Lợi ích của địa phương từ rừng cộng đồng trên các vùng đồi núi của NepalCommunity forestry has been in existence for about 20 years in Nepal, but there are few observations on changes in resource use since its inception. This study aims to investigate how farmers affected by community forestry have adapted to its introduction and whether their livelihood options have changed in response to changes in forest accessibility and forest product availability. A total of 309 households were interviewed in eight forest user groups in the middle hills of Nepal and results on reported changes in forest product collection and livestock numbers analysed. Reported forest product collection has increased since the introduction of community forestry, while livestock numbers have decreased. The main findings of the study are that, despite controlled access to the forest, collection rates have increased slightly in the majority of income groups and households. The number of trees on private land has increased, while the size of livestock herds has decreased. As long as the farmers can continue to adapt without adversely affecting their livelihoods they will support this forestry management system leading to an increased chance of the long term success of common property forest management in Nepal. D 2006 Elsevier B.V. All rights reserved.
Local benefits from community forests in the middle hills of Nepal Bhim Adhikari, Frances Williams, Jon C. Lovett * Centre for Ecology, Law and Policy, Environment Department, The University of York, Heslington, York, YO10 5DD, UK Received 26 October 2004; received in revised form 5 October 2005; accepted 8 November 2005 Abstract Community forestry has been in existence for about 20 years in Nepal, but there are few observations on changes in resource use since its inception. This study aims to investigate how farmers affected by community forestry have adapted to its introduction and whether their livelihood options have changed in response to changes in forest accessibility and forest product availability. A total of 309 households were interviewed in eight forest user groups in the middle hills of Nepal and results on reported changes in forest product collection and livestock numbers analysed. Reported forest product collection has increased since the introduction of community forestry, while livestock numbers have decreased. The main findings of the study are that, despite controlled access to the forest, collection rates have increased slightly in the majority of income groups and households. The number of trees on private land has increased, while the size of livestock herds has decreased. As long as the farmers can continue to adapt without adversely affecting their livelihoods they will support this forestry management system leading to an increased chance of the long term success of common property forest management in Nepal. D 2006 Elsevier B.V. All rights reserved. Keywords: Community forestry; Adaptive management; Forest product collection; Livelihoods 1. Introduction In the last few decades community-based manage- ment of common pool resources (CPRs) has become an important land-use policy in many countries, with Nepal being one of the pioneers in introducing this system of resource management. The involvement of people in the management of local natural resources, and therefore the need to understand their attitudes towards the management systems, developed from the realisation that traditional top-down management techniques were not solving the problem of over- exploitation and subsequent environmental degrada- tion. This led to the development of community based natural resource management (CBNRM), with local communities in control of the management of their local natural resources. CBNRM programmes are based on the fact that people are likely to have more interest in conserving a forest or other natural resource that is near them, than central government or private institutions (Ostrom, 1990; Nugent, 1993; Uphoff, 1993; Wade, 1988; Bromley and Cernea, 1989; 1389-9341/$ - see front matter D 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.forpol.2005.11.002 * Corresponding author. Tel.: +44 1904 434063; fax: +44 1904 432998. E-mail address: JL15@york.ac.uk (J.C. Lovett). Forest Policy and Economics 9 (2007) 464– 478 www.elsevier.com/locate/forpol Oakerson, 1992; Tang, 1992; Bardhan, 1993; Tsing et al., 1999). Local communities also have a greater understanding of the unique conditions in their area that affect the natural resource and so can adapt their management techniques accordingly and more effec- tively than a centrally controlled management plan (Agrawal and Gibson, 1999; Agrawal, 2001; Twy- man, 2000). Analysis of the foundation of communi- ty-based regimes in developing countries has shown that local institutional arrangements including cus- toms and social conven tion s des igned t o induc e cooperative solutions can overcome the problems of collective action and help achieve efficiency in the use of such resources. Moreover, local collective action can be instrumental in finding rules for allocation of the resource between different users in a way that is seen as equitable by the users themselves (Meinzen- Dick and Knox, 1999). An increasing number of scholars advocate that decentralized collective man- agement of CPRs by t heir users could be an appropriate system for overcoming the dtragedy of the commonsT (Ostrom, 1990; Berkes, 1989; Wade, 1988; Jodha, 1986; Baland and Platteau, 1996) and it is being appli ed in many countries (e.g. Kumar and Kant, 2005; Masozera et al., in press; Rosyadi et al., 2005; Sunderlin, in press). The wide recognition that local communities are able to manage resources efficiently, equitably, and sustainably has led to the devolution o f natural resource management from centralized government control to local user groups in Nepal and other South Asian countries. One particular form of CBNRM is community forestry (CF) which was introduc ed in the 1970s in Nepal so that local communities could be involved in managing and utilizing the forests in a sustainable way (Chapagain et al., 1999; Fisher, 1995). Community forestry started with the 1976 National Forest Plan which enabled limited areas of government forests to be handed over to local control. The 1988 Forestry Master Plan extended this policy and declared that all accessible forests in the hills areas of Nep al s houl d be ha nd ed ov er to local communities from national control by the Forestry Department (Soussan et al., 1998). Over the last twenty-five years, the Nepalese government has been transferring property rights in forests from the state to forest user groups (FUGs), which are legally recog- nised groups for managing local forests under a community-based property rights regime. The Forest Act 1993 has granted secure property rights and put local people in charge of managing the forests, with technical support from the Forestry Department. About 997,077 ha of state managed forests have been already handed over to 12,560 FUGs benefiting 1401,361 user households (HMGN/CPFD, 2003). CF occupies about 28% of the potential community forest (POCF) area of Nepal. Community forestry is now based on FUG control of forest resources, with operational plans being designed and implemented by the FUG with support from the forestry department (Rejal and Petheram, 2001). As control of the forests has been transferred back to the community, a sense of ownership has been re- established (Springate-Baginski et al., 1998) and there is a general consensus that introduction of community forests in Nepal has been a success in terms of forest protection and regeneration (Dev et al., 2003; Richards et al., 2003). Many FUGs closed all access to the forests when they were initially handed over in order to protect the remaining forest resources and allow time for regeneration. They also restricted use of the forest to members of the FUG. According to Yadav et al. (2003), forest resources at approximately 75% of study sites were deteriorating before the introduction of community forests and management by FUGs, whereas the forests at all the sites are now improving where community forestry is well estab- lished. Protection, sound management and controlled harvesting were seen as the main reasons for this change. Activities like planting were integrated into forest management, and rules for protection were closely related to the process of improving the resource base of community owned forests. All FUG members are aware of the relationship between forest condition, species composition and type of harvesting that needs to be pract iced. Based on the existence of management institutions, together with the present condition of community forests, it appears that management of community forests is being done on a sustainable basis by the forest user groups (Branney and Yadav, 1998; Dev et al., 2003 ). Though there is an upsurge of empirical literature on biophysical and institutional aspects of resource management, very few studies examine the changes of resource use and control patterns after the institutional changes of forest management. In the context of B. Adhikari et al. / Forest Policy and Economics 9 (2007) 464–478 465 participatory forest management in South Asia, there is a tendency to assess impact of community-based resource management in terms of biophysical and institutional change rather than effects on villager’s livelihoods (Das, 2000). In Nepa l, many studies have assessed the state of country’s forests since the introduction of community forestry (Branney and Yadav, 1998), but there has been little focus on how the people affected by this management regime have adapted to it. Res tricted access to the forest has been reported to affect forest users, especially those who rely on the forest for their livelihoods, and there is some evidence that community forestry has not been successful in improving the livelihoods of poorer groups in the community (Malla, 2000). However, it is unclear whether reported forest product collection will have increased or decreased as many studies have stated that closed management regimes have restricted access to the forest (Springate-Baginski et al., 1999; Graner, 1997), but recent literature has stated that the forests are becoming more accessible after manage- ment responsibility was handed over to the local communities (Poffenberger and McGean, 1998). If the forests are still operating under a closed regime, a reduced collection rate can be expected, but if the forests are now more accessible an increased collec- tion rate is anticipated due to improved forest health. It is these apparent variations in access to the forest and to forest products that has led to this study. This study aims to explore whether the changes in management have affected resource use and collection. More specifically, we asked: (1) whether the amount of fuel wood, leaf litter, fodder, grass and thatching material coll ected have incre ased since the introduction of community forestry, (2) whether the number of trees on private land have increased due to intervention of CF, (3) whether a household’s socio-economic and demographic characteristics in- fluenced collection patterns, and (4) whether live- stock numbers change d following introduction of community forestry. Identifying changes in resource use due to the introduction of community forestry and how this has affected livelihood options will enable decision makers to adjust the management process leading to a more successful and sustainable management regime. The rest of the paper is organised into four sections. The following section deals with m ethodological issues, including survey methods, study sites and data collection. Section three presents results of the analysis and discussions are provided in section four. The paper concludes with some policy implications. 2. Study sites, survey and data collection methods The survey was carried out in two districts in the middle hills of Nepal, Kavre Palanchok and Sindhu Palanchowk. The two districts rely on subsistence agriculture and forest resources to sustain local livelihoods. They represent a typical forest dependent district in the mid-hills. The middle hills form the central belt of Nepal running from east to west across the country between the Himalayan range in the north and the Ganges River plain in the south. The altitude of the middle hills varies between 1000 and 3000 m and the area is a m ixture of agricultural land, grassland, shrub land and forest. Trees are also grown outside the forests, on the edges of fields, around houses and on kharbari land (areas of thatc hing grassland where trees are also grown). The majority of the population in this area are subsistence farmers with livestock playing an important part in their livelihood options (Adhikari, 2003). Community forestry has been in existence in this area for up to twenty years, and therefore provides a suitable case study in which to analyse the change in resource use patterns that have occurred. The study villages and FUGs from these districts for collection of household survey data and other FUG specific information were selected on a range of criteria. The focus of the field survey was to assess the quantity of fores t product collection before and after the intervention of community forestry. In this respect, it was necessary to select only those FUGs which have already reached a mature stage in terms of benefit distribution (they must be at the stage of providing forest products to the community) and which are operating in accordance with the forest management plan. In order to meet these conditions, the following criteria were considered for the selection of FUGs: (1) age of forest user groups i.e. FUGs officially handed over to the community at least five years before this study; (2) nature of forest resources (representative forest types); (3) FUGs at the stage of substantial harvesting and benefit sharing; (4) ethnic- B. Adhikari et al. / Forest Policy and Economics 9 (2007) 464–478466 ity/caste representation; (5) FUGs representing differ- ent income groups; (6) apparent degree of succes s of intervention; and (7) apparent degree of success of user groups. Furthermore, the selection was also made on the basis of the magnitude of institutional build-up that has taken place over the last five years in these villages. On the basis of these characteristics, four FUGs in each district were selected for the fieldwork. The survey was undertaken in September 2000. All households in the village were divided into income groups based on wealth ranking during the participa- tory rural appraisal method. This gave four different income groups—very low, low, m iddle and high. Important factors in the ranking were the amount of land owne d, the amount of off-farm income, food sufficiency a nd the quality of land within each household. A total of 20% of randomly selected households were surveyed from each income group within these FUGs. A total of 330 randomly selected households were surveyed. Care was taken to select an equal percentage of respondents from each stakeholder group to avoid a bias towards certain groups in the village. The household questionnaire was designed to elicit forest use information from the respondent households. It covered information on four general areas before and after CF intervention: (1) demographic information; (2) land holding, tenure and off-farm production systems; (3) natural resource management and utilization; and (4) household awareness/partici- pation in community forest management. A total of twenty-one questionnaires were excluded from the final analysis, as they were incomplete. In some households, we encountered a problem in obtaining the quantity of forest product harvested before the forest was handed over as a community forest because respondents had to recall events from the distant past. However, recall data from household survey was cross-checked by group discussions with both forest users and members of the executive committee, records of the forest users committee, and key-informant interviews. Care was taken to avoid unrealistic estimation because the accuracy of recall information drops when people are asked to remember events in the distant past (Bernard et al., 1984). However, there is still a possibility that recall survey might over estimate (or under estimate) the annual collection by households. In the absence of base-line studies, there were no other methods to check reliability and accuracy of the data, other than crosschecking with other households with similar socio-economic attrib- utes living in close proximity and consensus from the group discussion and key informant interviews. The data concerning the reported change in resource use and livestock and tree nu mbers before and after community forestry were analysed using paired sample t-tests to test the significance of the changes within income groups and household sizes. The ratio of reported forest product collection between community forests and private land was also analysed in addition to graphs, pie charts and histograms to show the resource use pattern by socio-economic characteristics of households. 3. Results The reported number of trees on private land has increased with the introduction of community forestry Fig. 1. Reported change in number of trees on private land. B. Adhikari et al. / Forest Policy and Economics 9 (2007) 464–478 467 when considered both by income group and by household size (Fig. 1). All the changes are significant at the 5% level, apart from the change in the low income group and the smallest household size (Table 1). There are an increasing number of trees from the lowest to the highest income group, with the very low income group having the least number as they own the least land, an average of 0.51 hectares per person. This increased to an average of 0.98 hectares in the high income group. An overwhelming majority of individuals do own land, with all but four respondents stating that they owned some land. The largest households reportedly have the highest number of trees on private land, while households of 5–8 people have the least. The analysis of the amount o f forest product collection before and after the introduction of com- munity forestry in the FUGs surveyed in this study shows that there has been an increase in collection of some key products, specifically the amounts of leaf litter, fodder and grass materials, although the amount of fuel wood has decreased (see Table 2). When the reported changes in forest product collection are considered (Fig. 2), fuel wood collection amounts for all income groups after community forestry vary between 80 and 100 head loads, with less variation between the groups than before the introduction of community forestry. There is considerable variation between different household sizes, with households of 5–8 people reportedly collecting the lowest amount of fuel wood. The amounts collected incre ase for the very low income group, but decrease for all other groups, while they decrease for the smallest house- holds, but increase for households with five or more people. The only significant changes at the 5% level are for the high income group and households with five to eight people. This may seem counterintuitive as the reported change in actual quantity is only 3.5 head loads, but the overall collection rate itself was very low compared to the other households, and therefore there is a significant change in amount of fuel wood collected before and after the introduction of community forestry. Fig. 3 shows that high income groups reportedly collect more leaf litter than the other income groups both before and after community forestry, and the largest households also collect more leaf litter than other households. Again, households of 5–8 people have the lowest reported collection rate. Leaf litter collection within all groups increased with community forestry except within the low income group where it decreased by just over one unit. The only significant changes are those within the very low and middle income groups, and households of five to eight people. The reported amount of fodder, grass and thatching material collected since the introduction of commu- nity forestry is highest in the middle and high income Table 1 Significance of paired sample t-tests for forest product collection and number of livestock before and after Community Forestry Income groups Household sizes Very low Low Middle High 0–4 5–8 N 8 Cattle numbers 0.007 0.002 0.009 0.004 0.012 0.000 0.050 Buffalo numbers 0.036 0.003 0.291 0.039 0.006 0.007 0.052 Goat/sheep numbers 0.000 0.000 0.052 0.001 0.011 0.000 0.001 Total livestock units 0.002 0.000 0.008 0.000 0.009 0.000 0.002 Number of trees on private land 0.022 0.090 0.000 0.000 0.068 0.000 0.000 Fuel wood consumption 0.097 0.197 0.270 0.021 0.429 0.036 0.517 Leaf litter consumption 0.002 0.964 0.032 0.200 0.282 0.004 0.104 Fodder/grass consumption 0.066 0.645 0.004 0.361 0.924 0.215 0.249 Total forest product consumption 0.004 0.621 0.003 0.803 0.743 0.051 0.097 Table 2 Total reported amounts of forest products collected by households before and after the Introduction of Community Forestry (in bhari) Before Community Forestry After Community Forestry Change in amount collected Fuel wood 31,395 29,429 À 1966 Leaf Litter 106,304 124,643 18,339 Grass/fodder 229,803 241,766 11,963 B. Adhikari et al. / Forest Policy and Economics 9 (2007) 464–478468 groups, although the reported amounts collected were fairly equal between all income groups apart from the highest before community forestry was introduced (Fig. 4). The same reported collection pattern between households is again observed, with middle sized households colle cting the least, and the largest households reportedly collecting the most fodder and grass. Reported collection has increased with community forestry apart from in the low and high income groups. None of the changes by household size are significant at the 5% level, while only the change in the middle income group is significant when the respondents are classified by their total income. Fig. 5 shows that reported collection rates for all groups apart from high income groups were similar before community forestry. Since its introduction the middle and high income groups are reportedly collecting more forest products overall. The collection rates between households again show the lowest reported collection for households of 5–8 people. Overall collection of forest products has increased throughout all household groups and all income groups apart from within the high income group, but the only significant changes are those for the very low and middle income groups. There is a general increase in forest production, with these products coming from both community forests and private land. Even though the general trend is towards an increase in forest product collection, there is great varia tion in amounts collected before and after community forestry when considered on an individual basis. One important activity that communities undertake after the handover of forest is that they completely close the forest for considerable period of time to allow for regeneration. A highly protective silvicultural regime, of the sort that is practiced in most CF areas, is more likely to lead to early closure of forest, which might reduce the Fig. 2. Reported change in fuel wood collection before and after community forestry. Fig. 3. Reported change in leaf litter collection before and after community forestry. B. Adhikari et al. / Forest Policy and Economics 9 (2007) 464–478 469 quantity of forest products to be harvested by the local population. Various authors have noted that the restriction posed on collection of various non-timber products after the institutional change actual ly hurting the poorer households whose livelihoods traditionally were closely linked to its collection (Soussan et al, 1998; Richards et al., 1999). In Fig. 6 positive amounts indicate the number of people for whom forest production collection has increased. Negative amounts illustrate the decreases in forest production collection since the introduction of community forestry. The large differences between the maximum and minimum changes in amount o f products collected shows that some individuals have greatly increased the amount they collect, while for others the amount they collect has fallen dramatically. Overall there is a fairly even distribution of individ- uals with an increased and a decreased collection amount. In the fuel wood graph the negative side dominates slightly, in agreement with the slight overall decrease in amount of fuel wood collected. For the other graphs the positive side dominates slightly, again in agreement with the overall increase in collection rates discussed above. Government supported tree-planting activities were one of the major forest rehabilitation initiatives in the early phase of CF program in Nepal. Nursery establishment and distribution of free seedlings were one of the major interventions in the late 1980s. Farmers planted trees on their private land because of available technical support such as free seedlings, planting methods and other extension services avail- able to them. This has resulted in an increase in the number of trees on private land which eases the pressure in adjacent community forests. Fig. 7 shows that slightly over half of the total amount of fuel wood collected comes from private land as does almost three quarters of the fodder, grass Fig. 4. Reported change in fodder/grass collection before and after community forestry. Fig. 5. Reported change in forest product collection. B. Adhikari et al. / Forest Policy and Economics 9 (2007) 464–478470 and thatching material co llected. Leaf litter is pre- dominately collected from the community forest, though this is still only just over half the amount collected in total. This emphasises the role that private land plays in supplying the community with the forest products they need, with over 60% of all forest products coming from private land and the community forest only supplying approximately 37% of the community’s total requirements for forest products. The fact that the number of trees on private land has increased by over 13,000 since the start of community forestry has contributed to the overall increase in the amount of forest products collected. The graphs in Fig. 8 of reported changes in livestock numbers show that numbers of cattle, buffalo and goats and sheep have all decreased with the introduction of community forestry, regardless of the income group of the respondent. All the decreases are statistically significant at the 5% level, apart from the changes in buffalo, goat and sheep numbers for the middle income group. The reported numbers of cattle, goats and sheep owned since the introduction of community forestry increase from the lowest to the highest income group, while numbers of bu ffalo increase from the low to high income group, but the very low income group reportedly owns more buffalo than the low income group. Before community forestry was introduced there did not appear to be any pattern to reported livestock ownership between the income groups. Fig. 6. Reported variation in forest product collection. B. Adhikari et al. / Forest Policy and Economics 9 (2007) 464–478 471 When the graphs of reported changes in livestock numbers against household size are considered (Fig. 9), livestock numbers increase from the smallest to the largest households both before and after community forestry. However, reported numbers have consistently decreased across all households with the introduction of community forestry. All the changes are significant at the 5% level, apart from the decrease in buffalo numbers for the largest households. 4. Discussion 4.1. Rules for forest product collection Rules and regulations for forest product allocation differ considerably from one FUG to another. How- ever, there are many similarities between forest user groups with respect to group functioning. Green product harvesting (mainly the green cut wood) is a collective activity and is carried out once a year in early February, a slack period for agricultural activ- ities. Although the distribution of green firewood varies between groups, in general it is distributed according to family size. Only a few of the FUGs considered in this study are reported to have an equal distribution system i.e. firewood is distributed equally to all user group members. In most FUGs, dead branches and fallen twigs can be collected throughout the year for domestic use. However, green firewood collection and animal grazing are prohibited in all community forests. Tree fodder collection in CF occurs primarily at the end of the dry season. Grass fodder can be collected throughout the year. Collec- tion of leaf litter from CF for animal bedding and mulching is a common practice in most of the FUGs surveyed in this study. NTFPs collection for subsistence needs, especially herbs and other medicinal plants, was also common practice under traditional systems of forest manage- ment. However, most of the FUGs do not allow the Fig. 7. Reported forest product collection from private land. B. Adhikari et al. / Forest Policy and Economics 9 (2007) 464–478472 individual collection of NTFPs for commercial purposes. In most cases, private collector parties are given exclusive collection rights. The income generated from such activities goes directly to the community fund which is later invested into community development activities such as irrigation canal construction and maintenance, drinking water schemes, providing small loans and initiating other activities. As already mentioned there is no restriction on collecting dead branches and twigs which constitutes a major portion of fuel wood collected by house- holds from community forests. Harvesting forest products is labour-intensive because people have to walk, search and spend time harvesting the products. Household size has a direct influence on the capacity of a household to harvest forest products when there is very low level of FUG restrictions on collecting products. In this case households with more labour would be expected to collect more products when there no restrictions on collecting certain products. Most households also collect forest products from their private land or private forests. In the case of community forests, the land is owned by the state but management and utilization are assigned to the local communities. In private forests, land and tree owner- ship rests with the private owners. Nearly all sample households with private land holdings (either khet or bari) have grown some trees on their private land. Most commonly these trees were grown around houses on the borders of househo ld land and on non-irrigated (bari) land. Very few trees were grown on irrigated land (khet) due to the high value placed on productivity of this land and the unwillingness to reduce yields through competition or shading from trees. These trees were usually species valued for their fodder, timber or fruit. Fig. 8. Reported change in livestock numbers by income group. B. Adhikari et al. / Forest Policy and Economics 9 (2007) 464–478 473 [...]... low income group now receiving more fuel wood from the community forest than they did previously through additional amounts they are permitted to collect from the community forest However, the allocation level from the forest is lower than the amount other groups previously collected from the forest Even though this amount is supplemented with fuel wood from private sources (Malla et al., 2003), the... collection from the community forest This is supported by the fact that the majority of leaf litter collection is from the community forest, rather than from private land The variation in reported fuel wood collection shows no trends between income groups, possibly because individuals are collecting their allocated amount from the community forest, and supplementing this amount with fuel wood from their... Reference to the Community Forestry Programme in Nepal, A Working Review, Nepal UK Community Forestry Project, Kathmandu, Nepal Das, J., 2000 Institutions and incentives in a Garhwal village: I Common property regimes in traditional societies Economic and Political Weekly, 4337 – 4344 (December 2) 478 B Adhikari et al / Forest Policy and Economics 9 (2007) 464–478 Dev, O.P., Yadav, N.P., Springate-Baginski,... this paper was produced The paper benefited from the comments of two anonymous reviewers References Adhikari, B., 2003 Property Rights and Natural Resource: SocioEconomic Heterogeneity and Distributional Implication of Common Property Resource Management SANDEE Working Paper No 1-03 South Asian Network for Development and Environmental Economics Kathmandu, Nepal Adhikari, B., 2004 Poverty, property rights... on private land must also be maintained to ensure that the proportion of litter collected from this source can also be sustained The collection of fodder, grass and thatching material has increased slightly, apart from the low and high income groups, with a large majority (73%) of these products being collected from people’s own private land This is consistent with the increase in the number of trees... 2003), the increase in the number of trees on private land for all income groups does not appear to have provided a sufficient number B Adhikari et al / Forest Policy and Economics 9 (2007) 464–478 of replacement trees to compensate for controlled fuel wood collection from the community forest The picture changes when variation in fuel wood collection is considered by household size In this case fuel... Forestry Processes Between 1994–1998 in the Koshi Hills Districts of the NUKCFP Nepal UK Community Forestry Project Malla, Y.B., 2000 Impact of community forestry policy on rural livelihoods and food security in Nepal Unasylva 51 (3), 37 – 45 Malla, Y.B., Neupane, H.R., Branney, R.J., 2003 Why aren’t poor people benefiting more from community forestry? Journal of Forest and Livelihood 3 (1), 78 – 90 Masozera,... litter collected to increase, notwithstanding the controlled access regimes of the community forests The data from this study also show that the poorest groups within the community are not consistently accessing less forest products than they did before community forestry as stated in other studies (Adhikari, 2003, 2004) Collection of leaf litter, grass and fodder is generally lower in the lowest two income... collected (Locke, 1999; Springate-Baginski et al., 1999; Soussan et al., 1998) This is especially true for poorer households, who are unable to replace the resources they did obtain from government forests with resources from private land, as they own little or no land themselves (Pokharel and Nurse, 2004) However, our findings support other studies that have found that once FUGs have been established... the least amount of forest products This is unexpected as it was anticipated that amount of forest products collected would increase from the smallest to the largest households due to their higher demands for fuel wood for cooking, and the increases in livestock ownership from the smallest to largest households, leading to increased need for fodder and leaf litter A possible explanation is that forest . Local benefits from community forests in the middle hills of Nepal Bhim Adhikari, Frances Williams, Jon C. Lovett * Centre for Ecology,. that local communities are able to manage resources efficiently, equitably, and sustainably has led to the devolution o f natural resource management from centralized government control to local. to local user groups in Nepal and other South Asian countries. One particular form of CBNRM is community forestry (CF) which was introduc ed in the 1970s in Nepal so that local communities could