also see Global Witness, 2002, 3) to as much as 8 million hectares by the end of the 1990s (Chan et al., 2001; also see Springer, 2011).1 Writing in 1999, an Oxfam legal advisor lamented the growing landlessness that accompanied this first concession boom, noting that ‘the current system is unable to cope with the pressure it has come under from the newly opened market economy and the growing population […] [F]or the first time in its history, Cambodia is experiencing a shortage of arable land’ (Williams, 1999, 2).
9The discourse of ‘anarchic encroachment’ (RGC, 1999) came into widespread use during this period as government officials sought to describe the rural land situation without directly implicating economic elites, whose activities were a key part of the problem, but who were also key political allies. Encroachment on state (e.g. forested) land, then as now, frequently involves the supervision, resources, and political connection of elites who offer access to land and/or work to the poor and socially vulnerable. This system held steady for a time, especially during the 1990s when the Khmer Rouge were still a military threat from the forests of the Thai border region (Le Billon, 2000), and elite resource patronage could be widely tolerated ‘as a necessary evil’ because it provided much-needed economic stability and development (Hughes, 2007, 840). But as the Khmer Rouge threat receded, neo-patrimonialism’s blatant unfairness––its ‘discriminatory enforcement of laws and regulations, discretionary provision of monopoly franchises, concessions and contracts, and diversionary collection of public revenues and disbursement of state lands, funds and employment’ (Cock, 2010, 263)––became increasingly disruptive. Seeking to place land development on a firmer legal basis, the government rewrote the Land Law in 2001.
10The 2001 Land Law is notable for its strategic orientation: much like that which the Leopard-Skin policy is currently attempting to achieve (see below) The 2001 Land Law is also notable for including provisions that recognise smallholder entitlements at both the household and community scale, but that also develop a flexible and powerful doctrine of state land ownership, particularly when it comes to the allocation of concessions. In the spirit of continuing the process of ‘reconstituting ownership over immovable property after the period of crisis from 1975 to 1979’ (Article 29), the new law thus reversed the 1992 Land Law’s exclusion of agricultural land from the legal definition of smallholder ownership, and outlined pathways for titling both individual smallholdings and indigenous communal lands (see also Grimsditch and Henderson, 2009; Adler and So, 2012). At the same time, however, it ended the establishment of new rights of possession, singling out ‘the private property of the state and public legal entities’ as a priority area where the practice of ‘encroachment’ via the establishment of possession rights needed to be stopped (Article 17).
11The new law also created a powerful doctrine of state land, strengthening the legal foundation for ongoing ‘public enterprise privatisation’ and positioning concessions as a key piece of the state’s development repertoire. ‘State public’ lands, which include areas of natural and cultural significance, as well as areas with infrastructure for general public use, were protected from alienation (Article 15), although they could be transferred out of state-public status ‘when they lose their public interest use’ (Article 16); Everything else––all state lands that were not state-public lands––were state-private lands by default, and eligible for allocation via concessions (Articles 48-62). Concessions could take two broad forms: economic land concessions (ELCs), aimed at cultivating large-scale investment, and social land concessions (SLCs), aimed at alleviating landlessness through the provision of surplus state land.
12ELCs were already in existence, but needed legal grounding and protection after the free-for-all of the 1990s. Building on a 1989 Council of Ministers’ Instruction (No. 3) and the 1994 Investment Law, the 2001 Land Law extended their maximum duration to 99 years, stipulated that ELC land be put to use within twelve months, and imposed a limit of 10,000 hectares per concessionaire. Existing ELCs over 10,000 hectares, of which there were roughly a dozen at the time (UNCOHCHR, 2007), were allowed to maintain their size if reduction ‘would result in compromising the exploitation in progress’ (Article 59). SLCs were a new invention, oriented explicitly towards replacing ‘encroachment’ with a more orderly and managed solution to landlessness and demographic expansion. After prohibiting further ‘encroachment’ on state land, Article 17 of the new law thus contained a provision for allowing ‘vacant lands of the State private domain [to] be distributed to persons demonstrating need for land for social purposes in accordance with conditions set forth by [a future] sub-decree.’ The new Land Law, anticipating a move that would be repeated in 2012-2013 (see below), thus sought to harness the allocation of state land to the aspirations of the landless and land-poor.
13Unfortunately, the geography of ELCs and SLCs has never approached anything like parity. Although the enabling sub-decree for SLCs was issued more than a year and a half before the one for ELCs, the SLC-granting process has continued to lag far behind that of ELCs. A German advisor to the Cambodian land sector recently put the number of SLCs at a few thousand hectares, lamenting that ‘as a gross summary, it has to be stated that 99 per cent of the distributed state land was handed over in long-term leases of up to 99 years to national and international investors, to the detriment of the rural poor, who got only a 1 per cent share’ (Müller, 2012, 3-4). This asymmetry was mirrored in the accompanying bureaucracy: the slow allocation of SLCs is often explained by its adherence to rules, while ELC allocation has proceeded apace (Un and So, 2011; Müller, 2012: 3; ADHOC, 2013). The same advisor quoted above noted that despite the extensive regulatory requirements governing ELCs––including social and environmental impact assessments, community consultations, legal efforts to avoid relocation, and formal land use planning, allin advance of signing a contract––these prescriptions have been ‘widely ignored’ (Müller, 2012, 6).
14The current landscape of Cambodian concessions thus poses a problem. On the one hand, the allocation of ELCs has been so extensive that many worry that Cambodia’s arable land is ‘all but gone’ (Titthara and Boyle, 2012a). On the other hand, much of this land – land that is located inside formal concessions but has yet to be actually alienated – is still being used by local communities. Even as the intersection of post-2001 neo-patrimonialism and the global land rush has created widespread (and putatively low-rent) corporate land access, land conflicts and entitlement losses for Cambodia’s rural poor and indigenous communities have become a growing political liability. It is into this breach that the Leopard-Skin policy has sought to step.
15In July 2011, one of Cambodia’s most famous tycoons delivered an unexpected lecture to his fellow members of the Cambodian senate. Publicised widely in the days that followed, Oknha Mong Reththy’s speech criticised the rash of evictions and legal violations that have plagued Cambodia’s land sector in recent years. Referring to ‘some investors’ who ‘claim that they need to evict people to develop their concessions’ Reththy said: ‘I disagree with this tactic––these people have been living on their land for generations. Where will they go when they’re kicked out?’ According to Chakrya and Sherrel (2011), he ‘urged investors to employ local residents on their concessions rather than hiring outsiders,’ and ‘called on the government to create a panel to verify that companies granted ELCs abided by their [legal] conditions.’ At the time he made this speech, Reththy was himself confronting land tenure-related business risk. Just days later, a prospective USD 115 million joint venture with France’s largest sugar company was reported to have fallen through (Hul, 2011); while the precise timing and details remain unclear, Reththy’s speech suggested that Cambodia’s business elite––many of whom rely on foreign capital for joint-venture partnerships––were feeling the need to distance themselves publicly from the scourge of land grabbing.
- 2 MAFF, ELC profile section, Mong Reththy Investment Cambodia Oil PalmCo., Ltd, ‘Legal Papers and Ri (...)
16Reththy’s comments were not just posturing, however. They were based on his experience developing his flagship MRICOP concession, an oil palm plantation that was named in an influential United Nations human rights report as Cambodia’s only ‘successful’ ELC (UNCOHCHR, 2007, 20). This project had its roots back in 1993, when Reththy purchased some land in northern Preah Sihanouk province with the help of a local intermediary (Pal, 2010, 233-234). Two years later, Reththy submitted a proposal to the Council for Development of Cambodia, requesting land on which to develop an oil palm plantation. By November 1995, he had been approved for an 11,000-hectare concession in Ta Ney village, in the southern part of Choeng Kou commune (Figure 8.1); by January 1996 the contract was signed.2 In March of 2000, Reththy signed a second contract, adding a second, smaller ELC of 1,800 hectares. Despite having areas, Reththy’s ELCs were not given precise locations; other than commune and village names, no concession maps existed at the time. As this case illustrates, the ELC therefore functioned in practice more like a general license to develop the northern part of Preah Sihanouk province.
FIGURE 8.1: Mong Reththy Project Area (hatch), Northern Preah Sihanouk Province.
Note: The precise location of his ELC was not specified, and is approximated here; granted ELC sizes are shown at lower left for reference.
- 3 Interview with village and commune officials, December 2011. Unless otherwise indicated, interview (...)
- 4 According to local officials, this study was carried out by a MRICOP associate designated personal (...)
17Soon after its initial contract was signed, MRICOP conducted a study with local officials officials to identify area residents who would be impacted by the plantation.3 This study, while not an Environmental and Social Impact Assessment (ESIA) of the type later mandated by law, created a list of local occupants and provided a basis for the negotiations that followed.4 Three compensation options were documented in different key informants’ accounts: (i) monetary compensation in exchange for their land; (ii) land-for-land exchange outside the concession area; and (iii) exclusion of the person’s land from the project.
18MRICOP’s ELC acquisition in the 1990s was characterised by a situation in which Reththy was viewed by many local residents (and even some officials) not only as an investor, but also as someone who carried state authority. Interviews with area residents who lived along the main north-south road (see Figure 8.1), and thus tended to experience the company’s land access efforts early in MRICOP’s history, tended to give rather critical accounts (also cf. Lang, 2000). While respondents took pains to avoid directly criticising Reththy himself, they nonetheless raised concerns about their unequal relationship with the company, a pattern typical of the Cambodian context more generally. In a few instances, residents explained that they felt pressure to accept cash compensation or land exchange simply by virtue of living near the plantation of a wealthy and powerful developer. They reported that people were told they could remain on and continue to cultivate their land as they normally would but were also warned about causing damage to the new plantation crops. In this context, rumours and innuendo were used to significant effect. One area resident reported deciding to sell after ‘hearing people say that my land was under a development project and would be taken’ anyway.
19The presence of the company also created a speculative land market geared towards providing MRICOP with land. Some residents reported the presence of speculators who used the name of the company to undermine residents’ faith in their own land tenure, thus acquiring land that would have otherwise sold at higher prices (or possibly have been kept). One villager explained that land brokers bought from people at lower prices and sold to MRICOP at higher prices (for example buying at USD 200 per hectare and selling for USD 700-800). Residents reportedly sold to speculators out of fear they would lose the land anyway. This pressure to sell was reportedly sometimes the result of collusion between speculators and government officials, or in some cases between officials and people claiming to be Reththy’s employees.
20These accounts point to a heterogeneous mix of property claims encountered by the company; this hardly accords with the notion of a state-owned countryside. In this context, farmers who lacked documentary proof of ownership were able to have their entitlements recognised. But these entitlements were also precarious. Rumours of living ‘under the Oknha’s development project’ were mobilised by rival elites, land brokers, and even some who claimed to be Reththy’s employees, and were used to convince reluctant smallholders to sell ‘before it was too late.’ While it is impossible to determine how often this took place, it is clear that various actors exploited the fears of poor land users, capitalising on the fact that the MRICOP concession area was geographically ill-defined. This is hardly a situation of prior state landownership. Rather, it is a case where locally existing landholdings were recognised to a point, but also rendered precarious through the granting of a concession to a powerful businessman over a general area.
- 5 This paragraph is based on interviews with residents of Reththy 1 Village, December 2011.
21At later stages, MRICOP’s land acquisition efforts were more positively experienced, especially when they were paired with the work opportunities and expanded infrastructure that accompanied the conversion of Kaev Phos––a coastal village in the concession’s western zone––into a port facility. During the building of ‘Oknha Mong’ Port in Kaev Phos in 2003‒2004 (see Figure 8.1), residents described having their pictures taken in front of their old homes, and receiving land transfer documents certified by local authorities with these pictures attached. This process of land exchange was the same regardless of the gender of the residents and was described by respondents as being widely accepted. One resident explained that locals had not only received new land but gained access to a health centre, a school, and better roads thanks to the company. Another resident of ‘Reththy 1’, a village built to resettle Kaev Phos’s former residents, described actually making money after he sold some of the land he received as part of the compensation process.5
22This diversity of responses highlights the issue of timing: negative responses were related to MRICOP’s land acquisition in the 1990s, whereas positive responses referred to events in the following decade, when MRICOP had more business relations with clients around the world, including European companies. Presumably to facilitate selling in European markets, MRICOP began the process of joining the Roundtable on Sustainable Palm Oil (RSPO) in the mid-to-late 2000s. While the company’s membership was initially delayed due to the land conflict reported by Lang (2000), it was granted in late 2011 after certifiers determined that the dispute had been resolved in an acceptable manner (Intertek-Moody, 2012, 12).
23Grandis Timber is a teak plantation company that has been developing an ELC in western Kampong Speu province since 2009. Catering to a small but growing group of sustainability-oriented ‘frontier’ investors,6 the company aims to turn its plantation into ‘an asset class [that] hits […] financial returns and is socially responsible and environmentally sustainable’ (Daniel Mitchell, manager, quoted in Becker, 2012). Most of its plantation land is a former logging concession, and the company emphasises its mix of reforestation, conservation, and economic development activities. The company is a joint venture between the SRP International Group and Danish and Swedish pension funds, and was certified by the Forest Stewardship Council in July 2013 (FSC, 2013).
24Like MRICOP, Grandis has attracted significant attention for its publicised belief that it is cheaper, as a feature about the company in the Phnom Penh Post (Becker, 2012) put it, ‘to treat people well and have them stay in their homes and work their small farms, rather than pushing them off the land.’ Grandis was featured as a field-trip destination by the International Woodfibre Association (IWA, 2011), and lauded by the Phnom Penh Post in the paper’s above-mentioned feature as a company that was successfully implementing the Prime Minister’s Leopard-Skin approach to development. Grandis’s emphasis on local entitlements also caught the attention of German cooperation (Gesellschaft für Internationale Zusammenarbeit, GIZ) land sector advisors, who have been working with the Cambodian government’s SLC program in recent years (Bickel and Löhr, 2011; Müller, 2012; Neef et al., 2013). This interest relates to the company’s policy of working around existing land users even if they do not have legal rights to the land (Müller, 2012) — a preference for regularisation over eviction that exemplifies the Leopard-Skin approach. As this case demonstrates, however, this is rarely a straightforward process.
- 7 Interview with commune official, Phnom Srouch District, December 2011.
25Grandis’s ELC is located in the foothills of western Cambodia’s Cardamom Mountains (Figure 8.2), a part of Kampong Speu province where permanent settlement is comparatively recent. During the 1980s the area was insecure and therefore sparsely populated, especially following a Khmer Rouge offensive in 1989 that scattered many residents and left Reaksmey Samaki Commune (to the north) almost entirely empty (see Gottesman, 2003).Repopulation began in significant numbers in the late 1990s, and especially in the early years of the twenty-first century, right around the time the 2001 Land Law was passed. Some settlers were drawn by the low density of farmsteads, others by reports that government officials were distributing land. Local officials identified areas where new settlement could occur, and in many cases helped settlers file applications for legal possession rights. One commune chief described the rapid but erratic expansion of settlements and farmland as producing ‘villages that stretched like rubber bands’, with residences and agricultural plots frequently separated by wide distances, and with few ways for authorities to regulate the geography of settlement and production.7
FIGURE 8.2: Grandis Timber Project Area, Western Kampong Speu Province.
Source: the authors
- 8 Interview, same as previous.
- 9 Despite being moved, the SLC fared better than many (cf. Neef et al., 2013), and was finalised in (...)
26In 2007 a local commune chief proposed a 1,200-hectare Social Land Concession to be established in Prey Torteng and Krang Deyvay villages, at the southern end of what eventually became the Grandis concession (Figure 8.2). The SLC was approved. But when, in early 2009, Grandis proposed using the same area for its own ELC, the SLC had not yet materialised.8 Ultimately, it was decided to move the proposed SLC to a different location nearby, and to develop the Grandis ELC in the area where the SLC had been proposed.9
27Setting its sights on full legality, social responsibility, and environmental sustainability, Grandis embarked on an effort to delimit its concession from locally used land, to carve out patches of good forest for conservation, and to begin the certification process with the FSC. A feasibility study conducted by company representatives and local authorities was finalised in late July 2009, and identified 682 hectares of landholdings, used by 310 families. These lands were mapped using GPS and demarcated with cement posts in order to show local residents that Grandis intended to respect their land boundaries, and, presumably, to help prevent encroachment on the company’s land. The survey also identified large areas of intact forest to be set aside for a variety of environmental uses. Excluding these forest lands and the 682 hectares of local landholdings, Grandis’s final proposal for its ELC came in at 9,820 hectares, just under Cambodia’s legal limit. A month after the proposal, the company signed an ‘Agreement in Principle’ with Cambodia’s Council of Ministers. On the last day of 2009, Grandis signed a concession contract with the government, giving the company the legal right to develop a teak plantation in the area identified by the survey.10 Land clearing began in early 2010.
- 11 Interviews with residents and local officials, Krang Deyvay Commune, December 2011.
28Many residents we spoke to described the company’s activities, including its ongoing presence, in positive terms. One resident of Krang Deyvey village, for example, noted that many locals, including those living inside the concession area, were working for the company. Although this was day labour work on the Grandis ELC, the combination of wages (USD 2.5 per day) and transportation (‘company trucks come into local villages every morning to pick up workers and drop them off in the evening’) suggested that it was a relatively good option compared to other local alternatives. A local official also praised Grandis for its provision of work opportunities, although noting the challenge of not enough skilled people locally to meet the company’s needs. This meant some skilled workers were recruited from other provinces.11
29The company has also allowed local residents to harvest tree stumps from its concession area. These are a valuable resource for charcoal production—an important source of cash, especially for land-poor and landless households. Taking wood for charcoal in other areas is classified as illegal and regulated by the forest administration with serious consequences for those found extracting wood.A local official we spoke to thus also praised Grandis’s policy on providing access to tree stumps, albeit that it is also a means of clearing the land and therefore of benefit to the company.
- 12 Interview with a Grandis employee, December 2011.
30Somewhat surprisingly, Grandis’s decision to respect existing land uses rather than legal land rights eventually brought the company into conflict with state authorities. Following the 2009 survey and corner marker placement, company staff drew up GPS-referenced paper maps to hand out to residents. These ‘farm map documents’ were not intended to carry legal meaning, but rather to formalise the company’s commitment not to encroach on villagers’ land;12 presumably they could also have been used as a form of prior agreement if cases of encroachment emerged in the future. However, cadastral officials worried that the documents could be (mis)interpreted as proof of legal tenure rather than simply as company-acknowledged limits on its own concession land. Faced with the charge of encroaching on the state’s authority to define and demarcate property, Grandis backed off its initial plan and decided not to distribute the documents.
31In a context of widespread land conflict marked by a mix of unequal power relations, insufficient land administration capacity, and the popular perception that land rights are a reform-unfriendly area, this paper has examined the possibility of ‘better practice’ within the modality of state land concessions. As the economic and political liabilities of land conflict are increasingly felt, both in Cambodia and elsewhere in the global south, a key question that remains is the extent to which this room to manoeuvre hinges on local contextual circumstances versus more general factors. The cases examined above provide a number of insights.
32Notwithstanding their individual differences, one of the most striking features of both the MRICOP and Grandis Timber cases is their family resemblance to the Land Law revision that took place in 2001, albeit with a key difference. As noted in Section 2, the 2001 Land Law moved in two directions simultaneously: it offered significant potential benefits to smallholders and communities in the form of titles, but it also strengthened the legal doctrine of state land by drawing a ‘line in the sand’ when it came to establishing new rights of possession. Cutting off the establishment of new possession rights (after 2001) was a key feature of the new Land Law, and was part of the larger project of strengthening the state’s legal claims to land as a way of replacing the allegedly ‘anarchic’ development of land with more governable and efficient means. The cases presented here show how the Leopard-Skin approach is pursuing a variant of this ‘line in the sand’ approach, but this time with respect to actual possession rather than legal possession rights. By acknowledging the de facto right to keep lands that are already under concession-area-households’ possession, MRICOP and Grandis are pursuing a version––from their perspective a much more practical one––of the possession rights compromise enshrined in the 2001 Land Law. The difference is that, unlike the 2001 Land Law (where legal respect for possession rights was limited by the state’s ability to formalise those rights in a timely manner), these companies are actively working on the ground. By developing the lands that come up to the borders of current smallholder plots, companies like MRICOP and Grandis can enforce the ‘line in the sand’ compromise in ways that the 2001 Land Law failed to.
33This shift from recognising legal to actual possession has major implications. The intention to mitigate conflict and avoid needless and harmful evictions should be acknowledged and built upon. As global supply chains (not only in palm oil and teak, but in timber, sugar, rubber, maize and other commodities) become increasingly transparent, tenure-related risk is growing; as it does, the business case for these types of steps increases. However, the ad hoc approach examined above is forced to rely on local power relations to regulate land-related negotiations between local users, companies, and state officials. This is a tough sell to communities and their advocates; while third-party certification may be able to alter incentives by increasing transnational oversight, the degree to which it can tip the balance towards equality is doubtful. One advantage of law is that it can, in theory, give equal rights to parties who, in practice, do not have equal power. Moving the basis of land-related negotiation from law to case-by-case negotiation may work well in particular cases (and as the MRICOP case shows, within the same case at particular times), but it depends on the benevolence of the powerful, and it further attenuates the already weak capacity of law to protect the most vulnerable members of society.
34The shift towards recognising (without legally recognising) local land possession seems, then, to be geared towards improving the legitimacy of the state land concession system at a time when it is facing increasing strain from both local communities and transnational investors (see e.g. Brinkley, 2013; de Leon et al., 2013). This essentially tactical dimension is important because it highlights the issue of timing in maintaining power asymmetry between local land users and would-be concessionaires. The German advisor quoted above in Section 2 described (in another part of the same paper) the rationale of the Cambodian Ministry of Land Management, Urban Planning and Construction (MLMUPC) when it came to Grandis’s significance: the ministry, he said, ‘was initially reluctant to implement Circular 02 [on regularisation], giving the reason of not wanting to encourage further encroachments. But MLMUPC [began] to pilot Circular 02 [in 2012] […] in Kampong Speu province where an institutional investor from Denmark [i.e. Grandis Timber] […] has already anticipated the ‘leopard skin’ feature of the Prime Minister’ (Müller, 2012, 11; emphasis added). The cases examined above show how Leopard-Skin development is attempting to prevent ‘further encroachments’: namely, by allocating concessions first, and conducting regularisation only in areas where smallholders are already surrounded.
35This approach is expanding. In May 2012, the Cambodian government expanded the ‘Leopard-Skin formula’ significantly with the issuing and implementation of a prime ministerial order (No. 01BB) on ‘strengthening and increasing the effectiveness of the management of economic land concessions’ (RGC, 2012a). Implementation has focused largely on demarcating rural landholdings, especially in concession-contested areas, and issuing titles or small-scale concession agreements to occupants (Titthara and Boyle, 2012b; ADHOC, 2014; Grimsditch and Schoenberger, forthcoming). While this process was initially praised for reorienting land titling towards rather than away from socially marginal communities, it has come under substantial criticism for its opacity, its uneven implementation, its inability to actually address the land conflicts it encounters, and its creation of social divisions in indigenous communities forced to choose between individual and communal titles (Titthara and Boyle, 2012b; Rabe, 2013; Milne, 2013; Beban, 2014; ADHOC, 2014). Notwithstanding these issues, the significance of the process is clear for the purposes of this paper. Order 01BB represents a model of Leopard-Skin development based on the allocation of state land to deserving recipients (‘in order to favour the conditions of land development’—RGC, 2012b), rather than the provision of title based on existing legal rights. The Order-01BB campaign was thus widely criticised as an election-year ploy––and it may well have been that––but it is also significant in that, much like the cases examined here, it acknowledged occupation-based rights in exchange for the acknowledgment of the state’s right to distribute land. In this move away from a rights-based model of property recognition, like the cases presented above, it exemplifies the reciprocity between public authority and property-making described by Sikor and Lund (2009). If the MRICOP and Grandis cases showed how private actors can blur the lines that surround this authority––substantially in the case of MRICOP, and merely in the form of a threat (exemplified by the farm map documents) in the Grandis Timber case––the Order-01 campaign illustrates the state’s efforts to claim the mantle of Leopard-Skin development clearly for itself.
36The cases of MRICOP and Grandis Timber raise the questions of where the line between ‘better’ and ‘good’ concession practice lies, and of the degree to which those concerned with reforming land governance in the interest of smallholders should pursue good concessions in addition to their efforts to promote alternatives. These are fraught questions, and we refrain from trying to answer them definitively here. Our research found evidence pointing in both directions, with significant improvements over time (in the MRICOP case) as well as significant proactive work carried out by both companies (especially by Grandis) to minimise community contestation up front. But the jury is still out; high-quality, responsible, and sustainable investments go beyond the short term (and beyond the issue of land access alone). What is needed most, both in the cases examined above and in other investments more broadly, are improved transparency to allow greater public scrutiny; free, prior, and informed consent of local land users; independent investigation; and effective grievance mechanisms—all of which are basic components of most guidelines and regulatory frameworks on land and investment governance. Third-party certification has provided some of this in our two case studies, and its impacts should not be minimised. But the limits of audit-based research are also significant: it is limited to occasional visits and forms of remedy (i.e. additional investigation, and ultimately denial or revocation of certification) that may not enable or inspire communities to voice concerns openly. Indeed, much like the community-based natural resource management (CBNRM) paradigm, without adequate transparency and local ownership, certification could end up legitimising enclosure and thus stifling expressions of discontent by affected communities (cf. Tubtim and Hirsch, 2004).
37Time will thus tell whether private sector efforts to ‘work around’ local land uses will produce truly sustainable development. While we would prefer to see an investment model that allows communities to deal with investors directly as empowered landowners, we realise the gulf between these positions is nontrivial. Acknowledging communities’ de facto rights not only in the present, but for future needs as well, entails a business model predicated on the community having a much greater role in representing the public interest. Current experiments at the ‘better practice’ end of the concession spectrum are likely to shape the contours of this debate for the coming years. In the meantime, their ability to sustain their investments and the communities in their midst’s remains to be seen.