Article
Tiger, Lion, and Human Life in the Heart of Wilderness: Impacts of Institutional Tourism on Development and Conservation in East Africa and India
Nilanjan Ghosha,# and Emil Uddhammarb
aMulti Commodity Exchange of India Limited, Mumbai, India
bDepartment of Government, Uppsala University, Uppsala, Sweden
#Corresponding author. E-mail: nilanjan.ghosh@gmail.com
Copyright: © Ghosh and Uddhammar 2013. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and distribution of the article, provided the original work is cited.
Abstract
This article tests the hypothesis on whether tourism is an important institutional factor in reconciling the conflicting goals of conservation and development. The study entails data from field surveys across protected areas including the Serengeti National Park and the Ngorongoro Conservation Area in northern Tanzania, and the Corbett National Park in northern India. With human development defined in terms of ‘stages of progress’ (SOP) delineated by the respondents themselves, the study finds indicative evidences of the validity of the posed hypothesis in the two nations, in varying proportions. Factors not related to tourism, like incomes from livestock, have affected development in Tanzania, though not in India.
Keywords: human development, stages of progress, conservation, tourism, community, Serengeti National Park, Ngorongoro Conservation Area, Corbett Tiger Reserve
Access this article online
Quick Response Code:
Website:
www.conservationandsociety.org
DOI:
10.4103/0972-4923.125750
INTRODUCTION
The apparent conflict between conservation and development in and around the protected areas of the developing world arises as the poor in those areas are reliant on forest resources (Dewi et al.
2005; Chan et al. 2007; Torri and Herrmann 2010). This leads to a decline in forests, much to the detriment of both flora and fauna. Man-animal conflict is also a special feature in these parts of the world. Wild animals cause losses to property, cattle, and even human life. Hence, in most cases, the human habitat in and around wilderness does not hold a very kind opinion about the wild predators. In most cases in the developing nations, protecting biodiversity has resulted in a shrinkage of traditional economic opportunities for the local population due to ensuing restrictions on cattle ranching, farming or fuel wood collection. People have often been evicted altogether from the protected areas
(Uddhammar 2006; Schmidt-Soltau 2010), thereby aggravating the conflict between conservation and traditional economic activities (Uddhammar and Ghosh 2009). This necessitates innovative thinking on new institutional arrangements that could reconcile conservation and development, and, in the best of worlds, make them benefit from each other.
However, the possible existence of a symbiotic relationship between humans and forests has been a matter of debate among scholars. One school strongly believes that forest resources can be put to use to help improve the livelihoods of the poor (Scherr et al. 2002; Dewi et al. 2005). There are others who believe that forests can provide only limited opportunity to contribute to poverty reduction (Wunder 2001). Part of the discrepancy between the conflicting views originates from the difference in assumptions about the institutional mechanisms for creating new opportunities for rural people to take advantage of forest resources (e.g., Sunderlin et al.
2005). Publicationsby several researchers like Agrawal and Clark (2001), Anuradha et al. (2001), Borrini-Feyerabend et al. (2003), and Greiber (2009) advocate specific institutional mechanisms to reverse the trade-off between conservation and development. An important entry-point of this article lies with an attempt to understand the nature of the impact of such a specific institutional mechanism as the exogenous stimulus on the endogenous conservation-development dynamics.
In this article, tourism is hypothesised as an important institutional variable affecting the trade-off between conservation and development. Duffy (2002) and Vanasselt (2003) feel that unregulated tourism can bring about major environmental losses, with marginal financial gains. Fennell (1999), Wearing and Neil (1999), and Ulfstrand (2003) are, however, optimistic. Kiss (2004), Zapata et al. (2011), and Uddhammar (2006) emphasise the need for necessary institutional structures for success of community-based tourism1 in promoting the dual objectives.
We draw our hypothesis from this ongoing debate in international literature, and pose it as: tourism as an institutional intervention can reverse the trade-off between conservation and development, thereby generating employment and income in the sector. In order to test this hypothesis, we have conducted surveys in Serengeti National Park (NP) and the Ngorongoro Conservation Area (CA) in northern Tanzania, and the Corbett Tiger Reserve in northern India. Eventual analysis has been carried out on the basis of primary data (mostly based on perception), as also some related secondary information. So far, despite the raging international debate, there are hardly any studies that test such a hypothesis for the developing world in two completely contrasting settings, which would lend further applicability to the posed hypothesis. An important aspect is the methodological issue, where we define development from a local well-being perspective, following Krishna (2004a, b), and conservation on the basis of a composite sighting index.
Such a methodology has not been adopted so far in order to test this hypothesis—this is an important contribution of this article to the literature base.
Apart from the methodological perspective, the contribution of this article to the literature is also the perspective it provides from its departure from neoclassical valuation frameworks based on which often such analyses are carried out (e.g., Beharry and Scarpa 2009; Guha and Ghosh 2009; Lange and Jiddawi 2009, among others). Here, the assessments of two comparable institutional frameworks have been conducted taking into consideration how institutional arrangements and tourism as a critical variable affect two target variables like conservation and development, in a social-ecological system (SES).
Selection of the study areas
The idea here was to find well-known and frequently visited tourist destinations in developing countries with prevalent nature-related tourism. If tourism benefits generated by global nature-related tourism trickle down to the local human population, it should be visible in these areas. Secondly, in many of these areas the pressure from a growing human population constantly threatens the biotopes and species of remaining wildlife. By covering protected areas in settings with different cultural and institutional backgrounds, the aim here was to reveal patterns that are of general applicability.
East Africa and India have some of the most widely known and precious inheritance of biodiversity on Earth. The unique fauna of India includes elephant, tiger, gaur, and other large mammals, while the unique variety of large mammals in East
Africa (lion, elephant, zebra, etc.) is no less renowned. At the same time, the poverty of the rural human population adjacent to the protected areas in these regions is often striking. Selecting protected areas in these countries provides the opportunity to examine whether protected areas with a strong capacity for tourism really can make a difference to the well-being of the people surrounding them, and also study the consequent impact of wildlife on human economy. Tourism in East Africa fluctuates between being the most important and the second- most important export product, and the safari destinations in the region are world-renowned. Around 90 % of travellers to East Africa are foreign tourists. The Serengeti-Ngorongoro zone in Tanzania is a typical case representing this phenomenon of high international tourism, as also the case of critical livelihoods of locals being linked to the tourism economy.
India is interesting not because international tourism to protected areas is conspicuous—international tourists comprise only 20 % of the total number of visitors; 80 % is domestic tourism, mostly from urban elites(Uddhammar 2006)—but because it has a conspicuous biodiversity that is both well known, to a large extent red-listed, and under severe pressure. In India, Corbett Tiger Reserve in the northern part of the country was chosen for the study. Interestingly, although most tourists come from within the country, more than half the total revenue derives from foreign tourists. Thus, the global connection with the Corbett park is quite strong (Uddhammar 2006).
Again, with a majority of employees in the camps and the park being recruited locally from the region, the local connection is also highly prevalent. Tourism, as such, is still emerging in the region, and has advancedonly in the new millennium (Uddhammar and Ghosh 2009). Therefore, the two cases from developing economies offer some interesting features to compare and contrast in the context of the hypothesis posed.
The article is divided into seven sections. In the second section, the hypothesis is explained in the context of social- ecological systems (SES) (Ostrom 2005, 2007). In the third section, the study sites are described in light of the variables described in SES. The fourth section briefly talks about the methodology used. In the following section, we present some descriptive statistics on the ‘stages of progress’ (SOP) which delineate development in this context. Since development and poverty have been defined by the respondent community, this also speaks a lot about the existing culture, tradition, and social norms of the community under consideration. It is in this context that wewould like to declare that no gender-based distinction has been made in this article, and the information has been reported as obtained from the field. The sixth and the seventh sections report the results of the Indian case and the Tanzanian cases, respectively. Finally, we end with the concluding remarks.
A SOCIAL-ECOLOGICAL SYSTEM: HUMAN WELL-BEING AND WILDLIFE CONSERVATION The hypothesis can be posed in the framework of the social- ecological system (SES), to better understand the interactions
between the properties of the ecosystems and the actions of human societies. In a given social-ecological system, one can identify a number of variables (Ostrom 2009), as presented in Figure 1.
In Figure 1, we can see that in the formalised flow of influence and use, the users’ use of resource units is the core activity affecting the outcomes, which provides feedback to the resource system (the ecosystem) via the outcomes. The dynamic part, i.e., the ‘interactive process,’ is represented by the arrow going from the users through the resource units resulting in the outcomes. By internalising the SES presented in Figure 1 in the context of this study, it may be noted that a special feature of the resource systems studied here is that they are inhabited by dangerous wildlife, that every year kill a number of people, cause damage to livestock, and destroy crops, and hence are not popular among the local human population. This makes these social-ecological systems unique and critical in the sense that institutions need to be developed to protect human lives and livelihoods.
From the SES perspective, the resource units in the Serengeti regions are characteristically almost identical to those of the Corbett Tiger Reserve. Typically, they involve humans, wildlife, tourists, and NGO groups. The Serengeti and Ngorongoro conservation zone is one of the earliest established national parks in sub-Saharan Africa. The late 1950s witnessed the excising of Ngorongoro from Serengeti National Park, proposed as a measure to accommodate the interests of the Maasai pastoralists. From an ecosystemic perspective, however, they can be considered to be integral components of the same ecosystem.
The dynamic interactive processes of the users (which, in this case, are the communities and the tourists), with the governance systems and the fauna species, result in outcomes related to the dynamic interrelationship between the dependent variables, fauna conservation, and human development. An
important part of these is the exchange between users, resulting in effective selling of resource units. This market exchange is essentially tourism, where tourism service providers have to
‘sell’ the services along with the sightings of wild animals, which are a major attraction of these protected areas.
The interactive processes are affected by governance systems. In the core zone of the Corbett Tiger Reserve, mandatory tourist guides are recruited locally from among villagers, and this arrangement has many advantages. While on the one hand, local people get employment and training, on the other, the community also gets a clear signal that wildlife is an asset to be conserved. Uddhammar (2006: 672) also notes that further efforts by the Ramnagar municipality and mayor to create the image of ‘tiger city’ have played a big role in increasing local awareness and appreciation of the park.
In Tanzania, access to the protected areas is curtailed, and penal clauses exist on infringement (Robinson 2011).
However, there are game reserves where licensed hunting takes place. Tourist hunting in Tanzania is regulated by the central government with little local input into quota-setting, block allocation, or management (Leader-Williams et al. 1996).
Revenues go to the central government with a proportion (approximately 20%) returned to the district councils in areas where hunting occurs.
Governance systems, on the other hand, have affected the interactive processes between resource units. Though human- wildlife conflicts in the Serengeti have been a traditional phenomenon, communities feel that most of these conflicts emerged as a result of wild animals being accorded a higher priority than human beings (Kideghesho 2010). However, as reported by Robinson (2011), that perception has been changing over the last few decades. While local communities have been actively involved in providing tourism services, there has also been a recent plan to establish Wildlife Management Areas (WMAs) in the buffer zones surrounding Serengeti National Park, out of which numerous benefits for the local communities can be envisaged in the form of tourism incomes and conservation (Kideghesho 2010: 240). This adds a distinctive dimension to the interactive process at the
‘resource system,’ where a ‘conflictual’ interaction between two critical resource units has been attempted to be transformed to a ‘symbiotic’ interaction, through conscious government policy measures.
Our interest here is primarily to explore if the interactions between resource units are mainly ‘symbiotic’ or ‘competitive’
(Ostrom 2007, 2009). For this purpose, ‘output’ (in terms of the SES) has been measured in four different ways. They are: 1) ‘stages of progress’ (SOP) out of (or into) poverty for communities around conservation areas based on primary data (Tables 2 and 3; regressions looking at specific factors behind this movement presented in the section ‘Measurement of conservation’ below; and in equations 4–10); 2) employment in the tourism sector based on primary data (Tables 4 and 7);
3) biodiversity (Tables 5 and 6); and 4) the coexistence of tourism and wildlife (Tables 5 and 6) based on secondary and primary data collected by us.
Figure 1
A formalised social ecological system (SES)
The main dependent variables are found in the ‘Outcome’ square, while the
‘Governance system’ and the ‘Resource system’ squares contain the main independent variables. The ‘Resource units’ are the units to be measured, and the ‘Users’ are the stakeholders involved. The resource system (a local ecosystem) influences the resource units (the kind of units, such as
farm products, wildlife, etc.) that can be used. Dotted arrows represent indirect influence (or feedback), while the solid arrows represent causal
mechanisms.
5HODWHG HFRV\VWHPV
*RYHUQDQFH
V\VWHP (FRQRPLF
VRFLDORUGHU
8VHUV 5HVRXUFHXQLWV 2XWFRPHV
5HVRXUFH V\VWHP
,QWHUDFWLYH
SURFHVVHV
A BRIEF DESCRIPTION OF THE STUDY AREAS The Serengeti ecosystem encompasses the 14,800 sq. km Serengeti National Park as well as game reserves surrounding the NP such as Grumeti, Maswa, Ikorongo, and Kijereshi, and
Table 1
Overview of definitions of poverty and definitions of stages out of poverty as defined in group discussions in villages and towns in Tanzania and India
Poverty First stage out of poverty Second stage out of poverty
Tanzania Serengeti National Park and Ngorongoro Conservation Area
Low income, i.e., less than 1,000 Tanzanian Shillings (TSH) per day (0.84 USD); bad housing (grass roof); none or less than 0.4 ha of land; no children;
no wife; no livestock
Land; ability to feed family; ability to send children to school; house with corrugated iron sheets for roof;
4 oxen; 5-10 diary cows; one wife;
few children (5-10); radio (in the case of towns)
2 tractors; more than 4 ha of land, brick house; 50 livestock; 5-7 wives;
more than 40 children; (particularly daughters, who have value, and can be
‘sold’ to purchase cows); ability to buy clothes (in the case of towns); enough capital to start a small business (in the case of towns)
India
Corbett National Park No livestock or only one; no land;
no house; no electricity; no job or income of INR 60 per day (USD 1.33); no medical service
Can afford to hire tractor; owns a few milch animals; owns bullocks;
earns between INR 70 and 135 per day as income (USD 1.5-3); 2-3 members of family are government employed; possesses a house;
possesses land<2 ha; can afford food for three meals per day; can provide for school
Electricity in house; sends children for higher education; owns a television set; owns a dish antenna; has water tank in house; has solar panels; has a pucca (brick) house; has 2-4 ha of land; can repay debts; owns a motorcycle
Source: Primary survey
Table 2
‘Stages of progress’ for households around Serengeti NP and Ngorongoro CA in 2007 as compared to 1997 Stages of
progress for households in 1997 (%)
Stages of progress for households in
2007 (%) Total (%)
Low Middle High
Low (63.1) 43 (24.2) 132 (74.1) 3 (1.7) 178 (100) Middle (36.2) 37 (36.3) 51 (50.0) 14 (13.7) 102 (100)
High (0.7) 0 (0) 2 (100) 0 (0) 2 (100)
Total (100) 80 (28.4) 185 (65/6) 17 (6.0) 282 (100) Percentages of each row within brackets. In the far left column, column percentages are presented within brackets
N.B.:
1. Low refers to ‘poverty’; Middle refers to ‘first stage out of poverty’; High refers to ‘second stage out of poverty.’
2. Out of the 300 respondents, 282 respondents provided valid responses (responses like “cannot answer” are not valid responses) of their present state, and the state that they were in around 10 years ago.
Source: Primary survey
Table 3
‘Stages of progress’ for households around Corbett NP in 2007, as compared to 1997
Stages of progress for household in 1997 (%)
Stages of progress for household in
2007 (%) Total (%)
Low Middle High
Low (30.9) 8 (13.6) 15 (25.4) 36 (61.0) 59 (100) Middle (51.8) 1 (1.0) 65 (65.7) 33 (33.3) 99 (100) High (17.3) 1 (3.0) 8 (24.2) 24 (72.7) 33 (100) Total (100) 10 (5.2) 88 (46.1) 93 (48.7) 191 (100) Percentages of each row within brackets. In the far left column, column percentages are presented within brackets
N.B.
1. Low refers to ‘poverty’; Middle refers to ‘first stage out of poverty’; High refers to ‘second stage out of poverty’.
2. Out of the 196 respondents, 191 respondents provided valid responsesof their present state, and the state that they were in around 10 years ago.
Source: Primary survey
Table 4
Socio‑economic profile of lodges in and around Corbett National Park, India in 2007
Socio-economic features of lodges
Stratified sample 15 (population size=25)
Number of people employed in tourism
sector in area 570
Number of people earning livelihood
from tourism sector in the area* 2,964 Percentage of the employed belonging to
the region 59
Percentage of managers belonging to the
region 52
Percentage of foreign tourists 7-10
Source: Survey results; *World Bank 2006
Table 5
Changes in wildlife and local human and tourism populations in the Corbett NP area
Factor change
1987-2007 Correlation with tourist visitors to Corbett NP
1987-2006
Human population* 1.2
Tourist visitors 4.5
Elephant (Elephas maximus) 3.9
Tiger (Panthera tigris) 2.1*** 0.738**
Sambar deer (Cervus
unicolor) 2.0
Cheetal deer (Axis axis) 1.7
*Garhwal district, Uttarakhand; average change 1981-1991 and 1991-2001;
**Significant at 0.01 level;
***1987-2006; Source: WII 1999; Jhala et al. 2008; NP=National park
open areas/community lands. The Ngorongoro Conservation Area covers an area of 8,300 sq. km. The Maasai Mara National Reserve in Kenya is also a part of this ecosystem (Figure 2). Many villages outside the Serengeti National Park participate in the community-based conservation programmes.
The National Park allocates up to 7% of its budget to support projects identified by villagers surrounding national parks.
This offers good opportunities to study the long-term effects on biodiversity as well as on human development in the area.
The Corbett NP (Figure 3) is located 250 kilometres northeast of Delhi and close to the city of Ramnagar in the state of Uttarakhand (formerly called Uttaranchal, when the survey was conducted). The park was created in 1936 and today has a total area—including buffer zones—of about 1,318 sq. km.
Human habitation is not allowed in the major core zone but some settlements existin the surrounding buffer zone. The park is owned by the Uttarakhand state government and managed by the Uttarakhand Forest Department.
Situated on the foothills of the Himalayas, Corbett NP is widely renowned as a tiger reserve with a rather successful history of conservation and natural resource management. The national park’s institutional history draws from varied sources:
the legacy of the colonial forester and conservationist Jim Corbett, international initiatives to save the tiger in the 1970s, the Indian government’s national level conservation programme through Project Tiger, the history of the Forestry Civil Service, and the interventions of various NGOs. The Corbett National Park and the Sonanadi area were included in the Corbett Tiger Reserve in 1991 (WII 1999). While community-based tourism initiatives and lodges were developed in the zone, most of the developments have occurred in the new millennium.
Corbett TR presents a unique case where the community’s relationship with the government (or forest department) has not been uniform. During our study, we found that in some of the villages where tourism had developed (e.g., Bhakrakot), the relationship seemed quite cordial, while tensionswere prevalent in others (e.g., in Laldhang, with respect to relocation).
AN OVERVIEW OF THE METHODS
A strategic method was used in selecting the villages for interviews and data collection. For each area, we selected some (two or three) villages close to the protected areas within the
Figure 2
Map of the Serengeti–Mara ecosystem, including the Tanzanian game reserves where, except in Ikorongo and Grumeti, licensed hunting
takes place Table 6
Changes in wildlife, livestock, local human, and tourist populations in the Serengeti‑Ngorongoro ecosystem, and their
correlations (Pearson’s r)
Population Factor
change 1997-2006
Correlation with tourist visitors to area 1988-2004 Local human
population (Ngorongoro) 1.52
Livestock (Ngorongoro) 1.35
Tourist visitors (Ngorongoro) 1.23*
Tourist visitors (Serengeti) 1.01*
Elephant (Loxodonta
Africana) (Serengeti) 1.81 0.826**
Buffalo (Syncerus
caffer) (Serengeti) 1.0
Lion (Panthera leo) (Serengeti) 1.41 0.658***
N.B: correlations were calculated for 1988 to 2004, during which period wildlife numbers have fluctuated;
*Between 1994 and 2004, the factor change is merely 1.07 due to a sharp drop in visitor numbers after terror bombings in the US in 2001. However, a long-term trend from 1966 to 2004 shows a factor change of almost 5 in Serengeti, and almost 7 in Ngorongoro.;
**Significant at 0.05 level;
***Significant at 0.01 level; Source: Ottichilo 1999; Reid et al. 2003
Table 7 Profiles of lodges surveyed
Socio-economic features of lodges Serengeti NP/
Ngorongoro CA
Stratified sample 13 (population size=23)
Number of people employed in sector
in area 1,650
Number of people earning livelihood
from sector in area* 8,085
Percentage of the employed belonging
to the region 64
Percentage of managers belonging to
the region 20
Percentage foreign tourists 87
Source: Survey results; *World Bank 2006 NP=National park; CA=Conservation area
Source: Emerton and Mfunda 1999
tourism ‘circuit’ (zone where tourism was more prevalent than in others), and some a considerable distance away from the protected area. We also selected two neighbouring towns, one within and one outside the tourism circuit. The differentiation between towns and villages was done based on the definition provided by Census of India 2001. In Tanzania, the definition of a town, as distinguished from a village, was obtained from the 2002 Population and Housing Census. Generally, towns are distinguished from villages on the basis of administrative, demographic, and infrastructural characteristics, and hardly on the basis of occupational patterns or dominance of the agricultural sector. The control cases for towns and villages were provided by those that were outside the tourism circuit.
This helped us to compare and contrast between regions with and without tourism and determine the exact impact of tourism on conservation and human development2.
Selection of households within each zone of villages/towns was done on the basis of complete enumeration or random (or systematic) sampling in cases where the total number of households was not too large. In cases of very large populations across large areas, stratification in terms of localities was created,
and then random (or systematic) samples were drawn (see Appendix 2 for details). In total, we interviewed 300 households in Tanzania, and 196 households in India. In both places, the data were collected during January-March 2007. We developed various indices as and when required, and econometric techniques were used to test for the relationship between variables.
Development defined in terms of ‘stages of progress’
matrix
Human development has been defined in the analysis through data collected from a ‘quasi-longitudinal’ survey, following the method used by Krishna (2004a) in villages in Rajasthan, India, to assess who escaped poverty, who became poor, and why.
Part of the method was to let the villagers themselves define poverty in preliminary discussions of ‘stages of progress.’ This was delineated by change in living conditions, and was defined in a two-step process. In the first stage, in each of the villages and towns selected around the protected areas, we assembled around 8–10 people for a presentation of our purposes and to provide them with information about the subsequent surveys.
Figure 3
A map of the Corbett Tiger Reserve
Source: http://www.corbett-national-park.co.in/corbett_national_park_map.html
We also had a focus group discussion with these respondents to find out their definition of poverty, and the first stage out of poverty. Then, we enquired about the next step ‘upwards’
in development.
With these results as a base, we formed three Weberian ideal types3 (but with mutually excluding categories for each item) that were presented to the informants as part of the survey conducted a few months later in the respective sites. This entailed the second stage of the field survey. Even though the definitions differed somewhat, the important thing here is that each community had the opportunity to define poverty and the steps for moving out of poverty themselves.
During the field survey that followed, we asked each respondent which of the three stages fitted his/her family’s living conditions: 1) 10 years ago; and 2) at the time of the interview. In doing so, we created what we call a quasi- longitudinal measure of the respondent’s living conditions.
From this data, a matrix of various positions of development could be constructed with movements from the possible positions 10 years ago to the present (see Appendix 1 for details).
Measurement of conservation
To measure conservation, on the other hand, a composite index was devised. This index measures wildlife sightings by respondents. To examine whether the sightings of some critical fauna had changed, respondents were asked whether the sightings of certain species (decided in consultation with the forest department, existing literature, and knowledgeable persons from the field) had increased, remained the same, or had diminished over time. A rating of +1 was given if the sighting had ‘increased’, 0 for ‘no change’, and -1 if the sighting had ‘decreased’.
We constructed a ‘fauna sighting change index’ based on weights given to each of the species and the score given by a respondent in terms of change in sighting. The weights were decided in consultation with the abovementioned stakeholders, taking into consideration the ‘rarity’ aspect of the species, and their importance in the context of tourism. In that sense, this could be considered as ‘informed arbitrariness’ with which the weights in India (as also in Tanzania) were assigned (see Appendix 3). Hence, a positive value of the index is an indicator of the increase in sighting, while a negative value is an indicator of a decline in the same. This measure was complemented by measuring the factor change in species, which was obtained based on secondary data.
Regression analysis
The hypothesis was simple here. We tried to determine which factors lead to coexistence between conservation and development goals. These institutional factors emerged from tourism and other sources of change. We determined the influence of these factors in the two study sites. The regression equations used were as follows:
Y = α1 + β1X1 + β2X2 + β3X3 + β4X4 + u (1) Z = α2 + β5X5 + β6X6 + β7X7 +ε (2) Y = α3 + β8D1 + β9D2 + Ω (3) The following are the interpretations of the symbols used: Y ≡ Stages of progress movement, as will be defined in course of this analysis and further explained in Appendix 1;
X1 ≡ Difference in income from sale of livestock between 1997 and 2007;
X2 ≡ Difference in income from sale of agricultural products between 1997 and 2007;
X3 ≡ Difference in income from park and tourism between 1997 and 2007;
X4 ≡ Difference in income by working in large cities between 1997 and 2007;
Z ≡ The fauna sighting change index;
X5 ≡ Change in importance of income from livestock4; X6 ≡ Change in importance of income from sale of agricultural products;
X6 ≡ Change in importance of income from working in major cities;
X6 ≡ Change in importance of income from tourism;
D1 ≡ Dummy variable for towns related to tourism;
D2 ≡ Dummy variable for villages related to tourism.
All these variables, which are perception-based observations of the community, were obtained from the two rounds of primary surveys, the first one being unstructured and the second one consisting of a structured questionnaire. For most of the variables (except for stages of progress, whose estimation has been explained in Appendix 1), the respondents were asked about their perception of whether a particular variable had changed, as mentioned above. The changes were ‘increase’ (denoted by +1), ‘decrease’ (denoted by -1), and ‘no change’ (denoted by 0).
In that sense, we are not looking for actual (absolute or relative) figures of change, but for perceptions of change.
Equation (1) attempts to find out the community’s perceptions of what sources of income have contributed to the change in their overall poverty status. This is a reflection of the changing relative importance of a source of income in determining the changes in the community’s developmental status. In equation (2), we intend to test whether change in relative importance of a particular source of income (particularly from tourism) has an impact on ‘fauna sighting change index,’ which is assumed to be a proxy of fauna conservation according to the community’s perception5. One needs to keep in mind that even if income from a source might have increased, it is not necessary that the relative importance of that source of income has increased vis-à-vis other sources. Fauna sighting might be more affected by the change in relative importance of a particular source of income rather than the change in income6.
Equation (3) tests whether communities close to tourism sites have witnessed better development defined in terms of ‘stages of progress,’ as compared to those far from tourism sites. The idea here is to examine the differences in the developments that have been encountered in zones that are associated with tourism against those that are not associated with the same. This
is done by considering two tourism dummy variables (one for towns, and the other for villages) as explanatory variables for
‘stages of progress.’ In order to vindicate the impact of tourism- related zones on development, by controlling the impacts of all other factors, and also removing any possibility of multicollinearity in the model (that may destroy our objective), all other explanatory variables are deliberately excluded. If the estimates of the coefficients of the dummy variables are found to be statistically significant, then one may safely argue the existence of indicative evidences on development being associated with tourism. In that case, equation (3), combined with equation (1), will further buttress the contention that tourism can be an enabling factor of development and can enable moving the community up the ladder of ‘stages of progress.’
DESCRIPTION OF RESULTS FOR
‘STAGES OF PROGRESS’
As we understand here, the most critical variable in this context is the ‘stages of progress.’ The results of the first and second rounds of discussions in the two cases are given in Tables 1, 2, and 3.
As can be seen in Table 1, the definition of ‘stages of progress’ from poverty hinge more on family relations in the context of Tanzania than in India. As an exposition,
‘no wife, no children’ is one of the indicators of poverty in Tanzania. In the Tanzanian case, 5–7 wives and 40 children was defined as part of the second step out of poverty. This indicates that family relations in Maasai communities are not only confined to social relations, but that they also have a clear material content for the men. This is, of course, not the case in India.
In the Indian case, government employment was mentioned as one of the components of the first stage out of poverty. It is interesting to note that in both the cases, being able to send children to school is an item in the first stage out of poverty, as is also the ownership of cattle. In the Serengeti/Ngorongoro case in Tanzania as well as in the Corbett case in India, the possibility of hiring a tractor is also a component of the first stage out of poverty. The second stage out of poverty, in both the cases, includes owning a brick house. Again, possession of about 4 ha of land also stands as a condition in both cases.
A description of the ‘quasi-longitudinal’ data produced by the surveys in the two countries is presented in the cross tabulation in Tables 2 and 3.
As can be seen from Tables 2 and 3, the upward movement has been more modest in the areas around the Ngorongoro CA and Serengeti NP in Tanzania, as compared to the Indian case. In Table 2, we see in the first row, that 74.2% of those in poverty in 1997 had moved up to the middle level ten years later, while only 1.7% had moved to the high level. In the Indian case, 25.4% moved from poverty to middle level, but 61%
moved two steps up. Also the reverse movement—falling into poverty—occurred to an extent in the Tanzanian cases; 36.3 % of those at the middle level in 1997 had fallen to poverty ten
years later. This has not happened to any considerable extent in India. However, 24.2% of those at the high level had fallen one step down to the middle level in the Indian case.
Also in the Tanzanian case, about 63% were in poverty ten years ago, while only 31% of those in the Indian case around Corbett NP were in that position at the time. The difference is also striking in 2007, where almost 50% in the Indian sample considered themselves to be placed in the ‘high’ living conditions category, while only 6% of the respondents in the Tanzanian case classified themselves thus. More than 28% in the Tanzanian sample considered themselves poor in 2007, while only 5% in the Indian sample did so.
As a more general observation, a large number of those in poverty in 1997 escaped from poverty in 2007 in both the cases.
In Tanzania, this figure is 75.9%, while in India this figure is 86.4%. One needs to bear in mind that the leap out of poverty is a big achievement in itself, and both regions have achieved it. An overview of how ‘stages of progress’ was quantified is given in Appendix 1. The matrix in Appendix 1 shows that the nine possible positions of development can be constructed with movements from possible positions 10 years ago to those of the present.
RESULTS IN INDIA Factors affecting ‘stages of progress’ in India
The regression results assessing the factors for ‘stages of progress’ movements in India are as given in the following equation.
Y = 0.0244 +0.085X1 X2+0.095 .X3+0.02.X4+
0.422 0.001 0.002
( ) ( ) ( ) 00.442 0.056180.00
= 187, 2= 0.12, . = 0.092
( ) ( )
n R adj R (4)
The figures within parentheses are the p-values of regression.
The result in equation (4) shows that the ‘difference in incomes from the sale of agricultural products’ and the ‘difference in incomes from tourism’ are statistically significant factors affecting ‘stages of progress’ (SOP) (at 5% levels). The difference in incomes from sale of agricultural products7 has essentially resulted from developments in agricultural marketing facilities and better infrastructure in and around the area, as also the processes of urbanisation affecting adjoining urban agglomerations like the town of Ramnagar. The growth of tourism has also been a prime factor in this context, as hotels, lodges, and eco-tourism initiatives have provided for a ‘ready’
market for agricultural products.
On the other hand, another important driver of the SOP has been income from tourism. We surveyed around 15 of the existing 25 lodges and found that more than 80% of the lodges surveyed came up after 2003. The lodges were mostly owned by urban residents of large cities like Delhi, or at times residents of the nearby town of Ramnagar. The lodges provided employment to the local population directly and indirectly. As a result, there was a decline in the population migrating away
from home to large cities in search of employment. Hence, the
‘difference in incomes earned by working in large cities’ has not made any significant contribution to SOP. A better profile of the lodges is provided in Table 4.
With nearly 59% of the overall employment in lodges coming from within the zone, and 52% of the managers being local inhabitants, it clearly goes to show that the lodges have primarily been run by the local population, as compared to Tanzania, where only 20% of the managers were local (as will be described later)—this is an interesting phenomenon to be noted. The other interesting feature to be noted here is that only 7–10% of the total number of tourists were of foreign origin, which is miniscule as compared to 90% of the same in Tanzania (Uddhammar and Ghosh 2009). This further justifies the contention that trained personnel who are employable as managers might not be in high demand in the Corbett zone since it as yet does not cater to international tourists to the same degree that Tanzania does.
Local educated people can serve the purpose of managing the pattern of tourismthat is domestic economy-centric.
As is evident from this discussion, a positive ‘difference in income from tourism’ has definitely resulted in an increase in SOP. As stated earlier, most of the lodges came up after 2003, and has resulted in a significant change in the standards of living of those employed by them. This has also substituted for incomes from other sources (like incomes from large cities, as was evident from our interviews), and has helped in supplementing other local sources of income (e.g., agriculture).
Therefore, one of the major drivers that led to ‘escape from poverty’ in 2007, as is noted from the results in Table 3, is development of tourism in the Corbett zone.
Changes in fauna sightings in Corbett Reserve: is tourism a determinant?
The critical species considered here include: tiger, elephant, barking deer, sambar, chital, leopard, nilgai, and wild boar, among others (Appendix 3; Table A.3.1).
Of all the 191 respondents, an overall negative value of the composite index was estimated for only two respondents, while all others reported a positive value. Interestingly, for the tiger, which is considered an ‘umbrella species’ in the zone, 188 respondents reported that the sightings had increased, while three respondents felt that the sightings had remained the same.
The respondents were also asked to state whether the importance of income sources had changed (increased, remained the same, or diminished). The results obtained were as follows:
Z= 0.033. = 0.0194.X5 X6 0.0397.X7+0.0299
0.12 0.314 0.064
− −
( ) ( ) ( ) .. + 0.62
= 0.055, . = 0.0344, = 191
8
2 2
0.016 X 0.00
R adj R n
( ) ( )
(5)
The regression equation (5) finds ‘change in importance of income from tourism’ as a significant variable, contributing positively to fauna sighting. The implication can be drawn
in the following manner. Households, which have been increasingly exposed to the tourism industry over time with the development of the sector in the Corbett NP, are exposed to higher sightings of species, as compared to those less exposed to the tourism industry. During the interactive sessions with the local people during both phases of interviews, the communities associated with tourism revealed having adopted a more positive outlook towards wildlife as animal sightings was what was driving the tourism industry. The increase in
‘fauna sighting change index’ is an indicator that animals were not treated with a negative mind set, as they used to be.
Rather, their presence was a welcome feature that helped the cause of tourism, thereby helping the community to generate more income out of tourism. This is where one might state that the perception of conservation (if a positive ‘fauna sighting change index’ is an indication) has only got better in the zones associated with tourism.
On the other hand, the variable ‘change in importance of income from working in major cities’ is a significant explanatory variable. The negative sign8 associated with this indicates that fauna sighting had generally diminished for those households who had an increased reliance on income from employment in major cities. A possible explanation of this can be that the ‘city-centric’ nature of these households make them less suitable for the natural species of the zone.
Hence, to summarise, while increased income from tourism implies higher fauna sightings, an increase in alternative income (from working in cities), implies a decrease in fauna sightings.
Are sightings higher in tourism-related villages?
We ran another regression with sighting index as the dependent variable, and with the two dummy variables related to tourism sites; one of these was for towns that supported tourism, and the other was for villages that supported tourism. The regression results are as follows:
191 11
2
= 2=
( ) ( ) ( )
0
0 0 00 0 00
, . ,
. . .
n R aadj R. 2=0 0. 9 Z = 0.0109 D1 + 0.106. D2 + 0.69
(6)
Our results show that indeed the sightings are higher in and around tourism-related villages, but not significantly so in tourism-related towns. This buttresses the results obtained in regression equation (5).
Perception-based results: verification with secondary observations
Most of what we have observed until now in the Indian case is based on perception, and this deserves to be verified with secondary level information as obtained from various other sources. The secondary information was obtained from WII (1999) and Jhala et al. (2008), analysed through ‘factor change’
in the various variables under consideration, and presented
in Table 5.
Table 5 shows that in Corbett NP in India, wildlife populations have expanded significantly, including that of the tiger. The annual number of tourist visitors has also increased considerably, albeit from relatively low levels. It is noteworthy that the annual number of visitors in the park has increased from 29,000 in 1986–1987 to 52,000 in 1997–1998, and finally to 120,000 in 2006–2007. The correlation between tiger—an important umbrella species of this ecosystem—and tourist numbers was 0.738, during the 1987–2006 period (Pearson’s r). Increased tourism and increased local human population did not hinder the increase in wildlife numbers. In fact, we find indicative evidence of better quality of park management leading to an increase in wildlife numbers. This, in turn, has led to the area becoming more attractive for tourists to visit, which, again in a circular turn of events, has led to better monitoring of wildlife by putting pressure on the park management to perform well and keep wildlife well protected.9
Thus, combining the perception-based survey and also these secondary observations that buttress the survey analysis results, we can draw the conclusion that for this protected area, efficient wildlife protection has worked side by side with tourism, resulting in the well-being of the surrounding local human population.
RESULTS IN TANZANIA Drivers of ‘stages of progress’
In Tanzania, an identical regression was run with SOP as the dependent variable, with the same explanatory variables as shown in equation 1 for the Indian case. The results are as follows:
Y= 0.047 +0.0314X1 X2+0.057 . +0.035 .X3 X4+
0.013 0.1 0.1 0.
( ) ( ) ( ) 113 0.0290.00
= 282, 2 = 0.065, = 0.0512
( ) ( )
n R adj.R (7)
In this case, ‘change in income from sale of livestock’ is the only statistically significant variable. It was also revealed from conversations during the initial pilot surveys that ‘livestock’
is an extremely valuable possession for inhabitants of the Ngorongoro area. The importance of livestock could be gauged from statements like “girls are important—they [can] be sold and I could get a cow.”10
However, we ran another regression to check whether the SOP movement is significantly better in villages where tourism is prevalent. The regression analysis gave a positive response to this concern.
Y D D
n R
= 0.01. 1+ 0.087. 2+ 0.046 = 282, = 0.03,
0.05 0.00 0.00
2
( ) ( ) ( )
aadj R. = 0.022 (8) There is a clear indication that the ‘stages of progress’
movement has been positive in zones where tourism exists.
This is prevalent for both villages and towns. However, in our previous reporting based on regression result (7), we did not find that tourism income is an important determinant of the ‘stages of progress.’ This may be because ‘livestock’ generally has evolved as an important component for income generation in the region, particularly after 1997, while tourism income (though prominent) might be concentrated in only a few villages and towns.
Drivers of fauna sightings
The critical species here are: lion, elephant, buffalo, wild dog, rhino, zebra, warthog, monkey, and fish, and the ‘fauna sighting change index’ has been constructed based on respective weights.
Here, the lion has emerged as the ‘umbrella species’ and has been given a weight of 0.25, while considering the ‘rarity’ aspect of zebra and rhino across space and time, both of them have been assigned a weight of 0.15 each (Appendix 3; Table A.3.2).
Interestingly, in Tanzania, out of 293 valid responses for changes in sightings, around 65 reported a negative ‘fauna sighting index’ value, reflecting a perception of decline in fauna sightings during the 10-year period, while 24 respondents revealed a score of ‘zero’ implying a state of no change in sightings. Two hundred and four respondents, i.e. 70% of the sample, reported an increase in sightings.
In fact, to find whether the importance of income from tourism has resulted in such a change, we attempted to run an identical regression as was attempted in equation (4). In the results, as given in equation (9), none of the variables are significant at 5% levels, though the tourism-related variable can be stated to be significant at 10% level of significance.
Z= 0.079 .X5 0.0018 .X6+ 0.068 .X7+ 0.041
0.12 0.533 0.1
( ) − ( ) ( ) .. + 0.43
= 0.043, . = 0.03, = 293
8
2 2
0.078 X 0.00
R adj R n
( ) ( )
(9)
The regression results, here, weakly exhibit some evidence of ‘change in importance of income from tourism,’ contributing positively to fauna sighting.
Fauna sightings in tourism-related areas
It has also been observed that fauna sighting is more in areas that are associated with tourism. This is shown in the following regression results.
n= R = ad
( ) ( ) ( )
0
0 00 0 00 0 00
, . ,
. . .
293 2 13 jj R. 2= 0 12. Z = 0.363.D1 + 0.422.D2 + 0.29
(10)
Now, if we combine the results obtained in equations (8) and (10), we find that there is clear indication that tourism- related areas reveal better animal sightings than other areas.
The results have also revealed a better movement in SOP than in areas not related to tourism.
What does the secondary data reveal?
In East Africa, significant population fluctuations have occurred in most species between the first and last measured figures.
However, we ignore that fluctuation and report on the overall trend during the period 1997–2006.As shown in Table 6, there has been a considerable increase in predators like lions over the period, while the numbers of elephants and buffalos have stayed more or less constant. All these species are important for tourism. Visitor numbers in the Serengeti-Ngorongoro area have not increased much during the 1997–2007 period (Table 6). However, a longer term trend reflects a large factor change in the number of tourists (see Table 6). On the other hand, the number of livestock owned in the Ngorongoro CA has increased sharply, as also the human population in the region. The increasing importance of livestock in the Tanzanian economy is recognised and was enhanced by the Agricultural and Livestock Policy 1997, where a host of incentives for livestock was provided. This policy shift might have been a driver of the livestock economy. With increasing human settlements around forest areas, the demand for manure, hides, and skins has been increasing. Apart from that, livestock is also a potential source of draught power for transport and cultivation activities. Interestingly, some respondents also felt that livestock are a potential guard against price rise.
In Table 6, we further find that during this period there was also a sharp increase in wildlife. The number of tourists visiting this site also increased. Therefore, we find a positive, high, and statistically significant correlation of tourist visitors with the elephant and lion population (Table 6). Although considerable fluctuations have occurred within the period, the figures give an indication of the long-term trends (Packer et al. 2005).
Hence, with tourism already at a very high level, with more than half a million tourists visiting the protected areas of East Africa every year, the importance of tourism has increased over the last two decades. This is almost five times that of the number of tourists visiting Corbett NP annually. It has remained at a very high level during the period 1997–2007. We may thus argue that with reference to the base period, the importance of tourism income has not changed; neither has it been responsible for changing the ‘stages of progress’ for the entire area as a whole during this phase. But, again, the regions associated with tourism have benefited more in terms of SOP, than other regions. There is no doubt that there has been a simultaneous expansion of wildlife, local human, and livestock populations.
Table 6 clearly reveals the positive correlation between wildlife populations and the number of tourists visiting the NP/ CA.
Again, we detect a strong possibility of a causal factor from good wildlife management leading to increased tourism visits, leading to better monitoring of wildlife, and also to increased pressure on wildlife authorities to maintain high standards of wildlife protection. This observation is supported by other research findings that concur that tourism in this area has positively affected wildlife (NINA 2007).
It further needs to be noted that tourism has had a sustained impact on the standards of living, while the domestic economy
has also benefited from it. Profiles of the lodges surveyed reveals this to a certain extent (Table 7).
The total number of people working for and earning a livelihood from the tourism sector in Serengeti NP/
Ngorongoro CA is almost three times that of those in Corbett NP in India. While a large proportion of employed personnel in the lodges come from the local area only, quite unlike in Corbett NP, only 20% of the managers are locals. The demand for more trained personnel from outside the region in Tanzania is prevalent primarily to cater to the international nature of tourism in the zone.
CONCLUDING REMARKS
The results of the analyses of the data in the two cases presented here have some differences and some similarities, though there seem to be indications of broad similarities in terms of the conclusions that we may draw. From the secondary data, we find that in both the cases, an increase in the number of key species such as lion, tiger, buffalo, and elephant has occurred parallel to a similar increase in tourist visitors. Factors such as the expansion of tourism and an increased human population in general (as shown in Tables 5 and 6 in terms of factor change) around protected areas have not affected wildlife negatively.
Rather, wildlife and tourism have expanded simultaneously.
In both the cases, respondents showed more awareness of the opportunities that tourism had created for them in terms of income and employment. As is evident from our regression results, we find that sightings generally have increased in regions where importance of tourism and tourism-related income are more or have increased over time.
The change in the standards of living (as reflected in the SOP movement) because of changes in incomes from tourism is more prominent in the Indian case, and the causality is not so prominent in the case of Tanzania, where lately, income from livestock has emerged as an important determinant for change in economic status. However, SOP movement in the Serengeti-Ngorongoro region has revealed an interesting characteristic of being more positively related to tourism-affected areas than other areas. On the other hand, one may even note that livestock herding has not affected conservation efforts in Tanzania, as is often expected.
In Tanzania, the other important aspect to be noted is that in our reference period, tourism was already at a high level of development, and not much factor change was noted even in the secondary data. But, the long-term changes definitely show that tourism has developed over time in a big way. In any case, these are general causal links that can be noted here. One plausible mechanism at work is that a rise in species count, or more specifically, sightings, might have attracted more visitors to the sites. Word of mouth information also quickly spreads via electronic media from those who visit the sites.
Another causal link at work is that of alternative land-use and biodiversity—a relationship not in the ambit of this article.
Only in parts of the Serengeti-Ngorongoro where low-yield cattle herding is practiced, wildlife and the local rural economy coexist. With farming, the human-wildlife conflict increases
