Curse or blessing ? Local elites in Joint Forest Management in India ’ s Shiwaliks

This article suggests that local elites play an instrumental role – either with positive or negative consequences – in shaping struggles for power over processes and outcomes of participatory forest management interventions, when implemented in communities characterized by social hierarchies. We show how the contrasting outcomes of joint forest management in two case study villages cannot be attributed to institutional reform, but appear to be caused largely by differences in the role assumed by local elites. The evidence indicates that institutional reform itself does not guarantee changes in the actual management of natural resources. Rather, vested interests at the local level and among State actors may continue to shape events while working within or beyond the new institutional landscape. On the basis of the results of our case studies, the article poses the hypothesis that a network theory of social capital could be a useful way of analyzing such diverse outcomes of similar institutional reforms implemented in relatively similar communities. We conclude by arguing that attempts at institutional reform at the level of the community in hierarchical societies should proceed with modest expectations, 708 Kulbhushan Balooni et al. and an eye for the incentives facing local elites and the implementing and facilitating State actors.


Introduction
The increased participation of local communities in managing the forests they live in the vicinity of and use represents a profound change in forest management policy over the last 30 years.The analysis of such approaches, called community forestry, community-based forest management, joint forest management etc. has elicited an enormous body of literature in the developing world.This literature indicates that, despite the changes in policy and rhetoric, only scattered and modest successes have appeared on the ground (e.g.Campbell et al. 2001;Blaikie 2006;Ribot et al. 2006).This has led not only to allegations of foot-dragging against donors and national policy makers and implementers, but also to calls for assessments that can better explain local-level outcomes (e.g.Agrawal 2001;Blaikie 2006).Recent studies indicate that such local-level outcomes are shaped by a plethora of factors.Studies within new institutional economics have focused on how characteristics of the forest resources system, the users, and institutional arrangements -for example, forest size, forest utilization pattern, group size and heterogeneity, and trust among the group -may explain why some settings are more conducive to local collective management than others (e.g.Agrawal 2007).Another body of recent studies focusing on the relations between communities and various State actors indicate that active support or, as a minimum, acceptance by the latter of attempts by local communities to establish authority and legitimacy as natural resources managers, is essential given the vast opportunities for powerful higher-level State actors to quench or nurture fragile local initiatives (Ribot 2004;Benjamin 2008;Lund and Treue 2008).These and other recent studies illustrate how processes of local participation in forest management can be construed as a continuous power struggle that unfolds in numerous ways and forums between external actors and internal community fractions over the processes and outcomes of forest management (Nightingale 2002;Pérez-Cirera and Lovett 2006).In relation to elites, there has been a tendency for studies on local participation in forest management to focus mainly on the curse of elites -the risk of elite capture (Oyono 2004;Iversen et al. 2006;Lund and Treue 2008;Dhakal and Masuda 2009).This article represents an attempt to contribute to the existing literature on how better to explain local-level outcomes.We do this by focusing explicitly on the role of power struggles among actors, in particular local elites, over the processes and outcomes of joint forest management (JFM).We present an analysis of the role of State actors, NGOs and, in particular, local elites in JFM, and how these actors may work for better or worse.The paper is based on an empirical study of a JFM program implemented in the Shiwalik hills in the state of Haryana in India.
JFM in the Shiwalik hills was one of the earliest such initiatives in India, the success of which led to the implementation of JFM programs throughout the country.It was initiated in the 1970s with the creation of Hill Resource Management Societies (HRMSs), as decentralized village institutions constituted to protect and manage forests with the participation of villagers living in and around them.The HRMSs were originally created as part of a government led integrated watershed development program that focused on arresting erosion of the hill slopes (Samra et al. 2002).Hence, the HRMS institutions were born with a dual focus on both forest and water management to rejuvenate the degraded hills.Improved management of natural forest and pasture lands, tree planting, and techniques to conserve soil and water resources were all integral parts of the work of the HRMSs (Chopra et al. 1990).Over the years, HRMSs were deemed a successful initiative and were implemented widely by the Forest Department over the Shiwalik hills (Arya and Samra 1995).Support for this process came from the Tata Energy Research Institute, now The Energy and Resources Institute (TERI).With financial backing from the Ford Foundation, India, TERI acted as a community facilitator and as a community-state liaison.As of today, TERI continues in this role as a mediator between the communities and the Forest Department.
JFM initiatives in the Shiwaliks are exhaustively documented as pioneering and successful JFM experiments (e.g.Mishra et al. 1980;Mittal et al. 1986;Chopra et al. 1990;Sarin 1996;Singh and Varalakshmi 1998).Initial studies focused on the development and participatory processes (Mishra et al. 1980;Mittal et al. 1986), and on analyzing the outcome of water and forest management (Chopra et al. 1990;Dhar 1994).Subsequent studies have focused on issues of sustainability (Gulati and Sharma 1998;Agarwal and Narain 2000) and on benefit-sharing (Sengupta et al. 2003).Finally, a number of action research studies have been conducted on the Shiwalik JFM processes that have contributed to the execution of programs and government policy (Dhar et al. 1990;Sarin 1996;Samra et al. 2002).These studies were undertaken mainly by people involved directly in the conceptualization, implementation, monitoring, and evaluation of JFM in the Shiwaliks, and generally analyze a mix of socioeconomic, institutional, legal, and technical issues in forest and water management by HRMSs.While this impressive body of studies provides rich and varied information about the process of implementing the JFM program, limited attention has been given to analyzing the role of State and community actors in the struggle for power over the process and outcomes of JFM.We feel that revisiting this first-of-its-kind decentralization initiative with this particular objective in mind may offer important lessons from both academic and policy viewpoints.
The paper proceeds as follows.Section 2 outlines the methods applied in the empirical study.The three following sections present the results of the empirical study.Section 3 presents background information about the two case study villages, 'Sahyogpur' and 'Sangharshpur'.Section 4 focuses on the processes and outcomes of JFM in Sahyogpur, whereas Section 5 does the same for Sangharshpur.Section 6 provides a brief discussion and poses a hypothesis for future research.Finally, Section 7 concludes.

Research method and data
The focus of this article is on understanding how the struggles for power over JFM processes and outcomes by State and community actors affect JFM outcomes.To achieve this, we report on a case study of two villages that we will call Sahyogpur (in Hindi 'Sahyog' means collaboration, and 'pur' means habitation) and Sangharshpur ('Sangharsh' means competition or conflict).The true names of the two communities in the Yamunanagar Forest Division in the state of Haryana are withheld.These two adjoining communities were selected purposively, based on the background knowledge of the area of one of the co-authors, Chetan Kumar.We chose Sahyogpur to represent a village with a high degree of internal cooperation and good relations with external actors.On the other hand, we chose Sangharshpur to represent the opposite state of affairs.
Initially group discussions were held with key informants to collect basic information about the villages, and the history of the HRMSs and JFM management.
The key informants included six to ten past and present executives of the new village institutions and the panchayat, two officials of the local Forest Department, and a TERI official.Secondary data on demographics, socioeconomics, and natural resources endowments was collected from the offices of the Forest Department, panchayat, and the new village institutions.An analysis of this data provided an initial overview of the issues under study.
On the basis of the first field visit and the data collected, a professional field investigator belonging to this region was hired to collect further data.Before beginning fieldwork, the field investigator was thoroughly briefed about the purposes and theoretical issues of our study, and provided a list of questions and a checklist.During this round of data collection, the field investigator interviewed past and present executives in Sahyogpur and Sangharshpur a second time.Further, interviews were conducted with people from the poorer sections of the villages, with a focus on understanding their role in and views about the new village institutions, the benefits they derived from the resources managed by these institutions, and patron-client relationships in the villages.In the end, these interviews constituted the main source of information about the intra-elite competition and conflicts discussed in this article.Data from all interviews have been recorded in detailed field notes in Hindi and verified through discussions between the main author and the field investigator.Data was also collected on the number of elections held for executive committees of the new village institutions and the panchayat, as well as landholdings of all the elected past and present executives of these institutions in Sahyogpur.From Sangharshpur, however, only limited data was available on the elections as the Sangharshpur HRMS has not been functioning since 2003.In Sangharshpur, we were informed that the HRMS records were with the Forest Department, but due to the disruptive situation in this village, we were unable to retrieve these records.Hence, for Sangharshpur, we will only present data based on the field notes from the interviews.

The case study villages
Table 1 presents some basic data about the two villages.Sahyogpur is larger than Sangharshpur in terms of geographical area, population, and number of households.In these two villages, forest lands under JFM are of approximately equal size, and both have water harvesting dams for irrigation of these lands.Sahyogpur has a larger grazing area (shamlats, or village commons) and other land.The JFM forests are of a similar type -sub-tropical moist deciduous -and are classified as protected forests by the Forest Department, implying that they are designated for village use.Both forests are in good condition and supply the villagers with a wide range of products and services.Among the economically important species found in the forests are: khair (Acacia catechu), the wood, bark and roots of which are in great demand and which yields a commercially valuable byproduct, a condiment called katha; shisham (Dalbergia sissoo), which provides valuable timber used for making furniture and; bhabhar (Eulaliopsis binata), hereafter referred to as fiber grass, which is used by the paper industry as raw material and locally for making of ropes.
The populations of Sahyogpur and Sangharshpur are dominated by the ethnic group Gujjar that in India is classified under the Other Backward Classes (OBC) category (see Table 1).Historically, the Gujjars have been identified with agriculture and cattle rearing in northern India.Sahyogpur and Sangharshpur have a small group of households belonging to the Scheduled Castes (SC), and a few Brahmin and Kshatriya households that are classified under the General Category (GC).Traditionally, Brahmins and Kshatriyas constitute the elite in India.In Sahyogpur and Sangharshpur, however, the sheer dominance of OBCs implies that the traditional Indian caste hierarchy is not a determining factor in defining the local elite.
Rather, personal wealth is the main distinguishing factor in defining the local elite.We assessed wealth based on ownership of cultivable land.The size of landholding is strongly correlated with wealth of farmers in the agrarian landscape in India.The definition of the local elite in the agrarian landscape in India is a relative concept, as landholding patterns and socio-economic and cultural characteristics (our case study villages are inhabited by people from a deprived community, as mentioned above) differ from region to region and even village-to-village.There are also other defining characteristics of the rural elite, such as possession of irrigated land, farm machinery or a large cattle herd, government employment, or education and the ability to speak in public forums.However, we chose to focus on land ownership as it appears to be correlated with personal wealth.Table 1 shows that the case study communities have small and equal shares of households owning more than 2 ha of land.The vast majority of the 101 landless households in the two communities are highly dependent on at least one land-owing family for their livelihood.The patron-client relationship between landless and landholding households has feudal traits, i.e. the landless are provided for by the patron and have almost no income of their own.This aspect is important in order to understand how the JFM processes play out, as we reveal later.Sangharshpur has more irrigated agricultural land than Sahyogpur.The households in Sahyogpur, which has a more undulating landscape with less agricultural potential, generally depend more on livestock than those of Sangharshpur.The share of households from Sangharshpur with income from formal employment in the private sector (in a nearby industrial town) and the government sector is higher than that in Sahyogpur, making the former comparatively betteroff.Sahyogpur households are, in turn, generally more dependent on the forest.

The new village institutions
HRMSs were established in Sahyogpur and Sangharshpur in 1990, when water harvesting dams were built near these two villages.Between 2001 and2003, Village Development Committees (VDCs) were established in both villages by the Department of Agriculture, Haryana, financed by a World Bank project.The mandate of these committees was focused on soil and moisture conservation.Finally, the Haryana Community Forestry Project, which was co-funded by the European Union and Haryana Forest Department, formed a Village Resource Management Committee (VRMC) in Sahyogpur in 2002 with the purpose of establishing and managing a village woodlot.Sangharshpur was not targeted by the Haryana Community Forestry Project, but in 2005 a Village Forest Committee (VFC) was formed in Sangharshpur under a project supported by the Japan Bank for International Cooperation.
All three village institutions, in both Sahyogpur and Sangharshpur, have similar executive committee structures, and eight to 12 members.The executive committees of the VDCs and VRMC were modeled on the HRMS structure.In Sangharshpur the executive committee members are the same for the HRMS and VDC.Villagers and concerned government agencies agree to this arrangement to overcome complexities and mitigate the costs of setting up different committees under each project.One notable difference among these institutions is that the HRMS is an autonomous society of villagers registered under the Societies Act, 1860, whereas other committees are established by villagers in consultation with respective government agencies implementing the projects.

Benefit sharing arrangement
The benefit sharing arrangement under the JFM program in India is the raison d'être for the power struggle over the process and outcomes of JFM between State and community actors.In this section, we will summarize some of the legal and practical issues surrounding benefit sharing, to provide a backdrop to the later sections.
In relation to forest access, the arrival of the HRMSs in the villages in Haryana meant both new restrictions and privileges.As an entity, the village has received increased legal access to forest products as well as improved irrigation systems.Increased restrictions on access to grazing areas are a notable exception.On sharing of forest produce from JFM areas, the JFM policy of the Government of Haryana says: "All members of HRMS(s) may be permitted to collect dry and fallen wood, fencing material, limited number of bamboo and poles free of cost from their respective [JFM] areas for their bona fide domestic use and not for sale.
[Previously] existing rights of members or non-members are not to be [overruled].The HRMS(s) shall regulate collection of the forest produce on a sustainable basis by framing rules for their members in consultation with the concerned Range Forest Officer" (Government of Haryana 1998).
For many of the non-wood forest products (NWFPs) used for subsistence, access is free for the needy.Landless people and widows are given concessions to collect fallen and dry wood and NWFPs like fruits.The bamboo basket makers are provided with permits to cut a specified number of bamboo sticks per household.In relation to the marketable products, such as fodder grass and, in particular, fiber grass and timber, however, the Forest Department is more restrictive.
The HRMSs are required to pay the Forest Department for access to fodder grass and fiber grass on an annual basis.As per the JFM agreement, all fodder grass shall be given to the HRMSs at the average price obtained in open auctions for the last three years before the advent of JFM (with a provision of an annual increment of 1%).For fiber grass, the HRMS has first right over collection and use of this grass from forests on an annual lease basis.The Forest Department can only auction away fiber grass from a JFM forest to private contractors if the HRMS fails to bid for it.Whether the lease is allocated to an HRMS or a private contractor, the Forest Department, being de jure owner of forest land, plays an important role in deciding the value of the lease, and the bidder has to abide by the terms and conditions specified by it (see, Poffenberger and Sarin 1995;Kumar and Vashisht 2005).
On the other hand, when an HRMS sells fiber grass through an open bidding, the net income shall be divided between the Government (25%) and HRMS (75%).The HRMS contributes 30% of its share towards plough back funds for further improvement and development of the JFM area, and these funds are retained by the HRMS and jointly operated by the Forest Department and HRMS functionary with dual approval of the General Body of the HRMS.The HRMS contributes another 10% towards kalyan kosh to support implementation of JFM in other areas, and these funds are managed by the Forest Department (Government of Haryana 1998).
Unlike fodder and fiber grass, the benefit sharing arrangement for timber favors the Forest Department.The JFM policy of the Government of Haryana reads: "After ten years of [JFM] to the satisfaction of conservation of forests of that area, the [Haryana Forest Department] shall share the income from timber, at the time of final harvest and after deducting the expenditure incurred [in] felling, transportation and auction, with the HRMSs [in the ratio of 70 (Government) and 30 (HRMS)] provided that HRMS shall contribute 30% of its share towards plough back fund …. .and another 10% towards [kalyan kosh]" (Government of Haryana 1998).
The HRMS fund, also called 'village development fund', consists of the HRMS's share of proceeds from sale of fiber grass and timber (minus deductions for plough back funds and kalyan kosh) and other various income sources.This fund is to be used for village development and community welfare as decided by the General Body of the HRMS, and is maintained as a separate bank account to be operated by the cashier and any of the other two office bearers of the HRMS.It is the responsibility of the Forest Department that the accounts are maintained properly by the HRMS.Some examples of village development and community welfare undertaken using the HRMS fund across the Yamunanagar Forest Division in Haryana are: buying communal utensils and a gas stove which can be used by anyone organizing feasts for marriages or other functions in their families, some minor road improvements, and improvements in schools or community halls.
While the JFM policy lays down unambiguous directives for sharing of benefits between the Forest Department and the HRMS, there are few procedures guiding benefit sharing within the community.The JFM policy merely mentions that the Forest Department has to "ensure that the most disadvantaged men and women members of the HRMS have access to forest produce for meeting bonafide needs of firewood, fodder, fiber, etc. by promoting consultative and accountable decision making by HRMS office bearers" and "to ensure equitable distribution of usufruct which may be in cash or kind" (Government of Haryana 1998).
In the case study villages, the benefits to the poor are mainly confined to forest products as the landless do not benefit from dam water.The dam water is the most valuable environmental service produced under the regime of the HRMSs in Sahyogpur and Sangharshpur.The water is mainly used for irrigation and has implied enhanced agricultural production.Both villages have seen many conflicts amongst elites, and between elites and poor villagers, over dam water use, which evidently reflects the differential economic benefits to elites and non-elites.The water is distributed among farmers on a rotational basis (10 hours to each farmer at a time) by a private contractor who is overseen by the HRMS office holders.This is a compelling reason for elites to run for office in the HRMS.To ensure equity, HRMS executive committees in the two case study villages some years ago decided to provide equal entitlement to dam water to all villagers, and not only landowners.As per the JFM policy, the HRMS has to manage water distribution from the water harvesting dams on an equitable basis, implying that all members are entitled to claim an equal share of water, irrespective of whether they own land or not.Hence, this decision brought the de facto management in line with the policy.Following the change in management, the landless and land-poor could sell water to large landowners, as Sengupta et al. (2003) reported for two other HRMSs.Under the former rotational system, the fee was fixed at Rs. 20 per ten-hour release of water.The change in entitlements, however, implied that large landholders often paid twice that rate when buying from landless and land-poor households.Outspoken resistance among the larger landholders eventually led to a reversal of the system.When raising this issue of water with poor villagers, they would describe at length their gratitude to the local elite for providing them with daily necessities like milk and fodder, and helping them during emergencies like providing money to get their children married.Between the lines, they expressed that the system of tradable water shares threatened to erode the very patron-client system which is fundamental to their livelihoods.
In the following two sections, we will turn to each of the two case study villages to attempt an analysis of the struggle for power over the processes and outcomes among internal and external actors.The struggle for power over the processes and outcomes of JFM is in itself a continuous process that may be expressed in all forms of structures and interactions among actors, from the informal ones, such as groups of kinship and everyday activities in the village, to the more formal committees and associated activities, such as meetings and general assemblies in the new village institutions.

Sahyogpur: state and intra-elite cooperation
Table 2 provides an overview of the executive members in the village institutions in Sahyogpur over the period 1990-2007.In 1990, the first of ten elections to the HRMS was held, whereas elections for the VDC and VRMC have so far been held only once, in 2001 and 2002.The tenure of the first HRMS executive committee in Sahyogpur was five years.This committee, however, was very inactive.Only in 1998, when Bhishma (to be presented later) was elected president, the HRMS started growing in influence and importance.Since then, six elections were held during 2000-2007, facilitated mainly by TERI.
The mere presence of the new village institutions implies that more people are given opportunities to engage in decision-making on natural resources management than in the past, when the village panchayat was the only forum.In fact the wellfunded HRMSs are encroaching on the traditional domain of the panchayats, the statutory and democratically elected village institution created to foster local self-government.The HRMSs are registered as legal entities, which gives them credibility and access to income to finance village development activities.In all, 92 different individuals, corresponding to 6% of the total population of the village, were elected at least once to an executive position.All castes were represented more or less according to their proportion of the village population.With women and landless executives constituting 35% and 21% of those elected at least once to new institutions respectively, Sahyogpur has a high representation of these groups when seen in the context of Haryana.The same pattern concerning representation of women and landless is repeated in the Sahyogpur panchayat (Table 2).The reason for this high level representation of women and landless in these two village institutions is the legal requirement that these groups are represented along with the facilitation of the election process by TERI.In addition, elected local elites often seek to conciliate people from their clan, as also the landless who are attached to their households, by supporting them as candidates for the executive committee.Sahyogpur is one of the few HRMSs in Haryana with the distinction of two women elected as the HRMS president in 1995 and 2000 elections.The president elected in 1995 was a literate woman who dared to speak up in large meetings, and interestingly enough, she was landless, so was the man elected as president in 1997.Both these landless presidents were, however, perceived as 'weak' presidents in the village, and during their tenure the HRMS achieved very little.Generally, women and landless executive members do not actively participate in HRMS activities.The former specifically because of the custom of the pardah or the gunghot in this region, which secludes women from men.
Clearly, sex, caste and wealth all matter to the opportunities of exercising one's mandate in the village institutions.So does the position one holds in the individual executive committee.Almost invariably, one or more of the three individuals holding the premier positions of president, secretary and treasurer, exercise a high degree of control over committee activities.The election mechanism is that the members of the HRMS are elected by the villagers on a neighborhood basis, and they in turn select the president, secretary and treasurer amongst themselves.Table 3 shows the landholdings of those chosen to the premier executive positions, and further substantiates that the new village institutions are controlled by the elites of Sahyogpur.Of 36 premier executive positions in the new village institutions from 1990 through 2007, 30 were held by landholders, and 18 by local elites.Comparing the characteristics of those elected to the three new institutions to those elected to the panchayat indicates an important difference.The average landholdings of people elected to the panchayat appears to be lower (Table 2), none of the five local elites elected from 1990 to 2007 were re-elected, and only one was elected president of the panchayat.What appears to be a discrepancy between the three new institutions and the panchayat could be related to differences in the electoral procedures.Elections to the panchayat are regularly held under the direct supervision of the State Election Commission, a constitutional body, using ballot boxes.Accordingly, the opportunities for local interference with panchayat elections are minimal as compared to the elections to the new grassroots development committees or 'users committees', where elections are generally based on consensus processes or voting by hand.On constitution of an HRMS, the JFM policy of Government of Haryana says: "An HRMS shall be constituted by its members of their own free will and consent" (Government of Haryana 1998).In his critical article on the proliferation of local user committees during the second wave of decentralization in less developed countries in the mid-1990s, Manor (2004, 194) observes that "user committee members are often selected by less-reliably democratic means than are most elected councils at lower levels [as in case of panchayats in Indiaour emphasis], or by undemocratic means".
Table 4 indicates the premier positions held in the three new institutions by six individuals who have been prominent figures in politics in Sahyogpur in the period 1990-2007.These six prominent individuals, Bhishma, Yudhishter, Bhim, Arjun, Nakul and Sahadev (the real names of villagers are withheld) together held 17 of 36 of the most coveted premier executive positions over the period 1990-2007.Bhishma is a very dominating figure in the village institutions of Sahyogpur.In addition to the premier positions shown in Table 4, he was elected ordinary VRMC executive and is the only villager who has held premier positions in all three new institutions.He has the largest landholdings of all elected local elites.During Bhishma's second tenure, the HRMS prospered and received an award for excellence.Likewise Yudhishtir was elected president of the VRMC in the one election held.In the HRMS he was elected secretary once, treasurer twice, and ordinary executive once.Bhim is the only landless among the six dominating elites.He is well respected by landless and local elites alike, perhaps because he is a retired army soldier (who often command respect in Indian rural society), and was elected president twice, secretary once, and ordinary executive of the HRMS once.He also served as a panchayat executive.
The new village institutions have managed to instate rules that establish relatively effective control of the utilization of natural resources in the area.The authority of the institutions have, however, been won without effort.In Sahyogpur, a few years ago, a herder from a neighboring village was caught grazing cattle at night in the forest, and had all his 15 heads held back by the executives.In accordance with the rules and regulations of Sahyogpur HRMS, the herder was asked to pay a fine of Rs. 125 (US$ 2.7) per head.Comparing the level of fine with the average wage of Rs. 158 (US$ 3.5) paid per person per day in Haryana (as of June 2010) under the Mahatma Gandhi National Rural Employment Program indicates that, if followed, the rules are relatively strict and should deter potential offenders.The herder promised the executives to pay the fine within two days but never did.Rather, he was again caught grazing his cattle inside the Sahyogpur forest.The executives again confiscated the cattle and reported the offence to the police.The police, however, did nothing, following which the executives complained to the Forest Department.When this too was in vain, the executives finally approached the Divisional Commissioner (the head of the district administration) who directed the police to register the complaint.It was then that the herder paid the fine.While the case was in progress, it led to a deteriorated relationship between Sahyogpur and its neighboring village.It was through the efforts and personal networks of the elite, particularly Bhishma, that the authority of Sahyogpur's institutions prevailed.Bhishma had very good rapport with the retired Forest Department official.The latter was appointed by TERI as the facilitator of the JFM initiatives in Sahyogpur and Sangharshpur.Bhishma was more or less chosen as 'spokesperson' or 'showcase leader' of JFM in the Yamunanagar Forest Division in Haryana by the TERI official, because of his wealth, leadership, and good interpersonal skills.Due to these important qualities of Bhishma and the initial success of Sahyogpur in developing collective action in protecting their forests, the TERI official used to recommend it for visits by Forest Department officials.Hence, Bhishma obtained good rapport with the Forest Department, and Sahyogpur became known as a prime example of successful JFM.Clearly, the relationship between Bhishma and the TERI official was beneficial to them both.The TERI official had a good showcase when receiving visitors, and could use the case of Sahyogpur to argue for his success in assisting the HRMSs.No doubt, the development of good relations between Bhishma and the external actors -and Sahyogpur's standing as a prime example of JFM -were important to the way the leaders were able to counter challenges to the authority of the local institutions.
One such challenge came from the 'timber mafia' -a group of people involved in illegal harvesting of valuable khair trees in the area.The village watchman along with the Forest Guard belonging to the Forest Department, came upon a band of people cutting trees at night, but were both seriously beaten up when trying to apprehend the offenders.The Forest Guard was admitted to a hospital with serious injuries from a blow on the head with an axe.The police registered the case but took no action -presumably because the timber mafia has connections with influential people.Later, it was suggested to the Forest Guard by his seniors to withdraw the complaint.The executives of Sahyogpur HRMS, however, persisted with the senior forest officials at the division level, and when this failed, they approached the Principal Chief Conservator of Forests (PCCF), i.e. the head of the State Forest Department.The PCCF promised that his sub-ordinates would be directed to immediately take action on this issue.In the end, no action was taken.Nevertheless, after this incident there has been no illegal timber harvesting in the forest.The evidence that the village elites were able and daring enough to bypass lower level police and forest officers to obtain the needed support, is an indication of the strength of the local leadership.In Indian society, such bypassing of lower level bureaucrats is detested and seen as an insult of those bypassed, implying that it carries a cost of deteriorated relationships and a risk of retaliation.
During interviews, most of the local elites strongly asserted their view points on forest management issues.For example, during our discussions with past and present executive members, the local elites acknowledged the Forest Department's efforts to promote JFM in their locality by naming individual officers, but did not hesitate to identify corrupt officers.The assertiveness of local elites, also in direct address to Forest Department officers, indicates their willingness to actively take on the role of defending the authority of their institutions against external actors.It also serves to manifest their control of the decision-making processes internally in the community.Further, a few of these local elites in Sahyogpur, such as Bhishma, are emerging as regional leaders.The position attained by Bhishma in a sense becomes self-reinforcing as it enables him to demand the attention and services from the Forest Department when needed, which further strengthens his standing in the village.
The growing assertiveness of local elites in Sahyogpur against the Forest Department is underpinned by unresolved issues over access to resources and sharing of benefits.Generally, most Shiwalik hill communities have benefited substantially from JFM through exploitation of NWFPs, such as fiber grass, despite royalty payments and control on utilization exercised by the Forest Department.However, profits have been dwindling due to rising extraction costs, declining prices, diminishing demand from paper mills for fiber grass due to substitution by timber from tree species which grow fast, and a downscaling of the paper mill industry in the area (Chaturvedi 2004;Kumar and Vashisht 2005).This has implied that the royalty for fiber grass has become a contentious issue between Sahyogpur HRMS and the Forest Department in recent years, and has led to an increased focus on the potential benefits from utilization of timber from the forest.Increasingly, local elites in Sahyogpur question the legitimacy of the Forest Department's powers to approve of any timber harvesting from their forest.In Sahyogpur, where the HRMS was instituted more than 15 years ago and hence, is entitled to get a share from timber exploitation (as per the JFM policy mentioned earlier), people demand that timber harvesting should commence.According to the president of Sahyogpur HRMS, both khair (Acacia catechu) and shisham (Dalbergia sissoo) are available in quantities that merit harvesting in many areas of the forest.Yet, the Forest Department resists timber harvesting.
According to the retired forest officer working for TERI, the reasoning behind the Forest Department's resistance to timber harvesting is that substantial earnings from khair and other valuable timber species would result in intervention by local political elements to capture the earnings.This politicization could in turn, according to the view of the Forest Department, jeopardize the stability of the HRMS institution.This official concern could, however, be questioned, as the Forest Department benefits from the current situation by maintaining control over the revenues from timber.There is a natural resistance in the Forest Department to share timber -the most valuable product of the forest land they own.More than the revenue from timber that the Forest Department gains, the reason for such resistance can be attributed to its inclination to prevent such benefits from reaching the community, as it will set a precedent for other similar demands in the future, which will weaken the department's authority.

Sangharshpur: state and intra-elite competition and conflicts
Whereas the village institutions in Sahyogpur have succeeded in initiating and maintaining a JFM regime, the story of Sangharshpur is one of conflict and failure.Over the last decade a few local elites have vied to control the HRMS, which eventually led to a reversal of JFM by the Forest Department.
The president of the most recently elected HRMS in Sangharshpur is a member of the local elite.Four local elite family members were last elected to the HRMS, whereas in the panchayat there has almost always been one elite at the helm (the president, or sarpanch), though none was ever re-elected.
Since 1990 all Sangharshpur HRMS presidents have been local elites, except in 1998, when a land-poor person was elected.Only the first president served the full tenure of five years (1990)(1991)(1992)(1993)(1994)(1995) and provided much needed leadership for activities of the HRMS in its nascent stage.During his tenure, regular meetings were held and attended by Forest Department officials, and rules and regulations were formulated for forest management and the protection and distribution of benefits including dam water.Accordingly, Sangharshpur HRMS was initially perceived as a highly promising institution.
In 1996, however, the Forest Department decided to lease fiber grass in the forest land managed by Sangharshpur HRMS in an open auction.Until then, fiber grass was auctioned by the HRMS and was considered a novel JFM development initiative (Lise 1997).The Forest Department gradually dominated the HRMS with the obvious intent of capturing rent.This rent-seeking by government officials undermined JFM (Lise 1997).The lack of benefits for the villagers protecting the forests sparked their discontent with the JFM program.Additionally, the differences among the few local elites trying to control the HRMS further strengthened the Forest Department's position.The community was divided over allegiance to these local elites.Naturally, the frontline Forest Department officials favored those among the local elites who would collaborate with them.
Conflicts among local elites surfaced openly during the election in 2003.The previous five elections held between 1995 and 2002 were based on consensus (oral vote).A conflict between two local elites competing for president in the 1999 election was avoided by the facilitation of TERI with an appeal for the good of the community.In 2003, the incumbent president Gaj Kumar ('Gaj' in Hindi means elephant'; 'Kumar' is a very common name in India, mostly used as a prefix in South India and suffix in North India) was challenged by Chunauti Kumar ('Chunauti' means challenge), who had considerable community support (the real names of villagers are withheld).A few villagers claimed that Gaj Kumar had embezzled HRMS funds.Villagers were divided over their allegiance to Gaj Kumar and Chunauti Kumar, making an election by consensus impossible.Candidates pressured the landless to take sides, some of whom decided to boycott the election process after having been subjected to physical harassments and threats.TERI and the Forest Department officials intervened without success.This situation continued for several months, and ultimately the elections were cancelled.This was in violation of the JFM agreement between the HRMS and the Forest Department, and the latter thus terminated the HRMS's lease rights to the forest and dam water.This was the beginning of the Forest Department's de jure control of forest management.Section 9(ii) of the JFM policy of the Government of Haryana on 'Dispute Arbitration/Termination of Agreement' says: "If an HRMS as a whole fails to comply with any of the conditions laid down [in a JFM agreement between Haryana Forest Department and HRMS], despite at least two warnings in writing by the Forest Officer, the [Divisional Forest Officer] shall be entitled to terminate agreement with the HRMS after due enquiry" (Government of Haryana 1998).
The institutional breakdown had a tremendous impact on resource management and use in Sangharshpur.Because the Forest Department auctioned dam water and fiber grass to a contractor rather than to the village, the rules and regulations instituted by the community for water distribution were ignored, and local elites appropriated the benefits.According to interviewees among the poorer segments of the village, it is not uncommon for the elites to bribe the contractor by giving him money or liquor, to distribute water in their favor.Under the Forest Department's terms, all income from water and fiber grass benefits the state treasury rather than the HRMS fund.Finally, the termination of free access to fodder from the forest implies that a household with four or five heads of cattle experiences increased costs in the range of Rs. 5000 (US$ 108) annually to purchase fodder.The concessions to the landless to collect fallen and dry wood and widows to collect NWFPs like fruits of amala (Embilica officinalis) and harad (Teminalia chebula) have, however, been continued even after the Forest Department took over.
Some villagers approached the Forest Department for assistance in reviving the HRMS.Unlike Sahyogpur, however, Sangharshpur failed to get the Forest Department's support.The division of the village community between the conflicting elites impeded the villagers' efforts to approach the Forest Department despite the efforts of TERI to break the ice.Later, in 2004, concerns for the losses of income from the forest compelled the panchayat to take an initiative with support from TERI.A young man from amongst the elite, Umeed Kumar (a relative of Gaj Kumar; 'Umeed' in Hindi means hope) was persuaded to take the lead in obtaining village lease rights to dam water and fiber grass by paying Rs. 249,000 (US$ 5395) to the Forest Department.In return for his efforts, the panchayat promised Umeed Kumar he would be the new HRMS president.That year, a new executive committee was constituted and Umeed Kumar became president.But the HRMS resurgence could not be sustained for long.The 2007 election saw the reappearance of the division among the elites.The incumbent Umeed Kumar wanted to continue as president, but another local elite member, Araujo Kumar (a relative of Chunauti Kumar; 'Araujo' means aspire), opposed him and contested the election.A general meeting failed to arrive at a consensus.The community was again divided, and poor villagers were harassed by people close to these elites to garner their support.
The 2003 and 2007 elections revealed that village politics evolved around two competing elite families traditionally dominating the village.On the one hand, were Gaj Kumar and Umeed Kumar (ex-presidents), and on the other hand, Chunauti Kumar and Araujo Kumar (aspiring presidents), a combination of older and younger generations.Lack of cooperation between these two fractions gave the Forest Department a convenient chance to gain control of the leasing of forest products and water to its own benefit.
Although elections in Sahyogpur have seen conflicts, they have been overcome by collective efforts.Once, when two people vied for the presidency and neither wanted to withdraw, the general assembly members simply elected a third person as president.In another case, an ex-president of the Sahyogpur panchayat, who was elected as an executive member in the 2005 election, was initially denied the presidentship by the generally assembly of the HRMS, because he had been involved in an hitherto unsettled case concerning the use of development funds during his tenure in the panchayat.Only after having settled his case with the panchayat and making apologies to the general assembly was he supported for the position of the president of the HRMS.

Discussion
The results have illustrated diverse outcomes of similar institutional reforms in what appear to be relatively similar neighboring communities.This implies that we need to look beyond institutional reform to explain outcomes.The evidence points to two differences between the communities that appear to constitute promising directions of further consideration.First, the difference in intensity of conflict internally in the communities appears to have been instrumental in shaping the direction of events.Second, the ability of key people -members of the elitesto manage external actors, in the sense of drawing upon them when needed to advance local priorities and fending them off when seeking to avoid unwanted interference, seems to have been equally important.Obviously, and what was also clear from the evidence, these two aspects are highly related and interlinked.
Recent studies on social capital and local leadership indicate that local leaders, or elites, can be a blessing to local participation in natural resources management, in the sense of playing a key role in the establishment of collective action to manage natural resources through their networks to external knowledge, resources and people (e.g.Bodin and Crona 2008).These studies exemplify recent developments in the literature on the network theory of social capital.Within this line of theory, Lin (1999, 35) defines social capital as "resources embedded in a social structure which are accessed and/or mobilized in purposive action".More generally, the literature on social capital features two viewpoints about the role of social capital in social processes.One strand, represented by Bourdieu (1990) among others, focuses on the role of social capital in supporting and reproducing a dominant class, whereas another strand, represented by Coleman (1988) and Putnam (1993) sees social capital as a collective asset that is useful to all members of the collective and enhances its ability to pursue strategic advantages in society.In the former, social capital is contributing to reproduction of elite capture, whereas in the latter, social capital can be seen as resources embedded in a community that its members -including elite members -can access and mobilize in pursuit of strategic advantages.Woolcock and Narayan (2000) distinguish between bonding and bridging social capital, where bonding is relations of reciprocity within the community that secure a minimum subsistence livelihood [see also Scott's (1976) concept of the moral economy of the peasant] and bridging is relations to more resourceful external actors.Woolcock and Narayan (2000) argue that the poor usually are characterized by more bonding and less bridging social capital, whereas (local) elites are characterized by more of the latter.We believe that this network theory of social capital provides a useful analytic lens to study and forward explanations for the observed varied outcome of institutional reforms in our case.
The diverse outcomes and the roles played by internal conflict and elites' management of external relations in shaping these outcomes in our case appears to fit well with a network theory of social capital.The build up of strong personal ties between Bhishma from Sahyogpur and external actors from TERI and the Forest Department seemed to make a difference to the ways in which Sahyogpur managed to fend off threats to its management initiative.The relations between Bhishma and the relevant external actors could clearly be interpreted as an indicator of a high level of bridging capital.Likewise, the internal conflicts and lack of ability to effectively engage external actors displayed by the elites in Sangharshpur might be explained by low levels of bonding and bridging capital, respectively.The external NGO assistance from the TERI official was important to keep the Forest Department's intervention attempts in check in Sangharshpur.When the influence of the Sangharshpur HRMS was declining, the TERI official persistently urged people to revive the organization, but saw his efforts hampered by conflicts among local elites.One executive of the non-functional Sangharshpur HRMS commented that "it was the TERI official's continued assistance that deterred the Forest Department from taking over our society [HRMS] in the past, although ultimately the Forest Department prevailed due to our inability to resolve our differences".Unfortunately, we are unable to take this investigation much further beyond this level of setting up of hypotheses, as we have not empirically investigated the nature of networks that the individual members of the elites are part of.
Both the case studies reveal that Forest Department officials played a veiled role in shaping JFM outcomes.They preferred to establish good relations to local elites in the villages to have some control over the HRMSs, as in the case of Sahyogopur.This strategy, however, could not function in Sangharshpur due to ongoing struggles between local elites over power over the HRMS leading to instability of the local regime.In relation to such 'problematic' villages, the Forest Department officials face strong incentives to recentralize control of the resources, to the detriment of villagers' access to resources, local public finances and the build-up of local institutional capacity.

Conclusions
This article has shown how local elites exert a tremendous influence over the opportunities for newly created village institutions to assert their authority and defend their interests against a variety of actors in the continuous power struggle over the processes and outcomes of JFM.In the context of the two case study villages, featuring a high degree of wealth heterogeneity and strong patron-client relations, local elites mattered to the degree that they constituted either a curse or a blessing for village institutions.In both case study villages, we note how the robustness of village institutions was affected by the degree to which the local elites cooperated rather than competed over power over the institutions, which in turn influenced outcomes.Literature on the role of elites in processes of local participation in forest management in the developing countries mostly focuses on the risk of elite capture.However, this study draws attention to the importance of looking beyond elite capture to analyze conflict and cooperation among local elites and how they in turn influence the sustainability of local institutions.The network theory of social capital appears a promising avenue for such studies of the role of local elites.
The patron-client relationship, traditional domination by local elites, and high perceived costs of participation are some of the underlying reasons for the important roles played by elites in the case study villages.These circumstances surrounding the poor in the study area are more or less structurally determined, and participatory processes can probably only influence them over a longerterm.In our case, the State actors did not interfere with this status quo.Rather, they focused on maintaining control over lucrative assets and only intervened to support the local processes when pressure was applied by the right people in the right place -by local elites.
We do not mean to say that the role of local elites, who, along with State actors, were instrumental in defining the course of events in JFM in this case, is flawless.Rather, we believe that the case stresses the point that under circumstances where elites are likely to be dominating the local political arena in the foreseeable future, their role and interests may need to be considered to ensure that the concept of JFM takes root at the local level.Further, the case illustrates that institutional reform at the local level -a la JFM -itself does not guarantee changes in actual management of natural resources.Rather, vested interests at the local level and among State actors may continue to shape events while working inside or outside the new institutional landscape.
To conclude, attempts at institutional reform at the community level, such as JFM, in hierarchical societies, such as rural India, should proceed with modest expectations and an eye for the incentives facing local elites and implementing and facilitating State actors.

Table 1 :
Salient features of the case study villages.
a Figures in parentheses in this column are the number of women executives elected at least once.b Figures in parentheses in this column are the maximum landholdings by local elites.c Executives elected to more than one institution are counted only once to obtain the number of executives elected at least once to any institution.

Table 4 :
Six local elites dominating the premier positions in the new village institutions inSahyogpur, 1990Sahyogpur,  -2007.   .
a Election held only once.