Do collective property rights make sense? Insights from central Vietnam

We draw on empirical results from three case studies of property rights change across forest and fisheries ecosystems in central Vietnam to investigate the circumstances under which collective property rights may make sense. A generic property rights framework was used to examine the bundles of rights and associated rights holders in each case, and to assess these arrangements with regard to their contextual fit, legitimacy and enforceability. The cases illustrate the interactions between private and collective rights to lands and resources, and the trade-offs inherent with different mixes of rights.


Introduction
Responding to the challenges of rural poverty and environmental sustainability requires a flexible mix of individual and collective property rights. Resourcebased activities shift, depending upon, among other things, household needs, local ecologies and market opportunities. For these reasons, conventional categorization or advocacy of private, collective or public rights rarely account for the complex realities found in particular places (Barry and Meinzen-Dick 2008;German and Keeler 2010). Many property rights arrangements tend to enclose specific areas or reduce some people's access to specific goods. Overlapping but differentiated 'bundles of rights' (Schlager and Ostrom 1992) and hybrid property regimes can offer a more effective lens for understanding property rights complexity. In the context of a mixed public-private or collective rights situation, such bundles of rights may be related to access, withdrawal, management, exclusion and alienation of resources, or parts of a resource, through time and space (Barry and Meinzen-Dick 2008). Farmers or fishers may advocate for part of a bundle of rights (extraction, for example) with other rights residing with the collective or the state (management or alienation, as an example). Sensitivity to circumstance or context reveals that individual, collective and public property rights each have merits (Evans et al. 2010). The challenge for the resource manager, donor or policymaker is to 'read' when and where different rights regimes may be appropriate to support poverty alleviation and sustainable rural livelihoods more generally.
Property rights in Vietnam make for an interesting case. Vietnam has moved from forms of collectivization and state ownership that began in the late 1950s to an ambitious 'renovation' program leading to individual land titling in the late 1980s (Do and Iyer 2008). The Doi Moi period (or 'renovation') aimed to transform a centralized, state-planned economic system into a more decentralized, market-oriented system whereby the private sector would become the main engine of growth 1 . One aspect of these reform policies was to devolve authority over production decisions to farmers and enterprises, and to establish property rights (for agricultural land and in some cases for individual households to manage forest areas) to encourage investment and provide a form of collateral for rural dwellers (Sunderlin et al. 2008). The majority of Vietnam's 90 million people have access to small amounts of land (1-2 ha), particularly in rural, agriculturefocused areas (where 72% of the population lives) (HDR 2009). Policy reforms in the 2000s (e.g. changes to the 2003 Land Law and Fisheries Law) recognized the role for collective rights, once again, to manage forest areas and fishing grounds. However, in the context of increasing privatization of land and marketization of rural production, the contextual fit, legitimacy and enforceability of collective rights has been uncertain.
Drawing on empirical results from three case studies of property rights change across forest and fisheries ecosystems in central Vietnam, we investigate the circumstances whereby collective property rights may make sense. Specifically, we seek to (i) understand the complex bundles of rights that exist across upland and lagoon ecosystems and the implications for livelihoods and environmental sustainability; (ii) examine the circumstances in which individuals and local institutions may seek to organize both for individual and collective property rights; and (iii) to reflect on the fit (social, ecological), enforceability and legitimacy of these collective rights arrangements. Insights from the cases highlight how the needs and aspirations of individuals and households do not easily conform to conventional property rights narratives (private vs. collective) or the implementation of policy prescriptions that emerge from these narratives. Results of the analysis contribute to common property theory by showing how local actors may choose to collectively manage and use natural resources (forest lands and aquatic resources in this case) as part of a broader strategy to obtain individual bundles of rights (which may include access, withdrawal, management, exclusion and alienation of resources, or parts of a resource) within the context of a collective rights policy framework.

Understanding property rights and the Vietnamese context
Property relations exist in multiple forms, and include laws, regulations, cultural norms, social values, social relationships or property practices. These arrangements are legitimized in the sense that the state or some other form of politico-legal authority (e.g. customary) sanctions them (Sikor and Lund 2009). Property rights define or delimit the range of privileges granted to individuals health care (HDI 2009). Agriculture remains an important source of employment and continues to contribute to Vietnam's economic growth (25% in 2000 compared with 21% in 2009) (VNB 2010). Export commodities include agricultural products such as rice, coffee, rubber, pepper, cashew nuts, and tea along with natural resources such as oil and marine products (VBN 2010). Market institutions are firmly established in rural Vietnam, although the state intervenes actively in land markets, supplies many inputs in agricultural production and plays a large role in many organizational activities (Markussen et al. 2010). or collective entities regarding specific assets or resources (Meinzen-Dick and Mwangi 2008;Mascia and Claus 2009). Rights to property (land, a set forest, individual trees, or a fishery area) may be held by an individual, shared by a group, or held collectively by multiple groups, and may be granted to user groups, villages, government agencies, and non-governmental organizations. While private property is something that all households aspire to gain access to, particularly in resource-dependent areas, there is also a role for collective rights, and a significant body of scholarship (Ostrom 1990) that offers evidence of its historical and current importance.
Commons scholars follow from the extensive research done by Ostrom (c.f., Ostrom 1990Ostrom et al. 1999) and others to illustrate the potential of collective rights. Commons research has shown how community groups are capable of managing resources given certain conditions and appropriate incentives. There are numerous cases of commons governance (c.f., the database held at Indiana University), with research demonstrating how community forestry or fisheries can work both for natural resource conservation and, in certain cases, to produce local economic benefits (c.f., Barsimantov et al. 2011 for a review). Management may happen without government support, in cases of high capacity or relative isolation, although increasingly it is recognized that government involvement is an essential component of commons management (Khumsri et al. 2008). The characteristics of a resource, levels of trust among actors, and rules in use also play an important role (Berge and van Laerhoven 2011). Most examples of common pool governance are small-scale and single-case. Yet from a policy perspective, linkages between tenure arrangements and tropical forest or fisheries conservation has become increasingly important as countries look to devolve forest and fisheries resources to local communities as a means of conserving forests and producing local economic benefits. Market-based conservation in collectively managed areas is also being applied, to varying degrees of success (c.f., Ricketts et al. 2004;McCauley 2006). Recent scholarship also illustrates how economic inequality can have a negative effect on resource governance outcomes (c.f., Andersson and Agrawal 2011).
Common property management does have its challenges. For example, it can be part of a general off-loading of responsibility to lower management units (see Berkes 2010). This type of management arrangement becomes particularly challenging where resources are transboundary and in areas with high poverty rates (Cox 2008). Even with a plethora of new policy supporting commons management and collective rights arrangements, getting the right fit between collective goods (conservation) and individual benefits (livelihoods) remains problematic (Marschke 2012). Much of the common property and collectiveaction literature has focused on the characteristics of the user group and attributes of the resources (Agrawal 2007). Thus, there is a need to consider the role of external influences, including the role of market drivers on tenure regimes and the costs of collective action (Barismantov et al. 2011), along with a need to draw on political economy processes to consider the social and political context in which Do collective property rights make sense? Insights from central Vietnam 5 communities operate (Clement 2010) to better understand when collective rights are a viable resource management option.
The strengths and weaknesses of individual rights are also contested. Most formalization processes around property rights (the transfer of rights from the state or collective to an individual or legal individual) involves a consolidation of all bundles of rights in a single owner (Meinzen-Dick and Mwangi 2008). The benefits of individual property rights are such that security will provide increased incentives for individuals to invest, allow use of titles as collateral, and facilitate market development (Meinzen-Dick and Mwangi 2008). Where land grabbing is an issue, this may give households a measure of assurance. Individual rights may also improve productivity through easing access to credit, since land is an excellent form of collateral when borrowing money from the bank (Markussen 2008). In socialist countries undergoing rapid transitions, land titling has led to increased food security, although crop variation may be an issue, depending upon state policies (Markussen et al. 2011). Improved transferability rights of land have also led to a significant increase in activity in both rental and sales markets. Such transactions have, on average, ensured that most farmers have ownership of small amounts of farm land (Meinzen-Dick and Mwangi 2008).
Evidence as to the benefits of private rights to land and natural resources is mixed. For example, experience from the past decades of land titling processes shows that a single policy to achieve an exclusive outcome does not always work: restrictive land ownership and transfer policies have been shown to limit poorer people's opportunities (Meinzen-Dick and Mwangi 2008). Poverty typically involves a number of capability and entitlement deprivations that limit freedoms (Jentoft et al. 2010), and limited access to land that can result from exclusionary land titling is a basic deprivation. Another issue facing many commons users is a process of creeping enclosure which results from incremental and cumulative privatization events rather than a single regulatory moment (Murray et al. 2010). In other words, as commons areas become increasingly enclosed or territorialized, there are less common areas available for people to access (to fish, to collect non-forest timber products, to graze animals). Enclosure stems from multiple policies that restrict communal land access: protected areas, forest plantations or individual transfer quotas in the fisheries. Evidence shows such arrangements can achieve some form of economic efficiency and business flexibility, but also result in a loss of flexibility for resource users who depend upon shifting among resources (fish, wildlife, forest products) or access to transboundary resources. These unintended consequences are particularly painful for poorer households (Murray et al. 2010).
In Vietnam, tensions around individual and collective rights are a source of uncertainty for a wide range of rural actors (state and non-state). Gaining access to forestland and forest products, as an example, can be challenging for villagers, since the Vietnamese state owns and controls vast swaths of forestland (Sunderlin et al. 2008). While forest cover has increased as a result of large-scale afforestation efforts, this has not led to poverty alleviation in heavily forested upland areas, particularly for ethnic minority villagers (Sunderlin and Ba 2005). Households do receive land title for agricultural land in and around the village, however, they cannot officially access forestlands unless participating in forest management programs whereby small forest blocks are allocated to households in exchange for protecting these plots from outside encroachment (Sikor and Nyuyen 2007;Clement 2010). Therefore, many claims are made in the informal realm since forest products 2 fill income gaps and complement other forms of income (McElwee 2010;Turner Forthcoming). This can be risky, since villagers do not know when government officials might start enforcing official policies i.e. banning informal land and forest uses. Government policy is shifting to acknowledge these access challenges through recent reforms. As an example, the new Forestry Law (2005) recognizes the need to increase local access to forest resources, particularly in terms of enabling community forestry (a village-held collective right), and allocating private property rights to households in forest areas or former forest areas that were owned by the state.
Aquatic resources can also be challenging to access, albeit for different reasons. The pace of change along Vietnam's coasts, deltas and lagoons has been intense, and has led to an enclosure of common spaces for fishing as aquaculture has intensified and coastal development has expanded generally 3 . This expansion of aquaculture ponds has brought a series of new pressures into coastal areas including higher competition for land and the threat of marginalizing rural communities from access to productive land and water resources (Bush et al. 2009;Tuyen et al. 2010). Aquaculture farmers and capture fishers are drawing on the same resources, particularly since fish farmers have historically been dependent on wild caught fish to use as feed for their aquaculture farms (Bush et al. 2009). The natural resource base, particularly the inshore fisheries, has been severely overexploited, with many high value fish resources having significantly declined (Pomeroy et al. 2009;FAO 2010). Policy reform in this sector aims to better regulate fisheries and aquaculture growth. For example, the Ministry of Fisheries has created a series of master plans and issued decrees and ordinances aimed at the preservation of marine resources (Boonstra and Nyugen 2010). The new Fisheries Law (2003) also recognizes the role of community members in the management of their natural resources (Armitage et al. 2011). Provincial and District authorities have used the law to enact specific decisions enabling community forms of management and to formally transfer rights to village level associations.

Approach, data collection and analysis
To understand where, if and when collective rights for natural resource management make sense, sensitivity to contextual variation and a diversity of property rights bundles is required. The analysis of cases in central Vietnam adapts a generic framework (c.f., Schlager and Ostrom 1992) in which rights categories (public, private, common), rights holders (state, individuals, corporations, groups), and rights bundles are emphasized (see Figure 1). Rights bundles overlap but include: 1) use rights relating to access, withdrawal and exploitation; 2) control or decisionmaking rights relating to management and exclusion; and 3) rights of 'alienation', such as those related to the right to rent (in whole or part), sell or transfer rights to others (Barry and Meinzen-Dick 2008;Meinzen-Dick and Mwangi 2008). The framework was used to organize data collection and analysis in three case study sites in central Vietnam to address the central question of when collective rights make sense, and to 'read' property rights regimes as they present themselves in complex social and ecological settings.
No framework can address the confounding diversity of property rights regimes. However, a systematic examination of the holders of property rights and bundles of rights is helpful in mapping the reality facing resource users and policymakers. We use this framework as a descriptive tool, with which we are better able to examine the complexity of individual and collective rights. To facilitate our analysis in these sites, we examine policy interventions and associated rights arrangements with regard to their fit, legitimacy and enforceability (Ostrom et al. 2002). Fit refers to the contextual appropriateness (social, economic, ecological) of the rights arrangement from the perspective of individual producers. This includes the fit of rules, rights and institutions with a) the social, cultural and economic practices of resource users, and b) the ecological characteristics of the resource system. For instance, the stationarity of forest resources or aquaculture outputs posses a different set of challenges than those associated with transboundary fish stocks. In both instances, the seasonality of access to resources (e.g. spawning periods, availability of non-timber forest products) further adds to the challenge of fit. Legitimacy involves both a material and subjective dimension, and thus, refers to the extent to which individuals (state and non-state) accept the rights arrangements as well as the actual codification or informal recognition of rights. Enforceability involves the pragmatic need for rights holders to protect their interests from others (non-local resource users) through monitoring and patrolling, and includes the ability of individual and community members themselves to enforce their rights, as well as the important role of formal authorities in securing rights arrangements. A multi-method and comparative case study approach was utilized in the context of a team project, "Governance and Management of Common Pool Resources in Vietnam 4 ". The core objective of this project was to identify when collective rights make sense to promote natural resource management and livelihood enhancement. We chose three specific rights allocation cases within Thua-Thien Hue province, central Vietnam to illustrate how collective rights may serve as a trigger for enhanced resource management and to consider the implications of such rights for resource governance ( Figure 2): (1) community forestry across the upland forests; (2) state forest re-allocation vis-à-vis temporary use rights to a group of households in the upland forests; and (3) territorial use rights for fishers and aquaculturalists in Tam Giang Lagoon. The case studies were chosen to highlight different resource systems, although each case is situated in a similar context of shifting policies emphasizing decentralization and differing forms of rights allocation. The three empirical cases are illustrative of interactions between private and collective rights to land and natural resources, and the tradeoffs inherent with different mixes of rights.
Case specific research questions and data collection were tailored for each site and examined how the different rights in the bundle shifted between various rights holders, and then addressed core issues of fit, enforceability and legitimacy. Data collection included semi-structured interviews, surveys (approx. 15% of households within six communes surveyed across the uplands and the lagoon), and participant observation ( Table 1). The research builds on findings and data from a series of investigations in the case study communities, other secondary literature and a review of specific resource governance arrangements (c.f., Tuyen et al. 2010aTuyen et al. ,b, 2011Armitage et al. 2011;Huynh et al. 2011;Ngo et al. 2011;Nguyen et al. 2011). The research also builds on more than a decade of experience within these areas, and during the more recent project period, extensive interaction. Results of the case analysis are provided in the following section.

Case one: community forestry
Community forestry in Vietnam emerged as a policy response to access challenges, and is meant to enable villagers to incorporate aspects of traditional rights and management systems, particularly in ethnic minority villages (Sunderlin and Ba 2005). This is an example of a collective management right that is allocated to the village level, enabling villages or groups of villages to enter into partnership with the State for the management of forest resources. This policy innovation is a departure from highly centralized forest policies that have characterized forestry management in Vietnam -a series of community forestry sites have now been established throughout Vietnam (Sunderlin and Ba 2005). A significant effort has gone into community forestry in the past number of years, with a number of international organizations supporting the designation and establishment of these sites (c.f. Ngo et al. 2011 for a review of international organizations working in uplands of Hue 5 ). By the end of 2007, over 10,000 villages were managing over 2,700,000 ha of forest and bareland area, of which nearly 2,000,000 ha were community forests (Ngai 2009).
The quality of such forest land allocation has been questioned (c.f., Sikor and Nuygen 2007;Ngo et al. 2011). Our first case draws insights from several community forestry sites within A. Loui District in central Vietnam. People here are reliant upon non-timber forest products (NTFPs), practice swidden and low land agriculture and may fish, do logging, work in a sawmill, or run small businesses (CPR Uplands Survey 2009; see also Volker and Waibel 2010). Forest-related income is important because households can only access small amounts of agricultural land (a maximum of 2 ha), either through a formal land title process, approval by commune officials, or illegal claiming. Households also aspire to plant, as an additional source of income, forest crops such as acacia trees that can be harvested after five to seven years. For these reasons, many claims to additional agricultural land, trees or swidden agricultural plots are made in the informal realm. While some activities are sanctioned (fishing in rivers, collecting NTFPs, firewood collection) by lowerlevel government units (the commune or district), other activities are prohibited and can be risky to do (logging, illegal saw mills, planting acacia in areas without land title). Around 6283 ha of natural forest were allocated for community forestry across 17 villages within this district between 2005 and 2009 (Table 2). While a small amount of this forest area was designated for individual households to manage and protect specific forest blocks, the majority of forests (nearly 97%) were designated for community forestry.
In this uplands context, community forestry agreements provide villagers access rights to a designated forest area, withdrawal rights to NTFPs (but not hunting or most logging activities), management rights (small amounts of land can be turned into agricultural or pasture lands, management may include tree thinning, pruning or patrolling), and exclusion rights (for those not following community forestry regulations or not belonging to the community forestry group). Alienation (selling) is not allowed within community forestry areas. Community forests designated to villages in this area are around 100 ha, with most forests being of medium to poor quality (in terms of forest density and diversity) (Duc et al. 2011). In reality, with or without the community forestry designation, villagers mainly use forests that are within a few hours walking distance from their homes to access NTFPs, cut small amounts of wood for home construction, to graze buffalo or for hunting. In cases where villagers go further into the forest, for logging or a specific type of NTFP collection, this involves multi-day, over-night trips that not all villagers are willing to do. Small amounts of individual land claims may also be found, for agricultural activities, within forest areas. Thus, while people do not officially have access to much forest, how they use forest areas depends somewhat on how carefully the government is monitoring activities and if villagers can get away with certain types of practices (i.e. clearing forest areas to plant acacia trees). Community forestry advocates argue that villagers need formal rights to forest areas (i.e. withdrawal, management, exclusion), since it is not clear how the government might use such land in the future (see Figure 1).
Despite this policy focus and the forest access needs of upland communities, the uptake of this particular collective right has been slow. In A Loui District, evidence from key informant interviews and focus groups point to a series of reasons for this situation. First, despite the relative stationarity of forest resources, the bundles of rights associated with the community forest agreements do not fit the prevailing economic and livelihood aspirations of the majority of individuals. Community forestry imposes an increased management burden upon villagers who are often interested in the potential to enhance their individual livelihoods through forest-based or agricultural activities within community forest areas. Generational differences also factor into individual decisions to participate in a collective enterprise. For example, in one case youth admitted that they had been hopeful that after the community forestry site was allocated, the area could then be allocated to individual households which would enable people to plant acacia or other crops. Meanwhile, elders reflected that they had previously managed NTFPs and forest areas through agreements between user groups whereby individual households all agreed to uses within a particular area (e.g. grazing land, resin trees, protecting specific tree species). Their interest in community forestry was to find ways of having individual management, as in the past, within a collectively managed area. Thus, in many ways the 'collective' interest to forest areas in these upland sites in Vietnam relates to individual aspirations (Table 3). Issues of legitimacy and enforceability were also highlighted by villagers. For instance, an obvious constraint to the utility of this particular collective rights arrangement has been the distance of the community forestry allocations from villages (an average of 1 km, but further in some cases) and their location on steep slopes, creating a disincentive and relatively high transaction costs for community members to monitor and patrol the forest. In several sites, villagers assumed that the forest quality was poor (Ngo et al. 2011). While this is often the case (c.f., Sunderlin et al. 2008), participatory forest assessments showed that in a few cases forest quality was far richer and diverse than expected. This suggests that people have not been adequately involved in site selection of community forests, and further highlights distance, access issues (i.e. limited walking paths into a dense forest area), and how an active role in managing the community forest is not seen as worthwhile. Moreover, community forest areas have not been clearly demarcated, and villagers feel that they cannot exclude outsiders from using their community forest area since this might create tensions with other villages. A few villagers have made individual claims in community forestry areas, cutting the natural forest and planting acacia trees. This reflects a problem with perceptions about the legitimacy of this particular collective rights arrangement. In one community forestry site, 20 ha have been claimed by two individuals living in the village in order to plant acacia trees: the total community forestry area is only 50.5 ha so this is a significant land claim (Ngo et al. 2011). This has led to tensions between villagers with regard to the utility of this particular collective rights arrangements, and frustration that the rules of the game are not enforced by the community forestry committee. District officials have not been interested to get involved in solving these village-level tensions. This case is informative since it helps us to consider the role of a collective rights allocation within an uplands context. The balance of benefits and responsibilities are not right in this case: government officials would like to off-load management responsibility to lower levels; villagers argue that it is impossible to patrol and exclude outsiders from remote forest areas. While we know that people draw on NTFPs, firewood and other forest products (c.f., Uplands survey 2009; McElwee 2010), this type of rights allocation will require some modification if it is to be effective, including local champions to advocate for this type of allocation, and strategies to plant and harvest rattan, bamboo or acacia in ways that benefit members of a community forest. On the surface, individual claims to forest land would appear a better fit for people in the uplands.

Case two: state forest re-allocation households
In central Vietnam, community forestry is not the only property rights reallocation model to enable household's greater access to forest and agricultural lands. Our second case examines the planned re-allocation of state forest land to individual households, the land claiming that ensued because of lack of follow through by state authorities, and the resulting institutional uncertainty. Within A Loui District the state-run forestry operation -the Forest Protection Management Board (FPMB) 6 -is in the process of transferring over 5825 ha (of 33,649 ha) to the district (this takes time since the pine has to be first harvested and sold by the FPMB). Once land has been cleared, the idea is for district authorities to then have individual communes measure and allocate up to 1.5 ha of land for poor households or nearly-poor households that lack productive land (Huynh et al. 2011). The transfer of land from the FPMB via the district and commune to individual households is expected to help households secure their land rights, and address rural poverty and environmental protection goals.
We examine this re-allocation process with specific reference to experiences within Phu Thuong village in Phu Vinh 7 commune. Within this commune, over 6 After the end of the Vietnam War, 1975, the government established "State Enterprises" to manage forest and forest lands. After forming as a State Enterprise in the late 1970s, pine trees were replanted in this area since A. Luoi District was subjected to defoliation campaigns during the war. State Forest Enterprises became Forest Protection Management Boards in the late 1990s. 7 Phu Vinh commune consists of four villages, three villages are lowland Vietnamese (Kinh) and one is comprised of an ethnic minority group, the Pa Co. Throughout the commune, 60% of households self-identify as poor (Uplands survey 2009). Livelihood activities are agricultural and resource based, with people planting wet rice and cassava, raising animals (including fish, in a few cases) and relying on NTFPs from the nearby forests. It appears that minority villagers (the Pa Co) practice shifting cultivation (a total of 10,471 m 2 / household or home gardening rather than other forms of crop cultivation (Uplands survey 2009). Other jobs, mainly available to lowland Vietnamese residents, include seasonal migration, selling one's labour, or operating a small home business. Pa Co households tend to have limited opportunities, often relying on a small monthly government stipend to help make ends meet. Acacia is something people are interested to plant. 70% of land is classified as forest land, and shifting cultivation does take place in forest areas (Uplands survey 2009). Only 2.5% of land is designated for agriculture and the rest being designated as non-agricultural land or unused land. Forest lands in this context are not available for access or use by local residents, other than for seasonally-differentiated NTFP collection or for harvesting small amounts of wood for cooking or home construction. Even the land that has been categorized as unused (257 ha in this case) has already been claimed (for at least 5 years, but often longer), by households that have planted crops or acacia trees (Huynh 2009) (see Table 4). This overall lack of land to expand agricultural activities or to plant tree crops is problematic for people in terms of expanding their livelihood base.
In early 2008, 44 ha (of a total of 591 ha designated for transfer to this commune) of pine forest were harvested within the commune. The transfer occurred near Phu Thuong village which consists of 96 Pa Co households 8 . As the pine forest plantation area was cleared by the protection forest management board, 'word' also came down that local households would be granted small amounts of land (between 1 and 1.5 ha). However, government officials did not immediately follow through on this process. Thus people heard that they may be granted rights to land, in the form of private property (see Figure 1), but nothing came of this. As such, some households began staking claims to this land within the 44 ha harvested area.
In mid-2008 the 44 ha of cleared pine forest became dotted with evidence of new economic activity. Brush had been cleared, small burns could be seen, acacia was freshly planted, a few fish ponds had been constructed, and simple living structures had been erected. Walking paths could be seen throughout the area.  planted acacia (31 out of 34 households), in part because the acacia market has emerged in the past 5 years. Acacia grows relatively quickly which is why acacia is a convenient way to stake a land claim. The remaining three households (of the 34) have planted either rubber, cinnamon or rice. Twelve of the 34 households have intercropped acacia with cassava, pineapple, tea, jackfruit, bamboo or banana (Ngu et al. 2011). This is not the only area where villagers have claimed land, as other areas of pine forest plantation have also been harvested in this district.
After several years of individual land claiming, a more collective bargaining process emerged in which district and local (commune, village) interests interact. District authorities became interested in designing a process that could ensure that poorer households are allocated land in an equitable, rather than ad hoc manner, as was the case in Phu Thuong village. Using Phu Thuong village as the case model, district, commune and village authorities, along with villagers, met multiple times in early 2011 to negotiate a process that could acknowledge the existing claims to the 44 ha and, at the same time, develop a collective process that could be implemented in other areas before land was claimed. After negotiation and discussions on management, those households that had claimed land (34 of the 96 households in this village) were granted temporary use rights from the District until the end of 2015. Table 5 highlights respondent reflections on this rights process.
This case is informative for several reasons. The case shows that in the absence of a clear transfer process from state to individual rights, problems of fit and legitimacy emerge very quickly, even where the resources of interest are stationary. In this case it would seem that a stronger allocation mechanism would in fact enable the transfer of rights to those in most need, as is the policy intent, and that it would also improve the process by which the right transfer would be legitimized. In the end, a hybrid arrangement emerged that reveals how in this context, individual rights can work, when embedded in a process that also supports a collective good. For example, the households I took the risk as I hope that I will at least get a portion of the acacia tree harvest even if I do not have any future rights to this land. Villager C Because the transfer process was not clear, we needed to stabilize the process and think about production. We hope the state will continue to work with us and ensure that if any household exceeds 1.5 ha as per our new regulations that the extra areas will be reallocated to other households to create a fair situation.

Commune official
We could not stop the land claiming as we were waiting for the district official.

District official
We are really happy with this process -we talked about rights with the commune and villagers and decided on appropriate rules. We really had not been sure what to do in terms of the land claiming, and hope that we can apply this process to other places. will maintain their temporary access, withdrawal and management rights in the former pine plantation (44 ha), thus enabling households to harvest their acacia trees. In an effort to move forward, however, when it comes time to harvest the acacia trees, individual households will retain at least 70% of the profit from the harvested trees, will voluntarily contribute 20% of their profit into a collective fund which will pay for collective work (e.g. protecting acacia trees, infrastructure development such as roads, and other costs related to harvesting), and will provide up to 10% of their profits as a management fee to the commune authority (i.e. for tree licenses, biomass management, general environmental management, and management of the acacia harvest). In the meantime, these 34 households have the right to exclude other villagers from harvesting their trees. These temporary rights cannot, however, be transferred. After 2015 the idea is for these individual plots to be re-allocated by the commune (post-acacia harvest) in a process that will consider the entire village and the land use needs of all households. This is intended to ensure even greater legitimacy and can be adapted to the situation at that point in time. As this process unfolds, enforcement by commune and district officials will be necessary or there will be a risk that further conflict will ensue.

Case three: rights allocation to fisheries association in Tam Giang lagoon
The Tam Giang Lagoon is the largest brackish-water lagoon in Southeast Asia, covering 22,000 ha and spanning 70 km of Hue's coastline 9 . Lagoon physiography makes it ideal for capture fishing and aquaculture activities. Property rights are complex because of the sheer number of fishing gears people use (approx. 32 different types, according to Mein 2006), the class of fishing gear people use (mobile and fixed), the mix of aquaculture practices and how people move between multiple fishing and aquaculture activities, and because the ecological system itself is spatially and temporally dynamic. People use stake traps, bottom nets, gill nets, and practice various forms of aquaculture (net enclosure, earth ponds; single and 9 Around 300,000 people, representing one third of the provincial population, live in the 33 municipalities surrounding the lagoon, with an estimated 100,000 people depending directly on smallscale capture fisheries and small-producer aquaculture and another 200,000 people depending on a range of related livelihood activities including coastal agriculture and occasional fishing or fishingrelated activities (Tuyen et al. 2010 mixed species) 10 . Mobile gear fishing continues to function, more or less, as an open access or 'first come first serve' activity. Fixed gear fishing gear began as an open access activity that, over time, has led to de facto titles or permits (stake traps are an example of fixed fishing gear with de facto rights). There are also land use titles granted for 20-year periods from the district level, mainly for aquaculture earth ponds near the village, plus permits granted for 5-year periods for some aquaculture earth ponds at the edge of the lagoon, and 1-year commune permits/ registration for fish corrals, converted net enclosures and a few earth ponds (Ta 2010). Given this mix of formal and informal use rights, multiple resource users are competing for the same space. Those households that have start-up capital and access to land are able to do aquaculture, whereas those that do not have start-up capital or access to land for ponds generally draw on forms of mobile fishing gear (gill nets and, more recently, bottom steel-frame nets). The lagoon is too densely used, and provincial government staff are too far removed to adequately monitor and regulate this situation. This lagoon is an ecosystem in transition -capture fisheries are at near collapse and no fisher is particularly optimistic about their future (Armitage et al. 2011). While villagers have been discussing declines of capture fisheries in the past few years (34% of fishers observe that fish species diversity has decreased a lot since 1999; 64% of fishers suggest that there has been a significant decrease in the total species caught since 1999), in 2011 fishers reported only catching very small species of fish and shrimp. Aquaculture as an effective alternative to capture fisheries is risky. Disease was already reported in the early 2000s, especially in semiintensive shrimp culture. By 2006 people began switching to mixed species and extensive shrimp culture in an effort to deal with disease outbreaks. In this context of increased resource pressure and conflict among fixed and mobile gear fishers, efforts to reconcile and (re)allocate rights depends on a careful reading of what is a decidedly complex and dynamic property rights setting.
A solution to such complexity is to allocate a collective spatial or territorial use rights for fishing 11 and aquaculture activities to a legally constituted social 10 Aquaculture yields the highest total benefit to local people, followed by capture fisheries, agriculture production and sea-grass production (Tuan et al. 2009). Yet, when examining total net benefit, capture fisheries has historically had the highest production since it is less risky than aquaculture and has lower production costs compared to both aquaculture and farming (Tuan et al. 2009). This may be changing, as fishers report being no longer able to catch crabs (2011) and particularly efficient mobile gear (steel net traps) was introduced into the lagoon in 2006 with negative ecological consequences. In 2009 fishers felt that a mix of fishing and fish farming were important for the future, with fish corals (20%) being seen as the most important activity in the future, followed by earth pond aquaculture (15%) and net enclosure (14%). However, people also talked about the importance of rice culture in the future (16%), followed by being a business person (5%), doing animal husbandry (3%), being a government worker (3%) and being a labourer (2%). 11 This process is often referred to as community fisheries within the Southeast Asian region (c.f., Pomeroy et al. 2008), although it is more often than not a co-management process with some form of state-village partnership that is promoted. For more details on TURFs in other regions, it is worth examining the experiences from Chile (c.f. Gelcich et al. 2010;Gallardo et al. 2011). organization (i.e. a fishing association unit) at the village or commune level. User groups are allowed to form socio-professional organizations (in accordance with the 2008 Law on Social Organization and Association, updated in 2010). In the Tam Giang lagoon, a network of Fishing Associations have emerged that are recognized by higher level authorities as having an important role to play in managing aquatic resources in the lagoon. There are now 50 Fishing Associations established at the village or sub-commune level with over 4500 members across the lagoon. For individual Fishing Associations to be granted formal fishing 'rights' over the lagoon, 75% of a village needs to belong to the fishing association and they need to show the District government their plans for fisheries management.
In 2009, the first ever (in Vietnam, to the best of our knowledge) territorial use right for fisheries (TURF) allocation of 967 ha was allocated to the Vinh Giang Fishing Association 12 . What is innovative in this case is that Fisheries Associations are given rights to fisheries and aquaculture activities within a designated area, and this right is allocated to a village organization rather than the village council in general. The TURF includes sub-zoning of lagoon surface area for different functions or purposes. Specific modifications in fishing activity are required as part of the collective rights allocation (access, withdrawal, management, see Figure 1), including a reduction in the number of stake trap units, restrictions on certain gear types (electric fishing), improved protection of breeding areas and seaweed beds, and appropriate aquaculture intensity by introducing mix-species culture techniques. In Vinh Giang, for example, the Fishing Association was able to reduce the number of stake traps from 89 to 56. Since this initial allocation, another six rights allocation have been granted to six Fishing Associations around the lagoon 13 . This rights allocation has made the difference of people using the lagoon in an ad hoc manner in comparison to people using the lagoon in a more managed manner. 12 In May 2009, 978 ha were designated for management from the district to the fishing association of Vinh Giang commune. The management plan for the TURF was developed by the fishery association (with facilitation support by university researchers), and includes the zoning of lagoon surface areas into different functional units with different purposes (Armitage et al. 2011). Zoning serves as a foundation for the development of regulations and the organization of self-managed groups aligned to specific resource use activities. Thus, the lagoon space in Vinh Giang has been divided into different sub-zones, including those for fixed gear fishing, waterway combined with mobile gear fishing, traffic areas, breeding areas and seasonally protected zones. Zoning is based on local experiences and knowledge of lagoon resources and on the current uses of the lagoon, but is consistent with Provincial and District plans for fishery exploitation. Key changes in fishing activity as required by the TURF include a decline in the number of fish corral units, restrictions on certain gear types (steel net traps), improved protection of breeding areas and seaweed beds, and introducing semi-intensive, mix-species aquaculture techniques. 13 Two of these rights allocations were facilitated by the project research team: 187 ha in Phu My in 2010 (main activities are fish corral and net enclosure) and 242 ha to another fisheries association in 2011 (main activities are fish corral, dai, gill nets and steel net stake traps). This latter allocation was tricky in that boundaries had to be agreed between two districts.
The collective rights allocation has emerged as one way to deal with the unplanned expansion of lagoon activities, in and around a village, to enable a detailed plan to control fishery exploitation and, at the same time, to maintain household livelihoods. In short, the rights allocation aims for a good fit with current conditions and local context because it matches the spatial scale of activity with the appropriate jurisdictional level. Fishing Associations not only manage the capture fisheries area, they are also responsible for household-based aquaculture activities in their community-designated area. The rights allocation thus provides a legitimate mechanism from the perspective of community members and officials to ensure access and withdrawal rights for Fishing Association members. Moreover, the collective rights allocation contains a significant focus on management, including the monitoring and patrolling functions that enable enforceability of the right. While Fishing Associations cannot control for all illegal activities, they can increase patrolling to avoid the theft of fish or prevent electric fishing, implement consistent stocking regimes in an area, and designate areas for protection or for seaweed production which can provide a form of income to support committee work (Armitage et al. 2011). See Table 6 for selected perspectives on the allocation of fishing rights.
This particular collective rights allocation reflects fit, legitimacy and enforceability in several additional ways. For instance, people move between forms of fishing and aquaculture (Tuyen et al. 2010), and property rights are overlapping and often contested. The collective rights designation, therefore, serves to give village level actors greater control over decision-making. At the same time, the collective right provides individual net enclosure and fish corral owners (forms of aquaculture) with greater individual security over use rights. The collective right also offers a mechanism to secure local interests against outside business interests or migrants coming into the area in search of resources that may shift through The rights allocation helps us to stabilize our capture fishing and aquaculture so as to stabilize our livelihoods. It helps improve the fishery production with balancing capture fishing, aquaculture and resource conservation. Villager C The commune now asks our opinion about managing lagoon resources.

Commune official
We do not have the time or the budget to always think about fishing or aquaculture in the lagoon, so it makes sense for village groups to focus more on this. District official This is a far more effective way to manage the lagoon as the village groups can work with us. time and space. The fishing right allocation does not prevent users who are nonfishing Association members from having access to waterways, but they do need to follow the management regulations. Longer-term, however, it is not clear if this collective rights allocation will contribute to poverty alleviation, particularly if fisheries declines continue. Whether Fishing Associations can mitigate or dampen the effects of severe stock declines or disease outbreaks in aquaculture ponds with their monitoring and management techniques is also an open question. This is an ecosystem in transition, and this collective right makes sense at least for the short-term.

Discussion and conclusion
In Vietnam, as elsewhere, individual and collective rights can be formal or informal (unwritten or codified). As a result, de facto rights that govern the day to day use of natural resources are often different from de jure rights that exist in formal legal documents (Mascia and Claus 2009). Actors may derive benefits from resources without holding property rights to them, such as deriving benefits from an agricultural field by way of occupation or market exchange even though they do not have any formal property rights (Ribot and Peluso 2003). In practice, the contextual diversity and variation in bundles of rights make property rights policy a challenge, leading to oversimplifications of appropriate strategies, or unintended consequences associated with implementation of natural resource management projects (Tanner 2007). It is important to understand everyday land and resource use practices in combination with formal policies; it is also worth recognizing the incentive structures for both individual and collective rights particularly from the perspective of local users who may need a mix of rights to secure and sustain their livelihoods. As our cases show, formal and informal rights form a continuum. Within such formal and informal rights, farmer and fisher agency through various (individual) claims and government responses to them (condoning, tolerating or actively supporting) play a key role in negotiating how mixtures of overlapping rights work in practice. The three cases examined here illustrate how efforts to resolve the challenges of rural poverty and environmental protection involve complex and dynamic property rights and linked resource-based livelihood activities. Support to existing collective rights is one strategy to address rural livelihood challenges, as are efforts to implement new collective rights arrangements. Potentially useful policy responses to the challenges of rural poverty have emerged in Vietnam, including our three cases of community forestry, the re-allocation of state land to individual households, and territorial use rights for fisheries.
Each of our cases represents different social and ecological situations, demonstrating the need for flexible governance arrangements beyond discrete property regimes (c.f., German and Keeler 2010). In the case of the state forest land re-allocation, this is an example of the individualization of forest land for conversion into agriculture and short-cycle agro-forestry such as acacia. The Do collective property rights make sense? Insights from central Vietnam 21 former pine plantation areas are no longer seen as forests: this is accepted by all parties involved, including the government. This is in stark contrast to our community forestry case where the starting point is that forest integrity needs to be maintained. Forest conservation is an important component of community forestry policy; this links to the lagoon rights allocation case in that conservation (or preventing further degradation or exploitation of the lagoon) is also an overriding concern. The title of our paper asks "Do collective rights make sense", and what we show is that there are no collective rights arrangements in a 'pure' sense, but rather a series of arrangements that, when working, foster collaborative action and strengthen resource conservation. This is taking place in a context where there is strong pressure for individualization. For these reasons, a mix of individual and collective rights is important for people who use natural resources in multiple ways, and move back and forth between spaces, ecosystems and administrative boundaries.
Whether collective rights will work in practice depends upon the extent to which they fit the context (e.g. social, economic, ecological), are considered legitimate, and are enforceable. As these three cases illustrate, policy uptake has not been consistent across these examples and getting the right fit is not so easy. It is one thing to have a policy; it is another thing to ensure such rights go from a blueprint approach to being contextually adapted. Most centrally, the three cases show how difficult it is to determine if collective rights make sense when individuals have a desire to increase their access to land in unambiguous ways. At a minimum there needs to be sensitivity to variation within collective rights arrangements, recognizing the variety of interests that people may have in any rights transfer process (Table 7).
As our three cases each demonstrate, the benefit of collective rights has to be such that there is a perceived collective and individual benefit. Only at this point is there a chance that the right will be enforced, whether through peer pressure, formal patrols or outside support. For instance, a major challenge with community forestry allocation appears to be getting the right fit: the balance between benefits and responsibilities is off, and villagers are not able to enforce their rights. There is an additional dimension here: when we talk about collective rights we are talking as well about the need for collaboration or collective action.
The government is concerned about resource preservation for which collective action is needed (agreement and legitimization of rules in use, implementing such rules, etc.) whereas farmers are more interested in direct benefits, which they see in agriculture or agro-forestry on individual plots. Since farmers can informally access forest areas in and around their villages, the burden of patrolling and enforcement is not seen as worthwhile at this point in time. Perhaps community forestry is something that might make sense at a future point, when there are less forest areas in general, or when trees can be harvested in certain parts of a community forestry area. This type of collective right needs to match people's aspirations and the effort they are willing to put into the management of natural resources.
Re-allocating state forest areas to individual households did not unfold as authorities envisioned. The district was not able to intervene with a coherent process in a timely manner, leaving a window in which people could claim the cleared pine forest plantation land. The ensuing land grab speaks to people's need for more land and the belief that one way to legitimize a claim is to be active on the landscape -this is reflected by the fact that 34 households sought to claim land. Those that claimed the cleared land took a risk, as it was unclear how this process might unfold (although there is a history in this area of district officials ignoring some local claims). The livelihood activities in the cleared pine forest land have two sides: the 'utility' of actual land use (growing trees, raising fish) and the political act of staking a claim. This has paid off, since these households now have individual (although temporary) use rights to this land and much of the proceeds of acacia harvest until the end of 2015. The commune also derives a benefit from this process in the form of a tax on harvested trees. The government is probably often hard put to intervene when farmers who are really short of land open up forest (or, for that matter, in the lagoon when really poor people engage in destructive fishing activities). There seems to be a moral argument that people have the right to livelihoods (and thus resources) that makes it hard to intervene strongly.
For these reasons, the forest land re-allocation appears to be a situation whereby the temporary lease rights are enforceable, legitimate (in part because those who took a risk could claim land) and contextually adapted. What is less clear is who did not dare to claim land, and how one would negotiate this process in another area to ensure poorer households (i.e. not those with labor or connections) could gain access to land. Ultimately, this process will lead to individual rights. Yet, the case is revealing because it was through a collective process that these individual rights have been clarified. This process prevented a potentially high-conflict situation from emerging, and it likely reflects a better fit with the norms and aspirations of an upland Pa Co dominated village.
In the lagoon case, the TURF may be an appropriate rights allocation because use rights are over-lapping and complicated. Fishing Association members indicate that the rights allocation is an opportunity for officials with limited resources to share the burden of management, thus reducing their transaction costs with communities and potentially ensuring a better success rate around contentious issues (this is also the case with community forestry, although here it is more difficult to get past policy and into practice). Fishing Association members are also clear about the incentives they see embodied in the collective rights allocation. Central among these incentives is: 1) the opportunity to more directly self-manage lagoon resources within their allocated territory; 2) expectations about stabilizing environmental conditions in the lagoon; 3) a more proactive approach to solving technical problems (e.g. stocking density); 4) more effective patrolling and increased collective effort to patrol fishing areas and prevent the expansion of destructive fishing practices (e.g. electric fishing); and 5) a reduction in conflict among fixed and mobile gear fishing activities (Armitage et al. 2011). Certain rights are enforceable, it is legitimate since 75% of a village needs to belong to a Fishing Association for a territorial use right to be granted, and this appears to be adapted to the local context. Most centrally, however, peoples' interest in collective rights to a spatially defined area of the lagoon is high because it creates an opportunity to secure individual aquaculture rights. Several fishing association members suggested that as capture fisheries decline, aquaculture will continue to be an important livelihood activity. Thus, there is the guarantee of sustaining individual benefits within this collective rights gain.
Evidence in each of these three cases suggests that individual rights are paramount to households. Yet, it also appears that a sole focus on individual rights makes less sense where complexity is particularly high or where the process of re-allocation or transfer is not recognized as legitimate. Thus, there are important policy insights that can be derived from these three cases (Table 8). There is a need for collaborative action and agreed rules in a context of weak government capacity to ensure conservation and an overlapping context of government policy promoting the decentralization of rights. Thus, there is a role for collective rights, if adapted to fit a local context and resource reality. Resource rights need to ensure that people will get enough benefit to be involved in the management of a resource. The limited ability to harvest trees is an issue for community forestry designation and this is in stark contrast to the collective rights for fisheries or aquaculture that enable harvesting of aquatic products. The cases remind us of the complexity found on the ground, and the inherent difficulty of clarifying bundles of rights to resources that different users may have at different times. Ultimately, enforceability, fit and legitimacy will determine if and how collective rights make sense, and provide a filter with which to analyze and read a rights situation within complex social-ecological systems.

24
Melissa Marschke et al. The line between collective and individual rights is often blurred in practice Individuals can come together to manage an area; those with collective rights may also opt for individual rights within a collective rights setting as in the lagoon case; collective processes can be used to allocate individual rights in ways that fit the context as in the upland state forest reallocation case Collective rights make sense when they are: (i) designed to fit context; (ii) provide users with access to resources they perceive as valuable; and (iii) they create space to enhance individual security (either of tenure or in terms of resource access) The TURF allocation is a collective right that addresses fishing, aquaculture, conservation and economic development activities simultaneously; the TURF balances collective and individual aspirations Achieving 'fit' and legitimacy takes time; Collective, individual and hybrid arrangements are likely to emerge through trial and error requiring flexibility and learning from different groups Property rights in many (most) situations are complex and are emergent; a tendency of outsiders (policy-makers, researchers) to understand these arrangements 'too quickly' (sensu Tanner 2007) constrains opportunity for successful policy design and uptake Community forestry policy is innovative, but it is difficult to get policy uptake (by officials and villagers) in this area, reflecting a need to adjust the current approach In the lagoon example, rights allocation has emerged in the context of a co-management process that has evolved over a decade or more, and has slowly built trust among various groups