Environmental Management

, Volume 42, Issue 3, pp 391–401

Evaluating Local Benefits from Conservation in Nepal’s Annapurna Conservation Area

Authors

  • Arian Spiteri
    • Alberta Tourism, Parks and Recreation
    • Department of Recreation, Park, and Tourism SciencesTexas A&M University
Article

DOI: 10.1007/s00267-008-9130-6

Cite this article as:
Spiteri, A. & Nepal, S.K. Environmental Management (2008) 42: 391. doi:10.1007/s00267-008-9130-6

Abstract

Protected areas are integral to the global effort to conserve biodiversity, and, over the past two decades, protected area managers have begun to recognize that conservation objectives are next to impossible to achieve without considering the needs and concerns of local communities. Incentive-based programs (IBPs) have become a favored approach to protected area management, geared at fostering local stewardship by delivering benefits tied to conservation to local people. Effective IBPs require benefits to accrue to and be recognized by those experiencing the greatest consequences as a result of the protected area, and those likely to continue extractive activities if their livelihood needs are compromised. This research examines dispersal of IBP benefits, as perceived by local residents in Nepal’s Annapurna Conservation Area. Results reported here are based on questionnaire interviews with 188 households conducted between September and December 2004. Results indicate that local residents primarily identify benefits from social development activities, provisions for resource extraction, and economic opportunities. Overall, benefits have been dispersed equally to households in villages on and off the main tourist route, and regardless of a household’s participation in tourism. However, benefits are not effectively targeted to poorer residents, those highly dependent on natural resources, and those experiencing the most crop damage and livestock loss from protected wildlife. This article provides several suggestions for improving the delivery of conservation incentives.

Keywords

Annapurna Conservation AreaIncentive-based programsProtected areasConservationNepalLocal communities

Introduction

The global need for biodiversity conservation and the important role protected areas play in achieving this goal are irrefutable; yet, throughout the world, conservation comes at a substantial cost for local people. Particularly for subsistence-based rural communities in developing countries, conservation often translates into social and economic hardships and protected areas are seen as a playground for wealthy urbanites or foreigners (Adams and Infield 2003; Balmford and Whitten 2003). Protected areas that foster such resentment have continually been shown to be unable to curb resource degradation, and, can in fact, accelerate the loss of biodiversity (Colchester 1997; Ghimire and Pimbert 1997; Honey 1999).

In an attempt to address these socio-economic and ecological limitations, protected area management approaches now commonly establish and distribute benefits to local communities (Salafsky and Wollenberg 2000; Bauer 2003; Gadd 2005). Benefits, such as social services, agricultural improvements, and tourism revenue sharing, are intended to reduce resource dependent livelihoods and act as conservation incentives by tying benefits to conservation success (Hutton and Leader-Williams 2003; Spiteri and Nepal 2006). Incentive-based programs (IBPs) have not, however, proven to be a panacea for biodiversity conservation or the wellbeing of rural communities (Barrett and others 2005; Krüger 2005). The blind faith placed on IBPs as a solution to conflicts between protected areas and people in the early 1980’s, has led to widespread use of this concept without fully understanding its potential limitations or requirements for success (Tutin 2002). In fact, recent discourse indicates many IBPs achieve neither conservation nor development goals (Noss and others 1999; Terborgh 1999).

Commonly cited failures in IBPs include the inability to distribute recognizable benefits to those most affected by and most likely to affect conservation activities (Gillingham and Lee 1999; Kiss 2004). Barriers to benefit distribution can be overcome by designing benefits to specifically address the needs, lifestyles, and concerns of target beneficiaries. However, even well designed benefits can remain unobtainable by certain members of a local or regional population. Groups or individuals with elevated social status, such as the wealthy and educated, or urban outsiders, tend to be best positioned to capture IBP benefits at the expense of the less powerful and most likely to experience hardships from conservation (Barrett and Arcese 1995; Brandon 1998; Walpole and Goodwin 2000; Timsina 2003). Concerns have also been raised over regional inequities, since replicating IBPs in all villages throughout a region proves difficult (Bookbinder and others 1998; Josiah 2001; Kiss 2004; Spiteri and Nepal 2006).

By examining local residents’ perceived benefits provided by IBPs in Annapurna Conservation Area (ACA), and local and regional distribution of the benefits, this research provides further insights into the limitations of IBPs in a protected area considered relatively successful at integrating social, economic, and ecological objectives. In particular, two main research questions are addressed in the article: (1) What do residents consider as benefits from conservation?; and (2) How are benefits distributed relative to residents’ perceived costs, participation in tourism, their demographic characteristics, and the proximity of their village to the main tourist routes. The results presented in this article are one component of a much larger study exploring IBPs in Nepal’s protected areas.

This study was conducted in the Himalayan Mountains of north central Nepal in ACA. The elevation in ACA’s 7629 km2 area ranges from 450 meters to over 7000 meters and supports several endangered species, such as snow leopard (Panthera unica), Himalayan black bear (Ursus thibetanus), red panda (Ailurus fulgens), musk deer (Moschus chrysogaster), and clouded leopard (Neofelis nebulosa). Suggestions to conserve the Annapurna region began in the late 1970’s, due to evidence of environmental degradation from population pressures and tourism impacts. The King Mahendra Trust for Nature Conservation (KMTNC) (recently renamed Nepal Trust for Nature Conservation and hereinafter referred to as the Trust), a national nongovernmental organization (NGO), initiated the Annapurna Conservation Area Project (ACAP) in 1986 as a small scale pilot project, at the request of the Nepalese government. Due to the success of the pilot project, the ACA project area was expanded significantly in 1989, and again in 1993 (Nepal 2000). The ACA was the first protected area in Nepal to allow local residents to live within the boundaries and maintain use of natural resources, and it became a new model for protected areas throughout the world (Wells 1994).

ACA is home to over 120,000 people. Most people, particularly in the northern part, are of Buddhist faith, while some Hindus are found along the lower valleys in the southern part. ACA is home also to a significant number of transient population and recent migrants who have moved to the region in search of better livelihood opportunities and employment prospects in tourism. Hindus such as Brahmin and Chetri occupy the upper position in the caste hierarchy, most Buddhist like Gurung and Thakali (who actually practice a mix of Buddhist and Hindu religious customs) occupy the middle position, followed by Kami, Damai who are mainly Hindu but belong to the lowest hierarchy in the Nepalese caste system. The majority of local residents depend on natural resources from the conservation area. This dependency has proven to be beneficial to building support for forest conservation, as many residents are concerned with the future availability of resources for use. Although uncommon, hunting of wildlife by local residents is primarily motivated by retaliation against or protection from wildlife damage (Sherpa and others 1986; Bajracharya and others 2005), as ACAP has no direct compensation program in place to offset the economic burden of wildlife damage.

The ACA continues to be managed by the Trust, with the ultimate goal of building capacity for a transfer of management responsibility to community-based institutions. Projects are determined through a participatory process with ACAP acting primarily as a matchmaker between local people and the expertise or funding of external NGOs or government organizations. Because of the future goal of local management, projects are only initiated if they can be locally maintained when external support is withdrawn. Through active participation, ACAP’s intent is for local people to become custodians of their natural resources so that ACA will be self-regulating and not require the army to enforce rules and regulations—an arrangement unlike those in national parks in the country.

ACA is the most popular trekking destination in the country. The Trust has been authorized to charge and collect entry fees from visitors at a current rate of Nepali Rupees 2000 (approximately US $30) per person, per entry. These fees provide the primary source of financing for operations in the conservation area. Tourism is spatially and temporally concentrated in villages and regions along the main trekking route providing only seasonal employment and economic opportunities (Parker 1997; Nepal and others 2002). Trekking tourism has more than doubled over the past two decades, but the political turmoil in Nepal has contributed to a substantial decline in tourist arrivals since 2001 (HMGN 2003). This decline has not only directly impacted hotels, restaurants, and guide businesses, but has also limited support for conservation and development activities in the protected area, and, in turn, threatens the foundation upon which the success of ACA has been built (Bajracharya and others 2005). In recent years, conservation offices in the southern portion of ACA have been forced to close down due to political instability and threats to employee safety, halting conservation and development activities in these areas (Bajracharya and others 2005).

Methods

Fieldwork was conducted in the villages surrounding the Jomsom Field Base, located in the northwestern region of ACA in the Lower Mustang District (Fig. 1). Political conflict in the ACA limited the selection of the study villages due to concerns over researchers’ safety in other regions. The area with the longest history of ACAP management was occupied by Maoist insurgents and the headquarters were destroyed by bombs in 2003. Although the Trust’s first project in ACA began twenty years ago, ACAP has been active in the Jomsom region for only fourteen years. The project has not been as well received in the northern districts of Mustang and Manang as in the other project regions (KMTNC 1997). Because of the program’s success in the southern regions, awards and evaluations highlighting the achievements of the project are primarily based on research conducted in the south (Parker 1997). This article provides insight into the achievements of IBPs in only the northern region of ACA due to significant differences in culture, topography, programs, and project history between the southern and northern portions of ACA.
https://static-content.springer.com/image/art%3A10.1007%2Fs00267-008-9130-6/MediaObjects/267_2008_9130_Fig1_HTML.gif
Fig. 1

 Map of Annapurna Conservation Area showing the locations of tourist (TV) and distant (DV) villages

Structured questionnaire interviews were conducted with 188 household heads, which provided the primary measure of local perceptions. The purpose of the survey was not only to identify differences in perceptions of IBPs based on a respondent’s demographic characteristics, but also to identify regional differences in perceptions between villages. Therefore, villages included in the study were purposively selected (Ward 1993) to allow comparisons to be made between villages along the main trekking route and promoted in pamphlets distributed by ACAP (i.e., tourist villages or TV), and those that are distantly located from the main trekking route (i.e., distant villages or DV). While DV outnumber TV in the region, DV are comprised of only a few households, and, therefore, more DV were included in the study to ensure a representative sample. Eight TV (Jharkot, Kagbeni, Kalopani, Kinga, Lete, Marpha, Ranipauwa, and Tukuche) and 15 DV villages (Chhairo, Chhayo, Chhongur, Chimang, Dhumba, Jhipra Deurali, Kunjo, Lubra, Naurikot, Parsyang, Polche, Sauru, Taglung, Tiri Gaur, and Titi) were selected within the boundaries of the conservation area. In each village, 20 to 30% of household heads were interviewed, with a total of 103 interviews in TV and 85 in DV villages.

The questionnaire was partially constructed by adapting questions used in previous research on the topic (Nepal and Weber 1993; Mehta and Heinen 2001; Walpole and Goodwin 2001). Questions were mostly structured to allow for quantitative analysis, and compiled information on natural resource use, damages caused by wildlife, perceptions of benefits, and demographic characteristics. Respondents were selected from every other household to ensure the sample was randomly selected. In the absence or unavailability of household heads, another adult in the house was selected. During interviews, two research teams asked the questions and recorded responses, each consisting of one male and one female to prevent the unintentional exclusion of women from the research. Results presented in this study represent the views of household heads only.

Data were analyzed using the Statistical Package for the Social Sciences (SPSS) Version 13.0. Likert scale questions were coded so positive responses corresponded with higher numerical codes. This process allowed for the creation of composite scales for reduced access to natural resources (hereinafter referred to as reduced access), and perceptions of benefit receipt (De Vaus 2002). A composite scale (0–10) reflecting a respondent’s perceived receipt of benefits was created by summing ranked responses from two questions on benefits from the protection of natural resources and conservation-related development activities (De Vaus 2002). Perceived household benefit from tourism was determined by asking respondents if their household benefited from tourism, and then dividing tourism benefit into direct benefits (from employment in the tourism industry) or indirect benefits.

Three variables were used to measure perceived costs experienced by a household as a result of the conservation area: reduced access; livestock losses; and crop losses. Composite scale scores for reduced access were calculated by summing weighted responses to ten questions regarding personal use of (five questions) and desire to use (five questions) specific resources. Measuring crop and livestock losses by monetary value proves unreliable due to inaccurate figures assigned by respondents who cannot remember their exact loss or exaggerate to increase compensation, or simply because livestock and crops intended for household consumption may have no measurable market value. Instead, the reported numbers of wildlife species responsible for crop losses and livestock depredation were used as measures of wildlife costs (after de Boer and Baquete 1998).

Differences between village categories and demographic variables, benefits identified, and level of tourism benefit were explored using chi-square (χ2) tests of independence. Cramer’s V was used to provide an indication of the strength of significant relationships (p < 0.05), with values ranging from 0 (no association) to 1 (perfect association). Relationships between type of tourism benefit and benefit perceptions were examined using Kruskal-Wallis tests. Costs were also individually subjected to bivariate analysis to identify differences between village categories using Mann-Whitney U tests (De Vaus 2002).

The relationships between costs and demographic variables, benefits and demographic variables, and benefits and costs were explored with multivariate statistics. As the data were not normally distributed, logistic regression was used to analyze relationships between multiple variables. Reduced access, livestock losses, crop losses, and benefit perceptions were all used as dependant variables in logistic regression, and, therefore, required transformation into dichotomous variables.

Hierarchical logistic regression was used to explore the significant relationships identified between perceived benefits and each of the costs, and to determine the effect of village category on the relationship. Hierarchical logistic regression is useful to determine if other independent variables account for a relationship between a dependent and independent variable. With the effect of village determined and controlled, a third block was entered to explore potential relationships between perceived costs and benefits. All significant (p < 0.05) relationships identified by regression results are presented with odds ratios, as odds ratios provide an indication of the strength of a relationship between two variables, whereas, significance does not (Tabachnick and Fidell 2001).

The demographic variables entered into the logistic regression analyses were: village category, gender, respondent’s resident status (native or immigrant), wealth (as measured by ability to meet livelihood needs after Mehta and Heinen 2001), education, age, occupation, and caste. Categorical demographic variables entered into logistic regression were transformed into dichotomous variables. Such collapsing of many categories into only two categories results in a loss of information, but is the simplest way of including these variables in the analysis (De Vaus 2002). Education and age were recoded to reflect low levels (primary or less) and high levels of education (more than primary), and younger (<45 years old) and older (>45 years old) respondents, respectively. Creating dichotomous categories from variables for occupation and caste would result in a loss of important differences between groups, so these variables were converted into dummy variables (De Vaus 2002). Occupation was a three-category variable (domestic work and agriculture, tourism, other) and is represented by two dummy variables (agriculture or otherwise, tourism or otherwise). For the purposes of this research, caste was treated as a three-category nominal variable and is represented by two dummy variables (low caste or otherwise, high caste or otherwise). Ordered ranking of castes is embedded in Nepali society, and, as such, provides a valid measure of a respondent’s social status.

Results and Discussion

Overall, the data represents 52% of male and 48% female respondents, without any statistical differences in gender between the two village types. The majority (84%) of respondents were born in the region now encompassed by the conservation area. Most of the respondents were between the ages 25 and 64 years old, with a moderate tendency towards younger (under 25 years) and older (over 65 years) respondents in DV. More than half of the respondents had little or no formal education; more in TV had a higher education level than those in DV. The main caste group in both villages was traders (middle caste), with significantly more low caste (i.e., Bishwakarma, Pariyar) respondents in DV and high caste (i.e., Chhetri, Brahmin) respondents in TV. A very strong relationship was found between occupation and village categories. Agriculture is the main livelihood activity in DV (66%), and no households were found to be employed in tourism. In TV, only 29% were engaged in agriculture and almost half of the respondents were employed in tourism. More respondents in DV (15.3%) than in TV (8.7%) reported difficulty in meeting livelihood needs.

Perceived Benefits

When asked to identify reasons in support of the conservation area, 46% of respondents suggest ACA benefits local people. To determine what local residents perceived as the benefits, respondents were asked in an open-ended question to provide a list of benefits received by their household and their community from the protection of natural resources and development projects. Responses were then grouped into six main categories of benefits: economic opportunities, social development programs, provision of resources for extraction, intrinsic values for resource protection, perks from participation in conservation and development, and livelihood protection through mitigation (Table 1). Except for participation benefits at the household level, no significant differences exist between villages and their recognition of specific types of benefits. More respondents in DV than in TV perceived participation as a benefit (χ2 = 6.050; p < 0.014).
Table 1

Household and community level benefits identified by respondents

Benefits

Total

Villages

p

TV

DV

χ2

Total household benefits

   Economic

17.6%

18.4%

16.5%

0.026

0.871

   Social development

40.4

43.7

36.5

0.730

0.393

   Extraction

54.3

53.4

55.3

0.013

0.910

   Conservation

16.0

15.5

16.5

0.000

1.000

   Participation

9.0

3.9

15.3

6.050a

0.014

   Mitigation

1.6

1.0

2.4

Total community benefits

   Economic

43.6

40.8

47.1

0.514

0.474

   Social development

69.7

69.9

69.4

0.000

1.000

   Extraction

58.0

52.4

64.7

2.400

0.121

   Conservation

35.1

38.8

30.6

1.052

0.305

   Participation

6.4

4.9

8.2

0.415

0.520

   Mitigation

5.3

3.9

7.1

Percentage based on total number of respondents, including those who gave no response. Percentages do not equal 100% because people were allowed to provide more than one response. Chi-square test results not provided when more than 20% of cells have expected cell frequencies less than 5. a Cramer’s V-values only provided for significant relationships; in this case the association is moderate (0.198)

Economic benefits from the conservation area are recognized by respondents for their community more so than their household. The development of transportation networks through road building is the most common economic benefit mentioned. Tourism is the main wage-generating economic activity in ACA; however, only 18% of respondents recognized household economic benefits from the protected area. Overall, 44% of respondents recognized benefits from tourism. Of those perceiving tourism benefits, 61% had a household member directly employed in tourism services. The remaining 39% recognized indirect benefits from tourism. Perceptions of benefits from tourism were substantially different between villages, with significantly more households in TV perceiving benefits from tourism than in DV (χ2 = 35.753, p < 0.0001), and having at least one person employed in the tourism industry (χ2 = 23.786, p < 0.0001).

Benefits from social development are the most frequently noted category of benefits at the community level. The main social benefits recognized by respondents include: sanitation management through organized village cleanup, garbage control and toilet construction, trail and bridge construction and maintenance, and access to drinking water. Social development projects are frequently mentioned as benefits resulting from ACA, as these projects provide the primary method to extend tourism revenue to individuals and communities not employed in tourism. As indicated by the response patterns, residents perceive social development benefits to accrue to the community more so than the individual, suggesting the communal nature of these benefits. While collectively, social development projects may make meaningful contributions to a community, unless individual households are positioned to take advantage of a project (i.e., able to send children to a newly constructed school), these developments may provide few recognized benefits to households (McIvor 1997; Archabald and Naughton-Treves 2001).

Wood, fuelwood, grass and fodder are the most frequent resources listed as benefits from conservation. Second to social development benefits, availability of resources for use is identified as a household benefit by 54% of respondents and as a community benefit by 58%. While current levels of use are insufficient to meet local people’s needs, restrictions on access are necessary as extraction levels required to fulfill needs may not be sustainable. However, deficiencies in current resource access may cause local people to have negative perceptions of the conservation area or illegally harvest resources to meet needs. In the study area, forests are too small to meet local needs and restrictions limit the amount of wood households can collect. To fulfill fuelwood needs, local people also use livestock dung and agricultural byproducts for cooking. People also indicated a desire to harvest medicinal herbs and live trees; however, at the time of this research, such activities were illegal. Conservation benefits at the household level were recognized by a fifth of the respondents and at the community level by a third. Reforestation and forest conservation in general are the most commonly cited benefit.

The categories of benefits identified by respondents are the same in all villages, with the exception of participation. Although 61% of respondents indicated participating in conservation and development committees, only 9% recognized participation as a benefit to their household, and 6% recognized it as a benefit to their community. Although the association is only moderate, DV did, however, recognize participation benefits at the household level significantly more than TV, suggesting the activities of ACAP have led to significant achievements in mobilizing remote communities.

Few respondents mentioned household or community benefits from mitigation efforts on the part of conservation area authorities. While ACAP has initiated a number of mitigation projects to reduce wildlife damages, such as providing financial support for constructing livestock pens in barbwire, residents do not recognize such efforts as benefits. Farmers must carry out personal mitigation measures by guarding crops overnight and scaring wildlife from fields during the harvest season. For the most part, reports indicate villagers do not kill offending wildlife; however, interviews with village leaders suggest retaliatory killing of wildlife did occur prior to a ban on firearms resulting from the political conflict. Also, reports of villagers killing monkeys were supported by sightings of observation platforms equipped with slingshots in crop fields. The conservation area does not provide a compensation scheme to cover damages caused by wildlife, resulting in a common complaint in ACA. Despite the complaints regarding wildlife conflicts, only 15% of respondents cite mitigation measures as necessary for their community. Given the widespread conflicts with wildlife, improvements to mitigation measures are necessary and could offer significant contributions to improving local attitudes toward wildlife.

Perceived Costs

Results show that almost all households collect resources from the conservation area (98%), and 53% indicate a desire to have increased access to resources. All respondents indicated collecting wood from the conservation area; however 91% would like to collect more. The current land available for livestock grazing appears to be sufficient, with only 33% wanting more access. Most significant is the apparent disconnect between demand and current levels of use of live trees, nontimber forest products/medicinal herbs, and wildlife. Few respondents admit to collecting these resources from the conservation area, but demand for increased access is notable: roughly 57% residents want access to live trees, 72% want nontimber forest product/medicinal herbs, and 19% want wildlife. Increased access to resources is needed mainly for supplies of fuelwood, construction material, and herbal medicine.

The overall mean for reduced access was 4.4, on a scale from zero to 10 (Table 2); more in DV than TV indicated this to be a problem. Results of logistic regression analysis (Table 3, top part) indicate no relationship between reduced access and village categories, and no demographic variables determined reduced access.
Table 2

Summary of costs identified by respondents

Specific costs

Total

Villages

χ2/Z

p

Cramer’s V

TV

DV

Reduced access to natural resourcesa,e

4.4

4.1

4.9

−2.102

0.036

Livestock lossesb,c

60.6%

51.2%

71.1%

5.788

0.016

0.203f

Crop lossesb,d

64.7%

55.1%

75.3%

6.755

0.009

0.212f

Results of a Mann-Whitney U tests and b Chi-square tests. Expressed as a percentage of people c raising livestock and d growing crops. e Mean of scale from 0 to 10. f Cramer’s V-values provided for significant relationships only; in both cases, the association is moderate

Table 3

Results of logistic regression showing the effect of demographic factors on perceived costs of conservation

Socio-economic variables

B

SE

Wald

p

Odds ratioa

Reduced access to natural resourcesb

   Village category (TV)

−0.36

0.38

0.91

0.341

   Gender (women)

−0.29

0.33

0.81

0.367

   Origin (migrant)

−0.42

0.47

0.80

0.371

   Wealth

−0.46

0.36

1.57

0.210

   Education

0.42

0.38

1.25

0.264

   Age

−0.10

0.35

0.08

0.783

   Occupation 1 (domestic or agriculture)

0.33

0.52

0.41

0.524

   Occupation 2 (tourism)

0.53

0.58

0.82

0.366

   Caste 1 (low)

0.61

0.42

2.05

0.152

   Caste 2 (high)

0.21

0.59

0.12

0.725

Livestock lossesc

   Village category (TV)

−0.54

0.41

1.78

0.182

   Gender (women)

−0.22

0.34

0.39

0.533

   Origin (migrant)

−0.43

0.50

0.75

0.388

   Wealth

−1.29

0.38

11.63

0.001

0.28

   Education

0.18

0.39

0.22

0.643

   Age

0.03

0.37

0.01

0.931

   Occupation 1 (domestic or agriculture)

1.39

0.57

5.87

0.015

4.00

   Occupation 2 (tourism)

1.66

0.66

6.25

0.012

5.25

   Caste 1 (low)

0.02

0.46

0.00

0.969

   Caste 2 (high)

−0.01

0.67

0.00

0.994

Crop lossesd

   Village category (TV)

−1.25

0.43

8.68

0.003

0.29

   Gender (women)

−0.24

0.39

0.38

0.536

   Origin (migrant)

−1.28

0.61

4.43

0.035

0.28

   Wealth

0.04

0.42

0.01

0.928

   Education

0.63

0.46

1.85

0.174

   Age

0.42

0.41

1.03

0.311

   Occupation 1 (domestic or agriculture)

1.57

0.64

6.07

0.014

4.80

   Occupation 2 (tourism)

0.49

0.80

0.38

0.540

   Caste 1 (low)

1.61

0.50

10.24

0.001

5.00

   Caste 2 (high)

0.38

0.78

0.24

0.628

B = regression coefficient, SE = standard error, Wald = Wald statistic, p = significance. The odds ratio is a measure of association between each independent variable and the dependent variable. When B is negative, odds ratio needs to be inverted to indicate odds. Equation to invert ratio (1/odds ratio). Overall fit of predicted to observed results = 63.1%, c 60.0%, d 74.5%

Roughly 85% of respondents raise livestock with no significant difference between village categories (χ2 = 1.69, p < 0.2). Overall, 61% of the people raising livestock experience predation on their animals by wildlife, but a moderate association was identified, with DV suffering more than TV (see Table 2). The number of wildlife species named as responsible for killing livestock by those raising livestock is an average of 0.7. Respondents from DV reported a higher number of wildlife species killing their livestock than respondents from TV (0.9, 0.6, respectively, Z = –2.646, p = 0.008). Logistic regression results show no relationship between livestock losses and village category, and only wealth and occupation showed significant effects (Table 3, middle part). The odds of poor respondents reporting high livestock depredation are 3.6 times higher than for the wealthy. The variables created for occupation show odds ratios of 4 for agricultural and domestic workers and 5.3 for respondents employed in tourism.

Crops are grown by 90% of respondents and represent the primary livelihood activity throughout ACA; roughly 95% in DV and 86% in TV grow crops. Overall, 65% of crop growers suffer losses caused by wildlife, with a moderate difference in crop losses between DV and TV villages; DV experience significantly more losses than TV (see Table 2). On average, 1.5 wildlife species were named responsible for crop damage. A higher number of wildlife species were reported as responsible for crop losses by DV than TV respondents (2.1, 0.9, respectively, Z = −4.861, p < 0.0001). This is confirmed by results of the logistic regression analysis (Table 3, bottom part). The odds of respondents from DV reporting high wildlife damage to crops compared to TV are 3.5, all other factors being equal. Origin, occupation and caste also had significant effects. Native residents were 3.6 times more likely to name many wildlife species responsible for crop damage than immigrants. Participation in agriculture or domestic chores leads to respondents being 4.8 times more likely to report high crop damage than those involved in other occupations. The odds of a low caste respondent reporting a high number of species responsible for crop damage are 5 times greater than for respondents from high castes.

Distribution of Benefits by Demographic Characteristics and Costs

When perceived benefits are examined across demographic variables, wealth was the only significant factor contributing to the variation (Table 4). The odds a wealthy respondent will report high levels of benefit are 2.2 times greater than for poor respondents, all other factors being equal. While regional inequities in the distribution of benefits are a common problem with IBPs (Bookbinder and others 1998; Josiah 2001; Kiss 2004), the results show ACA does not display similar discrepancies in benefit distribution between the two types of villages within the protected area. Although gender and caste have proven to be significant determinants of benefit distribution in other IBPs around the world (see for example, Ghimire and Pimbert 1997), these demographics proved to be insignificant in ACA. These results possibly suggest ACA has made significant achievements in transcending typical gender and caste barriers in their IBPs.
Table 4

Results of logistic regression showing the effects of demographic factors on perceived benefits of conservation

Socio-economic variables

B

SE

Wald

p

Odds ratioa

Village category (TV)

−0.02

0.40

0.00

0.960

Gender (women)

0.36

0.35

1.07

0.301

Origin (migrant)

−0.96

0.53

3.28

0.070

Wealth

0.80

0.38

4.48

0.034

2.22

Education

0.29

0.39

0.55

0.457

Age

−0.19

0.37

0.25

0.616

Occupation 1 (domestic or agriculture)

−0.27

0.55

0.25

0.615

Occupation 2 (tourism)

−0.31

0.61

0.26

0.608

Caste 1 (low)

−0.38

0.46

0.67

0.413

Caste 2 (high)

0.25

0.64

0.15

0.699

B = regression coefficient, SE = standard error, Wald = Wald statistic, p = significance. a The odds ratio is a measure of association between each independent variable and the dependent variable. When B is negative, odds ratio needs to be inverted to indicate odds. Equation to invert ratio (1/odds ratio). Overall fit of predicted to observed results = 62.6%

Past research has shown that tourism is often criticized as an IBP for its inability to provide benefits throughout a community and an entire protected area (Barrett and Arcese 1995). However, results of this study show that an individual’s type of benefit from tourism (none, indirect, direct) has no relationship with their levels of perceived overall benefit including community benefits (χ2 = 1.513, p < 0.5). Figure 2 shows that benefit perceptions remain constant regardless of the type of involvement in the tourism industry. While the economic benefits received through direct employment or participation in tourism are higher in TV, tourism benefits have been extended indirectly to other areas. Approximately one-third of respondents recognize indirect benefits from tourism, indicating that residents realize the development and conservation projects conducted in their community derive from the revenue from tourist entry fees.
https://static-content.springer.com/image/art%3A10.1007%2Fs00267-008-9130-6/MediaObjects/267_2008_9130_Fig2_HTML.gif
Fig. 2

Relationship showing the effects of type of tourism benefit on overall perception of benefits from conservation. Perceived benefits scale is based on composite scores

In order to determine if benefit distribution reflects conservation costs, the relationship between perceived level of benefit and each cost (reduced access, livestock losses, crop losses) was investigated using bivariate simple logistic regression. The only significant relationship (negative) found was between crop losses and benefit perceptions; as the number of wildlife species named responsible for crop losses decreases by one, the odds that the respondent will perceive high levels of benefit are 1.3 (Table 5, Block one). In other words, benefits are perceived to be less among those experiencing more crop losses.
Table 5

Results of hierarchical logistic regression showing the effects of perceived costs on conservation benefits

Blocks

B

SE

Wald

p

Odds ratioa

Block one

   Crop losses

−0.29

0.10

8.38

0.004

0.75

Block two

   Crop loss

−0.29

0.11

7.44

0.006

0.75

   Village category (TV)

−0.05

0.35

0.02

0.877

Block three

   Crop losses

−0.30

0.12

6.64

0.010

0.74

   Village category (TV)

−0.06

0.35

0.03

0.866

   Livestock losses

0.03

0.20

0.02

0.888

   Reduced access to natural resources

−0.01

0.08

0.02

0.897

B = regression coefficient, SE = standard error, Wald = Wald statistic, p = significance. a The odds ratio is a measure of association between each independent variable and the dependent variable. When B is negative, odds ratio needs to be inverted to indicate odds. Equation to invert ratio (1/odds ratio). Overall fit of predicted to observed results = 57.1%

By controlling for village category, the relationship between crop losses and benefits remained significant, and no relationship was identified between village category and benefit perceptions (Table 5, Block two). This indicates differences in perceptions of benefit receipt cannot be accounted for by location of the villages (i.e., on or off the main tourist trail), and benefits are not distributed according to crop losses throughout the study area. Reduced access and livestock losses showed no relationship with benefit perceptions (Table 5, Block three); therefore, we conclude that IBPs do not target those suffering the greatest consequences of conservation.

Conclusion

Overall, results of this study indicate that local people recognize benefits from ACA. Perceptions of intrinsic conservation values are not commonly associated with protected area residents in developing countries (Muller-Böker and Kollmair 2000; Balmford and Whitten 2003). Yet conservation benefits in ACA were recognized by a fifth of the respondents and at the community level by a third. Despite the lack of recognition of intrinsic values of wild animals, findings indicate an interesting trend in ACA, where local residents recognize the intrinsic value of conservation efforts in their area. This may suggest educational efforts by ACAP have been somewhat successful. Study findings also indicate that the households which are facing difficulties meeting livelihood needs are less likely to perceive benefits from conservation and development. Findings indicate that the gap between the rich and poor is accentuated by IBPs in ACA; however, the recognition of even some benefits such as social development by the underprivileged indicates their wellbeing has been boosted. On a spatial scale, benefits are distributed equally across the region regardless of a village’s location relative to popular tourist routes. ACA has successfully overcome the common shortcoming of IBPs by designing and distributing benefits to all villages within the protected area. These results contradict typical criticisms of IBPs in the literature (Pandit and Thapa 2004; Gadd 2005). Not only are benefits perceived equal across all villages, but those not positioned to participate directly in tourism still perceive benefits from ACA.

However, results presented in this study show IBPs are not targeted specifically to those experiencing the greatest costs from protected wildlife. In order for IBPs to address local concerns over restrictions and policies imposed by protected area regulations, addressing the costs borne by local residents must be a priority. To date, little has been done to mitigate the costs faced by residents of ACA. ACAP needs to make donors and NGOs aware of the extent of the problem, and encourage funding support for effective solutions.

The benefits available to residents could be increased by maximizing benefit retention in the region. ACAP has been successful in retaining tourism revenue by preventing the devaluation of the market from competitive pricing. By further regulating tourism in support of local participation in direct and indirect employment opportunities, ACAP would be able to increase the economic benefits available to individuals. Partnerships should be further encouraged between local producers of goods and tourism service providers to integrate the local economy. Strengthening educational efforts to inform tourists of their contribution to livelihoods by buying locally, and encouraging minimal imported goods would further contribute to local revenue retention. In addition, ACAP should impose minimum residency requirements for guiding and porter operations in the ACA region. Currently, most guiding companies are based in the capital city, Kathmandu, and employ guides from all over Nepal. By requiring a certain percentage of guides to be from the ACA region, employment is captured locally. Targeted training programs would help to ensure the availability of locally qualified guides and porters.

Critics of IBPs suggest the source of IBP failures rests in their inability to reach marginalized residents of protected areas (Ghimire and Pimbert 1997). While this study does support such criticism, the fact that marginalized residents recognize any benefit at all from IBPs provides an indication that these programs can make a difference—in the absence of IBPs, marginalized residents would be worse-off. The reliance of IBPs on social development benefits is also viewed as an obstacle to success (Brandon 1998); however, IBPs in ACA prove that social development benefits can, in fact, be an effective tool to extend benefits to communities and households otherwise excluded from tourism benefits. ACAP’s success at extending benefits to communities throughout the region indicates regional distribution inequities stemming from IBPs that rely on tourism as a primary source of project funding can be overcome.

While the importance of tourism as a direct provider of income and employment opportunities is recognized by local residents, indirect benefits such as social development are equally recognized. Thus, conservation benefits are not necessarily always perceived in purely economic terms, and that integration of social development benefits in conservation plans are critical to their successful implementation.

The achievements ACAP has made in the distribution of recognizable IBP benefits do not necessarily indicate benefits act as sufficient incentives to generate positive attitudes and behavior toward natural resource conservation. For IBPs to make significant contributions to conservation objectives, IBPs must generate support for conservation by establishing a recognized link between benefits and conservation among beneficiaries. This link is essential to the success of incentive-based conservation. One of the first steps in exploring this link is in finding out what types of benefits of conservation exist and how they are distributed (relative to costs). This study has done just that. The next step is to examine how service delivery can be improved to better reach those involved in unsustainable activities and the impact ultimately of these services to on the ground conservation behaviors.

Acknowledgments

Funding support for this project was provided by the Social Sciences and Humanities Research Council of Canada (Grant No. 410-2003-0136). We would like to thank the staff of the Trust for providing relevant information, Damodar Khadka for assisting with the design and application of the survey, and our field research assistants, Prem Kumar and Alka Gauchan, for their commitment and hard work. Gail Fondhal, Ray Chipeniuk, and Jeremy Mackenzie also provided valuable assistance throughout the research. Thanks are due also to Stella Nepal for preparing the ACA map presented in Fig. 1, and Jill Belsky, Harold Goodwin, and one anonymous reviewer for their comments to improve this manuscript. Most of all, we thank the people in the villages in the Jomsom area for welcoming us to their communities and offering their time and insights.

Copyright information

© Springer Science+Business Media, LLC 2008