THE ECONOMIC VALUE OF
THE MOUNTAIN GORILLA PROTECTED FORESTS
(The Virungas and Bwindi Impenetrable National Park)
Photo: Jose Kalpers
FINAL REPORT
Based on original fieldwork conducted in 2002-2003 supplemented by additional fieldwork in 2005
Submitted to the International Gorilla Conservation Programme (IGCP), Nairobi, Kenya
by Richard Hatfield and Delphine Malleret-King
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
LIST OF CONTENTS
List of Annexes ix
List of Acronyms x
Acknowledgements xi
1 INTRODUCTION 1
1.1 Management and threats ............................................................................................ 1
1.2 Study aims and objectives ......................................................................................... 2
1.3 Study Area description .............................................................................................. 2
2 OPPORTUNITY COST OF THE PROTECTED FORESTS: The economic value of
agriculture at the forest edge 4
2.1 Opportunity cost ........................................................................................................ 4
2.2 Section objective ....................................................................................................... 4
2.3 Methodology.............................................................................................................. 4
2.4 Analysis of production systems ................................................................................. 4
2.5 Section organisation .................................................................................................. 4
2.6 Main characteristics of farming systems around the Virunga and Bwindi forests .... 4
2.6.1 Factors of production ......................................................................................... 5
2.6.1.1 Labour............................................................................................................ 5
2.6.1.1 ............................................................................................................................ 5
2.6.1.2 Land ............................................................................................................... 5
2.6.1.3 Capital/Technology ....................................................................................... 6
2.6.1.4 Soil quality..................................................................................................... 6
2.6.2 Production systems ............................................................................................ 7
2.6.2.1 Crops grown .................................................................................................. 7
2.6.2.2 Crops consumed vs. sold ............................................................................... 8
2.6.2.3 Livestock ......................................................................................................10
2.6.2.4 Livestock as a source of income ...................................................................10
2.6.2.5 Woodlot use ..................................................................................................11
2.6.2.6 Honey ...........................................................................................................11
2.7 Quantitative presentation of the costs and income of farming systems ....................12
2.7.1 Net farming benefits .........................................................................................12
2.7.1.1 Crops.............................................................................................................12
2.7.1.2 Livestock ......................................................................................................14
2.7.1.3 Woodlots.......................................................................................................16
2.7.1.4 Total net income ...........................................................................................17
2.7.2 Opportunity cost of the forests to local communities .......................................18
2.7.3 Whose opportunity cost? ..................................................................................19
2.8 Future trends: main factors impacting the opportunity cost of forests in the future .20
2.8.1 Soil fertility and erosion ...................................................................................20
2.8.2 Agricultural productivity ..................................................................................20
2.8.3 Lack of economic alternatives ..........................................................................20
2.8.4 Potential returns to capital ................................................................................21
2.8.5 Population growth ............................................................................................21
3 TOTAL ECONOMIC VALUE OF THE VIRUNGA & BWINDI PROTECTED
FORESTS (PNV, PNVi-Sud, MGNP & BINP) 22
3.1 Forest benefits ..........................................................................................................22
3.2 Forest costs ...............................................................................................................22
3.3 Net forest benefits (conservation value) ...................................................................22
3.4 Section organisation: ................................................................................................23
3.5 Direct use: the value of products extracted from the forest ......................................23
3.5.1 Current direct-use benefits: forest products ......................................................23
ii
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
3.5.2 Potential direct-use benefits: timber .................................................................24
3.6 Indirect (non-extractive) use: the value of gorilla-based tourism .............................24
3.6.1 Introduction and section objectives ..................................................................24
3.6.2 Estimation of gorilla tourism benefits: the Travel Cost Method (TCM) ..........24
3.6.3 Methodology.....................................................................................................25
3.6.4 Results ..............................................................................................................26
3.6.4.1 Economic value of viewing gorillas to international visitors .......................26
3.6.4.2 Economic impacts of gorilla tourism............................................................29
3.6.4.3 Summary: the economic value of gorilla viewing ........................................36
3.7 Non-use: benefits and costs derived from the forests’ existence ..............................43
3.7.1 Introduction: types of value and cost ................................................................43
3.7.2 Section objectives .............................................................................................43
3.7.3 Estimation of forest non-use benefits-costs: Contingent Valuation .................43
3.7.4 CVM Design .....................................................................................................44
3.7.5 The CVM questionnaires ..................................................................................45
3.7.5.1 Local CVM questionnaire ............................................................................46
3.7.5.2 National CVM questionnaire ........................................................................46
3.7.5.3 International CVM questionnaire .................................................................46
3.7.5.4 The Gishwati, Rwanda CVM questionnaire .................................................47
3.7.6 Results ..............................................................................................................47
3.7.6.1 Local level ....................................................................................................47
3.7.6.2 National level................................................................................................57
3.7.6.3 International level .........................................................................................59
3.7.6.4 The case of Gishwati Forest .........................................................................61
3.7.6.5 International ecological benefits ...................................................................63
3.8 Valuation summary: the short-term and long-term worth of the protected forests...64
4 POLICY CONCLUSIONS, BENEFIT-COST DYNAMICS & CONSERVATION
OPTIONS 68
4.1 Policy conclusions ....................................................................................................68
4.2 Benefit-cost dynamics and trends .............................................................................68
4.2.1 Time preference: the impact of limited planning horizon on forest benefit
levels 68
4.2.2 Opportunity cost: the cost to local communities of the protected forests. ........69
4.2.3 Agricultural productivity: intensification or expansion? ..................................69
4.2.4 Soil erosion and decreasing fertility .................................................................69
4.2.5 International gorilla tourism .............................................................................70
4.2.6 International non-use value ..............................................................................70
4.2.7 Returns to investment into rural communities ..................................................70
4.3 Scenarios and outcomes............................................................................................71
References 75
iii
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
List of Figures
Fig. 2.1: Percentage of households per country using the different tools 6
Fig. 2.2: Percentage of households using: fertilisers, insecticides, other inputs or fallow 7
Fig. 2.3: Percentage of households growing identified crops by country 8
Fig. 2.4: Percentage of households with livestock and/or poultry 10
Fig. 2.5: Percentage of household using, selling their own wood and percentage of households
buying wood. 11
Fig. 2.6: Percentage of households producing honey 12
Fig. 3.1: International gorilla visitors by park 2000-2005 26
Fig. 3.2: Distribution of international gorilla visitor expenditure per person, by geographical
grouping 29
Fig. 3.3: Annual gross direct and indirect revenues accruing to Buhoma community (2004-5)
34
Fig. 3.4: Source from which local tourism-related revenues are derived 34
Fig. 3.5: Distribution of benefits derived from gorilla tourism (annual basis) 37
Fig. 3.6: Distribution of gross benefits between international, national and local levels (annual
basis – see Table 3.9) 37
Fig. 3.6a: Long-term value of forest benefits and costs from differing perspectives 40
Fig. 3.7: Actual vs. potential international gorilla visitors by park 41
Fig. 3.8: Actual vs. potential long-term gorilla-viewing value according to different discount
rates (stakeholder perspectives), 2001-02 vs. 2004-05 42
Fig. 3.9: Distribution of WTP bids (willingness-to-pay for forest protection) amongst
residents living within 5 kms. of a protected forest park (US$). 50
Fig. 3.10: Type of WTP bid from local residents, by park 50
Fig. 3.11: Distribution of sample WTA bids (US$) 53
Fig. 3.12: Distribution of forest protection WTP bids for Kabale, Uganda residents (note:
mean ‘other’ bid = US$2.46) 58
Fig. 3.13: Distribution of forest protection WTP bids for Goma, DRC residents note: (mean
‘other’ bid = US$1.56) 58
Fig. 3.14: Perceived change in water regime by Gishwati farmers 61
Fig. 3.15: Perceived change in soil fertility among Gishwati farmers 62
Fig. 3.16: Annual distribution of gross use and non-use benefits from Virunga and Bwindi
forests (excluding international non-use value). 64
Fig. 3.17: Annual distribution of gross benefits from Virunga and Bwindi forests including
international non-use value. 65
iv
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Fig. 3.18: Annual bet benefits and losses from the Virungas and Bwindi forests according to
local, national, and international level excluding international non-use value (US$
millions) and when forest opportunity costs are assigned to the local level. 66
Fig. 3.19: Annual net benefits and losses from the Virungas and Bwindi forests according to
local, national, and international level including international non-use value (US$
millions) and when forest opportunity costs are assigned to the local level. 66
Fig. 3.20: Annual bet benefits and losses from the Virungas and Bwindi forests according to
local, national, and international level excluding international non-use value (US$
millions) and when forest opportunity costs are now assigned to the national level as
opposed to local level. 67
Fig. 3.21: Annual bet benefits and losses from the Virungas and Bwindi forests according to
local, national, and international level including international non-use value (US$
millions) and when forest opportunity costs are now assigned to the national level as
opposed to local level. 67
Fig. 4.1: Scenarios depicting future impact of economic dynamics on 6 key benefit-cost
components 72
v
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
List of Tables
Table 2.1: Average household size, number of members of the households working on the
farm and the number of dependents per active household member 5
Table 2.2: Casual agricultural worker wage rates 5
Table 2.3: Percentage of households renting at least one cultivated plot 5
Table 2.4: Average surface cultivated (in ha.) per household and active household members 6
Table 2.5: Percentage of households using fertilisers and insecticides by crop 7
Table 2.6: Average number of crops grown per household 8
Table 2.7: Percentage of crop value derived from sale 9
Table 2.8: Importance of the crop in terms of household total crop value (median rank of sales
+ home consumption) 9
Table 2.9: Percentage consumption vs. sale in relation to income provided by animal
husbandry 10
Table 2.10: Percentage value animal product vs. animals in relation to total income provided
by livestock 10
Table 2.11: Average size of woodlot 11
Table 2.12: Total annual crop income, costs & net income per ha (US$) 12
Table 2.13: Pearson's correlation between net income and distance from park* 14
Table 2.14: Total annual animal husbandry income, costs & net income per hectare when
animals are kept 14
Table 2.15: Annual net income per ha of woodlot (US$) 16
Table 2.16: Summary of annual net income per ha. of woodlot (US$) 16
Table 2.17: Average annual net household income per hectare (US$) 17
Table 2.18: Annual opportunity cost of the protected forests in terms of lost production 18
Table 2.19: Net present value of forest opportunity cost (US$ millions) 19
Table 2.20: Number of farming households that the parks represent, assuming 50% cultivable
area 19
Table 2.21: Percentage of households identifying lack of land and soil fertility as a constraint
to farming development 20
Table 2.22: Engagement and encouragement of children in agriculture 21
Table 2.23: Returns to capital investment (ROI) in farming (US$) 21
Table 3.1: Summary of forest uses, with corresponding economic value type, estimation
method and data source/location 23
Table 3.2: Average annual visitor numbers and origin by park 27
Table 3.3: Annual value to international visitors of viewing gorillas according to 3 model
estimates 28
vi
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Table 3.4: Distribution of international visitors’ per person and annual expenditure between
local, national, regional and international levels based on base year visitor levels
(2001-2) 30
Table 3.5: Economic contribution of gorilla-viewing to the national economies of Uganda
and Rwanda: a comparison between the current study and a previous study 31
Table 3.6: Annual gross direct and indirect revenues accruing to Buhoma community (2004-
5) 33
Table 3.7: Main benefit and cost impacts cited in Buhoma from tourism revenue 35
Table 3.8: Annual value of gross benefit & impacts attributed to gorilla viewing (base year)
36
Table 3.9: Long-term value of gorilla-viewing benefits by stakeholder level (US$ millions) 39
Table 3.10: Capitalized value of total benefits & in-country benefits 40
Table 3.11: Potential annual amounts and net present value (NPV) of benefits from gorilla
viewing among stakeholders, at full capacity 41
Table 3.12a: Weighted scores for benefit categories by park 48
Table 3.12b: Weighted score for cost categories by park 48
Table 3.13: Overall attitude towards forest after rankings of benefits and costs, by location 49
Table 3.14: Average willingness-to-pay (WTP) bid for forest protection by residents
surrounding parks, according to park and distance away from park 51
Table 3.15: Willingness-to-pay (WTP) for forest protection by residents surrounding parks,
by parish/location 51
Table 3.16: Annual WTP (willingness to pay) for forest protection amongst local residents,
according to 3 different estimates (US$) 53
Table 3.17: NPV (net present value) of local level WTP under differing discount rates (US$)
53
Table 3.18: Average willingness-to-accept (WTA) compensation bids for forest costs borne
by residents surrounding parks, also expressed as % of annual household income 54
Table 3.19: Willingness-to-accept (WTA) the cost of lost opportunities and direct costs by
residents surrounding parks, by parish/location 55
Table 3.20: Annual WTA (willingness to accept) compensation for costs to local residents
incurred by presence of the forests 56
Table 3.21: NPV (net present value) of local level WTA under differing discount rates (US$
millions) 56
Table 3.22: Attitudes of town residents towards the forest 57
Table 3.23 Overall attitude towards the protected forest 57
Table 3.24: NPV (net present value) of forest protection WTP (US$ millions) at regional level
59
vii
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Table 3.25: Mean percentage distributions for Environmental Attitude items 59
Table 3.26: Pro-environmental attitude and type of bid response for WTP (willingness to pay)
for mountain gorilla protection* 60
Table 3.27: The importance of different motivations for protecting mountain gorillas 60
Table 3.28: NPV (net present value) of the non-use of mountain gorilla habitat to international
citizens 61
Table 3.29: WTP for reafforestation by Gishwati farmers according to perceived change in
waterflow regime 62
Table 3.30: WTP for reafforestation by Gishwati farmers according to perceived change in
soil fertility 62
Table 3.31: NPV (net present value) of forest protection WTP 63
Table 3.32: NPV (net present value) of carbon sequestration by Virunga and Bwindi forests
(US$ millions) 64
Table 4.33: Summary of short-term (annual) and long-term value (NPV) accruing to direct-
use, indirect-use, and non-use forest benefit & cost components (US$ millions) 65
Table 4.34: Short-term (annual) and long-term (NPV) distribution of forest net benefits
between international, national, and local levels (US$ millions) 66
viii
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
LIST OF ANNEXES
Annex 1: Notes on sampling design, sampling strategy, and field methodology
Annex 2: Notes on prices, exchange rates and computation of farm income
Annex 3: Farm Production Systems questionnaire
Annex 4: Crop mix and diversity by country, parish, and distance zone
Annex 5: The Travel Cost and Contingent Valuation questionnaires
Annex 6: Travel Cost demand estimation results for international gorilla-park visitors sample
Annex 7: WTP/WTA model specification and estimation, and basis for extrapolation to the
wider population
ix
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
LIST OF ACRONYMS
BINP Bwindi Impenetrable National Forest, Uganda
CARE-DTC CARE Development-Through-Conservation programme, BINP
CV Contingent Valuation (economic research)
CVM Contingent Valuation Method (economic research)
IGCP International Gorilla Conservation Programme
MGNP Mgahinga Gorilla National Park, Uganda
NPV Net present value (long-term economic worth)
PNV Parc National des Volcans, Rwanda
PNVi-Sud Parc National des Virunga – Southern Sector (Mikeno), DRC
TCM Travel Cost Method (economic research)
WCS Wildlife Conservation Society
WTP Willingness-to-pay (for environmental benefits)
WTA Willingness-to-accept (compensation for environmental costs)
x
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
ACKNOWLEDGEMENTS
We would like to thank the entire of staff of the IGCP country offices in DRC, Rwanda,
Uganda and Kenya for all their support and assistance during the course of this study.
Our thanks also go to the wardens and staff of BINP, MGNP, PNV, and PNVi-Sud National
Parks, as well as the staff of Buhoma and Mgahinga Community Campgrounds, for their great
cooperation, hospitality and assistance in the field. In addition, our thanks go to the
enumerators and supervisors in the field, who worked willingly and diligently, and with
patience.
Gratitude also goes to the numerous respondents who patiently answered the surveys, and
especially to the Buhoma community, Bwindi, for the extended information they gave us.
This work was made financially possible by the Howard G. Buffett Foundation, whose
support is gratefully acknowledged.
xi
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
THE ECONOMIC VALUE OF THE VIRUNGA AND BWINDI PROTECTED FORESTS
1 INTRODUCTION
Situated in the Albertine branch of Eastern Africa’s Great Rift Valley, at the border confluences of
Uganda to the north, Rwanda to the south, and DRC (Democratic Republic of Congo) to the west, lie
two isolated and protected afromontane forests – Virunga and Bwindi - part of a once more extensive
forest tract that now supports one of the highest human population densities in Africa. A Pleistocene
refugium unique in its levels of endemism and species richness, the Albertine Rift is considered one
of the most bio-diverse areas of the world and commonly considered to be one of the top 20 of the
Global 200 priority areas for biodiversity; moreover, the Virunga-Bwindi region, centrally located
within the Albertine Rift, represents one of the area’s biodiversity centres (Lanjouw, 2003).
Of particular significance is the fact that the two forests contain the world’s only remaining natural
populations of mountain gorilla (Gorilla beringei beringei), currently numbered at around 650
individuals. Despite the existence of long-running conservation programs that have assisted
protection, generated substantial tourism income, and enhanced local appreciation of the forests, their
long-term viability remains threatened by land pressure exacerbated by large-scale political conflict.
It is common belief in many circles that the long-term interests of not only global but national and, in
particular, local society will be better served by conserving rather than converting these forests. As
such, they remain undervalued.
One reason for this is that forests tend to be undervalued on a global scale. It is increasingly
recognised that forests provide significant benefits which to date have received little attention, either
due to lack of knowledge or difficulty in quantification. Primary amongst these is the value of the
ecological services provided by the forests – for example, the benefits to agricultural production of
climate control; regulation of water flow; and soil retention; or the wider benefits of atmospheric
pollution control. Other less obvious but highly-valued benefits include amongst others: biodiversity
value (including flora, fauna, and invertebrates); aesthetic value; value to future generations; and
ethical value.
A second reason involves the social context within which the forests exist. Since the relationship
between economic and ecological systems is necessarily dictated by human perception, the value of a
resource is subject to the socio-political-economic landscape – the forces of which can mitigate in
favour of, or against, conservation both in the short-term and the long-term. The social context
surrounding the Virunga and Bwindi forests does not favour long-term conservation. The challenge
for conservationists is to explore mechanisms that meet conservation goals while gaining long-term
approval within the social and economic context.
1.1 MANAGEMENT AND THREATS
The two forests are administered as national parks. Bwindi Impenetrable Forest National Park (BINP)
borders DRC but lies wholly in south-west Uganda, whilst 35 kilometres to the southwest, the
Virunga forest is shared by three countries, resulting in three adjacent national parks: Parc National de
Volcans (PNV) in Rwanda; the Mikeno section of Parc de Virunga Sud (PNVi-Sud) in DRC; and
Mgahinga Gorilla National Park (MGNP) in Uganda.
The most visible activity has been the success of mountain gorilla viewing tourism, which has
developed into a world-wide attraction since the early 1980s, at times accommodating up to 15,000
annual visitors paying premium viewing fees of up to $375 per day. There is no doubt that gorilla
tourism has generated significant and tangible benefits for the countries involved. However, land
pressure immediately around the forests is high and growing, due to a combination of poverty and
population growth. With the perception that the main beneficiaries of the forests are international
1
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
tourists and the government, resentment by the surrounding population, who lack land and have to
face crop raiding by wildlife harboured by the forests, triggers both animosity and poaching.
Redressing of this imbalance is vital to sustainable conservation of the forests: increased rules,
regulations and policing can assist in the shorter term, however, the final and necessary tenant of any
successful property regime is that it is perceived to be socially just.
Exacerbating this situation has been widespread political conflict and instability, including the 1994
Rwandan genocide; the 1996 and 1998 civil wars of DRC – that still render tourism closed; and the
highly publicised killing of tourists in BINP in 1999. In addition to negatively impacting tourism, the
conflicts have further contributed to land pressure and natural resource conversion by: creating
uncertainty and therefore short-term gain; the resettling of displaced populations on the only available
land – national parks and reserves; and increasing poverty. Such developments have increasingly
made the existence of protected areas more difficult to justify politically.
At the same time, the relative stability experienced in the region since 2000 has resulted in
significantly improved visitor levels and gorilla revenues, showcasing its potential value and
economic role in the region.
1.2 STUDY AIMS AND OBJECTIVES
This study seeks to:
Provide policy-makers with information on the short- and long-term benefits of the two
protected forests, against the costs borne by their existence
Ascertain the distribution of costs and benefits between local, national and international levels
Examine the short-term and longer-term trade-offs between competing land uses and
stakeholder levels, and their implications for conservation policy
Provide a suitable framework for economic analysis that can be applied to other similar areas in
the greater region as well as for further studies pertaining to the habitats of the mountain gorilla.
The study comprises three main components:
Objective 1: To estimate the current total economic value of farm production being carried out
around the forests
Objective 2: To estimate the total economic value of the four protected forests parks, including
distribution of costs and benefits amongst stakeholders
Objective 3: To identify important economic dynamics and explore the implications for land-
use and conservation policy
1.3 STUDY AREA DESCRIPTION
The BINP was created in 1991 and protects one of the most biodiverse afro-montane forests of the
world, with a number of endemic trees, plants and birds (Cunningham, 1996). The park consists of
32,100 ha of rugged land, steep narrow valleys bordered by hill crests of 1200 m of altitude in the
north an 2600 m in the south. Before being established as a National Park in 1991, the area was set up
as forest reserve in 1932 and a wildlife reserve in 1964. Throughout this period, timber was exploited
and it is estimated that about 30% of the forest was cleared between 1954-91. Due to its unique
characteristics and to its richness, it was declared a UNESCO World Heritage List in 1995 (Wild and
Mutebi, 1996).
The Virunga forests cloaks the chain of six spectacular volcanoes that comprise the Virunga range, or
massif. The Parc de Volcans portion in Rwanda after several excisions covers 160 sq. kms. of higher
altitude forest. Originally an area of 340 sq. kms. at the western end of the Virunga chain, it was
protected in 1925, becoming the first national park in Africa; then incorporated into the larger Albert
National Park in 1929 (now the Parc des Virunga in DRC), before becoming the Parc des Volcans in
1960. On the DRC side the forest park covers an area of 250 sq. km. attached to the southern section
2
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
of the much larger Parc des Virunga (PNVi), and is known as the Mikeno section. The Mgahinga
Gorilla National Park (MGNP) covers an area of approximately 27 sq. km. on the Ugandan side.
3
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
2 OPPORTUNITY COST OF THE PROTECTED FORESTS: The economic value of
agriculture at the forest edge
2.1 OPPORTUNITY COST
Opportunity cost is the foregone benefits or the opportunity lost from undertaking an activity or
approach. Most households in the Virunga and Bwindi region depend on farming for their livelihood.
The decision to protect forest areas through the establishment of national parks will thus have an
opportunity cost in terms of farming activities since the majority of households depend on farming for
their livelihood – up to 90% in Rwanda, for example (Bizimungu, 2002).
2.2 SECTION OBJECTIVE
The objective of this section is to determine the economic value of farming within a 5 km. zone
around the Virunga and Bwindi forests, and on this basis estimate the opportunity cost of the National
Parks.
2.3 METHODOLOGY
The main cost and income derived from the land around the four parks were investigated through
household surveys. The net benefits from farming are calculated using the major incomes and costs
from the main land-based activities - farming, animal husbandry and woodlands. Other uses were
minimal and were not taken into consideration (see Annex 1 for detailed description and further notes
on ‘Production Systems’ methodology; Annex 2 for notes on prices, exchange rates, and calculations;
Annex 3 for the ‘Production Systems’ survey questionnaire).
2.4 ANALYSIS OF PRODUCTION SYSTEMS
The results of the surveys were analysed using basic statistics and multiple factor analysis such as
ANOVA for ordinal data.
The data was aggregated at different levels to identify whether significant variations across these
levels were detected:
national level (Uganda, Rwanda, DRC)
parish (location) level
zone level (km. distance from park from 1-5)
regional level (Virunga region versus Bwindi region).
2.5 SECTION ORGANISATION
Results are organized in three main sections:
Description of the main characteristics of the farming systems around the Virunga and Bwindi
forests
Quantification of farming system costs and income
Other factors impacting the evolution of the cost of opportunity
2.6 MAIN CHARACTERISTICS OF FARMING SYSTEMS AROUND THE VIRUNGA AND BWINDI
FORESTS
This section emphasises the economic livelihoods of the population surrounding the protected forests
rather than its social demographics, since the latter was the central concern of a recent IGCP-
facilitated socio-economic baseline survey (IGCP/WCS/CARE, 2004).
4
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
2.6.1 Factors of production
Production factors in farming activities are composed of labour, land and capital/technology. The
following results are based on the data collected through households surveys carried out in August
and September (during the long dry season) 2002.
2.6.1.1 Labour
Average farming household size is shown in Table 2.1 and varies from 4.9 to 6.7 members. In most
households the household members participate to farming activities on average with a ratio of 2
dependents per active household member on the farm. Extra labour is occasionally hired in peak times
or to help on major tasks, labour intensive work particularly during harvest or for heavy work such as
digging trenches. Wage rates are depicted in Table 2.2. The average daily cost of labour is similar
across the three countries.
Table 2.1: Average household size, number of members of the households working on the farm
and the number of dependents per active household member
Household Number Active member/
size active Household
members
Rwanda 6.5 3.1 0.5
DRC 6 3.4 0.6
Uganda 6 3.2 0.6
Mgahinga 4.9 3.5 0.6
Bwindi 6.7 3.6 0.6
Virunga 6 3.5 0.6
Total 6.2 3.2 0.5
Table 2.2: Casual agricultural worker wage rates
Local currency USD
Country Rwanda DRC Uganda Rwanda DRC Uganda
Daily cost of labour 300 150 1200 0.6 0.54 0.67
2.6.1.2 Land
In the Virunga and Bwindi areas land can be acquired and owned. A number of households however
rent fields to cultivate. These are rented for a season or for a year. Around Bwindi 60% of the
interviewed households rented at least one plot (Table 2.3).
Table 2.3: Percentage of households renting at least one cultivated plot
Rwanda DRC Uganda Virunga Total
Mgahinga Bwindi Total
% household 38.9 22.8 3.8 60 24.1 26.7 28.4
The average surface cultivated per household (Table 2.4) varies between 2.4 ha. around Bwindi and
0.7 ha. on the DRC side of Virunga, with between 0.8 and 0.2 ha. per active household member. The
size cultivated vary significantly across countries (ANOVA: F=12.340, p 0.849
Two significant but opposing results occur: in Rwanda crop net income increases with distance from
the Park; whereas in DRC it decreases. The reasons for this are unclear. One possibility is that
cultivation occurs up to a higher altitude on the Rwandan side, negatively impacting crop yields due
to colder conditions and/or more shallow/inferior soils. Conditions may work conversely on the
opposite (DRC) side of the range, rainfall in particular. No significant results were found for Uganda.
2.7.1.2 Livestock
As discussed earlier, income from livestock comes from sales and consumption of livestock itself
although home consumption represents a small portion of the total income from livestock (see section
2.6.2.4). Costs associated with livestock are antibiotics, veterinary fees, and shelter built for the
livestock. Very few households interviewed bought fodders or feed for their livestock, which are
generally fed on the households plot verges.
Average income only took consideration of households with livestock. This is not the case when total
income per ha. is calculated.
Table 2.14 presents the average net income from livestock by country; distance zone; parish / secteur
/ localite; and national level. ANOVA tests were carried out to investigate for which situations net
income varied significantly.
Table 2.14: Total annual animal husbandry income, costs & net income per hectare when
animals are kept
Rwanda Total income Net income Total cost N
ANOVA (parish) NS NS NS 50
Parish 1 2.051 0.842 1.209 9
2 139.056 137.683 1.373 17
3 30.647 28.355 2.293 13
4 89.088 172.588 .9 16.5 11
ANOVA (Zones) NS NS F=3.26 p= 0.01 50
Zone 1 143.548 142.160 1.388 13
2 29.776 28.243 1.533 13
3 32.748 30.679 2.069 14
4 221.494 208.449 13.044 7
5 276.000 234.763 41.237 2
Rwanda All 97.2 92.3 4.9 50
14
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
DRC Total income An Net income Total cost N
ANOVA (Parish) NS NS NS
Parish 1 65.341 61.492 3.849 17
2 22.672 22.672 0.0 11
3 122.449 122.449 0.0 2
4 0 0 0 0
5 0 0 0 0
ANOVA NS NS NS
(zones)
Zone 1 21.418 21.418 0.0 6
2 57.464 56.357 1.106 5
3 42.414 41.261 1.153 6
4 37.859 28.970 8.889 5
5 93.186 92.119 1.067 8
Total 51.322 35.03 2.181 30
DRC
Uganda Total income Net income Total cost N
ANOVA (Parish) F=2.6 p=0.05 F=2.8 F=6.8 p 0.005)
The model correctly predicted 79% of outcomes, implying that sample responses were non-
random. However, the model had far greater difficulty predicting ‘yes’ responses as opposed to
‘no’ responses (7% vs. 97% correct)
Level of bid offer ($5, $10, $20) and formal education level as a whole had a significant effect
on probability of ‘yes/no’ response (p=0.05 and 0.003 respectively). However no single specific
level of education (no school; primary school; secondary school; university) was found to be
significant factor explaining response. While this may be somewhat unexpected, the lack of
strong relationship is likely influenced by a number of other possible factors, for example, civic
environmental education carried out in both Rwanda and Uganda in recent years; the forest
multi-use program in BINP and MGNP; a general lack of environmental education within the
formal education system; and traditional knowledge of the forest value
Surprisingly, response was found to independent of household income – which might be
expected to be a major determinant of WTP. This suggests other factors coming into play: for
example, respondents may have responded according to willingness to pay with less regard for
ability to pay; ‘yea-saying” (giving the correct answer); or, it may be that forest protection is no
more of a priority for those who are better off (i.e. it is not a ‘luxury good’ that induces greater
spending as income rises). Alternatively, while income levels may vary relative to respondents,
income levels are low (lack significant variation) in absolute terms (sample income mean is
US$305 with standard deviation of $452, with maximum level of income $3744. Otherwise it is
possible that other factors not included in the model - random or otherwise - may be exerting a
greater influence on response than the income effect – for example, plausibility of the CV
scenario, payment mechanisms, who-pays, etc.
Mean WTP from the model was estimated to be US$3.90. This compares with US$4.62
computed for the sample mean (see Table 3.14). Thus, having taken into account the
distribution of the results from a statistical viewpoint, the model produces a similar but more
conservative estimate.
Cases for which the model predicted least well (“outliers”) were tabulated. The results showed
a total of 12 outliers. However, no concentration in one particular zone or parish is evident.
Aggregating the results for mean WTP over the number of households in the 5 km. zone
surrounding each park (see Annex 7) then yields an estimate for annual value of the forest to
that population. Table 3.16 gives results for taking three possible values for average WTP
(overall sample; the more conservative model estimate; and sub-sample for each park) and
52
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
assuming a population density of 400 per sq. km. for the rich volcanic soils of the Virunga
massif and 200 per sq. km. for the poorer soils surrounding BINP.
Table 3.16: Annual WTP (willingness to pay) for forest protection amongst local residents,
according to 3 different estimates (US$)
Estimate1: Estimate 2: Estimate 3: park sub-sample
overall sample model
Park All parks All parks PNV PNVi-S MGNP BINP
annual WTP 4.62 3.90 2.91 2.76 7.47 5.48
no. households* 53,548 53,548 18,462 14,666 4898 15,522
sub-total 53,724 40,478 36,588 85,061
total annual WTP 247,392 208,837 215,851
*
see Annex 7, assuming population density of 400/sq.km. for MGNP, PNV & PNVi-S and
200/sq.km. for BINP
However, as discussed earlier, actual value depends discount rate. Table 3.17 expresses total worth
(or net present value) over a 29-year time horizon for the three estimates of annual value, according to
discount rate assumed. Thus for the middle annual value (park sub-sample), for example,
‘conservation’ value may be $6.26 million (NPV assuming a 0% discount rate), while the ‘societal’
value may be $3.27 or $2.02, depending on whether a 5% or 10% ‘social’ discount rate is used to
compute NPV. Alternatively, NPV is US$1.07 when a ‘commercial’ discount rate of 20% is applied,
while under a discount rate of 30% - more applicable to a rural subsistence economy - forest worth
falls to US$0.72 to local residents (see section 3.6.4.3 for further discussion).
Table 3.17: NPV (net present value) of local level WTP under differing discount rates (US$)
Annual NPV under differing discount rates (US$ millions)
value*
Estimate: 0% 5% 10% 15% 20% 30% 40%
1. overall sample 247,392 7.17 3.75 2.32 1.62 1.23 0.82 0.62
2. model 208,837 6.06 3.16 1,96 1.37 1.04 0.70 0.52
3. park sub-sample 215,851 6.26 3.27 2.02 1.41 1.07 0.72 0.54
*
from Table 3.16
3.7.6.1.6 Costs of the forests: WTA (willingness-to-accept) loss by local
residents
The breakdown of sample WTA bid responses is shown in Fig. 3.11 with 94% positive responses
(after discarding protest bids) - 27% accepting the initial $30 bid; 25% the follow-up $60 bid; and
refusing these, 30% gave an open-ended bid averaging $696 while 6% gave a zero-bid – the latter
indicating that compensation was not necessary, implying no costs are incurred nor opportunities
forgone by the existence of the forest. ‘Missing’ or protest votes constituted 13% of the sample.
Local WTA
zero bid
protest bid 6%
13%
open bid
$60 ($696)
25% 29%
$30
27%
Fig. 3.11: Distribution of sample WTA bids (US$)
53
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
WTA levels of compensation for the different parks are illustrated in Table 3.18, with BINP having
the highest average value of US$526. BINP’s result is significantly different from that of MGNP
($164), PNV ($80) and PNVi-S ($70) (F=20.22, p=0.000, Tukeys HSD homogenous subsets
alpha=0.05). In addition, WTA when expressed as a percentage of household annual income is again
significantly higher for BINP. As discussed earlier under Contingent Valuation (CV) methodology
(section 3.7.3), these figures represent loss experienced by the forest, principally crop damage, and
more significantly, the opportunity cost of foregoing forest assets (principally potential land for
cultivation and forest resources). The relatively high cost for BINP is reinforced by the benefit-cost
rank scores (see section 3.7.6.1), which show that BINP residents perceive exclusion of forest
resources and negative effect on development as important costs imposed on them by the gazettement
of the forest. This result is emphasised by the fact that the gazettement of and subsequent loss of
access to BINP in 1991 is likely to be fresh in residents’ minds, relative to the 1925 gazettement of
the Virunga parks.
Results also indicate that WTA does not vary significantly with distance from park (F=).455,
p=0.769) i.e. within the 5 kms. zone those further away do not value the lost opportunities that might
have been gained from the forest any less than those living close to the forest. This is contrary to
what might be expected, that is, the WTA of those living closer to the forest would be higher due (a)
greater loss from crop damage by wildlife and (b) a stronger sense of “lost use” of the forest since
they are the ones living immediately adjacent. However this was not found to be the case in this
study. Other factors at play may include the fact that those closer to the forest are often more recent
settlers (e.g. in Rwanda) and therefore feel less proprietorial over the forest, or that those living
adjacent to the forest are using its resources more often which would serve to decrease the cost of
being excluded.
Also depicted in Table 3.18 is the level of “no bids”. These reflect respondents who genuinely
experience no loss from the exclusion of forest resources. In addition, number of “protest votes” are
shown – those who experience a cost but bid zero believing for example that the scenario outlined
could not happen in real life (e.g. that a government would actually compensate them, etc.). This
difference was ascertained in the survey by asking a follow-up question if the respondent gave a zero-
bid as to the reason for this. PNV displayed the anomaly of greatest number of respondents who
perceive no forest cost, as well as the highest number of protest bids – the latter mainly due to
scepticism of a scenario involving compensation.
Table 3.18: Average willingness-to-accept (WTA) compensation bids for forest costs borne by
residents surrounding parks, also expressed as % of annual household income
BY PARK: average WTA as % Sample size No. zero No. protest
WTA bid of household bids bids
(US$) income
BINP 526.03 1.69 247 19 6
MGNP 164.11 0.28 104 - 13
PNV 70.00 0.25 115 24 48
PNVi-S 80.39 0.51 174 - 27
Overall 264.41 0.87 640 43 94
BY
DISTANCE:
1 km zone 241.26 0.79 131 8 17
2 km zone 207.79 0.68 138 8 22
3 km zone 399.77 1.31 140 7 22
4 km zone 271.81 0.89 136 11 19
5 km zone 313.69 1.03 95 9 14
54
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Table 3.19 below portrays WTA by location surrounding parks. Although significant differences
exist across locations as a whole (F=29.70, p=0.000), this is due to significantly higher WTA results
for only two locations - Kinaba and Mpungu, adjacent to BINP (Tukeys HSD homogenous subset
test). Such differences can only be interpreted in the context of the individual location.
Table 3.19: Willingness-to-accept (WTA) the cost of lost opportunities and direct costs by
residents surrounding parks, by parish/location
Park Parish.location WTA bid No. +ve No. zero No. protest
(US$) bids bids bids
BINP Kalangara 169.15 30
Kinaba 1213.17** 25 3 1
Kitojo 519.33 14 16
Masya 63.73 14
Mpungu 2962.96** 18
Mukono 96.89 30
Nteko 67.52 18 2
Remera 171.67 15 3
Rubimbwa 323.26 30
Rubuguri 134.77 34
subtotal 526.03 228 19 6
MGNP Gisozi 57.45 17 1
Gitenderi 69.31 22
Rukengi-Gisozi 210.17 32 5
Rukengi-Mabugo 233.32 33 7
Subtotal 164.11 104 13
PNV Bisate 59.69 20 11
Kabatwa 45.36 10 4 11
Kagano 47.33 34 11 10
Muhingo 142.75 12 15
Rutamba 95.72 15 9 1
subtotal 70.00 91 24 48
PNVi-S Gikoro 86.27 22 3
Kanombe 44.27 50
Kibumba 253.12 3 22
Mugora 68.92 24 1
Nkokwe-Kabaya 52.08 26
Nkokwe 119.69 49 1
subtotal 80.39 174 27
Overall total 264.41 597 43 94
** significantly different (Tukeys HSD homogenous subset test alpha=0.05)
3.7.6.1.7 Costs of the forests: extrapolation to the wider population
As in the case of WTP (previous section 3.7.6.1), the viability of extrapolating the WTA survey
results to the wider population were tested by use of a logistic regression model. Details of the model
designation and estimation are contained in Annex 7. The results show:
The logistic regression model is an appropriate model (-2 Log Likelihood test p=0.006)
The model’s estimates fit the data at an acceptable level (Hosmer & Lemeshow Goodness-of-Fit
test p=0.6596 > 0.005)
The model correctly predicted 80% of outcomes, implying that the sample responses contained
some element of random response. The model also had great difficulty in predicting ‘no’
responses (1% correct). However, ‘no’ responses to WTA for the sample were relatively scarce
(n=43 out of 734)
55
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Level of bid offer ($30, $60, $548) had a positive but insignificant effect on probability of ‘yes’
response, possibility due to large variation in response (high standard errors (SE)). On the other
hand, education level overall had a positive and significant effect on the probability (although
no one education level was significant), while household income level was found to be
independent of yes/no response (as was the case for WTP). This suggests that other factors are
more important in determination of yes/no: firstly, WTA – unlike WTP - is not dependent on
income; secondly, ‘free-riding’ is likely to influence results, where respondents elicit inflated
bids with the hope of greater financial gain. In addition, as discussed in section 3.7.4 (CVM
Design), responses may have been skewed by the fact that compensation in general is
uncommon and therefore creates a non-realistic scenario.
Mean WTA from the model was estimated to be US$250.32. This compares with US$264.41
computed for the sample mean (see section 3.7.6.1above). Thus, having taken into account the
distribution of the results from a statistical viewpoint, the model produces a similar but more
conservative estimate.
Using results for each of the mean WTA bids (overall sample, model estimate, and park sub-
sample) and aggregating over the number of households in the 5 km. zone surrounding each
park (see Annex 7), yields the annual value of forest costs to local residents living within that
area as shown in Table 3.20.
Table 3.20: Annual WTA (willingness to accept) compensation for costs to local residents
incurred by presence of the forests
Estimate 1: Estimate 2: Estimate 3: park sub-sample
overall sample model
All parks All parks PNV PNVi-S MGNP BINP
individual annual WTA 264.41 250.32 70.00 80.39 164.11 526.03
(US$)
no. households 53,548 53,548 18,462 14,666 4898 15,522
sub-total 1.29 1.18 0.80 8.17
total annual WTA 14.16 13.40 11.44
(US$ millions)
Estimate 3 is essentially a weighted mean taking into account differences in WTA by park, and falls
below both the values based on the overall sample mean ($14.16 million) and the more conservative
model estimate ($13.40 million).
However, as discussed in section 3.6.4.3, actual value depends discount rate. Table 3.21 expresses
total worth (net present value or NPV) over a 29 year time horizon for the three estimates of annual
value, according to discount rate assumed. Thus for the middle annual value (model estimate), for
example, NPV for forest costs is valued at $375.2 million assuming a 0% discount rate, while the
‘societal’ value may be $199.63 or $124.71 million, depending on whether a 5% or 10% ‘social’
discount rate is used. Alternatively, NPV is US$66.59 million when a ‘commercial’ discount rate of
20% is applied, while under a discount rate of 30% more applicable to a rural subsistence economy
forest costs to local residents fall to US$44.64 million (see section 3.6.4.3 for further discussion on
significance of NPV and discount rates).
Table 3.21: NPV (net present value) of local level WTA under differing discount rates (US$
millions)
Annual NPV under differing discount rates (US$ millions)
value*
Estimate: 0% 5% 10% 15% 20% 30% 40%
1. sample 14.16 396.48 210.96 131.78 92.51 70.37 47.17 35.40
2. model 13.40 375.20 199.63 124.71 87.55 66.59 44.64 33.50
3. park sub-sample 11.44 320.32 170.43 106.47 74.74 56.85 38.11 28.60
*
from Table 3.20
56
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
While such values imply significant costs to residents from the protected forests, it is worth repeating
the caution that WTA values are prone to exaggeration, particularly in the case of unique
environmental goods (Carson, 1991) (see section 3.7.4 for further discussion).
3.7.6.2 National level
3.7.6.2.1 Attitudes towards the forest: benefits and costs
Tables 3.22 and 3.23 summarise the views of both Goma and Kabale residents towards the forest. For
Goma 53% felt that the forest had a direct effect on their lives, while 47% felt it did not. Overall,
53% of Goma respondents considered the forest to be beneficial, while 39% considered it bad, or
were indifferent. By contrast, 70% of Kabale respondents felt that the forest affected their lives, with
16% responding negatively (with 14% of views unknown). In addition 69% perceived the forest as
beneficial overall, and only 9% saw it as negative or were indifferent (22% views unknown).
Table 3.22: Attitudes of town residents towards the forest
Does the forest affect your life?
Total
No Yes unknown
Count 18 20 38
Goma
TOWN % 47% 53% 100%
Count 13 56 11 80
Kabale
% 16% 70% 14% 100%
Count 31 76 11 118
Total
% 26% 65% 9% 100%
Table 3.23 Overall attitude towards the protected forest
overall attitude
Total
good bad indifferent unknown
Count 20 5 10 3 38
Goma
% 53% 13% 26% 8% 100%
TOWN
Count 55 2 5 18 80
Kabale
% 69% 3% 6% 22% 100%
Count 75 7 15 21 118
Total
% 64% 6% 12% 18% 100%
3.7.6.2.2 Value of the forest: benefits
Asked if they would be willing to pay for forest protection, 96% of Kabale respondents bid (with 72%
bidding positively), while 47% of Goma respondents bid, with only 5% bidding positively. Figs.3.12
and 3.13 show the respective breakdown of bids between protest bids (ejected from analysis), set bids
($5 and $10) and, on declining these, open bids – either positive (‘other’) or zero. There was a
significant difference between overall mean WTP bids for Kabale ($3.82) and Goma ($0.08 or
effectively $0). Asked for a zero bid reason, answers giving ‘no gain’ or ‘no concern’ for the forest
were considered bone-fide zero bids i.e. no value, while other answers were considered protest bids
i.e. not based on value. Protest bids in Goma were split 67% to 33% between respondents feeling
they were ‘not responsible’ for forest management and citing ‘poverty’ i.e. lacking ability to pay.
There were significant differences between the two towns in education level (completion of secondary
vs. primary years 1-4 respectively, p=0.098) and household income ($64.17 vs. $117.53, p=000) i.e.
higher education level in Kabale, but higher income level in Goma. However, responses rates for
Goma were relatively low for income level (34%) within a small sample (n=38), diluting the strength
of the result.
57
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
KABALE, UGANDA
protest
4% $0
24%
other
33%
$5
$10 18%
21%
Fig. 3.12: Distribution of forest protection WTP bids for Kabale, Uganda residents (note: mean
‘other’ bid = US$2.46)
GOMA, DRC
$0
protest 42%
53%
other
5%
Fig. 3.13: Distribution of forest protection WTP bids for Goma, DRC residents note: (mean
‘other’ bid = US$1.56)
3.7.6.2.3 Extrapolation to the wider population
Econometric models were not employed to check for the consistency of results and extrapolation due
to the limited sample size. Extrapolation of these results to a wider population is somewhat
hazardous, raising the question of who is represented by the survey: urban Ugandans; Ugandans living
in the region; or all Ugandans? Assuming that the Virunga massif and Bwindi forests are too
removed to hold value for the average Ugandan, a more reasonable assumption might be that such
values apply to the population of the districts adjoining the two parks MGNP and BINP (population
1.08 million). Taking average household size for communities surrounding MGNP and BINP
(average 6.0, see Table 2.1), and applying the sample WTP mean of $3.82 yields an overall annual
forest value estimate of $687,600. In the absence of data, and applying the same assumptions
(adjoining districts’ population 500,000 with average household size of 6.5), a corresponding estimate
of $293,846 is arrived at for Rwanda. However, it should be noted that such estimate are tentative,
for reasons cited earlier related to small sample size.
The results of calculating NPV (net present value) over a 29-year period based on the above estimates,
for differing discount rates are given in Table 3.24.
58
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Table 3.24: NPV (net present value) of forest protection WTP (US$ millions) at regional level
Annual NPV under differing discount rates (US$ millions)
value ($)
0% 5% 10% 15% 20% 30% 40%
Uganda (MGNP & BINP) 687,600 19.25 10.24 6.40 4.49 3.42 2.29 1.72
Rwanda (PNV) 293,846 8.23 4.38 2.73 1.92 1.46 0.98 0.73
DRC (PNVi-S) 0 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Total
27.48 14.62 9.13 6.41 4.88 3.27 2.45
3.7.6.3 International level
Results appear below. Methodology is adapted from that of Kotchen and Reiling (2000). A limited
survey of 120 respondents was planned, however, due to survey implementation problems
encountered by the enumerator the final sample size executed was just 26, clearly too limited to infer
results to a wider population with any confidence – particularly one of global significance. Therefore
the main contribution of this section should be regarded as (1) outlining a methodology for exploring
non-use value amongst “global citizens” and (2) indicating the potential magnitude of international
non-use value of mountain gorilla habitat.
3.7.6.3.1 Effect of environmental attitude on response and willingness to pay
for mountain gorilla habitat protection
There are 3 main advantages to considering environmental attitude viz a viz willingness to pay:
Controls for bias amongst respondents for or against environment
Explores how attitude explains responses (willingness or unwillingness to pay; protests)
Explores how attitude affects willingness-to-pay amount
Table 3.25 shows scores for the questions determining ‘environmental attitude’. Each question is
scored 1-5 depending on response from ‘strongly agree’ to ‘strongly disagree’ with higher scores for
pro-environment responses. Average score per respondent was 18.87 out of a maximum pro-
environment possible score of 25.
Table 3.25: Mean percentage distributions for Environmental Attitude items
Statement Average STA SWA U SWD STD* (%)
score**
a. We are approaching the limit of the number of 4.04 42 19 38 - -
people they earth can support
b. The so-called ‘ecological crisis’ facing humankind 3.25 12 23 23 12 27
has been greatly exaggerated
c. Humans were meant to rule over the rest of nature 4.27 - 4 15 31 50
d. When humans interfere with nature it often 4.12 58 19 8 8 8
produces disastrous consequences
e. The balance of nature is strong enough to cope with 3.19 27 8 15 19 31
the impacts of modern industrial nations
Average respondent score 18.87
* STA = strongly agree; SWA = somewhat agree; U = unsure; SWD = somewhat disagree; STD =
strongly disagree
** 5 = maximum pro-environment score; 1 = minimum pro-environment score
Assigning each individual score to one of 3 approximately equal-size categories – weaker, moderate,
and strong pro-environment attitude – allows consideration of the effect of attitude on bid type, and
willingness to pay (Table 3.26). It would normally be expected that the ‘weaker attitude’ would
59
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
contain more ‘zero-bids’ and lower average willingness-to-pay. The opposite is true in this case, most
likely due the limited sample size. The overall average WTP per respondent was US$12.43 per year.
Of interest, 80% of zero-bids were due to the fact that respondents did not consider themselves
responsible for protection of mountain gorilla habitat, while only 20% considered the cause a low
priority. By contrast the reasons for protest-bids varied according to attitude: those with ‘weaker’
environmental attitudes cited income restriction, while those with ‘stronger’ attitude were only willing
to pay for multi- as opposed to single-species conservation (despite awareness of the conservation
status of the mountain gorilla); or were already donating.
Table 3.26: Pro-environmental attitude and type of bid response for WTP (willingness to pay)
for mountain gorilla protection*
Weaker Moderate Stronger Overall
attitude* (n=5) attitude (n=11) attitude (n=10)
Yes bids 2 5 4 42%
Zero bids 1 6 3 39%
Protest bids 2 0 3 19%
Average bid 21.67 13.64 4.6 12.43
(US$)
* Weaker = attitude score 20.
3.7.6.3.2 Motivations giving rise to non-use value
Also of interest is the exploration of which of the 5 recognised non-use values provide the stronger
motivations for protecting the mountain gorilla. Table 3.27 indicates that ‘existence’ value and
‘ethical’ value are the most important, with higher average scores (2.69 and 2.65 respectively out of a
possible maximum of 3) and % rating as ‘important’ (80 and 76% respectively). ‘Option’ value is
least important, which together with the relatively moderate scores for ‘altruistic’ and ‘bequest’ value,
imply that possible future visiting value (questions a, b and c) is less important than mountain
gorillas’ right to exist (questions d and e) when considering their value.
Table 3.27: The importance of different motivations for protecting mountain gorillas
Statement Average Importa Slightly Not
score* nt (%) importa importa
nt (%) nt (%)
a. I may want to see a mountain gorilla in the 2.08 38 31 31
future (option value)
b. I enjoy knowing other people can enjoy them 2.15 38 38 24
(altruistic value)
c. I enjoy knowing future generations will enjoy 2.31 54 23 23
them (bequest value)
d. I enjoy knowing mountain gorillas exist even 2.69 80 8 12
if no-one ever sees one (existence/intrinsic value)
e. All endangered species have a right to exist 2.65 76 12 12
(ethical/moral value)
Means are calculated from coding where ‘important’ = 1, ‘slightly important = 2, ‘not important’ = 3.
Maximum average score = 3.
3.7.6.3.3 Extrapolation of results to the wider population
It should be re-emphasised that extrapolation of results is extremely hazardous due to the very limited
sample size involved. However, based on the sample average WTP of US$12.43 per household
applied to a conservation-conscious, ‘developed’ world population of 150 million households
(US=250 + EU=250 + Australia/Japan=100 million, and assuming average household size of 4), the
annual non-use value of mountain gorillas computes at some US$1865 million per year. The net
present value of this amount over a 29 year period is depicted in Table 3.28 for different discount
60
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
rates. The substantial magnitude of the figures coupled with the limited sample size suggests a
significantly reduced estimate is warranted. Thus, in the interests of conservatism, for the purpose of
further analysis it will be assumed that the real benefits are one-tenth of estimates derived from the
sample. The adjusted results appear in the bottom row of Table 3.28.
Table 3.28: NPV (net present value) of the non-use of mountain gorilla habitat to international
citizens
Annual NPV under differing discount rates (US$ millions) over
value 29 years
(US$
millions)
0% 10% 20% 30% 40%
International non-use
value 1864.5 52,206 17,352 9265 6210 4660
Adjusted estimate 186.5 5221 1735 926 621 466
3.7.6.4 The case of Gishwati Forest
3.7.6.4.1 Attitudes towards and willingness-to-pay for forest ecological
services at the local level
Respondents in two locations farming on flat land at or near the edge of the previous Gishwati forest,
before it was felled in 1996 to make land available for returning refugees, were surveyed. Given the
exposed slopes above them, and lack of tree replacement on them, farmers were asked about the
changes – before 1996 versus the present - they perceived in two types of ecological services:
waterflow regime (decrease in flow year-round; decrease in dry-season flow; and increase in wet-
season flow); and soil fertility (increased fertility, possible through siltation; or decreased fertility,
through erosion). Respondents were then asked to respond to suggested WTP bids for investment into
re-afforestation of the previous forest area.
Figs. 3.14 and 3.15 summarise combined results for the 2 locations (n=58) and show that respondents
perceived a marked change in water regime, principally increased wet season runoff (66%) and gully
formation (‘other, 19%), and a decline in soil fertility.
change in water regime
10%
3% no change
19%
2% annual reduction
dry season reduction
wet season increase
66% other
Fig. 3.14: Perceived change in water regime by Gishwati farmers
61
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
change in soil fertility
0% 17%
no change
3% improved fertility
decreased fertility
80%
Fig. 3.15: Perceived change in soil fertility among Gishwati farmers
Results for corresponding WTP (willingness to pay) are illustrated in Tables 3.29 and 3.30
respectively, for each location. While no significant difference exists the locations in income level
(F=0.701, p=0.406), distance from the former forest of fields farmed by respondents varied markedly
(F=20.951, p=0.000). Overall average WTP for Gatagara-Busoro (US$9.96) was also significantly
different from that for Mukamira ($4.88) (F=9.807, p=0.003) - although both values are higher than
the corresponding value for communities surrounding BINP, MGNP, PNV and PNVi-S (see section
3.7.6.1.4). For water regime a marked difference in WTP between those perceiving negative changes
($10.33) versus those not was evident ($5.15) in the case of Gatagara, while for Mukamira relatively
little difference in WTP bids existed between those perceiving change versus not.
Table 3.29: WTP for reafforestation by Gishwati farmers according to perceived change in
waterflow regime
WTP according to perceived change
Annual Average
No Dry season Wet season Other (new WTP
reductio
change reduction increase gullies)
n
Gatagara- 4.12 20.62 - 10.77 7.50 9.96
Busoro (n=1) (n=1) - (n=18) (n=8) (n=28)
Area
3.09 20.62 1.03 5.15 2.06 4.88
Mukamir
(n=5) (n=1) (n=1) (n=20) (n=3) (n=30)
a
3.26 20.62 1.03 7.81 6.05 7.33
Overall average
(n=6) (n=2) (n=1) (n=38) (n=11) (n=58)
Table 3.30: WTP for reafforestation by Gishwati farmers according to perceived change in
soil fertility
WTP according to perceived change
No change in Improvement in Deterioration in Total
soil fertility soil fertility soil quality
Gatagara- 5.15 10.31
Busoro (n=2) (n=1) 10.33 9.96
Area (n=25) (n=26)
4.30 20.62
Mukamira
(n=6) (n=1)
62
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
4.35 4.88
(n=23) (n=24)
4.51 15.46
Total
(n=8) (n=2)
7.46 7.33
(n=48) (n=50)
3.7.6.4.2 Extrapolation of results to wider population
Again, an econometric model was not utilised to extrapolate results due to limited sample size. One
would expect that willingness-to-pay for forest protection/reafforestation amongst Gishwati residents
to be higher than for the areas surrounding the protected gorilla forests, other things being equal,
given that the latter have not experienced that scale of the negative effects. Accordingly, average
WTP for Gishwati ($7.33) was found to be significantly different from that of the zones surrounding
BINP, MGNP, PNV, PNVi-S ($4.62) (t=3.070, p=0.002). The difference ($2.71) may be interpreted
as the value of ecological restoration to a household in a degraded (deforested) scenario, and as such,
provide a more precise indication of the value of ecological services. Following the methodology of
section 3.7.6.1.5 and applying this difference to the households n=53,548) in the 5 km. zone
surrounding the Virunga massif and Bwindi forests yields an annual value of $145,115 for ecological
services in that zone, which translates into NPV 9net present value) as portrayed in Table 3.31.
Table 3.31: NPV (net present value) of forest protection WTP
Annual NPV under differing discount rates (US$
value ($) millions) over 29 years
0% 10% 20% 30% 40%
Ecological services 145,115 4.06 1.35 0.72 0.48 0.36
3.7.6.5 International ecological benefits
Of the many perceived regional and global ecological benefits derived from African tropical forests,
focus typically remains on two: the future potential for new drugs and medicines; and carbon
sequestration – the value of forests in storing carbon which under deforestation would be released into
the atmosphere as CO2 thereby contributing to global warming.
3.7.6.5.1 Pharmaceutical benefits
In a review of the potential value of pharmaceutical discoveries in African forests, Norton-Griffiths
concludes few data exist for Eastern Africa, and that other studies suggest that “the benefits are likely
to be modest and difficult to capture”, citing for example analysis by Pearce et al (1993) that suggests
a range of values between $0.01 and $ 21 per hectare. Furthermore, Africa is considered poor in
biodiversity compared to South America and Asia, while individual countries are not as unique in
flora, suggesting a likely competitive market with relatively low prices (Norton-Griffiths, 1995).
However, as acknowledged earlier, the Albertine Rift and Virunga-Bwindi forests do contain high
levels of species richness and endemism, and as such may compare favourably to other global sites in
terms of biodiversity.
3.7.6.5.2 Carbon sequestration
Analysis by Nordhaus (1991), Scneider (1991) and the World Bank (1991), gives a range of possible
values for carbon sequestered in tropical forests between $1500-3500 per hectare per year, while
Brown (1992) and Pearce et al (1993) give values of $320-1600 per hectare as the net global costs of
damage incurred by converting tropical forests to agricultural use (Norton-Griffiths, 1995).
3.7.6.5.3 Extrapolation to the Virunga and Bwindi forests
Taking the average of the more conservative above estimate ($960 per hectare) yields an annual value
for the Virunga forests of US$396,480 ($129,600, $240,000 and $26,880 respectively for PNV,
PNVi-S, MGNP) and $316,800 for Bwindi Impenetrable forest, for an annual total of $713,280.
63
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Computing NPV (net present value) yields the economic worth of this flow of benefits at different
discount rates, depicted in Table 3.32.
Table 3.32: NPV (net present value) of carbon sequestration by Virunga and Bwindi forests
(US$ millions)
Annual NPV over 29 years under differing discount
value ($) rates (US$ millions)
0% 10% 20% 30% 40%
713,280 20.0 6.6 3.5 2.4 1.8
3.8 VALUATION SUMMARY: THE SHORT-TERM AND LONG-TERM WORTH OF THE PROTECTED
FORESTS
The distribution of short-term (annual) values amongst all benefit components covered in this report
are illustrated in Figs. 3.16 and 3.17, while the corresponding long-term (NPV) results are
summarised in Table 3.33. Taken as a whole, they represent the total economic benefit and cost
components of the combined forests of BINP, MGNP, PNV and PNVi-S. In addition, the distribution
of these components amongst the international, national (non-local) and local levels is summarised in
Table 4.34, and illustrated in Figs. 3.18 and 3.19.
For the purpose of this analysis, Table 4.33 highlights five main conclusions:
1. The non-use value of mountain gorilla habitat to the international community accounts for 90%
of total annual benefits derived from the Virunga and Bwindi protected forests.
Beyond this dominant value:
2. The majority (91%) of remaining benefits are flowing from international gorilla tourism.
3. There is a total positive net benefit flow from the forests of US$9.9 million each year.
4. Benefits of relatively equal magnitude are accruing to international and national levels, whilst
those at the local level are suffering losses of similar magnitude mainly due to lost opportunities
from forest exploitation.
5. Time preference i.e. the importance of present vs. future gratification (represented by discount
rate) substantially impacts the net worth of forests (represented by net present value (NPV)).
Int'l: Tourism
1%
3%
4%1% Nat'l: Tourism
6%
Local: Tourism
37%
Local: use & non-use
National: non-use
48%
Local: ecological services
Int'l: ecological services
Fig. 3.16: Annual distribution of gross use and non-use benefits from Virunga and Bwindi
forests (excluding international non-use value).
64
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Int'l: Tourism
Nat'l: Tourism
4% 5% Local: Tourism
1%
0%
0%
Local: use & non-use
National: non-use
Local: ecological services
90% Int'l: ecological services
International: non-use
Fig. 3.17: Annual distribution of gross benefits from Virunga and Bwindi forests including
international non-use value.
Table 4.33: Summary of short-term (annual) and long-term value (NPV) accruing to direct-use,
indirect-use, and non-use forest benefit & cost components (US$ millions)
Sectio Benefit & Cost components Ann- NPV under differing discount rates over 29-year
n ual period
value
0% 5% 10% 15% 20% 30% 40%
GROSS BENEFITS / LEVEL
3.5 Local direct use value
- - - - - - - -
3.6.4.3 Tourism (indirect use) valuea:
International 8.7 252.3 131.7 81.5 57.0 43.3 29.0 21.7
National 11.2 324.8 169.6 114.7 73.4 55.7 37.3 28.0
Local 1.3 37.7 19.7 12.2 8.5 6.5 4.3 3.3
3.7.6.1 Local direct, indirect & non-use
forest valueb 0.2 6.3 3.3 2.0 1.4 1.1 0.7 0.5
3.7.6.2 Regional indirect and non-use
forest valuec:
Uganda 0.7 19.2 10.2 6.4 4.5 3.4 2.3 1.7
Rwanda 0.3 8.2 4.4 2.7 1.9 1.5 1.0 0.7
3.7.6.3 International non-use value 186.5 5221 2778 1735 1218 927 621 466
3.7.6.4 Local ecological services 0.2 4.1 2.2 1.3 0.9 0.7 0.5 0.4
3.7.6.5 International ecological
services 0.7 20.0 10.6 6.6 4.7 3.5 2.4 1.8
Total benefits 209.8 5894 3130 1962 1370 1043 699 524
Total benefits less int’l non-
use 23.3 672.6 351.7 227.4 152.3 115.7 77.5 58.1
GROSS COSTS
3.7.6.1, Local level forest costsd
2.7.2 13.4 375.2 199.6 124.7 87.5 66.6 44.6 33.5
NET BENEFITS
Net benefits (benefits-costs) 196.4 5518 2930 1838 1283 976 654 491
Net benefits excluding int’l
non-use value 9.9 297.4 152.1 102.7 64.8 49.1 32.9 24.6
a
see Table 3.9
b
local WTP for forest protection, middle estimate (park sub-sample) – see Table 4.17. Some overlap
with ‘local tourism value’
c
national WTP for forest protection – see Table 4.24
d
local WTA compensation for forest costs, middle estimate (model) – see Table 4.21
65
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Table 4.34: Short-term (annual) and long-term (NPV) distribution of forest net benefits between
international, national, and local levels (US$ millions)
Net benefits accruing to: Annual NPV(net present value) under differing discount rates
value (US$ millions) over 29 years
(US$
millions)
0% 5% 10% 15% 20% 30% 40%
International level
non-use 186.5 5221 2778 1735 1218 927 621 466
use 9.4 272.3 142.3 88.1 61.7 46.8 31.4 23.5
total 195.9 5493.3 2920.3 1823.1 1279.7 973.8 652.4 489.5
National level (non-local) 12.2 352.2 184.2 123.9 79.8 60.6 40.6 30.4
Local level (within 5
kms) -11.7 -327.1 -174.4 -109.2 -76.7 -58.3 -39.1 -29.4
Total 196.4 5518 2930 1838 1283 976 654 491
Total excluding int’l non-
use value 9.9 297.4 152.1 102.7 64.8 49.1 32.9 24.6
9.4
-11.7
International gain
National gain
Local loss
12.2
Fig. 3.18: Annual bet benefits and losses from the Virungas and Bwindi forests according to
local, national, and international level excluding international non-use value (US$ millions)
and when forest opportunity costs are assigned to the local level.
12.2
-11.7
National gain
Local loss
International gain
195.9
Fig. 3.19: Annual net benefits and losses from the Virungas and Bwindi forests according to
local, national, and international level including international non-use value (US$ millions)
and when forest opportunity costs are assigned to the local level.
66
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
The above analysis, however, assumes that forest opportunity costs accrue to local communities
surrounding the forests. However, it can be argued that opportunity costs in fact should be assigned to
the State i.e. national level (see discussion Section 2.7.3). If this is the case, Figs. 3.13 and 3.14 will
be modified with the following results, reflecting a relatively small overall aggregate benefit (+
US$0.5 million per year) accruing at in-country.
1.7
-1.2 International gain
National loss
Local gain
9.4
Fig. 3.20: Annual bet benefits and losses from the Virungas and Bwindi forests according to
local, national, and international level excluding international non-use value (US$ millions)
and when forest opportunity costs are now assigned to the national level as opposed to local level.
1.7
-1.2
International gain
National loss
Local gain
205.3
Fig. 3.21: Annual net benefits and losses from the Virungas and Bwindi forests according to
local, national, and international level including international non-use value (US$ millions)
and when forest opportunity costs are now assigned to the national level as opposed to local level.
67
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
4 POLICY CONCLUSIONS, BENEFIT-COST DYNAMICS & CONSERVATION
OPTIONS
4.1 POLICY CONCLUSIONS
Four main conclusions emerge from the study results:
From an economic standpoint, the Virunga and Bwindi forests should be protected since they
are generating significant benefits over and above the costs borne by local communities.
There should be a commitment to increasing mountain gorilla tourism, since it is the driving
force of tangible benefits realized from the forests, and is operating at under-capacity.
The standard of local communities’ livelihood must be increased, since communities are
experiencing losses from the existence of the forests or, at best, no gain. Two avenues are
available: (a) increased benefits derived from the tourism/forests; and/or (b) increased
development investment into the local communities.
In addition to current benefits, international non-use value of the forests represents a huge
potential source of funding.
4.2 BENEFIT-COST DYNAMICS AND TRENDS
This report identifies seven important elements, the dynamics of which hold major ramifications for
the longer-term future of the Virunga and Bwindi forests. An understanding of each is vital to
achieving sustainable conservation:
1. Time preference: the impact of limited planning horizon on forest benefit levels.
2. Opportunity cost: the cost to local communities of the protected forests.
3. Agricultural productivity: intensification or expansion?
4. Soil erosion and decreasing fertility
5. International gorilla tourism
6. International non-use value
7. Returns to investment into rural communities
Each of these is now discussed in turn. Section 4.3 then incorporates consideration of these factors
into a series of numerical scenarios to project and illustrate the impact of each into the future.
4.2.1 Time preference: the impact of limited planning horizon on forest benefit levels
The significance of considering time preference in the estimation of benefits and costs over time, as
reflected by different discount rates (see e.g. Table 3.32 and 3.33 above), lies in its explanation of
how the same resource can be valued differently by one individual or institution versus another. From
the perspective of policy-making this has two major implications:
Simply put, a resource’s value depends on which rate is applied. In this way, when valuing the
magnitude of expected or realized benefits a conscious decision can be made as to which
discount rate is appropriate. This typically involves balancing ideal goals for society against
practical realities. Decisions based on ideals may be economically unsustainable given the
realities of the human environment – for example, too costly for local communities; conversely,
those based on present-day needs may serve to over-deplete the resource to the detriment of the
future society.
It provides a framework, once a decision has been made, for assessing the degree to which
different stakeholders will benefit from increased revenues, according to their time preference.
For example, assuming local communities operate under 30% time preference, the current
benefits of $400,000 per year from Table 3.9 are worth $4.3 million. If policy changes allowed
revenues to be doubled, value would increase to $8.6 million. However, this increase in
revenue may also allow present needs to be better met, freeing up extra funds for investment
68
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
into the future. If sufficiently significant, the effect may result in lowered time preference. If,
for argument’s sake, time preference was lowered from 30% to 20% as a result, the new value
of the resource at the local level can be calculated as $13.0 million – a further increase in value
of $4.4 million over and above the initial revenue increase. This concept of “double benefit”
therefore becomes more significant the higher the time preference under which a particular
stakeholder is operating. It also emphasises the economic significance of development and
poverty alleviation.
4.2.2 Opportunity cost: the cost to local communities of the protected forests.
The opportunity cost of the forests is the value of the forests forgone by not converting them to their
next best use – assumed to be smallholder agriculture. This cost is borne by the communities farming
around the forests since it is assumed they would be the beneficiaries of any conversion.
In theory, any future decrease in the productivity of existing farmland e.g. due to soil deterioration,
will result in decreased opportunity cost of forests since farmland will be deemed to be less valuable.
This may appear to be a positive result for forests, since their relative value is increased. However in
practice such gains are dampened by a secondary effect: lower productivity will increase the pressure
on farmers to realize immediate benefits at the expense of future gains i.e. increase farmers’ time
preference. This in turn has the effect of increasing the value of converting the forest; that is,
decreasing the forests’ value over time.
Conversely, any future increase in per hectare productivity should result in a relative decrease in the
value of forests, since farmland gains in relative value, thereby creating greater pressure for forest
conversion. And while the secondary effect of decreased desire for immediate benefits may slacken
i.e. decreased time preference, in practice it can be expected that pressure to convert forest into farms
will remain, in the interests of progress.
4.2.3 Agricultural productivity: intensification or expansion?
As discussed in Section 2, increased agricultural benefits can be achieved by increasing land
productivity (intensification); or increasing land area (expansion). As available new land disappears,
intensification occurs as a matter of necessity. However, intensification usually requires the
substitution of capital for labour (section 2.8.2). The results from Section 2 showed that for the
Virunga region, all available land is cultivated every season, without possibility of rest. Coupled with
the low level of fertilizer use and other improvements, the prognosis points to the falling land
productivity in the face of lack of capital investment. For the Bwindi region, fallow is still widely
used as a method of replenishing soil fertility, however, it can be expected to follow the trend in
Virunga over time – if not already, since soils are known to be poorer than the rich volcanic soils of
Virunga. The challenge for conservation of the Virunga and Bwindi forests is to increase local
standards of living through improved productivity of existing land holdings, without increasing
pressure for forest conversion. Under such circumstances the following would be necessary
conditions in presenting a robust economic argument in favour of continued forest protection:
gains in the productivity of existing farmland were fuelled/funded through forest
interests/benefits
farmers were aware of this fact
the magnitude of these gains approached those of converting forest to farms
4.2.4 Soil erosion and decreasing fertility
The results of Section 2 showed decreasing soil fertility and fears of soil erosion as major concerns for
the future and, indeed, were being experienced in some areas already. As discussed earlier, decreases
in land productivity result in lowered opportunity costs of the forests dampened by increases in time
preference (see previous section 4.2.1). For the purpose of analysing dynamics at the local level, it
will be assumed that these effects cancel each other out i.e. that declining soil fertility on existing
farmland neither increases nor decreases pressure to convert forest to farm land. However, it is
69
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
acknowledged that from a national perspective, such decline serves to lessen the benefits, and
therefore the wisdom, of forest conversion.
4.2.5 International gorilla tourism
The results of Section 3 showed that international gorilla tourism accounts for the majority (at least
91%) of tangible benefits being derived from the forest (Table 4.33). They also show that gorilla
tourism is operating at under-capacity (section 3.6.4.3). Gorilla tourism has averaged 25% of
maximum capacity dictated by the availability of habituated gorilla groups and visitor group size over
the past decade, with a current level of 39%. The analysis of this report suggests that current annual
gorilla viewing benefits of $21.2 million are worth $54.4 million at full capacity, increasing the
resource’s net worth from $615 to $1576 million (Tables 3.9 and 3.11). Thus, untapped potential
exists; however, the main hindrance to realization of this potential is political instability.
4.2.6 International non-use value
The results of Section 3 also suggested that the non-use value placed on mountain gorilla habitat by
international citizens – motivated by biodiversity, future option, and ethical reasons - represents easily
the single largest source of benefits (90%) derived from the Virunga and Bwindi forests (Table 4.33).
This has long been the benefit stream into which many conservation NGOs have traditionally tapped
in order to raise funds for conservation. However, the magnitude of this value in relation to the other
benefit and cost components in the case of the mountain gorilla suggest significant funding potential.
4.2.7 Returns to investment into rural communities
This report has illustrated the importance of improving the livelihoods of local communities
surrounding the Virunga and Bwindi forests, and identified the main need as capital availability
and/or investment. The report has also identified significant sources of potential capital flowing from
forest benefits at the international level. In addition, the mechanism for meeting the twin objectives
of forest conservation and community development has been outlined in section 4.2.3 above. It
appears that these conditions can be met, namely: (1) the forests are able to generate sufficient funds
(2) such benefits could fuel rural benefits to a degree that more than offset the costs of the forests’
existence (3) mechanisms are required whereby local communities are aware of these links.
In this context, one of the most significant results of analysis into farm production systems (Section 2)
was the potential return on investment, which implied that each $1 invested into agriculture could on
average be expected to generate $6.40 in return (Table 2.23). Put differently, an annual investment
into local communities within the 5 km. zone surrounding the forests equal to the costs of the forests’
existence ($13.4 million – see Table 4.33) represents a per hectare investment of $209 (based on
53,548 households with an average cultivated holding of 1.2 hectares – see Table 2.4), for an expected
annual return of $1338 per hectare. This compares with an expected annual return of $436 per hectare
from converting forest to crops given current productivity (see Table 2.17)– a 307% superiority
factor.
The additional benefits can be summarised as follows:
Secondary income effects in the local economy. Applying the national income multiplier of
0.45 (section 3.6.4.2) implies resulting secondary income benefits within the local economy of
$2.88.
A decrease in time preference i.e. increasing emphasis on future as opposed to present needs.
This in turn, has the effect of significantly increasing forest value since forest benefits accrue
through time (see section 4.2.1 above for illustration of this).
Creation of employment in a labour-abundant environment.
Possible income diversification, both within and outside agriculture.
Increased tax revenue. Using the national government revenue multiplier of 0.26 (section
3.6.4.2) implies further potential secondary benefits in the longer run of $1.66.
70
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
In summary, this suggests that a $1 investment into local agriculture can be expected to return a total
of $10.94 plus the non-monetary benefits of employment creation, income diversification, and more
sustainable planning horizon.
4.3 SCENARIOS AND OUTCOMES
The following scenarios illustrate the impacts of the dynamics discussed in section 4.2 on six benefit-
cost components:
• gorilla tourism monetary benefits accruing to the international, national, and local levels
(columns 1-3);
• the opportunity cost of the forests to local residents (column 4);
• the value of tourism revenue at the local level (column 5);
• and the economic multiplier effect of tourism revenue at the local level, based on the potential
returns to agriculture (column 6).
All amounts are net present value, or worth, over a 29-year period. It is important to note that
columns 1-3 portray value from the perspective of a policy maker and, accordingly, are based on a
time preference rate of 10%, while columns 4-6 portray value from the standpoint of subsistence-level
rural communities, and are based on a time preference of 40%. Thus for example, columns 3 and 5
depict the same thing: the value of annual gorilla tourism benefits to the local level over time.
However, column 3 portrays the value of those benefits as perceived by policy makers; whilst column
5 portrays the value of those same benefits from the perspective of local communities, who have more
immediate priorities and thus place less value on future benefits.
It should be noted that these scenarios below are based on the original 2002 study, and have not been
modified in the light of the subsequent 2005 data, since the 2005 results do not change the broad
conclusions. However, two differences between the 2002 analysis and 2005 data can be kept in mind:
first, the value of tourism to local communities is assumed to be 5% as per the original 2002, although
the subsequent 2005 research indicated that the real amount was less than 5%. Second, gorilla
revenues – and therefore visitor expenditure – nearly doubled from 2002 to 2005. As indicated,
however, this fact does not change the broad conclusions made below.
Notes accompany each scenario. The 5 scenarios are ordered as a progressive sequence, the outcomes
of which are largely self-explanatory. In particular, they illustrate the effects of different income
scenarios, primarily as they relate to the local level. The outcomes suggest that given a significant
increase in the share of benefits together with a sustained growth in tourism, increased revenues from
gorilla tourism are potentially useful contributors to the improvement of rural livelihoods. However,
additional investment into the local communities is likely to prove the real engine of development,
and by extension, sustainable forest conservation.
71
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Fig. 4.1: Scenarios depicting future impact of economic dynamics on 6 key benefit-cost
components
Scenario 1: Current status
Benefits and costs ($million)
150
115
100 81
50 34
12 1 4
0
intl natl local opp local l multr
cost
Scenario 1 notes:
Worth over 29 year period
Current levels of benefits/costs with no change over time
Scenario 2: No change in benefits
Benefits and costs
150
115
($million)
100 81
50 39
12 1 4
0
intl natl local opp local l multr
cost
Scenario 2 notes:
Opportunity cost increasing at 4% annually due to population growth
Local time preference (discount rate) rising at 10% every 10 years due to increase in opportunity
costs
Result: increase in opportunity cost to communities of $5 million
72
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Scenario 3: Increase in gorilla tourism
132
Benefits and costs
150
103
($million)
100
50 39
15 5
1
0
intl natl local opp local l multr
cost
Scenario 3 notes:
Tourism increasing at 5% per year until 75% capacity, level thereafter
Benefit gains at all levels
Minimal gain from local perspective ($1 million gain in local multiplier effect)
Opportunity cost and time preference still rising due to minimal gains
Scenario 4: Gorilla tourism increase +
increase in local share
Benefits and costs
150 123
($million)
93
100
53
50 34 28
6
0
intl natl local opp local l multr
cost
Scenario 4 notes:
Same increase in tourism
Local share increased from 5 to 25% of tourist expenditure at park level (increase of $1.6
million per year at equal expense of int’l and national shares)
No increase in opportunity cost and time preference improving due to significant benefits
73
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
At local level conservation benefits and multiplier effects ($6 + $28 million) now equal the
opportunity cost of the forests
Scenario 5: tourism increase + local share
increase + investment into rural economy
Benefits and costs
150 123
106
($million)
93
100
56
50 34
13
0
intl natl local opp local l multr
cost
Scenario 5 notes:
Same as scenario 4, with additional annual investment of $2 million into local communities
(matching local gorilla tourism receipts under Scenario 4)
Local level time preference improving due to significant benefits, further increasing value of
benefits received
Substantial benefits flowing to local level when viewed from either policy or local perspective.
74
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
REFERENCES
Bizimungu, Francois (2002) Warden, ORTPN (Rwanda National Parks and Tourism Office),
Ruhengeri, Rwanda, August 2002, personal communication
Brown, G. Jr. and W. Henry (1993) ‘The viewing value of elephants’ in ‘Economics and Ecology:
New frontiers and sustainable development’ Ed. E. Barbier, Chapman and Hall, London
Carson, R. (1998) ‘Valuation of tropical rainforests: philosophical and practical issues in the use of
contingent valuation’, Ecological Economics 24 (1998) pp15-29
Cunningham, A.B. (1996) ‘People, parks and plant use: recommendations for multiple use zones and
development alternatives around Bwindi Impenetrable N.P., Uganda’, People and Plants working
paper 4, UNESCO, Paris
IGCP (2000) ‘Analysis of the economic significance of gorilla tourism in Uganda’, Report by
Environmental Monitoring Associates (EMA) Ltd., International Gorilla Conservation Programme
(IGCP), Nairobi, Kenya
IGCP/WCS/CARE (2004) ‘The Socio-Economic Status of People living near protected areas in the
Central Albertine Rift’, Albertine Rift Technical Reports, Vol. 4, IGCP Nairobi. International Gorilla
Conservation Programme (IGCP) Nairobi, and Wildlife Conservation Society New York, CARE
Regional Office Nairobi
IIED (1997) ‘Part 3: Economic Valuation Techniques’, in ‘Eco-Valuation Manual’, IIED, London
Kotchen, M.J. and S.D. Reiling (2000) ‘Environmental attitudes, motivations, and contingent
valuation of nonuse values: a case study involving endangered species’, Ecological Economics 32
(2000) pp93-107
Lanjouw, A. (2003) Programme Director, IGCP, Nairobi Kenya, March 2003, personal
communication
Namara, A., McNeilage, A., Franks, P., Blomley, T. Infield, M., Malpas, R., & Donaldson, A.
(forthcoming) ‘Bwindi Impenetrable and Mgahinga Gorilla National Parks in Uganda: has 15 years of
ICD programming achieved conservation impact?’ Unpublished report, Institute of Tropical Forest
Conservation, Kabale, Uganda.
Sikoyo, G.M. (1995) ‘Economic Valuation of the Multiple Use Forests: the case of Bwindi
Impenetrable National Park (BINP), Uganda’, MSc. Dissertation, University of Edinburgh, U.K.
Wild, R. and J. Mutebi (1996) ‘Conservation through community use of plant resources: establishing
collaborative management at Bwindi Impenetrable and Mgahinga Gorilla National Parks, Uganda’,
People and Plants working paper 5, UNESCO, Paris
75
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
LIST OF ANNEXES
Annex 1: Notes on sampling design, sampling strategy, and field methodology
Annex 2: Notes on prices, exchange rates and computation of farm income
Annex 3: Farm Production Systems questionnaire
Annex 4: Crop mix and diversity by country, parish, and distance zone
Annex 5: The Travel Cost and Contingent Valuation questionnaires
Annex 6: Travel Cost demand estimation results for international gorilla-park visitors sample
Annex 7: WTP/WTA model specification and estimation, and basis for extrapolation to the wider
population
76
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Annex 1: Notes on sampling design, sampling strategy, and field methodology
The fieldwork built upon the experience of the Socio-economic survey carried out by IGCP in May-
June, thus most of the interviewers and supervisors selected to do the data collection for the Economic
Valuation had been employed in the socio-economic survey.
The number of questionnaires conducted around each park varied in relation to individual park size,
as follows: BINP=253; MGNP=117; PNV=164; PNVi-Sud=201. Questionnaires were designed to
complement the information already gathered during the socio-economic survey. Questionnaires were
translated in French for DRC and Rwanda and in English for Uganda. The interviewers then
translated them directly in local languages.
Sampling frames and strategies varied according to the survey carried out.
1. Production systems and household contingent valuation (local level) surveys
The sampling frame for these two surveys was based on analysis carried out by the
consultants on the socio-economic data gathered in May-June by IGCP. The analysis related
to the most frequent agricultural production carried out in the different parishes/sectors or
localités, as the production systems survey needed a representation of the various agricultural
systems.
The secteurs/localité/parish chosen on the basis of this analysis can be found in Tables 2.14
and 2.15 of the main report.
The households were chosen no further away than 5 km from the park. The limit of 5 km was
chosen on the basis of:
• The socio-economic survey that showed that beyond 5km the households did not have
strong links to the forest.
• The time available.
The minimum number of interviews per field assistant per day agreed upon was 3 household
sampled for the production systems questionnaire (4 if possible) and 5 households for the
contingent valuation.
To obtain an acceptable number of interviews, the field assistants in Rwanda felt it was
necessary to increase their number of days work. Thus each assistant worked 6 days
collecting a minimum of 3 production systems interviews and 5 Contingent Valuation
interviews per day. As they were working 6 days in each of the secteurs chosen, the working
zone of the 6th day was determined randomly within the 5km from the park.
Table 4: Number of assistants and number of days
Number of
Country Number of FAs days worked
per FA
Rwanda 2 6
DRC 4 5
Uganda MGNP 2 4
Uganda -BINP 6 5
77
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
The consultants ensured the supervising for the first 2 days in Rwanda, MGNP and BINP
with a local supervisor taking responsibility for the remaining days. Because of the
impossibility of the going in the field in DRC, the supervision was delegated throughout.
Supervisors made regular visits to the FA to help them in their work and collect the
questionnaire. This was not felt necessary in the Mgahinga area as the fieldwork was only
carried out for 4 days and the level of education of the FAs ensured good quality of work.
Respondents were sampled according to a "random walk" (see Fig. 1). The field assistants had
the two questionnaires (household Contingent Valuation, and the production system). They
administered them separately on different households. Every other households selected was
interviewed with CVM, and every other household was interviewed with their production
system questionnaire (see Fig. 1).
The method was to interview a respondent working on every third or fourth plot and alternate
the two questionnaires. The xth number of plot was determined according the average field
size in each parish/secteur/localité estimated on the basis of the results of the socio-economic
survey.
Fig. 1: Plot selection by the field assistants (every 3rd field alternating questionnaire). The
route would be the same when the FAs will select every 5th field alternating questionnaires.
PS CVM
Start
F6
F3 F2
F7 F5 F4 F1
F8
F9 CVM
F10
PS CVM F11
F18 PS
F17 F16 F15 F14 F13 F12
F19
F20
F21 CVM
F22
F23
PS
F24
78
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Explanation for the FAs:
Walk along the park at the set distance (0m, 1000m, 2000m, 3000m, 4000m from the Park),
on field F3 or F5 interview a household (CVM). Carry on walking along the park. On field F6
or F10, interview a household (PS).
Then turn left. After 3 or 5 fields (F9 or F15), interview a household (CVM).
Carry on. After 3 or 5 fields(F12 or F20) interview a household (PS)
Then turn right. Interview a household after 3 or 5 fields (F15 or F25), interview a household
(CVM)
Carry on, interview a household after 3 or 5 fields (F18 or F30). Interview a household (PS).
Then turn left. Interview a household after 3 or 5 fields (F21 or F35) etc….
The path will not be straight but it does not matter as long that the starting point is at 0
distance from the park. The FAs will interview the person who works in the nearest field to
the one they are supposed to select. And then carry on the process, but resuming counting
from the plot they were supposed to select. They will have to take record of their steps when
going perpendicular from the Park (away from the park) to make sure they stay in the band of
a 1000 m wide.
The sampling strategy was planned as described above, however in DRC, the fieldwork
coincided with a time when people were not working in their fields. It was thus decided for
the FAs to walk the “random walk” the first day and not the names of the households who
should be interviewed and then interview them in the village the following 4 days. It was also
agreed that the FAs would select more than the amount of households needed in case they
were not to be found when back at the village.
2. Sampling strategy for additional contingent valuation surveys
• local level
• national level
• international level
• Gishwati farmers
FAs were employed to carry out additional Contingent Valuation interviews in order to (1)
increase sample size for local residents surrounding the forest and (2) collect data for national
. One extra FA was employed for 2 days in the three countries to concentrate on the CV
interviews. They were asked to do a transect walk in 2 sectors, interviewing a target number
of 10 respondents in each of the five distance zones from the forest (1-1000m, 1-2000m, etc.
up to 5 kms. from the forest).
Table 5. Number of FAs, data collection days, and locations
Number of
Country Number of FAs days worked
for CV only per FA
Rwanda PNV 1 2
DRC PNVi-Sud 1 2
Uganda-MGNP 1 2
Uganda-BINP 3 3
National 2 1.5
International 1 3
Gishwati 1 2
The sampling strategy for towns was to conduct street questionnaires in different physical
socio-economic sectors, for 2 hours in each sector. Sectors were determined from
consultations with IGCP field staff. 1 FA was utilized for 1 day in Goma; and 2 days each for
Kabale.
79
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
The sampling strategy at international level involved surveys of international beach visitors in
the Mombasa, Kenya area.
As many households as possible were interviewed in the area of Gishwati. One interviewer
from IGCP sampled farming households for two days. The farming households had to have
been cultivating around the forest before it was cleared, or households needed to be
cultivating or have cultivated on the deforested area. 60 households were interviewed.
3. Sampling strategy for the travel cost questionnaire
As many tourists as possible were interviewed at the entrance of the parks for the duration of
the fieldwork. The tourists were interviewed by the one of the consultants after their return
from the gorilla visit. The guides had warned them of the study which was being carried out
which increased the awareness and willingness to be interviewed.
Table 6: Number of days of data collection
Number of days
Country worked
Rwanda PNV 5
Uganda MGNP 2
Congo PNVi- -
Sud
BINP BINP 4
80
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Annex 2: Notes on prices, exchange rates and computation of farm income
Income incorporates both traded and non-traded production.
Prices used to evaluate the economic benefits in the following analysis are farmgate prices which are
the prices that farmers sell their crops at their farm, costs were evaluated at the market prices. Most of
the crops mainly potatoes, sorghum, maize beans and animal products are grown for the local and
national market and a large part is grown for subsistence. Except for cash crops such as pyrethreum,
coffee, tea (for export) and in some instances vegetables which are for mainly sold. Prices vary
according to the season. The only fixed price is the price of the pyrethreum in Rwanda and which
might be distorted however using the market prices is easier and more appropriate in this instance
than looking for shadow prices? The price of tea and coffe might be distorted also due to the situation
of monopoly of the buying companies? Still easier to use market prices or farmgate prices.
The study being carried out across three countries, a common denominator to be able to compare the
economic benefits of the farming activities chosen was the USD and the following exchange rates
were used (these exchange rates were the ones at the time of the field work).
1USD=485 Frw
1USD=1800 Ugandan Shillings
1USD= 275 Fcongolais
The household crop value was calculated as follows:
Total crop value HH=h = P (Qs+ Qc) where:
Qs: Quantity sold in Kg.
Qc: quantity consumed by the household (h)
P: farmgate price: the farmer obtains without factoring in transport (see Table A1 below)
Table A1: Farmgate average crop prices
Crop/Price Rwanda DRC Uganda Rwanda DRC Uganda
(Frw) (Fc) (Ush) (USD) (USD) (USD)
Irish Potatoes 21 16 126 0.04 0.06 0.07
Beans 76 36.5 348 0.16 0.13 0.19
Maize 53 22 206 0.11 0.08 0.11
Sorghum 78 33 219 0.16 0.12 0.12
Wheat 64 390 0.13 0.22
Millet 350 0.19
Peas 103 579 0.21 0.32
Vegetables 15 14
8 0.03 0.01
Sweet potatoes 5 138 0.02 0.08
Yam/Cassava 83 0.05
Colocase 8 0.03
Tea 150 0.08
Coffee 494 0.27
Pyrethrum 50 0.1
During the survey crop prices were investigated for the four different seasons (2 wet and 2 dry
seasons) to account for price variation during the year. The above prices are an annual average per
country. They represent the farmgate prices i.e. price obtained ex-farm without transport.
81
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Annex 3: Questionnaire on production systems
Date: …………….. Field Assistant: ………………………………………. Interview
number:………….
I. HOUSEHOLD
I.1 Plot location: (parish):…………………………Zone: 1 2 3 4
I.2. Name of head of household:…………………………………………………
I.3. How many people live together in the household?………………
II. LAND
II.1. How many plots do you cultivate? …………
II.2. How many m2/proportion of football pitch are your plots? Plot 1……… Plot 2. …….. Plot
3………. Plot 4……… Plot 5……
II.3. Which plots are fallow (resting for more than 1 season) ?………..
II. 4. This year do you cultivate a larger or smaller area than: How many ha/m2 larger or smaller?
In 2001 How many In 2000 How many 5 years ago How many
ha/m2? ha/m2? ha/m2?
Larger/small Larger/smaller Larger/smalle
er r
II.5. Do you own all the plots or fields you cultivate? Yes…. No…..
II.6 If no how many of these field /plots do you rent? ….. ……Which plot(s) are rented: 1 2 3 4 5 6
78
II.7. How much rent do you pay per year per plot? ……………………………………………..
III. WORK
III.1. How many members of the family work or worked on the farm? (Note: write the number in
the squares)
This In 2001 In 2000 5 years ago
year
III.2. Do you or have you ever employ extra people to help? Yes….. No……,
III.3. If Yes: Agricultural 2001-2002 2000-2001 1999-2000 5 year
year ago
a. How many did you employ?
b. Which month did you employ them?
c. How many d/w/m did you employ them d/w/m: d/w/m: d/w/m: d/w/m:
for? (Note: day (d), week (w), month (m))
d. How much does a person helping cost d/m/t: d/m/t: d/m/t: d/m/t:
per day/month/ task (t)?
82
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
III.4 Do you encourage your children to farm? Yes….No…
Why?…………………………………………..
……………………………………………………………………………………………………………
………
III.5. How many children do you have of working age? ………. How many are farmers?
…………………
III.6. If some members of the household don't farm, what do they
do?…………………………………………..
IV. ACTIVITIES AND TOOLS
IV.1. Which months of the year is there the most work and why? What agricultural activities are
carried out during these
months?…………………………………………………………………………………………………
…….
……………………………………………………………………………………………………………
………….
IV.2. Which months of the year is there the least work and why? What agricultural activities are
carried out during these
months?…………………………………………………………………………………………………
……….
……………………………………………………………………………………………………………
……………
IV.3. Which tools do you use to prepare the land and to plant, to weed, or to harvest the crops:
Tools/machine How many do you How many years does it Price per unit?
own? last? (if rented: how much does it
(If rented: how many (if rented: how many cost per tool per day/month)
do you rent?) days/months did you rent
it?)
1.
2.
3.
4.
5.
83
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
V. AGRICULTURAL PRODUCTION
V.1. Which crops did you plant?
This year In 2001 In 2000 5 year ago 10 years ago
1
2.
3.
4.
5.
6.
V.2. What do you from your fields in what quantity per year? (Note: if baskets are more appropriate
please use it as a measure and note how many kgs in a basket or how many basket in a bag)
Crops How many Kgs How many Kgs did How many Kgs did How many Kgs did
have you sold this you sell in 2001? you sell in 2000? you sell 5 years
year ? ago?
1 Kgs/Bags/basket: Kgs/Bags/basket: Kgs/Bags/basket: Kgs/Bags/basket:
2. Kgs/Bags/basket: Kgs/Bags/basket: Kgs/Bags/basket: Kgs/Bags/basket:
3. Kgs/Bags/basket: Kgs/Bags/basket: Kgs/Bags/basket: Kgs/Bags/basket:
4. Kgs/Bags/basket: Kgs/Bags/basket: Kgs/Bags/basket: Kgs/Bags/basket:
5. Kgs/Bags/basket: Kgs/Bags/basket: Kgs/Bags/basket: Kgs/Bags/basket:
6. Kgs/Bags/basket: Kgs/Bags/basket: Kgs/Bags/basket: Kgs/Bags/basket:
(Note: How many kgs is a bag or basket specify for each
crop?………………………………………………………)
V.3. What price do you sell your crops? (Note: If the prices vary per season (S), note the different
prices)
Crops 1. 2. 3 4. 5. 6.
What price do S1: S1: S1: S1: S1: S1:
you sell
Basket/Kg/Ba S2: S2: S2: S2: S2: S2:
g?
(Note 2: Please for each price specify whether it is for a kg , bag or basket for each crop)
V.4. What and how much do you consume of your crops:
Crops How much does the family eat per week (w) or month (m)? For how many months in
the year?
1. Kgs/Bags /basket per w/m: Number of months:
2. Kgs/Bags/basket per w/m: Number of months:
3. Kgs/Bags/basket per w/m Number of months:
4. Kgs/Bags/basket per w/m: Number of months:
5. Kgs/Bags/basket per w/m: Number of months:
6. Kgs/Bags/basket per w/m: Number of months:
(Note: How many kgs is a bag, or in baskets specify for each
crop?………………………………………………...)
V.5. For each crop how many kgs of seed do you buy? How many kgs do you keep to plant? (Per
Year)
84
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Crop How many Kgs/bags of seeds do you buy How many kgs/bags/baskets of your
per year? What does it cost you per Bag/Kg? own crop to you keep to plant?
1. Kgs/bags: Price:………. Kgs/bags/baskets:
Bag/Kg
2. Kgs/bags: Price: ………. Kgs/bags/baskets:
Bag/Kg
3. Kgs/bags: Price: ………. Kgs/bags/baskets:
Bag/Kg
4. Kgs/bags: Price: ………. Kgs/bags/baskets:
Bag/Kg
5. Kgs/bags: Price: ………. Kgs/bags/basket:
Bag/Kg
6. Kgs/bags: Price: ………. Kgs/bags/basket:
Bag/Kg
(Note: How many kgs is a bag, and in a basket specify for each
crop?………………………………………………)
V.6. How much is planted with the following crop? (m2 or proportion of a football pitch)?
Crops 1 2 3 4 5 6
How much?
m2/ proportion
(Note: specify for each crop whether m2, ha is taken as a measure)
V.7. Which crops mix, on the same plots (interplanting)? plot 1…………………………
………………………………………. ……….. Plot 2
………………………….…………………………………
Plot 3………………………………………………………Plot
4………………………………………………….
V.8. After having planted your main crop (specify which:…………………) , which crop do you
plant 2nd or is it left fallow?……………………………. Which crop do you plant 3rd or is it left
fallow?…………..…………….. ……….., which crop do you plant 4th or is it left
fallow?……………….. …………………………………………
V.9. a. If you leave the land fallow, how long do you leave it fallow for? months (m)/years
(y)…………….………
b. How long did you leave the land fallow on average 5 years ago? m/y……………10 years ago?
m/y……………
V.10. If you buy fertiliser, (name of product:…………………………Price per kg/bag……………..)
specify for which
crops:……………………………………………………………………………………………………
………..
V.11. If you buy pesticides (name of product:…………………… Price per kg/bag/l ……… ..…..) or
other chemical, specify for which
crop?……………………………………………………………………………………..
85
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
V.12. How many kgs/bags of fertilisers or other chemicals did you buy in total?
Inputs How many Kgs did How many Kgs did How many Kgs did
How many Kgs did
you buy this year ? you buy in 2001? you buy 5 years
you buy in 2000?
ago?
Fertiliser Kgs/Bags: Kgs/Bags: Kgs/Bags: Kgs/Bags:
Pesticides Kgs/Bags/l: Kgs/Bags/l: Kgs/Bags/l: Kgs/Bags/l:
Other chem.. Kgs/Bags: Kgs/Bags: Kgs/Bags: Kgs/Bags:
(Note: how many kgs is a bag of fertiliser:……………………of pesticide……………of
other…………………)
VI. DYNAMICS OF
AGRICULTURE
VI.1. For the last 5 to 10 years:
Increase Stayed the Decreased
d same
Has your income:
Has the quantity produced from your
fields:
(Note: tick the appropriate box)
VI.2. If your income or production has decreased, could you tell us
why?…………………………………………..
……………………………………………………………………………………………………………
…………….
VI.3. a. Do you try to increase the production on your fields? Yes…. No….
b. How? (change in crops, rotation, tools, fertilisers, manure
etc.)………………………………………………
……………………………………………………………………………………………………
……………..
VI.4. What do you feel the main constraints on farming are?
……………………………………………………….
……………………………………………………………………………………………………………
…………….
VI. 4 a. If crop raiding is mentioned
Ask: How many times it happened this year? Per
w/m/y…………………………………………………………
How many times it happened in 2001? Per w/m/y ………………………….
How many times in 2000? Per w/m/y …………….
Which crops are mainly raided?
……………………………………………………………………………..
86
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
What proportion of which crop did you loose this year?
…………………………………………………… ………………..And last
year?………………………………………………………………………………
VI. 4. b. If erosion is mentioned:
Ask: How many of your fields suffer from erosion?…………………..
What do you do to protect your field from
erosion?……………………………………………………….
How much time per month/year does it take to maintain?
Hours/days……………………………………..
If plant trees: What proportion of your field does it
take?…………………………………………………
VI.5. Which of these problems is your biggest problem for farming? Which is the smallest problem
for you? Please rank all of the following problems by order of importance (1: largest problem to 7:
smallest problem)
Ran Rank Ran Rank
k k
Soil fertility ……. Access to market …….. Storage …… Other (specify): ……
.
Prices of ……. Wildlife ……. Water (for crop) ……
inputs .
Lack of land ……. Prices of the crop …….. No one to sell ……
to .
87
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
VII. ANIMAL
HUSBANDRY
VII.1 Which animals do you have and how many?
Livestock How many do you How many did you have in How many did you have in
have? 2001? 2000?
1. Cows Calves: Adults: Calves: Adults:
2.
3.
4.
5.
6.
VII.2 How much do you sell per year? How many of your livestock do you consume each year?
Livestock Sold per week/month/year Consumed per week/month/year
1. w/m/y : w/m/y :
2. w/m/y : w/m/y :
3. w/m/y : w/m/y :
4. w/m/y : w/m/y :
5. w/m/y : w/m/y :
6. w/m/y : w/m/y :
VII.3. If you produce milk:
How many litres do you sell? w/m ………For how many months per year? …… . …Price
per litre?………
How many litres does your family consume? d/w/m…. …… How many months per
year?………………..
VII.4. If you produce eggs:
How much do you sell? w/m………… How many months per year? ……………Price per
egg?…………
How many eggs your family consumes? w/m …………… How many month per year?
………………..
VII.5. Do you buy feed or fodder for your animals? Yes……No…..…
VII.6. If Yes: For which animals to you buy food for?……….. ………………….
How many Bags/kgs of feed do you need to buy per year?…………….What is the price?
Kg/bag……….
How many bundles of fodder do you buy per
year?……………………….Price/bundle?…………………
VII.7. Where do you take your livestock to eat (forest, own fields, MUZ)?
…………………………………
How many months per year in the field? ………..……. in the Forest ?…………In the
MUZ?…………
88
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
If you could not feed your animals in the forest, where would you feed
them?……………………………
VII.8. How much do you need to spend (in Ugandan shillings) every year for your animals?
Livestock Veterinary fees Medicine Other (specify)
1.
2.
3.
4.
5.
6.
VII.9. What are your main problems for animal husbandry, specify the most and the least
important?…………..…
..….………………………………………………………………………………………………………
………………
VIII.
HONEY:
VIII.1 Do you produce honey? Yes …… No……
VIII.2. If yes: a. How many Litres/kgs of honey do you get per year? L/kgs……………. …………..
b. How much money you get from your honey per week/month/year?……………………..
……
c. How many hives do you have? ……….. .. Did you build them yourself? Yes…No….
d. How much does it cost to buy a hive? ………………
How long does a hive last usually?……………
e. How much time does it take to build one hive?
Hours/days…………………………………………………
f. Where are your hives (field, forest, MUZ…)?
..………………………………………………………………
IX.
WOODLOTS:
IX.1. Do you have woodlots? Yes……No…….
IX.2. If Yes: a. How many m2/ or what proportion of your land/football pitch?………………..
b. Do you plant trees regularly?…………………………….
c. After how many years can you start harvesting the wood? …………
………………………………
89
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
d. What type of trees have you got in your woodlots?
……………………………………………………
e. What do you use the trees for? …………… ……………… ……………………………..
f. How many bundles (small) do you harvest every day/week / month from your
woodlot?………………
How many bundles (big) do you harvest every day/week / month from your
woodlot?……………………..
How many poles do you harvest every year from your woodlot?……………………..
IX.3. Is it enough for your household? Yes…….. No…….
IX.4. If no: a. How many bundles (small) do you buy per w/m?………How many bundles (big) per
w/m?………
b. How many bundles (small) do you collect elsewhere than on you land?
w/m……………How many bundles (big) do you collect elsewhere than on your land?……
IX.5. Do you sell some of your wood? Yes……No…….
IX.6. If yes:a. How many bundles (small) per w/m/y?…………….. At what price?…………..
How many bundles (big) per w/m/y?…………….. At what price?…………..
How many poles per year?…………….. At what price?…………..
IX.7. Since you have started using your woodlot, you are able to harvest : the same…….less…….
more……
FOREST PRODUCT USES:
X.
STAKE
S
X.1. Do you use stakes? Yes…… No…. X.2. If Yes: For which
Crop?…………………………………….
X.3. a. How many bundles of stakes do you use per year?………………………………...
b. How long do the stakes last? months/years ……………How many stakes are there per
bundles?…………
c. How many bundles do you buy per year?…………………………….
X.5. If you collect/harvest stakes: a. How many bundles do you collect/harvest per year?
…………………………
b. Where do you collect/harvest the stakes (own fields, other plantation, forest, MUZ….)?
………………………
(if forest or MUZ is mentioned: How many bundles come from the forest?……… …..from the
MUZ?…………)
90
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
c. How many people from the household go to collect/harvest the stakes together ?
……….………………………
d. How long does it take to collect/harvest enough for your crop? days/hours
……………………………….
X. 6. Which wood are the stakes made of mainly? ……………………………………
XI. BAMBOO
XI.1 Do you use bamboo? Yes….. No……
XI.2. If yes: What do you use bamboo for? …… ………………………………………………….
How many bundles (constuction/baskets) do you use per w/m/y? ………
How many bundles (stakes) do you use per w/m/y? ………
Do you buy bamboo? Yes…No.... If yes:How many bundles (stakes) per w/m/y? …………
How many bundles (construction) per w/m/y? …………………
How long does a bamboo (construction) lasts? m/y…………………
Price of a bundle (construction)?………………………Price of a bundle
(stakes) ?…………………….
If you make baskets: how many baskets do you make per bundle?………. How many baskets do you
make per month?……………How long does it take to make one basket?…………….How much do
you sell 1 basket?…….
XI.3. Do you collect/harvest bamboo? Yes….No…..
XI.4. If yes:a. How many bundles (const.) do you collect/harvest per w/m /y?………
How many bundles (stakes) do you collect/harvest per w/m/y?…………………………..
b. How many times do you go and collect/harvest bamboo per
w/m/y?……………………………………
c. How long does it take to collect the amount you need? hours/day ……………..
d. Usually how many people of the family go and collect bamboo at the same time?………..
……………..
e. Where do you go and collect it (forest, own, MUZ etc.)?
…………………………………………………………
f. If forest or MUZ, how many bundles (stakes) do you collect in the forest?……………In
MUZ?……………….
If forest or MUZ, how many bundles (const.) do you collect in the forest?……………In
MUZ?……………….
XI.5. Do you sell bamboo? Yes… No…..
XI.6. If yes: a. How many bundles (const.) do you sell per w/m/ y?…..……How many bundles
(stakes)?w/ m/y…..
91
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
b. Price you sell a bundle (const)?…………(Note: how many sticks in a bundle……..)
Price you sell a bundle (stake)?………. ( Note: how many sticks in a bundle……..)
c. Since you have started collecting bamboo, is it more easy…… the same ……. more difficult
…….. to find?
d. In your opinion, is the situation going to: Improve………get worse………stay the same…….?
Why?…
……………………………………………………………………………………………………………
……………
XII. BUILDING POLES OTHER THAN BAMBOO
XII.1. Do you buy poles for building? Yes….No….
XII.3. If yes: How many have you bought this year? Poles (big)………… Poles (small)……....
XII.4. Do you collect/harvest poles? Yes….. No…..
XII.5. If yes: a. How a many poles (big) have you collected/harvested this year?………….. Last
year?……….
How a many poles (small) have you collected/harvested this year?………….. Last
year?……….
b. Where do you collect/harvest the poles from (forest, own plantation, MUZ
etc.)?…………………………
c. How many poles (big) come from the forest this year?……………How many from the
MUZ?…….
d. How many poles (small) come from the forest this year?……………How many from the
MUZ?…….
e. How many people of the family go and collect the building poles at the same
time?………………………
f. How long does it take to collect the amount you need?
Hours/days..……………………………………..
XII.6. Do you sell poles for building? Y…. N…..
XII.7. If yes: a. How many poles (small) do you sell per w/m/ y?……….…………At what price a
pole?…
b. How many poles (big) do you sell per w/m/ y?……….…………At what price a pole?…
c. Since you have started collecting poles, is it: more easy…… the same ……. more difficult
…….. to find?
d. In your opinion, is the situation going to: Improve………get worse………stay the same…….?
Why?…
92
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
……………………………………………………………………………………………………………
……………
XIV. MEDICINAL PLANTS:
XIV.1. Do you use medicinal plants? Yes… No….
XIV.2. If yes Which plants mainly? 1……………2………………3…………….4……………
XIV.3.: How much do you use per year/week/month?
Plants 1 2. 3. 4.
How much do you w/m/y: w/m/y: w/m/y: w/m/y:
use per w/m/y?
What is the price
of:
(Note: you will need to find the unit of measure and specify for each plant: grams, bundles etc.)
XIV.4. Do you sell medicinal plants? Y…N…..
XIV.5. If yes: a. Which plant to you sell mainly?
…………………………………………………………………
b. Where do you collect them (forest, MUZ, own woodlots, own garden
etc)?……………………………
c. How much money do you make selling the plants per
w/d/m/y?…………………………………………………..
d. How many people of the family collect the plants at the same
time?………………………………………..………
e. How long does it take to collect enough plants per d/w/m/y)?
days/hrs…………………………………………….
XIV. 6. Since you have started collecting plants, is it: more easy…… the same ……. more difficult
….. to find?
XIV.7. In your opinion, is the situation going to: Improve………get worse………stay the
same…….?
Why?……………………………………………………………………………………………………
………….
93
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
XVI. WATER
XVI.1. How much water do you need per day? litres/jericans……………. (Note: how many litres in a
jerrican….………) Are you part of the GFS? Y..N…
XVI.2. Do you buy water? Yes……No…… For how many months in the year?…………
XVI.3. If Yes: How many litres/jerricans per day/week? …………. Price of a
litre/jerrican?…………………….
XVI.4. Do you collect water? Yes….No…. For how many months in the
year?…………..
XVI.5. If yes: Where (forest, MUZ, tap, own…)?………………………….
XYI.6. If you collect water from the forest:
a. Which months do you collect water from the
forest?…………………………………….……………..
b. How many litres/jerricans do you collect per day/week?
………………..………………………………
c. How many people of the house usually go to collect the water per day?………………….
d. How long does it take for one person get enough water? hours……………………….
e. If you could not collect water from the forest what would you have to
do?………………………………
f. How far would you have to go? hours walk………………..
94
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Annex 4: Crop mix and diversity by country, parish, and distance zone
Figs. A1, A2 and A3 illustrate crop choice and diversity for each country, by sampled parish/location
on the left and by km. zone 1-5 on the right. Potatoes (PDT), beans, maize, sorghum are the most
widespread crops. Although similar crops are grown across parishes, their importance varies –
presumably according to micro-climate and/or soil differences. Crop mix varies only slightly
according to distance from park. Results show that cash crops (pyrethrum for Rwanda, tea and coffee
for Uganda, and vegetables such as leeks and carrots in DRC) are grown in localised areas.
For Rwanda (Fig. A1), potatoes are grown by more than 80% of the households and beans by more
than 60% except in zone 5. Although grown in all zones pyrethrum is the fact of one parish mainly.
PDT PDT
100 100
Beans Beans
90 90
Maize Maize
80 80
Sorg Sorg
70 Wheat 70
Wheat
60 Peas 60 Peas
50 Veg % 50 Veg
%
40 pyrethre 40 pyrethre
30 30
20 20
10 10
0 0
1 2 3 4 Rw anda
1 2 3 4 5 Rwanda
Localites Zone
Fig. A1: Percentage of households cultivating identified crops by locality and zone in Rwanda
In DRC (Fig A2), beans, maize and sorghum are produced by most of the households. Vegetables are
also important particularly in localite 3 (Kibumba) where more than 80% of households produce
mainly leeks, cabbages and carrot. Sweet potatoes and colocase are also grown in DRC. This does
not vary across the zones.
100 100
PDT PDT
90 90
Beans Beans
80 80
Maize Maize
70 70 Sorg
Sorg
60 60 Veg
% Veg %
50 50 Sweet pot.
Sweet pot.
40 40 Coloc
Coloc
30 30
20 20
10 10
0 0
1 2 3 4 5 DRC 1 2 3 4 5 DRC
Localites Zones
Fig. A2: Percentage of households growing identified crops by locality and zone in DRC
Beans, potatoes, maize, sorghum, and millet are the staple crops in Uganda (Fig. A3). This varies
across parishes as well as zones in some instances. Sweet potatoes are an important crop around
Bwindi (parishes 3 to 6). Cash crops such as tea and coffee are confined to one parish, while yam and
cassava are much more prevalent in Uganda than Rwanda and DRC.
95
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
100 P DT 100 P DT
B eans
90 B eans
90
M aize
M aize
80 80
S org
S org
70 W heat 70 W heat
M illet
60 M illet
60
P eas
P eas
50 50
S weet pot. %
% S weet pot.
40 Y am c as 40 Y am cas
tea
30 tea
30
coffee
coffee
20 20
10 10
0 0
1 2 3 4 5 6 Uganda 1 2 3 4 5 Uganda
Pa rish Zone s
Fig. A3: Percentage of households growing identified crops by parish and zone in Uganda
96
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Annex 5: The Travel Cost and Contingent Valuation questionnaires
1. TRAVEL-COST QUESTIONNAIRE FOR GORILLA TOURISTS
The park authorities are conducting an economic valuation of this park in order to better justify the
importance of protecting the gorillas for the future. One part of this involves how much tourists like
yourself have spent on coming to see the gorillas. For this purpose, they would very much appreciate
knowing the following information:
1. Was your safari to Africa a package deal i.e. all expenses included? Yes ___ No ___
2. Did you book through an overseas company, or a local company? ________________
If local, country? ______________
3. If Yes to (1), how much? Amount _____ Currency_____
4. If Yes to (1), did this amount include the air fare? Yes ___ No ___
5. If Yes to (3), how much? Amount _____ Currency _____
6. If Yes to (1), how much extra will you spend over & above the safari price?
Amount _____ Currency _____
7. If No to (1), how much will you spend on your safari? Amount _____ Currency _____
how much did you spend on the air fare? Amount _____ Currency _____
how much extra will you spend over & above the safari + airfare?
Amount _____ Currency _____
8. How much time did it take from your home until you reached Africa? _____ hrs
9. How many days will you spend on safari? _____ days
10. How many days will you spend in this park? _____ days
11. What is the main reasons you are visiting this park? 1st__________ 2nd ___________
3rd __________ 4th ___________
12. Which hotel are you staying at? Name _____________ Town/place _________
13. Was this park your first choice for seeing gorillas? Yes ____ No ____
14. If No to (11), which was your first choice? Park name _______________
15. How many days will you spend in this country? ____ days
16. The park would to develop other activities such as hiking trails & climbs so that visitors can
observe the forest, other unique wildlife, and high mountain views. If these were already available
now, how much more time would you have liked to spend in this park?
None ____ One more day _____ Several more days _____ Other (specify)_____
17. In which country do you live? __________ Is that where you travelled from? Yes ___ No __ 18.
What is your profession? ________________
19. What is your annual income? Amount _________ Currency _________
20. Number in your household? _____
On behalf of the park authorities, thank you very much for your assistance. Your information will be
of help in safeguarding the future of this park.
97
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
2. CV QUESTIONNAIRE FOR PRODUCTION SYSTEMS/HOUSEHOLDS (local
level)
Sector _____________ Date ________ Name of Interviewer ____________
[Distance from Park _____________________]
(zone: 0-1000m 1000-2000m 2000-3000m 3000-4000m 4000-5000m)
These questions concern your attitude towards the forest (Bwindi):
A. How is the forest (incl. wildlife) important to your life (good things) – in order of importance?
No.1 ____________ No.2 ______________ No.3 _____________ No.4 _____________
No.5 ____________ No.6 ______________ No good ways ______________
B. Suppose that the forest was going to be cut down because the Park Authorities could not afford to
keep up the cost of looking after it, and also that you were NOT going to benefit from what was taken,
nor gain land. If the Government proposed that the forest could be saved if every household in the
country paid Ush 18000 every year, would you be willing to pay this annual fee?
Yes = 1 No = 2
C. If Yes to (B), would you be willing to pay an annual fee of Ush 36000 to protect the forest?
Yes = 1 No = 2
D. If No to (B), would you be willing to pay an annual fee of Ush 9000 to protect the forest?
Yes = 1 No = 2
E. If No to (D), what is the maximum you would be willing to pay each year to protect the forest?
Ush_______________
F. If (E) is zero, why do you not want to pay anything? _________________________________
G. How does the forest (incl. animals) impact your life badly – in order of importance?
No.1 ____________ No.2 ______________ No.3 _____________ No.4 _____________
No.5 ____________ No.6 ______________ No bad ways ______________
H. The conversion of the forest into a National Park means that you have given up or lost values like
land for agriculture, wood, building poles, bamboo, and other benefits. Suppose you were to be
compensated for the benefits you lose or give up because the forest is a Park. If you were given Ush
54,000 every year from now on, would this equal the value of the benefits you are giving up?
Yes = 1 No = 2
98
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
I. If No to (H), would Ush 108,000 compensation every year equal the value you are giving up?
Yes = 1 No = 2
J. If No to (I), state any other amount Ush ______________that you would need to receive in
compensation every year, in order to equal the value you are giving up.
K. Now a general question: overall, do you think it is good for the forest to be there, or not?
Good = 1 Not good = 2
For research analysis purposes, we would appreciate if you can give us the following:
L. How many people in your household? __0-3 __4-6 __7-9 ___other (specify)
M. What is your educational level? 0 = none 1= 1-4 primary 2 = 5-6 primary
3 = 1-4 post-primary 4 = secondary 5 = university
N. How much was your production last season? Potatoes _______________ (kg / bags)
Sorghum _______________ (kg / bags)
Maize _________________ (kg / bags)
Wheat _________________ (kg / bags)
Beans __________________(kg/Bags)
Millet__________________(kg/Bags)
Other __________________(kg / basket / bags)
O. Does anyone in your household earn wages? Yes = 1 No = 2
P. If Yes, what is the value of wages per month? Ush _____________
Q. Do you own the land you work on, or rent?
Own = 1 Rent = 2 Part own/part rent = 3
99
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
3. CV QUESTIONNAIRE FOR CITIES/TOWNS (national level)
City/Town ___________________
These questions concern your attitude towards the PNV / Pvi / Mgahinga forest:
A. Does the forest affect your life? Yes = 1 No = 2
• If Yes, terminate survey
• If No, go to (B)
B. In your opinion, what are the Good things the forest gives – in order of importance?
No.1 ________ No.2 _________ No.3 ___________ No.4 ___________ No.5 ________
No.6 ________ No.7 _________ No.8 ___________ No good things ____
C. In your opinion, what are the Bad things the forest brings – in order of importance?
No.1 ________ No.2 _________ No.3 ___________ No.4 ___________ No.5 ________
No.6 ________ No.7 _________ No.8 ___________ No bad things ____
D. Overall, do you think it is good for the forest to be there, or not?
Good = 1 Not good = 2
E. If you heard that the forest was going to be cut down because the Park could not afford to keep up
the cost of looking after it, or that if everybody in the country paid ______________ every year, then
the forest could be saved, would you be willing to pay this annual fee?
Yes = 1 No = 2
F. If Yes to (E), would you be willing to pay an annual fee of _________ to protect the forest?
Yes = 1 No = 2
G. If No to (E), would you be willing to pay an annual fee of __________ to protect the forest?
Yes = 1 No = 2
H. If No to (G), what is the maximum you could afford to pay each year to protect the forest?
_______________ (state amount + currency)
I. If (H) is zero, why do you not want to pay anything? _______________________
[continued on page 2……]
100
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
[continued from page 1…...]
For research analysis purposes, we would appreciate if you can give us the following:
P. How many people in your household? __0-3 __4-6 __7-9 ___other (specify)
Q. What is your educational level? 0 = none 1= 1-4 primary 2 = 5-6 primary
3 = 1-4 post-primary 4 = secondary 5 = university
R. What is the monthly value of the crops you produce? ___________ (state amount + currency)
S. Does anyone in your household earn wages? Yes = 1 No = 2
T. If Yes, what is the value of wages per month? ___________ (state amount + currency)
Thank you very much for your assistance.
101
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
4. INTERNATIONAL VISITORS CV QUESTIONNAIRE (non-gorilla)
Can you spare 5 minutes to help us with a questionnaire? We are working on behalf of an
international conservation organization based in Kenya called the International Gorilla Conservation
Program (IGCP). We are conducting a survey amongst international visitors relating to conservation
of the endangered mountain gorillas.
Background
Are you aware of the status of the Mountain Gorilla? [endangered – yes – no] [threatened – yes – no]
There are only 600 mountain gorillas left in the wild, living in 2 small forests – one in Uganda, the
other shared by 3 countries – Uganda, Rwanda, and Congo (DRC). The forests are surrounded by
farmland, and are under pressure by poor people in search of land and resources.
1. If a special fund for the conservation of mountain gorillas was established in your country under a
respected organization such as the World Wildlife Fund (WWF), with the money be spent directly on
conservation efforts to secure the future of the Mountain Gorillas, would you be willing to pay 30
Euros / US$ (whichever applies) every year into this fund?
Yes = 1 No = 0
2. If Yes to 1, would you be willing to spend 60 Euros / US$ per year? Yes = 1 No = 0
3. If No to 1, would you be willing to spend 15 Euros / US$ per year? Yes = 1 No
=0
4. If No to 3, what amount would you be willing to spend? ___________Euros / US$
5. If Zero to 4, why not? _______________________________________
The following questions relate to the reasons why you would be willing/not willing (as applies)
[present 1st]
Important (I) Slightly imp (SI) Not
Imp (NI)
6a. I may want to see a Mountain Gorilla in the future. I SI NI
6b. I enjoy knowing other people can enjoy them. I SI NI
6c. I enjoy knowing future generations will enjoy them. I SI NI
6d. I enjoy knowing mountain gorillas exist even if no-one
ever sees one. I SI NI
6e. All endangered species have a right to exist. I SI NI
The following questions relate to your attitude towards the environment
[STA = strongly agree SWA = somewhat agree U = unsure SWD = somewhat disagree STD =
strongly disagree]
7a. We are approaching the limit of the number of people they earth can support.
STA SWA U SWD STD
7b. The so-called ‘ecological crisis’ facing humankind has been greatly exaggerated
STA SWA U SWD STD
7c. Humans were meant to rule over the rest of nature
STA SWA U SWD STD
7d. When humans interfere with nature it often produces disastrous consequences
STA SWA U SWD STD
7e. The balance of nature is strong enough to cope with the impacts of modern industrial nations
STA SWA U SWD STD
Finally some general questions
8. Which country do you live in? ______________________
102
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
9. What is your age? ______yrs
10. Do you mind indicating the annual income of your household? ____________Euros / US$ /
other
Thank you for your time and help
103
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
5. CV QUESTIONNAIRE FOR GISHWATI FARMERS
(Distance from cleared forests ________ kilometres)
1. Are you a farmer? Yes ___ No ___ IF YES, continue IF NO, do not continue
2. Do you own or rent the land you farm? Own = 1 Rent = 2 Own & Rent = 3
The following questions concern the changes that have happened since the forests were cut
down in this area in 1996:
A. According to you, has there been a change in the flow of water in streams? Yes = 1 No = 2
B. If Yes to (A), what changes ? decrease in waterflow year-round = 1
decrease in dry season waterflow only = 2
increase in wet-season waterflow = 3
C. According to you, has there been a change in soil quality? Yes = 1 No = 2
D. If Yes to (C), what changes? better soil quality = 1 worse soil quality = 2
Suppose that the Government started a project to replant trees on those areas where the forests were
cut down since 1996. This would be in order to stop soil erosion, decrease any flooding, and increase
the amount of water flowing in streams in the dry season. Suppose all those who live and farm
around these replanted forests had to pay an annual contribution to the project to cover its annual
costs. Assume that the project was successful.
E. Would you be willing to pay an annual fee of ____ Frw __________ for these benefits provided
by the project? Yes = 1 No = 2
F. If Yes to (E), would you be willing to pay an annual fee of ____ Frw _________ for these benefits
provided by the project? Yes = 1 No = 2
G. If No to (E), would you be willing to pay an annual fee of ____ Frw _________ for these benefits
provided by the project? Yes = 1 No = 2
H. If No to (G), what is the maximum you could afford to pay each year for the benefits provided by
the project? ____ Frw ___________
I. If (G) is zero, why do you not want to pay for this service? _________________________
For research analysis purposes, we would appreciate if you can give us the following:
J. How many people in your household? 0-3__ 4-6__ 7-9__ other (specify) ____
K. What is your educational level? 0 = none 1= 1-4 primary 2 = 5-6 primary
3 = 1-4 post-primary 4 = secondary 5 = university
L. What is the monthly value of the crops you produce? ___________ (state amount + currency)
M. Does anyone in your household earn wages? No = 2 Yes = 1
If Yes, amount per month? _______________ (add currency)
104
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Annex 6: Travel Cost demand estimation results for international gorilla-park visitors sample
c.s. = consumer surplus
pp = per person
1. MODEL 1: Groups Model: US, Europe (EUR), and Pacific (PAC)
Y = 0.02304 – 0.00000744X X = when Y=0 (US, EUR, PAC)
Mean expenditure per US visitor = $2216
So total US expenditure per year (based on 1515) = $3,357,240
Mean consumer surplus per US visitor =( (3162-2216)*0.00606)/2 = $2.866 per 1000 US pop
(250 million) =$716,500 total US consumer surplus per year
Alternative calculation: total US c.s. per year = ((3162-2216)*1515)/2 = $716,595
Pp cs = $473
Mean expenditure/revenue per EUR visitor = $891
So total EUR revenue per year (based in 3218) = 2,867,238
Mean consumer surplus per EUR visitor = (3162-891)*0.01287)/2 = $14.614 per 1000 Eur pop
250 million =$3,653,500 million total EUR consumer surplus per year
Alternative calculation: total EUR c.s. per year = ((3162-891)*3218)/2 = $3,654,039
Pp cs = $1135.50
Mean expenditure/revenue per PAC visitor = $1055
Total exp per year based on 1443 = $1,522,365
Mean consumer surplus per PAC visitor = ((3162-1055)*0.01924/2) = $20.269
Total population: 75 million. Total PAC consumer surplus per year based on 1443 = 1,520,175
Alternative calculation: ((3162-1055)*1443)/2 = 1,520,200
Pp cs = $1053.50
2. MODEL 2: Whole sample
Y = 80.73 –0.023X; mean 1107; n=105; sd 1106; not-sign
Mean exp per visitor on gorillas = $1107 for 6176 visitors =6,836,832
Total c.s. pp = ((3510-1107)*55)/2 = $66,082.50
Mean cs pp = 66082.5/105 = $629.36
So total cs per year based on 6176 visitors = $3,886,927
Mean expenditure per visitor on gorillas= $1145
So total expenditure per year for --- intl visitors are 8133 ((US1750 + EUR4350) representing
75% of international visitors) = $9,312,285 total expenditure per year
Mean consumer surplus = (((3773-1145)*50)/2)/50 = $1314
So total consumer surplus based on 8133 visitors per year = $10,686,762
2. MODEL 3: Whole sample (log-linear)
Y = 318 –39.40X log X mean” 6.6684 ($788) X=8.0725 ($3203) when Y=0; n=100
Total sample cs = $66,412.50
Mean cs = $632.50
Total annual cs based on 6176 visitors = $3,906,320
105
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Annex 7: WTP/WTA model specification and estimation, and basis for extrapolation to the
wider population
(Reference: standard advanced statistics texts e.g. Garson, G. David (2001) ‘Guide to Writing
Empirical Papers, Theses, and Dissertations’ (print ISBN 0-8247-0605-6] Marcel Dekker.com)
The following linear model - that is, that the probability of accepting a bid is a dependent on the level
of the bid offered, household income, and education level - was specified in order to explore various
factors influence on the acceptance/non-acceptance of an offered bid. Making the common
assumption that the distribution for CV responses is non-normal (see ---) requires that a maximum
likelihood estimation (MLE) technique be used. In addition, since it was assumed that the underlying
relationship between bid levels and acceptance is logistic in nature, binomial logistic regression was
chosen as a suitable estimation method over the ‘logit’ and ‘probit’ alternatives, both of which yield
the same conclusions but are more difficult to interpret.
P(y) = A + b1X1 + b2X2 + b3X3
Where:
P(y) = probability of accepting a bid offer
A = constant term
X1 = bid offer (US$)
X2 = household income (US$)
X3 – education level (categories 1-7)
And b1, b2, b3 are the coefficients of X1, X2, X3 respectively
The solution yielded the following results:
• "-2 Log Likelihood" significant at 0.05 (p=0.001), rejecting the null hypothesis that none of
the independents are linearly related to the log odds of the dependent.
• Hosmer and Lemeshow Goodness-of-Fit significance at 0.05 (p=0.1492), indicating that the
model is an acceptable fit (failing to reject the null hypothesis that there is no difference
difference between the observed and predicted values of the dependent.
• Bid function coefficients for each variable (B), Wald statistic, significance level (Sig), and
odds-ratio (Exp(B)) as follows:
------------------------ Variables in the Equation ------------------------
Variable B S.E. Wald df Sig R
Exp(B)
BID$OFFR(1) - .5142 .2748 3.5016 1 .0613 -.0545
.5980
HH$INCOM .0002 .0002 1.0363 1 .3087 .0000
1.0002
EDUCN 17.57225 .0035
.1223
EDUCN(1) -1.9740 1.4404 1.8782 1 .1705 .0000
.1389
EDUCN(2) -1.7095 1.4356 1.4179 1 .2338 .0000
.1810
EDUCN(3) -1.3859 1.4343 .9337 1 .3339 .0000
.2501
EDUCN(4) -1.5715 1.5198 1.0692 1 .3011 .0000
.2077
EDUCN(5) -.5465 1.4409 .1438 1 .7045 .0000
.5790
106
Economic Valuation of the Virunga and Bwindi protected forests
IGCP 2007 Hatfield & Malleret-King
Constant .4076 1.4494 .0791 1 .7785
• Odds ratio (Exp(B)) example for bid offer variable (‘bid$offr’): Since a ‘1’ value in the
dependent corresponds to a ‘Yes’ bid, and the ‘bid$offr’ lower value = $5 with higher value =
$20, and given that an odds ratio less than 1 corresponds to decreases, the odds ratio of
0.5980 indicates that with a unit increase in bid$offer (i.e. from $5 to $20) the odds of a ‘Yes’
bid are multiplied by 0.5980, which is a 40.20% (1-.5980) decrease.
• Mean WTP computation holding other variables at mean value: P(0.5) = 0.4076 – 0.5142X –
1.9740 (median education level (1)) + 0.061 (average hh income = $305) = US$3.90
WTA equation logistic regression
------------------------ Variables in the Equation ------------------------
Variable B S.E. Wald df Sig R
Exp(B)
WTA$OFFR(1) .2404 .2557 .8838 1 .3472 .0000
1.2717
HH$INCOM .0005 .0002 4.0919 1 .0431 .0690
1.0005
EDUCN 10.6968 .0398
5.0577
EDUCN(1) 3.3456 13.5028 .0614 1 .8043 .0000
28.3789
EDUCN(2) 3.6976 13.5017 .0750 1 .7842 .0000
40.3505
EDUCN(3) 4.1410 13.5012 .0941 1 .7591 .0000
62.8635
EDUCN(4) 3.9913 13.5117 .0873 1 .7677 .0000
54.1240
EDUCN(5 4.5615 13.5018 .1141 1 .7355 .0000
95.7314
Constant -5.5621 13.5000 .1698 1 .6803
Extrapolation of results to wider population: Local Level:
Park Park SurroundingSample Sample No. No.
perimeter sample population average households households
(kms) population @ 200 & household @ 200/sq @ 400/sq
400/ sq km
area (sq km) size km density km density
density
PNV 60 300 60,000 / 6.5 9231 18,462
120,000
PNVi-S 44 220 44,000 / 6 7333 14,666
88,000
MGNP 60 60 12,000 / 4.9 2449 4898
24,000
BINP 104 520 104,000 / 6.7 15,522 31,044
208,000
Compares to (Sikoyo, 1995): Bwindi households: 17,900 for all surrounding parishes (approximates
above estimate at population density 200 people / sq. km.
107