1

Caste and Social Justice in Common Property

Forest Management in Nepal•

Abstract: This paper presents an empirical assessment of the role caste, and other household socio

economic characteristics, on both collective action and use of common property resources

in Nepal. Primary data are drawn from a survey of 308 households belonging to 8

different forest user groups in the mid hills of Nepal. Econometric evidence shows that

households belonging to the lower caste groups have limited access to the decision-

making body of forest user groups. These results indicate that the empowerment and

increasing the participation of the lower caste and marginalized groups within the

community forestry program are critical to increase their access to local level decision-

making body and achieving social equity objective of community forestry program.

JEL classification: 012,013,Q23

Keywords: Nepal, Caste, Common Property, Forest

1. Introduction

• We would like to thank Dr E Somanathan for very useful discussion.

2

Recently a flourishing body of literature has suggested that community management of

common pool resources (CPRs) can be a viable way to achieve environmental, economic

and social development goals (Ostrom 1990, Baland and Platteau 1996). A large

theoretical literature shows that communities can develop mechanisms limiting extraction

from CPRs (see Sethi and Somanathan 1996 for a discussion) and environmental

conferences such as the 1992 Earth Summit and the United Nations Conference on

Environment and Development have taken the position that “sustainable development”

requires community management of resources (Leach, et al 1999). Evidence of successful

self-governance of natural resources by the users themselves has engendered considerable

optimism that turning responsibility over to organized user groups at the local level will

improve the efficiency, equity, and sustainability of the resource base (Meizen-Dick and

Knox, 1999). An increasing number of scholars advocate that decentralized collective

management of CPRs by their users could be an appropriate system for overcoming the

'tragedy of the commons' situation (Ostrom, 1990; Berkes, 1989; Wade, 1988; Jodha,

1986; Baland and Platteau, 1996).

Recognizing the high potential of community-based approaches to natural resource

management (NRM), the past decade has witnessed an increasing emphasis on

community-based resource management, with a realization that local communities are

better equipped to manage these crucial resources in a sustainable way. In Nepal, as in

other developing countries, the development of community-based resource management

has led to devolution of forest management from centralized government control to local

user groups. In the case of community forestry (CF) in Nepal, local forest user groups

3

(FUGs) retain the forest management and utilization under a community-based property

rights regime which has become the main operational strategy for the CF program since

the late nineties. This process has been continuing since the mid-1980s when national

forests were progressively handed over to FUGs. Under this program, local villagers

came together as a FUG and the government handed over certain sections of national

forests as community forests, in which the community had traditional use rights.

While the implementation of the CF programme has succeeded in halting the ongoing

trend of deforestation, evidence on equity and distributional benefits from CF

management is scant in quantity and somehow contentious (Richards et al., 1999;

Springate-Baginski et al., 1999; Adhikari, 2005, Adhikari et al., 2004; Kumar, 2002).

Researchers have described the positive impacts of the CF programme on forest

regeneration and improvement of the biophysical condition of forests, but at the same

time they have questioned the equity and welfare implications of this institutional change

(Branny and Yadav, 1998, Das, 2000; Kumar, 2002). Recently, international donors,

government policy makers and academics are increasingly concerned as to why CF

programs do not have the same level of success and desired impact on the livelihood of

the poorest sections of the community for whom the programmes were targeted.

Socially constructed differences can lead to inequitable access to decision-making and

participation in local level collection action in many villages of the developing world

where local communities are more heterogeneous. In Nepal, the most contested issue

with regard to equity aspects of collective action is the role of caste. Indeed, there is an

4

increasing realisation that scheduled and lower caste groups are disproportionately

represented in the decision-making process. This results in marginalization of these

communities from both social and economic point of view.

The caste system and its attendant practices have been outlawed and declared punishable

offences long way back, but these laws are difficult to implement. According to Article

11(4) of the Constitution of the Kingdom of Nepal 1990, "no person shall be

discriminated as untouchable or debarred from attending public places or using things of

public utility on the basis of caste. Such an act shall be punishable by law." Further, there

is a strong national debate is going on to bring lower caste into the mainstream national

politics after the recent political moment. But Janajatis (various ethnic groups) and

lower-caste people such as Dalits still live in conditions of great poverty and social

disadvantage, forming the most impoverished segment of the society. Although

untouchability exists in its extreme form especially in the hills and mountains but the

sense of untouchability has been found less in the urban areas of the country.

With regard to management of community property resources in Nepal, gender and caste

debate is increasingly being the focus of the discourse in empowerment, poverty

alleviation and natural resource management. About, 42 per cent of the population of the

country is below poverty line; of them 35 per cent belongs to untouchable caste groups.

In his study aimed to analyze community-based organization (CBOs) and local

democracy in Nepal, Dhakal (2006) pointed out that, ‘the males and high caste groups

are, for instance, dominant in terms of number and influence in CBOs compared to low

5

caste groups and women and that the CBOs, which are usually considered to be a key

mechanism for social inclusion, can be liable to elite capture.’ Although a number of

development interventions try to address the issues of caste and social exclusion, the

access to broader political and economic system by socially and economically backward

communities in Nepal is still questionable. In the context of natural resource

management, the debate is that households belonging to higher caste groups often have

privileged access to village leadership and decision making process of the local

commons. Further, power relations within the community and how they are contested and

reproduced over time is getting proper attention when it comes about access to lower

caste households to the resource at the community level (Nightingale, 2002).

In this paper we address the issue of caste and exclusion in collective action using a case

study of eight different community forest users groups (FUGs) in the middle hills of

Nepal. The paper is raising the following questions for empirical analysis. Is there any

relationship between lower caste and other household socio-economic characteristics

(i.e., education and wealth status) on the probability of being elected as a member of the

FUG executive committee (forest management decision-making body)? We, therefore,

propose to integrate the available knowledge with certain additional variables which have

local and contextual significance in order to explore how attributes of households, with

focus on caste, influence the access and the management of the local community forests.

To study the role of lower caste (and other socio-economic variables) in determining the

composition of the executive committee of the CF, we estimate a probit model to assess

6

the role of lower caste and the other explanatory variables on the probability of becoming

member of the executive committee of the CF. Besides the focus on caste, we control for

the role of participation of the household in community gatherings. We, thus, included in

the set of explanatory variables the number of attended meetings regarding community

development issues at the village level.

The paper is organized as follows. In the following section, we provide some accounts of

caste and access to decision-making with special emphasis on community-based

development intervention and resource management. In section three, we discuss

methodological issues, survey design and data collection. Section four reports on the

econometric models and section five provides the results obtained from the analysis.

Finally, the paper concludes in section six with major conclusions and policy

implications.

2. Caste, Forest User Groups and Access to Decision Making

Like other parts of south Asia, Nepal also demonstrates a distinct social stratification

based on the caste system and encompasses a wider socio-economic heterogeneity

(Adhikari and Lovett, 2006). The dogma of the caste system though, still remains a

contentious issue, continues to act as an obstacle in the development of indigenous

communities belonging to the lower social class. The caste system in Nepal can be traced

back to the migration immigrants from the Gangetic plains and the ascendancy of the

7

Hindu religion. These happened during the early Christian era. The system has been

modified to accept within its fold many ethnically Tibeto-Burman groups that were

indigenous to the region apart from the Indo-Aryan speakers from the plains. An

important part of the caste system is that lower caste people are considered to be

untouchable and they face a multitude of disadvantages in land endowment, socio-

economic marginalisation, and participation in decision-making process and employment

opportunities (Lawati, 2005). Usually, there is no marital relation with untouchable, and

the upper castes do not allow them to enter their house. In the western part of Nepal, they

also practice sprinkling water (act of purification) when the upper caste touches someone

belongs to untouchable caste. Further, the low-caste people are deprived of utilizing most

of the temples, funeral places, drinking water taps and wells, restaurants, shops and other

public places where their admission is strictly prohibitive.

Outcastes –who treated as untouchables are basically Dalits who are engaged in different

professional activities such as Kamis (Iron-smiths), Sunars (Gold-smiths), Lohars (Iron

smiths), Vishwakarmas (V.K.), Sobh, Snehi Nepali, etc. They have been named after the

professions they are found to be engaged in. For instance, one who is engaged in works

related to gold is a Sonar; the person who makes things out of copper is a Tamta; on who

is engaged in iron-related works is a Lohar, one who is engaged in making utensils from

wood is a Chunara; one who constructs good from bamboo is called a Parki; cloth

weaver a Koli and Sarki for the person engaged in leather work. Pradhan and Shrestha

(2005) posit that there is no agreement concerning the exact number of Dalit castes in

Nepal making it difficult to estimate the exact population. For example, the National

Dalit Commission lists 28 Dalit castes, whereas the 2001 Census lists only 16 Dalit

8

castes. The largest Dalit group is Kami (blacksmiths) with 30% of the Dalit population,

followed by Damai (tailor cum musicians) 13%, Sarki (cobblers) 11%, and Chamar

(sweepers) 9% (Pradhan and Shrestha, 2005).

Caste and social dominance generally govern the beliefs and practices about rights and

responsibilities, powers and privileges with respect to the different resource management

activities (Singh, 2004). It should be noted that decisions related to access, distribution,

management, and responsibilities with regard to local level resource management are

rooted in social and cultural institutional mechanisms (Bhatia, 1997).

Caste discrimination is being a strong determining factor of exclusion in various forms of

collective action. For instance, in his study of 39 traditional canal systems in India, Baker

(1997) concluded that upper-caste farmers are always in better position to use more water

as they are located at the head-end of irrigation water supply. Further, Tiwary (2006)

systematically explored water resource access structure and highlighted the cumulative

inequality that scheduled caste group’s face in rural India. Caste does not only matter for

participating in collective action but also other economic activities such land ownership

and asset formation. For example, lower caste people from the Terai region (Madhesis) of

Nepal face difficulties in getting their citizenship certificates, without which they are

unable to buy land, get loans from banks, and apply for government jobs. Pradhan and

Shrestha (2005, pp. 15) pointed out that, ‘while the Janjatis, in general, do not face such

discriminations, they, like the Dalits and Madhesis, experience political and cultural

exclusion, which in many cases translates into economic exclusion. Representation of

9

lower caste people and ethnic groups in the judiciary in Nepal is just 2.4 % and the civil

service is about 2.3% (Pradhan and Shrestha, 2005). Further, only 15.15% of the top

party leaders are Janjatis, and only 15.76% are Madhesi and there are no Dalits among

the party leaders . The disadvantages of ethnic and minority groups is also reflected in

the human development index (HDI) as hill castes, such as the Brahmins and Chhetris

and some better off ethnic groups, such as the Newars, Thakalis, and Byasis, have a

higher HDI than most other ethnic groups and Dalits. Further as Dalits have little

decision-making powers, attend fewer trainings and meetings, and are not able to capture

benefits from various community development activities such as credit or savings

schemes (Pradhan and Shrestha, 2005).

Given the socio-economic hierarchy posed by caste inequality, poorer households usually

affiliated to lower caste groups do not necessarily benefit as much as the relatively richer

households from higher caste background in deriving benefits from local development

intervention. For instance, ethnic minorities and indigenous people, who represent a

significant number of landless households marred with poverty, have tended to lag

behind in benefiting from any community-based development initiatives in Nepal.

Although there is a gradual breaking down of social barriers between higher and lower

castes, it still remains influential despite the fact that this is officially declared non-

existent. Although the caste system is strong social institution both in Nepal and India,

however there is a difference when comparing caste systems in these two countries. For

instance, India has a much stronger affirmative action program and preferential politics as

the government provides reservations for government jobs and of university seats in

10

programs of higher education for people hailing from scheduled castes, scheduled tribes

and other backward communities. As a result, the issue of case is much more politicized

and perhaps a little less discriminatory. Nonetheless, even with the rapid urbanization

and education of India's largely rural, agrarian population, the effects of caste are slowly

being eradicated, but meet up with stiff resistance in many parts of the country.

As discussed earlier, one of the concerns in community-based management of forests in

South Asia is the contentious debate about the level of representation of lower caste

households in the decision making authority. A caste with numerical preponderance,

occupying a comparatively higher position in the local ritual hierarchy, possessing a

greater share of the village land and also having greater contacts with the outside world is

generally regarded as ‘dominant’ in the community (Singh, 2004; Srinivas, 1959).

Beteille (1983) points out that in Indian villages access to local common-property

resources is often restricted to privileged caste groups. The outcasts or schedule castes are

often among the poorest of the poor and are frequently excluded in the decision making

process. With regard to joint forest management in India, Hildyard et al (1998) argued

the structural dominance enjoyed by more powerful groups in joint forest management is

very evident. They argue that participation requires wider processes of social

transformation and structural change to the system of social relations through which

inequalities are reproduced and that without this transformation marginalized caste

groups could not get enhanced access to the resource base.

11

Graner (1997) argued that the shift in forest management from state control to

community ownership in Nepal, has not necessarily helped poor people belonging to

lower caste groups, but have often worked to their disadvantage. In many cases, the poor

are not included in the forest management decision-making body (i.e. forest users

committee), as fuel wood sellers and other occupational households are not effectively

represented in the operational regime. Adhikari’s (2003) study reveals that the nature of

representation in the executive committee of the FUG is more biased towards the better

off as only 15.7 per cent female were represented in the FUG committee and the rest of

are all men usually from the upper caste. Thapa et al. (1998) posits that even if a person

from the lower caste group is represented on the committee, they are often not listened to

and have less bargaining power at the community meetings and assemblies. In terms of

resource exploitation, Adhikari et al (2004) observed that lower caste households are

benefiting less from community forests than households belonging to higher caste groups.

Singh (2004) noted that the dominant caste generally leads in regulating water

management affairs in India. In his study on power, equity and conflicts in South Asia,

Bhatia (1997) pointed out that, ‘while some conflicts in community-based resource

management are caused by disagreements related to access and distribution of

community forestry resources, in many cases the causes are actually based on class, caste,

gender, and power but manifest themselves in the guise of community forestry’.

However, this position is also contested by other scholars. For instance, a study

conducted by Sharma (2002) concluded that the use of forest products from the

community forests by wealth and caste fails to reveal any discrimination against the poor

12

or lower caste people. Sharma further argued that community forestry still remains a

viable development strategy for securing basic needs as it can contribute to poverty

reduction because the poor and low caste households have easy access to forest products.

Pokharel (1997) also argued that community forestry has become a vehicle in bringing

change in social processes empowering poor and disadvantaged members of the

community as the program is able to raise awareness among disadvantaged and

marginalized members of the group in matters related to inequality, social injustice and

their exclusion from social and political processes including the benefits from

mainstream development. He further argued that, ‘community forestry has made a

significant contribution to increasing people's participation in many marginalized sections

of the community because their representation has increased in users committees and

many of them have begun to voice their concern in meetings and assemblies’. In this

context it is interesting to examine how the caste system influences the participation of

rural households in managing the local commons especially how caste influences access

to resources and leadership status of households in the regulated form of common

property resource management

3. Survey and Variables

Data are drawn for a survey conducted in 2000 in two districts in the middle hills of

Nepal, Kavre Palanchok and Sindhu Palanchowk. The middle hills comprise the central

area of Nepal with a mixture of agricultural and forested land. The majority of the

population in this area are subsistence farmers, depending on the surrounding agricultural

and forested land, with livestock playing an important part in their livelihood options

13

(Adhikari, 2003). The middle-hills run from east to west across the center of the country,

sandwiched between the low laying Gangetic plains (Terai) and the snow-capped

Himalayan Mountains. Altitude in the middle-hills ranges from 300 m in river valleys to

5,000 m on hill tops. Land uses in this region are categorized as cultivated land, non-

cultivated inclusions, grasslands, forestland, shrub lands and other types of land use.

Community forestry has been in existence in the middle hills for up to twenty years, and

therefore provides a suitable environment in which to study the resource use patterns by

gender and caste since the introduction of the forestry system has been in existence for

some time. Since there is a fair degree of variability within the two districts in terms of

bio-physical and socio-economic characteristics, deliberate selection of sites was

considered likely to yield more information on the nature of project impact for a

minimum level of study resources than would a strictly random selection of sites (Collett

et al., 1996).

While villages formed the second stage of sampling, selection of FUGs formed the third

stage of the sampling frame. The next task was to select forest user groups (FUGs) from

the two districts under consideration. In order to address the research questions posed in

the previous section, four Forest User Groups (Saradadevi FUG, Jyala Chiti FUG,

Mahavedsthan FUC and Thuli Ban FUG) in Kavre Palanchok districts and four FUGs

(Gaurati FUG, Shree Chhap FUG, Janghare FUG and Karki Tar FUG) in Sindhu

Palanchowk district were selected.

14

Since the main focus of this study was to assess the use of forest products by households

and assess farm and non-farm endowments of households, households are the final

sampling units in the sampling design. Within these FUGs a sample of households was

selected for interview based on their wealth ranking. This wealth ranking was

constructed after carrying out a participatory rural appraisal exercise within each village

that ranked households according to criteria that the villagers themselves considered

important in assessing a household’s position within the village (Adhikari, 2003). Key

to the ranking was the amount of land owned, the amount of off-farm income, food

sufficiency and the quality of land within each household. Households were therefore

selected for interview to ensure that a representative sample of households from each

wealth group was chosen. A total of 330 interviews were conducted, with twenty one

questionnaires being excluded from the final analysis because they were incomplete.

The survey focused on obtaining demographic information, land holding and tenure

information, use and management of the community forest and awareness and more

importantly participation and membership of households according to the caste of sample

households. We concentrated more on the household’s level of participation in the

management and implementation of the community forest.

The Tole (hamlet) level meetings were held separately for men and women in a smaller

and homogenous group to check their level of participation in the forest users committee.

This helps to arrive at a consensus on the perceptions of the group on issues like

community forestry, their awareness of and participation in the functioning of the FUG,

15

role and membership of the committee and operational rules of the FUG, management of

the forest, distribution of products, and management of community funds and so on.

Group discussion was also helpful to discuss and define a consensus on the perceptions of

the group. Gender issues, benefit and cost sharing practices, decision making processes,

participation and representation in users’ committee, caste and gender representation in

users’ committee (apex body of FUG), nature of existing conflicts, ownership over

agreed upon decisions, and institutional procedure were assessed by group discussion. In

general, the group meetings sought to uncover the specific resource needs and opinions of

women, occupational castes and ethnic minorities including issues of welfare, equity, and

each group’s participation in forest resource management.

[TABLE 1 – ABOUT HERE]

From the survey a set of variables was selected for inclusion in the econometric model.

The variables used in this analysis are summarized in Table 1 which defines the

explanatory variables incorporated in the econometric analysis. Table 2 reports the

summary statistics.

[TABLE 2 – ABOUT HERE]

4. Analysis

To analyze the role of lower caste status, other socio economic variables and the level of

participation on both the probability of becoming a member of the FUG committee , we

estimates the role of the set of explanatory variables in determining the probability of that

specific household having one of their members in the FUG committee. We will refer to

16

this equation as the “membership equation.” Therefore, socio economic characteristics

(rich, education, age and lower caste), participation and the labour units per household

are also assumed to be determinants of the memberships in the common property forest.

The membership equation is given by

Mi= wi′δ+ηI (1)

Where Mi is a binary variable that is equal to 1 if the household has a member in the FUG

committee and 0 otherwise, wi is a vector of covariates for household i, δ is a vector of

coefficients for the membership equation and η is the random disturbance.

We assumed that the extraction equation is linear. To test for the presence of collinearity,

the computation of the variance inflation factors (VIF) was implemented. We found that

the VIF values of all the explanatory variables were between 1.03 and 1.2. Thus, no

evidence of collinearity was found.

5. Empirical results

Table 4 reports the econometric results for both the membership equation and the

extraction equation1. The former is specified as follows:

Prob [Memtype=1] = F (constant, lower caste, rich, age, education, participation,

labour) (2)

17

. Heterogeneity can lay an important role in this analysis (Edmonds, 2002). To control for

institutional and other unobservable forest user groups effects on fuelwood collection and

FUG membership we estimated the models both with and without fixed effects. To

appreciate the qualitative difference between these two models, we present in table 3 both

results. The Huber/White/sandwich estimator of variance was used. Further, for the 308

households belonging to the 8 different FUGs that manage a specific forest, we allowed

observations to be not independent within group (although they still are independent

between groups). Therefore, the observations are independent across groups (clusters) but

not necessarily independent within the group. From the inspection of the latter we found

that the probability of becoming a member of the FUG committee is sensitive to socio

economic variables. The estimated coefficient for lower caste is negative and statistically

significant (at 1% level). This implies that belonging to the caste of the untouchables is

negatively correlated with the probability of being elected as a member of the FUG

committee. This supports the argument that untouchable households are marginalized and

are not part of the decision making and governing body of the forest user groups.

Relatively better off households, instead, have a positive estimated coefficient

(statistically significant at the 5% level). This observation is similar to that of Singh

(2004) who argued that leadership in water management issues is exercised by the senior

group from the dominant caste in India.

More educated households are also more likely to become part of the FUG committee.

Indeed, the estimated coefficient for education is positive and statistically significant at

the 1% level. Age, instead, negatively affects the probability of membership. However,

18

the estimated coefficient is statistically not significant. The level of participation in

community meeting seems to play an important role in determining the probability of

FUG membership. Indeed, the estimated coefficient is positive and highly significant.

Households that allocate time to participate to community meetings that relates to the

forest management, have better chances to become part of the decision making body.

Labor units per household, instead has been found highly not significant.

The calculation of the marginal effects highlighted the relative importance of the

variables lower caste and rich. The former displayed a marginal effect equal to -0.1 and

the latter equal to 0.07. The marginal effect calculated for the variable education was

0.075. Therefore, among the socioeconomic characteristics, lower caste the largest

impact on the probability of membership in the FUG. The level of participation has been

found to have the largest impact of all the explanatory variables (0.16).This result stresses

the important of household participation in community meeting.

[TABLE 3 – ABOUT HERE]

5. Conclusions and Policy Implications

Community-based management of common pool resources has been gaining momentum

in many developing countries due to their contribution on rural livelihoods and

environmental conservation. However, there is a strong debate among scholars as to

whether collective action is more equitable in heterogeneous society with a number of

socio-economic differences among actors and that these types of cooperative action

19

empower the marginalized and poorer sections of the society in question. This paper

focused on some of these issues related to collective action taking a case study of

community forestry in Nepal.

Forest is considered to be crucial resource for livelihood security of the poor. Collective

action towards management of forest resources is legitimized by the government by

providing forest management ownership and authority to the local people. In Nepal, it

was expected that community-based management could reconcile the goals of social

justice, equity, development, empowerment and sustainability of forest resources. Policy

documents related to community forest explicitly recognize social equity and poverty

alleviation as fundamental objectives. Although the CF program in Nepal has been able

to meet some of these requirements (such as improvising the resource base, watershed

management and supply of forest products to the rural households), access to decision

making by lower caste and marginalized households has not been a universal result.

This study reveals that caste is playing an important role in local level collective such as

being elected as a member of governing body of FUG. Econometric evidence shows that

while richer and more educated households are more likely to be part of the governing

body the untouchables are marginalized. Their participation in the FUG committee and

the decision making process is less likely.

These results indicate that institutions of CF management are so far unable to recognize

that there are different types of resource users having different economic and social

20

status, perspectives, knowledge systems, values, understanding and objectives (Anderson

et al., 1998). CF intervention, at least in the sites covered by this study, has not taken into

account the issue of caste although there are few initiatives to increase the women and

marginalized caste into the decision-making body. Although the program itself has

emphasized the provision for equality in opportunity in all aspects of management, the

decision-making process as well as in sharing cost and benefits. Nevertheless, given the

inherent socio-economic inequality in village communities, the equality of opportunity as

envisioned by the program did not necessarily reflect equitable outcomes.

Local elites are found to be advantaged in both accessing the decision making committee

and extracting benefits from the forest. The overall observation is that despite the

enormous potential of CF to significantly improve livelihoods and poverty reduction in

the Mid-Hills, the poverty alleviation agenda set by the government through the CF

program is unlikely to be achieved unless intervention related to empowering and

increasing the participation of the lower caste and marginalized group within the

community forestry program are properly materialized. Despite its focus on forest

management in Nepal, this study can contribute to the broader ongoing debate on

common pool resources. Indeed, the results stress the important role that lower caste and

local elites may have. Caste inequality, as well as other socioeconomic characteristics,

could undermine the success of collective action unless local institutions are strong

enough to tackle inequity in access and opportunities at the local level.

21

References

Adhikari, B. 2005.Poverty, Property Rights and Natural Resource: Understanding Distributional

Implications of Common Property Resource Management. Environment and Development

Economics, Vol 10, pp. 7-31.

Adhikari, B., Di Falco, and J.C Lovett (2004) Household Characteristics and Forest Dependency:

Evidence from Community-Based Forest Management in Nepal. Ecological Economics, Vol. 48

(2) pp. 245-57

Adhikari, B. J.C. Lovett (2006) Institutions and Collective Action: Does Heterogeneity Hinders

Community-Based Resource Management? Forthcoming Journal of Development Studies.

Volume 78, Issue 1, Pages 5-15

Anderson, J., J. Clement, and L.V. Crowder (1998) “Accommodating Conflicting Interests in

Forestry: Concepts Emerging from Pluralism”. Unasylva, Vol. 194 (49) pp. 3-10.

Baker, M. (1997) Persistence, Transformation and Demise within the gravity flow Irrigation

Systems (Kuhls) of Kangra Valley, Himachal Pradesh, India. Workshop on `Co-operative

Management of Water Resources in South Asia" Centre for India and South Asia Research

(CISA), University of British Columbia, in Vancouver, Canada, December 15-17, 1997.

Baland, J. and J. Platteau (1996) “Halting Degradation of Natural Resources: Is there a Role of

Rural Communities?” New York: FAO, Oxford University Press.

Berkes (ed) (1989) “Common Property Resources - Ecology and Community Based Sustainable

Development”. London: Belhaven Press.

Beck, T. and C. Nesmith (2001) “Building on Poor People’s Capacities: The Case of Common

Property Resources in India and West Africa”. World Development, Vol. 29 (1) pp. 119-33.

22

Byron, R. N. (1991) “Cost-Benefit Analysis and Community Forestry Project”. In D. A. Gilmour

and R. J. Fisher (eds.) Village, Forests and Foresters. Sahayogi Press, Kathmandu, pp 163-80.

Bhatia, A. (1997) Power, Equity, Gender, and Conflicts in Common Property Resources in the

Hindu Kush-Himalayas. Issues in Mountain Development (1997/7), ICIMOD, Kathmandu,

Nepal.

Branney, P., G.C. King & Y.B. Malla (1994) “Mid-term Review of Nepal’s Hill Community

Forestry Project. Washington, DC. USA. World Bank.

Bromley, D.W. (1991) “Environment and Economy: Property Rights and Public Policy”. Oxford:

Basil Blackwell.

Bromley, D.W. and D.P. Chapagain (1984) “The Village against the Centre: Resource Depletion

in South Asia”. American Journal of Agricultural Economics, Vol. 66, pp. 868-73.

CPFD/ HMGN Database (2000) Department of Forest, Kathmandu, Nepal.

Collett, G., R. Chhetri, W.J. Jackson, and K.R. Shepherd (1996) “NACFP: Socio-Economic

Impact Study”. ANUTECH Pty Ltd, Canberra, ACT, Australia

Cooke, P. (1996) “The Effects of Environment Goods Scarcity on Non-Farm Labour Allocation”.

Environment and Development Economics, Vol. 3, pp. 443-69.

Das, J. (2000) “Institutions and Incentives in a Garhwal Village – I: Common Property Regimes

in Traditional Societies”. Economic and Political Weekly, December 2, pp. 4337-4344.

Dasgupta, P. (1993) “An Inquiry into Well-Being and Destitution”. Oxford: Clarendon Press.

Dhakal, S. (2006) Politics beyond the Political Sphere: Community Based Organizations and

Local Democracy. Proceedings of the 19th European Conference on Modern South Asian Studies

(ECMSAS), 27–30 June 2006, Leiden, Amsterdam, the Netherlands.

23

Gauld, R. (2000) “Forestry: Construction in the Philippines”. Development and Change, Vol. 31

pp. 229-254.

Graner, E., 1997. The Political Ecology of Community Forestry in Nepal. Saarbrucken, Verlag

fur Entwickungspolitik.

Gordon, H.S. (1954) “The Economic Theory of a Common Pool Resource: The Fishery”. Journal

of Political Economy, Vol 62, pp. 124-42.

Hildyard, N., Hegde, P., Wolverkamp, P., and Reddy, S. (1998) Same Platform, Different Train:

The Politics of Participation. Corner House Briefing 04, The Corner House Station Road

Sturminster Newton Dorset DT10 1YJ UK

Jodha, N.S. (1986) “Common Property Resources and the Rural Poor in Dry Regions of India”.

Economic and Political Weekly (EPW). Vol .21 (27), pp. 169-181.

Kumar, S (2002) “Does “Participation” in Common Pool Resource Management Help the Poor?

A Social Cost-Benefit Analysis of Joint Forest Management in Jharkhand, India”, World

Development, Vol. 30 (5) pp. 763-782.

Lawati, M., (2005) Towards A Democratic Nepal Inclusive Political Institutions for a

Multicultural Society, Sage Publication, New Delhi

Mahat, T.B.S., D.M. Griffin and K.R. Shepherd (1987) “Human Impacts on Some Forests in the

Middle Hills of Nepal, Part 3: Forests in the Subsistence Economy of Sindhupalchowk and

Kavrepalanchowk”. Mountain Research and Development, Vol 7, pp. 111-134.

Meinzen-Dick, R. and A. Knox (1999) “Collective Action, Property Rights and Devolution of

Natural Resource Management: A Conceptual Framework”. Paper Presented at the International

Workshop on Collective Action, Property Rights and Devolution of Natural Resource

Management, Exchange of Knowledge and Implications for Policy, June 21-25, Philippines.

24

http://www.cgiar.org/capri/status.htm#devolution

Ostrom, E. (1990) “Governing the Commons”. Cambridge, UK: Cambridge University Press.

Pokharel, B. (1997) Community forestry and people's livelihoods. Available at

http://www.livelihoods.org/post/Docs/Pokharel.rtf.

Pradhan, R. and A. Shrestha (2005) Ethnic and Caste Diversity: Implications for Development.

Working Paper Series No. 4, Asian Development Bank, Nepal Resident Mission, Kathmandu.

Richards, M., K. Kanel, M. Maharjan and J. Davies (1999) “Towards Participatory Economic

Analysis by Forest User Groups in Nepal”. ODI, Portland House, Stag Place, London, UK.

Sharma, A.R. (2002) Community Forestry from Wealth and Caste Perspective:

Elvira Graner in The Dock. Presented at "The Commons in an Age of Globalisation," the Ninth

Conference of the International Association for the Study of Common Property, Victoria Falls,

Zimbabwe, June 17-21, 2002.

Singh, N. (2004) Water management traditions in rural India: Valuing the unvalued. Department

of Land and Water Resources Engineering, Royal Institute of Technology, Sweden.

Springate-Baginski. O., J.G. Soussan, O.P. Dev, N.P. Yadav and E. Kiff (2000) “Community

Forestry in Nepal: Impacts on Common Property Resource Management”. School of

Environment and Development, Series No. 3, University of Leeds, UK.

Wade, R. (1988) “Village Republics: Economic Conditions for Collective Action in South India”.

Cambridge: Cambridge University Press.

Somanathan, E., Prabhakar, R., and Singh, B. (2002) Collective Action for Forest Conservation:

Does Heterogeneity Matter? Indian Statistical Institute, New Delhi, India.

Srinivas, M. N. 1959. The dominant caste in Rampura. American Anthropologist 61: 1-16

25

Tiwary, R. (2006) Explanations in Resource Inequality: Exploring Scheduled Caste Position in

Water access structure. International Journal of Rural Management, Vol .2 (1): 2: 85-106.

Wyatt-Smith, J. (1982) “The Agricultural System in the Hills of Nepal: The Ratio of Agricultural

to Forest Land and the Problem of Animal Fodder. APPROSC Occasional Paper 1. Agricultural

Projects Services Centre, Kathmandu, Nepal.

26

Table 1 Definition summary statistics for the explanatory variables

Variable Name Variable definition Mean Std Deviation Min Max

Membership Dummy for membership in the committee (if member =1, 0 otherwise)

0.18 0.38 0 1

Lower caste Dummy for lower caste (if untouchable caste =1, 0 otherwise)

0.12 0.32 0 1

Education Average education of adult family members (per capita total number of school years)

3.97 2.3 0.14 10.2

Rich Dummy for wealth (if rich =1, 0 otherwise) 0.29 0.45 0 1

Age Age of the household head 43.7 12.89 22 84

Participation the number of attended meetings regarding community development issues

4.32 5.047 1 36

Labor Days allocated for fuel wood collection per member of the household

1.15 0.77 0.07 5.33

Table 2 Correlation matrix

rich lower caste education age labour participation

Rich 1

lower caste -0.1587 1

education 0.1389 -0.2845 1

Age 0.1805 -0.002 0.2167 1

labour -0.1042 0.1029 -0.2542 -0.1743 1

participation 0.1255 -0.1346 0.0333 0.0006 0.0463 1

27

Table 3 Probit equation results

Variables Coeffs Std Err Coeffs Std err (A)

Without FUG fixed effects

(B) With fixed effects

Lower caste -0.6*** 0.017 -0.79*** 0.1932 Rich 0.32** 0.14 0.45*** 0.1916 Education 0.38*** 0.013 0.0377*** 0.0154 Age -0.3 0.35 0.10 0.171 Labor 0.003 0.21 0.1479 0.195 Participation 0.85*** 0.15 0.86*** 0.179 Constant -0.32 1.38 -1.63*** 0.381

Model (A) Log Likelihood; -91; Wald chi2 (6) = 402.76; Prob > chi2 = 0; Pseudo R2 = 0.33 – Model (B): -91; Pseudo R2 = 0.3536 Significance levels are denoted by one asterisk (*) at the 10 % level, two asterisks (**) at the 5 % level, three asterisks (***) at the 1 percent level ; N=308;Note: Robust standard errors are used.