Dissertation of Hsing

Dissertation of Hsing
On Sustainable Use of Renewable Resources in Protected
Areas as an Instrument of Biodiversity Conservation:
A Bioeconomic Analysis
INAUGURAL-DISSERTATION
zur
Erlangung der Doktorwürde
der
Wirtschaftswissenschaftlichen Fakultät
der
Ruprecht-Karls-Universität Heidelberg
vorgelegt von
Hsing-Sheng Tai
aus Taiwan
Mai 2002
Eidesstattliche Erklärung
Hiermit erkläre ich, Hsing-Sheng Tai, daß meine bei dem Promotionsausschuß der
Wirtschaftswissenshaftlichen Fakultät der Ruprecht-Karls-Universität Heidelberg
eingereichte Dissertation mit dem Thema: „On sustainable use of renewable resources
in protected areas as an instrument of biodiversity conservation: a bioeconomic
analysis“:
1.
2.
3.
4.
von mir selbständig angefertigt wurde und andere Quellen und Hilfsmittel als die
angegebenen nicht benutzt wurden,
daß die Dissertation weder in dieser noch in einer anderen Form einer anderen
Fakultät vorgelegt worden ist,
daß die Dissertation weder als Ganzes noch Teile daraus anderweitig als
Prüfungsarbeit bei einer akademischen oder Staatsprüfung verwendet worden ist
und
daß von mir keine, von einer anderen Prüfungsbehörde zurückgewiesene
Dissertation oder in einem sonstigen Prüfungsverfahren als Prüfungsteil
verwendete Arbeit vorgelegt worden ist.
Hualien, Taiwan, im Mai 2002
Abstract
The objective of this dissertation is to provide a theoretical framework for answering
the question, whether and under which biological and socio-economic conditions the
sustainable use of wild species in or around protected areas is an adequate strategy for
biodiversity conservation. To do this, the dynamic interaction between the use of wild
species, management of protected areas, population levels of the utilized species and
poaching is investigated. A nonlinear bioeconomic model with two state variables
(resource stock, management capital stock) and two control variables (harvest rate,
investment rate) is developed on the basis of the traditional bioeconomic model and
optimal control theory. The model identifies eight fundamental factors that influence
the equilibrium population levels of the utilized species. In sum, the lower the discount
rate, the depreciation rate of management capital, the poaching coefficient and the cost
coefficient of investment are, and the higher the intrinsic growth rate of species, the
non-consumptive value coefficient of species, the efficiency coefficient of
management capital and the gross profit coefficient of harvest are, the higher the
equilibrium population levels of harvested species will be.
At the theoretical level, our model suggests that, apart from the usually discussed
intrinsic growth rate, discount rate and price/harvest cost ratio, more factors should be
taken into account when considering the impact of harvest on resource stock.
Moreover, contrary to the conclusion of the Clark model and to the popular belief, our
model demonstrates that, other things being equal, the higher the gross profit
coefficient of harvest (termed as price/harvest cost ratio in the context of the Clark
model) is, the greater the equilibrium resource stock will be. This conclusion is
consistent with that drawn by the Swanson model, which first considered the important
I
role of management in resource harvest problem. However, compared to the Swanson
model, our model provides a more deliberate and extensive modeling for investigating
the resource harvest and management problem.
According to the model results, the eight parameters can be evaluated as indicators for
assessing the feasibility of a sustainable use project as a conservation strategy before or
when it is applied in specific sites. An assessment procedure is then developed and
applied for the case study with reference to several conservation programs in the A-LiShan area of Taiwan. The results of the assessment procedure are in principle
consistent with what happened in the reality. The case study shows that, some factors
that are newly introduced in our model, namely the non-consumptive value coefficient
of species, the efficiency coefficient of management capital and the cost coefficient of
investment explain the differences of the performance of various conservation
programs in the area concerned.
II
Contents
1 Introduction
1
1.1 The Problem concerned and the objective of dissertation………………………..
1.2 Methods…………………………………………………………………………..
1.3 Contents of the dissertation………………………………………………………
2 Biodiversity: Concept and its loss rate
2.1 Definition of biodiversity………………………………………………………..
2.2 Measurement of biodiversity and some indicators………………………………
2.3 The loss rate of biodiversity……………………………………………………..
3 The nature of protected areas as an instrument of biodiversity
conservation
3.1 Definition and classification of protected areas: The IUCN system……….……
3.2 Economic theories justifying the existence of protected areas: theories with
reference to nonrivalry and nonexcludability……………………………….…...
3.3 Economic theories with reference to uncertainty and irreversibility: option
value, quasi-option value and the Safe Minimum Standard……………….…….
3.4 The Perrings and Pearce Model………………………………………………….
3.5 Concluding remarks……………………………………………………………..
4 State of protected areas and the debate on sustainable use of
renewable resources in and around protected areas as an
instrument of biodiversity conservation
4.1 Effectiveness of protected areas: a global perspective…………………………...
4.2 Current problems of protected areas……………………………………………...
4.3 Defining ‘sustainable use of renewable resources’………………………………
4.4 The debate on sustainable use of renewable resources in and around
protected areas as an instrument of biodiversity conservation…………………...
4.4.1 Background of the debate…………………………………………………..
4.4.2 Perspectives of the sustainable use approach………………………………
4.4.3 The Community-Based Conservation (CBC)………………………………
4.4.4 Perspectives of the preservation approach…………………………………
I
1
4
5
7
7
9
11
15
15
18
19
21
27
28
28
31
35
37
37
38
41
43
4.5 A case study: the national park system of Taiwan……………………………….
4.5.1 Introduction to the national park system of Taiwan………………………..
4.5.2 Management issues…………………………………………………………
4.5.3 Effectiveness of the national park system…………………………….……
4.5.4 Current problems of the national park system……………………………...
4.5.5 Prospects of the national park system……………………………………...
5 Economic models of species extinction and biodiversity loss
5.1 The Gordon model.………………………………………………………………
5.2 The Clark model………………….. ………………..…………………………...
5.3 The Swanson model……………………………………………………………..
5.4 Concluding remarks……………………………………………………………..
6 Use of renewable resources, poaching and anti-poaching: a
simple bioeconomic model with one state variable and two control
variables
6.1 Introduction……………………………………………………………………...
6.2 The model………………………………………………………………...……...
6.3 Uniqueness of the steady state solution………………………………………….
6.4 Stability of the steady state solution……………………………………………..
6.5 Phase diagram analysis…………………………………………………………..
6.5.1 Phase diagram (X, h)……………………………………………………….
6.5.2 Phase diagram (X, E)……………………………………………………….
6.6 Comparative static analysis……………………………………………………....
6.7 A special case of the simple model………………………………………………
6.8 Concluding remarks and policy implications…………………………………….
Appendix 6.1………………………………………………………………………...
Appendix 6.2………………………………………………………………………...
7 Management capital, use of renewable resources, poaching and
anti-poaching: a bioeconomic model with two state and two
control variables
7.1 Introduction………………………………………………………………………
7.2 Management capital……………………………………………………………...
7.3 The extended model……………………………………………………………...
7.4 Uniqueness of the steady state solution……….…………………………………
II
46
46
50
52
54
58
59
59
60
62
69
71
71
72
79
80
82
82
84
87
90
94
97
98
100
100
101
103
107
7.5 Stability of the steady state solution……………………………………………...
7.6 Comparative static analysis……………………………………………….……...
7.7 Concluding remarks and policy implications…………………………………….
Appendix 7.1. ……………………………………………………………………….
8 Management capital, use of renewable resources, poaching and
anti-poaching: a general bioeconomic model
109
112
115
118
120
8.1 The general model……………………………………………………………….. 120
8.2 Existence of the steady state solution……………………………………………. 123
8.3 Phase diagram analysis: computer simulation…………………………………… 124
8.4 Comparative static analysis: computer simulation………………………………. 129
8.5 Policy implications of the comparative static analysis with regard to the gross
profit coefficient of species and the poaching coefficient: two examples………… 133
8.5.1 Debate on conservation and consumptive use of the African elephant……. 133
8.5.2 Conservation and consumptive use of wildlife in Taiwan………………… 137
8.6 Concluding remarks and some implications for conservation policy…………... 140
8.6.1 Some remarks …………………………………………...…………………. 140
8.6.2 Policy implications with regard to the intrinsic growth rate……………….. 142
8.6.3 Policy implications with regard to some other parameters………………… 144
Appendix 8.1………………………………………………………………………... 148
9 Case study: sustainable use and conservation of renewable
resources in Danayiku Nature Park at Shan-Mei, Taiwan
9.1 Background.………………………………………….……………………….….
9.2 Project history and evolution……………………………………………………..
9.3 Resource use……………………………………………………………………...
9.4 Performance of the Danayiku Nature Park……………………………………….
9.5 Ecological, economic and social benefits………………………………………...
9.5.1 Ecological benefits…………………………………………………………
9.5.2 Economic benefits………………………………………………………….
9.5.3 Social benefits……………………………………………………………...
9.6 Negative impacts…………………………………………………………………
9.7 Comparison of different community-based conservation projects in the
151
151
155
156
157
166
166
166
168
168
A-Li-Shan area: an assessment procedure………………………………………. 169
9.7.1 The assessment procedure……………………………………………….… 169
III
9.7.2 A comparison of different community-based conservation projects in the
A-Li-Shan area…………………………………………………………… 171
9.8 Some challenges to DNYKNP at Shan-Mei……………………………………... 176
10 Conclusions, policy implications and limits in applicability of the
theoretic model
10.1 Study conclusions……………………………………………………………….
10.1.1 Conclusions of the theoretic models………………………………..…….
10.1.2 Conclusions of the case studies…………………………………………...
10.2 Policy implications……………………………………………………………...
10.3 Limits in applicability of the theoretic model and recommendations for further
research………………………………………………………………………….
References
178
178
178
181
182
185
187
IV
Chapter 1
Introduction
1.1 The problem concerned and the objective of dissertation
The concept ´biodiversity´ refers to the variety and variability within
living organisms and the ecological complexes in which they occur. This
term encompasses the diversity of life at all levels of organization, ranging
from the gene, organism, and species levels to the community and ecosystem
levels. Biologists usually define biodiversity in terms of gene, species and
ecosystem diversity, and believe generally that it is an extremely critical
factor of ecosystem health and ecological stabilization of the earth (Wilson,
1992).
Due to habitat destruction, overexploitation and poaching, the loss of
biodiversity may constitute currently one of the most serious environmental
problems for human beings, and has attracted widespread public concern.
Based on extrapolations of measured and predicted rates of habitat destruction, and estimates of species richness in various habitats, some evaluations
about the loss of biodiversity suggest that a possible loss of between 15 and
50 percent of the worlds total species will occur over the next century, if
currently measured trends of habitat loss persist (Wilson, 1992). All these
estimates about current and future extinction rates should be interpreted
with very considerable caution, because they involve high degree of uncertainty. Nonetheless, it is hard to doubt that human beings is inducing mass
loss of biodiversity. What such kind of mass extinction means precisely for
the welfare of human beings is, to great extent, still uncertain. But it is
certain that human beings will take much more risk of losing the life-support
system of the earth than before, and human society is suffering considerable loss in economic value, including use and non-use value, of biodiversity
from mass extinction. Conservation initiatives are therefore required to be
precautionary enough to prevent further mass loss of biodiversity.
To protect biodiversity, one of the most critical approaches is establishing
legally or privately designated protected areas. As a form of environmental
regulation, the maintenance of protected areas is even, from the point of view
of many conservation biologists, the only effective instrument for conserving
biodiversity. Up to the year 1996, 30350 protected areas are known to have
been designated worldwide, covering 8.83 per cent of total land area of the
earth (IUCN, 1998). Many assessments indicate that in those parts of the
1
world that have established protected area networks, some degree of success
in preserving certain proportion, if not majority of the biodiversity in a country has been achieved. However, it has become increasingly evident that the
identification, selection, establishment and management of protected areas
are worldwide involved in many problems that need to be solved.
Many problems are threatening the survival of the existing protected area
networks throughout the world. First, many protected areas are too small
or too fragmented to effectively maintain the minimum viable population of
some species in the long run. This problem will become increasingly evident
as habitats outside protected areas become more and more degraded. Secondly, most protected areas have been acquired and created on a haphazard,
but not scientific basis, depending on the availability of fund and land, because that socio-economic and political factors, but not ecological factors,
are often the most important considerations in the establishing and siting
of protected areas. This leads to the unbalanced representative of various
ecosystems in protected areas at all levels, and raises a number of concerns
about the ability of existing protected area networks alone to protect biodiversity adequately (Primack, 1998).
Moreover, besides adequate identification and selection, effective biodiversity conservation requires also adequate management of protected areas,
since many factors with reference to management issues are threatening the
biodiversity and ecological health of protected areas. A list of major threats
faced protected areas include logging, mining, cattle grazing, poaching, cultivation, introduction of exotic species, excessive tourism, pollution, corruption
of park staff and insufficient funding for management. To great extent, most
of these threats have to do with the interest conflicts between protected areas
and local residents living in or near protected areas. Usually, after protected
areas are established, local communities are precluded from exploiting natural resources they need, as they traditionally have practiced. In many cases,
this has resulted in confrontation between local communities and park authorities, illegal exploitation of resources in protected areas, and sometimes
leads to refusal of local residents to establish new protected areas or to expand
existing protected areas. In the long run, protected areas can survive only
when they are supported, or at least tolerated by local communities. And
unless local communities can benefit from protected areas, there will be no
long-term incentive to support the existence of protected areas. This may be
the most serious problem which existing protected area networks are faced.
In addition, as a result of the prevailing insufficient funding for protected
areas and corruption of park staff, some conservationists question also the
2
ability and the willingness of central governments to conduct effectively the
traditional top-down preservation approach (or the so called U.S. national
park model) followed by most of the protected areas around the world. This
query holds especially for the developing countries.
In sum, the present protected area networks need to be adjusted and
expanded on a scientific basis to include a more complete pallet of various
ecosystems and thus to safeguard most of biodiversity in the long run (MacKinnon, 1997). The present inadequate management practice of many existing protected areas needs to be improved (Brandon, 1997). All these aims
requires the support of the interest groups, whatever they are local communities, private organizations or national governments, which bear the cost
derived from the existence of protected areas. It follows that the traditional
´fence and police´ policy of protected areas, which emphasizes the strict
protection of habitats but easily results in the hostility of affected interest
groups toward protected areas, may be insufficient to reach the previous aims
and should be reconsidered. An alternative approach, which enables people
to benefit from the maintenance of protected areas in a sustainable manner
without substantially harming biodiversity, must be found to supplement the
strict preservation approach.
In recent years, many conservationists and scholars have promoted an
incentive-oriented approach, namely, that people are allowed to use wild renewable resources in protected areas or in buffer zones around protected
areas. In some cases, local communities are also authorized to management
natural resources and human activities in protected areas. This alternative
strategy is often called the sustainable use approach. Due to self-interest,
it is expected that more protected areas, whether existing or new, will be
accepted or even designated actively by people under such an approach
(IUCN/UNEP/WWF, 1980, 1991). Numerous initiatives have been implemented around the world, and many relevant researches have been conducted
to investigate the results of the sustainable use approach and their implications for both general conservation policy and specific protected area policy.
However, most of the researches about this topic are based on case studies
and relative few works have been conducted in a theoretical and rigorous
way, especially in the way of economic rationale.
The objective and major task of this dissertation is, based on rigorous
modeling, to investigate the dynamic relation between use of wild renewable
resources, management of protected areas and biodiversity conservation, and
thereby to afford a general framework for answering the question, whether
3
and under which biological and socio-economic conditions the sustainable
use strategy of wild renewable resources in and around protected areas is
an appropriate instrument for biodiversity conservation. We intend that the
general analysis in this dissertation may hold for all cases in both developing
and developed countries. This may help to build a solid scientific basis for
rethinking and modification of the current conservation policy.
1.2 Methods
In this dissertation the dynamic relation between use of wild renewable
resources, management of protected areas and biodiversity conservation and
the relevant policy implications will be investigated by the development of
bioeconomic models and by the application of the optimal control theory. To
do this, a simple bioeconomic model with one state and two control variables
will be firstly constructed on the basis of the traditional bioeconomic model,
namely, the Clark model (Clark, 1973, 1976). Afterward, the simple bioeconomic model will be extended and thereby a more complex model with two
state and two control variables can be developed. Finally, a general model,
which represents a generalized version of the previous extended model, will
be completed. The necessary conditions for optimum will be derived. The
uniqueness and stability properties of the steady state solution of the models,
the relevant phase diagram analysis, and the comparative static analysis will
also be presented.
In the general model, in which the interaction between control and state
variables is so complex that phase diagrams can not be obtained through
the analytic method, a numerical method with computer simulation will be
applied to draw the relevant phase diagrams. With the assistance of numerical method, the final outcomes of the variables of the model under different
scenarios will also be demonstrated.
In addition to the theoretical models, two empirical case studies based on
the experiences from the conservation practice in Taiwan are investigated.
The first one refers to the national park system of Taiwan which represents
a typical top-down preservation approach following the U.S. national park
model. The other one studies the Shan-Mei Community Conservation Project
which emphasizes the sustainable use of renewable resources and represents
an example of the bottom-up approach, or the so called community-based
conservation (CBC) (Western and Wright, 1994). A comparison of these two
case studies affords critical, though incomplete, evidences for assessing the
relative performance of the two fundamental conservation approaches under
specific socio-economic conditions in Taiwan. The ability of the theoretical
4
models to predict possible outcomes of a sustainable use conservation project
will be examined simultaneously through the deliberate investigation into the
Shan-Mei Community Conservation Project and the other community-based
conservation projects in the A-Li-Shan area in Taiwan. The findings of the
case studies could be applied to the countries or regions which have similar
conditions like Taiwan.
1.3 Contents of the dissertation
Following this introductory chapter, chapter 2 explains first some of the
key concepts of biodiversity. It then provides background material on the
issues about the measurement methods and some conceptual indicators for
biodiversity. Finally, it chronicles the present state of biodiversity and its
loss rate according to the results of some scientific assessments.
In chapter 3, the definition and classification of protected areas will be
first introduced. Thereafter, the nature of protected areas as an instrument
of biodiversity conservation will be explored from both biological and economic aspects. Thus we will review a few economic theories regarding nature
and biodiversity conservation and discuss the economic rationale justifying
protected areas as an instrument of biodiversity conservation.
Chapter 4 describes the limited success and present problems of existing protected area networks. While extensive efforts have been successful at
preserving some types of habitat, two major problems of present networks
can be identified, namely, insufficient representative and inadequate management. It follows that the traditional preservation approach of the protected
areas alone may be insufficient to safeguard biodiversity, and conservation
communities may need to consider the sustainable use approach which is assumed to possess the potential to solve these problems simultaneously. The
definition of sustainability regarding the use of wild renewable resources will
be addressed. The debate between the ´preservation approach´ and ´sustainable use approach´ will also be discussed. As an example, the case study
with reference to the national park system in Taiwan will be investigated to
evaluate its performance and to address its problems.
As an introduction into the theoretical modeling, chapter 5 reviews briefly
three important economic models regarding biodiversity loss, including the
Gordon model (Gordon, 1954), the Clark model (1973, 1976) and the Swanson model (1994). This is followed by a discussion about the policy implications of these models. This review affords a direction for the modeling in
the later chapters.
5
In chapter 6, a bioeconomic model with on state variable (resource stock)
and two control variables (harvest rate, management effort) will be first constructed on the basis of Clark´s bioeconomic model and optimal control theory under the assumption that people are allowed to use renewable resources
in or around protected areas. Thereafter, the simple model of chapter 6 will
be extended and thereby a more complex model with two state variables (resource stock, management capital) and two control variables (harvest rate,
investment rate) can be developed in chapter 7 to investigate deliberately
the dynamic development process of resource stock, management capital,
harvest and poaching activity. The necessary conditions for optimum are
derived. The uniqueness and stability properties of the steady state solution
of the models and the comparative static analysis will be demonstrated. The
policy implications of the models will also be addressed.
Then, a general model which represents the generalized version of the
previous extended model will be developed in chapter 8. A computer simulation of the model is conducted through the use of numerical method. The
results of the computer simulation of the general model may help offer more
arguments for judgement of the current conservation policies.
Chapter 9 gives an example of how the community-based conservation,
an important variant of the sustainable use approach, can work well under
specific biological and socio-economic conditions. With the application of
the analysis framework afforded by the theoretical models, we investigate
deliberately the Danayiku Nature Park at Shan-Mei, Taiwan. The analysis
shows us, which conditions contribute to the success of the Danayiku Nature Park at Shan-Mei, and which result in the failure of some other similar
community-based conservation projects in the A-Li-Shan area of Taiwan. It
follows a discussion about the policy implications of the findings from the
case study.
The concluding chapter 10 synthesizes the findings and shortcomings of
our theoretical models, and tries to offer some suggestions for modification
of current conservation policy and some possibilities for future research.
6
Chapter 2
Biodiversity: Concept and its loss rate
2.1 Definition of biodiversity
The term biodiversity, a special terminology representing biological diversity, is used to describe the variety and variability within living organisms
and the ecological complexes in which they occur. It encompasses all species
of plants, animals, microorganisms and the ecosystems and ecological processes of which they parts. The emerging concern about biodiversity reflects
the empirically based recognition of the fundamental interconnections within
and among these various levels of ecological organizations, and the general
belief that biodiversity is an extremely critical factor of ecosystem health and
ecological stabilization of the earth (Wilson, 1992). To quantify the measurement of biodiversity and thereby to facilitate the management of biodiversity,
it is necessary to disentangle some of the separate elements of which biodiversity is composed. Biologists usually define biodiversity in terms of genes,
species and ecosystems (WCMC, 1992).
Genetic diversity refers to the range of variation within and between populations of organisms, or more precisely, it is the sum of genetic information
contained in the genes of individuals of plants, animals and microorganisms. Ultimately, this again resides in variations in the sequence of the four
base-pairs that, as components of nucleic acids, constitute the genetic code
(WCMC, 1992). Wilson (1992) estimated that there are about 1017 different genes distributed across the world´s biota, refraining from entering into
differences within organisms of any given species. Considering the fact that
each species is made up of many organisms, the total number of different
genes will be then far more. For example, the worldwide about 10,000 ant
species have been estimated to comprise 1015 living individuals at each moment of time (Wilson, 1988). However, each of the estimated different genes
does not make an identical contribution to overall genetic diversity because
of different functions of various genes (WCMC, 1992). Moreover, we do not
know even the number of existing species, and respectively the number of
existing individuals within a given species. No practical tools up to now
are available to evaluate these factors. Given these problems, it seems that
genetic diversity is not yet applicable to the evaluation of both biodiversity
loss and conservation programs.
7
Species diversity refers to the number and variety of species. A species
is generally defined as populations within which gene flow occurs under natural conditions, although this definition may not work well for some species
(Brown, 1993). Because the definition about what a species is differs considerably between various groups of organisms, species cannot be recognized and
enumerated by biologists with perfect precision. Moreover, a straightforward
count of the number of species affords only a partial indication of biological
diversity, since species make different contributions to overall diversity, depending on the extent to which they differ from each other. Generally, the
more different a species is from any other species, the greater its contribution
to overall biological diversity. Furthermore, different species play different
ecological roles, and thereby have different effects on community structure
and overall biodiversity. For example, a keystone species whose activities
govern the well-being of many other species apparently makes a greater contribution to the maintenance of biodiversity than a species on which no or
only few species wholly depend (WCMC, 1992).
It is evident that the number of species in different taxonomic groups
at a site, or the so called ´species richness´, is not a perfect indicator for
biodiversity and even for species diversity. However, probably because of the
lack of knowledge and the difficulty with quantifying biodiversity at genes
and ecosystems levels, and the fact that species are the primary focus of evolutionary mechanisms, biodiversity has in practice been presented primarily
in terms of species richness, although we do not know the true number of
species existing on earth as well, even to the nearest order of magnitude.
Roughly 1.4 million species of all kinds of organisms have been formally described. Approximately 57,000 are vertebrates, 250,000 are vascular plants
and bryophytes, and 750,000 are insects. The remainder include a complex
array of invertebrates, fungi, algae, and microorganisms (Wilson, 1992). Wilson (1988) estimated that there are totally 5 to 30 million species on earth.
Some biologists, such as Terry Erwin (1988), have put forward even higher
estimates, up to 50 million. Of the different taxonomic groups, plants and
vertebrates, as well as a few other groups such as butterflies, are relatively
well known. For poorly studied fungi and microorganisms, the estimates of
overall species numbers are probably inadequate (Wilson, 1992). For example, it has been estimated that as many as 1.5 million species of fungi may
actually exist, with 69,000 known species (Hawksworth, 1991). One survey
of the marine ecosystems estimated that the total unexplored new species
could well reach upwards to 10 million (WRI, 1994).
Ecosystem diversity is defined as the variety of habitats, biotic commu8
nities and ecological processes in the ecosystems. While it is possible to
define what is in principle meant by genetic and species diversity, there is
no unique definition and classification of ecosystems at the global level, because ecosystems differ from genes and species in that they explicitly include
abiotic components, being determined by the physical environment, such as
climatic, edaphic and topographic condition. Even though various weightings can be ascribed to these different factors when estimating the diversity
of particular areas, there is no one definitive index for measuring ecosystem diversity. The quantitative measurement of diversity at the ecosystem,
habitat or community levels remains therefore problematic (WCMC, 1992).
2.2 Measurement of biodiversity and some indicators
To make the concept biodiversity operational and reduced to measurable quantities, some measures of biodiversity have been suggested by scientists. While biodiversity is very commonly used as a synonym of species
richness, the number of species alone can be highly misleading as a measure
of biodiversity, since it fails to consider the different facets of biodiversity,
as discussed in the previous section. During recent attempts to discuss biodiversity, scientists have developed some concepts and methods to measure
biodiversity more precisely, in the sense that, instead of a straightforward
counting of numbers of species, these measurements consider explicitly the
extent to which species differ from other species.
Based on the ´genetic distance´ data originating from DNA-DNA hybridization method, scientists try to develop measures that reflect precisely
those characteristics that define the difference between various biological
units, whether they are genetic material, sub-species, communities or ecosystems (Weitzman, 1992; Eiswerth and Haney, 1992; Solow et al., 1993).
The genetic distance data represent differences between the DNA of various species, and hence provide information about differences at the genetic
level. Furthermore, these data afford an indication for higher-taxon diversity
as well, because the genetic distance between species that belong to different,
higher taxa tends to be greater. Weitzman (1992) applied the criterion of
genetic distance between species to develop a measure of biodiversity, which
simultaneously considered the probability of the extinction of species. Solow
et al. (1993) constructed a similar measure. However, in addition to the
genetic distances between species, the factor of species richness is also considered by them. The measure introduced by Eiswerth and Haney (1992)
is also based on genetic distance and species richness, but, unlike Weitzman
and Solow et al. do, it does not take the factor of the probability of the
extinction of species into account.
9
These methods discussed here are scientifically consistent measures of
biodiversity, and they could in principle be used to assess conservation priorities at any level. In practice, however, these measures require substantial
information from the biological sciences in large-scale problems. This limit
their usefulness to a great extent. At most, they may be useful for extremely
small-scale problems under given data base. In addition, while these measures emphasize the genetic properties and the endangered status of species,
a critical facet of species, namely their ecological role in supporting the functioning and resilience of ecosystems is neglected.
At the species level, two measures, namely species richness and species
diversity, are usually applied in practice. Species richness, an important
dimension of biodiversity, refers to the number of species existing in an area.
Species diversity indices are derived by weighting species by some measure of
their importance, such their abundance, productivity, or size (Orians, 1994).
At ecosystem level, as discussed in the previous section, ecosystem diversity is very difficult to define and measure. Given the complexities of the
numerous components of ecosystem diversity, some measures of ecosystem
level diversity are introduced. Based on significant differences in flora, fauna,
vegetation structure, and physical attributes such as climate, Udvardy (1975)
developed a system of biogeographic analysis for terrestrial ecosystems. He
divided the world into eight terrestrial biogeographical regions, and these
eight regions are further subdivided into 193 provinces which may be very
useful for assessing the effectiveness of the protected area network in protecting various ecosystems. Similarly, to establish priorities to conserve the most
important areas, two policy oriented methods, the ´ecological hotspots´ (Myers, 1988) and the ´mega-diversity countries´ (Mittermeier & Werner, 1990)
are developed, by the use of lists of plant species or other taxa to identify
biologically rich biogeographic areas or countries.
Because measuring and monitoring all facets of biodiversity are very difficult, conservation biologists have proposed some indicators at species level
as a shortcut whereby attention is focused on one or a few species to monitor
and solve biodiversity conservation problems. These indicators which can be
easily monitored includes the so called umbrella species, flagship species, keystone species and biodiversity indicator (Simberloff, 1998; Caro & O´Doherty,
1999). The concept of umbrella species, defined as a species that requires a
large range of habitat so that protecting it will automatically protect many
other species, have been applied to depict the type of habitat or size of
the area for protection, though significant ignorance about how many other
10
species could be saved under the protection umbrella of the target species
still exists. Tiger is a well-known example for umbrella species which played
a critical role in the designation of protected areas. The flagship species, usually a charismatic large vertebrate such as the giant panda, have been used
to attract public concern and thereby promote conservation campaign. The
keystone species, such as elephant, plays a pivotal role in the ecosystem and
its activities have great impacts on the well-being of many other species, so
that protecting keystone species contributes also greatly to the conservation
of many other species and the maintenance of the health of the ecosystem.
It is notable that an umbrella species is not necessarily an adequate flagship
or keystone species, and vice versa. Therefore, management regimes of two
indicator species can conflict. Furthermore, intensive management of an indicator species does not necessarily imply successful conservation of the rest
of the communities to be indicated or protected, since they do not receive
similar treatment (Simberloff, 1998). In particular, as Caro and O´Doherty
(1999) suggested, these indicators are not necessarily adequate biodiversity
indicator species. In addition to the concept of umbrella species, flagship
species and keystone species, biologists usually apply the biodiversity indicator, namely the number of species in a well-known taxonomic group as an
indicator for the number of the species in poorly-known taxonomic groups,
for assessing the overall status of biodiversity in a given area. Once the biodiversity indicator species have been identified, their absence can be used as
a sensitive indicator for the absence of other species in the same area. By the
application of this concept, areas with high biodiversity could be identified
more easily and then designated for protection.
The previous discussion apparently suggests that no single measure or indicator can capture all facets of the complex concept biodiversity. It follows
that, rather than attempting to develop an universal indicator for biodiversity, a system of multiple indicators which assesses different facets of biodiversity may be a more practical solution to the measurement problem of
biodiversity under given ignorance about relevant scientific knowledge. Reid
et al. (1993) developed a set of 22 indicators for biodiversity which are used
to assess the diversity of wild species and genetic diversity, the diversity at
the community/habitat level, and the diversity of the domesticated species.
Such an indicator system may help, though not perfect precisely, capture the
full view of biodiversity status.
2.3 The loss rate of biodiversity
In this section, we briefly review the current assessments with reference
to the loss of global biodiversity which has resulted in urgent concern about
11
biodiversity conservation. Probably because of the lack of knowledge and
the difficulty with quantifying biodiversity at genes and ecosystem levels,
the problem of biodiversity loss has in practice been presented primarily in
terms of species loss, although we do not know the true number of species
existing on earth as well, even to the nearest order of magnitude. And it
is evident that loss of other dimensions of biodiversity, though difficult to
quantify in the manner of a universal indicator, may be greater still (Ehrlich
and Daily, 1993).
Current estimates with reference to the loss of biodiversity taking forms of
extinction of species are mainly based on the ecological relationship between
area and number of species. The fundamental relationship between the size
of an area and the number of species it supports, is an empirical generalization of the theory of island biogeography first developed by MacArthur and
Wilson (MacArthur and Wilson, 1967). Originated from the observations
using island data, the theory states that the size of an area and of its species
number tend to have a predictable relationship, depending on various types
of ecosystems. It implies that fewer species are able to exist in a number
of small habitat fragments, like the islands in a sea of human-dominated
landscapes, than in the original unfragmented habitat, and this can result
in the extinction of species. This relationship is commonly presented in the
following functional form:
S = cAz
where S denotes the equilibrium number of species that should persist in
a given habitat area, A represents the size of the area, and c and z are
constants whose values depend on habitat type. Both c and z are positive
constants, and most of the estimates with reference to the theory of island
biogeography gave a value of z between 0.20 and 0.35 in many groups of
organisms (Meffe and Carroll, 1994). This suggests that we can predict the
reduction in numbers of species as the area of habitat decreases, if some
estimates about c and z have been made.
The deforestation of tropical forests is commonly considered as a major
cause of global biodiversity loss, since tropical forests support the majority
of terrestrial species. As a result, based on extrapolations of measured and
predicted rates of habitat destruction in tropical forests, and estimates of
species richness in various habitats, some estimates of current rates of global
species extinction have been undertaken by applying the theory of island
biogeography. These estimates suggest that a possible loss of between 15
and 50 percent of the world´s total species will occur over the 21st century,
12
if currently measured trends of habitat loss persist (see Table 2.1). Wilson
(1988) argues that, due to the current destruction of tropical rain forests and
setting aside from the moment extinction due to the destruction of other
habitats, both the per-species rate and absolute loss in number of species
would be about 1, 000 to 10,000 times the historic rate of extinction. Current
extinction rates thus appear to be far higher than the so called ´natural´ or
´background´ rates.
Table 2.1. Estimates of the current rates of
Estimate of species loss
Basis
15-20% by year 2000
Forest area loss
50% by year 2000
Forest area loss
33% in 21st century
Forest area loss
25% in 21st century
Forest area loss
5-15% by year 2020
Forest area loss
0.2-0.3% per year
Forest area loss
2-8% by year 2015
Forest area loss
Source: WCMC (1992) and references.
species extinction
Source
Lovejoy (1980)
Ehrlich & Ehrlich (1981)
Simberloff (1986)
Raven (1988)
Reid & Miller (1989)
Ehrlich & Wilson (1991)
Reid (1992)
Even on the best available present knowledge, these estimates involve high
degree of uncertainty, because of the ignorance of the total number of species
and their distributions, the patterns of habitat loss, and the effects of deforestation on species (Myers, 1994). Moreover, a straightforward count of the
number of extinct species only provides a partial indication of biodiversity
loss, since species that differ widely from each other in some respect by definition contribute more to overall diversity than those which are very similar,
and the different ecological importance of various species could have a direct
effect on community structure, and thus on overall biodiversity (WCMC,
1992). Besides, some scientists assert that estimations of species loss based
on extrapolations of deforestation and on the theory of island biogeography
are misleading, because these estimations fail to take the possible significant
amount of biodiversity after deforestation into account (Lugo et al., 1993).
Assumptions and estimations with reference to the rate and extent of
habitat loss have also raised the uncertainty when estimating the loss rate
of species. As Table 1 shows, relatively new estimates suggested somewhat
more conservative calculations of the rate of species loss. This may partly reflect the fact that, after a period of rapid deforestation in tropical forests, the
pace of forest clearance have slowed down in the early 1990s. However, the
pace of forest destruction in amazonian forests has again accelerated significantly in recent years (Laurance, et al., 2000). Therefore, these conservative
13
estimations might be modified in the future. In any case, all estimates about
current and future extinction rates should be interpreted with very considerable caution under given high degree of uncertainty. Nonetheless, these
estimates appear to provide a useful approximation of the degree of threat
to the global biodiversity during this period. It is hard to doubt that human
being is inducing mass loss of biodiversity.
14
Chapter 3
The nature of protected areas as an
instrument of biodiversity conservation
Protected areas are without doubt the focus of the current conservation
activities. As one of the most important instruments of biodiversity conservation, it deserves our research into the question why the existence of
protected areas can be justified in our crowded planet. In this chapter, we
will explore the nature of protected areas from both biological and economic
aspects. Some important economic theories regarding nature and biodiversity preservation will be briefly reviewed to explore the economic rationale
justifying protected areas as an instrument of biodiversity conservation.
3.1 Definition and classification of protected areas: The IUCN
system
One of the most critical approaches in protecting biological communities is establishing legally or privately designated protected areas. Based on
the agreement at the Fourth World Congress on National Parks and Protected Areas, The World Conservation Union (IUCN) (IUCN, 1994) defined
a protected area as ´An area of land and/or sea especially dedicated to the
protection and maintenance of biological diversity, and of natural and associated cultural resources, and managed through legal or other effective means´.
Once land comes under protection, decisions must be made regarding how
much human disturbance will be allowed. In order to be able to categorize
protected areas, IUCN (1994) has developed a system of classification for
protected areas that ranges from minimal to intensive allowed use of the
habitat by humans. The following categories are arranged in ascending order
of human use permitted in the area:
Ia. Strict nature reserve: protected areas managed primarily for preserving representative examples of biological diversity for scientific study,
education, environmental monitoring, and maintenance of genetic variation.
Ib. Wilderness area: large areas of unmodified or slightly modified wilderness managed primarily for recreation, for subsistence economic activities, and for protection of natural process.
15
II. National park: large protected areas with outstanding scenic beauty
and ecological importance, designed primarily to preserve unique natural beauty and resources, and maintained for scientific, educational
and recreational purpose. Extractive use of resources in national parks
is in principle not allowed.
III. Natural monument: smaller protected areas designed primarily for
preservation of unique natural areas with specific natural or cultural
significance.
IV. Habitat/species management area: strict nature reserves which require
active management intervention, managed primarily for maintenance of
the characteristics of the community. Controlled harvesting is allowed
in some cases.
V. Protected landscape/seascape: protected areas managed mainly for the
maintenance of areas in which people and the environment interact in
a harmonious way. They include natural areas that have undergone
considerable human transformation. Nondestructive use of resources is
permitted.
VI. Managed resource protected areas: protected areas managed primarily
for sustainable use of natural resources, in a manner that ensures the
long-term protection and maintenance of biodiversity.
The categorization is based on the primary management objective of protected areas. Of these categories, the first six can be considered as true
protected areas, with the objective managed mainly for protection of biodiversity. The main management objective of areas in category VI is not
protection of biodiversity, but they can play an important role in conserving
biodiversity, since they are usually much larger than strict protected areas,
since they still contain many or even most of their original biological diversity, and since strictly protected areas are often surrounded by managed
resource protected areas (primack, 1998). In addition, protected areas are
usually managed for multiple objectives. According to the IUCN classification system and the priority assigned to relevant management objectives, a
categorization with more detailed management objectives is made by Phillips
and Harrison (1999), as Table 3.1 demonstrates.
16
Table 3.1. Potential primary management objectives of
type of protected areas
Objectives
Ia
Ib
II
III IV
Scientific research
1
3
2
2
2
Wilderness protection
2
1
2
3
3
Preserve species and
genetic diversity
1
2
1
1
1
Maintain environmental
services
2
1
1
NA 1
Protection of natural/
cultural features
NA NA 2
1
3
Tourism and recreation NA 2
1
1
3
Education
NA NA 2
2
2
Sustainable use of
natural ecosystems
NA 3
3
NA 2
Maintain cultural/
traditional attributes
NA NA NA NA NA
1 = Primary objective
2 = Secondary objective
3 = Acceptable objective
NA = Objective not applicable
Source: Phillips and Harrison (1999), p. 15.
various
V
VI
2
3
NA 2
2
1
2
1
1
1
2
3
3
3
2
1
1
2
In the field of conservation biology, two approaches have been widely
used to conserve biodiversity. The maintenance of protected areas is a part
of so called in situ or on-site preservation, which leaves biological communities and populations in the wild, whereas the ex situ or off-site approach
involves permanent collections of species in zoos, botanical gardens and the
preservation of seeds and other genetic material in a controlled environment
such as germplasm banks (Primack, 1998). It is generally agreed that in situ
approach is the most effective way, even the single way in the long run to preserve biodiversity (Primack, 1998), because we do not have enough resources
or knowledge to maintain the majority of the world´s species in captivity. As
Woodruff (1989) pointed out, in situ preservation of biodiversity is far more
cost-effective than ex situ preservation, although the latter has an important
role to play when in situ approach fail. Moreover, only in natural communities are populations large enough to conserve relatively complete heritable
base, and only within natural communities are species able to function adequately as a part of the complex ecosystems and continue the process of
evolutionary adaptation to the changing environment.
17
Thus, for the sake of conserving biodiversity, the in situ approach, namely
leaving biological communities in the wild, is extremely necessary from the
biological point of view. However, of the de facto existing undisturbed or
relatively undisturbed wilderness, why should we give some certain areas
special protected status, i.e., declaring them as strictly protected areas and
leaving them out of almost all development considerations? What is the
economic rationale justifying strictly protected areas as an instrument of
biodiversity conservation? To answer these questions, we may turn to several
economic theories dealing with nature and biodiversity preservation.
3.2 Economic theories justifying the existence of protected
areas: theories with reference to nonrivalry and
nonexcludability
From the point of view of economics there are primarily four special
characteristics associated with wilderness and biodiversity, i.e., nonrivalry,
nonexcludability, uncertainty and irreversibility. With reference to nonrivalry and nonexcludability, Sherman (1989) showed that the main reason
for the degradation of natural areas is that there is an underlying disparity
between the private and social costs and benefits of wilderness use and conservation. Much of the benefits associated with wilderness exhibit nonrivalry
and nonexcludability, such as the existence value derived from simply knowing that a certain wilderness area or a certain species exists, even though
people will never truly see or use it (Krutilla, 1967). The problem with nonrival goods is that the market cannot set an efficient price for them. When
goods are nonexcludable, there are problems of externalities. This mix of
public goods and externalities problems results in significant market failures,
and these market failures make it much more difficult for people to appropriate the benefits of protecting wilderness. As a result of market failures,
there is a bias toward conversion and development of wilderness. The effect
of this bias is that a smaller amount of areas is protected or left in natural
state than would be the case if there was a full accounting of all the social
benefits and costs associated with each alternative land use.
According to the theory of island biogeography, this result implies generally that a smaller amount of biodiversity is left in natural state than would
be the case if market failures would not exist. To bridge this gap between
suboptimal and optimal provision of biodiversity, additional provision of protected areas will be necessary. This requires usually government intervention,
although private non-government organizations play as well an important role
18
in this task.1 In any case, the establishment of strictly protected areas is in
this context used as an instrument to counterbalance the effects of market
failures and helps increase the amount of remaining biodiversity to an amount
much closer to the socially optimal allocation.
3.3 Economic theories with reference to uncertainty and
irreversibility: option value, quasi-option value and
the Safe Minimum Standard
The effect of uncertain and irreversible decisions compared to certain and
reversible ones in the environmental field have long been addressed in the
economic literature. Dealing with these two properties, several approaches
have been suggested. One possible approach of handling uncertainty and
irreversibility is to explicitly introduce the concept of ´option value´, which
is defined as the value, in addition to expected consumer´s surplus from
actually using a good, that arises from retaining an option to a good for
which demand and/or supply is uncertain (Weisbrod, 1964; Bishop, 1982).
For example, when addressing the issue about whether converting a piece
of rain forests into a farm, the option value of this piece of rain forests will
be that consumers are willing to pay more than the expected consumer´s
surplus derived from converting rain forests into a farm, so that they can
ensure that they can make use of this piece of rain forests later on, given
that their tastes in the future are uncertain. In this context, option value
could be interpreted as a risk aversion premium (Bishop, 1982).
Dealing with the demand side uncertainty, i.e., consumers are uncertain
about their future demand, Schmalensee (1972) and Henry (1974) argued
that the net option value may be positive, negative or zero for a risk averse
individual, because of the fact that preservation, as well as development, can
bring risks. While the demand-side option value is indeterminate, Bishop
(1982) maintained that supply-side option value, i.e., option value associated
with uncertainty about whether a good will be available when consumers
want to use it, would unambiguously be positive for risk-averse individuals,
if both the utility function and income are certain. However, Freeman (1985)
and Johansson (1988) asserted that, under certain assumptions, the sign of
the supply-side option value is indeterminate. Finally, in any case, option
value depends on the attitude of people toward risk. In sum, it seems that
the existence of option value is not a definitive argument for the preservation
1
For example, through privately funded efforts to conserve biodiversity in the USA
and elsewhere, The Nature Conservancy has established many protected areas and made
a substantial contribution to biodiversity conservation (Grove, 1988; Primack, 1998).
19
of biodiversity and wilderness, especially in many developing countries, in
which the future income level is highly uncertain.
In the context of an irreversible development decision where information
about the future consequences of development would be available with time,
Arrow and Fisher (1974) developed the concept of the ´quasi-option value´.
When an irreversible development is undertaken, future alternatives become
limited. As a result, there is a value in delaying a decision that involves
irreversible effects whose values are not known. This value can be defined
as the gain from being able to learn about future benefits that would be
precluded by development by delaying an irreversible decision (Fisher and
Krutilla, 1985), or as a conditional value of information, conditional on a
particular choice of first-period development, i.e., that the development is
postponed initially (Fisher and Hanemann, 1987). As Fisher and Hanemann
showed, so long as there is a non-zero probability of such information becoming available, a positive quasi-option value always exists for the alternative
of avoiding an irreversible development.
In the context of nature and biodiversity preservation, quasi-option value
refers very specifically to the value that biological resources and wilderness have as resources of information that is not yet discovered. Just like
usual preservation benefits, quasi-option value is, at least partly, nonrival
and nonexcludable. The owner of a piece of land will hardly take quasioption value into consideration when developing the land, and this fact leads
to market failures and underprovision of wilderness and biodiversity. In this
connection, the maintenance of strictly protected areas could be again interpreted as an instrument for correcting market failure bias.
An alternative approach suggested to dealing with uncertainty and irreversibility is known as the Safe Minimum Standard (SMS) (Ciriacy-Wantrup,
1952; Bishop, 1978). In essence, the SMS approach applies a modified version of the ´minimax´ criterion in game theory. In its strict sense, applying
the minimax criterion involves choosing the alternative that minimizes the
maximum possible losses that will arise when making the wrong decision.
The SMS approach which uses the modified version of the minimax criterion
also suggests minimizing the maximum possible loss, but only when the costs
of doing so are not unacceptably high (Bishop, 1978).
The SMS approach is especially applicable for the cases in which resources
can be irreversibly depleted. Many biological resources, such as plant and
animal species, have a ´minimal viable population´. These resources are
renewable, if their population levels are greater than those of the minimal
20
viable population. However, they are subject to irreversible depletion, if their
populations go below the minimal viable populations. The species in question
may have little known value today but its future value may be significant, e.g.,
as a gene and food source, or functioning as an important keystone species
which has a great influence on many other species within the same ecosystem.
The irreversible depletion of these resources may cause enormous future social
or economic losses. Using the SMS approach implies that such alternative
involving irreversible potential loss should be avoided unless the costs of
doing so are unacceptably large. Therefore, in the context of biodiversity
conservation, the maintenance of strictly protected areas can be viewed as
setting physical safe minimum standards to safeguard biodiversity, at least
partly, and thereby to prevent from potential enormous loss in the future.
3.4 The Perrings and Pearce Model
In a paper addressing conservation of biodiversity and the relevant policy
instrument issues, Perrings and Pearce (1994) provided a similar, but more
detailed biological and economic rationale for the use of physical standards.
They asserted that the problem of biodiversity loss is especially associated
with the ecological threshold effects. The erosion of biodiversity is a process
with the special characteristic of irreversibility. If the loss of biodiversity
goes on, certain ecological threshold will sooner or later be reached, at which
ecosystems are on the edge of losing their ability of self-organization and
their ability as the life support system of the earth. In this context the
ecological threshold can be defined as the critical values for populations of
organisms or biogeochemical cycles. Once, for example, some populations of
organisms are already on the edge of threshold, a marginal depletion of these
populations will eventually result in the collapse of ecosystems and enormous
costs for human beings. As Perrings and Pearce showed, the existence of
the ecological threshold effects has important implication for the choice of
appropriate instruments dealing with biodiversity conservation.
Let us demonstrate this at hand of a simple model. For convenience
of comparison, we consider first the normal case with no threshold effects.
Defining w as a strictly positive vector of market input costs, r a non-negative
vector of biological resources, and q output. The total economic cost of
exploiting biological resources includes usually the components of private
cost and external cost:
T EC = C + E
where C = C(w, q) denotes private cost function which is assumed to be
continuous, differentiable and increasing in both w and q, and E = E(r(q))
21
represents external cost function which is assumed to be increasing in q. The
external cost function can be neither differentiable nor continuous, as we will
address later. Again, defining R = R(p, q) as the private revenue function
under given price p of the output q and Πp as the marginal net private profit
(private revenue minus private cost), then it is evident that the necessary
condition for privately optimal output level is
∂R
∂C
=
∂q
∂q
and the necessary condition for socially optimal output level is
∂Πp
∂E ∂r
=
.
∂q
∂r ∂q
These two conditions are satisfied at the output levels qp∗ and qs∗ respectively, as figure 3.1 shows. In this case, the privately optimal output level
will diverge from the socially optimal output level, suppose that there is no
environmental regulation.
Now let us consider the case with ecological threshold effects to show how
the introduction of threshold effects change the nature of the problem. As
defined previously, the existence of ecological threshold effects means that,
once ecosystems or populations are depleted beyond these thresholds, the life
support system will eventually collapse and result in enormous costs. This
accordingly implies, as Perrings and Pearce argued, the discontinuity or at
least the non-smoothness of the external cost function. Such a case is demonstrated in figure 3.2 in which the external cost function E is discontinuous at
certain output level, and is thought to increase dramatically at that point to
certain much higher level, and then go on to increase with the output level. It
is obvious that, at the discontinuity point of the external cost function where
p
∂E
, the first order condition for social optimum will not hold. In this
< ∂Π
∂q
∂q
case, the socially optimal output level will be qs∗ in figure 3.2, at which the
threshold and the corresponding discontinuity of the external cost function
happen.
22
C(w,q)
R(p,q)
Costs
Revenue
E(r(q))
0
q
E
Costs
Revenue
R-C
q
0
q s*
q
*
p
Figure 3.1. Private and social optima: continuous external cost function.
Source: Perrings and Pearce (1994), p.16.
23
C(w,q)
R(p,q)
Costs
Revenue
E(r(q))
0
q
E
Costs
Revenue
R-C
0
q
q s*
q *p
Figure 3.2. Private and social optima: discontinuous external cost function.
Source: Perrings and Pearce (1994), p.17.
From the policy perspective, it is necessary to consider the possible policy instruments which can protect society against the costs of exceeding the
threshold. One approach widely suggested is setting physical restrictions
around thresholds, just like the rationale of the safe minimum standards.
Nonetheless, instead of the quantitative restriction the safe minimum standards approach implies, Perrings and Pearce suggested an alternative instrument based on price incentive - special environmental levies or fines which
are extremely severe. Their idea is as follows. If the social costs derived from
threshold effects are higher than the maximum private net benefit derived
24
from exceeding the thresholds, an arbitrary penalty between the two can be
introduced and enforced to guarantee that the output level will not exceed
qs∗ , the socially optimal output level, as figure 3.3 shows. The line 0abc in
figure 3.3, which denotes the private cost function derived from charges and
penalties, includes two components: environmental charges for the expected
social costs derived from the use of biological resources below the threshold
level of economic activity (0a), and a severe penalty (ab) when exceeding the
threshold level of economic activity. The penalty must be high enough, at
least equal to the maximum private net benefit of using biological resources,
so that the private output level will not exceed the threshold value as a result
of the self-interest motive. In addition, considering the factor of uncertainty
in measuring private benefit, a penalty much higher than the maximum private net benefit can offer an additional margin for safety.
Costs
External costs
Revenue
Threshold
costs
Pvt cost of
Penalty
c
b
charges plus
penalties
Maximum
private net
R-C
benefit
a
0
q s*
q *p
q
‘Standard’
Figure 3.3. Penalty function: thresholds known with certainty. Source:
Perrings and Pearce (1994), p.24.
The case considered above implicitly assumes that the ecological threshold
level of output is known with certainty. Under the premise that the penalty
25
can be effectively enforced, it can be guaranteed that the output level will not
exceed that of the social optimum. However, the uncertainty as a result of the
ignorance about ecological threshold values of biological resources usually still
exists. In this case, even a severe environmental penalty can not guarantee
that output level will not exceed the threshold level. What we can do is
setting standards, on which the penalty function is based, conservatively
relative to the thresholds which are assumed to be or are designed to protect,
as figure 3.4 demonstrates in which the penalty function 0abc lies on the left
side of the discontinuity at qs∗ . In this way the risk of exceeding ecological
thresholds can be reduced.
Costs
External costs
Revenue
Threshold
costs
Pvt cost of
Penalty
c
b
charges plus
penalties
Maximum
private net
R-C
benefit
a
0
‘Standard’
q s*
q *p
q
Figure 3.4. Penalty function: thresholds not known with certainty. Source:
Perrings and Pearce (1994), p.25.
In any case, as Perrings and Pearce concluded, a judgement with reference
to the socially acceptable margin of safety in the exploitation of biological
resources is required, whether the policy instrument is based on the physical
restrictions (safe minimum standards) or on the economic incentive (environmental charges and penalties) associated with the enforcement of those
26
physical restrictions. In the context of biodiversity conservation, their argument justified the appeal for the preservation of at least part of biodiversity,
and accordingly justified the existence of strictly protected areas as physical
standard for safeguarding biodiversity.
3.5 Concluding remarks
As a critical instrument of biodiversity conservation, the existence of protected areas can be justified from both the biological and economic perspectives. From the point of view of biology, the designation of protected areas
is the most effective way, even the single way in the long run to preserve
large area of wilderness and biodiversity. It is also more cost-effective than
the ex situ preservation approach. From the point of view of economics,
the establishment and maintenance of protected areas can be viewed as an
instrument for counterbalancing the effects of market failures and helping
increase the amount of protected biodiversity to an amount much closer to
the socially optimal biodiversity. In addition, it can be regarded as setting
physical standards to safeguard biodiversity and thereby to avoid potential
enormous costs in the future resulted from biodiversity loss.
27
Chapter 4
State of protected areas and the
debate on sustainable use of renewable
resources in and around protected
areas as an instrument of biodiversity
conservation
In this chapter we will first briefly review the effectiveness and problems
of the present protected area network from a global perspective. The severe
problems faced by protected area network aroused the widespread debate
within conservation communities on the questions, whether the traditional
preservation policy of the protected areas alone is sufficient to safeguard biodiversity, and whether the sustainable use approach could protect biodiversity more effectively than the preservation approach. To explore this crucial
topic which has to certain extent dominated the direction of conservation
policy at regional, national and international levels, the debate between the
preservation approach and the sustainable use approach will be investigated
in details. The definition of sustainability regarding the use of wild renewable resources will accordingly be addressed. Finally, as an example of the
previous discussion, a case study with reference to the national park system
in Taiwan will be investigated to evaluate its performance and to address its
problems.
4.1 Effectiveness of protected areas: a global perspective
The extent of the global protected area network has increased steadily
in the last several decades. Up to the year 1996, 30,350 protected areas are
known to have been designated worldwide, covering a total land area of 13.23
millions square kilometers. This amount represents 8.83% of the total land
surface of the world (IUCN, 1998).2 Of these protected areas, 90% are strictly
protected (IUCN categories Ia-V), covering 73% of the protected surface, and
10% are partly strictly protected (IUCN category VI), representing 27% of
the protected surface (Green and Paine, 1999). However, substantial variations of the protected percentage can be observed between countries, biomes,
biogeographical realms and biogeographical provinces (WCMC, 1992). Moreover, landscape usually encompasses large area of a fairly uniform habitat
2
It is worth noting that this percentage includes many marine protected areas or protected areas which encompass a marine component.
28
with only a few small areas of rare habitat types. It seems that, for the sake
of conserving biodiversity, including representatives of all the habitats in a
system of protected areas will be more important than only preserving large
areas of the common habitat type (Primack, 1998). These facts complicate
substantially the assessment about the effectiveness of protected area networks to conserving biodiversity. The conclusions that can be drawn from
the high level of aggregation at such as the biogeographical realms or biome
level is limited. A more accurate picture of biodiversity conservation can
only be obtained from an analysis at lower level or even from case studies.
Some investigations have been carried out to assess how effective the
limited protected area networks are in maintaining biodiversity. At national
level, for example, a survey of 25 mostly temperate countries indicated great
variation in the extent to which listed threatened plant species discovered
in protected areas, i.e., from 35-40% in Spain to 100% in Czechoslovakia
(WCMC, 1992). Siegfried (1989) found that, with only 6% of the land area
of the region, the protected areas in southern Africa protect respectively
more than 90% of amphibian, reptilian, avian and mammalian species native
to the region. Sayer and Stuart (1988) found that, in 11 of the 12 large
tropical African countries, the majority (at least 75%, and generally well
over 80%) of the native bird species have populations inside protected areas.
In Taiwan, 55% of all vascular plant species (2200 of 4000) and 50% of the
bird species (200 of 400) have been found to exist within the boundaries
of Kenting National Park, which covers only 0.5% of the total land area of
Taiwan (COA and DNP, 1992). These figures indicate that in those parts
of the world that have established protected area networks, some degree of
success in preserving many, if not most of the species in a country has been
achieved.3
On the other hand, while the number of species existing in protected
areas is important as an indicator of the ability of protected areas for biodiversity conservation, the occurrence of species in protected areas is no guarantee of their long-term security. Many protected areas are too small or
too fragmented to effectively maintain the minimum viable population of
some species in the long run (Primack, 1998). These problems will become
increasingly evident as habitats outside protected areas become more and
more degraded. As the theory of island biogeography predicts, under plausible assumptions, about half of all species will in the long run inevitably go
extinct, if 90% of all habitat area are destroyed, leaving protected areas as
3
For more examples see the summary of the relevant papers made by Primack (1998),
pp. 404-406.
29
isolated ´biological islands´of natural or semi-natural habitat (Wilson, 1992).
Moreover, as Daily and Ehrlich (1995) argued, only considering the aspect of
species extinction may greatly underestimate the rate of loss of organic diversity as a whole. They asserted that, to maintain the ecosystem services on
which human economic system depends, the conservation goal should be the
preservation of a minimum ratio of natural to human-dominated habitat in
all habitat types. It supports the point of view that protected area networks
should try to ensure the representatives of as many types of ecosystems as
possible.
From the perspective of representative, at the biome level, there are substantial differences between the extent to which they are protected. Of the
14 major biomes, the subtropical/temperate rainforest/woodlands are relatively better protected, with 10.29% under protection. In opposition to this,
only 0.98% of the temperate grasslands and 1.12% of the lake systems are
under protection. Compared with the 10% target planned for the protection
of biomes at the Fourth World Parks Congress, almost all biomes remain
underpresented in the present protected area networks (Green and Paine,
1999).4 Dinerstein and Wikramanayake (1996) found that, lowland moist
tropical forests are greatly underpresented in terms of coverage by large protected areas in Indo-Pacific region, while most of the large protected forests
are lower-mountain or mountain forests. Globally, the marine and coastal
regions are poorly represented in the protected area networks and emerge as
an obvious gap (McNeely et al., 1990).
In another way, Udvardy (1975) divided the world into eight terrestrial
biogeographic regions. These eight regions are further subdivided into 193
provinces. A survey suggested that 50 of the regions have a protected coverage more than 10%, 34 have a protected coverage between 5% and 10%,
and the other possess less than 5% of the land under protection or have no
protected areas (WCMC, 1992). Studies of protected areas coverage at regional and national levels provide a similar result. The case studies of Sattler
(1992) about Queensland, Australia, of Turner et al. (1992) about Canada,
of Barnard et al. (1998) about Namibia, and of Powell et al. (2000) about
Costa Rica showed that all these countries have not yet achieved an ecologically representative network of protected areas. To reach this goal, for
example, Queensland require an addition of 3.6 million hectares of national
park. Klubnikin (cited in Scott et al., 1987) concluded that in California,
95% of the alpine habitats are in reserves, but less than 1% of biologically
4
A detailed list of these biome and their protected percentages see Green and Paine
(1999).
30
rich riverbank is protected. It is here worth noting that, the countries addressed above have established well-managed protected area networks, and
have generally good reputation within conservation communities. For most
of the countries, it is plausible that the situation could not be better, though
rigorous assessments are still required. At least from the point of pure biological view, all above studies indicated that, while extensive efforts have
been successful at preserving some types of habitat, many regions are still
virtually ignored.
4.2 Current problems of protected areas
Protected areas are the cornerstone of the modern conservation movement. While some degree of success in preserving certain proportion, if not
majority of the biodiversity has been achieved in those parts of the world
that have established protected area networks, it has become increasingly
evident that the identification, selection, establishment and management of
protected areas are worldwide involved in many problems that need to be
solved.
From the perspective of an ideal model of protected area network, strictly
protected areas should be large enough to maintain the minimum viable
population of some critical species, selected and created on a rigorously scientific foundation to maintain the representative of various habitat types,
connected via a network of corridors to avoid isolation of populations, and
surrounded by buffer zones to mitigate the impacts of human activities from
the world outside protected areas. Compared with these ideal standards,
some problems of the present protected area system emerge. First, as discussed previously, many protected areas are too small or too fragmented to
effectively maintain the minimum viable population of some species in the
long run. Green and Paine (1999) indicated that 59% of the total protected
areas throughout the world are smaller than 1000 hectares, and only 6% are
larger than 1000 km2 in size.
Next, the present protected area system suffers from the problem of unbalanced representative of various ecosystems in protected areas. From this
point of view, the protected area networks in almost all parts of the world is
apparently inadequate for the protection of a representative proportion of the
ecosystem. This bias can trace back to several factors. Historically, the initial purpose of many protected areas was to protect spectacular scenery and
provide recreational resources (Dixon and Sherman, 1991). Up to now, many
national parks tend to be selected for their aesthetic value and recreational
31
opportunity rather than for their biological richness. Ecological considerations are dominated by scenic and recreational values, probably because individuals can benefit more directly from these values (Krautkraemer, 1995).
Moreover, protected areas are established typically in those regions which
are unfavorable for farming, settlement and other human activities. Socioeconomic and political factors, not ecological factors, are often the most important considerations in the selection of the sites of protected areas (WCMC,
1992; Sattler, 1992). Thus, most protected areas have been acquired and created on a haphazard, but not scientific basis, depending on the availability of
fund and land. In addition, as Shafer (1999) indicated in the example of the
US national parks, the absence of biological considerations in the past may
partly reflect the fact of scientific ignorance. Only until 1960s, the concept
of representative on a systematic basis began to evolve (Eidsvik, 1992). This
leads to the unbalanced representative of various ecosystems in protected areas at all levels, and raises a number of concerns about the ability of existing
protected area networks alone to protect biodiversity adequately (Primack,
1998).
Moreover, besides adequate identification and selection, effective biodiversity conservation requires as well adequate management of protected areas,
since many factors with reference to management issues are threatening the
biodiversity and ecological health of protected areas. Almost all these factors
are, directly or indirectly, after all human factors, even though in some cases
in which seemingly natural factors have caused the problems. For example,
the overpopulation of deer and thereby caused overgrazing in many protected
areas is caused in fact by human beings, because major predators in those areas were eliminated (Primack, 1998). A list of major threats faced protected
areas include logging, mining, cattle grazing, poaching of wildlife, cultivation, introduction of exotic species, excessive tourism, pollution, corruption
of park staff and insufficient funding for management (Dixon and Sherman,
1991; Shah, 1995; Primack, 1998). Protected areas in different countries
are faced different threats. In developing countries, the major threats to
protected areas include logging, cattle grazing, poaching, fire, cultivation,
insufficient funding, corruption and excessive tourism (Machlis and Tichnell,
1987). The major threats in developed countries are often mining, exotic
species, tourism and pollution (Dixon and Sherman, 1991; Mitchell, 1994).
To great extent, most of these threats have to do with the interest conflicts between protected areas and local residents living in or near protected
areas who bear mostly the opportunity cost resulted from the existence of
protected areas. In the last several decades, most of the newly established
32
protected areas, especially in developing countries where large area of wilderness still exists, have followed the strict preservation model (or the so called
U.S. national park model) which emphasizes keeping ecosystems functions
in their natural state and minimizing any possible human interference. To
do this, after protected areas are established, local communities are usually
precluded from exploiting natural resources they need, as they traditionally
have practiced, or they are even forcibly resettled without compensation. In
many cases, this has resulted in confrontation between local communities
and park authorities, illegal exploitation of resources in protected areas, and
sometimes leads to refusal of local residents to establish new protected areas
or to expand existing protected areas. This has become increasingly evident
especially because the rapidly increasing populations in developing countries
need more land and resources to survive. Some authors even accused the
preservation model that it is an implicit form of second wave of colonialism (Adams and McShane, 1996). Leaving aside the debate about whether
such an accusation is fair or not, the fact is that political pressure increased
dramatically in developing countries to change the concept of conservation
(Kramer and van Schaik, 1997). Recently, this has partly led to the marked
decrease in numbers of the newly established protected areas in the tropics
as a whole, and to near zero in some countries of tropical Asia and Africa
(Terborgh and van Schaik, 1997). It must be recognized that, in the long
run, protected areas can survive only when they are supported, or at least
tolerated by local communities. And unless local communities can benefit
from protected areas, there will be no long-term incentive to support the
existence of protected areas. This may be the most serious problem which
existing protected area networks are faced. In addition, as a result of the prevailing insufficient funding for protected areas and corruption of park staff,
some conservationists question also the ability and the willingness of central
governments to conduct effectively the traditional top-down preservation approach followed by most of the protected areas throughout the world. This
query holds especially for the developing countries in which the commitment
for conservation has been strongly weakened because of the severe financial
reality. Many protected areas in developing countries are in fact the so called
´paper park´ where protected areas are designated but never implemented.
In sum, as a result of the problems of insufficient size and inadequate
representative, the present protected area networks need to be adjusted and
expanded on a scientific basis to include a more complete pallet of various
ecosystems and thereby to maximize the protected biodiversity in the long
run (MacKinnon, 1997). To effectively protect biodiversity, the present inadequate management practice of many existing protected areas should be
33
improved (Brandon, 1997). All these require the support of the relevant
interest groups which bear the cost derived from the existence of protected
areas, whatever they are local communities, private organizations or national
governments. It follows that the traditional preservation model of protected
areas, which emphasizes the strict protection of habitats but easily results in
the hostility of affected interest groups toward protected areas, alone may be
insufficient to achieve the previous aims, and the general conservation policy
should therefore be reconsidered. An alternative approach, which enables
people to benefit from the maintenance of protected areas in a sustainable
manner without substantially harming biodiversity, must be found to supplement, or to substitute for the strict preservation approach under some
circumstances.
In the last two decades, many conservationists and scholars supported the
concept that protected areas should be part of a larger portfolio of sustainable
resource use, namely protected area network should include the component of
strictly protected core reserves, and the component of areas in which limited
resource use is permitted. They promoted accordingly an incentive-oriented
approach, namely, that people are allowed to use wild renewable resources
in protected areas or in buffer zones around protected areas. In some cases,
local communities are also authorized to management natural resources and
human activities in protected areas. This alternative strategy is often called
the sustainable use approach. Due to self-interest, it is expected that more
protected areas, whether existing or new, will be accepted or even designated
actively by people under such an approach, if people really benefit from the
maintenance of protected areas (IUCN/UNEP/WWF, 1980; 1991). Numerous initiatives have been implemented around the world, and many relevant
researches have been conducted to investigate the results of the sustainable
use approach and their implications for both general conservation policy and
specific protected area policy.
On the other hand, the introduction and application of the sustainable use
approach in conservation practice, and the fact that this approach has greatly
influenced the conservation policy of the so called mainstream conservation
organizations such as Worldwide Fund of Nature (WWF, formerly known as
World Wildlife Fund) and World Conservation Union (IUCN) (Kramer and
van Schaik, 1997), have induced widespread debate on the question, whether
and under which conditions the sustainable use approach is appropriate as
an instrument for biodiversity conservation. This is just what we concerned
about in this dissertation. Before the details of the debate are discussed, we
should turn our attention to the definition of the concept sustainable use of
34
renewable resources.
4.3 Defining ´sustainable use of renewable resources´
The use of renewable resources for material or immaterial purposes is an
old tradition of human beings and the basis of human civilization. However,
the sustainable use of renewable resources as an instrument of biodiversity
conservation is a new concept. To avoid misunderstanding in the discussions
throughout the remaining part of this dissertation, it is critical to define
precisely the relevant terminologies.
First, the term ´renewable resources´, both in the customary usage of the
conservation community and in this dissertation, refers to the wild biological
resources. Therefore, some non-biological renewable resources such as water,
and biological resources which are domesticated, are not within the scope of
our discussion.
Next, the term ´use´ has reference to the different ways in which wild
biological resources are utilized by human beings. Usually, it encompasses
the consumptive and the non-consumptive use of wild biological resources.
The consumptive use, or the extractive use, refers to the direct utilization of
parts or products of organisms of biological resources such as meat, skin or
wood. It includes often hunting of wildlife, fishing, logging and gathering of
plants and of other nontimber products. In the way of the non-consumptive
use, no parts or no individual will be taken, directly and intentionally, from
their population. Some losses in the population may happen as a result of the
indirect impact of human activities, for example in the case of ecotourism,
but generally to a minor extent. Both consumptive and non-consumptive
use are important issues and are within the scope of this dissertation. However, in the context of biodiversity conservation, we are mainly concerned
with the consumptive use, as most members of the conservation community
do, because of the easily understood reason that, compared with the nonconsumptive use, consumptive use of wild biological resources is generally less
compatible (at least seemingly) with conservation of biological resources.
Now let us turn to the definition of sustainability in the context of the biological resources utilization, or more precisely the following question: when
is the use of a certain species or a certain population sustainable? The
World Conservation Strategy (IUCN/UNEP/WWF, 1980), the well known
document which first promoted the sustainable use of renewable resources
as an instrument for conservation, defined that the use of wild biological resources is sustainable if their wild populations are not significantly affected.
35
Under this definition, natural resource stock is viewed as a certain kind of
capital, and resource user takes only away the interest derived from the natural capital. The deficiency of this definition is, that any consumptive use
will inevitably lower the population of utilized species, but even though the
population is greatly affected, it does not necessarily imply that the use is
unsustainable.
To correct such a deficiency, we may define alternatively that the use of
biological resources is sustainable if the production of resources can balance
with their harvest . But this definition possess as well deficiency in that,
while harvest can equal production at many different population levels, the
population can be reduced to such a low level that local or global extinction
could happen, once a natural disaster destroys the remaining population.
And even though the population level remains stable, it could lose its ecological role to maintain the essential ecological processes and services, or lose
its significance as an useful resources to meet human needs under such a low
population level (Bennett and Robinson, 2000a). Therefore, as Bennett and
Robinson asserted, the definition of sustainability should, in addition to the
resource itself, take wider management goals with reference to resource users
and ecosystems into account.
We apply in this dissertation the definition of sustainability suggested
by Bennett and Robinson (Bennett and Robinson, 2000b), because their
proposal includes more complete criteria which assess the different aspects of
sustainability from biological, ecological and economic perspectives. The use
of renewable resources is sustainable, if the following criteria can be satisfied:
1. Utilized species cannot show a consistent decline in their populations.
2. The populations of utilized species cannot be reduced to such levels
that they are vulnerable to local extinction.
3. The populations of utilized species cannot be reduced to such levels
that their ecological roles in the ecosystem is impaired.
4. The populations of utilized species cannot be reduced to such levels
that they lose their significance as useful resources to human users.
While criteria 1 and 2 can be applied on an in principle unambiguously
scientific basis, it is evident that criteria 3 and 4 are to certain extent ambiguous and therefore can only be applied in a general way. These criteria cannot
36
provide a clear-cut answer to the question about sustainability. However,
they are undoubtedly critical foundation for the assessment of sustainability.
4.4 The debate on sustainable use of renewable resources
in and around protected areas as an instrument of
biodiversity conservation
4.4.1 Background of the debate
During the last two decades, the sustainable use strategy of renewable
resources as an instrument of biodiversity conservation has increasingly become an important component of the policy of the mainstream conservation
organizations. The IUCN and WWF gave more particular emphasis to the
need to increase the number of protected areas in the IUCN´s Categories
V, Protected Landscapes/Seascapes, and VI, Managed Resource Protected
Areas, in which the sustainable use strategy of renewable resources plays a
critical role in conserving biodiversity. They also emphasized the important
role of non-governmental organizations, private companies, individuals, local communities and indigenous peoples in managing protected areas. These
changes within the conservation organizations´ policy reflect a growing sophistication in their understanding of the relationship between protected areas and the human societies in which they exist. On the other hand, these
changes have simultaneously induced one of the most serious controversy regarding conservation issue within conservation community. It involves the
following questions: is sustainable use of renewable resources possible and
appropriate in and around protected areas? Could such a strategy really
help to protect biodiversity more effectively, or just on the contrary, it would
degrade biodiversity more rapidly? To answer these questions, two often
conflicting approaches dominated the debate.
The traditionally prevailing model of protected areas, which originated
from the U.S. National Parks System in the nineteenth century, is often
called the preservation approach. This approach asserts that more progress
will be made toward maintaining biodiversity on the remaining wilderness of
the world by establishment of strictly protected parks and reserves under the
management of national government or international organizations (Noss,
1991). By drawing boundaries around specific areas, absolute banning of
consumptive use of natural resources and effective legal enforcement, human
interference would be minimized and protected areas would be preserved in
their natural, at best non-inhabited state. This view is mainly founded on
the historical experiences that harvest of natural resources has generally led
to serious degradation of biodiversity. Furthermore, it doubts the possibility
37
of success of the sustainable use model under the current insufficient management capacity in many places, and especially under the socio-economic
conditions that prevail in developing countries, in which a large proportion
of the world´s remaining intact biological community exists (Brandon et al.,
1998; Oates, 1999; Madhusudan and Karanth, 2000).
While the preservation approach has, to great extent, successfully safeguarded a small proportion of the remaining wilderness, it potentially antagonized a section of the human community denied access to natural resources,
especially the local community. Consequently, poaching, degradation of resources, and local hostility against protected areas and park authorities increased. In many cases, the interest conflicts between local communities
and management authorities have prevented existing protected-area networks
from enlargement. Under the threat of habitat fragmentation, ecological isolation, poaching and other factors which will greatly degrade the existing
protected area networks, some scientists, conservationists and conservation
organizations began to question the long-term viability of these isolated biological islands. They argue that the protected-area system should be expanded and, owing to the fact that the preservation model may have reached
the limit of its ability, there must be an economic incentive for conservation.
In contrast to the protectionist philosophy, they advocate that more wilderness and biodiversity will be conserved by developing sustainable use strategy
of renewable resources in those wildlands that are not strictly protected, or
in buffer zones around strictly protected areas. The underlying philosophy is
that people will protect what they receive value from, especially when they
receive tangible financial benefits (McNeely, 1988; Western and Wright, 1994;
Freese, 1998).
To help realize the diverse dimensions of this controversy, we summarize
the fundamental arguments for and against the two positions of the debate
in the following subsections.
4.4.2 Perspectives of the sustainable use approach
The primary concern of the proponents of sustainable use approach is,
that the existing protected areas are too small or too fragmented so that,
as the theory of island biogeography predicts, they may prove incapable of
maintaining most of the biodiversity in the long run. Even in the existing
protected areas, scientists have shown that the maintenance of protected areas cannot guarantee the long-term survival of all, or most of the wild species
initially present. Extinctions still occur in protected areas, also in strictly
protected reserves, largely as a result of inadequate area (Lovejoy et al.,
38
1986). To mitigate biodiversity loss derived from habitat degradation and
fragmentation, there are two possible ways. The first one is to create much
larger strictly protected areas. This is hardly realistic under current circumstances because of the associated high economic and social costs. Another
option suggests that, through the creation of buffer zones around strictly protected areas in which sustainable use of renewable resources is allowed, the
total protection effect would be similar to outright expansion of strictly protected areas but without, or with less economic and social costs. And what
the most important is, that regulated, sustainable use strategy will provide
important economic incentives for local people to accept and maintain the
existence of protected areas (Shaw, 1991). If local communities which bear
part of the costs derived from the maintenance of protected areas never benefit from protected areas, they will not support the expansion of existing
protected area network, unless we can accept the undemocratic way usually
used in the past when creating new protected areas, especially in developing
countries.
A stronger position is asserted by the proponents of the so called ´use-itor-lose-it´ strategy. As David Western, the former Director of Kenya Wildlife
Service (KWS), and R. M. Wright indicated, that the most important conservation issue is how to deal with the vast majority of the earth´s land
surface where are not protected by parks and the interests of local communities prevail (Western and Wright, 1994). If conservation agencies do
not encourage local communities to conserve biodiversity by sustainable use
of natural resources in natural and seminatural ecosystems, the remaining
wilderness will sooner or later be converted into human-dominated land and
therefore lose the biodiversity on it, since wilderness as a land use pattern
financially cannot compete with other profitable land use options such as
monocrop agriculture. Instead of focusing on strictly protected areas as the
preservation approach traditionally does, this position tends to emphasize
the importance of conservation actions outside of these areas. Once natural resources and their sustainable use has become integrated into people´s
way of life, the areas with rich biodiversity will be well protected at minimal
costs and biological resources will have a habitat larger than the existing
protected areas. Janzen (1994) asserted that the use-it-or-lose-it approach
can probably protect 80-90 percent of tropical terrestrial biodiversity on a
land surface of 5-15 percent of the tropics, in sharp contrast with the result
of the conserved 10-30 percent of biodiversity on 1-2 percent of the lands
when the preservation approach still dominates the conservation policy. Furthermore, sustainable resource use may potentially support conservation in
the way that harvesting on one site may relieve pressures to harvest on other
39
sites with higher conservation priority (Freese, 1998). It is here worth noting
that many conservationists support the sustainable use approach, not just
because of the moral or public relations´ considerations, but because they
believe that it is the best strategy under prevailing conservation constraints,
especially in developing countries.
While any use of biological resources has inevitably direct impacts on
their populations and involves a loss of biodiversity (Robinson, 1993), some
proponents of the sustainable use approach argued that, through the creation
of economic incentive and thereby enlarged habitats and better protection,
there are many cases in which the sustainable use of wild species has in
fact supported the conservation of the target species, such as the well-known
examples of white-tailed deer and snow goose (Medellı́n, 1999). These two
species have today much higher levels of total population than those of the
period before sustainable use strategy was applied. Moreover, ecosystem
conservation has in many cases benefited from specific sustainable use programs. For example, the waterfowl management and fostering programs in
North America has greatly contributed to the conservation of wetland and
biological resources living there (Medellı́n, 1999). Similar examples can be
found in many southern African countries where sustainable use of wildlife
resources comprises an even better form of land use than other options. And
this has greatly promoted wildlife conservation and habitat protection simultaneously in those countries (Prins et al., 2000).
In addition to the pragmatic considerations discussed above, some people
support the idea of sustainable use as a result of the moral considerations of
justice and human rights. As Adams and McShane (1996, P.xviii) claimed
´conservation has long operated on the comfortable belief that Africa is a
paradise to be defended, even against the people who have lived there for
thousands of years´. Strictly protected areas have come under criticism for
the fact that, after protected areas are created, local people are usually precluded from utilizing natural resources on which their livelihood depends.
They ignored the potential benefit of sustainable resource use for local people. Sometimes traditional preservation approach is accused of being a style
of imperialistic (Gadgil, 1992) or colonial operation (Adams and McShane,
1996). Pimbert and Pretty (1997), and Ghimire and Pimbert (1997) claimed
that the main reason of the failure of the protected areas is that western
norms and practice of conservation science ignores the fact that local communities should be in a better position to assess what is good for them than
western scientists. Some people stake out an extreme position that all protected areas should be open to some kind of use (Wood, 1995; Ghimire and
40
Pimbert, 1997). In short, this view tries to achieve the goal of ´win-win´,
namely, both people and nature are winners. These considerations about
justice and human rights have especially reference to the issue of indigenous
peoples´ self-determination and territorial control.
Indigenous peoples currently inhabit and claim a land area of between
20% and 30% of the earth´s surface, in which most of the wilderness, and
most of the existing and potentially new protected areas of the world are embedded. This is as large as four to six times more territory than is included
in the strictly protected areas of the earth. And as recent judicial decisions
in many countries showed, indigenous peoples may reassert ownership of significantly larger land than they have at present (Stevens, 1997). The land
use patterns on this large territory are of great importance from either the
point of view of human rights or of biodiversity conservation. Some conservationists asserted that, from the perspective of justice and human rights,
sustainable resource use by indigenous peoples should be allowed, even in
strictly protected areas. Moreover, through sustainable use of natural resources and co-management of protected areas in which indigenous peoples
share resource ownership and responsibility, conservation will be more likely
effective, because, apart from the effect that protected areas will be accepted
more easily, indigenous peoples have possessed necessary ecological knowledge, and developed social and cultural mechanisms that regulate resource
use to sustainable levels, or at least mitigate its impact on the environment to
the acceptable extent. Conservation actions might be conducted more costeffective by indigenous people in a decentralized way than the traditionally
state-centric approach. This optimistic attitude toward indigenous peoples´
role in conservation is often called the modern version of ´the ecologically
noble savage´ (Redford, 1991).
4.4.3 The Community-Based Conservation (CBC)
Conservation thoughts and policies have changed to great extent in the
1980s in response to the call that local communities should be involved in
conservation rather than being ignored, and that more attention should be
paid to the positive roles of sustainable resource use and economic incentives
in conservation. This has led to the popularity of a special form of the
sustainable use strategy, the Community-Based Conservation (CBC). The
idea is so popular and so widely accepted that it sometimes becomes the
synonym of sustainable use approach in developing countries, though, not
only at community level, sustainable use approach includes in fact a wider
range of conservation initiatives at other levels. As a result of its importance
41
in conservation practice in developing countries, it is here worthy of briefly
investigating the idea of community-based conservation.
In the broadest sense, Western and Wright (1994, p.7) maintained that
´community-based conservation includes natural resources or biodiversity protection by, for and with the local community´. Adams and Hulme (2001, p.13)
defined the community-based conservation as ´those principles and practices
that argue that conservation goals should be pursued by strategies that emphasize the role of local residents in decision-making about natural resources´. In
practice, conservation goals can be pursued by community-based conservation in three ways: (1) permitting local residents living in or near protected
areas to participate in resource use and management policy; (2) transferring ownership or user rights over natural resources to local residents; and
(3) benefiting local community through conservation (Hackel, 1999). This
includes a wide range of various conservation initiatives with different titles, such as community wildlife management, collaborative management,
community-based natural resource management, and integrated conservation
and development programs5 (Adams and Hulme, 2001).
According to the criteria of objectives, ownership/tenure status and management characteristics, Barrow and Murphree (2001) identified three primary types of community conservation approach:
1. Protected Area Outreach: The primary objective of this approach is
conserving ecosystems and biodiversity. The land and natural resources
of protected areas are owned legally by state, and all decision-making
with reference to the management of protected areas are made by state
alone. However, it seeks to share the benefits derived from protected
areas, such as entrance fee, with local community, and thereby resolve
the conflicts between protected areas and local communities in a mutually agreeable manner. Extractive use of renewable resources is in
principle not allowed. Local communities play a passive participatory
role in management practice of protected areas.
2. Collaborative Management (or Co-management): In addition to the
primary objective of conserving ecosystems and biodiversity, this approach pursues also some local livelihood benefits through sustainable
5
Integrated conservation and development program (ICDP) refers to the social and economic development program that offers local community alternatives to natural resource
use (Brandon and Wells, 1992). It usually focuses on the improvement of infrastructure
and seldom involves the utilization of renewable resources.
42
use of natural resources. The land and natural resources of protected
areas are owned and managed legally by state, but some user rights and
management responsibilities of certain resources are devolved upon local communities to achieve conservation as well as livelihood improvement objectives. Controlled extractive use of renewable resources which
is agreed by conservation authorities and local communities is allowed.
Local communities share certain management responsibilities of protected areas and of natural resources with conservation authorities and
thereby play a more active role than in protected area outreach, but
conservation authorities still play a predominant role in deciding management policy.
3. Community-Based Conservation: The primary objective of this approach is sustainable management and use of natural resources by local communities. Through devolution of the user or property rights of
natural resources to local communities, the land and natural resources
are owned and managed either de jure or de facto by local communities, although state generally retains some control of last resort over
land and natural resources. Sustainable extractive use of renewable
resources is allowed and encouraged. Local communities play an active
and predominant role in management practice of protected areas.
All these ideas have great influences on conservation policies and practice
throughout the world. Of the three primary types of community conservation
approach, we are particularly interested in community-based conservation
and collaborative management which emphasize the role of sustainable use
of natural resources as an instrument for conservation. Whether these models
are as effective as they are said to be by their proponents has become the
focus of the debate.
4.4.4 Perspectives of the preservation approach
The point of view that utilization of natural resources is in principle compatible with conservation objective has been criticized by proponents of the
preservation approach. They asserted that, given current ignorance about
biological and ecological knowledge, and given the prevailing socio-economic
circumstances in most countries, the sustainable use approach will result in
substantial losses of biodiversity. Maybe the use strategy is politically correct
and intellectually appealing, but it is in most cases less effective than it is said
to be. There is growing scientific literature that question the probability of
successfully achieving sustainability in practice. For example, after reviewing
43
some historical experiences, Ludwig et al. (1993) asserted that, once natural
resources are opened to utilization, they are usually overexploited to extinction or to the population level of collapse. In the following we briefly review
the primary factors that contribute to overexploitation.
Some species are particularly vulnerable to harvest as a result of their
special biological attributes. For example, it has long been recognized that
long-lived and slow-reproducing species, such as primates, elephants, whales,
sharks and tropical hard woods have low intrinsic growth rates and are particularly vulnerable to harvest (Mangel et al., 1996; Gullison, 1998; Bennett and
Robinson, 2000b). The mammals in tropical forests are also easily overexploited because of their low standing biomass (Robinson and Bennett, 2000).
Species whose behavior allow easy harvest, that do not have the ability to
recolonize hunted area, or that are intrinsically rare are highly vulnerable
to harvest (Bennett and Robinson, 2000b). In addition, uncertainty which
emerges from the inherent stochasticity of ecosystems and from human ignorance about biological and ecological knowledge makes the task of sustainablly utilizing resources more difficult than it is expected to be. Many cases
show that the fluctuations in resource stock is so great and unpredictable
that it is usually too late until significant decrease in resource stock is detected, and this significantly increases the risk of extinction (Ludwig et al.,
1993; Lavigne et al., 1996).
At ecosystem level, the use approach is often criticized that it rarely
consider the impact of exploiting target species on all the other species living
within the same ecosystem and on the ecosystem itself as a whole. There
would be enormous impact especially when the target species in question
is a keystone species. For example, the exploitation of certain tree species
in old-growth forests are usually not compatible with the conservation of
the other non-harvested species whose survival depend on the integrity of
forests (Struhsaker, 1998). Furthermore, the economic incentive behind the
use strategy usually favors the increase of those more highly valued species
and lead to the simplicity of ecosystems (Freese, 1997). Simplicity means
that through modifying biological or abiotic factors, such as eliminating those
species that have competitive or predatory relationship with the highly valued
species, the environment will become more feasible for the production of
target species at the expense of other components of biodiversity.
In addition to biological and ecological factors, some fundamental socioeconomic factors play an important role in influencing sustainability of natural resource use. Clark (1973) demonstrated in his pioneering paper that low
44
growth rate of renewable resources, high discount rate and high price/cost ratio of harvest are the primary factors contributing to overexploitation. The
usually prevailing economic, social and institutional conditions in many
countries, such as high interest rate, poverty and uncertainty about tenure
and resource market, create high discount rates that are disadvantageous to
sustainable use of renewable resources (Freese, 1997). With reference to the
problem of price/cost ratio, certain new economic model such as the Swanson model6 (1994) and empirical cases such as trophy hunting in southern
African countries (Child, 2000) show that high price/cost ratio, in contrast
to the result of the Clark model, can help conserve utilized species by offsetting the opportunity costs of competing land use options under appropriate
property rights and policy environment. However, Freese (1997) maintained
that an unusually high price/cost ratio may create a destabilizing environment against sustainable use of renewable resources, such as the numerous
examples of the collapse of coastal fisheries throughout the world.
Based primarily on the arguments discussed above, many question the
feasibility of the community-based conservation projects in achieving conservation objectives. Songorwa et al. (2000) examined the four premises underlying the community-based conservation: (1) that national governments are
willing to devolve user rights, property rights and management responsibilities for natural resources to local communities; (2) that local communities are
willing to participate in resource management; (3) that local communities are
able to management resources; and (4) that conservation is compatible with
rural economic development. After reviewing existing literature regarding
the practice of community-based conservation in Africa, they take the skeptical view that, in the African context, the four premises are problematic
and the approach is less effective than it is expected to be by its advocates.
Community-based conservation is expected to work only in areas with large
piece of wilderness where big populations of wildlife still exist and human
population density is low. Similar views are shared by Noss (1997), Spinage
(1998), Hackel (1998) and Oates (1999). In addition to the arguments discussed above, Noss emphasized the problems of ethnic diversity as a result of
immigration and of the lack of conservation ethic that challenge the practicability of grass-root conservation initiatives. Spinage blamed the fallacy of the
ecologically noble savage hypothesis for its influence on misleading conservation policy. Hackel addressed the problems of poverty, long-standing economic stagnation and rapid population growth in Africa, and concluded that
the idea of community-based conservation is oversold, because it is unlikely
that such idea can be applied generally in rural Africa. Oates denounced
6
The Swanson model will be deliberately investigated in chapter 5.
45
the claim of some proponents that central governments of many developing
countries are not willing to and/or not capable of safeguarding their protected area network. He recognized that many protected areas in developing
countries did become increasingly ineffective as a result of the serious economic, social and political problems. However, that is the flaw of rather use
approach than of preservation approach. He questioned whether, with much
lower management capacity and without public authority, local communities
could perform conservation actions more effectively than central governments
under the same circumstances. Based on his empirical studies in developing
countries, he concluded finally that, in contrast to numerous failures of the
community-based conservation initiatives, the traditional top-down preservation model can work relatively well in developing countries, even in the
face of the serious economic, social and political problems.
A synthesis of the above arguments leads to the skeptical attitude of
many people toward sustainable use approach as an instrument of conservation. They assert therefore that the preservation approach should still be
the cornerstone of all conservation strategies (Kramer et al., 1997; Brandon
et al., 1998; Oates, 1999). This does not mean that the use approach should
not be applied in any case, but that more emphasis and resources should be
devoted to the protection, instead of use, of protected areas.
4.5 A case study: the national park system of Taiwan
4.5.1 Introduction to the national park system of Taiwan
Taiwan, with a land area of 35,570 square kilometers, is a subtropical/tropical island located between cool-temperate Japan to the north, subtropical south China to the west, and tropical Philippines and Indo-Malayan
islands to the south, a location which is just on the mixed edge of several different biogeographical regions. The island is dominated by rugged, forested
mountains with more than two hundred peaks over 3,000 meters. All these
climatic, biogeographical and topographic characteristics support a highly diverse flora and fauna communities (COA and DNP, 1992). More than 4,000
species of vascular plant, of those a quarter being endemic to Taiwan, and a
spectrum of 6 forest types have been found existing in Taiwan. In addition,
Taiwan has a fauna world with 62 species of mammals, 500 species of birds,
95 species of reptiles, 32 species of amphibians, 150 species of freshwater
fish, and an estimated 50,000 insect species (of those more than 400 species
of butterflies) (COA, 1997).
On the other hand, the rapid economic growth, high population density (over 600 people per square kilometer) and illegal harvesting of natural
46
resources have significantly degraded almost all coast areas, flatland and
slopeland below 500 meters in the past fifty years, and are still threatening
the remaining intact wilderness. To effectively protect the natural environment of Taiwan, 58 strictly protected areas have been designated in last two
decades by the central government of the Taiwan, Republic of China. Of these
there are 6 national parks (NP), 18 natural reserves (NR), 11 wildlife refuges
(WR), and 23 national forestry natural protected areas (NFNPA) (Lu, 1999).
This system covers an area of about 451,951 hectares, or about 12.6% of the
land surface of Taiwan (Table 4.1). As Table 4.1 shows, the national park
system, comprising 70% of the total surface of the strictly protected areas, is
the cornerstone of the nature conservation in Taiwan. Moreover, in contrast
to the de facto ´paper park´ status of NR, WR and NFNPA, the national
park system generally possesses sufficient staff, budget, and relatively complete management institutions. We therefore, as an example for illustrating
the arguments stated in Sections 4.1 and 4.2, concentrate our discussion on
the performance and problems of national park system in Taiwan.
Table 4.1 Protected areas in Taiwan
Types of protected
NP
NR
WR NFNPA Total
area
Numbers
6
18
11
23
58
Coverage/ha
322,207 63,279 11,714 82,654 451,951
% of land surface
9.0
1.80
0.30
2.30
12.6
of Taiwan
% of all protected
70.12
13.77
2.55
17.99
100
areas
Source: Lu (1999), p.66.
The establishment of the national park system in Taiwan can be traced
back to early 1970´s. As a result of the international conservation movement
and the growing environmental consciousness of citizens, the National Park
Law was promulgated in 1972. However, it was not until 1981 that the
authority which is responsible for establishing and supervising national park
system, the National Park Department, was established in the Ministry of
Interior (DNP, 1999). Thereafter, the first national park in Taiwan, the
Kenting National Park, was designated in 1984. The number of national
Parks increased rapidly in 1980´s and early 1990´s. Up to now, the system
includes 6 national parks and accounts for nearly 9% of the total land surface
of Taiwan (see Table 4.2 and Figure 4.1).
47
Table 4.2 List of National Parks in Taiwan
Name of National Major Protected Features
Area (ha.)
Park
Kenting National Marine Ecosystem, tropical
17,731
Park
coastal rainforest, waterfowl,
(land)
migratory birds, butterflies,
14,900
limestone caves, cliffs.
(marine)
Yushan National Virgin forest, wildlife, rare
105,490
Park
species of flora and fauna,
high peaks and mountainous
terrain.
Yangmingshan
Butterflies, birds, amphibians, 11,456
National Park
volcanic topography,
meadows, hot springs,
waterfalls.
Taroko National
Virgin forest, wildlife, marble 92,000
Park
gorge, cliffs, high mountains.
Shei-Pa National Virgin forest, wildlife, rare
76,850
Park
species of flora and fauna,
high mountains.
Kinmen National historical battlefields,
3,780
Park
traditional villages, natural
scenic areas, flora and fauna
of island.
Source: COA and DNP (1992), p.17; DNP (1999), p.26.
48
Date of
Designation
1.1.1984
4.10.1985
9.16.1985
11.28.1986
7.1.1992
10.18.1995
1
6
0
10KM
1.Yangmingshan National Park
2.Shei-Pa National Park
3.Taroko National Park
2
4.Yushan National Park
3
5.Kenting National Park
6.Kinmen National Park
The Tropic of Cancer
4
Japan
0
Mainland China
TAIWAN
50KM
Lanyu
5
Philippines
Figure 4.1. National Parks in Taiwan. Source: DNP (1999), P.2.
49
4.5.2 Management issues
Like most national park systems throughout the world, the national park
system of Taiwan aims primarily at protecting natural and cultural resources.
And under the premise that the above aim is achieved, national parks can
help achieve some minor aims. More specifically, the goals of the national
park system of Taiwan include (COA and DNP, 1992):
1. protecting ecologically significant areas
2. conserving gene pools
3. promoting scientific research and environmental education
4. providing nature-based recreational opportunities
5. promoting local economic development through developing tourism that
is compatible with nature conservation.
The primary aim of almost all national parks is conserving biodiversity.
Only the Kinmen National Park, the first cultural national park in Taiwan,
was established primarily for protecting historically and culturally important
sites. The park system follows principally the strict preservation model of
the U.S. National Parks. Therefore, the extractive use of natural resources in
national parks is generally not allowed, with the exception that some limited
land use forms, such as grazing, farming, fishing (only in marine area) and
use of hot springs may be allowed under the supervision of the park authority,
if they have long existed before the park is designated. In any case, harvest
of wildlife is absolutely forbidden. In order to achieve the multiple objectives
of park system which are in certain cases conflicting with each other, the area
within the boundary of every national park is divided into several different
management zones. These includes (Lin, 2000):
(1) General Protection Areas: they are land or marine areas in which certain land use forms have long existed before the park is designated.
Limited land use under the supervision of the park authority is allowed
to continue to mitigate the potential conflicts between local residents
and park authorities. They accounts for 30.36% of total area of the
park system.
50
(2) Recreation Areas: these are areas in which recreational facilities are
allowed to be established and large numbers of tourists are allowed to
enter. They accounts for 0.69% of total area of the park system.
(3) Cultural/Historic Preservation Areas: these are areas with special cultural and/or historic significance in which tourist use is limited. They
comprise only 0.12% of total area of the park system.
(4) Special Scenic Areas: those strictly protected areas with special or
spectacular natural scenery. Any development activities are forbidden.
Tourism is allowed, but its impact on natural environment must be
kept to a minimum level. They accounts for 10.74% of total area of the
park system.
(5) Ecological Protection Areas: those strictly protected areas with unique
ecological value. Any development activities in and human access to
these areas are forbidden, with the exception that scientists and hikers
with special permits are allowed to enter these areas. Hikers should
stay on existing trails. They comprise 58.10% of total area, and are
the cornerstone of the in situ conservation in national park system.
According to the National Park Act, the department of National Parks in
Construction and Planning Administration, Ministry of Interior, is the authority responsible for selecting and managing national parks. To effectively
manage individual national park, each park is supervised by a park headquarter which consists one police corps and five administrative divisions,
including the planning, the construction, the conservation, the tourism and
the interpretation education division. In general, the central government of
Taiwan, Republic of China, advocates the in situ conservation in national
parks, and equip park headquarters with sufficient funds and young, highly
qualified staff. This contributes to the result that the department of National
Parks and park headquarters have become the most active public authorities
in safeguarding biodiversity in Taiwan.
Table 4.3 tabulates the budget of the national park system during the
time period 1982 to 2000. The budget devoted to the national park system
by central government has steadily increased since the first park was planned.
This reflects partly the rapid increase in the number of national parks during
1986 to 1996. However, even after the number of parks ceased to increase in
1996, the total budget still slightly increased until the fiscal crisis of the central government changed the trend in 2000. Of the many budget categories,
51
staff salary and investment in equipments and infrastructure constitute the
majority of the total budget. The steady increase of expenditures on staff
salary reflects that both staff and average pay have increased in succession, or
generally speaking, that the human resources of the park system have significantly improved in the last two decades. Similarly, the generally increasing
expenditures on investment in equipments and infrastructure reflects that
the park system is improving in physical capital. All these factors contribute
to the relatively well performance of the park system in recent years.
Table 4.3 Budget of the national park system of Taiwan (Unit: NT$1,000)
Staff salary Investment in equipments
Others
Total
and infrastructure
1982
2,508
0
2,710
5,218
1983
2,640
0
2,710
5,350
1984
13,801
101,000 101,057
215,858
1985
40,686
239,330
87,452
367,468
1986
92,816
415,710 152,386
660,912
1987
69,955
400,480 202,369
672,804
1988
93,702
628,428 234,410
956,540
1989
117,523
829,041 286,851 1233,415
1990
130,757
1074,679 362,110 1567,546
1991
194,519
1211,041 449,349 1854,909
1992
297,583
1054,861 410,730 1763,174
1993
348,733
1116,150 647,381 2112,264
1994
444,001
1211,916 405,813 2061,730
1995
481,028
1315,790 379,209 2176,027
1996
544,125
1086,095 449,835 2080,055
1997
601,425
1292,226 427,188 2320,839
1998
625,020
1664,187 332,698 2621,905
1999
677,816
1636,608 335,400 2649,824
2000
983,514*
1804,736* 661,068* 3449,318*
*Figures including the budget of one and a half year.
Source: EYROC (1981-1999) and personal calculation.
4.5.3 Effectiveness of the national park system
In this subsection we make use of some indicators to evaluate the effectiveness of the national park system in conserving biodiversity. For some
rare plants and animal species, especially mammals with large body size,
national parks have become their last refuges under the strict preservation
policy. Table 4.4 shows the number of plants species which have been found
52
to exist in national parks. As Table 4.4 indicates, each national park, except
the Kinmen National Park, protect more than one fourth of the total about
4,000 species of plants existing in Taiwan. As a result of the overlapping
of some species, it is unknown whether national parks protect most of the
plants species existing in Taiwan. But it is hardly probable that national
parks safeguard only a minor proportion of total plants species.
Table 4.5 tabulates the number of animal species which have been found
to exist within the boundary of national parks. Leaving aside the animal
species found in Kinmen National Park, the other five national parks protect
most of the mammals, birds and amphibians species, and about half of the
reptiles and fresh fish species. In addition, 1,105 sea fish, 616 Mollusk, 327
coral and 295 crustacen species have been found in marine area of the Kenting
National Park (CPA, 2000).
In the past five decades, habitat degradation and illegal harvesting of wild
plants and animals has severely reduced populations of any valuable species
to an endangered level. Some species, such as clouded leopard (Neofelis nebulosa brachyurus) and Formosan flying fox (Pteropus dasumallus formosus)
are suspected to have gone extinct for a long time. Until after the national
parks have been designated in succession and the National Park Act effectively enforced by park headquarters, the depressing situation has, at least
within the boundary of national parks, to great extent changed. Some observations and reports in the field showed that populations of wildlife, including
almost all of the mammal species with large body size and rare bird species
such as Formosan black bear (Ursus thibetanus formosanus), Mikado pheasant (Syrmaticus mikado) and Swinhoe´s pheasant (Lophura swinhoii), have
significantly recovered in national parks (The Nature, 1991; 1995b). Based
on the standing observations of scientists and park staff, it seems to be reasonable to assert that national parks have effectively safeguarded the flora
and fauna communities within park boundaries, although more rigorously
scientific works are still needed to evaluate the long-term effectiveness of
national park system in conserving biodiversity.7
Equipped with sufficient funds and highly qualified staff, park headquarters have actively dampened the illegal activities within national parks (Sung,
1999). Table 4.6 tabulates the number of illegal activities detected by park
headquarters in 1990s. No clear-cut conclusions with reference to the trend
of illegal activities can be drawn from the statistics. However, even under the
severe pressures resulted from economic development and increasing human
7
Some exceptions to this assertion, however, will be discussed in the next subsection.
53
population, illegal activities are as ever kept under control by park headquarters to an acceptable level, except in a few cases discussed later.
One of the important tasks of the national park system is to provide
interpretation education service for tourists which aims at raising public environmental consciousness. Table 4.7 shows the statistics of the relevant
briefing and guided tour attendance in 1990s. Both the briefing and guided
tour attendance have dramatically increased, almost year by year. It is difficult to measure the precise effects of interpretation education service on the
formation of public environmental consciousness, but the more environmental friendly behavior of the tourists in national parks in recent years might
be partly attributed to the active promotion of environmental education initiated by park authorities.
4.5.4 Current problems of the national park system
While certain degree of success in preserving majority of the floral and
faunal species and in protecting some types of habitat have been achieved by
the national park system of Taiwan, a critical investigation into the system
shows that, like almost all of the protected area systems throughout the
world, the selection, establishment and management of national parks in
Taiwan are involved in some problems which need to be solved.
First, from the perspective of representative, the national park system
should try to encompass the representatives of as many types of ecosystems
as possible. However, under the circumstances that lowland, coastal and
marine areas have long been intensively cultivated, settled or utilized, some
ecosystem types in these areas, including lowland forests, wetlands, coastal
regions and marine ecosystems, are virtually underpresented in national park
system, while most parts of the national parks are located in lower-mountain
or mountain forests. This is recognized by the Department of National Parks
to be one of the current deficiencies of the national park system. To improve
the representative, they plan to establish a few new national parks which
include wetland and marine parks (The Nature, 2000).
Next, although the effective enforcement of the National Park Act has
brought most illegal activities, such as poaching, logging, mining and cultivation, under control, the existing national parks are generally suffered from
some management problems, in particular the excessive tourism and hostility
of local people toward national parks. Some popular and spectacular sites
in national parks attract several millions of tourists every year, and this has
54
severely degraded some important habitats in the parks. A well-known example is the destruction of coral reefs in Kenting National Park caused by
pollution and human disturbance brought by about four millions of tourists
per year (The Nature, 1996; Dai et al., 1998, 1999). Similarly, about 12
millions of tourists per year visit the Yangmingshan National Park which is
located near the metropolitan city Taipei, and seriously degrade some parts
of the park (Lin, 2000). Table 4.8 demonstrates the rapidly increasing pressure of tourism putted upon the national park system in recent years. From
1992 to 1999, the number of tourists visiting national parks has dramatically
increased by 78.13%.
The solutions to the problem of excessive tourism may be limiting tourist
numbers and limiting access to fragile habitats, but such measures will probably, at least in the short term, reduce tourism revenues and arouse more
severe hostility of local people toward national parks than before. Indeed,
as a result of the strict protection policy of national parks, the conflicts
between local residents and park authorities have long lasted. Although limited land use under the supervision of the park authority is allowed in general
protection areas if these land use forms have long existed before parks are
designated, local residents living in and around national parks still claim that
their traditional land and natural resource user/property rights have suffered
from the designation of national parks. All national parks have therefore, to
various extent, experienced the protest of local residents (The Nature, 1993,
1994a, 1994b, 1995a, 1999; Sung, 1999; Huang, 1999). In particular, the indigenous people played a primary role in the protest initiatives, because the
majority of national parks is located in the traditional territory where indigenous people live and lay claim to. They claimed the land tenure, respect for
the indigenous culture, relaxation of land use limitations and legal hunting,
fishing and gathering rights of renewable resources in national parks.8 Meanwhile, national park authorities persist in their strict protection policy, and
these conflicts led finally to the failure of the efforts to expand the existing
national park system. Under severe protest of indigenous people, the Department of National Parks stopped designating two planned new national
8
It is here worth noting the fact that not all indigenous people oppose the designation
of national parks. Some people consider that national parks effectively prevent the modern
market economy from invading indigenous communities in and around national parks, and
thereby, maybe without intention, help protect their traditional culture and land tenure
(Huang, 1999; Ming-I Gu, personal communication, 4.17.2001). Without national park
or similar protected area and under the pressure of the predominant modern civilization,
as the example of the planned Lanyu National Park shows, indigenous people might have
lost their culture, natural resources and land tenure more rapidly than expected in the
situation if national park is successfully designated (Huang, 1999).
55
parks, including the Lanyu National Park (The Nature, 1994b; Huang, 1999)
and Nandan National Park (Huang, 1999). In addition, whether the planned
Chilanshan National Park can be successfully established under protest, remains still to be seen. In short, except the successful designation of the
Kinmen National Park in 1995, all plans for new parks proposed after middle 1990s were forced to been abandoned or deferred by protests initiated by
indigenous people.
Table 4.4 Numbers of plant species in national parks of Taiwan
National Park Pteridophyta Angiospermae Dicotyledons Monocotyledons Total
Kenting
194
3
898
143
1,238
Yushan
254
18
726
146
1,144
Yangmingshan
181
2
747
294
1,224
Taroko
223
18
747
294
1,183
Shei-Pa
223
19
706
155
1,103
Kinmen
36
2
351
153
542
Taiwan area
582
28
na
na
na
Source: Lin (2000), p.26; CPA (2000), p.143.
Table 4.5 Numbers of animal species in national parks of Taiwan
National Park
Mammals Birds Reptiles Amphibians Fresh fish Sea fish
Kenting
15
184
35
14
30
1105
Yushan
34
154
17
12
3
0
Yangmingshan
14
77
35
20
12
0
Taroko
31
147
30
14
16
0
Shei-Pa
32
97
14
6
16
0
Kinmen
8
281
13
5
na
9
Total animal species
43
260
57
25
42
na
∗
in national parks
% of animal species
69%
65%
44%
86%
50%
na
∗
of Taiwan
*Figures excluding the animal species of Kinmen National Park.
Source: Lin (2000), p.26; CPA (2000), p.142.
56
Table 4.6 Numbers of detected illegal activities in national parks of
Taiwan
Squatter Cultivating Hunting/ Lumbering Polluting
Fishing
Environment
1992
215
80
26
246
81
1993
69
43
58
123
9
1994
45
79
85
121
65
1995
47
55
69
64
33
1996
92
114
26
71
17
1997
124
104
53
84
49
1998
116
78
18
54
7
1999
90
59
24
39
32
*Including vending, dumping garbage, arsoning, extracting coral, and
trespassing and driving into protected areas.
Source: CPA (2000), pp.128-129.
Table 4.7 Briefing and guided tour attendance in national parks
of Taiwan
Briefing attendance Guided tour attendance Total
1992
404,722
139,553
544,275
1993
409,072
124,678
633,750
1994
414,567
158,900
573,467
1995
444,246
174,288
618,534
1996
452,111
176,312
628,423
1997
517,699
211,659
729,358
1998
578,257
387,643
965,900
1999
584,349
346,593
930,942
Source: CPA (2000), pp.114-115.
Table 4.8 Number of tourists and entrance fee
revenue of national parks in Taiwan*
Number of Tourists Entrance fee revenue
(Unit: NT$1,000)
1992
4845,661
26,033
1993
5805,332
26,036
1994
6811,596
21,343
1995
7437,589
28,730
1996
6429,216
27,083
1997
8525,707
27,878
1998
8290,840
28,443
1999
8631,763
36,448
57
Others∗
Total
1,467
747
851
923
977
2,087
1,534
752
2,115
1,049
1,246
1,191
1,297
2,501
1,807
996
*These figures do not include all tourists, because people
are allowed to visit some parts of national parks without
paying entrance fees.
Source: CPA (2000), pp.102-103.
4.5.5 Prospects of the national park system
The young national park system of Taiwan is, to great extent, successful
in protecting biodiversity within park boundaries, at least at its beginning
stage between the year 1984 and 2000. Its success can be attributed to the
two pivotal factors: the support of the central government and the strict
protection policy. While the support of the public authority is in any case
necessary for the maintenance of such a large protected area system, however,
the strict protection policy, as a critical factor contributing to safeguarding
biodiversity, ironically hindered the expansion of the national park system,
and thereby hindered the possibility of protecting biodiversity out of range
of the existing park system. From the long-term perspective, the existing
national park network need to be enlarged to include a more complete pallet
of various ecosystems and thereby to maximize the protected biodiversity.
To effectively protect biodiversity, the current management problems in some
national parks should be solved. All these require the support, or at least
the tolerance of the local residents, in particular of the indigenous people,
who virtually bear the majority of the costs derived from the designation
of national parks. The question is, instead of the strict protection policy, is
there a better alternative under current circumstance in Taiwan?
To resolve the conflicts between local communities and park authorities,
the central government proposed to modify the National Park Act, so that
indigenous people have the legal rights to utilize the renewable resources in
national parks (The Nature, 2000). The possible modification of current conservation policy aroused the controversy between proponents and opponents
of the sustainable use strategy, like the prevailing debate within conservation
communities throughout the world (Liu, 2000; Chang, 2001). Furthermore,
some conservationists proposed that the idea of co-management could be applied in national parks. In any case, whether and under which conditions
use of renewable resources is feasible will be the core of all policy issues in
national park system of Taiwan. This is also what we are interested in and
will focus on throughout the dissertation.
58
Chapter 5
Economic models of species extinction
and biodiversity loss
In chapter 4 we have identified some primary problems of existing protected area systems. All these problems indicate that protected areas cannot
exist in isolation from the social and economic system in which they have
been created. While their major task is conserving biodiversity, they are
confronted with the same problems which biological communities generally
are faced. To remedy the problems of protected areas, it is therefore necessary to investigate and grasp the fundamental causes of species extinction
and biodiversity loss.
In this chapter we will briefly investigate three important economic models which have been derived with regard to the optimal utilization of renewable resources and to the problems of species extinction and biodiversity
loss, including the Gordon model, the Clark Model and the Swanson model.
These models provide the general analytical framework for most of the analysis that handle the same issues, also for our models presented in the following
chapters. Moreover, the investigation into these models will provide critical
insight for the rethinking of the prevailing conservation policies.
5.1 The Gordon model
The Gordon model (Gordon, 1954) may be, though indirectly, the origins
of the economic theory of species extinction. In his paper Gordon examined
how property rights influence the extent to which fish population is exploited,
especially under the so called ´sole owner´ regime and the ´open access´
regime. The latter means that no one, de facto and/or de jure, owns the
natural resources, and access to resources is open to all. The well-known
conclusion drawn by the Gordon model is, that the equilibrium population
level under the sole owner regime will be higher than the one under the open
access regime, since, under open access, the resource will still be exploited
by any new entrants until that total revenues gained from harvesting equal
total harvest costs when profits are totally dissipated, while the sole owner
will choose to maximize his profits and thereby maintain a higher population
level as a result of the self-interest motive. It follows that the phenomenon of
overexploitation can be attributed to the inadequate property rights. What
in the Gordon model is particularly noticeable is the fact that it is not possible
59
that fish population is exploited to extinction, irrespective of under open
access or under sole owner regime, as long as the harvest costs are positive.
Even though at zero price of harvest effort, the sole owner will choose to
harvest the ´maximum sustainable yield (MSY)´, and thereby preserve a
positive population. This conclusion contradicts obviously the extinction
phenomenon which is occurring worldwide.
Gould (1972) has made some supplements within the framework of the
Gordon model. He pointed out that, under open access regime, extinction of
a species is possible, if (1) the species has a minimum viable population size,
or (2) the price of resource is so high that the price of the last remaining
unit of resource before extinction is still higher than the marginal harvest
costs. In sum, in the context of the Gordon model, the open access regime
significantly raises the extinction risk of exploited species, while the sole
owner regime and the equivalent profit maximization behavior never leads to
species extinction.
5.2 The Clark model
The Gordon model is a purely static model, and has therefore neglected an
important factor that significantly influences the use of renewable resources,
namely, the time factor. In his pioneer work Clark (1973, 1976) demonstrated
that a dynamic analysis will change the basic conclusion of the Gordon model,
so that exploitation to extinction may, under certain biological and economic
conditions, appear as the optimal policy, even to the profit-maximizing sole
resource owner.
Now let´s demonstrate the Clark model by the use of a version of the
model suggested by Pearce and Turner (1990). Clark suggested that it is
plausible to assume that the sole owner of the resource prefers present to
future revenues, and would try to discount future revenues at certain rate
and maximize the present value of the profits derived from his harvest yields.
The profit maximization behavior of the resource owner can be formulated
as follows9 :
Z ∞
Max
[p − c(x)] he−δt dt
0
·
s.t. x = F (x) − h, x(0) given
9
(5.1)
For simplicity, the following variables are assumed to be subscripted for time throughout this chapter.
60
where x: resource stock level
p: constant unit price of resource
c(x): unit harvest cost when resource stock level=x with c0 (x) < 0
h: harvest rate
δ: instantaneous rate of discount
F (x): natural reproduction rate of the resource stock with F (x) ≥ 0
and F 00 (x) < 0.
The following optimal condition for the optimization problem can be then
derived:
F 0 (x) −
c0 (x)F (x)
= δ.
p − c(x)
(5.2)
If we define
R = [p − c(x)] F (x) = [p − c(x)] h
(5.3)
in a stationary state, R can then be interpreted as the sustainable rent at
population level x. Equation (5.2) can be rewritten as
1 dR
·
= p − c(x)
δ dx
(5.4)
which states that the marginal profit derived from an increase in the current
harvest must equal the present value of the marginal future loss when x is
optimal.
According to equation (5.4), it can be shown that the optimal resource
stock level is lower, the higher the discount rate and the resource price, and
the lower is the cost per unit harvest. In addition, Clark (1976) proved that
a zero equilibrium population level would be optimal, if an immediate profit
can be made from harvesting the last remaining individual of the population,
and the discount rate is more than two times as large as the growth rate of the
endangered population. Moreover, the open access regime is only a special
case of the Clark model, namely, in which the sole owner adopts an infinite
discount rate and thereby sets a zero value on future revenues. Therefore,
an extremely high discount rate implies also indefinite property rights.
It follows, within the framework of the Clark model, that the combination of a high price-cost ratio, high discount rate, low growth rate of species
and overexploitation that results from the former three factors are the fundamental causes of species extinction. Its policy implication is straightforward.
61
While the growth rate of species is biologically given and not easy to be increased, we may manipulate the three variables resource price, harvest cost
and discount rate to prevent from the extinction of species. Means such as
building definite property rights (reducing the discount rate), prohibitions
of wildlife trade (reducing the demand for resource and thereby reducing
resource price) and criminalisation of hunting (raising the harvest cost) are
therefore frequently recommended as the necessary and adequate policies for
species conservation.
5.3 The Swanson model
While the Clark model can certainly be applied to some cases of species
extinction resulted from overexploitation, it is evident that, instead of overexploitation, habitat degradation and conversion, such as conversion of rain
forests to ranches, is the only cause in most cases that results in mass extinction, or the so called biodiversity loss. To remedy this deficiency, Swanson
(1994) developed a generalized model that explains extinction as a human
decision process of the naturally biological resources to retain in the assets
portfolio. As a part of the human assets portfolio, wild biological resources
must generate high enough flows of benefits to compete with the man-made
assets. Otherwise, they will be converted to different forms of man-made
assets until the returns between various assets are equilibrated. The different rates of return between assets, as the Swanson model indicates, are the
fundamental forces which drive numerous species to extinction.
Compared to the Gordon and the Clark model, the Swanson model is
characterized by two special features. First, it introduces an important factor
in the model, namely, the so called base resources (habitats in natural state,
water, etc.) on which the subsistence of all wild species depend,10 and thereby
endogenizes the allocation decision of base resources to wild species, while the
Gordon and the Clark model are in principle based on the fishery economics,
and thus assumed implicitly that base resources, namely the sea in the case
of the fishery, are exogenously given and costless. Next, it asserts that many
wild species require not only base resources, but also an another important
form of ancillary resource, namely, the so called management services, which
are mainly used for institution-building, such as the establishment of property
rights. The Swanson model endogenizes thereby the investment decision
process about the amount of management services which are allocated to
10
For example, once a piece of rain forest is converted into cattle farms, almost all the
wild species on this land will not survive. Whether retaining a piece of rain forest in its
natural state, is a decision made by human society or land owner.
62
some given species. These features imply that the flow of benefits derived
from a given species depends on three factors: harvest rate, level of base
resources and level of management services.
Based on the Clark model, the bioeconomic model of species extinction
can be modified to investigate the important role of the base resources for
the survival of species. To do this, the Swanson model can be developed step
by step as follows. First, a new variable R is introduced in the model which
represents the level of base resources retained for the survival of biological
communities. The traditional logistic growth function of species which depends only on resource stock level, F (x), must be revised to adopt the new
variable R, so that we have a new logistic growth function of species F (x; R)
which depends on both the resource stock level x and the amount of base
resources R allocated to given species. This function implies that an increase
in R would lead to an upward shift of the natural growth function of wild
species, and vice versa, as shown in Figure 5.2. Such an upward shift will raise
the overall growth potential of wild species at every population level. However, the allocation of base resources to wild species, for example returning
a cattle farm to wilderness, causes the resource owner to incur opportunity
cost which can be expressed as the potential flow of benefits available from
the best alternative use of the base resources. For each period, the incurred
opportunity costs will be those resulted from the market value of base resources, ρR R, multiplied by the interest rate, r. Together with the gross
profits received from harvesting target species, the maximization behavior of
the resource owner can be formulated as follows:
Max
Z
0
∞
[ph − c(x)h − rρR R] e−rt dt
·
s.t. x = F (x; R) − h
(5.5)
where x: resource stock level
p: constant unit price of resource
c(x): unit harvest cost when resource stock level=x with c0 (x) < 0
h: harvest rate
r: interest rate
R: amount of base resources
ρR : unit price of base resources
F (x; R): natural reproduction rate of the resource stock with
F (x; R) ≥ 0, Fxx (x; R) < 0 and FR (x; R) > 0.
63
This is a dynamic model with one state variable x and two control variable
h and R. With the inclusion of the additional control variable R in the model,
it yields accordingly an additional first-order condition regarding the optimal
investment in base resources:
R∗ :
µ · FR
=r
ρR
(5.6)
where µ denotes the shadow price of the resource stock in the steady state.
This condition states that the marginal value product derived from an allocation of base resources to production of species must equal the incurred
marginal costs in the steady state, or in other words, the resource owner
will retain base resources for growth of the target species only to the extent
that the species can offer a competitive rate of return from retaining base
resources.
By the application of the optimal condition, let´s examine what would
happen if the price of the target species rises. On the one hand, this would
raise the incentive to exploit species, but on the other hand this implies that
the shadow price of species in the steady state µ would increase. An increase
in µ would raise the relative rate of return for this species, as shown by the left
hand side of (5.6), and thereby raise the optimal level of investment in base
resources for the species, as shown in Figure 5.1. More base resources would
be then made available to the species because of its increased investmentworthiness. This would lead to an upward shift of the growth function of
species, as demonstrated in Figure 5.2, and thereby decrease the extinction
risk of the species. At the same time, numerous other species would benefit
from the increased allocation of base resources, a situation in which biodiversity is better conserved than ever. In the reverse case, if the price of species
falls, a smaller amount of base resources would be allocated to the species as
a result of its investment-unworthiness. The extinction risk of the harvested
species and of all the other species would therefore rise, a situation in which
biodiversity is worse conserved than ever.
Therefore, while overexploitation is one of the primary factors contributing to species extinction in some cases, most extinction cases of numerous
unknown species are in fact the direct result of the habitat degradation and
conversion which can be regarded as the active removal of base resources by
resource owner as a result of the investment-unworthiness of these species.
Instead of overexploitation, it is active conversion of base resources into manmade assets that threatens most species, or biodiversity of the world today.
64
Returns on
assets
Marginal value product
of base resources
Rate of return on
capital (r)
R*
R*’
Base resources (R)
Figure 5.1 Optimal allocation of base resources. Source: Swanson (1994),
p.63.
65
Growth rate
Growth function
(same species/increased base resources)
Resource stock
Figure 5.2 Upward shift of the growth function of the species resulted from
increased base resources. Source: Swanson (1994), p. 61.
Furthermore, as Swanson indicated, the management factor is just as
important for the survival of species as base resources. Even when substantial base resources have been allocated to certain species, management
services, especially the establishment of property rights, are still required for
their maintenance and protection. To demonstrate the importance of management in the decision process of the resource use, Swanson introduced a
new variable M in the model which represents the level of the management
services devoted by resource owner for protecting and managing renewable
resources. The previous logistic growth function of species F (x; R) which
depends on both the resource stock level x and the amount of base resources
R must be revised to adopt the new variable M, so that we have now a new
logistic growth function of species F (x; R, M) which depends additionally
on, except the resource stock level x and the amount of base resources, the
level of the management services. And just like the influence of R on growth
function of species, an increase in M would raise the overall growth potential
of species, and vice versa. After considering the opportunity cost incurred
by the devotion of management services, the Swanson model can be modified
again as follows:11
11
·
In the original model, Swanson used an equation of motion for resource stock x =
66
Max
Z
∞
0
[ph − c(x)h − rρM M − rρR R] e−rt dt
·
s.t. x = F (x; R, M) − h
(5.7)
where M: level of management services
ρM : unit price of management services
F (x; R, M): natural reproduction rate of the resource stock with
F (x; R, M) ≥ 0, Fxx < 0, FR > 0 and FM > 0.
This is a dynamic model with one state variable x and three control
variable h, R and M. With the inclusion of the additional control variable
M in the model, it yields accordingly an additional first-order condition
regarding the optimal investment in management services:
M∗ :
µ · FM
= r.
ρM
(5.8)
This condition states that the marginal value product derived from allocation
of management services to production of species must equal the incurred
marginal costs in the steady state, or in other words, the resource owner will
invest necessary management services for the growth of the target species
only to the extent that the species can offer a competitive rate of return
from this investment.
What condition (5.8) implies for the optimal allocation of management
services and for the extinction risk of species, is in principle the same as the
condition (5.6) does. In short, an increase in resource price would raise the
relative rate of return for this species and thereby raise the optimal level
of management services for the species. More management services would
be then made available to the species because of its increased investmentworthiness. This would lead to an upward shift of the growth function of
species and thereby decrease the extinction risk of the species. Simultaneously, many other species would also benefit from the increased allocation of
management services, so that biodiversity is better conserved than before.
On the contrary, if the price of the harvested species falls, a smaller amount
F (x) − h and a utility function S(h; R, M ). However, we make here a modification by
·
using the equation of motion for resource stock x = F (x; R, M ) − h and the revenue
function ph that accord with what are used in (5.5).
67
of management services would be allocated to the species as a result of its
investment-unworthiness. The extinction risk of the harvested species and
of all the other species existing in the same habitat would therefore rise, a
situation in which biodiversity is worse conserved than ever. It follows that
the usual open access state of poorly managed resources is in fact the result
of the disinvestment in management services by resource owner as a result
of the investment-unworthiness of these species.
To sum up, within the framework of the Swanson model, the only fundamental cause driving the extinction process is that the species is not worth
retaining in the assets-portfolio of the resource owner. If the species can generate only a relative low rate of return and therefore cannot compete with
the man-made assets, the resource owner would have no incentive to invest in
the growth potential of the species. Many species are depleted to extinction,
because the necessary base resources for their survival are converted to other
forms of use which afford a higher rate of return. Some other species are
subjected to inadequate property rights, such as open access regime, because
they are not worth investing management services. Finally, some species
with extremely high price-cost ratios and low growth rate are endangered,
since the owner is willing to harvest the total population to cash in them
and invest the funds in other forms of assets which have a higher growth
rate than the species, as the Clark model demonstrated. Therefore, in total
there are three alternative routes to extinction: base resource conversion,
inadequate management and overexploitation. However, they are not really
fundamental, but merely the proximate causes for observed extinction. In
other words, they are merely different ´routes´ to extinction rather than the
fundamental cause resulting in extinction. The only fundamental cause of
extinction is the investment-unworthiness of the species. In this connection
the Swanson model affords a more generalized explanation regarding the loss
of biodiversity than the Clark model.
The policy implications of the Swanson model are as follows. To remedy
the loss of biodiversity resulted from the investment-unworthiness of wild
biological resources, human society needs to develop effective approaches
to capture the values of the wild species given current human preferences,
including consumptive and non-consumptive values. With regard to the consumptive value, as Swanson asserted, this would imply in many cases that
we should try to increase the difference between resource price and harvest
costs, rather than to minimize or even to eliminate it. In other words, the
use of wild biological resources could be an effective instrument for conserving biodiversity and should be promoted. Swanson maintained therefore that
68
conservation policies of demand destruction and supply criminalization which
try to minimize the price-cost ratios, such as the prohibition of wildlife trade
and hunting, are misleading. With reference to the protected area policies,
the Swanson model implies that the ´fence and police´ policy prevailing in
many protected areas may be misguided and should be reconsidered. For the
sake of conserving biodiversity, we should seriously consider the question,
whether wild biological resources in protected areas should be utilized to a
greater extent than today. According to the Swanson model, this may in
the long run raise the incentive of governments and local residents to invest
more resources in the management of protected areas or even to expand the
existing protected area networks, so long as the use of biological resources
can afford a high enough rate of return from this investment.
The example of South Africa may support the rationale of the Swanson
model. As a result of the promotion of the sustainable use of wildlife, especially of safari hunting and tourism, many farmers ceased cattle farming,
returned their farms to natural habitats that are suitable to the survival of
wildlife (increasing base resources for wild species, in the terminology of the
Swanson model), and thereby built private reserves, because wildlife use is
in many cases a form of land use that is more profitable than the traditional
farming. The management efforts and devotion of land for wildlife subsistence invested by private individuals and enterprises has resulted in steady
increase in wildlife populations (Grootenhuis and Prins, 2000) and increase
in privately protected habitat area which is even greater than the total land
area under the control of the National Park´s Board (Hearne and Mckenzie,
2000). This contributed not only to the conservation of wildlife, but also to
the conservation of biodiversity.
5.4 Concluding remarks
The basic theme of the sustainable use approach is that, through creation
of economic incentives, people will invest more resources in management of
protected areas and thereby reduce simultaneously the illegal exploitation of
renewable resources and the amount of wildlands which would be converted
to land use of non-conservation alternatives. It is clear that, with the extensive explanation for the fundamental cause of biodiversity loss, the Swanson
model provides an excellent theoretical foundation for the promotion of the
sustainable use strategy as an instrument of biodiversity conservation, and
is certainly a feasible starting point for further theoretical modeling. However, the Swanson model is still a rough model in the sense that, except
asserting the importance of endogenizing the factors of land use competition
69
and management, it did not deliberately investigate the dynamic interaction between the resource stock, base resources, management and harvest
when considering the resource use and management issues. With its simple
analysis, it suggested that a higher resource price/harvest cost ratio will enhance the investment-worthiness of the resource, and thereby contribute to
the growth potential of the resource stock. However, it neglected the effect
that an increased resource price/harvest cost ratio will simultaneously enhance the incentive of the resource owner to exploit resources and to cash
in, as demonstrated by the Clark model. Therefore, whether an increased
resource price/harvest cost ratio will lead to a net increase or decrease of the
equilibrium resource stock, is a question that needs to be addressed. Finally,
the Swanson model also did not consider the problem of illegal harvest which
plays an important role in the resource management issues of the real world.
Hence, more deliberate modeling is needed.
The sustainable use approach could be applied both inside and outside
protected areas. The primary objective and task of this dissertation is to investigate whether and under which conditions the sustainable use approach
is a feasible instrument for biodiversity conservation in existing protected
areas. While the positive effect of the sustainable use approach on land use
decisions outside existing protected areas is in principle uncertain and cannot
be measured precisely, the use approach will result in direct impact on the
population levels of the harvested species. Considering the fact that overexploitation in existing protected areas in exchange for future, unpredictable
conservation benefits outside protected areas will hardly be accepted, the
use approach should be applied in protected areas with particular caution.
Hence, rather than concentrating on issues of land use competition, as Swanson did, we will confine the dissertation to addressing the related harvest
and management issues of the sustainable use strategy applied in protected
areas. Based on the findings of the Clark and the Swanson model, several
new models will be developed in the next chapters.
70
Chapter 6
Use of renewable resources, poaching
and anti-poaching: a simple
bioeconomic model with one state
variable and two control variables
6.1 Introduction
During the last two decades, one of the most serious controversy regarding
the conservation issue involved the following questions: is sustainable use of
renewable resources possible and appropriate in and around protected areas?
Could such a strategy really help to protect biodiversity more effectively, or
just on the contrary, it would degrade biodiversity more rapidly? To answer
these questions, two often conflicting approaches, the preservation and the
sustainable use approach, dominated the debate, as discussed in chapter 4.
In addition to the two somewhat extreme positions, many conservationists
may agree that what is the best strategy will depend on a variety of biological, economic and social conditions at the specific site. There are seemingly
no clear-cut answers, but, based on case studies arising from different circumstances, some general patterns and themes emerge (see Freese, 1997; Bennett
and Robinson, 2000).12 These case-study-based observations regarding harvest of renewable resources furnish the conceptual framework for generating
and testing hypotheses and represent a crucial first step. Nonetheless, they
are mostly fragmentary and lack theoretically consistent foundation. There
are relatively few rigorously theoretical analysis to address the implication
of the sustainable use strategy for biodiversity conservation. The existing
analytical models were mainly developed by biologists and thereby based
principally on biological points of view.13 Therefore, it is necessary to develop models which incorporate economic rationale into the existing models
of harvest.
In this context, Skonhoft and Solstad (1996) analyzed the conflict between illegal hunting and wildlife management within east Africa´s national
12
A well-known example is that species with low intrinsic rates of population increase
tend to be more vulnerable to harvest. However, this is not the only factor which would
affect vulnerability of species (Bennett and Robinson, 2000).
13
See, for example, McCullough (1984) and Robinson (2000).
71
parks in a bioeconomic model with one state variable (wildlife stock) and
one control variable (anti-poaching effort). They concluded that, if the local
people have the legal rights to reap the benefits of wildlife, more wildlife
stock would be safeguarded, because people have the incentive to invest in
wildlife stock. Hence, they argued for a shift of conservation policy from
protectionism toward utilization of wildlife.
The Skonhoft and Solstad Model is in principle a variant of the Clark´s
bioeconomic model (Clark, 1973; Clark and Munro, 1975). In this chapter we will further develop and adequately modify the existing models, in
particular the Clark model and the Swanson model, and thereby to investigate deliberately the dynamic development process of resource stock, management efforts, harvest and poaching activity, under the assumption that
people are allowed to use renewable resources in protected areas. To do this,
a bioeconomic model with one state variable (resource stock) and two control
variables (harvest rate, management effort) will be first constructed on the
basis of the traditional bioeconomic model and the optimal control theory.
An extended model which based on the simple model of this chapter will
be then developed in next chapter. What we are concerned about is the
question, whether and under which conditions the use of renewable resources
in protected areas could really help to induce more human investment in
biodiversity conservation and thereby protect them more effectively, as the
optimistic conclusion of the Skonhoft and Solstad Model , or just on the contrary, this would lead to species extinction more rapidly. The models may
help to offer more detailed answers for these questions and therefore more
arguments for the judgement of present conservation policies. The dynamic
interaction between control variables and state variable will be investigated
deliberately. The policy implications of the models will also be discussed.
6.2 The model
In this section a nonlinear bioeconomic model with one state variable (resource stock) and two control variables (harvest rate, management effort) is
employed on the basis of the traditional bioeconomic model and the optimal
control theory. The necessary conditions for the optimal policy are also derived. The uniqueness and stability properties of the steady state solution of
the model, and the relevant phase diagrams will be presented in the following
sections, respectively.
As ever, poaching14 is a serious problem so that anti-poaching become
14
In many cases, poaching often means the illegal hunting of wildlife. However, for con-
72
the most important management issue of many, if not most, park authorities, especially in developing countries. To investigate the dynamic development process of resource stock, management, harvest and poaching activity
resulted from the application of the sustainable use approach in protected
areas, the present analysis will commence with addressing the economic decision faced by poachers. Milner-Gulland and Leader-Williams (1992), based
on experiences of Luangwa Valley in Zambia, addressed the link between the
economic decision faced by poachers, law enforcement and poaching activity.
They found that several factors will influence the extent to which wildlife are
illegally exploited. A greater anti-poaching effort and the following greater
detection rate plays the most crucial role in reducing poaching activity. In
addition, higher penalties and higher opportunity costs of poaching as a job
also work in the same direction. On the other hand, a greater stock of wildlife
will induce more poaching activity. In what follows, we will transform the
analysis of Milner-Gulland and Leader-Williams into a formal poaching function which is prepared for the later construction of models.
Let p(Y ) represents the inverse black-market demand function of illegally
exploited resources, where Y is the poached resource stock and pY (Y ) < 0.
M(X, E) denotes the costs expended by poachers in exploiting per unit of
resource stock, where X is the stock of renewable resources in a given protected area and E is the management effort of park authorities. It is assumed
that a larger resource stock means a greater resource density, and thereby reduces the poaching costs, i.e., MX < 0. On the contrary, more management
efforts increase the difficulties of poaching activity and also the poaching
costs, i.e., ME > 0. In addition, a detection function q(E) is here introduced
to represent the probability that poaching activity is successfully detected,
where qE > 0 stresses the fact that more management efforts strengthen law
enforcement and increase the detection rate. Finally, the severity of penalty
given once the poacher has been caught, the variable V representing the fine
for per unit poached resource stock, can also have an important effect on
poaching activity.
Poachers do not own the legal property rights of resources, or in other
words, natural resources are for poachers in an open access state. Therefore,
it is reasonable for poachers to exploit resources as possible as they can, i.e.,
rather than to base on intertemporal profit maximization, their behavior is
driven by the motive of exploiting resources until rent dissipates, as Gordon
venience, poaching will be used to mean the illegal exploitation of any kinds of renewable
resources throughout this dissertation, including hunting of wildlife, fishing, logging and
gathering of plants and of other nontimber products.
73
(1954) suggested. We define
π = [1 − q(E)] [p(Y ) − M(X, E)] Y − q(E)Y V
(6.1)
where π is the expected profit function of total poachers, which is the difference between expected revenues and expected costs. Expected revenues
are given by the left term and are determined by multiplying successful harvest, [1 − q(E)] Y , by gross profits per unit of harvest, [p(Y ) − M(X, E)].
Expected costs derived from law enforcement are given by multiplying detection rate, total harvest and fine per unit of harvest.
Poachers are attracted to exploit resources in protected areas in response
to potential profits, and exploitation proceeds until their effort earns its opportunity cost. Their behavior determines the collectively optimal poaching
level. The problem is straightforward and easily solved by setting expected
profit function equal to zero. It yields
[1 − q(E)] [p(Y ) − M(X, E)] − q(E)V = 0.
(6.2)
This relationship defines an implicit function Y (X, E, V ). The implicit function describes the behavior of poachers associated with poaching, given a
resource stock, management effort of park authorities and fine level. It can
be analyzed by the following procedure.
First, we differentiate the implicit equation (6.2) with E. After a routine
calculation, it yields
µ
¶
1
dY
qE (E)(p(Y ) − M(X, E) + V )
=
(6.3)
ME (X, E) +
,
1 − q(E)
dE
pY (Y )
and from given assumptions it is obvious that
dY
<0
dE
(6.4)
under the fact that p(Y ) must be greater than M(X, E).15 Again, we differentiate the implicit equation (6.2) with X and V , respectively, and it yields
dY
MX (X, E)
=
>0
dX
pY (Y )
(6.5)
dY
q(E)
=
<0
dV
pY (Y ) [1 − q(E)]
(6.6)
and
15
Otherwise, poaching activity will not exist.
74
under given assumptions.
The results (6.4) and (6.5) show that, other things being equal, a higher
level of management effort of park authorities will reduce the poaching. This
is because more management effort increase the unit exploitation costs of
poachers and the detection rate. On the other hand, a higher resource stock
will increase the poaching, as a result that higher resource stock reduces the
unit exploitation costs of poachers. These results will be applied later for
constructing a new bioeconomic model. In addition, a higher fine level will
increase the exploitation costs of poachers, and thereby reduce the poaching.
We consider now the following scenario: certain people or certain organizations16 are given the legal rights to exploit renewable resources in specific
protected areas or in buffer areas around protected areas, and they are authorized to manage natural resources and human activities in those areas.
Let X = X(t) represent the biomass of renewable resources at time t. The
population dynamics of the resource is assumed to be a pure compensation
logistic function of the biomass level, i.e., the natural net growth function of
the resource stock, F (X),17 is a strictly concave function of the biomass level
with the following properties
F (X) > 0 for 0 < X < X,
F (0) = F (X) = 0,
F 00 (X) < 0,
(6.7)
where X represents the carrying capacity of the environment.
Harvest rate h(t) denotes the resource stock which is harvested by the
resource owner at time t. The poaching function Y (E(t))18 represents the
resource stock which is illegally exploited by poachers at time t, where E(t)
is the management effort invested by resource owner and, as equation (6.4)
showed,
Y 0 (E) < 0
(6.8)
asserts the fact that a higher level of management effort will reduce the
poached quantity of resources. In addition, it is assumed that
Y 00 (E) > 0
16
(6.9)
They might be individuals, local communities, private enterprises, park authorities,
central governments or non-governmental conservation organizations.
17
For convenience, the time notation will henceforth be omitted wherever possible.
18
Rather than Y = Y (X, E), it is assumed here that the poaching rate depends only
on management effort level. A complete poaching function will be introduced in the next
chapter for modifying the present model.
75
which explains the phenomenon of diminishing returns of management effort.
Both h and E are nonnegative.
In addition to natural factors, the population dynamics of the resource
is also subject to human interference. When legal harvest and poaching are
introduced, the equation of motion for an exploited resource stock X can be
stated as
·
X≡
dX
= F (X) − Y (E) − h.
dt
(6.10)
Now let us consider the functional form of the utility function. It has
long been recognized that wildlands and renewable resources may produce a
range of valued products and services, and in turn provide benefits to human
society (Barbier, 1992; Aylward,1992). Barbier (1992) suggested that the
total economic value of wildlife and wildlands may comprise use value, which
includes direct value, indirect value and option value, and non-use value. Under this definition the direct use value may includes benefits from harvesting,
recreation, tourism, genetic material, education etc. Hence, as a function of
harvest, the gross harvest profit function U(h) usually used in the traditional
bioeconomic model captures obviously only part of the total economic value
and even only part of the direct use value, namely the consumptive value in
the narrow sense. We use here the term ´consumptive value´ to refer to the
direct use values provided by directly harvesting renewable resources. For
convenience of analysis, we divide the economic value of renewable resources
into consumptive value and non-consumptive value which refers to all forms
of values that can not be classified as consumptive value. In other words, the
resource stock remains in principle intact when it provides non-consumptive
value.19 In what follows, a new utility function
U(h, X) = U(h) + V (X)
(6.11)
which consider simultaneously the consumptive value and non-consumptive
value is here introduced. The utility function is assumed to be additively
separable for the sake of technical simplicity. The gross harvest profit function U (h), which represents the total harvest revenues less the total harvest
costs and comprises the consumptive value component of the utility function, depends on the harvest rate and is strictly increasing and concave in h,
19
In some cases, for example usually happened in tourism, even non-consumptive use
has potentially negative impacts on resource stock, although these impacts often can not
be observed directly.
76
namely
U 0 (h) > 0,
U 00 (h) < 0.
(6.12)
Under the assumption of U 0 (h) > 0, we can avoid the possibility of saturation in harvest for extremely large harvest level and the relevant technical
complexity (Boyce, 1995). The concavity of U (h) derives from the both assumptions of decreasing marginal revenues in harvest and increasing marginal
harvest costs. Next, the non-consumptive utility function V (X) which comprises the non-consumptive value component of the utility function depends
on the level of the resource stock. We assume that V (X) can be measured
in monetary terms and is strictly increasing and concave in X:20
V 0 (X) > 0,
V 00 (X) < 0.
(6.13)
It is here worth noting that, due to the public good characteristics of many of
the non-consumptive values, the introduction of non-consumptive value does
not mean that resource owner is able to appropriate the total economic value
of renewable resources. In general, international conservation organizations
may tend to take all values into consideration. National governments may,
to certain extent, fail to capture part of the ecological function, existence
and option value. And for private ranges or parks, it is likely that ecological
function, option and existence value would not be taken into account in the
process of decision-making.
The resource-management problem of the resource owner is to choose the
harvest rate and the management effort to maximize the net present profits
derived from protecting and harvesting the resource, subject to the constraint
of biological dynamics. This task can be formally expressed as
Z ∞
[U (h) + V (X) − C(E)] e−rt dt
Max
0
·
s.t. X = F (X) − Y (E) − h
(6.14)
where C(E) and r denote the management cost function and the instantaneous discount rate, respectively. The management costs function owns the
properties that
C 0 (E) > 0,
C 00 (E) > 0,
20
(6.15)
For example, one can imagine that the more spectacular the wildlife population in a
national park is, the more tourists will be induced and in turn the more income generated.
However, for per one unit of additional resource stock, the marginal utility will be decreasing. The same statement can be applied to cases in reference to other non-consumptive
values, such as ecological functions, option value and existence value.
77
which implies that C(E) is strictly increasing and convex in E.21 The discount rate is assumed to be constant and 0 < r < 1.
Next, the problem is analyzed by the application of the maximum principle. The current-value Hamiltonian of our case is
H = U(h) + V (X) − C(E) + λ [F (X) − Y (E) − h]
(6.16)
where λ is the current-value costate variable associated with the state variable X which gives the imputed marginal value of the resource stock. Assuming an interior solution, the first-order necessary conditions describing
the optimization problem are given by equations (6.17)-(6.19) together with
(6.10):
·
∂H
= U 0 (h) − λ = 0
∂h
(6.17)
∂H
= −C 0 (E) − λY 0 (E) = 0
∂E
(6.18)
λ = rλ −
∂H
= [r − F 0 (X)] λ − V 0 (X).
∂X
(6.19)
Under given assumptions, the Hamiltonian is apparently concave in the
state variables X and in the control variables h, E. The second order
conditions are therefore satisfied. Along the optimal trajectory, the equation
(6.17) means that the imputed marginal value, or the shadow price of an
extra resource stock, λ, must be equal to the marginal profit of harvesting
renewable resources, U 0 (h). The equation (6.18) implies that the marginal
gain derived from an extra unit of management effort, −λY 0 (E), must equal
the marginal cost of management input, C 0 (E). Finally, equation (6.19)
asserts that the summation of the change rate of the shadow price of the
·
resource stock, λ, the marginal non-consumptive value of the resource stock,
V 0 (X), and the gain derived from the marginal growth rate of the resource
stock, F 0 (X)λ, must equal the opportunity cost when resource owner goes
on to keep one unit of resource stock, rλ. The term rλ represents the forgone
´interest´.
21
According to author´s personal experiences in Taiwan, the assumptions of the concavity of U (h) and the convexity of C(E) are reasonable, because the market of renewable
resources is usually relatively small and the market of manpower for conservation, especially at lower level, is typically localized.
78
6.3 Uniqueness of the steady state solution
In the following sections, we will investigate the properties of the steady
state solution of the dynamic problem. Suppose that a steady state exists,
the steady state solution can be determined under the conditions that the
·
resource stock and the shadow price of the resource are constant, i.e., X =
·
λ = 0. We can first verify the uniqueness of the steady state solution by
proposition 1.
Proposition 1 Under given assumptions with regard to the poaching function Y (E), the utility function U (h) + V (X), the management cost function
C(E) and the natural dynamics of the resource stock F (X), the dynamic
system (6.10) and (6.19) possess an unique steady state solution.
·
·
·
Proof. In steady state, X = λ = 0. First, λ = 0 gives
λF 0 (X) + V 0 (X) = rλ
by applying the equation (6.19). Under given assumptions, we know that
both F 0 (X) and V 0 (X) are monotonously decreasing in X. Therefore, the
function λF 0 (X) +V 0 (X) is also monotonously decreasing in X. It is obvious
that one unique X ∗ exists which satisfying the condition λF 0 (X ∗ )+V 0 (X ∗ ) =
rλ where X ∗ denotes the steady state solution of the resource stock X.
From (6.18), we know that
C 0 (E) = −U 0 (h)Y 0 (E).
(6.20)
Consider E as an implicit function of h. After substituting E = E(h) for
equation (6.20), and differentiating (6.20) with h, it yields
[C 00 (E) + U 0 (h)Y 00 (E)]
dE
= −U 00 (h)Y 0 (E).
dh
(6.21)
Under given assumptions, it can be easily shown that
dE
< 0.
dh
·
Applying E = E(h) for the case when X = 0, it implies that
F (X ∗ ) − Y (E(h)) − h = 0
79
(6.22)
in steady state. By defining
Ω(h) = F (X ∗ ) − Y (E(h)) − h
(6.23)
and differentiating Ω(h) with h, we obtain
Ω0 (h) = −Y 0 (E)E 0 (h) − 1 < 0,
(6.24)
because E 0 (h) < 0, as equation (6.22) shows. Therefore, Ω(h) = 0 possesses
an unique solution h∗ . Since E = E(h), E 0 (h) < 0, λ = U 0 (h) and U 00 (h) < 0,
it is obvious that both E(h) and U 0 (h) are monotonously decreasing function of h. It follows that E ∗ = E(h∗ ) and λ∗ = U 0 (h∗ ) are also unique.
Consequently, the steady state solution (X ∗ , λ∗ , h∗ , E ∗ ) is unique.
6.4 Stability of the steady state solution
Proposition 2 Under given assumptions with regard to the poaching function Y (E), the utility function U (h) + V (X), the management cost function
C(E) and the natural dynamics of the resource stock F (X), the unique steady
state solution of the dynamic system (6.10) and (6.19) is saddle point stable,
if F 0 (X) [r − F 0 (X)] − [−F 00 (X)λ − V 00 (X)] [−Y 0 (E)E 0 (λ) − h0 (λ)] < 0.
Proof. By use of the result λ = U 0 (h) of the equation (6.17), we can
define h as an implicit function of λ. Since
dh dλ
=1
dλ dh
or
dh 00
U (h) = 1
dλ
,it can be easily verified that
dh
< 0.
dλ
(6.25)
Again, by applying the result −C 0 (E) = λY 0 (E) of the equation (6.18), we
can define E as an implicit function of λ. Since
dE dλ
=1
dλ dE
80
and
dλ
−C 00 (E)Y 0 (E) + C 0 (E)Y 00 (E)
=
> 0,
dE
Y 00 (E)
it yields
dE
> 0.
dλ
(6.26)
After substituting h(λ) for h and E(λ) for E in equation (6.10), we observe
now the following dynamic system of equations:
·
X = F (X) − Y (E(λ)) − h(λ)
(6.27)
·
λ = [r − F 0 (X)] λ.
·
·
For the purpose of later analysis, we differentiate X and λ with X and λ,
respectively. This yields
·
∂X
= F 0 (X)
∂X
(6.28)
·
∂X
= −Y 0 (E)E 0 (λ) − h0 (λ)
∂λ
·
∂λ
= −F 00 (X)λ − V 00 (X)
∂X
·
∂λ
= r − F 0 (X).
∂λ
A Taylor expansion of the dynamic system (6.27) at (X ∗ , λ∗ ) gives then
à · ! µ
¶µ
¶
X
F 0 (X ∗ )
X − X∗
−Y 0 (E)E 0 (λ∗ ) − h0 (λ∗ )
=
·
−F 00 (X ∗ )λ∗ − V 00 (X ∗ )
λ − λ∗
r − F 0 (X ∗ )
λ
(6.29)
·
·
·
·
X ∂X
∂λ
, after the results of (6.28) are applied for
substituting ∂∂X
, ∂λ , ∂X
and ∂∂λλ
√
2
in (6.29). Hence, the two eigenvalues r± r2 −4A can be easily derived where
A = F 0 (X ∗ ) [r − F 0 (X ∗ )] − [−F 00 (X ∗ )λ∗ − V 00 (X ∗ )] [−Y 0 (E)E 0 (λ∗ ) − h0 (λ∗ )].
If A < 0, it is obvious that one of the eigenvalues is positive, and the other
one is negative. Therefore, the steady state solution (X ∗ , λ∗ ) is a saddle
point.
81
The saddle-point stability property of the steady state solution means
that, given the initial value of the resource stock, it will always be possible
for the resource owner to choose an optimal initial value of the harvest rate
and the management effort which are on the stable trajectories that converge
to the steady state equilibrium of the dynamic system.
6.5 Phase diagram analysis
6.5.1 Phase diagram (X, h)
In this section, we conduct a phase diagram analysis to investigate the
qualitative properties of the solution of the dynamic problem. Let us observe
first the phase diagram on the (X, h) plane.
First, we differentiate the necessary condition (6.17) with t and obtain
·
·
λ = U 00 (h)h.
(6.30)
Applying the results of equations (6.17) and (6.30) in (6.19), it yields that
·
h=
1
[U 0 (r − F 0 ) − V 0 ] .
U 00
·
(6.31)
·
By means of setting h = 0, the isocline h = 0 is the curve
U 0 (r − F 0 ) − V 0 = 0
(6.32)
as a result of the assumption U 00 < 0. A total differentiation of (6.32) gives
then
U 0 F 00 + V 00
dh
| · = 00
dX h=0 U (r − F 0 )
(6.33)
Under given assumptions, it is clear that the term U 0 F 00 +V 00 is negative. The
denominator term U 00 (r − F 0 ) is positive when X is extremely small, and it is
negative when X is big enough because F 0 is monotonously decreasing and
W 0 monotonously increasing in X. Accordingly, the gradient of the isocline
·
h = 0 on the (X, h) plane is negative on the left of Xr and positive on the
right of Xr , where Xr denotes the resource stock so that F 0 (Xr ) = r (see
Figure 6.1). It is worth noting that, in our case, the equilibrium resource
stock must be greater than Xr , because of the fact that the non-consumptive
value of the resource is taken into account.
82
Similarly, by the use of (6.10), it is obvious that
·
X = F (X) − Y (E(h)) − h.
·
(6.34)
·
By setting X = 0, the isocline X = 0 is the curve
F (X) − Y (E(h)) − h = 0.
(6.35)
A total differentiating of (6.35) yields
dh
F 0 (X)
|· =
.
dX X=0 1 + Y 0 (E)E 0 (h)
(6.36)
·
It is clear that the gradient of the isocline X = 0 is strictly decreasing in X.
The two isoclines determine then the unique steady state solution (X ∗ , h∗ ).
Next, we examine the properties of the points which are not on the isoclines. As Figure 6.1 shows, the phase plane (X, h) can be divided into four
isosectors by the two isoclines. By differentiating equation (6.31) with X, it
yields
·
dh
−U 0 (h)F 00 (X) − V 00 (X)
=
<0
dX
U 00 (h)
(6.37)
under given assumptions. This means that, in the region on the left of the
·
isocline h = 0, h tends to increase, with a symbolic upward vertical pointing
·
arrow. On the contrary, in the region on the right of the isocline h = 0, h
tends to decrease, with a symbolic downward vertical pointing arrow. Again,
by differentiating equation (6.34) with h, it gives
·
dX
= −Y 0 (E)E 0 (h) − 1 < 0.
dh
(6.38)
Therefore, with a rightward horizontal pointing arrow, X tends to increase in
·
the region below the isocline X = 0. And, with a leftward horizontal pointing
·
arrow, X tends to decrease in the region above the isocline X = 0. These
results outlined above show that the equilibrium point is a saddle point. The
two solid trajectories in Figure 6.1, which denote the two stable trajectories,
converge to the equilibrium point. Corresponding to each initial resource
stock level, a unique corresponding value of harvest rate could be chosen
on the stable trajectories. Hence, the following conclusion can be drawn
83
that, on the optimal dynamic path, the resource stock and the harvest rate
increase over time, if the initial resource stock level is less than the steady
state resource stock. Contrarily, if the initial resource stock level is higher
than the steady state resource stock, the resource stock and the harvest rate
decrease over time on the optimal dynamic path. The economic meaning
of this result is clear. The more resource stock people have, the more they
would harvest without influencing the long-run survival of the resource, or in
other words, in order to reach the steady state, people would harvest more
than the steady state harvest rate when the resource is in relative abundance,
in the sense that the resource stock is greater than the steady state resource
stock. And they would harvest less than the steady state harvest rate when
the resource is relatively scarce, in the sense that the resource stock is less
than the steady state resource stock.
h
•
h=0
h*
•
X =0
X
X*
Xr
Figure 6.1 Phase diagram on the (X,h) plane.
6.5.2 Phase diagram (X, E)
We observe now the phase diagram on the (X, E) plane. A transformation
84
of the equation (6.18) yields
λY 0 (E) = −C 0 (E).
(6.39)
Differentiating (6.39) with t and after a few transformations, we obtain
·
[−C 00 (E) − λY 00 (E)] E
λ=
.
Y 0 (E)
·
(6.40)
·
By substituting this result for λ in equation (6.19) and substituting U 0 (h)
for λ, it gives
·
E=
[[r − F 0 (X)] U 0 (h) − V 0 (X)] Y 0 (E)
.
−C 00 (E) − U 0 (h)Y 00 (E)
(6.41)
Considering h as an implicit function of E and substituting h(E) for h in
(6.10), it yields
·
X = F (X) − Y (E) − h(E).
(6.42)
We are now interested on the dynamic system (6.41) and (6.42). By means
·
·
of setting E = 0, the isocline E = 0 is the curve
[r − F 0 (X)] U 0 (h(E)) − V 0 (X) = 0
(6.43)
under given assumptions. A total differentiating of (6.43) gives
dE
U 0 (h)F 00 (X) + V 00 (X)
| · = 00
.
dX E=0 U (h)h0 (E) [r − F 0 (X)]
Under given assumptions, it is clear that the term U 0 (h)F 00 (X) + V 00 (X) is
negative. The denominator term U 00 (h)h0 (E) [r − F 0 (X)] is negative when
X is extremely small, and it is positive when X is big enough because F 0
is monotonously decreasing in X. Accordingly, the gradient of the isocline
·
E = 0 on the (X, E) plane is positive on the left of Xr and negative on the
right of Xr , where Xr denotes the resource stock so that F 0 (Xr ) = r (see
·
·
Figure 6.2). Similarly, by setting X = 0, the isocline X = 0 is the curve
F (X) − Y (E) − h(E) = 0.
(6.44)
A total differentiating of (6.44) gives
dE
F 0 (X)
|· = 0
.
dX X=0 Y (E) + h0 (E)
85
(6.45)
·
The gradient of the isocline X = 0 is negative when F 0 (X) > 0, and then
strictly increasing in X. It becomes positive when F 0 (X) < 0. The two
isoclines determine the unique steady state solution (X ∗ , E ∗ ).
Again let us examine the properties of the points which are not on the
isoclines. As Figure 6.2 shows, the phase plane (X, E) can be divided into
four isosectors by the two isoclines. By differentiating equation (6.41) with
X, it yields
·
dE
[−F 00 (X)U 0 (h) − V 00 (X)] Y 0 (E)
=
>0
dX
−C 00 (E) − U 0 (h)Y 00 (E)
(6.46)
under given assumptions. This means that, in the region on the left of the
·
isocline E = 0, E tends to decrease, with a symbolic downward vertical
pointing arrow. On the contrary, in the region on the right of the isocline
·
E = 0, E tends to increase, with a symbolic upward vertical pointing arrow.
Again, by differentiating equation (6.42) with E, it gives
·
dX
= −Y 0 (E) − h0 (E) > 0,
dE
(6.47)
as a result of h0 (E) < 0, according to (6.22). Consequently, with a leftward
horizontal pointing arrow, X tends to decrease in the region below the iso·
cline X = 0. And, with a rightward horizontal pointing arrow, X tends to
·
increase in the region above the isocline X = 0. These results outlined above
show that the equilibrium point is a saddle point. The two solid trajectories
in Figure 6.2, which denote the two stable trajectories, converge to the equilibrium point. Corresponding to each initial resource stock level, a unique
corresponding value of management effort could be chosen on the stable trajectories. Hence, the following conclusion can be drawn that, on the optimal
dynamic path, the resource stock increases over time while the management
effort input decreases, if the initial resource stock level is less than the steady
state resource stock. Contrarily, if the initial resource stock level is greater
than the steady state resource stock, the resource stock decreases over time
on the optimal dynamic path while the management effort input increases.
The economic meaning of this result is as follows. The more resource stock
people have, the less management effort would be needed, or in other words,
in order to reach the steady state, people would devote less management
effort than the steady state management effort level when the resource is
in relative abundance, in the sense that the resource stock is greater than
86
the steady state resource stock. And they would invest more management
effort than the steady state management effort level when the resource is
relatively scarce, in the sense that the resource stock is less than the steady
state resource stock. With reference to the poaching rate, we know that it
will increase as the management effort input decreases, and vice versa. It follows that the poaching rate will change in the same direction as the resource
stock does.
E
•
X =0
E*
E& = 0
X
X*
Xr
Figure 6.2 Phase diagram on the (X,E) plane.
6.6 Comparative static analysis
In order to investigate the influence of the exogenous variables on the
equilibrium resource stock, we conduct here a comparative static analysis by
introducing a particular specification for the natural growth function of the
resource stock F (X). We specify
F (X) = ρX(1 − X)
87
(6.48)
where the exogenous coefficient ρ denotes the intrinsic growth rate of the
resource stock. This means that the growth rate of the resource stock will
approach ρ when the resource stock level is extremely small, i.e.
F (X)
=ρ
X→0
X
lim
(6.49)
(Wacker and Blank, 1998), or in other words, the intrinsic growth rate of
the resource stock is the highest growth rate that a population can reach, if
it is not subject to food, space, resource competition and predation. This
specification also implies that the carrying capacity of the environment X
equals 1 and XMSY equals 12 , where XMSY denotes the stock level which can
afford the Maximum Sustainable Yield. The fundamental concavity property
of the Hamiltonian of the optimization problem and the uniqueness together
with the stability of the steady state solution remain unchanged under this
specification.
The specific version of the simple model results in the following dynamic
system after some rearrangements:
·
X = ρX(1 − X) − Y (E(λ)) − h(λ)
(6.50)
·
λ = (r − ρ + 2ρX) λ − V 0 (X).
The system encompasses two endogenous variables, X and λ, and two exogenous variables, r and ρ. By taking the total differential of the system (6.50),
it yields
¶ µ
µ
¶
dX ∗
ρ − 2ρX
−Y 0 E 0 − h0
·
(6.51)
2ρλ − V 00 (X) r − ρ + 2ρX
dλ∗
µ
¶ µ
¶
dρ
−X + X 2 0
=
·
λ − 2Xλ −λ
dr
By application of Cramer´s rule, the following results of the comparative
static analysis can be derived:
A
dX ∗
=
dr
|J|
dX ∗
B
=
dρ
|J|
88
where
J=
µ
ρ − 2ρX
−Y 0 E 0 − h0
2ρλ − V 00 (X) r − ρ + 2ρX
¶
,
A = λ(−Y 0 E 0 − h0 ) > 0
and B = (−X + X 2 )(r − ρ + 2ρX) − (λ − 2Xλ)(−Y 0 E 0 − h0 ).
Under the assumption of the proposition 2, we know that |J| < 0. It is also
clear that A > 0 under given assumptions. It yields then
dX ∗
< 0.
dr
(6.52)
0
From (6.19), we know that r − ρ + 2ρX = V λ(X) > 0 in steady state. The
term −X + X 2 = − F (X)
< 0. By applying these results, it can be easily
ρ
∗
shown that B < 0, if X < 12 or X ∗ = 12 , and the sign of B is ambiguous , if
X ∗ > 12 . This gives the following results:
dX ∗
1
1
> 0, if X ∗ < or X ∗ = ,
dρ
2
2
and
(6.53)
dX ∗
1
is ambiguous, if X ∗ > .
dρ
2
The outcome of (6.52) shows that an increase in the discount rate r will
lower the equilibrium resource stock. This is because a higher discount rate
raises the opportunity cost of holding resource stock, as the term rλ of (6.19)
shows, and constitutes motives for stock disinvestment. (6.53) demonstrates
that an increase in the intrinsic growth rate of the resources ρ will raise the
equilibrium resource stock, if the equilibrium resource stock is smaller than or
equals to the maximum sustainable yield stock level. The underlying reason
for this result is clear. The maximum sustainable yield stock level equals 12
under the special specification. If X ∗ < 12 , an increase in ρ will raise the
overall level of the marginal growth rate of the resources F 0 (X ∗ ) = ρ − 2ρX ∗ ,
and thereby raise the return of keeping resource stock. Hence, the resource
owner will be willing to hold a higher resource stock. On the other hand, if
X ∗ > 12 , an increase in ρ will lower the overall level of the marginal growth
89
rate of the resources, and cause a disincentive for keeping resource stock.
This leads to an ambiguously total effect.
6.7 A special case of the simple model
We investigate in this section a special case of the simple model, namely,
that the non-consumptive value of the resource will not be taken into account.
In this case, the utility function will have the functional form U(h), while
all other functions remain unchanged. It follows that the resource owner has
the resource management problem
Z ∞
Max
[U(h) − C(E)] e−rt dt
0
·
s.t. X = F (X) − Y (E) − h.
(6.54)
The relevant first-order necessary conditions can be easily derived:
∂H
= U 0 (h) − λ = 0
∂h
(6.55)
∂H
= −C 0 (E) − λY 0 (E) = 0
∂E
(6.56)
·
X = F (X) − Y (E) − h
·
λ = rλ −
∂H
= [r − F 0 (X)] λ.
∂X
(6.57)
(6.58)
These conditions have the same economic meaning as those presented in
section 6.2. Under given assumptions, it can also be easily verified that one
unique steady state solution exists, and the steady state solution is saddle
point stable.
Proposition 3 Under given assumptions with regard to the poaching function Y (E), the utility function U(h), the management cost function C(E)
and the natural dynamics of the resource stock F (X), the dynamic system
(6.57) and (6.58) possess an unique steady state solution.
90
Proof. See appendix 6.1.
Proposition 4 Under given assumptions with regard to the poaching function Y (E), the utility function U(h), the management cost function C(E)
and the natural dynamics of the resource stock F (X), the unique steady state
solution of the dynamic system (6.57) and (6.58) is saddle point stable.
Proof. See appendix 6.2.
Next, by applying the same method used in section 6.5, the phase diagrams on the X-h plane and on the X-E plane can be derived, respectively
(see Figure 6.3 and 6.4). The phase diagrams are similar to those of the
·
simple model, except that the isocline h = 0 in figure 6.3 is a vertical line
·
through X ∗ which parallels the h-axis, and the isocline E = 0 in figure 6.4 is
a vertical line through X ∗ which parallels the E-axis. These differences can
be attributed to the condition (6.58) which implies that
F 0 (X ∗ ) = r.
(6.59)
It is worth noting that, in this special model, the steady state resource stock
X ∗ is always less than the stock level which can afford the Maximum Sustainable Yield (MSY), because of the condition (6.59). In other words, the
resource stock is worth being preserved, from the point of view of the resource
owner, only when it´s marginal growth rate can compete with the discount
rate. Obviously, this is not possible for a resource stock level greater than
XMSY , where XMSY denotes the stock level which can afford the Maximum
Sustainable Yield. Except these differences, figure 6.3 and 6.4 demonstrate
similar interactions between X and h and between X and E on the stable
trajectories, as demonstrated by figure 6.1 and 6.2.
91
h
•
h=0
h*
•
X =0
X
X*
X MSY
X
Figure 6.3 Phase diagram on the (X,h) plane.
92
E
•
•
X =0
E=0
E*
X
X*
X MSY
Figure 6.4 Phase diagram on the (X,E) plane.
Finally, let us examine the comparative static effects of the special model
by using the specification of (6.48). Under this specification, the equilibrium
resource stock level
X∗ =
ρ−r
,
2ρ
(6.60)
as a result of the condition (6.59). Hence, the equilibrium resource stock
depends on the two exogenous variables r and ρ. It can be easily verified
that
∂X ∗
1
= − < 0, and
∂r
2ρ
∗
∂X
2r(1 − ρ)
> 0.
=
∂ρ
4ρ2
(6.61)
These outcomes mean that an increase in discount rate will lower the equilibrium resource stock level, since the marginal growth rate of the resource stock
F 0 (X) must compete with a higher discount rate in steady state, and only a
decrease in the equilibrium resource stock level can lead to a higher marginal
growth rate. On the other hand, a higher intrinsic growth rate of the resource
93
stock ρ will raise the overall level of F 0 (X) = ρ − 2ρX, if X < 12 = XMSY .
This is exactly the case as our model, since X ∗ = ρ−r
< 12 in our special
2ρ
model. Other things being equal, the resource owner will be willing to keep
a higher equilibrium resource stock level, because the resource, as a kind of
natural capital, becomes more productive.
6.8 Concluding remarks and policy implication
In this chapter we have developed a nonlinear bioeconomic model with one
state variable (resource stock) and two control variables (harvest rate, management effort) to investigate the dynamic development process of resource
stock, harvest rate, management effort input and poaching rate, under the
premise that people are allowed to use and manage renewable resources in
or around protected areas. The economic motive of the poachers is analyzed to construct the basis of the poaching function. Under some general
assumptions with regard to the poaching function, the utility function, the
management cost function and the natural dynamics of the resource stock,
it can be verified that the steady state solution of the dynamic problem is
unique and saddle point stable. The optimal time paths of the resource stock,
harvest rate and management effort input are also depicted in relevant phase
diagrams. By the help of a specification with reference to the natural growth
function of the resource stock, we can identify two critical exogenous variables, the discount rate and the intrinsic growth rate of the resource stock,
which will influence the equilibrium resource stock level. In addition, a special case of the model, which considers only the consumptive value of the
resource, is also examined.
From the point of view of the conservation policy, the implication of the
theoretical model is as follows. First, the quantity of the equilibrium resource stock depends on the intrinsic growth rate of resource and on the
discount rate. The smaller the discount rate is, the higher the equilibrium
resource stock level will be, and vice versa, while the comparative static effect
of the intrinsic growth rate on the equilibrium resource stock is ambiguous,
if the equilibrium resource stock is greater than the maximum sustainable
yield stock level. However, the special case of the model in which the nonconsumptive value of the resource is not considered demonstrates an unambiguous comparative static effect of the intrinsic growth rate of resource.
Therefore, these two variables are good indicators for the assessment of the
sustainable use strategy in specific cases. The use approach might potentially be more appropriate in a case with low discount rate and high intrinsic
growth rate of resource than another cases with high discount rate and low
intrinsic growth rate of resource.
94
Next, the impact of the sustainable use approach on conservation is
double-edged, in the sense that the sustainable use approach will not necessarily result in a higher stock level of renewable resources. On the one
hand, the sustainable use approach will theoretically contribute to better
management of protected areas, decrease of poaching activity and the following increase in resource stock, if the initial resource stock is less than
the equilibrium resource stock and no or only little management capacity
exists initially.22 On the other hand, if the initial resource stock is greater
than the equilibrium resource stock, the use approach will inevitably lead
to a decrease in resource stock through the adjustment of harvest rate until
the equilibrium resource stock is reached, even though the resource owner
will, to certain extent, invest simultaneously in management to control the
poaching activity. Moreover, as the special case of our model predicts, it is
notable that the equilibrium resource stock level is always smaller than the
stock level which can afford the maximum sustainable yield. Certainly, in
some cases, this equilibrium resource stock level X ∗ will not be necessarily
very small in comparison with the carrying capacity X, as the left-skewed
logistic growth function in Figure 6.5 shows.23 However, it is also absolutely
possible that the equilibrium resource stock is small enough in some cases
that people will be seriously concerned about the viability of the resource
and the loss of it´s ecological functions. Even in the case of our original
model, in which the non-consumptive value of the resource is taken into account so that the equilibrium resource stock is not necessarily smaller than
the maximum sustainable yield stock, such possibility cannot be excluded,
if the non-consumptive value of the resource is negligible. Apparently, in
order to decide whether the use approach is appropriate in specific cases, we
need to know, at least roughly, the relevant functional forms and parameter
values, and simulate the possible scenarios before any decision is made. A
similar trial will be conducted after the model presented here is extended in
next chapter.
22
An implicit but important premise for this conclusion is, that the use strategy can
really generate positive or at least zero discounted net profit. Otherwise the use approach
will not work at all.
23
Such cases could happen if the resource stock increases at a relatively steady marginal
growth rate, and the marginal growth rate will decrease apparently only when the resource stock comes near the carrying capacity of the environment. It happens usually in
species in which animals do not breed until relatively late in life. Robinson and Redford
(1991) suggested that, generally, wildlife species reach their maximum productivity when
population densities are close to the range of 65% to 90% of carrying capacity.
95
F(X )
X
X * X MSY
X
Figure 6.5 Left-skewed logistic growth function.
96
Appendix 6.1
·
From (6.58), we know that if λ = 0, it gives
r − F 0 (X) = 0.
(A.6.1)
Therefore, the steady state solution of the resource stock X must satisfy the
condition
F 0 (X ∗ ) = r
(A.6.2)
where X ∗ denotes the steady state solution of the resource stock X. Since
F 0 (X) is a monotonously decreasing function of X, it is obvious that X ∗ is
unique. From (6.56) and (6.55), we know that
C 0 (E) = −U 0 (h)Y 0 (E).
(A.6.3)
Consider E as an implicit function of h. Substituting E = E(h) for equation
(A.6.3), and differentiating (A.6.3) with h, it yields
[C 00 (E) + U 0 (h)Y 00 (E)]
dE
= −U 00 (h)Y 0 (E).
dh
Under given assumptions, it can be easily shown that
dE
< 0.
dh
(A.6.4)
Applying E = E(h) for equation (6.57), it implies that
F (X ∗ ) − Y (E(h)) − h = 0
in steady state. By defining
Ω(h) = F (X ∗ ) − Y (E(h)) − h
and differentiating Ω(h) with h, we obtain
Ω0 (h) = −Y 0 (E)E 0 (h) − 1 < 0,
because E 0 (h) < 0, as (A.6.4) shows. Therefore, Ω(h) = 0 possesses a unique
solution h∗ . And since E = E(h), E 0 (h) < 0, λ = U 0 (h) and U 00 (h) < 0, it is
obvious that both E(h) and U 0 (h) are monotonously decreasing function of h.
It follows that E ∗ = E(h∗ ) and λ∗ = U 0 (h∗ ) are also unique. Consequently,
the steady state solution (X ∗ , λ∗ , h∗ , E ∗ ) is unique.
97
Appendix 6.2
By use of the result of the equation (6.55), we can define h as a function
of λ:
h = (U 0 )−1 (λ) = h(λ),
(A.6.5)
and it can be easily verified that
dh
< 0.
dλ
(A.6.6)
Again, by applying the equation (6.56), we define E as a function of λ:
E = E(λ).
(A.6.7)
Since
dλ
−C 00 (E)Y 0 (E) + C 0 (E)Y 00 (E)
=
> 0,
dE
Y 00 (E)
it yields
dE
> 0.
dλ
(A.6.8)
After substituting h(λ) for h and E(λ) for E in equation (6.57), we observe
now the following dynamic system of equations:
·
X = F (X) − Y (E(λ)) − h(λ)
(A.6.9)
·
λ = [r − F 0 (X)] λ.
·
·
We differentiate then X and λ with X and λ, respectively. This yields
·
∂X
= F 0 (X)
∂X
·
∂X
= −Y 0 (E)E 0 (λ) − h0 (λ)
∂λ
·
∂λ
= −F 00 (X)λ
∂X
·
∂λ
= r − F 0 (X).
∂λ
98
(A.6.10)
A Taylor expansion of the dynamic system (A.6.9) at (X ∗ , λ∗ ) gives then
Ã
·
X
·
λ
!
=
µ
F 0 (X ∗ )
−Y 0 (E)E 0 (λ∗ ) − h0 (λ∗ )
−F 00 (X ∗ )λ∗
r − F 0 (X ∗ )
¶µ
X − X∗
λ − λ∗
¶
(A.6.11)
·
·
·
·
X ∂X ∂λ
, after the results of (A.6.10) are applied for substituting ∂∂X
, ∂λ , ∂X and ∂∂λλ
in (A.6.11). Moreover, we know that in √steady state F 0 (X ∗ ) = r , as (A.6.2)
2
shows. Hence, the two eigenvalues r± r2 +4A can be easily derived where
A = −F 00 (X ∗ )λ∗ [−Y 0 (E)E 0 (λ∗ ) − h0 (λ∗ )] > 0 under given assumptions. It is
obvious that one of the eigenvalues is positive, and the other one is negative.
Therefore, the steady state solution (X ∗ , λ∗ ) is a saddle point.
99
Chapter 7
Management capital, use of renewable
resources, poaching and anti-poaching:
a bioeconomic model with two state
and two control variables
7.1 Introduction
The capital theory plays an extremely important role in investigating the
management problem of the renewable resources. We can consider the link
between capital theory and management of the renewable resources in three
different aspects. First, it has long been recognized that the exploitation
problem of the renewable resources can be analyzed in a capital-theoretic
framework (Scott, 1955; Clark and Munro, 1975; Clark, 1976). This recognition is founded on the fact that renewable resource stock can be treated
as a capital stock, in the sense that it can generate a consumption flow,
i.e. harvest, over time, and current harvest will influence the resource stock
level, the potential for regeneration and the future harvest possibilities of the
resource stock (Clark and Munro, 1975).
Next, it is notable that the harvest rate is subject to the available capital stock, both physical and human, utilized in exploiting the resource stock
in investigating the harvest problem of the renewable resources. Therefrom
the problem of capital accumulation arises, especially in the fishery. The
usual single-state-variable (resource stock) and single-control-variable (harvest) model was then extended to a model which consider explicitly two
capital stocks (Smith, 1968: Smith, 1969). In a dynamic linear-in-controlvariables model involving two state variables (resource stock, capital stock)
and two control variables (harvest rate, investment rate), Clark, Clarke and
Munro (1979) depicted the dynamic path of an optimal policy and showed,
that the variable denoting the capital stock can be eliminated from the analysis and the model can be reduced to the usual single-state-variable model if
capital investment is assumed to be perfectly reversible. In other words, the
single-state-variable model is in fact a special case of the general two-statevariables model. Similar attempt was followed by Boyce (1995) in a nonlinear
two-state variable, two-control variable model with irreversible investment.
He found that the nonlinearity in control variables has great impact on the
optimal harvest and investment policy.
100
While the problem of capital accumulation utilized in exploiting resource
stock has been studied deliberately by economists for a long time, an important dimension of the link between capital theory and management of the
renewable resources remains almost intact. That is the problem of capital
accumulation utilized in protecting renewable resources when the problem of
poaching and anti-poaching is concerned. Among the comparatively few articles in economic literature dealing with the relevant management problem of
natural resources, Skonhoft and Solstad (1996) regarded the anti-poaching
effort as a flow variable and hence reduced the problem to a single-statevariable model. Katz (2000) argued that the existence of social capital, a
technical terminology applied in modern sociology, plays an important role
in natural resource management. She is probably one of the few economists
who explicitly use the terminology ´capital´ and treat a certain kind of capital as an critical factor in the management practices of natural resources.
However, her analysis is in principle based on case studies and lacks an internally consistent theoretic foundation.
In this chapter, rather than investigating the problem of capital accumulation utilized in exploiting resource stock, we will focus on the problem
of capital accumulation utilized in protecting renewable resources. To do
this, a new state variable ´management capital´ will be introduced and the
simple model of chapter six will be extended to a model with two state variables (resource stock, management capital) and two control variables (harvest
rate, investment rate). The dynamic interaction between control variables
and state variables will be investigated deliberately. By the assistance of a
comparative static analysis, the influence of exogenous parameters on the
equilibrium resource stock will be studied. The policy implications of the
model for the conservation issues will also be discussed.
7.2 Management capital
Both the Skonhoft-Solstad model and our simple model in chapter six
regard the management effort for protecting renewable resources as a flow
variable. Undoubtedly, such treatment avoid to great extent the technical
complexity. Nonetheless, it also neglects the fact in conservation practices
that management authorities usually need certain kinds of capital assets
to offer a management service flow through time which can contribute to
the conservation of natural resources. Hence, in addition to the traditional
stock variable ´resource stock´ in bioeconomic models, we may consider a
new stock variable to help modeling the interaction between the two stock
variables, i.e. the resource stock and the capital stock utilized in protecting resource stock. For convenience of discussion, we introduce here a new
101
terminology ´management capital´, which encompasses all kinds of capital
assets that are necessary for conservation practices. Under the concept of
the management capital, three important components could be classified as
follows.
First, among the various types of capital assets, the conventional physical
capital is undoubtedly necessary for conservation practices. This usually includes buildings, vehicles, equipments and sometimes aircraft. The measurement of the physical capital stock, investment and depreciation is relatively
easy. The physical inputs component of the flow variable ´management effort´ in the simple model can be then viewed as an investment in physical
capital.
Secondly, no one can deny the fact that human capital plays a critical
role in conservation practices. Similar to that Romer (1990) suggested, we
define human capital as all forms of intangible knowledge, know-how and
other human skills that are rivalrous and excludable, in the sense that they
are inherently tied to the physical object, i.e. the human body. In contrast, certain forms of knowledge or know-how can be neither rivalrous nor
excludable, since they are stored on paper or computer systems. They will
be then separated from the rival component of knowledge (or know-how) and
classified into an another sort of capital. Under this definition, the human
capital stock may be measured in criteria such as the years of formal education, on-the-job training and experiences (Romer, 1990). And a proportion
of the labor inputs component of the flow variable ´management effort´ in
the simple model can be treated as an investment in human capital.24 What
remains in the labor inputs devoted to management can be viewed as an
investment in institution capital which is investigated afterward.
Finally, it has long been recognized that a successful resource management requires an adequate institutional base. Under the term institution, we
may identify the two essential components, namely property rights (Bromley, 1994; Lant, 1994; Swallow and Bromley, 1995) and organizational issues
(Wade, 1987; Murphree, 1994; Bromley, 1994), which can substantially influence the outcomes of resource management. While these two factors are
viewed as exogenous variables in most of the resource-management-relevant
economic literature, we asserts here that institution in the field of resource
management arises and develops mainly on account of the intentional actions
24
It is notable that some routine and nonprofessional labor inputs make little contribution to the accumulation of human capital, or in other words, the depreciation rate of this
component of the human capital is quite high.
102
taken by people who react to market and/or other non-market incentives,
such as financial profits and/or non-use value. Hence, institution should
be treated as an endogenous rather than exogenous variable in our model.
Furthermore, developing new and modifying existing institution requires resource inputs and hence incurs costs. These resource inputs can be viewed as
an investment in the formation of a special component of the management
capital, i.e. the institution. In addition, for simplicity the knowledge or
know-how which is neither rivalrous nor excludable will be included in this
category of management capital, if we explain their essence in a wider sense
that they are a necessary component of institution.25 It should be here recognized that, up to now, it is extremely difficult to measure the stock of the
´institution capital´ quantitatively, while certain kind of qualitative ranking
may be possible. However, we asserts that the institution factor, as a concept,
is so important that it should not be neglected just because of the difficulty
of measurement. Besides, the measurement of the investment in institution
capital is relatively easy. It can be measured in criterion such as the labor
inputs in relevant institution-building issues, though some ambiguity arises
when we try to differentiate the labor inputs in relevant institution-building
issues from the labor inputs in human capital investment. For example, that
a park officer executes authority can contribute simultaneously to his personal experiences and the development of institution. But in practice, at
the aggregate level of management capital, it is not necessary to differentiate
them from each other.
In sum, we argue in this chapter that it is the existence of the stock variable management capital, rather than the flow variable management effort,
that can have positive impacts on the management of renewable resources.
And people invest intentionally in management capital in reaction to certain
market and/or non-market incentives. It follows that management capital
stock and relevant investment are treated as endogenous variables in our
extended model.
7.3 The extended model
In this section, a nonlinear bioeconomic model with two state variables
(resource stock, management capital) and two control variables (harvest rate,
investment rate) is developed on the basis of the simple model in chapter six.
The necessary conditions for the optimal policy are derived. The uniqueness
25
For example, a data-bank for the distribution of renewable resources in a specific
region is critical in developing property rights. An another example is that formally
internal instructions can contribute to the organizational operation of a park authority.
103
and stability properties of the steady state solution of the model will also be
presented in section 7.4 and 7.5, respectively.
In comparison to the simple model, two fundamental modifications are
introduced to develop the extended model. First, rather than the flow variable management effort, it is the stock variable management capital that can
affect the management of renewable resources. And people invest intentionally in management capital in reaction to certain market and/or non-market
incentives. As a result, the endogenous variable K(t) denoting the stock of
management capital at time t is introduced.26 The equation of motion for
the management capital is
·
K(t) ≡
dK(t)
= I(t) − δK(t)
dt
(7.1)
where I(t) represents the investment rate in management capital at time
t and 0 < δ < 1 is the depreciation rate of management capital which is
assumed to be constant.
The second modification refers to that, according to the analysis in section
6.2, a more complete poaching function Y (X, K) is introduced here that the
poaching Y depends on both the resource stock X and the management
capital stock K. For simplicity, we express the poaching function as an
additively separable function of the two stock variables:
Y (X, K) = W (X) − Y (K).
(7.2)
The function W (X) represents the resource stock which is illegally exploited
and possesses the properties
W 0 (X) > 0,
W 00 (X) > 0.
(7.3)
The convexity assumption of W (X) asserts that, with an increasingly marginal
poaching rate, a higher resource stock will induce a higher poaching rate. The
function Y (K) denotes the resource stock which is potentially ´rescued´ from
poaching activity because of the devotion of management capital. It is notable that Y (K) could be greater than W (X). If this happens, it means that
resource owner takes active actions to increase resource stock, for example
by re-introducing species individuals in protected areas. It is assumed that
Y (K) is concave so that
Y 0 (K) > 0,
26
Y 00 (K) < 0.
(7.4)
As in chapter six, the time notation will henceforth be omitted wherever possible.
104
This means that a higher level of management capital will reduce the poaching rate, but it is limited by a decreasingly marginal effect for per unit of
additional management capital.
Now let us consider the same scenario as in section 6.2, that certain
people or certain organizations are given the legal rights to exploit renewable
resources in specific protected areas or in buffer areas around protected areas,
and they are authorized to manage natural resources and human activities in
those areas. All assumptions and properties with regard to the natural population dynamics of the resource and the utility function remain unchanged,
as shown in section 6.2.
In addition to natural factors, the population dynamics of resource is also
subject to human interference. When legal harvest and poaching factor are
taken into account, the equation of motion for an exploited resource stock X
can be stated as
·
dX
= F (X) − W (X) + Y (K) − h.
(7.5)
X≡
dt
The resource management problem of the resource owner is to choose
the harvest rate and the investment rate to maximize the net present utility
(in monetary terms) derived from protecting and harvesting the resource,
subject to the constraint of dynamics of the resource stock and management
capital. This task can be formally expressed as
Z ∞
Max
[U(h) + V (X) − C(I)] e−rt dt
(7.6)
0
·
s.t. X = F (X) − W (X) + Y (K) − h
·
K = I − δK
where C(I) and r denote the investment cost function and the instantaneous
discount rate, respectively. The investment cost function owns the properties
that
C 0 (I) > 0,
C 00 (I) > 0,
(7.7)
which implies that C(I) is strictly increasing and convex in I.27 The discount
27
For resource owner, the marginal cost of the investment in physical capital remains
probably constant, because a resource management authority constitutes only a tiny fraction of the whole physical capital market, for example the Jeep market. However, the
marginal cost of the investment in human and institution capital is likely to be increasing,
because the market of manpower for conservation is small and, especially at lower level,
is typically localized.
105
rate is assumed to be constant and 0 < r < 1. The optimization problem
has hence two state variables, resource stock and management capital stock,
and two control variables, harvest rate and investment rate, both of which
are to be chosen optimally over time.
Next, the problem is analyzed by the application of the maximum principle. The current-value Hamiltonian of our case is
H = U (h) + V (X) − C(I) + λ [F (X) − W (X) + Y (K) − h]
+µ [I − δK]
(7.8)
where λ is the current-value costate variable associated with the state variable X which gives the imputed marginal value of the resource stock, and
µ is the current-value costate variable associated with the state variable K
which gives the imputed marginal value of the management capital stock.
Assuming an interior solution, the first order necessary conditions describing
the optimization problem are given by equations (7.9)-(7.12) together with
(7.1) and (7.5):
·
λ = rλ −
·
∂H
= U 0 (h) − λ = 0
∂h
(7.9)
∂H
= −C 0 (I) + µ = 0
∂I
(7.10)
∂H
= [r − F 0 (X) + W 0 (X)] λ − V 0 (X)
∂X
(7.11)
∂H
= (r + δ)µ − λY 0 (K).
∂K
(7.12)
µ = rµ −
Under given assumptions, the Hamiltonian is apparently concave in the
state variables X, K and in the control variables h, I. The second order
conditions are therefore satisfied. Along the optimal trajectory, the equation
(7.9) means that the imputed marginal value, or the shadow price of an extra
resource stock, λ, must be equal to the marginal profit of harvesting one unit
of renewable resources, U 0 (h). The equation (7.10) implies that the shadow
price of an extra management capital stock, µ, must equal the marginal
cost of investment in management capital, C 0 (I). Equation (7.11) indicates
106
·
that the change rate of the shadow price of the resource stock, λ, plus the
marginal non-consumptive value of the resource stock, V 0 (X), must equal the
opportunity cost when resource owner goes on to keep one unit of resource
stock, [r − F 0 (X) + W 0 (X)] λ. The opportunity cost includes the component
of forgone ´interest´, rλ, and the loss derived from marginal poaching rate as
a result of the increased resource stock, W 0 (X)λ, minus the gain derived from
the marginal growth rate of the resource stock, F 0 (X)λ. Finally, the equation
(7.12) implies that the change rate of the shadow price of the management
·
capital stock, µ, plus the gain derived from the reduction of the poaching rate
as a result of the devotion of an extra unit of management capital, λY 0 (K),
must be equated with the opportunity cost when resource owner devotes
one unit of management capital stock to the protection of resources. The
opportunity cost comprises the component of the ´interest cost´, rµ, and
the capital loss as a result of the depreciation, δµ.
7.4 Uniqueness of the steady state solution
In the following sections, we will investigate the properties of the steady
state solution of the dynamic problem. Suppose that a steady state exists,
the steady state solution can be determined under the conditions that the
resource stock, the management capital stock, the shadow price of the resource and the shadow price of the management capital are constant, i.e.,
·
·
·
·
X = K = λ = µ = 0. First, the uniqueness of the steady state solution can
be verified by proposition 5.
Proposition 5 Under given assumptions with regard to the poaching function W (X) − Y (K), the utility function U(h) + V (X), the investment cost
function C(I) and the natural dynamics of the resource stock F (X), the dynamic system (7.1), (7.5), (7.11) and (7.12) possess an unique steady state
solution.
·
·
·
·
·
Proof. In steady state, X = K = λ = µ = 0. First, λ = 0 gives
λF 0 (X) − λW 0 (X) + V 0 (X) = rλ
(7.13)
by applying equation (7.11). Under given assumptions, both F 0 (X) and
V 0 (X) are monotonously decreasing and W 0 (X) is monotonously increasing
in X. Therefore, the function λF 0 (X)−λW 0 (X)+V 0 (X) is also monotonously
decreasing. It is obvious that one unique X ∗ exists which satisfying the
condition λF 0 (X ∗ ) − λW 0 (X ∗ ) + V 0 (X ∗ ) = rλ.
107
Next, after applying the results of equations (7.9) and (7.10) in (7.12),
µ = 0 yields
·
(r + δ)C 0 (I) − U 0 (h)Y 0 (K) = 0.
(7.14)
·
In addition, K = 0 in steady state implies that
I = δK
(7.15)
by applying equation (7.1). After substituting δK for I in (7.14), h can be
defined as an implicit function of K. A total differentiation of equation (7.14)
with K yields then
δ(r + δ)C 00 (I) − U 0 (h)Y 00 (K) = U 00 (h)h0 (K)Y 0 (K)
(7.16)
and it can be easily shown that
h0 (K) < 0
(7.17)
under given assumptions with regard to C(I), U (h) and Y (K).
Finally, by applying the results of the previous discussion about X ∗ and
·
h(K) in equation (7.5), X = 0 gives
F (X ∗ ) − W (X ∗ ) + Y (K) − h(K) = 0.
Then we define a function
Θ(K) = Y (K) − h(K) + A
where A = F (X ∗ ) − W (X ∗ ) is a constant. It follows that
Θ0 (K) = Y 0 (K) − h0 (K) > 0
by the application of (7.17) and of given assumption with regard to Y (K).
Hence, Θ(K) is monotonously increasing in K and Θ(K) = 0 has a unique
solution K ∗ . According to equations (7.15) and (7.17), this result ensures that
the steady state solution of I and h is unique. It follows that λ∗ = U 0 (h∗ )
and µ∗ = C 0 (I ∗ ) are also unique. Consequently, the steady state solution
(X ∗ , K ∗ , h∗ , I ∗ , λ∗ , µ∗ ) of the dynamic system is unique.
It is here worth noting the equilibrium resource stock. The special case
of the simple model in chapter six indicated that the equilibrium resource
stock level is always smaller than the stock level XMSY which can afford the
108
maximum sustainable yield, because of the condition F 0 (X ∗ ) = r, as (6.59)
showed. However, according to (7.13), the equilibrium resource stock in the
extended model is determined by the condition [r − F 0 (X ∗ ) + W 0 (X ∗ )] λ −
V 0 (X ∗ ) = 0. In comparison to (6.59), two new factors, i.e. W 0 (X ∗ )λ and
V 0 (X ∗ ) can also influence the steady state resource stock. The marginal
poaching effect, W 0 (X)λ, leads to a decrease in the equilibrium resource
stock, since greater resource stock induces more poaching and thereby increases the cost of holding on resource stock. The marginal non-consumptivevalue effect, V 0 (X), results in an increase in the equilibrium resource stock,
because greater resource stock raises the non-consumptive value, and thereby
increases the benefit derived from holding on resource stock as an asset. The
interaction between these two effects makes it difficult to determine whether
the equilibrium resource stock is greater or smaller than XMSY . Finally, it
depends on the relative strength of these two effects. In any case, it is in
the extended model possible that the equilibrium resource stock is greater
than XMSY , if renewable resources can generate sufficiently great marginal
non-consumptive value so that the marginal non-consumptive-value effect
dominates the steady state solution.
7.5 Stability of the steady state solution
Now, we concentrate on the stability of the steady state solution. By
applying the result λ = U 0 (h) of the equation (7.19), h can be defined as a
function of λ:
h = (U 0 )−1 (λ) = h(λ).
(7.18)
it can be easily verified that
dh
< 0.
dλ
(7.19)
Again, by applying the result µ = C 0 (I) of the equation (7.10), I can be
defined as a function of µ:
I = (C 0 )−1 (µ) = I(µ).
(7.20)
dI
> 0.
dµ
(7.21)
it also can be shown that
109
After substituting h(λ) for h in (7.5) and I(µ) for I in (7.1), we observe
now the following dynamic system of equations:
·
X = F (X) − W (X) + Y (K) − h(λ)
(7.22)
·
K = I(µ) − δK
·
λ = [r − F 0 (X) + W 0 (X)] λ − V 0 (X)
·
µ = (r + δ)µ − λY 0 (K).
The Jacobian of the dynamic system (7.22) is stated as
 ·
·
·
· 



J =


∂X
∂X
∂X
∂K
∂X
∂λ
∂X
∂µ
∂K
∂X
∂K
∂K
∂K
∂λ
∂K
∂µ
∂λ
∂X
·
∂µ
∂X
∂λ
∂K
·
∂µ
∂K
∂λ
∂λ
·
∂µ
∂λ
∂λ
∂µ
·
∂µ
∂µ
·
·
·
·
·
·
·
·






(7.23)
and after some routine calculations it yields

F0 − W0
Y0
0
−h0

0
−δ
0
I0
J =
00
00
00
0
0
 −λ(F − W ) − V
0
r−F +W
0
00
0
0
−λY
r+δ
−Y
The value of the determinant can be derived:


.

|J| = δ(r + δ) [λ(F 00 − W 00 ) + V 00 ] h0
−δ(r + δ)(F 0 − W 0 )(r − F 0 + W 0 )
−λ [λ(F 00 − W 00 ) + V 00 ] Y 00 h0 I 0
+λ(F 0 − W 0 )(r − F 0 + W 0 )Y 00 I 0
− [λ(F 00 − W 00 ) + V 00 ] (Y 0 )2 I 0 .
Next, we define a function G
¯ ·
¯ ∂ X ∂ X·
¯
G = ¯¯ ∂X· ∂λ·
∂λ
∂λ
¯ ∂X
∂λ
as
¯ ¯
¯ ¯
¯ ¯
¯+¯
¯ ¯
¯ ¯
·
∂K
∂K
·
∂µ
∂K
·
∂K
∂µ
·
∂µ
∂µ
The value of G can be derived as follows:
¯
¯
¯
¯
¯
¯
¯ + 2¯
¯
¯
¯
¯
·
·
∂X
∂K
∂X
∂µ
∂λ
∂K
∂λ
∂µ
·
·
(7.25)
¯
¯
¯
¯
¯
¯
G = (F 0 − W 0 )(r − F 0 + W 0 ) − [λ(F 00 − W 00 ) + V 00 ] h0
−δ(r + δ) + λY 00 I 0 .
110
(7.24)
(7.26)
The local stability property of the dynamic system (7.22) is determined by the
eigenvalues of the Jacobian J. According to the general formula developed by
Dockner (1985), we can calculate the eigenvalues of the Jacobian associated
with the optimal control problems with two state variables:
ξ 1,2,3,4 =

³ r ´2
r
±
2  2
"µ ¶
# 12  12
2

G
G
− ±
− |J|

2
2
(7.27)
where r is the discount rate and |J| is the determinant of J. And Dockner
(1985) showed that two of the eigenvalues are positive and the other two
negative, i.e. the steady state solution is saddle point stable, if |J| > 0 and
G < 0. By applying this result, the local stability property of the dynamic
system (7.22) can verified under further assumptions.
Proposition 6 Under given assumptions with regard to the poaching function W (X) − Y (K), the utility function U(h) + V (X), the investment cost
function C(I) and the natural dynamics of the resource stock F (X), the
unique steady state solution of the dynamic system (7.22) is saddle point
stable, if U 0 (F 00 − W 00 ) + V 00 < U 00 (F 0 − W 0 )(r − F 0 + W 0 ).
Proof. Since h = (U 0 )−1 (λ) = h(λ) and λ = U 0 (h) according to (7.18)
dh dλ
and (7.9), and since dλ
= 1, it can be easily verified that dh
= U 001(h) because
dh
dλ
dλ
= U 00 (h). The condition U 0 (F 00 − W 00 ) + V 00 < U 00 (F 0 − W 0 )(r − F 0 + W 0 )
dh
0
00
00 )+V 00
> (F 0 −W 0 )(r−F 0 +W 0 ) or equivalently
can be then written as U (F −W
U 00
as [λ(F 00 − W 00 ) + V 00 ] h0 > (F 0 − W 0 )(r − F 0 + W 0 ).
Let us first determine the sign of |J|. If [λ(F 00 − W 00 ) + V 00 ] h0 > (F 0 −
0
W )(r − F 0 + W 0 ), it can be easily shown that the sum of the first two terms
on the right-hand side of (7.25)
δ(r + δ) [λ(F 00 − W 00 ) + V 00 ] h0 − δ(r + δ)(F 0 − W 0 )(r − F 0 + W 0 ) > 0
is positive because of δ(r + δ) > 0. And the sum of the third and fourth
terms
−λ [λ(F 00 − W 00 ) + V 00 ] Y 00 h0 I 0 + λ(F 0 − W 0 )(r − F 0 + W 0 )Y 00 I 0
is also positive since Y 00 < 0 and I 0 > 0, according to (7.21). Finally, the fifth
term − [λ(F 00 − W 00 ) + V 00 ] (Y 0 )2 I 0 cannot be negative since F 00 < 0, W 00 > 0
and V 00 < 0. Accordingly, we can show that |J| > 0.
111
Next, we investigate the sign of G. Under the same premise, it is obvious
that the sum of the first two terms on the right-hand side of (7.26) is negative.
The third term −δ(r + δ) is clearly negative. And the fourth term λY 00 I 0 is
also negative under given assumptions. Hence, G < 0. The results |J| > 0
and G < 0 guarantee that two of the eigenvalues of J are positive and the
other two negative (Dockner, 1985). The steady state solution of the dynamic
system (7.22) is therefore saddle point stable.
The saddle-point stability property of the steady state solution implies
that, given the initial value of the resource and management capital stock
which are close to the steady state, it will always be possible for the resource
owner to choose a pair of optimal initial values of the harvest rate and the
investment rate which are on the stable trajectories that converge to the
steady state equilibrium of the dynamic system.
7.6 Comparative static analysis
We conduct here a comparative static analysis to investigate the influence of permanent changes in the exogenous parameters on the equilibrium
resource stock. To study the impact of as many parameters as possible, a
special version of the extended model will be considered by introducing particular specifications for F (X), W (X), V (X) and Y (K). In what follows, we
specify
F (X)
W (X)
V (X)
Y (K)
=
=
=
=
ρX(1 − X)
αX
βX
γK.
(7.28)
The exogenous coefficient ρ denotes the intrinsic growth rate of the resource
stock and, as discussed in section 6.6. The constant coefficient α represents
the marginal poaching rate and is an indicator for measuring the intensity
of poaching activity. The constant coefficient β denotes the marginal nonconsumptive value of resource stock. Finally, the constant γ is a efficiency
coefficient which measures the effect of one unit extra management capital on
the poaching rate. The linearity assumption with regard to W (X), V (X) and
Y (K) is somewhat unrealistic, but it allows us to study the effects of more
exogenous parameters than before. It can be easily verified that the fundamental concavity property of the Hamiltonian of the optimization problem
and the uniqueness together with the stability of the steady state solution
remain unchanged under these specifications.
112
The specific version of the extended model results in the following dynamic system after some rearrangement:
·
X = ρX(1 − X) − αX + γK − h(λ)
(7.29)
·
K = I(µ) − δK
·
λ = (r − ρ + 2ρX + α) λ − β
·
µ = (r + δ)µ − λγ.
The system encompasses four endogenous variables, X, K, λ, µ, and six
exogenous variables, r, α, β, γ, δ, ρ. By taking the total differential of the
system (7.29), it yields


 
ρ − 2ρX − α γ
−h0
dX ∗
0
∗ 


0
−δ
0
I0 

 ·  dK∗ (7.30)

2ρλ
0 r − ρ + 2ρX + α
0   dλ 
0
0
−γ
r+δ
dµ∗


dρ


 dr 
−X + X 2 0 X 0 −K 0


 dα 

0
0
0 0 0
K 




·
= 

dβ
λ − 2Xλ −λ −λ 1 0
0  


 dγ 
0
−µ 0 0 λ −µ
dδ
By application of Cramer´s rule, the following results of the comparative
static analysis can be derived.
Proposition 7 Under special specifications of (7.29) with regard to the poaching function, the utility function and the natural dynamics of the resource
stock, and under given assumptions with regard to the investment cost func∗
tion, dX
> 0 if the equilibrium resource stock is less than or equals to the
dρ
∗
is ambiguous if the equilibmaximum sustainable yield stock level, and dX
dρ
rium resource stock is higher than the maximum sustainable yield stock level.
Proof. See appendix 7.1.
The outcome of proposition 7 shows that an increase in the intrinsic
growth rate of the resources ρ will raise the equilibrium resource stock, if
the equilibrium resource stock is smaller than or equals to the maximum
113
sustainable yield stock level. The underlying reason for this result is clear.
As in section 6.6 discussed, the maximum sustainable yield stock level equals
1
under the special specification. If X ∗ < 12 , an increase in ρ will raise the
2
overall level of the marginal growth rate of the resources F 0 (X ∗ ) = ρ − 2ρX ∗ ,
and thereby raise the return of keeping resource stock. Hence, the resource
owner will be willing to hold a higher resource stock. On the other hand, if
X ∗ > 12 , an increase in ρ will lower the overall level of the marginal growth
rate of the resources, and cause a disincentive for keeping resource stock.
However, the first term in the right-hand side of the first equation of (A.7.5)
δ(r +δ)(X −X 2 ) (r − ρ + 2ρX + α) is positive. This leads to an ambiguously
total effect.28
Proposition 8 Under special specifications of (7.29) with regard to the poaching function, the utility function and the natural dynamics of the resource
stock, and under given assumptions with regard to the investment cost func∗
∗
tion, the following comparative static effects are derived: dX
< 0, dX
< 0,
dr
dα
∗
∗
∗
dX
dX
dX
> 0, dγ > 0 and dδ < 0.
dβ
Proof. See appendix 7.1.
Proposition 8 demonstrates some unambiguous comparative static effects.
First, an increase in the discount rate r and in the marginal poaching rate α
will lower the equilibrium resource stock, and vice versa. This is because a
higher discount rate raises the opportunity cost of holding resource stock, as
the term rλ of (7.11) shows, and constitutes motives for stock disinvestment.
The same rationale applies also to the effect of a permanent change in the
marginal poaching rate. On the contrary, an increase in the marginal nonconsumptive value of resource stock β raises the benefit of holding resource
stock, as (7.11) demonstrates, and encourages the resource owner to hold a
higher stock level. The coefficient γ measures the efficiency of management
capital against the poaching activity, and a higher γ implies that management capital can dampen poaching more effectively than before and thereby
28
Up to now, empirical studies generally support the conclusion that species with low
intrinsic growth rate are less resilient to harvest (e.g., Bodmer, 1995a; Bodmer, 1995b;
Bodmer et al., 1997a; Bodmer et al., 1997b; Bodmer and Puertas, 2000; Fa et al., 1995;
Fitzgibbon et al., 1995; Clayton and Milner-Gulland, 2000; Lee, 2000; Peres, 2000). This
may support the conjecture that the term δ(r +δ)(X −X 2 ) (r − ρ + 2ρX + α) in the righthand side of the first equation of∗ (A.7.5) is big enough so that, in any case, we can obtain
an unambiguous total effect dX
dρ > 0.
114
contribute to a greater resource stock. Finally, an increased depreciation rate
of management capital δ will reduce the equilibrium resource stock. The intuition is that, other things being equal, a higher depreciation rate causes
more depreciation of the given management capital stock, and this in turn
weakens the strength of anti-poaching action and results in a smaller equilibrium resource stock. With reference to the comparative static effects of
exogenous variables on the equilibrium management capital stock, appendix
7.1 shows that it is not possible to determine the signs of these effects as a
result of the ambiguous term ρ − 2ρX − α in every expressions of (A.7.6).
7.7 Concluding remarks and policy implications
In this chapter we address the important role played by management
capital in conservation praxis. We argue that it is the existence of the stock
variable management capital, rather than the flow variable management effort, that can have positive impacts on the management of renewable resources. People invest intentionally in management capital in reaction to
certain market and/or non-market incentives. Accordingly, in comparison to
the traditional one-state-variable bioeconomic models in which no man-made
capital exists, and the two-state-variable bioeconomic models which investigate the problem of capital accumulation utilized in exploiting resource stock,
we introduce a new stock variable to help modeling the interaction between
the two stock variables, i.e. the resource stock and the capital stock utilized in protecting resource stock. A nonlinear bioeconomic model with two
state variables (resource stock, management capital) and two control variables (harvest rate, investment rate) is then developed on the basis of the
simple model in chapter six, and under the premise that people are allowed
to use legally the renewable resources in or around protected areas. In addition to the management capital, the important modification with regard
to the functional forms of the poaching function is also made. Under some
assumptions with reference to the poaching function, the utility function,
the investment cost function and the natural dynamics of the resource stock,
it can be verified that the steady state solution of the dynamic problem is
unique and saddle point stable. By introducing a specific version of the extended model, we can identify six critical exogenous parameters, the discount
rate, the intrinsic growth rate of the resource stock, the marginal poaching
rate, the marginal non-consumptive value of the resource stock, the efficiency
coefficient of the management capital and the depreciation rate of the management capital, which will influence the equilibrium resource stock level.
Some critical comparative static effects are thereby found.
Several important outcomes of the extended model and the relevant im115
plications for conservation policy are here worth noting. First, the special
case of the simple model in chapter six concludes that the equilibrium resource stock level is in any case smaller than the maximum sustainable yield
stock level. However, this conclusion can not be applied in the extended
model (also not in the simple model). In fact, it is difficult to know whether
the equilibrium resource stock is greater or smaller than XMSY in the extended model as a result of the introduction of the marginal poaching effect and the marginal non-consumptive-value effect, as discussed in section
7.4. In any case, it can happen in the extended model that the equilibrium
resource stock is greater than XMSY , if renewable resources can generate
sufficiently great marginal non-consumptive value so that the marginal nonconsumptive-value effect dominates the steady state solution. For some cases
in which the species have a maximum sustainable yield stock level close to
the carrying capacity, this implies that the equilibrium resource stock level
may be quite high, if the non-consumptive value of the species are highly
appreciated, and/or if some other conditions discussed later are appropriate.
Certainly, under some inappropriate conditions, the possibility can not be
excluded that the equilibrium resource stock is small enough in some cases
that people will be seriously concerned about the viability of the resource
and the loss of its ecological functions. What these conditions are will be
investigated later.
By the application of the comparative static analysis in section 7.6, some
important parameters affecting the equilibrium resource stock and sustainability are identified. Of the six parameters, the intrinsic growth rate of
species is a well-known biological factor. The outcome of the comparative
static analysis shows that an increase in the intrinsic growth rate will raise the
equilibrium resource stock, if the equilibrium resource stock is smaller than
or equals to the maximum sustainable yield stock level. In the cases which
the equilibrium resource stock is greater than the maximum sustainable yield
stock level, the comparative static effect is ambiguous. Nonetheless, empirical studies generally support the conclusion that species with low intrinsic
growth rate are less resilient to harvest. Accordingly, we may generally conclude that an increase in the intrinsic growth rate will raise the equilibrium
resource stock, and vice versa.
The comparative static analysis addresses also the comparative static effects of the other parameters on the equilibrium resource stock. In sum, the
lower the discount rate, the marginal poaching rate and the depreciation rate
of management capital, and the higher the marginal non-consumptive value
and the efficiency coefficient for the management capital is, the higher the
116
equilibrium resource stock will be. Accordingly, we can use these parameters as indicators for evaluating the success probability of a sustainable use
project before or when it is practiced. The sustainable use strategy may potentially be more appropriate in sites with more positive indicators, namely
high marginal non-consumptive value, intrinsic growth rate and efficiency coefficient for the management capital, and low discount rate, marginal poaching rate and depreciation rate of management capital, than those sites with
less positive indicators.
117
Appendix 7.1
By keeping all exogenous variables constant except a certain one, the
equation (7.30) can be rewritten as
JMi = di , i = ρ, r, α, β, γ, δ
where


ρ − 2ρX − α γ
0
−h0

0
0
I0 
−δ
,
J =

2ρλ
0 r − ρ + 2ρX + α
0 
0
0
r+δ
−γ
dX ∗ dK ∗ dλ∗ dµ∗ 0
,
,
,
)
dρ dρ dρ dρ
dX ∗ dK ∗ dλ∗ dµ∗ 0
,
,
,
)
(
dr dr dr dr
dX ∗ dK ∗ dλ∗ dµ∗ 0
,
,
,
)
(
dα dα dα dα
dX ∗ dK ∗ dλ∗ dµ∗ 0
,
,
,
)
(
dβ dβ dβ dβ
dX ∗ dK ∗ dλ∗ dµ∗ 0
,
,
,
)
(
dγ dγ dγ dγ
dX ∗ dK ∗ dλ∗ dµ∗ 0
,
,
,
)
(
dδ dδ dδ dδ
Mρ = (
Mr =
Mα =
Mβ =
Mγ =
Mδ =
(A.7.1)
(A.7.2)
(A.7.3)
and
dρ
dr
dα
dβ
dγ
dδ
=
=
=
=
=
=
(−X + X 2 , 0, λ − 2Xλ, 0)0
(0, 0, −λ, −µ)0
(X, 0, −λ, 0)0
(0, 0, 1, 0)0
(−K, 0, 0, λ)0
(0, K, 0, −µ)0 .
(A.7.4)
The steady state value of X, K, λ and µ are denoted by asterisks. By application of Cramer´s rule, the comparative statics can be derived:
Mij =
|Jij |
|J|
118
where Mij is the jth element of vector Mi , and Jij is the matrix J with
its column j substituted by the vector i. As a result of the assumption
of proposition 6, it is clear that |J| > 0. Therefore, the signs of Mij are
determined by the signs of |Jij |. Some routine calculations yield that
(A.7.5)
|Jρ1 | = δ(r + δ)(X − X 2 ) (r − ρ + 2ρX + α)
0
0 2
−δ(r + δ)h (λ − 2Xλ) + I γ (λ − 2Xλ)
|Jr1 | = δ(r + δ)λh0 − I 0 µγ (r − ρ + 2ρX + α) − I 0 γ 2 λ
|Jα1 | = −δ(r + δ)X (r − ρ + 2ρX + α) + δ(r + δ)λh0 − I 0 λγ 2
|Jβ1 | = −δ(r + δ)h0 + I 0 γ 2
|Jγ1 | = δ(r + δ)K (r − ρ + 2ρX + α) + I 0 λγ (r − ρ + 2ρX + α)
|Jδ1 | = (r − ρ + 2ρX + α) [−I 0 γµ − (r + δ)γK] .
From (7.29), we know that r − ρ + 2ρX + α = βλ > 0 in steady state. And
(7.19) and (7.21) show that h0 < 0 and I 0 > 0, respectively. By applying
these results, it can be easily shown that |Jρ1 | > 0, if X < 12 or X = 12 ,
∗
and the sign of |Jρ1 | is ambiguous , if X > 12 . This applies also to dX
, since
dρ
|J| > 0. Similarly, it is obvious that |Jr1 | < 0, |Jα1 | < 0, |Jβ1 | > 0, |Jγ1 | > 0
∗
∗
∗
∗
and |Jδ1 | < 0. It follows that dX
< 0, dX
< 0, dX
> 0, dX
> 0 and
dr
dα
dβ
dγ
dX ∗
< 0.
dδ
In the same way, the following determinants can be expressed as
£
¡
¢
¤
(A.7.6)
|Jρ2 | = γI 0 2ρλ X 2 − X − (λ − 2Xλ) (ρ − 2ρX − α)
0
0
0
|Jr2 | = I λγ (ρ − 2ρX − α) + 2ρλI µh
+I 0 µ (ρ − 2ρX − α) (r − ρ + 2ρX + α)
|Jα2 | = I 0 γ [λ (ρ − 2ρX − α) + 2ρλX]
|Jβ2 | = I 0 γ (ρ − 2ρX − α)
|Jγ2 | = −2ρλ2 I 0 h0 − 2ρλγKI 0
−λI 0 (ρ − 2ρX − α) (r − ρ + 2ρX + α)
|Jδ2 | = [K(r + δ) + I 0 µ] [(ρ − 2ρX − α) (r − ρ + 2ρX + α)]
+2ρλK(r + δ)h0 + 2ρλµI 0 h0 .
The term ρ−2ρX −α exists in all of the above expressions. As a result of the
fact that it is not possible to determine the sign of the term ρ − 2ρX − α, the
∗
∗
∗
∗
∗
∗
, dK
, dK
, dK
, dK
and dK
are ambiguous,
comparative static effects dK
dρ
dr
dα
dβ
dγ
dδ
if no further assumption is applied.
119
Chapter 8
Management capital, use of renewable
resources, poaching and anti-poaching:
a general bioeconomic model
In this chapter we will further investigate the interaction between use of
renewable resources, management capital accumulation, resource stock and
poaching activities in a more general model. To do this, the extended model
of chapter 7 will be generalized in the sense that, instead of applying the
additively separable poaching function W (X) − Y (K) and utility function
U(h) + V (X) for sake of technical simplicity, a general poaching function
W (X, K) and utility function U (h, X) will be introduced in this chapter.
The necessary conditions for the optimal policy will be derived. The existence property of the steady state solution of the model will also be presented.
By application of computer simulation, the relevant phase diagrams and the
impacts of exogenous parameters on the equilibrium resource stock will be
studied. The implications of the model for conservation policy will be addressed in section 8.5 and 8.6.
8.1 The general model
In comparison to the extended model, two fundamental modifications are
here introduced to develop the general model. First, a more general poaching
function W (X, K), rather than the additively separable poaching function
in the extended model, is applied here that the poaching rate depends on
both resource stock and management capital stock. The function W (X, K)
represents the resource stock which is illegally exploited and possesses the
following properties
WX > 0, WXX > 0, WK < 0, WKK > 0, WXK < 0
(8.1)
2
and WXX WKK − WXK
> 0.
The convexity assumption of W (X, K) in X implies that, with an increasingly marginal poaching rate, a higher resource stock will induce a higher
poaching rate. The convexity assumption of W (X, K) in K means that a
higher level of management capital will reduce the poaching rate, but it is
limited by a decreasingly marginal effect for per unit of additional management capital.
120
The second modification involves the functional form of the utility function, namely, the additively separable utility function in the extended model
is here replaced by the general utility function U(h, X),which simultaneously
considers the consumptive value and non-consumptive value of renewable resources. We assume that U(h, X) can be measured in monetary terms and
is strictly increasing and concave in both X and h
UX > 0, UXX < 0, Uh > 0, Uhh < 0, UXh > 0
(8.2)
2
and Uhh UXX − WXh
>0
where UX represents the marginal non-consumptive utility generated by an
additional unit of resource stock, and Uh denotes the marginal gross harvest
profit from an additional unit of harvest which in turn equals the differential
between the unit resource price and the unit harvest cost. In addition, it is
assumed that
lim UX (h, X) = 0
(8.3)
X→X
, meaning that the marginal non-consumptive utility of resource stock equals
zero when the resource stock approaches the carrying capacity.
Following the previous modifications, the equation of motion for the exploited resource stock X must also be modified, if legal harvest and poaching
activities are taken into account:
·
X≡
dX
= F (X) − W (X, K) − h.
dt
(8.4)
The functional forms of the other functions, including the equation of motion
for management capital,
·
K = I − δK
(8.5)
and the investment cost function, C(I), and the relevant assumptions remain
unchanged. The meanings of the notations δ and r and their properties
remain also unchanged.
Now let us consider the same scenario as considered in section 7.3, that
certain people or organizations are given the legal rights to exploit renewable
resources in specific protected areas or in buffer areas around protected areas,
and they are authorized to manage natural resources and human activities in
121
those areas. The resource management problem of the resource owner is to
decide the optimal harvest rate and the investment rate to maximize the net
present utility (in monetary terms) derived from protecting and harvesting
the resource, subject to the constraint of dynamics of the resource stock and
management capital. This problem can be formally stated as
Z ∞
Max
[U (h, X) − C(I)] e−rt dt
(8.6)
0
·
s.t. X = F (X) − W (X, K) − h
·
K = I − δK.
The optimization problem has therefore two state variables, resource stock
and management capital stock, and two control variables, harvest rate and
investment rate, both of which are to be chosen optimally over time. Next,
the problem can be analyzed by the application of the maximum principle.
The corresponding current-value Hamiltonian is
H = U(h, X) − C(I) + λ [F (X) − W (X, K) − h] + µ [I − δK]
(8.7)
where λ is the current-value costate variable associated with the state variable X which gives the imputed marginal value of the resource stock, and
µ is the current-value costate variable associated with the state variable K
which gives the imputed marginal value of the management capital stock.
Assuming an interior solution, the first order necessary conditions describing
the optimization problem are given by equations (8.8)-(8.11) together with
(8.4) and (8.5):
∂H
= Uh − λ = 0
∂h
(8.8)
∂H
= −CI + µ = 0
∂I
(8.9)
λ = (r − FX + WX )λ − UX
(8.10)
·
·
µ = (r + δ)µ + WK λ.
122
(8.11)
Under given assumptions, the Hamiltonian is apparently concave in the
state variables X, K and in the control variables h, I. The second order conditions are therefore satisfied. Along the optimal trajectory, equation (8.8)
shows that the imputed marginal value, or the shadow price of an extra resource stock, λ, must be equal to the marginal gross profit of harvesting one
unit of renewable resources, Uh . The equation (8.9) means that the shadow
price of an extra management capital stock, µ, must equal the marginal cost
of investment in management capital, CI . Equation (8.10) implies that the
·
change rate of the shadow price of the resource stock, λ, plus the marginal
non-consumptive value of the resource stock, UX , must be equal to the opportunity cost when resource owner goes on to keep one unit of resource
stock, (r − FX + WX )λ. The opportunity cost includes the component of
forgone ´interest´, rλ, and the loss derived from marginal poaching rate as a
result of the increased resource stock, WX λ, minus the gain derived from the
marginal growth rate of the resource stock, FX λ. Finally, the equation (8.11)
indicates that the change rate of the shadow price of the management capital
·
stock, µ, plus the gain derived from the reduction of the poaching rate as a
result of the devotion of an extra unit of management capital, −WK λ, must
be equated with the opportunity cost when resource owner devotes one unit
of management capital stock to the protection of resources. The opportunity
cost comprises the component of the ´interest cost´, rµ, and the capital loss
as a result of the depreciation, δµ.
8.2 Existence of the steady state solution
Now let us verify the existence property of the steady state solution by
virtue of proposition 9.29
Proposition 9 Under given assumptions with regard to the poaching function W (X, K), the utility function U (h, X), the investment cost function
C(I) and the natural dynamics of the resource stock F (X), the dynamic
system (8.4), (8.5), (8.10) and (8.11) possess at least one steady state solution.
·
·
·
·
Proof. In steady state, X = K = λ = µ = 0. Consider both h and K as
29
The proof of the proposition 9 is inspired by the discussion on existence property of
the steady state equilibrium of the two-state-and-two-control-variables model developed
by Li and Lőfgren (1998).
123
·
an implicit function of X. By dividing λ by λ, we can define a function
·
UX (h(X), X)
λ
Γ(X) = = (r − FX + WX (X, K(X))) −
.
λ
Uh (h(X), X)
(8.12)
Furthermore, let X be the resource stock level which satisfies the condition
−
·
FX (X ) − WX (X , K(X )) = r. When λ = 0, it also implies that Γ(X) = 0,
−
−
−
and the following result can be derived from (8.12):
FX (X ∗ ) − WX (X ∗ , K(X ∗ )) = r −
UX (h(X ∗ ), X ∗ )
<r
Uh (h(X ∗ ), X ∗ )
where X ∗ denotes the equilibrium resource stock level. Since FX is monotonically decreasing and WX monotonically increasing in X, it can be easily
verified that X ∗ > X .
−
(h(X),X)
Finally, at X = X , we know Γ(X) = − UUXh (h(X),X)
< 0. When X ap−
proaches the carrying capacity X, it can be shown that Γ(X) = (r − FX +
WX (X, K(X))) > 0, since it is assumed in (8.3) that limX→X UX (h, X) = 0.
Under the assumption that Γ(X) is a continuous function of X, the results
that Γ(X) < 0 at X and Γ(X) > 0 at X guarantee that there must be at
−
least one X ∗ ∈ (X , X) which satisfies the condition Γ(X) = 0. Therefore,
−
the dynamic system possesses at least one steady state solution.
8.3 Phase diagram analysis: computer simulation
As generally recognized, it is not possible to analytically depict phase
diagrams for nonlinear differential equations in models involving multiple
state and control variables because of the interdependence of the variables
(Li and Lőfgren, 1998). Hence, with the help of computer simulation, we apply numerical methods in this section to conduct the phase diagram analysis
and to investigate the qualitative properties of the solution of the dynamic
problem. To do this, a modified version of the computer program originally
developed by Martin Quaas (personal communication) is completed by using MAT HEMAT ICA (see appendix 8.1). 30 The basic procedure of the
computer simulation is demonstrated as follows.
30
I thank Martin Quaas, of the Interdisciplinary Institute for Environmental Economics
of University of Heidelberg, Germany, for his generosity to share his idea with me. All
remaining errors are mine alone.
124
First, the following functional forms are specified:
F (X) = ρX(1 −
1
X
)
100
(8.13)
1
U (h, X) = 100X 2 h 2
1 2
C(I) =
I
40
X2
W (X, K) =
2K
where ρ = 1. In addition, we specify a discount rate r = 0.05 and a depreciation rate of the management capital δ = 0.5. By using the similar technique
applied in sections 7.6.1 and 7.6.2, we can transform the differential equation
system (8.4), (8.5), (8.10) and (8.11) into the following system:
·
X = F (X) − W (X, K) − h
(8.14)
·
K = I − δK
·
·
1
h =
((r − F´(X) + WX )Uh − UX − UhX X)
Uhh
·
1
((r + δ)C´+ WK Uh ).
I =
C´´
Under these specifications, the steady state solution (X ∗ , K ∗ , h∗ , I ∗ ) of the
differential equation system (8.14) can be obtained, with X ∗ = 52.1514,
K ∗ = 202.945, h∗ = 18.253 and I ∗ = 101.472. The Jacobian of (8.14)
evaluated at the steady state solution can also be obtained:


−0.3
0.0330176 −1
0
 0

−0.5
0
1
.
(8.15)
J =
 −1.26

0.0577807 0.35
0
−2.67538 0.55
1.52879 0.55
The eigenvalues are ζ 1 = 1.27642, ζ 2 = −1.22642, ζ 3 = 0.815694 and
ζ 4 = −0.765694, with respective eigenvectors e1 = (0.66, −0.49, 0.12, −0.86),
e2 = (0.38, −0.70, 0.33, 0.51), e3 = (−0.03, −0.60, 0.02, −0.80) and e4 =
(−0.12, −0.96, −0.09, 0.25). Therefore, the differential equation system (8.14)
is saddle point stable.
Next, we determine a neighboring point s0 , which is located on the
convergentPsaddle point trajectory, to the steady state solution by setting
s0 = s∗ + 4i=1 gi ei , where s∗ = (X ∗ , K ∗ , h∗ , I ∗ ) is the steady state solution,
and gi are arbitrary tiny numbers, for example 0.01. To ensure that s0 is
125
located on the convergent saddle point trajectory, we specify gi = 0 for i
with ζ i > 0. Using s0 as the initial point at t = t0 , and solving the differential equations (8.14) by numerical method, we can obtain all the points on
the convergent saddle point trajectory from t0 to tn in a way of time reverse
where t0 > tn , and finally draw them on the phase diagrams. Similarly, the
convergent saddle point trajectory from an another direction can be drawn
by specifying gi as a negative tiny number, for example −0.01. Accordingly,
both the phase diagram projected on the (X, h) plane conditional on the optimality of the other variables and the phase diagram projected on the (K, I)
plane conditional on the optimality of the other variables can be depicted,
as figures 8.1 and 8.2 show, respectively.
The two trajectories ´o.p.´ in Figure 8.1 and 8.2, which represent the
stable saddle point trajectories, converge to the equilibrium point. It means
that, on the (X, h) plane, corresponding to each initial resource stock level,
an unique corresponding value of harvest rate could be chosen on the stable
trajectory. It follows that, on the optimal dynamic path, the resource stock
and the harvest rate increase over time, if the initial resource stock level is
less than the steady state resource stock. On the other hand, if the initial resource stock level is higher than the steady state resource stock, the resource
stock and the harvest rate decrease simultaneously over time on the optimal
dynamic path. The economic meaning of this outcome can be explained as
follows. The more resource stock people have, the more they would harvest
without influencing the long-run survival of the resource, or in other words,
in order to reach the steady state, people would harvest more than the steady
state harvest rate when the resource is in relative abundance, in the sense
that the resource stock is higher than the steady state resource stock. And
they would harvest less than the steady state harvest rate when the resource
is relatively scarce, in the sense that the resource stock is less than the steady
state resource stock.
Similarly, on the (K, I) plane, corresponding to each initial management
capital stock level, a unique value of investment rate could be chosen on the
stable trajectories. On the optimal dynamic path, the management capital
stock increases while the investment rate decreases over time, if the initial
management capital stock level is less than the steady state management
capital stock. On the contrary, if the initial management capital stock level
is higher than the steady state stock level, the management capital stock
decreases over time on the optimal dynamic path while the investment rate
increases. The economic meaning of this result can be explained as follows.
126
The less management capital stock people have, the more investment would
be needed in order to accelerate capital accumulation and thereby to reach
the steady state. In other words, people would devote less investment than
the steady state investment level when the management capital is in relative
abundance, in the sense that the management capital stock is greater than
the steady state capital stock. And they would invest more than the steady
state investment rate when the management capital is relatively scarce, in
the sense that the management capital stock is less than the steady state
capital stock.
40
35
.
h=0
30
h
25
X=0
20
15
10
5
O.P.
20
40
X
60
80
100
Figure 8.1. Phase diagram on the (X, h) plane
250
.
I=0
200
I
150
100
50
O.P.
.
K=0
50 100 150 200 250 300 350 400
k
Figure 8.2. Phase diagram on the (K, I) plane
With reference to the time path of the poaching rate, we know that, ceteris
paribus, it will decrease when the management capital stock increases, and
127
vice versa. However, the poaching rate depends also on the resource stock
level. Thus, unlike the interaction between resource stock and harvest rate
or between management capital and investment rate, the development trend
of the poaching in general can not be determined when the resource and
management capital stock varies. It depends mainly on the initial conditions
with regard to resource and management capital stock. To understand this,
we may consider the following different scenarios.
First, if the initial resource stock is in relative abundance but management
capital is scarce, the resource stock will then decrease while management capital stock increase. It follows that the poaching rate will decrease according
to the assumptions of (8.1). On the contrary, if the initial resource stock
level is relatively low but management capital is in relative abundance, the
resource stock will then increase while management capital stock decrease.
This will induce a higher poaching rate. In an another scenario that both the
initial resource and management capital are scarce, they will increase simultaneously. An increase in resource stock will induce more poaching, whereas
an increase in management capital curtails poaching rate. Therefore, the net
effect on poaching rate is ambiguous and depends on the relative strength of
the effects derived from resource and management capital stock. It is possible that the building of management regime effectively contributes to the
growth of resource stock and simultaneously effectively dampens poaching
activity to such a low level so that it results in a net effect of a decrease in
poaching rate. However, it may also happen that a conservation project is
so successful that the recovery of resource stock induces more poaching than
before, although the devotion of management capital has to certain extent
slowed down the growth of poaching activity. A similar but reverse rationale
can be applied to the case in which both the initial resource and management
capital stock level are relatively high.
The previous discussion suggests that, as an indicator for evaluating the
success of a management regime, the equilibrium resource stock may be more
appropriate than the poaching rate. The conclusion outlined here is opposed
to the point of view expressed by Lewis and Phiri (1998). They suggested in
a case study in Zambia that wire snare counts, a proxy for poaching activity,
can be treated as an indicator for evaluating the success of the communitybased conservation projects. However, we asserts here that a reduction in
poaching does not necessarily imply the success of a conservation project, if
no or only few renewable resources exist in reserves and poachers are therefore
not interested on them at all. Likewise, the variation of the management capital stock does not represent the success or failure of a conservation project.
128
For example, a conservation project may fail, because inappropriate exogenous conditions dominate the project and lead to a low equilibrium resource
stock level, while the management capital stock increases as a result of the
low initial management capital stock.31 After all, the renewable resources
themselves, rather than the poaching rate, management capital, investment
or harvest rate, are the ultimate concern of the conservation policy. From
the aspect of conservation, whether the equilibrium resource stock can satisfy the ecological criteria of a sound ecosystem, is the ultimate indicator for
judging the extent of success of a conservation policy.
8.4 Comparative static analysis: computer simulation
In addition to the phase diagram analysis, we can also make use of computer simulation to conduct a comparative static analysis, and thereby to
investigate the influence of permanent changes in the exogenous parameters
on the equilibrium resource stock. To study the impact of as many parameters as possible, some special parameters will be introduced in the functional
forms of F (X), W (X, K), U (h, X) and C(I). In what follows, we specify
X
F (X) = ρX(1 −
)
(8.16)
100
1
1
U (h, X) = 2βX 2 τ h 2
1 2
σI
C(I) =
2
αX 2
.
W (X, K) =
γK
The exogenous parameter ρ, as usual, denotes the intrinsic growth rate of the
resource stock. The functional form of F (X) shows that the carrying capacity
of the environment equals 100. The parameters α, β, γ, σ and τ represents
exogenous social, economic, cultural, natural or institutional factors which
influence marginal poaching rate, marginal non-consumptive value of the
resource, marginal efficiency of the management capital, marginal cost of
investment in management capital and marginal gross profit of the harvest,
respectively. For convenience, we name these parameters α, β, γ, σ and τ
the poaching coefficient, the non-consumptive value coefficient, the efficiency
coefficient of management capital, the cost coefficient of investment and the
gross profit coefficient of harvest, respectively.
The marginal poaching rate, WX , measures the impact of one extra unit
resource stock on the poaching rate, and it can be influenced by many exogenous factors. For example, compared to protected areas located in the
31
These exogenous conditions will be investigated in the next section.
129
heart of the Amazon rain forests, an area of easy access (say because it is
in close vicinity to a highway) would have an overall higher marginal poaching rate, or in other words, the poaching function would have a greater α
coefficient.32 As a result of the same geographic factor, it might also have
an overall lower marginal efficiency of the management capital, WK , which
measures the influence of one extra unit management capital on the poaching
rate, or a smaller γ. The marginal non-consumptive value of the resource,
UX , which represents the influence of one extra unit of resource stock on human utility level, is usually dominated by aesthetic or cultural factors. For
example, the panda bear has generally much higher overall level of marginal
non-consumptive value, or greater β, than most kinds of snake. Other things
being equal, beautiful scenery of a national park often contributes to overall
higher level of marginal non-consumptive value of the wild species living in
that park. Similarly, some institutional factors have great impact on the
marginal gross profit of the harvest, Uh , which denotes the gross profit (unit
resource price minus unit harvest cost) derived from one extra unit of harvest. For example, an absolute ban on hunting and trading of wildlife, if it
is successful, usually leads to an overall lower level of marginal gross profit
of the harvest of wild species, or a smaller τ , because the resource owner
cannot use and sell their resources, so that the marginal gross profit of harvesting wildlife will be lower or even become zero. Finally, the parameter σ
determines the overall level of the marginal cost of investment in management capital, CI . In a country with higher price level or higher unit labor
costs, for example, the parameter σ would be potentially greater and leads
to a higher overall level of the marginal cost of investment in management
capital.
Together with the discount rate r and the depreciation rate of the management capital δ, the differential equations system encompasses four endogenous variables, X, K, h, I, and six exogenous parameters, r, δ, α, β, γ,
σ, ρ and τ . After repeated tests under different parameter combinations, the
results of computer simulation show unambiguously that, other things being
equal, the steady state resource stock will increase, when β, γ, ρ and τ are
increased, and when r, δ, α and σ are decreased (see Table 8.1). Apart from
the two new parameters σ and τ introduced in this chapter, all results of the
computer simulation are consistent with those found by comparative static
analysis in section 7.6.
32
Another example is, as Hurt and Ravn (2000) indicated, poaching pressures are positively correlated with the human population density throughout Africa. The more people
living in the vicinity to wilderness areas, the more heavy the poaching pressure.
130
Table 8.1 Results of the comparative static analysis: computer
simulation.
r δ α β γ σ ρ τ
Comparative static effect on − − − + + − + +
equilibrium resource stock
With reference to the comparative static effect of the cost coefficient of
investment on the equilibrium resource stock, the model result is understandable. A decrease in the cost coefficient of investment will reduce the overall
level of the marginal cost of investment in management capital, thereby induce more investment in management capacity, and in turn result in a higher
level of equilibrium resource stock. Of the new comparative static results in
this section, what especially worth noting is the impact of an variation of
the gross profit coefficient of harvest on equilibrium resource stock. Our
model suggests that, other things being equal, an increase in the gross profit
coefficient of harvest will lead to a higher equilibrium resource stock level.
This result is consistent with what asserted by Swanson (1994), and contrary
to the conclusion drawn by Clark (1973). In his model, Swanson asserted
that an increase in the price/harvest cost ratio (unit resource price divided
by unit harvest cost) will contribute to the growth of the population of the
harvested species, and this conclusion is the reverse of Clark´s. The reason
for this difference is, that both Swanson´s and our models take the factor of
the evolution of management capacity into account, while the Clark model
did not. In the context of the Clark model, an increase in the price/harvest
cost ratio enhance the incentive to harvest resources, without a concomitant
increase in the resource stock resulting from the devotion of a higher management capital. However, in the context of our model, a higher gross profit
coefficient of harvest results in increased shadow price of resource stock λ,
and this in turn raise the shadow price of management capital stock µ, as
equation (8.11) indicates. Therefore, more capital will be devoted to the
management of resources, and finally result in a higher equilibrium resource
stock.
Can the conclusion be drawn from our model results that a higher gross
profit coefficient of harvest (or a higher price/harvest cost ratio in the context
of the Clark model and the Swanson model) can contribute to the conservation of harvested species? It does not necessarily, since the model result
is based on the premise of ´other things being equal´. As discussed previously, the gross profit coefficient of a harvested species is determined by many
exogenous factors. Sometimes, we may try to manipulate the gross profit coefficient in favor of our policy objectives through rearranging the institutional
131
framework which dominates the gross profit coefficient. However, in the real
world, there is usually a certain correlation between different parameters,
especially between the gross profit coefficient and the poaching coefficient.
For example, in order to raise the gross profit coefficient of wildlife, we may
lift the ban on wildlife hunting and trading prevailing in many countries.
But lifting the ban will probably influence the poaching coefficient, usually
raise it simultaneously, because both legal resource owner and poachers are
motivated to harvest wildlife after the ban is removed. Therefore, we should
consider the situation in which both the gross profit coefficient and the poaching coefficient are increased or decreased. Analytically, the final result must
be ambiguous, since the increased gross profit coefficient raises, and the increased poaching coefficient reduces the equilibrium resource stock, and we
do not know exactly which effect will dominate the final outcome. Nonetheless, with the help of computer simulation, we can make simulations under
different scenarios to explore, when the equilibrium resource stock will increase or decrease if the gross profit coefficient and the poaching coefficient
vary simultaneously and in the same direction.
As an example, let us make use of the functional specifications of (8.16),
and specify ρ = 1, δ = 0.5, r = 0.05, β = 50, σ = 0.05 and γ = 2 for running
simulation program. Table 8.2 demonstrates some of the simulation results
under different τ -α combinations. For example, under the initial scenario
τ = 0.1 and α = 0.1, it yields a equilibrium resource stock X ∗ = 58.79. When
τ and α are increased to 3 and 0.2 respectively, the equilibrium resource stock
increases also to 61.78. In fact, as long as τ is increased to 3 and α is smaller
or equal to 0.5, the equilibrium resource stock will increase, compared to
the initial scenario. On the other hand, when τ is increased to 3 and α
is increased to 0.6 or more, the equilibrium resource stock will be smaller
than that under the initial scenario. The ´break even´ point for α, in the
sense that at which the equilibrium resource stock remains unchanged, lies
somewhere between 0.5 and 0.6. In general, we can conclude that, whether
the equilibrium resource stock will increase or decrease depends on the extent
to how τ and α positively correlate to each other.33 If the increase in τ is
accompanied by only slight increase in α, it will result in a higher equilibrium
resource stock. Contrarily, if the increase in τ is accompanied by substantial
increase in α, it will possibly lead to a decrease in the equilibrium resource
stock.
33
It is evident that the comparative static effect of a shift of τ will be unambiguous, if
there is negative correlation between τ and α.
132
Table 8.2 Computer simulation results under different
τ = 0.1 τ = 0.5 τ = 1 τ = 1.5 τ = 2
α = 0.1 58.79
61.32
62.07 62.43
62.66
α = 0.2 55.34
59.22
60.39 60.96
61.32
α = 0.3 52.56
57.51
59.00 59.74
60.21
α = 0.4 50.17
56.00
57.78 58.66
59.22
α = 0.5 48.03
54.64
56.68 57.69
58.33
α = 0.6 46.10
53.39
55.60 56.79
57.51
α = 0.7 44.33
52.23
54.71 55.94
56.73
α = 0.8 42.69
51.14
53.81 55.15
56.00
α = 0.9 41.15
50.11
52.96 54.39
55.31
α=1
39.72
49.13
52.15 53.67
54.64
τ -α combinations
τ = 2.5 τ = 3
62.83
62.96
61.58
61.78
60.54
60.80
59.63
59.93
58.79
59.15
58.02
58.42
57.30
57.73
56.61
57.09
55.97
56.47
55.34
55.88
It seems that the somewhat ambiguous conclusion can help policy-makers
consider conservation issues in their decision making process, if they know the
correlation between τ and α. But in the real world it does not help so much,
because in many cases no one knows exactly, ex ante at least, the extent to
which how τ and α positively correlate to each other. Just therefrom many
prevailing controversy about conservation policy arise. In next section, we
will illustrate the policy implications of our comparative static analysis with
two important empirical examples.
8.5 Policy implications of the comparative static analysis
with regard to the gross profit coefficient of species
and the poaching coefficient: two examples
8.5.1 Debate on conservation and consumptive use of the African
elephant
During the last two decades, the consumptive use and trading of wildlife
were one of the most controversial conservation issues. The experiences of the
African elephant conservation typify many elements of the ongoing debate
surrounding sustainable use approach versus preservation approach. Hence
we use the example here to demonstrate what our model would imply for
this case.
During the 1980s, the populations of the African elephant throughout
Africa has experienced a massive slaughter to an unacceptable extent. Driven
by the extraordinary high price of the elephant ivory in the international
wildlife products market, poachers slaughtered more than half of the elephant
population in Africa within ten years. According to estimates, poaching has
133
led to a significant decrease in African elephant population, from about 1.3
million in 1979 to six hundred thousand in 1989 (ITRG, 1989). This raised
serious public concern about the survival of the African elephant. To save the
African elephant, conservation organizations initiated worldwide campaigns
aimed at banning the ivory trade. Finally, after some attempts to regulate the
ivory trade had failed to dampen poaching activities, the general assembly
of the Convention on International Trade in Endangered Species of Flora
and Fauna (CITES) decided in 1989 that the African elephant are up-listed,
from CITES Appendix II, which allows controlled and monitored trading, to
Appendix I which absolutely bans any trading of species concerned (Duffy,
2000).
The logic behind the ban on ivory trade is easy to understand. By destroying the market for ivory, the ivory price will dramatically fall, and this,
together with the lack of access to legal market, in turn constitutes a disincentive for poachers to poach again. In fact, at least in the beginning several
years after ban was imposed, the ban seemed to have, to certain extent,
achieved its objectives. The ivory price has dramatically fallen, poaching
activities decreased (Duffy, 2000), and population of the African elephant
remained stable.
However, the influence of the trade ban on the poaching rate is not as
clear-cut as it was supposed to be by the proponents of the ban. Research by
the IUCN African Elephant Specialist Group suggested that, it was mainly
the increased efforts on law enforcement, rather than the ban per se, which
contributed to the decline of poaching at the beginning stage of ban (Dublin
et al., 1995). In addition, Zimbabwe claimed that poaching has virtually rose
after the ban as a result of a higher illegal market ivory price (Duffy, 2000).
Meanwhile, some active opponents of the ivory ban, mostly the Southern
African countries such as South Africa, Zimbabwe, Botswana and Namibia,
continued to argue that the ban has virtually punished them for their sound
wildlife conservation policy. They asserted that, through wise consumptive
use of elephant in which the trading of ivory played a critical role, they have
protected their elephant populations and wilderness more effectively than
those countries which banned the consumptive use of elephant, and thereby
they virtually conserved a sound and stable elephant population, even in the
1980s.34 In fact, Zimbabwe, as one of the most active proponents of the
34
For example, in 1989, the Kruger National Park in South Africa culled 350 elephants to
prevent overpopulation. The park authority sold the ivory and hides, and earned US$2.5
million which constituted 10% of its annual budget (Rasker et al., 2000).
134
sustainable use of wildlife, has increased its elephant population since 1981
(Rasker et al., 2000). Using their successful experiences in wildlife conservation, these countries continued to call for lifting of the total ban on ivory
trade. In 1997, they succeeded in down-listing their elephant populations to
CITES Appendix II, which meant that a restricted ivory trade with Japan
was allowed in 1999 under the premise that only legally harvested ivory can
be traded and the harvest should be sustainable. The pro-ban conservationists criticized the restricted lifting of ban because of the fear of losing control
of ivory trade and the concomitant resurgence of poaching. They argued
that, given the circumstances that it is extremely difficult to differentiate
illegal ivory from legal ivory, the illegal ivory trading will continue by way of
laundering under the cloak of the legal trading (Duffy, 2000).
The brief retrospect about the conservation of the African elephant shows
how complex the issue in the reality is. The central question is, whether the
ban on ivory trade, as a whole, is advantageous or disadvantageous to the
conservation of the African elephant, or more precisely, whether the ban
results in an increase or a decrease in elephant population. We are here
especially interested in the implication of our model for this question.
In Africa, except in some private reserves, the elephant populations are,
de jure, owned by state. Hence we can treat state government as the resource
owner in this case. The ban has succeeded in reducing ivory price (a smaller
τ in our terminology) and thereby reducing poaching (a smaller α) almost
throughout Africa (although its influence on poaching is not as significant
as it is supposed to be). For the countries with very low management capacity and in which access to elephant is virtually open, the decline of the
overall poaching level is undoubtedly a good news for elephant conservation.
However, for those countries which actively manage their elephant populations for consumptive or non-consumptive use, they are faced a smaller α,
advantageous for elephant conservation on the one hand, and simultaneously
a smaller τ , disadvantageous for elephant conservation on the other hand.
According to our model findings, the ´net´ increase or decrease in elephant
population derived from the ban will be ambiguous, unless we know, at least
roughly, how τ and α correlate to each other. In reality, it is usually difficult to obtain these informations because it needs extensive field studies.
Therefore, given current available informations, neither the ban nor lifting
of the ban can be unambiguously justified. Any change of 180 degrees of the
current policy will take too many risks. Some experiments with regard to
lifting the ban are worth being done, but only on an adaptive, trial and error
basis. In any case, decisions about whether and where the ban should be
135
lifted depend primarily on the correlation between ivory price and poaching
rate.
In addition, our model suggests that we may try to raise τ , while keep α
unchanged, to get a greater equilibrium resource stock. From this perspective, the safari hunting in Africa is a good alternative which will promote
elephant conservation through consumptive use. According to the recent
study made by Hurt and Ravn (2000), sport hunters usually pay a licence fee
of US$10,000 for a single elephant, and the total price of an average 21 day
safari hunting, where elephant is the main trophy, is more than US$40,000.
In general, the greatest cash return on a single elephant from consumptive
use is usually the licence fee paid by sport hunters. Therefore, the introduction of safari hunting substantially enhances the overall level of the marginal
consumptive value of elephant, or in other words raises τ . Most importantly,
the high price holds only for the elephant owner. For poachers, their cash
return from killing an elephant remains unchanged, and this guarantees that
the poaching coefficient α remains also unchanged. In this case, the introduction of safari hunting will unambiguously lead to a greater equilibrium
elephant population. This may, at least partly, explains why South Africa
and Zimbabwe, as two of the most active proponents of the sustainable use of
elephant and major safari hunting destinations, have a healthy and increasing
elephant population. Therefore, we may conclude that, from the perspective
of elephant conservation, safari hunting is a feasible policy option.
In fact, the same rationale holds also for other game species. As a result
of the boom of wildlife use, especially safari-hunting, almost every private
ranches and reserves in South Africa have re-introduced wildlife (Hurt and
Ravn, 2000), and this has resulted in steady increase in wildlife populations
(Grootenhuis and Prins, 2000) and increase in privately protected habitat
area which is even greater than the total land area under the control of the
National Park´s Board (Hearne and Mckenzie, 2000). Similar trend with reference to populations of wildlife species and land area devoted to wildlife use
can also be found in Zimbabwe where more than 30% of the country´s land
area are devoted to some form of wildlife use (Kock, 1996). Our model and
the experiences in Southern Africa clearly demonstrate the safari hunting´s
conservation component. The more lucrative safari hunting is, the more it
can contribute to the conservation of the utilized species, and most importantly, to the conservation of habitats and biodiversity sharing habitats with
the target species. According to our model, any consumptive use strategy
of renewable resources, which can enhance the overall level of the marginal
consumptive value of resources and simultaneously keep poaching unchanged
136
(or even reduce poaching), will be adequate option for conservation policy
instrument, provided the other parameters discussed in section 8.4 are in
principle positive or at least neutral from the perspective of conservation.
8.5.2 Conservation and consumptive use of wildlife in Taiwan
Taiwan had once a rich and diverse fauna world. But in the past several
decades, severe habitat degradation and poaching have significantly reduced
population levels of almost all wildlife species, especially those of traditionally important game species. Given that more than 50% of the land area
of Taiwan is still covered by healthy forests, poaching is, to great extent,
responsible for the disappearance of wildlife. For Taiwanese, the so-called
´wild meat´, which means the meat and any eatable parts of wildlife, are
valuable delicacies. The prices of wild meat are usually much higher than
those of domestic animals. The high prices drive many people, indigenous
or non-indigenous, to go hunting for commercial purpose, while some indigenous people retain their hunting tradition for subsistence or cultural purposes. Given the de facto open access state of wild species, those commercial
hunters rapidly harvested almost all of the wildlife resources of Taiwan.
To save wildlife from going extinction, an absolute ban on hunting of
wildlife was imposed by the central government in 1973. In 1989, the more
strict Wildlife Conservation Law, which follows the strict preservation model
and may be one of the most strict wildlife law throughout the world, was
enacted. According to the Wildlife Conservation Law, all wildlife species35
should be classified into two categories: (1) Protected Species, which include endangered, rare, valuable and other conservation-deserving species;
(2) General Wildlife, which include all wildlife species not included in Protected Species (WCL, 1994: Article 4). All wildlife species and their products
are prohibited from being disturbed, abused, hunted, killed, traded, exhibited, displayed, owned, imported, exported, raised or bred, except in some
special cases (WCL, 1994: Article 16, 18, 21 and 24).36 Hence, any use of
35
The Article 3 of the Wildlife Conservation Law defines wildlife as ´any animal living in
a natural habitat, including mammals, birds, reptiles, amphibians, fish, insects and other
kinds of animals.´ (WCL, 1994: Article 3).
36
The Wildlife Conservation Law permits hunting or utilization of wildlife species only
in the following cases: ´(1) when population size exceeds the carrying capacity of the area;
or (2) for academic research or educational purposes and with proper approval from the
NPA.´(WCL, 1994: Article 18) and ´(1) danger to public safety or human life; (2) damage
to crops, poultry, livestock or aquaculture; (3) being a disese vector of zoonoses or other
pathogens; (4) danger to the safety of air transportation; (5) for traditional cultural or
ritual hunting, killing or utilization needs of Taiwan aborigines living in reserved areas;
137
wildlife for commercial or subsistence purposes is virtually prohibited by the
Wildlife Conservation Law.
The enactment of the Wildlife Conservation Law and the concomitant increased efforts on law enforcement against poaching has succeeded in damping poaching during the last decade. Populations of most of the important
game species have, to certain extent, recovered (Pei, 2001). However, the absolute ban on wildlife use also induced the protest of the indigenous people
who call for legal hunting rights for both economic and cultural purposes.
The fact that some indigenous subsistence hunters were treated as poachers
by the Wildlife Conservation Law further intensified the conflicts between
indigenous people and government authorities, though usually only slight
punishments were inflicted on these hunters because of their special identity.
Given that protest continues to intensify and that populations of some important game species have apparently recovered, there are recently more and
more people who suggest modifying the Wildlife Conservation Law (and/or
the National Park Act) to allow legal hunting and utilization of wildlife, and
thereby to promote both conservation and sustainable development projects
initiated by indigenous communities (Liu, 2000). On the other hand, like the
debate prevailing throughout the world, many people are averse to this idea
because they fear the concomitant resurgence of poaching if the hunting ban
is lifted (Chang, 2001).
Should the ban on wildlife hunting and utilization be lifted in Taiwan?
Again, we assert here on the basis of the model findings that the answer
to this question is not simple yes or no. It depends, except the other six
parameters discussed in section 8.4, primarily on the influences of the policy
options on the marginal gross profit of harvest of wildlife and on the poaching
rate. If it involves the reopening of hunting and of trading of wildlife meat
and products, poaching will probably boom again while the marginal gross
profits of harvest of wildlife are increased, because it is practically impossible to differentiate illegal from legal hunted wild meat and wildlife parts,
regardless of in the market or on the dishes of the so-called ´wild meat restaurant´.37 According to author´s personal observations and of Chang (2001),
even under the control of the Wildlife Conservation Law, wild meat restaurants are still very popular in the remote country, and the ´raw material´ of
their delicacies are, to great extent, illegal hunted wildlife. Therefore, it is
(6) other reasons approved by the authorities.´ (WCL, 1994: Article 21). With reference
to cultural or ritual hunting, each indigenous tribe is allowed to hunt twice in a year.
37
´Wild meat restaurant´ is a special kind of restaurant in Taiwan which sell primarily
delicacies made from wild meat and wild species of plants.
138
reasonable to conjecture that, once legal trading of wildlife meat and parts is
reopened, the poaching rate will significantly increase under the cloak of the
legal trading. Given the current circumstance that the management capacity
of areas out of national parks is deficient and access to wildlife is practically
open, the negative impact on wildlife populations will be inevitable in national forests and nature reserves. For those community based conservation
projects aimed at sustainable use of wildlife and initiated by local people,
the significant increase in poaching rate will reduce their success probability,
although, according to our model, the higher marginal gross profit of harvest
of wildlife may compensate the negative effect derived from increased poaching rate. But even legal hunting and trading of wildlife can contribute to
an increase in wildlife populations in some private protected areas, it cannot
offset the population loss occurred in national forests and nature reserves,
because the great majority of the wilderness remains under the control of
the state.38 It follows clearly that, from the perspective of wildlife conservation, the combination of reopening hunting and trading of wildlife meat and
parts may not be an adequate policy option under current circumstances of
Taiwan.
To promote community based conservation projects, or to mitigate the
conflicts between indigenous people and state authorities, we may consider
the policy option of legalizing sport hunting (and fishing), while the other
forms of consumptive use of wildlife are still not allowed. As discussed
in subsection 8.5.1, sport hunting can significantly enhance the marginal
gross profit of harvest of target species, while the poaching rate remains
unchanged.39 This, as a whole, will result in an increase in equilibrium
population of the target species, and simultaneously promote the protection
of habitat. In Taiwan, some wildlife species which are resilient to harvest
and have relatively abundant populations, such as Formosan wild boar (Sus
scrof a taivana), Formosan hare (Lepus siensis f ormosanus), Formosan
giant flying squirrel (P etaurista granis) and Formosan white-faced flying
squirrel (P etaurista lena), can be adequate game species. Some heavily
harvested big game species, such as Formosan Reeve´s muntjac (Muntiacus
reevesi micrurus), Formosan sambar (Cervus unicolor swinhoei) and Formosan serow (Naemorhedus swinhoei), are also valuable and adequate game
38
The state-owned wilderness in Taiwan can be roughly divided into the following three
categories: national parks, nature reserves and national forests. In general, only national
parks are practically strictly protected. Nature reserves and national forests, de jure,
should also be protected but practically not as a result of the low management capacity
of the authorities concerned.
39
If those poachers, usually skilled hunters, can be transformed into legal hunting guides
or rangers through legalizing sport hunting, the poaching rate will probabily decrease.
139
species, but we still need scientific assessments to identify their population
levels before they are hunted. From the perspective of wildlife conservation,
sport hunting should be encouraged, and may be the only feasible form of
sustainable wildlife use in today´s Taiwan.
8.6 Concluding remarks and some implications for
conservation policy
In this chapter we have investigated the relevant properties of the general
model. By use of the computer simulation, the relevant phase diagrams and
the impacts of exogenous parameters on equilibrium resource stock have been
studied. The comparative static effects of the general model are similar to
those of the extended model presented in chapter 7, which is a special case
of the general model. The implications of the comparative static analysis
for specific species conservation policy have also been explored with two
examples. In addition, we assert here that the comparative static effects of
the general model can be used to assess the feasibility of the sustainable use
strategy applied in specific areas. We will show how this works in the case
study of the next chapter.
8.6.1 Some remarks
There are still some points that are noteworthy. First, like the policy implication of the simple model, the impact of the sustainable use approach on
conservation is double-edged in the sense that the sustainable use approach
will not necessarily result in a higher stock level of renewable resources and/or
management capital than before. In some cases, it can cause a lower stock
level of renewable resources and/or management capital. Moreover, the impact of the sustainable use approach on conservation is here far more complex
than that in the simple model. Figure 8.1 demonstrates, that the sustainable use approach will theoretically contribute to an increase in resource stock
through the adjustment of harvest rate and management capital stock, if the
initial resource stock is smaller than the equilibrium resource stock. Otherwise, if the initial resource stock level is higher than the equilibrium resource
stock level, the use approach will inevitably lead to a decrease in resource
stock until the equilibrium resource stock is reached. The time path of the
accumulation of management capital exhibits a similar pattern, as figure 8.2
shows. On the one hand, the management capital stock will increase through
the adjustment of investment rate, if the initial stock level is lower than the
equilibrium stock level. On the other hand, if the initial management capital
stock level is higher than the equilibrium stock level, the management capital
140
stock will decrease steadily until the equilibrium stock level is reached. From
the previous discussions it follows that the sustainable use approach will not
necessarily result in a better conservation status of renewable resources.
How the accumulation of the management capital influence the growth
of the resource stock is a crucial problem in which we are interested. The
general model demonstrates that an increase in management capital will, ceteris paribus, reduce the poaching rate and thereby contribute indirectly to
the accumulation of resource stock, but it does not guarantee an simultaneous increase in resource stock. On the contrary, a decrease in management
capital will cause a higher poaching rate and thereby contribute indirectly to
the detriment of resource stock, but it does not necessarily lead to an simultaneous decrease in resource stock. Whether the resource stock will grow or
decrease depends on the relative size of the initial resource stock compared
to the equilibrium resource stock, and on the corresponding adjustment of
the harvest rate. This conclusion does not imply that the accumulation of
the management capital does not play any role in the accumulation process
of the resource stock. It means that the accumulation of management capital influences but can not definitely determine the accumulation process of
the resource stock. The previous discussion suggests that the variation of
management capital stock is not an appropriate indicator for evaluating the
performance of the sustainable use approach as a conservation instrument.
We are also interested in the question how the poaching activity will develop after the sustainable use approach is applied. The discussion in section
8.3 exhibits that in the case which initial resource stock level is lower than
and management capital stock is higher than equilibrium stock level, and
in the case which initial resource stock level is higher than and management
capital stock is lower than the equilibrium stock level, the development trend
of poaching rate can be derived. However, in the cases which both initial resource stock and management capital stock is greater than the equilibrium
stock, and which both initial resource stock and management capital stock
is smaller than the equilibrium stock, the development trend of poaching
rate is ambiguous. In any case, as section 8.3 concluded, the variation of
the poaching rate is not an appropriate indicator for evaluating the success
of the sustainable use approach. From the aspect of conservation, whether
the equilibrium resource stock can satisfy the ecological criteria of a sound
ecosystem is the ultimate indicator for judging the extent of success of a conservation policy. Under the general premise that the more the equilibrium
resource stock closes to the carrying capacity, the better it would be for the
whole ecosystem, we can use the equilibrium resource stock as an indicator
141
for judging the extent of success of the use approach as a conservation instrument, and accordingly turn our attention to the factors which affect the size
of the equilibrium resource stock and the corresponding policy implications.
8.6.2 Policy implications with regard to the intrinsic growth rate
By the application of the comparative static analysis, some important parameters affecting the equilibrium resource stock are identified. Of the eight
parameters, the intrinsic growth rate of species is a well-known biological factor. The outcome of the comparative static analysis shows that an increase
in the intrinsic growth rate will unambiguously raise the equilibrium resource
stock. Based on this conclusion, some policy implications with reference to
conservation can be drawn.
At the individual species level, the higher the intrinsic growth rate of the
species is, the more appropriate the use approach will be for the management
of the species. For species with low intrinsic growth rates, a specially cautious
attitude toward the harvest problem must be taken. In general, long-lived
and slow-reproducing species, such as primates, elephants, whales and sharks,
have low intrinsic growth rates and may be particularly vulnerable to harvest
(Mangel et al., 1996).40 Moreover, the intrinsic growth rate is affected in
principle by two factors, i.e. Body size and Phylogeny. Larger animals tend to
have lower rates (Eisenberg, 1980), and, as a group, primates and carnivores
have generally lower intrinsic growth rates than expected from body size,
whereas ungulates and rodents have higher rates for their body size (Robinson
and Redford, 1986; Bennett and Robinson, 2000b).41
Next, as a whole, various habitat types can be characterized by different
reproductive capacities of total biomass. Table 8.3 illustrates estimates of the
mean net primary production and mean plant biomass for various habitat
types of the world.42 Under the premise that the production/biomass ratio,
just like the intrinsic growth rate at the individual species level, can be
treated as an appropriate indicator for the reproductive rate of the whole
plant biomass of specific habitats, we may conclude that, from the viewpoint
40
An example for the mahogany tree with low rate of increase and the possible fate of
its sustainable harvest see Gullison (1998).
41
For more detailed estimates about the intrinsic growth rate of individual species of
neotropical animals see Robinson and Redford (1991), and Robinson and Bennett (2000).
42
Stiling (1992) defined the concept net primary production as gross primary production
minus energy lost by plant respiration. And gross primary production is equivalent to the
energy fixed in photosynthesis. For more details about net primary production see Stiling
(1992) and Ricklefs (1990).
142
of plant conservation, the higher the production/biomass ratio of habitats
is, the more appropriate the use approach will be for the management of
the habitats. A comparison of the production/biomass ratios of the earth´s
major ecosystem types shows that, in general, the production/biomass ratio
is inversely related to the overall degree of forest cover. For instance, open
grasslands, including Savannah and temperate grassland, have much higher
production/biomass ratios than forest ecosystems. This may suggest that,
compared to open grasslands, forest ecosystems are particularly vulnerable
to overharvesting of plant communities. Such a general comparison neglects
the considerable differences in reproductive rates of plants between specific
species and regions, but this deficiency does not override the general patterns
it reveals.
With reference to the total faunal biomass and the relevant reproductivity of the earth´s major ecosystem types, there is nowadays no similar
general estimates like those of plant biomass, but some estimates in tropical
ecosystems regarding game biomass, in which people are especially interested,
reveal also specific patterns. A comparison of large mammal biomasses43 at
various tropical ecosystems made by Robinson and Bennett (2000) showed
that, in general, the overall standing mammalian biomass has a negative relation to the degree of forest cover, and this difference in mammalian biomass
can in principle be accounted for by the difference in ungulate biomass. It
lacks definite estimates regarding the overall reproductive rate of mammalian
biomass. However, based on the estimates of the components of the mammalian communities in tropical ecosystems44 , some important facts emerge.
The mammalian biomass of the tropical forests, especially those in tropical
rain forests, encompasses a much higher proportion of primates which have
generally low intrinsic growth rates, than open grasslands or habitats with a
mosaic of forest and grassland do. Contrarily, the mammalian communities
of the open grasslands are made up of almost only ungulates and rodents
which have generally much higher intrinsic growth rates than primates have.
Therefore, the conjecture that the weighted mean intrinsic growth rate of
the total mammalian biomass of tropical ecosystems is inversely related to
the degree of forest cover should be reasonable. This may suggest that, compared to tropical open grasslands, tropical forest ecosystems are particularly
vulnerable to overharvesting of mammalian communities. Again, it is notable that such a general conclusion neglects the considerable differences in
43
The large mammals are defined as the species with over 1 kg adult body mass. This
definition includes most of the important game species.
44
For more detailed data about the components of the mammlian biomass see Robinson
and Bennett (2000), pp. 17-18.
143
reproductivity between specific species and regions, though this deficiency
does not override the general patterns it reveals.45
8.6.3 Policy implications with regard to some other parameters
The comparative static analysis addresses also the comparative static effects of the other parameters on the equilibrium resource stock. In sum,
the lower the discount rate, the poaching coefficient, the cost coefficient of
investment and the depreciation rate of management capital, and the higher
the non-consumptive value coefficient, the gross profit coefficient of harvest
and the efficiency coefficient of management capital are, the higher the equilibrium resource stock will be. Accordingly, we can use these parameters as
indicators for evaluating the success probability of a sustainable use project
before or when it is brought into practice. The sustainable use strategy
may potentially be more appropriate in sites with more positive indicators
than those sites with less positive indicators. Based on this conclusion, the
following important policy implications emerge.
For convenience of discussion, we divide roughly all the countries or regions into two categories: the developed and developing countries (regions).
The two groups have markedly different features in some of the previous parameters. First, compared to the developed countries, the developing countries are generally characterized by higher levels of discount rate as a result
of the prevailing poverty (Bodmer et al., 1997b), the uncertainty with regard
to tenure, markets and population levels of exploited species (Freese, 1997),46
and high levels of inflation rate. As Clark (1973) demonstrated in his pioneer
paper, a high discount rate has a destructive effect on the sustainable use
of resources, since the relatively slow-reproductive resources will potentially
never generate a competitive return on owner´s investment, and then the
rapid depletion of the resources becomes a rational option. Given the generally prevailing high discount rate in developing countries, it is difficult for the
species with low intrinsic growth rates that the use strategy can create strong
enough incentives so that biological over-harvesting would not happen.47
45
For more examples discussing the differences in wildlife production between various
tropical forest types see Hart (2000) and Peres (2000).
46
As Freese (1997) explained, uncertainty with regard to tenure, markets and population levels of exploited species increases the risk premium of holding resource stock and
raises accordingly the discount rate. The developed countries have similar but less serious
problems than developing countries, because property rights are generally secured and
population levels of exploited species are, though not perfectly, far more known than in
developing countries.
47
Milner-Gulland and Mace (1998) illustrated an example of a tragically very high dis-
144
In addition, as a result of poverty and deficiency of legal enforcement, the
natural resources in developing countries are generally under a much greater
pressure of poaching activity than in developed countries. This implies that
the developing countries have generally a higher poaching coefficient. Moreover, it is in principle undoubted that the non-consumptive value of renewable resources is much more appreciated by the people in developed countries
than in developing countries. It follows that the non-consumptive value coefficients of the utility function in the developed countries should be generally
higher than those in developing countries.
In sum, compared to the developing countries, the developed countries
have some more appropriate socio-economic conditions when we are concerned about the feasibility of the sustainable use approach as a conservation
instrument. In other words, given the social and economic conditions, the
success probability of the sustainable use approach might be relatively low in
developing countries which are the focus of the current conservation practice
and of the debates about conservation issues. This might imply that a more
conservative attitude toward the application of the sustainable use approach
should be taken in developing countries, and the enthusiasm of some international organizations, local people and conservationists for the use approach
should be questioned. The previous policy implication holds especially for
the tropical rain forests where the overall reproductive rate of faunal and floral communities is low. If we additionally consider the fact that tropical rain
forests species are generally characterized by high diversity and low densities
(Owen, 1992), it is worthwhile being particularly cautious of reconsidering
the use approach to prevent from local or even global extinctions of species.
Contrarily, in developed countries a more active attitude toward the application of the sustainable use approach might be taken to supplement the traditional preservation approach.48 In tropical open grasslands where the overall
reproductive rate of plant and mammalian biomass is high, depending on the
socio-economic circumstances, a neutral attitude could be taken. Generally
speaking, as a result of the relatively low probability of success in developing
countries, the use approach can hardly create systems of a scale sufficient
to preserve large portions of ecosystems in developing countries which are
the focus of the current international conservation campaign.49 At most, it
count rate happened in the Ache, an indigenous group living in Paraguay. The high
discount rate led rapidly to the depletion of natural resources, after they received legal
title to their reservations in 1988. According to author´s personal experiences, this is not
an accidental example. The indigenous people in Taiwan have similar problems.
48
A retrospect about the successful experiences of sustainable harvest of wildlife in North
America see Shaw (1991).
49
The experience of South Africa may be an exception. As a result of the boom of
145
could play a supplementary role in the whole conservation policy. Whether
this implies that we should re-emphasize the importance of the preservation
approach as several decades ago did and apply it more intensively when creating new protected areas, or that another alternative approach should be
developed, is a critical question worth investigating.
Certainly, the previous general conclusion neglects the considerable differences in socio-economic and biological conditions between various countries,
regions and habitat types. In certain circumstances the sustainable use approach can succeed even in developing countries and in cases in which the
intrinsic growth rate of the harvested species is low, such as the successful use and conservation of the African elephant in Southern Africa (Roth,
1997a). For some species with high intrinsic growth rates, such as Wild boar
and feral pigs, the use approach becomes an old tradition around the world
and even a necessity to control the damage to agriculture they cause (Roth,
1997b), and the populations remain relatively abundant almost irrespective
of under which socio-economic circumstances.50 Therefore, whether the use
or the preservation approach is appropriate, depends always on the site- and
species-specific conditions. Nonetheless, the general conclusion affords a fundamental direction for the rethinking of the conservation policies.
the wildlife use, especially safari-hunting, about 4,000 private ranches and reserves have
devoted totally over 80,000 km2 land area to wildlife, compared with less than 10,000 km2
in 1979. This is also remarkable for its extent, compared with the total land area of about
28,000 km2 under the control of the National Park´s Board (Hearne and Mckenzie, 2000).
50
There are a few exceptions to this rule. Some of the localised Asian pig species are
certainly overharvested so that their populations decrease to a alarmingly low levels. For
detailed list of these species see Roth (1997b).
146
Table 8.3 Primary production and plant biomass of the earth´s
major ecosystem types
Ecosystem Type
M.N.P.P.1 M. B.2 P/B Ratio3 (%)
Continental
Tropical rain forest
2200
45.0
4.89
Tropical seasonal forest
1600
35.0
4.57
Temperate evergreen forest
1300
35.0
3.71
Temperate deciduous forest
1200
30.0
4.00
Boreal forest
800
20.0
4.00
Woodlands and shrubland
700
6.0
11.66
Savannah
900
4.0
22.50
Temperate grassland
600
1.6
37.50
Tundra and alpine
140
0.6
23.33
Desert and semidesert scrub
90
0.7
12.86
Cultivated land
650
1.0
65.00
Swamp and marsh
2000
15.0
13.33
Lake and stream
250
0.02
1250.00
Marine
Open ocean
125
0.003
4166.00
Upwelling zones
500
0.02
2500.00
Continental shelf
360
0.01
3600.00
Algal beds and reefs
2500
2.0
125.00
Estuaries
1500
1.0
150.00
Source: Columns 1-3 From Whittaker (1975). Column 4 from personal
calculation.
Note 1. M.N.P.P.=Mean Net Primary Production. Units are dry grams
per square meter per year.
Note 2. M.B.=Mean Biomass. Units are kilograms per square meter
per year.
Note 3. P/B Ratio=M.N.P.P./M.B..
147
Appendix 8.1: Computer program
ρ = 1;
δ = 5/10;
r= 5/100;
α = 1;
β = 50;
σ = 1/20;
γ = 2;
τ = 1;
f[x ] := ρ x(1 − x/100) ;
u[x ,h ] := 2β xˆ (1/2) τ hˆ(1/2) ;
c[i ] := (1/2) σ iˆ2;
w[x ,k ] := α (xˆ2) / (γk) ;
dx[x ,k ,h ,i ] =f[x] −w[x,k] −h;
dk[x ,k ,h ,i ] =i−δk;
dh[x ,k ,h ,i ] =
(D [u [x,h] , h] (r-D [f [x] , x] + D [w [x,k] , x]) − D [u [x,h] , x] − D [u [x,h] , h, x] dx [x,k,h,i]) /
D[u [x,h] , {h,2}] ;
di[x ,k ,h ,i ] = (D [c [i] , i] (r+δ) + (D [u [x,h] , h] D [w [x,k] , k])) /D[c [i] , {i,2}] ;
NSolve[{dx [x,k,h,i] == 0, dk [x,k,h,i] == 0, dh [x,k,h,i] == 0, di [x,k,h,i] == 0}, {x,k,h,i}]
%[[3]] //N
xs=x/.%18 [[3]]
ks=k/.%18 [[4]]
hs=h/.%18 [[1]]
is=i/.%18 [[2]]
jm={{D[dx [x,k,h,i] ,x], D[dx [x,k,h,i] ,k], D[dx [x,k,h,i] ,h], D[dx [x,k,h,i] ,i]},
{D[dk [x,k,h,i] ,x], D[dk [x,k,h,i] ,k], D[dk [x,k,h,i] ,h], D[dk [x,k,h,i] ,i]},
{D[dh [x,k,h,i] ,x], D[dh [x,k,h,i] ,k], D[dh [x,k,h,i] ,h], D[dh [x,k,h,i] ,i]},
{D[di [x,k,h,i] ,x], D[di [x,k,h,i] ,k], D[di [x,k,h,i] ,h], D[di [x,k,h,i] ,i]}}/.
{x→xs, k→ks, h→hs, i→is};
TableForm[jm]
evs=Eigevalues[jm]
148
evcts=Eigenvectors[jm]
e1=evcts[[2]]
e2=evcts[[4]]
lsg=Fuction[{x0, k0, h0, i0},
solc=NDSolve[{
x´[t] ==dx[x [t] , k [t] , h [t] , i [t]],
x[0] ==x0,
k´[t] ==dk[x [t] , k [t] , h [t] , i [t]],
k[0] ==k0,
h´[t] ==dh[x [t] , k [t] , h [t] , i [t]],
h[0] ==h0,
i´[t] ==di[x [t] , k [t] , h [t] , i [t]],
i[0] ==i0,}, {x, k, h, i}, {t, -7.5, 0}];
scx[t ] :=Evaluate[x [t] /.solc [[1, 1]]] ;
sck[t ] :=Evaluate[x [t] /.solc [[1, 2]]] ;
sch[t ] :=Evaluate[x [t] /.solc [[1, 3]]] ;
sci[t ] :=Evaluate[x [t] /.solc [[1, 4]]] ;
Table[{t, scx [t] , sck [t] , sch [t] , sci [t] }, {t, -7.5, 0, 1/100}]];
lsg[xs, ks, hs, is] ;
Clear[xn1, hn1] ;
xn1=Join[Table[{x, Evaluate[f [x] − w [x,ks]]}, {x,0,xs-1/100,1/100}],
Table[{x, Evaluate[f [x] − w [x,ks]]}, {x,0,xs-1/100,xs+1/100,1/100}],
Table[{x, Evaluate[f [x] − w [x,ks]]}, {x,0, xs+1/100,100,1/100}]];
hnlfunc[y ] :=Evaluate[(2x (r-D [f [x] ,x] + D [w [x,k] ,x]) − f [x] + w [x,k]) /.{x → y, k → ks}]
hnl=Table[{x, hnlfunc [x] }, {x, 1, 100, 1/100}] ;
Clear[kn1, in1] ;
kn1=Join[Table[{k,δk}, {k, 0, ks-1/100, 1/100}] ,
Table[{k,δk}, {k, ks-1/100, ks+1/100, 1/100}] ,Table[{k,δk}, {k, ks+1/100, 400, 1/100}]];
inlfunc[x ] :=Evaluate[-D [u [x,h] , h] D [w [x,k] , k] /σ (r+δ))/.{x → xs, h → hs, k → x}]
in1=Table[{k, inlfunc [k] },{k, 1/1000, 400, 1/100}] ;
Clear[sc]
sc[h1 ,h2] :=sc[h1,h2] =lsg[xs+h1 e1[[1]] +h2 e2[[1]] ,
ks+h1 e1[[2]] +h2 e2[[2]] , hs+h1 e1[[3]] +h2 e2[[3]] ,
is+h1 e1[[4]] +h2 e2[[4]] ;
149
xh[h1 ,h2 ] :=Transpose[{Transpose [h1,h2]] [[2]], Transpose[sc [h1,h2]] [[4]]}]
Show[
ListPlot[xn1, PlotRange→ {{0, 100}, {0, 40}}, PlotJoined→True,
DisplayFunction→Identity, Axes→False, Frame→True,
FrameLabel→ {´´x´´, ´´h´´}],
ListPlot[hn1, PlotRange→All, PlotJoined→True,
DisplayFunction→Identity],
ListPlot[xh[1/300, 1/300], PlotJoined→True, Plotstyle→Hue[0] ,
DisplayFuction→Identity],
ListPlot[xh[−1/100, −1/100], PlotJoined→True, Plotstyle→Hue[0.8] ,
DisplayFuction→Identity],
.
Graphics[Text[StyleForm[´´x=0´´, FontSize→10, FontWeight→´´Bold´´],
{30,22},{1,1},Frame→True, DisplayFunction→Identity],
.
Graphics[Text[StyleForm[´´h=0´´, FontSize→10, FontWeight→´´Bold´´],
{51,35},{0,1},Frame→True, DisplayFunction→Identity],
Graphics[Text[StyleForm[´´o.p.´´, FontSize→10, FontWeight→´´Bold´´],
{37, 5},{1,1},Frame→True, DisplayFunction→Identity],
DisplayFunction→ $DisplayFunction];
Ki[h1 ,h2] :=Transpose[{Transpose [sc [h1,h2]] [[3]], Transpose[sc [h1,h2]] [[5]]}];
Show[
ListPlot[kn1, PlotRange→ {{0, 400}, {0, 250}}, PlotJoined→True,
DisplayFunction→Identity, Axes→False, Frame→True,
FrameLable→ { ´´k´´, ´´I´´}],
ListPlot[in1, PlotRange → All, PlotJoined → True, DisplayFunction → Identity] ,
ListPlot[ki[1/120, 1/120], PlotJoined→True, Plotstyle→Hue[0] ,
DisplayFuction→Identity],
ListPlot[ki[−1/30, −1/30], PlotJoined→True, Plotstyle→Hue[0.8] ,
DisplayFuction→Identity],
.
Graphics[Text[StyleForm[´´k=0´´,FontSize→10, FontWeight→´´Bold´´],
{50,44},{1,1},Frame→Ture, DisplayFunction→Identity],
.
Graphics[Text[StyleForm[´´I=0´´,FontSize→10,FontWeight→´´Bold´´],
{165,220},{0,1},Frame→Ture, DisplayFunction→Identity],
Graphics[Text[StyleForm[´´o.p.´´,FontSize→10,FontWeight→´´Bold´´],
{400, 60},{1,1},Frame→True, DisplayFunction→Identity],
DisplayFunction→ $DisplayFunction];
150
Chapter 9
Case study: sustainable use and
conservation of renewable resources in
Danayiku Nature Park at Shan-Mei,
Taiwan
In this chapter we will investigate a community-based conservation program in Danayiku Nature Park at Shan-Mei, Taiwan, which was initiated
to promote sustainable use and conservation of renewable resources. As the
most successful community-based conservation project in Taiwan, Danayiku
Nature Park at Shan-Mei has attracted intensive attention of public media, and was usually highly praised as an excellent model for conservation
and rural community development. Many rural communities, indigenous or
non-indigenous, are therefore enthusiastic about imitating this successful experience. What we here particularly concerned about is the question, which
exogenous factors contributed to the success of Danayiku Nature Park. Based
on extensive field studies between 1999 and 2002 and the theoretical framework developed in previous chapters, some assessments about the feasibility
of the sustainable use strategy in Shan-Mei and in vicinal several indigenous
communities can be made. The basic theme is, that the success of Danayiku
Nature Park can be attributed to some feasible exogenous conditions that
other communities do not possess. This explains to a great extent why sustainable use strategy scored a success in Shan-Mei, while similar projects in
vicinal communities failed or did not work so well like Shan-Mei did.51
9.1 Background
The 2,200-hectare (ha) Shan-Mei, named ´Saviki´ in Tsou, is an indigenous village of the Tsou52 located in the A-Li-Shan Township of the Chia-Yi
County, a remote montane area of southwest Taiwan (approximately 120.4◦
east and 23.2◦ north). It is located in the west part of the A-Li-Shan Mountain Range, and is not too distant from the A-Li-Shan National Scenic Area,
one of the popular destinations for tourists, and the Yushan National Park
(see figure 9.1). Elevations of the village vary between 400 and 1200m. The
51
Some accidental factors, such as the personalities of community leaders, may have
influenced the outcomes of community-based conservation projects. But these are beyond
the scope of our discussion.
52
Tsou is one of the ten officially recognized indigenous tribes of Taiwan.
151
village includes about 800 ha. aboriginal reserves and 1400 ha. national
forest lands which are covered mainly by subtropical forests. Almost all of
the residents are Tsou people. According to Li and Tang (1999), there were
about 530 residents in Shan-Mei, but only half of them practically lived in
the village in 1999. Therefore, like almost all indigenous villages in Taiwan,
Shan-Mei was faced the severe problem of population drain caused by scarce
employment opportunities.
The main occupation of the village residents is farming, especially bamboo plantation supplemented by seasonal job opportunities provided by tea
plantation. Compared to the Chinese people mainly living on the plain,53
the income level and standard of living of the village residents are relatively
low. The construction of Tai-18 highway passing through A-Li-Shan Mountain Range led to extensive tea and beetle nut plantation in this area since
1980s. The prosperous tea plantation in the last two decades had once significantly improved the standard of living of village residents, because it needed
intensive labor inputs and thereby provided many seasonal job opportunities
for indigenous people. However, the collapse of prices of tea and bamboo in
recent years again put pressure upon the local economy. Nowadays, village
residents usually still have to find part time jobs to make ends meet. Getting into debt is a usual phenomenon. To great extent, all these factors are
responsible for the prevailing social problems throughout indigenous communities in this area, such as poverty, unemployment, low education level and
excessive drinking.
In spite of the prevailing economic and social problems, residents of ShanMei accidentally preserved some of the Tsou traditions, while the other Tsou
communities rapidly lost their cultural traditions in the last several decades
(Li and Tang, 1999). Compared to other Tsou communities, Shan-Mei preserved stronger tribal identity. Individuals are more obedient to the communal decisions and in principle willing to work voluntarily for the community.
Chief and elders of Shan-Mei still have great influence on the process of
decision-making about public affairs. The traditional fishing territories were
still well-defined until poaching boomed. All these have significantly influenced the outcome of the conservation project in the future.
For Shan-Mei, the construction of the Tai-18 highway brought both opportunities and challenges. On the one hand, based on its beautiful scenery
and the advantage of adjoining the A-Li-Shan National Scenic Area and the
53
The Chinese people refer in this chapter to those who originally stem from mainland
China.
152
Yushan National Park, Shan-Mei has good chance to develop tourism after
the traffic infrastructure was improved.54 On the other hand, for poachers
coming from outside world, it became more easily to have access to the natural resources of Shan-Mei that can be found primarily in Danayiku. Danayiku
is the valley of the Danayiku river, a tributary of the Zeng-Wen river, which
originates from the Central Mountain Range (see figure 9.1). The 18 kilometers long Danayiku river is famous among Tsou people for its abundant
wildlife and fresh water fish resources. For residents of Shan-Mei, it was the
most important origin of protein. However, since the end of the World War
II, the Chinese poachers first heavily reduced wildlife populations, and then
the subsistence hunters of Tsou also depleted wildlife resources in Danayiku.
Even under such situation, the fresh water fish remained abundant until
1970s. The use rights of fresh water fish resources in Danayiku belonged to
five major families of Shan-Mei, including Du, Chuang, An, Wang and Yang.
Each family owned certain part of the Danayiku river, and had the right
to manage and use fresh water fish resources. It is interesting to note that
Danayiku river was not only private property. Sometimes communal fishing festival was held by using traditional fishing methods, and the harvests
were allocated to all community individuals. Therefore, Danayiku river was,
to certain extent, also the community common. The whole property system worked well until the construction of the Tai-18 highway at the middle
of 1970s. Thereafter, Chinese poachers came into Shan-mei, used modern
technology like chemical poison and electrofishing, and thereby significantly
reduced fish populations. The local residents also learned to use modern
technology. They soon exhausted all biological resources of the Danayiku
river. At the beginning of 1980s, the Danayiku river was virtually dead, and
the traditional property right system was broken down (Cheng-Sheng Gau;
Cheng-An Chuang, pers. comm.).
54
In fact, there were private enterprises that were interested in the potential of developing tourism in Shan-Mei and wanted to purchase lands. But these proposals were refused
by villagers because of their consensus of retaining autonomous development (Cheng-Sheng
Gau, the Village Chief of Shan-Mei, pers. comm.).
153
TAIWAN
Study
area
十
•A- Li-Shan
23 ° 30′N,120° 4 0 ′E
Tai-18 Highway
•
Yushan
National
Park
Da-Ban
Zen-Wen River
•Li-Chia
•Shan-Mei
Danayiku River
•Hsing-Mei
•Cha-Shan
KM
•Shan-Ming
0
10
20
Figure 9.1 Location of the study area.
154
9.2 Project history and evolution
In 1985, primarily driven by poverty prevailing in Shan-Mei, some leaders
and young people of the village tried to find an alternative for improving the
standard of living. Considering the proven fact that agriculture had failed to
generate enough income for residents, they turned their attention to the possibility of developing tourism. A team named ´Tourism Research Group´ in
1985, which later was transformed into the ´Tourism Promotion Committee´
in 1987, was authorized to assess different alternatives. At the community
assembly held in October, 1989, after years long severe debates and discussions within community, they finally decided to initiate a community-based
conservation program which aimed at developing ecotourism based on the
sustainable use of natural resources, especially the fresh water fish resources
of Danayiku (Li and Tang, 1999; Cheung-Mei Yang, the former general secretary of SMCDS, pers. comm.).55
The conservation program included several important components. First,
five major families that owned the property rights of fishing field in Danayiku
agreed to donate their traditional rights to the Shan-Mei village. Secondly,
Danayiku was designated as a protected area managed by Shan-Mei, and any
exploitation of wildlife and fresh water fish resources is banned in Danayiku,
as long as it is not permitted by the management authority. Thirdly, anyone
who violates the ban on hunting and fishing will be fined US$286 to 1429,56
and a double fine will be imposed on community leaders who violate the ban.
Fourthly, all male residents older than 18 and younger than 50 years old were
obligated to patrol Danayiku by turns, in principle one patroller in daytime,
and two at night (Li and Tang, 1999; Cheng-Sheng Gau, pers. comm.).
In November, 1989, about 13,000 kooye minnow, V aricorhinus barbatulus
(P ellegrin), the traditionally most valued fish species by Tsou and most popular fish species found in Danayiku, were re-introduced in Danayiku. The intensive manpower input and strict enforcement of ban worked. Poaching still
lasted, but decreased significantly. The kooye minnow population recovered
dramatically. In only two years, the population increased to an estimated
total number of 1.5 million in Danayiku (Cheng-Sheng Gau, pers. comm.).
Meanwhile, villagers worked voluntarily for improving the infrastructure of
55
The reason why a community-based conservation program was chosen had primarily
to do with the personal experiences and ideas of community leaders. It is here omitted
because it is beyond the scope of our discussion. For relevant details see Li and Tang
(1999).
56
In 2002, one US dollar exchanges approximately for 35 NT dollar.
155
the village and of Danayiku. To more effectively handle the increased management issues, the Shan-Mei Community Development Society (SMCDS)
was founded in June, 1993. SMCDS are authorized to execute the conservation program in Danayiku, and also responsible for tourism and community
development. Most adults of villagers are members of SMCDS, and have the
right to participate in the process of decision-making (Cheung-Mei Yang,
pers. comm.).
In August, 1994, Danayiku was opened to develop sport fishing. Anglers were charged for angling in Danayiku. At the same time, SMCDS
succeeded in persuading the owners of about 200 ha. private lands to abandon their farms and bamboo plantation in Danayiku, and leave the lands
recovering from human interference (Cheng-Sheng Gau, pers. comm.). In
January, 1995, SMCDS declared the opening of the Danayiku Nature Park
(DNYKNP). The park includes the area that extends 3 KM on both sides of
the 18 KM long Danayiku river, and a section of 6 KM long riverside area
along the Zeng-Wen river. Thereafter, tourists were allowed to enter into
Danayiku after an entrance fee was charged. The main tourist attraction
is the spectacular kooye minnow population which can hardly be found in
wilderness of Taiwan. The revenues derived from sport fishing and entrance
fee are used to pay the management costs, mainly staff salaries, and investment costs in physical capital. If there is a surplus, those money will be
distributed to finance welfare works, including old-age pension, grants for
students and subsidies for marriage, birth and death. Part of revenues was
also spent to promote preservation and development of the culture of the
Tsou (Wen, 2000; Cheung-Mei Yang, pers. comm.).
The success of the conservation program in DNYKNP soon attracted public attentions. Numerous official awards and reports of public media made it
much more famous than in the beginning years. Both governmental donations
and tourists began to pour in. Since 1998, the number of tourists increased
dramatically, and this significantly improved the financial foundation of SMCDS (see section 9.3). Nowadays, DNYKNP is generally recognized as the
most successful community-based conservation project in Taiwan.
9.3 Resource use
Based on the historical experiences, the Tsou people of Shan-Mei believe
that sustainable use of renewable resources is reasonable by nature and should
be encouraged. Accordingly, all interviewers speak bluntly that, rather than
conservation motive, it was economic motive that drove Shan-Mei to initiate
conservation project aimed at the sustainable use of renewable resources. The
156
target species that is protected and harvested is kooye minnow, V aricorhinus
barbatulus (P ellegrin). Kooye minnow, which is called ´Luska aku´ in Tsou
meaning ´real fish´, is highly valued by the Tsou people, and was one of
the main protein origins of the Tsou. It is also one of the most important
game species of fresh water sport fishing in Taiwan. The market prices of
kooye minnow are extremely high, varying from US$22.9 to 38.6 per kilogram
(Cheng-Sheng Gau; Cheng-An Chuang, pers. comm.).
Three kinds of consumptive use of kooye minnow are allowed in DNYKNP.
First, the section of 6 KM long riverside area along the Zeng-Wen river is
designated as the sport fishing zone. The fishing season begins in November
and ends in May. Anglers have to pay a charge between US$1.43 to 2.86 per
hour, depending on number of anglers. Sport fishing is currently the most
important form of consumptive resource use. In addition, infant fishes were
sometimes caught for sale to aquatic breeder. Usually once each year, adult
fishes will be caught, and then distributed to elders of the village. But these
two forms of use play a so minor role that they almost can be neglected.
Fish viewing has been the primary non-consumptive use form of kooye
minnow since the opening of DNYKNP. In fact, it is by far more important
than any of the consumptive uses because it generates currently more than
90% of the total revenues of DNYKNP. In DNYKNP, about 20 ha. land
are designated as the fish viewing zone and opened to tourists. Tourists
currently have to pay an entrance fee of US$2.86 on weekend and holidays
and of US$2.29 on weekdays (for child US$1.71 and 1.14, respectively). As
a result of the dramatic increase in number of tourists in recent years that
usually exceeded the carrying capacity of DNYKNP, an upper limit of daily
tourist number might be introduced in the future.
9.4 Performance of the Danayiku Nature Park
Since 1989, villagers of Shan-Mei have actively protected natural resources
in Danayiku, developed use strategies aimed at capturing both consumptive
and non-consumptive values of resources, and devoted manpower and physical investment to improving management capacity. The case is therefore
suitable for being analyzed under the theoretical framework developed in
previous chapters. In the following discussion, we attempt to briefly describe
the performance of DNYKNP in terms of fish population level, harvest quantity, physical capital stock, investment in physical capital, labor input, tourist
number, and revenues of DNYKNP.
Figure 9.2 demonstrates the trend of kooye minnow population levels in
DNYKNP in the last years. The original statistics about fish population are
157
transformed by author into estimates of population biomass (kg), based on
the premise suggested by Cheng-Sheng Gau (pers. comm.) that the mean
weight of fish (including adult and infant fish) is 40 gram. After 13,000
fishes were re-introduced in 1989, the total population biomass reached a
level of 60,000 kg in two years, and remained stable until 1996. In August,
1996, the typhoon Herb destroyed the whole A-Li-Shan area with a rainfall
of more than 1,000 mm. at one night. DNYKNP lost about two thirds of
its fish population as a result of the flood. Nonetheless, the fish population
recovered rapidly thereafter. In recent two years, it increased to the level of
slightly more than 60,000 kg, because fish population spilled from the core
conservation area into the upstream area of DNYKNP (Cheng-Sheng Gau,
pers. comm.). Several interviewers asserted that the current fish population
has approached the level of natural state like several decades ago, but not yet
totally recovered (Cheung-Mei Yang; Yue-Mei Chuang; Cheng-Sheng Gau;
Cheng-An Chuang, pers. comm.). The reason why fish population does
not totally recover can be attributed partly to the fact that fish population
spilled from the core conservation area into the unprotected section of the
Zeng-Wen river, and partly to the fact that poaching, controlled by insider
of village, still lasted (anonymous interviewer; pers. comm.). In general,
according to the population trend of fish, we can assert that DNYKNP has
succeeded in protecting its target species. Moreover, previous statistics reveal
that the intrinsic growth rate of kooye minnow is extremely high, although
currently there is no relevant scientific assessment that can precisely measure
the intrinsic growth rate of kooye minnow.
158
70000
60000
Fish Population Biomass (kg)
50000
40000
30000
20000
10000
0
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
Figure 9.2 Fish population biomass of DNYKNP. Source: Cheng-Sheng
Gau (pers. comm.) and personal calculation.
Figure 9.3 shows the harvest quantity of kooye minnow. The original
statistics about revenues of sport fishing are transformed by author into estimates of harvest quantity, based on the assumption provided by SMCDS
that fishing fee averages US$1.43 per hour and each angler has an average
harvest of 600 gram each day (8 hours). Before 1994, all consumptive uses
of kooye minnow were banned. After sport fishing and other forms of consumptive uses were allowed in 1994, the harvest quantity increased steadily.
The trend was temporarily interrupted by the typhoon Herb. As a result of
the serious loss caused by the typhoon, all consumptive uses were forced to
be stopped in 1997. However, as fish population recovered, harvest increased
again in recent years. In sum, the conclusion drawn from figure 9.2 and 9.3
is apparently consistent with that drawn from our theoretical models, that
harvest and resource stock correlate positively to each other.
Based on the fact that the kooye minnow population steadily remains
at a high, relatively stable level, and that the annually harvested quantity
constitutes only less than 2% of the total population biomass so that fish
population can easily recover from the effects of harvest, it can be asserted
that the current use form of kooye minnow is sustainable.
159
900
800
Harvest Quantity of Fish (kg)
700
600
500
400
300
200
100
0
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
Figure 9.3 Harvest quantity of fish of DNYKNP. Source: SMCDS
(1994-1999; 2001), Cheng-Sheng Gau (pers. comm.) and personal
calculation.
Figure 9.4 describes the development trend of the physical capital stock of
DNYKNP. These statistics are calculated by the author, based on the annual
investment in physical capital provided by SMCDS and on the assumption
that the depreciation rate of physical capital stock equals 0.1 in ordinary
years, and equals 0.5 in the year when the typhoon caused great capital loss.
In general, the physical capital stock increased steadily since 1993, except
that there was a significant decrease in 1996 caused by the typhoon Herb.
Figure 9.5 shows the trend of the annual investment in physical capital.
The investment reached its top in 1994 when DNYKNP was prepared for
opening for the public. Investment in 1997 increased significantly in response
to the great loss of capital stock occurred in 1996. The significant increase
in investment in 1999 reflects the fact that management capacity must be
improved in response to the dramatic increase in the number of tourists since
1999 (see figure 9.7). Thereafter, investment decreased gradually again. In
sum, the long-term trend of investment seems to be declining while the total
capital stock increased. Hence, the development trends of physical capital
stock and investment seems to be consistent with those drawn from our
theoretical models, that investment and management capital stock correlate
negatively to each other. But a definite answer to this question still needs
long-term observation in the future.
160
25
Physical Capital Stock (million NT$)
20
15
10
5
0
1993
1994
1995
1996
1997
1998
1999
2000
2001
Figure 9.4 Physical capital stock of DNYKNP. Source: SMCDS (1994-1999;
2001), Cheng-Sheng Gau (pers. comm.) and personal calculation.
8
Investment in Physical Capital (million NT$)
7
6
5
4
3
2
1
0
1993
1994
1995
1996
1997
1998
1999
2000
2001
Figure 9.5 Investment in physical capital of DNYKNP. Source: SMCDS
(1994-1999; 2001) and Cheng-Sheng Gau (pers. comm.).
As discussed in section 7.2 about management capital, it is extremely
difficult to measure the human and institutional components of management
161
capital stock. What we can do is simply calculating the manpower input as
an indicator for the investment in human and institutional capital. Figure
9.6 demonstrates an increasing trend of the manpower input of the conservation project that formally paid by SMCDS. However, these statistics do not
reflect the fact that many voluntary manpower were devoted to the conservation project in the beginning stage before 1994, because even rough statistics
about voluntary manpower input are not available.57 In fact, since the opening of DNYKNP in 1994, all manpower inputs in DNYKNP were paid by
SMCDS, and thereby voluntary works were virtually almost totally transformed into formal works. It follows that it is difficult to judge the long-term
trend of the investment in human and institutional capital without credible
data resources.
45000
40000
Formal Manpower Input ( hour/person)
35000
30000
25000
20000
15000
10000
5000
0
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
Figure 9.6 Formal manpower input of DNYKNP. Source: Cheung-Mei Yang
(pers. comm.) and personal calculation.
The reputation of DNYKNP as a successful conservation project and the
famous, spectacular fish population attracted numerous tourists in recent
57
For example, villagers remember that there were often more than ten persons that
voluntarily patrolled DNYKNP at one night in the beginning stage. Meanwhile, intensive
voluntary manpower was devoted to improving infrastructure, and villagers spent much
time in meeting and other organizational issues. Currently, no more voluntary workers are
devoted to patrolling, and those voluntary manpower devoted to infrastructure improvement and organizational issues have significantly decreased.
162
years. Figure 9.7 shows that the tourist number dramatically increased since
1999. This also brought DNYKNP significant increase in revenues since 1999,
as shown by figure 9.8.58
250000
Tourist Number
200000
150000
100000
50000
0
1994
1995
1996
1997
1998
1999
2000
2001
Figure 9.7 Tourist number of DNYKNP. Source: SMCDS (1994-1999; 2001)
and Cheng-Sheng Gau (pers. comm.).
58
The statistics demonstrated by figure 9.7, figure 9.8 and table 9.1 are calculated on
the basis of the fiscal year. For example, the fiscal year 1999 began in July, 1998 and
ended in June, 1999. However, in response to the modified accounting system, the fiscal
year 2000 included 18 months, namely began in July, 1999 and ended in December, 2000.
Since 2000, the fiscal year begins in January and ends in December.
163
700000
600000
Revenues of DNYKNP (US$)
500000
400000
300000
200000
100000
0
1994
1995
1996
1997
1998
1999
2000
2001
Figure 9.8 Revenues of DNYKNP. Source: SMCDS (1994-1999; 2001),
Cheng-Sheng Gau (pers. comm.) and personal calculation.
Table 9.1 analyses the financial performance of SMCDS in fiscal year
2000. The main sources of revenues were entrance and parking fee of tourists,
which made up of 90.86% of the total revenues of DNYKNP. In fact, part of
the item ´other revenues´ were contributed by fish viewing tourism. Hence,
non-consumptive value undoubtedly played a much more important role than
consumptive value which contributed only 2.67% to the total revenues of
DNYKNP. Governmental donations used primarily for physical capital investment had played an important role in the beginning stage. Nonetheless,
as a result of the dramatic increase in tourism revenues, the proportion of
self-financing to donor funding has significantly increased in recent years,
so that donor funding made up of only 13.92% of the total revenues of SMCDS in fiscal year 2000. These revenues were primarily (about 71% of total
expenditures) spent on staff salaries, administration costs and physical capital investment, but also lots of money (about 29% of total expenditures)
were spent for supporting social welfare and cultural and communal development. In sum, there was a slight deficit of US$25,229 in fiscal year 2000. If
the expenditures on social welfare and cultural and communal development
that are not necessary for maintaining conservation project, and the governmental donations were not taken into account, the conservation project
itself would virtually generate a net surplus of US$59,258. According to the
rough estimates of the fiscal year 2001, with a total revenue of more than
164
US$572,000 from DNYKNP and a decreased physical capital investment of
US$80,000, the financial surplus of the conservation project itself will undoubtedly approach to the level of about US$286,000. Given the fact that
physical capital investment is decreasing after infrastructure has been greatly
improved, and staff salaries remains relative stable, it follows that, as long
as the tourist number remains stable at the level of the year 2000, or even
slightly less than that of the year 2000, the financial self-sustainability of the
conservation project will be guaranteed. In fact, to maintain financial selfsustainability, SMCDS even doesn´t need to attract so many tourists like the
year 2000, because the expenditures on salaries, administration and capital
investment will also, though not proportionally, decline when tourist number
decreases. In any case, the financial self-sustainability of SMCDS can almost
totally be attributed to the high non-consumptive value of kooye minnow
population.
Since 1989, the intensive patrolling has effectively brought poaching under
control. Outside poachers were scarcely detected (Cheung-Mei Yang, pers.
comm.), but inside poaching, as anonymous interviewer reported, still exists.
The quantity of the illegally harvested fish biomass is unknown.
Table 9.1 Financial performance of SMCDS in fiscal year 2000
Net revenues/expenditure
in US$
REVENUES OF DNYKNP
441,637
Entrance fee
375,907
Sport fishing fee
11,782
Parking fee
25,348
Other revenues
28,600
GOVERNMENTAL DONATIONS
71,429
EXPENDITURES
−389,724
Staff salaries
−127,034
Administration costs
−106,773
Social welfare
−55,205
Traditional dance team
−54,410
Other subsidies
−46,302
PHYSICAL CAPITAL INVESTMENT
−148,571
SURPLUS/DEFICIT
−25,229
Source: SMCDS (2001) and personal calculation.
165
9.5 Ecological, economic and social benefits
9.5.1 Ecological benefits
Since 1989, the whole ecosystem of Danayiku has benefited from the
strict protection measures against poaching and other illegal activities. First,
about 200 ha. private farms and bamboo plantation lands in Danayiku
were returned to nature. The Danayiku river also returned to life. Although the protected target species is kooye minnow, no special measures are
taken to discriminate other species. All endemic fresh water fish species of
Danayiku have virtually shared the benefits of protection measures, especially
Acrossocheilus paradoxus Gunther and Zacco barbata. The populations of
these two species have significantly increased in recent years. Similarly, the
national forest covering Danayiku valley are more effectively protected. Illegal logging that usually occurred in the last several decades ceased at last.
The once disappeared wildlife gradually returns to Danayiku valley from the
nearby Yushan National Park. Currently, presence of many big mammalian
species are often reported, including Formosan white-faced flying squirrel,
Formosan giant flying squirrel, Formosan macaque, Formosan black bear,
Formosan wild boar, Formosan Reeve´s muntjac, Formosan sambar, Formosan serow and Chinese leopard cat (Cheng-Sheng Gau, pers. comm.).
9.5.2 Economic benefits
In the following we will briefly describe the economic benefits of DNYKNP
in terms of income creation, full time and part time job creation, and dividends to the community. The success of DNYKNP brings Shan-Mei huge
amount of income. For example, together with the governmental donations, the whole conservation program created a cash inflow of approximately
US$524,286 for SMCDS in the fiscal year 2000. About half of the US$524,286
were directly transformed into the income of villagers through the ways of
staff salaries, social welfare expenditures, pay for the traditional dance team
and other subsidies. Part of the other half of the cash inflow which were
expended for administration costs and capital investment also raised the income level when villagers contracted, or are employed to participate in these
works. The revenues of SMCDS directly created some full time and part
time job, currently including 12 full time employees, 4 full time rangers, 4
part time rangers, 30 part time dancers and some temporary workers. It is
worth noting that, except some jobs at manager level, most of the employees
and rangers of SMCDS have terms of office, usually two years. Villagers take
turns at working for SMCDS. This is an equitable solution for those who
don´t have any job.
166
For most of the villagers, what more important are the added values
brought by tourism. As a result of the remote location of Shan-Mei, most
tourists have to find accommodation or at least have a meal during stay in
Shan-Mei. They usually also buy souvenir and need certain kinds of services.
According to the experiences of villagers, an average added value of at least
one dollar will be brought by per one dollar revenue of DNYKNP. It implies
that, with the total revenues of more than US$571,429 in 2001, DNYKNP
generated an added value of at least US$571,429 for the whole village in one
year. Though precise figures are not available, the total financial benefit
of DNYKNP is undoubtedly huge for the small village with less than 600
population, especially from the perspective of indigenous community with a
relative low standard of living.
To date, two hotels in Shan-Mei operate at their full capacity with total 130 beds.59 They directly create about 10 full time jobs and some part
time jobs, depending on business, and indirectly create some job opportunities for farmers who supply agricultural products to them. More than 50
villagers work in snack bars, restaurants and souvenir shops. Most of them
work part time, namely only on weekends and holidays when tourists pour
in. But even these part time jobs earn them a higher income than any jobs
they can find elsewhere, according to personal survey of the author. 5 tour
guides operate interpretation tour, charging a fee of US$22.9-28.6 per tour.
Some skilled hunters recently began to operate sporting hunting for tourists
in the area around Danayiku valley, aiming at the hunting of Formosan wild
boar, Formosan white-faced flying squirrel and Formosan giant flying squirrel. In general, except a few teachers and government officers, almost all
villagers directly or indirectly depend on DNYKNP for their living (ChiaPing Chuang; Cheng-Sheng Gau, pers. comm.). The most significant sign
of the huge economic benefits is that population of Shan-Mei, especially the
younger generation, begins to return to and resettle in the village. This is
a phenomenon that hardly can be found in another indigenous communities
throughout Taiwan.60
According to the resolution of SMCDS, some revenues of SMCDS will
be made available for communities, used primarily for social welfare and
communal development. In this way, the revenues arising from conservation
59
As a result of the high occupancy rate of these two hotels on holidays, four hotels outside Shan-Mei with total 800 beds currently depend primarily on tourists visiting Danayiku
for doing their business.
60
Even many Chinese people wanted to do business in Shan-Mei, but were refused by
SMCDS (Cheung-Mei Yang; Cheng-Sheng Gau, pers. comm.).
167
project can therefore be more equally distributed amongst villagers. This,
together with other economic benefits, have contributed to some of the social
benefits that have been obtained because of the conservation project.
9.5.3 Social benefits
The social components of the conservation project can be judged according to the following criteria, including health, education, institution, infrastructure, excessive drinking, family problems, social security, and culture
preservation and development. The higher income level may have positive
influence on health conditions and formal education level of villagers, but
the effect is presumably negligible in the short run. Currently, no credible
data sources are available. One interesting phenomenon is that some young
people organized a small group to educate themselves through reading and
discussion. From the perspective of institution, the conservation project provided chances for villagers to participate in decision-making process of public
affairs. The community is developing management, marketing and business
skills, and thereby improving its management capacity. Through steady investment, the infrastructure of the community are significantly improved.
Unemployment dramatically declined, and this led to the decrease in excessive drinking and in other family problems (Cheung-Mei Yang, pers. comm.).
The financial support of SMCDS enhanced an initial social security system.
It also promoted preservation and development of the Tsou culture, and reconstructed the self-confidence of the Tsou people of Shan-Mei in their own
cultural tradition (Wen, 2000).
9.6 Negative impacts
Except positive influences, the conservation program also has negative
impacts on ecological and human system. From the ecological point of view,
excessive tourism, especially on holidays, constitutes the most serious threat
that gradually degrades the original ecosystem of DNYKNP. All interviewers
agree that the current tourist number on holidays has far exceeded the carrying capacity of DNYKNP, and it keeps growing. This will reduce the tourism
potential, and injure the reputation of DNYKNP as a conservation model.
Furthermore, in pursuit of economic benefits brought by tourists, some land
owners in Danayiku began to do their business outside the zone where is
designated as legal business area. This will further cause more damage to
ecosystem and tourism potential. However, no adequate measures are taken
to stop the trend. Fortunately, all these negative impacts are restricted to the
area of about 20 ha. that is opened for tourism. Therefore, rather than ecological health, what people really concerned about is the damage to tourism
168
potential, because, once tourism potential was significantly reduced, the financial foundation of SMCDS, and in turn the whole conservation project
would be threatened with collapse.
Another issue with reference to ecological health is the problem of fish
feeding. To attract tourists, fish feeding was allowed at three specific sites
since the opening of DNYKNP. Some ecologists were seriously concerned
about this problem. Recently, considering the possible negative effects, all
fish feeding was ceased.
With reference to the negative economic and social impacts, what people
most concerned about is the problem of wealth distribution. The introduction
of tourism and market mechanism led to a greater gap of wealth between rich
and poor families than a decade ago. Those who are quickly accustomed to
free market and learned modern business skill got a greater proportion of
the economic benefits created by the conservation program. Furthermore,
people began to complain that, in pursuit of economic benefits, villagers
gradually became selfish and utilitarian. Today, most villagers is unwilling
to work voluntarily for the community without any pay (Chia-Ping Chuang,
pers. comm.). In the long run, whether these phenomena will influence the
identity of villagers, and in turn influence the performance of the conservation
program, remains to be seen.
9.7 Comparison of different community-based conservation
projects in the A-Li-Shan area: an assessment procedure
9.7.1 The assessment procedure
The success of DNYKNP at Shan-Mei has inspired several indigenous
communities in the A-Li-Shan area, including Hsing-Mei, Li-Chia, Da-Ban
and Cha-Shan (see figure 9.1), to initiate similar conservation projects in
1990s aimed at the sustainable use of fresh water fish resources. In addition,
the Shan-Ming Township located in the Nan-Zi-Shen river basin (see figure
9.1) independently initiated its fresh water fish conservation program in 1990.
To date, only DNYKNP at Shan-Mei can be ranked as a successful conservation program. Hsing-Mei and Li-Chia are advancing in protecting their fresh
water fish resources, but still have to struggle for the maintenance of their
conservation programs whose fate remains to be seen. The conservation programs in Da-Ban, Cha-Shan and Shan-Ming have apparently failed. Based
on these facts, some questions emerge: Which factors affect the feasibility of
the sustainable use strategy as a conservation instrument? Is there a general framework that can explain, partly at least, why some community-based
conservation programs succeeded while some other failed?
169
Prescott-Allen and Prescott-Allen (1996) suggested an systematic assessment procedure that evaluates the sustainability of uses of wild species. However, this procedure can be used only after the use program of wild species
has been brought into practice. It is therefore an ex post assessment procedure. Instead of an ex post assessment procedure, an ex ante assessment
framework will be developed in this section to assess, or even forecast the
feasibility of a sustainable use strategy before it is brought into practice. In
the following discussion we will demonstrate how this systematic procedure
works.
As discussed in chapter 8, we have identified eight factors that influence
the equilibrium resource stock of a sustainable use program, including the
non-consumptive value coefficient (β), the efficiency coefficient of management capital (γ), the intrinsic growth rate of utilized species (ρ), the gross
profit coefficient of harvest (τ ), the discount rate (r), the depreciation rate
of management capital (δ), the poaching coefficient (α) and the cost coefficient of investment (σ). We have also demonstrated that, other things
being equal, the equilibrium resource stock will increase, when β, γ, ρ and
τ are increased, and when r, δ, α and σ are decreased. In general, the more
the equilibrium resource stock closes to the carrying capacity, the better it
would be for ecosystem health, and, from the ecological point of view, the
use program would be more likely to be sustainable. Accordingly, we can
use the equilibrium resource stock, and in turn the eight factors as indicators
for judging the feasibility of the sustainable use strategy as a conservation
instrument.
In reality, it is impossible for almost all practical cases to precisely estimate the size of the eight parameters without extensive scientific research.
We also don´t know the precise functional forms of relevant functions and
the precise comparative static effects of the eight parameters on equilibrium
resource stock. Since information is never complete, therefore, instead of
executing a computer simulation, what we practically can do may be giving
them a rough ranking based on the findings of field study. An equal weight
is then given to the eight parameters by assuming that they have equal influence on equilibrium resource stock.61 The formal procedure is as follows.
The eight parameters are first classified into five different ranking according
to their size, including ´very high´, ´high´, ´middle´, ´low´ and ´very low´.
For the parameters β, γ, ρ and τ , we define ´very high´ as ´highly positive´
61
In fact, some parameters may play a more important role than others do. For example,
according to historical experiences, the intrinsic growth rate of species may have greater
influence on equilibrium resource sock than some of the other factors.
170
in the sense that a very high β, γ, ρ or τ has a highly positive influence on the
equilibrium resource stock, and thereby can highly contribute to sustainable
use of resources. Similarly, ´very low´ is defined as ´highly negative´ in
the sense that a very low β, γ, ρ or τ has a highly negative influence on the
equilibrium resource stock, and thereby can seriously threaten the sustainable use of resources. The ranking ´high´, ´middle´ and ´low´ are defined
as ´positive´, ´neutral´ and ´negative´, respectively. A numerical ranking
1, 2, 3, 4 and 5 is then given to the evaluations ´highly positive´, ´positive´,
´neutral´, ´negative´ and ´highly negative´, respectively. The smaller the
ranking of a parameter is, the more positive influence it would have for equilibrium resource stock and sustainable use of resources. Contrarily, for the
parameters r, δ, α and σ, we define ´very high´ as ´highly negative´ in the
sense that a very high r, δ, α or σ has a highly negative influence on the
equilibrium resource stock, and thereby can seriously threaten the sustainable use of resources. The ranking ´high´, ´middle´, ´low´ and ´very low´
are hence defined as ´negative´, ´neutral´, ´positive´ and ´highly positive´,
respectively. The numerical ranking from 1 to 5 can be then given to the
evaluations from ´highly positive´ to ´highly negative´.
After all ranking about the eight parameters of a specific case are given,
an arithmetical average of these ranking can be obtained. The final score
represents the overall status of a sustainable use project. The smaller the
score is, the greater the equilibrium resource stock would be, and the more
feasible a sustainable use strategy would be as a conservation instrument. In
the following we will apply this assessment procedure for assessing the feasibility of different sustainable use programs in the A-Li-Shan area, Taiwan,
as a conservation strategy.
9.7.2 A comparison of different community-based conservation
projects in the A-Li-Shan area
Table 9.2 demonstrates the results of the assessment procedure with regard to the six community-based conservation programs, including ShanMei, Hsing-Mei, Li-Chia, Da-Ban, Cha-Shan and Shan-Ming. The ranking
are based on the findings of the field study executed during 1999 and 2002.
Now let´s explain the reasons underlying these ranking.
The target species of all the six conservation programs are fresh water
fishes, especially kooye minnow. As the experience of Shan-Mei showed,
kooye minnow has a very high intrinsic growth rate (ρ). Therefore, all the
ranking of the six conservation programs are 1 (highly positive) in the column
´ρ´. Next, as a result of the extremely high market prices of kooye minnow,
171
the gross profit coefficient of harvest (τ ) of all cases are extremely high.
According to our survey, there is no significant difference in the expensive
sport fishing fee charged by these villages. It follows that all the ranking of
the six conservation programs are also 1 (highly positive) in the column ´τ ´.
In addition, all six conservation programs are threatened with highly active
poaching. Most of poachers are insiders of villages and, without continuous
protection actions against poaching, the fresh water fish resources would be
rapidly depleted to the level of local extinction (Cheng-Sheng Gau, pers.
comm.). A ranking of 5 (highly negative) is hence given to all six villages in
the column ´α´. Finally, the six villages have similar economic conditions.
Compared to ordinary Chinese communities in Taiwan, their income level
and standard of living are relatively low. Getting into debt is a prevailing
phenomenon among them. The villagers have therefore relatively strong
incentive to cash in natural resources. But most of them are not really so
poor that they would deplete natural resources irrespective of the possible
negative impacts. Based on these observations, we believe that a ranking of
4 (negative) for the discount rate (r) will be reasonable. With reference to
the depreciation rate of management capital (δ), no information or facts can
be used to assess its status. Hence, it is here excluded from the assessment
procedure.
Table 9.2 Results of the assessment procedure
ρ τ α r δ β γ σ Average ranking
Shan-Mei
1 1 5 4 * 1 1 2
2.14
Hsing-Mei 1 1 5 4 * 5 4 2
3.14
Li-Chia
1 1 5 4 * 5 4 2
3.14
Da-Ban
1 1 5 4 * 5 4 3
3.29
Cha-Shan 1 1 5 4 * 5 4 4
3.43
Shan-Ming 1 1 5 4 * 5 5 4
3.57
1: highly positive. 2: positive. 3: neutral. 4: negative
5: highly negative.
*unknown
There are significant difference between the non-consumptive value coefficient (β) of different villages. Among these villages, Danayiku of Shan-Mei
originally owns the most abundant fish population (Cheng-Sheng Gau, pers.
comm.). The beautiful scenery of Danayiku makes it an attractive tourist
destination and significantly promotes the non-consumptive value of kooye
minnow. Shan-Mei is therefore given a ranking of 1 (highly positive) in this
item. Hsing-Mei, Da-Ban, Cha-Shan and Shan-Ming don´t have the spectacular fish population and outstanding scenery like Shan-Mei does. These lead
172
to the very low non-consumptive value of kooye minnow. Li-Chia has nearly
similar natural conditions like Shan-Mei, but its remote location offsets its
natural advantages.62 The five villages are therefore given a ranking of 5
(highly negative) in this item.
Natural topography and human road system lead to significant difference between the efficiency coefficient of management capital (γ) of different
conservation programs. In the case of Shan-Mei, there is only one road to
the entrance of DNYKNP and nowhere in DNYKNP is accessible by car
to poachers. Once the only entrance is controlled, the whole DNYKNP is
in principle safeguarded. In other words, the efficiency coefficient of management capital of Shan-Mei is very high and should be ranked as 1 (highly
positive). In Hsing-Mei, Li-Chia, Da-Ban and Cha-Shan, the river is at many
sites easy accessible by car or by walking as a result of the relative complex
road system. This leads to a relative low efficiency coefficient of management
capital. The four villages can be ranked as 4 (negative) in this item. The
worst situation occurs in Shan-Ming. A highway was constructed along the
river, and virtually everywhere is easy accessible by car. This is why it is
ranked as 5 (highly negative) in the column ´γ´.
About the cost coefficient of investment (σ), there is no significant difference in the investment cost of physical and human capital between different
villages. The major difference exists in the investment cost of institutional
capital as a result of the different combinations of tribes. For example, almost
all the villagers of Shan-Mei, Hsing-Mei and Li-Chia are the Tsou people. The
homogeneity of villagers greatly reduces the communication costs of building
management institution. The low investment cost in institution capital leads
to the ranking of 2 (positive). Contrarily, Cha-Shan and Shan-Ming primarily consist of Tsou, Bunun63 and Chinese people (Li and Tang, 1999). The
complex population combinations make it very difficult to reach consensus
when discussing public affairs (Li and Tang, 1999), and significantly raise
the communication costs of building management institution. We therefore
rank them as 4 (negative) in this item. The village Da-Ban primarily consists of the Tsou people that stem from two different communities that are
historically often hostile to each other (Li and Tang, 1999). Its communication costs are probably higher than Shan-Mei, but lower than Cha-Shan and
Shan-Ming. It is classified into the rank 3 (neutral).
62
Shan-Mei is about one car hour distant from Chia-Yi, the most important city in this
area. Compared to Shan-Mei, tourists have to drive another two hours to reach Li-Chia.
Almost no one will do that to see what they can see in Shan-Mei.
63
Bunun is also one of the ten officially recognized indigenous tribes in Taiwan.
173
Based on previous discussions, an average ranking of each village can
be calculated. In sum, Shan-Mei owns many positive, but only two negative exogenous conditions. This helps it obtain the lowest average ranking
of 2.14 (approximately positive) among all villages, which implies that the
biological, economic and social conditions are generally adequate to the application of the sustainable use strategy. Compared to Shan-Mei, Hsing-Mei
and Li-Chia have lower non-consumptive value of resources and lower efficiency of management capital. They both get an average ranking of 3.14
(approximately neutral), which is not particularly in favor of or against a
sustainable use strategy. Compared to Hsing-Mei and Li-Chia, Da-Ban and
Cha-Shan have a higher cost coefficient of investment. They obtain an average ranking of 3.29 and 3.43, respectively, that are probably somewhat
disadvantageous to sustainable resource use. Finally, under the existing conditions of low efficiency of management capital and high cost coefficient of
investment, Shan-Ming obtain the highest average ranking of 3.57.
The results of the assessment procedure imply that, without intensive
outside (governmental or non-governmental) support that can offset some
disadvantageous exogenous conditions, the use strategy would probably fail
in the sense that it cannot maintain a relatively high resource stock level,
and would not be feasible as a conservation instrument in the cases of DaBan, Cha-Shan and Shan-Ming. For Hsing-Mei and Li-Chia, the exogenous
conditions are not especially good or bad. Only Shan-Mei owns the biological,
economic and social conditions that are generally adequate to the application
of the sustainable use strategy. These conclusions are surprisingly consistent
with what really happened in the past years. Only Shan-Mei has succeeded in
maintaining a high resource stock level that approaches the carrying capacity.
Hsing-Mei and Li-Chia are advancing in protecting their fish resources, but
the results are not especially good. Their fish population levels are not high
enough so that sport fishing is allowed only in one week each year. As a
result of the scarcity of revenues caused by the relatively low population
level, they still have to struggle for the maintenance of their conservation
programs whose fate remains to be seen. The conservation programs of DaBan, Cha-Shan and Shan-Ming did not last for a long time and have soon
failed.
Some critical conclusions can be drawn from the findings of the case study.
First, sustainable use of wild species is not omnipotent recipe for resolving conservation problems (and development problems). As our case study
shows, even under the same conditions of intrinsic growth rate of species,
consumptive value of species, discount rate and poaching pressure, use strat174
egy worked well in few cases. It doesn´t work in other cases , because a
variety of exogenous factors also can influence the outcomes of conservation
programs. It follows that it is highly dangerous to judge the feasibility of a
sustainable use program by using only one or only few indicators, including
the intrinsic growth rate of species, the discount rate and the price/cost ratio
of harvest, as some people were used to do in the past.64 To evaluate the
feasibility of the sustainable use of wild species as a conservation strategy, it
always needs an overall assessment about relevant biological, economic and
social conditions at the site concerned.
Secondly, in the previous cases at least, the results of our assessment
procedure are in principle consistent with what happened in the reality. We
therefore argue that the assessment procedure can be applied for sustainable
use programs before they are brought into practice. For those cases with
low or very low average ranking, sustainable use strategy can be encouraged.
For those cases with high average ranking, more caution should be taken,
or they simply should be stopped. This may help reduce the probability
of resource overexploitation and environmental degradation caused by the
failure of sustainable use projects.
Thirdly, at the level of economic theories, both the Clark model and
the Swanson model demonstrated the important roles played by the three
factors when discussing resource use problems and relevant conservation issues, including the intrinsic growth rate of species, the discount rate and
the price/cost ratio of harvest (the gross profit coefficient of harvest in our
terminology). However, our model and the case study show that, rather
than the previous three often discussed factors, they are non-consumptive
value of resources, efficiency of management capital and investment cost in
management capital that explain the difference of the performance of various use programs in this case. Hence, these newly introduced factors also
play an important role, because they sometimes can determine the fate of the
sustainable use project. We conclude that it is worth while extending bioeconomic models to investigate problems of management capital accumulation,
non-consumptive value of wild species and illegal harvest. Such modifications help us have more insight into the complex resource use problems and
relevant conservation issues.
Finally, our case study shows that, under prevailing natural, economic
and social conditions in Taiwan, the probability of success of a sustainable
64
For example, Caughley argued that the consumptive use of African elephant is not a
feasible conservation strategy because of the low intrinsic growth rate (Caughley, 1993).
175
use program initiated by indigenous communities is relatively low, if villages
in the A-Li-Shan area can represent the ordinary indigenous communities
in Taiwan. According to author´s personal observation, there are only few
indigenous communities that have general conditions that can obtain the
ranking like Shan-Mei has. At the policy level, this implies that some popular thoughts about resource use, conservation and indigenous people should
be reconsidered. In recent years, more and more people argue that, from both
perspectives of human rights and conservation, the property rights of natural
resources in some protected areas should be returned to indigenous communities, and decentralized, community-based conservation programs would work
better than traditional protected areas managed by the central government.
From the perspective of human rights, whether the property rights of natural
resources should be returned to indigenous communities is a question of value
judge and beyond the scope of our discussion. But from the perspective of
conservation, we assert that the performance of community-based conservation programs is in general not so good as it is supposed to be. Compared to
the National Park system and other nature reserves managed by the central
government, the success probability of community-based conservation programs is relatively low so that they can protect only small fragments of the
whole ecosystem. Based on the case study in chapter 4 about the National
Park system of Taiwan, it is clear that, under current circumstances, only the
central government has the capacity to support and manage a large, systematic protected area network in Taiwan. It is unreasonable to suppose that
indigenous communities can deal with the problems like poaching, institution
building and financial deficit more effectively than the central government.
9.8 Some challenges to DNYKNP at Shan-Mei
Some problems still threaten the long-term sustainability of DNYKNP.
First, the majority of DNYKNP is national forest land owned de jure by
the central government, although villagers of Shan-Mei have de facto the use
rights of some natural resources, such as fresh water fish, since a long time.
The ambiguity of property rights of natural resources has troubled ShanMei, since all charges in DNYKNP is virtually illegal, and, once the friendly
attitude of the government toward DNYKNP changed some day, the project
would probably fail if charges were not allowed. Secondly, the current mass
tourism would significantly reduce the tourism potential of DNYKNP, if no
adequate measures are taken to control the number of tourists. Finally, as
a result of the huge economic benefits brought by tourism, SMCDS seemed
to have gradually lost the control of individual business behavior of villagers
in DNYKNP. Most villagers began to worry about that excessive business
176
activities will degrade the natural environment of DNYKNP, and in turn
damage its tourism potential and its good reputation as a conservation model.
Whether Shan-Mei can overcome these challenges in the future remains to
be seen.
177
Chapter 10
Conclusions, policy implications and
limits in applicability of the theoretic
model
In recent years, the sustainable use of wild species in protected areas or in
buffer zones of protected areas is usually promoted as an alternative conservation strategy. This dissertation has focused on the relevant resource harvest
and management issues. What this research especially concerned about is the
question, whether and under which biological and socio-economic conditions
the sustainable use of wild species is an adequate strategy for biodiversity
conservation. To provide a general theoretical framework for answering this
question, several models were developed which investigate the dynamic interaction between the harvest of wild species, management of protected areas,
population levels of the utilized species and illegal harvest. The case study
involving the community-based conservation projects in the A-Li-Shan area
of Taiwan was addressed to test the applicability of the theoretic models.
This final chapter offers study conclusions, policy implications and some
comments involving the limits in applicability of the theoretic model and
recommendations for future research.
10.1 Study conclusions
10.1.1 Conclusions of the theoretic models
At the theoretic level, our models enriches the analytical framework of
the traditional harvest model of renewable resources by developing a bioeconomic model with two state and two control variables. Compared to the
Clark model which considers only the harvest problem, we additionally consider the problems of the resource management and illegal harvest. On the
other hand, both the Swanson model and our models take the factor of the
evolution of the management capacity into account. However, rather than
also considering issues of land use competition, as Swanson did, we confine
the models to addressing the related harvest and management issues of the
sustainable use strategy applied in given protected areas, while the Swanson
model did not handle the problem of illegal harvest. Furthermore, we provide a more deliberate modeling for relevant issues than the Swanson model
by introducing a new state variable, namely the management capital, and
178
thereby regarding the evolution of management capacity as a process of capital accumulation. This is the characteristic that differentiates our models
from almost all other bioeconomic models which either treated the management factor as a flow variable, as Swanson did, or addressed the issue of
accumulation of the capital utilized in harvesting renewable resources. In
the following the results of the theoretic models are briefly summarized.
The uniqueness of the steady state solution of the simple model in chapter
6 and of the extended model in chapter 7 can be confirmed. Under specific
assumptions, it can also be verified that the steady state solution of the
simple model and of the extended model is saddle point stable, while in
chapter 8 only the existence of the steady solution of the general model can
be proved. A special case of the simple model which did not consider the
non-consumptive value of resources possesses the properties of uniqueness
and saddle point stability without using similar assumptions in the simple
model.
The phase diagrams of the simple model can be derived, while those of
the extended model, as generally recognized, cannot be directly obtained via
analytical method as a result of the complex interaction between multiple
state and control variables. By the application of the computer simulation,
the phase diagrams of the general model were derived. The outcomes showed
that, on the optimal dynamic path, the resource stock and the harvest rate
will increase over time, if the initial resource stock is lower than the steady
state resource stock level. On the other hand, if the initial resource stock
is higher than the steady state resource stock level, the resource stock and
the harvest rate will simultaneously decrease over time. In addition, on the
optimal dynamic path, the management capital stock will increase while the
investment rate will decrease over time, if the initial management capital
stock is lower than the steady state management capital stock level. On the
contrary, if the initial management capital stock is higher than the steady
state stock level, the management capital stock will decrease while the investment rate will increase over time.
The poaching (illegally harvested quantity of resources) depends on both
the resource stock and management capital stock level. Thus, unlike the clear
interaction between the resource stock and harvest rate or between the management capital and investment rate, the development trend of the poaching
in general cannot be determined when the resource and management capital
stock simultaneously vary. It depends mainly on the initial conditions of the
resource and management capital stock. Different scenarios that leads to
different results were demonstrated in section 8.3.
179
Under the specific assumptions used to prove the saddle point stability
of the steady state solution of the simple model and of the extended model,
some critical comparative static effects of exogenous parameters on the equilibrium resource stock can be derived. By using the computer simulation, the
results from the comparative static analysis of the general model confirmed
the comparative static effects found in the simple model and the extended
model. To sum up, the lower the discount rate, the poaching coefficient, the
cost coefficient of investment and the depreciation rate of management capital, and the higher the intrinsic growth rate of species, the non-consumptive
value coefficient, the gross profit coefficient of harvest and the efficiency coefficient of management capital are, the higher the equilibrium resource stock
level will be.
Of these comparative static results, what especially worth while noting
is the impact of an variation of the gross profit coefficient of harvest on the
equilibrium resource stock. Contrary to the conclusion of the Clark model
and to the popular belief, the general model demonstrated that, other things
being equal, the higher the gross profit coefficient of harvest (termed as
price/harvest cost ratio in the context of the Clark model) is, the higher the
equilibrium resource stock level will be. This conclusion is consistent with
that drawn by the Swanson model. The reason underlying this difference in
model results is, that both Swanson´s and our model take the factor of the
evolution of management capacity into account, while the Clark model did
not. In the context of the Clark model, an increase in the price/harvest cost
ratio will enhance the incentive to harvest resources, without a concomitant
increase in the resource stock resulting from the devotion of a higher level
of management capital. However, in the context of our model, a higher
gross profit coefficient of harvest will induce more capital devoted to the
management of resources, and an improved management capacity will finally
result in a higher equilibrium resource stock level.
Although both the Swanson model and our general model agree with the
comparative static effect of an variation of the gross profit coefficient of harvest on the equilibrium resource stock, there are some differences between
the results of the two models. First, the general model clearly demonstrated
that, on the optimal time path, the harvest rate and the equilibrium resource stock vary in the same direction, and the investment rate and the
equilibrium management capital stock vary in the opposite direction, while
the Swanson model did not depict the interactions between these variables.
Next, the Swanson model cannot study the comparative static effects of some
other parameters such as the poaching coefficient, the cost coefficient of in180
vestment, the depreciation rate of management capital, the non-consumptive
value coefficient of species and the efficiency coefficient of management capital, because it treated management as a flow variable and therefore cannot
handle the relevant problems of capital accumulation, and it did not take the
problems of poaching, anti-poaching and non-consumptive value of species
into account.
As a result of the usually observed correlation between the gross profit coefficient of harvest and the poaching coefficient in the real world when certain
conservation policies are brought into practice, the impact of a simultaneous
variation of these two parameters on the equilibrium resource stock has also
been addressed. Although the net effect is in many cases ambiguous, such
discussion help investigate the origin of the controversy involving the conservation and the consumptive use of certain wild species, such as in the case
of the African elephant.
The size of the equilibrium resource stock depends on a variety of factors, as previously discussed. In a special case of the simple model which
did not consider the non-consumptive value of resources, the equilibrium resource stock level is always lower than the stock level which can afford the
maximum sustainable yield. Once the factors of the non-consumptive value
and poaching were taken into account, as the simple, the extended and the
general model did, the equilibrium resource stock level will not be necessarily
lower or higher than the maximum sustainable yield stock level.
10.1.2 Conclusions of the case studies
The case study involving the national park system of Taiwan showed that
the national park system is successful, at least at its beginning stage between
the year 1984 and 2000, in the sense that it has in principle effectively safeguarded biodiversity within park boundaries. Its success can be attributed to
the two pivotal factors: the political and financial support of the central government and the strict protection policy. But nowadays the strict protection
policy, as a critical factor contributing to safeguarding biodiversity, ironically hindered the planned enlargement of the national park system, because
it also intensified the conflicts between park authorities and local communities. Whether the national park system of Taiwan can overcome this problem
by means of some innovative approaches, such as the co-management, or by
modifying the strict preservation policy, remains to be seen.
The case study involving the Danayiku Nature Park at Shan-Mei, Taiwan
was conducted to explore the performance of a typical community-based conservation project, an important form of the sustainable resource use strategy.
181
Another five similar conservation projects were also addressed so that a comparison between different cases can be done. An assessment procedure based
on the findings of the general model was developed. The assessment showed
that only Shan-Mei possesses the biological, economic and social conditions
that are generally adequate to the application of the sustainable use strategy.
This explains to a great extent why sustainable use strategy scored a success
in Shan-Mei, while similar projects in vicinal communities failed or did not
work so well like Shan-Mei did.
Based on the findings of the Clark model and the Swanson model, three
fundamental factors, including the intrinsic growth rate of species, the discount rate and the price/cost ratio of harvest (the gross profit coefficient of
harvest in our terminology), were usually discussed when resource use problems and relevant conservation issues were concerned about. However, the
case study showed that, rather than the three often discussed factors, they
are non-consumptive value of resources, efficiency of management capital and
investment cost of management capital that explained the difference of the
performance of various use programs in this case. These findings confirmed
that it is worth while extending traditional bioeconomic models to study
problems of management capital accumulation, non-consumptive value of
wild species and illegal harvest. Such modifications help us have more insight
into the complex resource use problems and relevant conservation issues.
10.2 Policy implications
The theoretic models and case studies offered some critical policy implications. First, as defined in section 4.3, the use of renewable resources
is sustainable, if the equilibeium populations of utilized species will not be
reduced to such levels that they are vulnerable to local extinction, that their
ecological roles in the ecosystem is impaired, and that they lose their significance as useful resources to human users. Following these criteria, our
models showed that the impact of the sustainable use approach on conservation is double-edged, in the sense that the sustainable use approach will not
necessarily result in a better conservation status of renewable resources in a
given protected area, because, depending on a variety of biological, economic
and social conditions, the equilibrium resource stock could be higher or lower
than the initial stock level, and, in the case when the equilibrium resource
stock is lower than the initial stock level, it can sometimes be reduced to such
a low level that the previous criteria cannot be satified. This reflects the fact
that, in the complicated real world there is no single, omnipotent approach
that can solve all the conservation problems in different cases throughout the
182
world. The sustainable use approach could, or could not be a feasible conservation strategy. To evaluate the feasibility of the sustainable use of wild
species as a conservation strategy and to reduce the risk of overexploitation,
it always needs an overall assessment about relevant biological, economic and
social conditions at the site concerned. It is highly dangerous to judge the
feasibility of a sustainable use program by using only one or only few indicators, for example by using only the intrinsic growth rate and/or the discount
rate.
As section 8.3 concluded, the variation of the poaching rate is not an appropriate indicator for evaluating the success of the sustainable use strategy.
Under the general premise that the more the equilibrium resource stock closes
to the carrying capacity, the better it would be for the whole ecosystem, we
may use the equilibrium resource stock as an indicator for judging the feasibility of the use approach as a conservation strategy. Accordingly, the eight
parameters affecting the equilibrium resource stock may be viewed as indicators for evaluating the success probability of a sustainable use project before
or when it is brought into practice. The sustainable use strategy may potentially be more appropriate in sites with more positive indicators, namely
high intrinsic growth rate of species, non-consumptive value coefficient, gross
profit coefficient of harvest and efficiency coefficient of management capital,
and low discount rate, poaching coefficient, cost coefficient of investment and
depreciation rate of management capital, than those sites with less positive
indicators. Based on this conclusion, some policy implications can be drawn.
At the individual species level, some long-lived and slow-reproducing
species, such as primates, elephants, whales and sharks, have generally low
intrinsic growth rates and are particularly vulnerable to harvest. At the
ecosystem level, compared to open grasslands or habitats with a mosaic of
forest and grassland, forest ecosystems, especially tropical forests, as a whole
are particularly vulnerable to overharvesting of plant and mammalian communities. Hence, a specially cautious attitude toward their harvest problems
should be taken, although this does not necessarily imply that, from the perspective of conservation, they should not be utilized in any case. Sometimes
the positive effects of some other feasible conditions on the resource stock
may compensate for the risk resulted from the low intrinsic growth rate.
At the national level, compared to the developed countries, the developing countries are generally characterized by high discount rate, high poaching pressure and low non-consumptive value of wild species. Given these
conditions, the success probability of the sustainable use approach might
183
be relatively low in developing countries. This might imply that a more
conservative attitude toward the application of the sustainable use approach
should be taken in developing countries, especially in the tropical rain forests
where the overall reproductive rate of faunal and floral communities is low.
Certainly, one thing we should bear in mind is that the previous general conclusion neglects the considerable differences in socio-economic and biological
conditions between various countries, regions and habitat types. Whether the
use strategy is appropriate, depends always on the site- and species-specific
conditions. Nonetheless, the general conclusions provide a fundamental direction for the rethinking of the current conservation policies.
Based on the discussion in section 8.4 about the usual correlation between
the gross profit coefficient of harvest and the poaching coefficient, we may
conclude that sporting hunting as a conservation strategy might be a feasible
policy option, for example in the cases of the conservation of the African
elephant and the wildlife of Taiwan, because it could enhance the gross profit
coefficient of harvest while the poaching coefficient would not be affected.
Finally, the case study in the A-Li-Shan area of Taiwan showed that, under prevailing natural, economic and social conditions in Taiwan, the success
probability of community-based conservation programs initiated by indigenous communities is relatively low, and, from the viewpoint of conservation,
the performance of community-based conservation programs is in general not
so good as it is supposed to be, if those villages studied can represent the
ordinary indigenous communities of Taiwan. Compared to the national park
system managed by the central government, community-based conservation
regime can hardly create systems of a scale sufficient to preserve large portions of ecosystems. It is clear that, under current circumstances, only the
central government has the capacity to support and manage a large, systematic protected area network in Taiwan. It is unreasonable to suppose
that indigenous communities (and other local communities) can overcome
the problems like poaching, institution building and financial deficit more
effectively than the central government. The argument that decentralized,
community-based conservation programs managed by indigenous communities would work better than traditional protected areas managed by the central government should be questioned. If the community-based conservation
model is promoted considering the reasons of rural community development
and/or social justice, intensive outside (governmental or non-governmental)
supports will be needed to offset some disadvantageous exogenous conditions
prevailing in most local communities of Taiwan, such as active poaching,
high discount rate, high investment cost of management capital and low
184
non-consumptive value of renewable resources.
10.3 Limits in applicability of the theoretic model and
recommendations for further research
Mainly based on the general model, the conclusions and policy implications presented in section 10.1 and 10.2 were drawn. However, one should
also be careful to realize that some features of the general model could limit
it´s applicability.
First, the comparative static and the phase diagram analysis of the general model were conducted by using a set of specific functional forms for
relevant functions and by applying the method of computer simulation. Otherwise, no unambiguous results can be yielded via analytical method in the
comparative static analysis, and, as generally recognized, it is not possible
to depict the relevant phase diagrams as a result of the complex interaction
between multiple state and control variables. It is not certain, whether some
another sets of specific functional forms will fundamentally change the model
results.
Secondly, the phase diagram analysis of the general model demonstrated
that the optimal paths are globally monotonic. However, in models involving
multiple state and control variables, it usually happens that some initial
conditions for the state variables can be found so that the optimal paths
are non-monotonic. In this case, it must be recognized that we are not sure
whether the optimal paths of the general model are always monotonic.
Thirdly, the influence of certain exogenously given policies, for example
the lifting of the hunting ban, on the gross profit coefficient of harvest and
the poaching coefficient was addressed in section 8.4 and 8.5. Nonetheless,
the two coefficients are also exogenous variables in the general model. A
modified model would be better, if it can endogenize the two coefficients
and study what would happen, when certain exogenously given policies are
implemented.
Fourthly, we confine the theoretic models to addressing the related harvest and management issues when the sustainable use strategy is applied in
given protected areas. However, the influence of the sustainable use strategy
on land use decisions outside existing protected areas is an another critical
dimension that is worthy of being investigated, from the viewpoint of both
the economic theory and conservation. The general model might be modified
185
in the future to simultaneously consider the interaction between resource use,
land use competition, evolution of management capacity and illegal harvest.
Fifthly, our model is a typical one-species model which considers only
the harvested species. The impacts of harvesting the target species on all
the other species living within the same ecosystem are neglected in such a
model. The impacts would be especially enormous when the target species in
question is a keystone species. From the perspective of biology, most of such
impacts cannot even be modeled by the economic theory. Maybe, the general
model could be modified in order to address the cases when two species
have competitive or predatory relationships, and they are simultaneously
harvested or only one species is harvested.
Sixthly, apart from the intrinsic growth rate of species, there are still
many special biological factors that can influence the outcomes of the resource use. For example, tropical rain forest species are easily threatened by
local extinctions, because they are generally characterized by high diversity
and low densities. Species whose behavior allows easy harvest, that do not
have the ability to recolonize hunted area, or that are intrinsically rare are
highly vulnerable to harvest. In addition, our model is a deterministic model.
However, it has long been recognized that uncertainty which emerges from
the inherent stochasticity of ecosystems and from human ignorance about
biological and ecological knowledge usually leads to overexploitation and significantly raises the risk of extinction. It is therefore necessary to develop
a stochastic model and thereby to study the influence of the uncertainty on
the optimal use of renewable resources and relevant management issues.
Finally, some directions are worthy of being studied in further research.
In this dissertation, the resource harvest and management problems were
investigated under the premise of the sole ownership. In fact, some other
ownership regimes, for example the co-management which means that local
communities and park authorities share the management and use rights of
resources, might be as important as the sole ownership regime, and it is therefore worthy of being addressed, whether different forms of ownership regime
will lead to different outcomes of resource harvest and management. Furthermore, we may attempt to obtain detailed data involving resource stock,
harvest rate, management capital stock and investment, thereby conduct a
econometric analysis, test the results of the general model and quantify the
interactions between these variables. This may help offer a more precise assessment foundation for the field work and reduce the risk of overexploitation,
before or when a resource use project is brought into practice.
186
References
[1] Adams, J. S. and T. O. McShane (1996) The Myth of Wild Africa.
University of California Press.
[2] Adams, W. and D. Hulme (2001) Conservation and Community:
Changing Narratives, Policies & Practices in African Conservation. In
D. Hulme and M. Murphree (Eds.), African Wildlife & Livelihoods:
The Promise & Performance of community Conservation. James Currey Ltd., pp. 9-23.
[3] Arrow, K. J. and A. C. Fisher (1974) Environmental preservation, uncertainty, and irreversibility. Quarterly Journal of Economics, 88: 312319.
[4] Aylward, B. (1992) Appropriating the value of wildlife and wildlands.
In T. M. Swanson and E. B. Barbier (Eds.), Economics for the wilds.
Earthscan, London, pp. 34-64.
[5] Barbier, E. B. (1992) Economics for the wilds. In T. M. Swanson and
E. B. Barbier (Eds.), Economics for the wilds. Earthscan, London, pp.
15-33.
[6] Barnard, P., C. J. Brown, A. M. Jarvis, A. Robertson and L. V. Rooyen
(1998) Extending the Namibian protected area network to safeguard
hotspots of endemism and diversity. Biodiversity and Conservation, 7:
531-547.
[7] Barrow, E. and M. Murphree (2001) Community Conservation: From
Concept to Practice. In D. Hulme and M. Murphree (Eds.), African
Wildlife & Livelihoods: The Promise & Performance of community
Conservation. James Currey Ltd., pp. 24-37.
[8] Bennett, E. L. and J. G. Robinson (2000a) Hunting for the snark. In
J. G. Robinson and E. L. Bennett (Eds.), Hunting for Sustainability in
Tropical Forests. Columbia University Press, New York, pp. 1-9.
[9] Bennett, E. L. and J. G. Robinson (2000b) Hunting for sustainability:
The start of a synthesis. In J. G. Robinson and E. L. Bennett (Eds.),
Hunting for Sustainability in Tropical Forests. Columbia University
Press, New York, pp. 499-519.
[10] Bishop, R.C. (1978) Endangered species and uncertainty: The economics of a Safe Minimum Standard. American Journal of Agricultural
Economics, 60: 10-18.
187
[11] Bishop, R.C. (1982) Option value: An exposition and extention. Land
Economics, 58(1): 1-15.
[12] Bodmer, R. E. (1995a) Susceptibility of mammals to overhunting in
Amazonia. In J. Bissonette and P. Krausmanm (Eds.), Integrating People and Wildlife for a Sustainable Future, pp.292-5. Bethesda, MD: The
Wildlife Society.
[13] Bodmer, R. E. (1995b) Managing Amazonian wildlife: Biological correlates of game choice by detribalized hunters. Ecological Applications
5: 872-877.
[14] Bodmer, R. E., J. F. Eisenberg, and K. H. Redford (1997a) Hunting
and the likelihood of extinction of Amazonian mammals. Conservation
Biology, 11: 460-6.
[15] Bodmer, R. E., J. W. Penn, P. Puertas, L. Moya I., and T. G. Fang
(1997b) Linking conservation and local people through sustainable use
of natural resources: Community-based management in the Peruvian
Amazon. In C. H. Freese (ed.), Harvesting Wild Species: Implications
for Biodiversity Conservation. The Johns Hopkins University Press,
Baltimore and London, pp. 315-358.
[16] Bodmer, R. E. and P. E. Puertas (2000) Community-based comanagement of wildlife in the Peruvian Amazon. In J. G. Robinson and
E. L. Bennett (Eds.), Hunting for Sustainability in Tropical Forests.
Columbia University Press, New York, pp. 395-409.
[17] Boyce, J. R. (1995) Optimal capital accumulation in a fishery: A nonlinear irreversible investment model. Journal of Environmental Economics and Management, 28: 324-339.
[18] Brandon, K. (1997) Policy and practical considerations in land-use
strategies for biodiversity conservation. In R. Kramer, C. P. van Schaik
and J. Johnson (Eds.) Last Stand: Protected Areas and the Defense of
Tropical Biodiversity. Oxford University Press, pp. 90-114.
[19] Brandon, K., K. H. Redford, and S. E. Sanderson (Eds.) (1998) Parks in
Peril: People, Politics, and Protected Areas. Island Press, Washington,
D. C..
[20] Brandon, K. and M. Wells (1992) Planning for people and parks: design
dilemmas. World Development, 20: 557-570.
188
[21] Bromley, D. W. (1994) Economic dimensions of community-based conservation. In D. Western and R. M. Wright (Eds.), Natural Connections: Perspectives in Community-based Conservation. Island Press,
Washington, D. C., pp. 428-447.
[22] Brown, K. (1993) Biodiversity. In D. Pearce (ed.), Blueprint 3: Measuring sustainable development. Earthscan, London, pp. 98-114.
[23] Caro, T. M. and G. O´Doherty (1999) On the use of surrogate species
in conservation biology. Conservation Biology, 13(4): 805-814.
[24] Caughley, G. (1993) Elephants and Economics. Conservation Biology,
7: 943-945.
[25] Chang, L.-S. (2001) On indigenous hunting tradition and the modification of the National Park Act. The Nature, 70: 98-101. (In Chinese)
[26] Child, B. (2000) Application of the Southern African Experience to
Wildlife Utilization and Conservation in Kenya and Tanzania. In H. H.
T. Prins et al. (Eds.) Wildlife Conservation by Sustainable Use. Kluwer
Academic Publishers, pp. 459-467.
[27] Ciriacy-Wantrup, S. V. (1952) Resource Conservation. Berkeley: University of California Press.
[28] Clark, C. W. (1973) Profit maximization and the extinction of animal
species. Journal of Political Economy, 81(2): 950-961.
[29] Clark, C. W. (1976) Mathematical Bioeconomics: The Optimal Management of Renewable Resources. John Wiley & Sons, Inc..
[30] Clark, C. W. and G. R. Munro (1975) The economics of fishing and
modern capital theory: A simplified approach. Journal of Environmental Economics and Management, 2: 163-180.
[31] Clark, C. W., F. H. Clarke, and G. R. Munro (1979) The optimal
exploitation of renewable resource stocks: Problems of irreversible investment. Econometrica, 47: 25-47.
[32] Clayton, L. and E. J. Milner-Gulland (2000) The trade in wildlife in
north Sulawesi, Indonesia. In J. G. Robinson and E. L. Bennett (Eds.),
Hunting for Sustainability in Tropical Forests. Columbia University
Press, New York, pp. 473-496.
189
[33] Council of Agriculture and Department of National Parks, Construction and Planning Administration, Ministry of Interior, Republic of
China (COA and DNP) (1992) Island of Diversity-Nature Conservation in Taiwan. COA and DNP, Taiwan, ROC.
[34] Council of Agriculture, Republic of China (COA) (1997) The Nature
Reserves in Taiwan. COA, Taiwan, ROC.
[35] Construction and Planning Administration, Ministry of Interior, Republic of China (CPA) (2000) Yearly Report of Construction and Planning Affair Indicator, Taiwan and Fuchien Area, Republic of China.
CPA, Taiwan, ROC.
[36] Dai, C.-F., K.-M. Kuo, Y.-T. Chen, and C.-H. Chuang (1998) Changes
of coral communities in Nanwan Bay, Kenting National Park: 19871997. Journal of National Park, 8(2): 79-99. (In Chinese)
[37] Dai, C.-F., K.-M. Kuo, Y.-T. Chen, and C.-H. Chuang (1999) Changes
of coral communities on the east and west coast of the Kenting National
Park. Journal of National Park, 9(2): 131-143. (In Chinese)
[38] Daily, G. C. and P. R. Ehrlich (1995) Population Extinction and the
Biodiversity Crisis. In C. A. Perrings, K.-G. Mäler, C. Folke, et al.
(Eds.), Biodiversity Conservation. Kluwer Academic Publishers, The
Netherlands, pp. 45-56.
[39] Dinerstein, K. and E. D. Wikramanayake (1996) Beyond ´Hotspots´:
How to Prioritize Investments to Conserve Biodiversity in the IndoPacific Region. In F. B. Samson and F. C. Knopf (Eds.), Ecosystem
Management. Springer Verlag, New York, pp. 32-45.
[40] Dixon, J. A. and P. B. Sherman (1991) Economics of Protected Areas.
Earthscan, London.
[41] Department of National Parks, Construction and Planning Administration, Ministry of Interior, Taiwan, Republic of China (DNP) (1999)
National Parks of Taiwan. DNP, Taiwan, ROC.
[42] Dockner, E. (1985) Local satbility analysis in optimal control problems with two state variables. In G. Feichtinger (ed.), Optimal Control
Theory and Economic Analysis, 2: 89-103.
[43] Dublin, H. T., T. Milliken and R. F. W. Barnes (1995) Four Years
After the CITES Ban: Illegal killing of Elephants, Ivory Trade and
190
Stockpiles. Gland, Switzerland: IUCN/SSC African Elephant Specialist Group.
[44] Duffy, R. (2000) Killing for Conservation: Wildlife Policy in Zimbabwe.
Indiana University Press, Bloomington & Indianapolis.
[45] Ehrlich, P. and A. Ehrlich (1981) Extinction. Random House: New
York.
[46] Ehrlich, P. R. and G. C. Daily (1993) Population extinction and saving
biodiversity. AMBIO, 22: 64-68.
[47] Ehrlich, P. R. and E. O. Wilson (1991) Biodiversity Studies: Science
and Policy. Science, 253: 758-762.
[48] Eidsvik, H. K. (1992) Strengthening Protected Areas Through Philosophy, Science and Management: A Global Perspective. In J. H. M.
Willison, C. Drysdale, T. B. Herman, et al. (Eds.), Science and Management of Protected Areas. Elsevier, Amsterdam, pp. 9-18.
[49] Eisenberg, J. F. (1980) The density and biomass of tropical mammals.
In M. E. Soulé and B. A. Wilcox (eds.), Conservation Biology: An Evolutionary Ecological Perspective. Sunderland, MA: Sinauer Associates.
pp. 34-55.
[50] Eiswerth, M. E. and J. C. Haney (1992) Allocating conservation expenditures: accounting for inter-species genetic distinctiveness. Ecological
Economics, 5(1): 235-250.
[51] Erwin, T. (1988) The tropical forest canopy: The heart of biotic diversity. In E. O. Wilson (Ed.), Biodiversity. National Academy Press,
Washington, pp. 123-129.
[52] Executive Yuan, Republic of China (EYROC) (1981) Budget of the
Central Government, Budget Year 1982, Republic of China, p. 198. (In
Chinese)
[53] Executive Yuan, Republic of China (EYROC) (1982) Budget of the
Central Government, Budget Year 1983, Republic of China, p. 213. (In
Chinese)
[54] Executive Yuan, Republic of China (EYROC) (1983) Budget of the
Central Government, Budget Year 1984, Republic of China, p. 204. (In
Chinese)
191
[55] Executive Yuan, Republic of China (EYROC) (1984) Budget of the
Central Government, Budget Year 1985, Republic of China, p. 225. (In
Chinese)
[56] Executive Yuan, Republic of China (EYROC) (1985) Budget of the
Central Government, Budget Year 1986, Republic of China, pp. 234235. (In Chinese)
[57] Executive Yuan, Republic of China (EYROC) (1986) Budget of the
Central Government, Budget Year 1987, Republic of China, pp. 230233. (In Chinese)
[58] Executive Yuan, Republic of China (EYROC) (1987) Budget of the
Central Government, Budget Year 1988, Republic of China, pp. 271277. (In Chinese)
[59] Executive Yuan, Republic of China (EYROC) (1988) Budget of the
Central Government, Budget Year 1989, Republic of China, pp. 290296. (In Chinese)
[60] Executive Yuan, Republic of China (EYROC) (1989) Budget of the
Central Government, Budget Year 1990, Republic of China, pp. 310315. (In Chinese)
[61] Executive Yuan, Republic of China (EYROC) (1990) Budget of the
Central Government, Budget Year 1991, Republic of China, pp. 309314. (In Chinese)
[62] Executive Yuan, Republic of China (EYROC) (1991) Budget of the
Central Government, Budget Year 1992, Republic of China, pp. 342347. (In Chinese)
[63] Executive Yuan, Republic of China (EYROC) (1992) Budget of the
Central Government, Budget Year 1993, Republic of China, pp. 447457. (In Chinese)
[64] Executive Yuan, Republic of China (EYROC) (1993) Budget of the
Central Government, Budget Year 1994, Republic of China, pp. 652666. (In Chinese)
[65] Executive Yuan, Republic of China (EYROC) (1994) Budget of the
Central Government, Budget Year 1995, Republic of China, pp. 585600. (In Chinese)
192
[66] Executive Yuan, Republic of China (EYROC) (1995) Budget of the
Central Government, Budget Year 1996, Republic of China, pp. 605619. (In Chinese)
[67] Executive Yuan, Republic of China (EYROC) (1996) Budget of the
Central Government, Budget Year 1997, Republic of China, pp. 612628. (In Chinese)
[68] Executive Yuan, Republic of China (EYROC) (1997) Budget of the
Central Government, Budget Year 1998, Republic of China, pp. 520532. (In Chinese)
[69] Executive Yuan, Republic of China (EYROC) (1998) Budget of the
Central Government, Budget Year 1999, Republic of China, pp. 317329. (In Chinese)
[70] Executive Yuan, Republic of China (EYROC) (1999) Budget of the
Central Government, Budget Year 2000, Republic of China, pp. 273255. (In Chinese)
[71] Fa, J. E., J. Juste, J. Perez del Val, and J. Castroviejo (1995) Impact of
market hunting on mammal species in Equatorial Guinea. Conservation
Biology 9: 1107-1115.
[72] Fisher, A. C. and J. V. Krutilla (1985) Economics of Nature Preservation. In V. K. Allen and J. L. Sweeney (Eds.), Handbook of Natural
Resource and Energy Economics, Volume I, Elsevier, pp. 165-189.
[73] Fisher, A. C. and W. M. Hanemann (1987) Quasi-option value: Some
misconceptions dispelled. Journal of Environmental Economics and
Management, 14: 183-190.
[74] Fitzgibbon, C. D., J. Mogaka, and J. H. Fanshawe (1995) Subsistence
hunting in Arabuko-Sokoke Forest, Kenya, and its effects on mammal
populations. Conservation Biology 9: 1116-1126.
[75] Freeman, A. M. (1985) Supply uncertainty, option price and option
value. Land Economics, 61: 176-181.
[76] Freese, C. H. (Ed.) (1997) Harvesting Wild Species: Implications for
Biodiversity Conservation. The Johns Hopkins University Press, Baltimore.
[77] Freese, C. H. (1998) Wild Species as Commodities: Managing Markets
and Ecosystems for Sustainability. Island Press, Washington, D. C..
193
[78] Gadgil, M. (1992) Conserving biodiversity as if people matter: A case
study from India. Ambio, 21: 266-270.
[79] Ghimire, K. B. and M. P. Pimbert (1997) Social Change and Conservation: An Overview of Issues and Concepts. In K. B. Ghimire and M. P.
Pimbert (Eds.) Social Change and Conservation. Environmental Politics and Impacts of National Parks and Protected Areas. Earthscan,
London, pp. 1-45.
[80] Gordon, H. S. (1954) The economic theory of a common property resource: The fishery. Journal of Political Economy, 62: 124-142.
[81] Gould, J. R. (1972) Extinction of a fishery by commercial exploitation:
A note. Journal of Political Economy, 80: 1031-1038.
[82] Green, M. J. B. and J. Paine (1999) State of the world´s protected
areas at the end of the 20th century. In S. Stolton and N. Dudley
(Eds.), Partnerships for Protection: New Strategies for Planning and
Management for Protected Areas. Earthscan Publications Ltd., pp. 1928.
[83] Grootenhuis, J. G. and H. H. T. Prins (2000) Wildlife utilisation: a
justified option for sustainable land use in African savannas. In H. H. T.
Prins, J. G. Grootenhuis and T. T. Dolan (Eds.) Wildlife Conservation
by Sustainable Use. Kluwer Academic Publishers, pp. 469-482.
[84] Grove, N. (1988) Quietly Conserving Nature. National Geographic,
174(January): 818-844.
[85] Gullison, R. E. (1998) Chapter 6: Will Bigleaf Mahogany be conserved
through sustainable use? In E. J. Milner-Gulland and R. Mace, Conservation of Biological Resources. Blackwell Science Ltd.
[86] Hackel, J. D. (1999) Community conservation and the future of Africa´s
wildlife. Conservation Biology, 13(4): 726-734.
[87] Hart, J. A. (2000) Impact and sustainability of indigenous hunting in
the Ituri Forest, Congo-Zaire: A comparison of unhunted and hunted
duiker populations. In J. G. Robinson and E. L. Bennett (Eds.), Hunting for Sustainability in Tropical Forests. Columbia University Press,
New York, pp. 106-153.
[88] Hawksworth, D. L. (1991) The fungal dimension of biodiversity: Magnitude, significance, and conservation. Mycological Research, 95: 641655.
194
[89] Hearne, J. and M. Mckenzie (2000) Compelling reasons for game ranching in Maputaland. In H. H. T. Prins, J. G. Grootenhuis and T. T.
Dolan (Eds.) Wildlife Conservation by Sustainable Use. Kluwer Academic Publishers, pp. 417-438.
[90] Henry, C. (1974) Investment decisions under uncertainty: The ´irreversibility effect´. American Economic Review, 64(6): 1006-1012.
[91] Huang, Y.-W. (1999) Aboriginal reserves policy in Taiwan´s national
parks- an institutional and spatial perspective. Journal of National
Park, 9(2): 182-198. (In Chinese)
[92] Hurt, R. and P. Ravn (2000) Hunting and its benefits: an overview of
hunting in Africa with special reference to Tanzania. In H. H. T. Prins,
J. G. Grootenhuis and T. T. Dolan (Eds.) Wildlife Conservation by
Sustainable Use. Kluwer Academic Publishers, pp. 295-313.
[93] Ivory Trade Review Group (ITRG) (1989) The Ivory Trade and the
Future of the African Elephant. Report to the Conference of the Parties
to CITES, Lausanne.
[94] IUCN (1994) Guidelines for Protected Area Management Categories,
CNPPA with the assistance of WCMC, IUCN, Gland and Cambridge.
[95] IUCN (1998) 1997 United Nations List of Protected Areas, prepared
by WCMC and WCPA, IUCN, Gland and Cambridge.
[96] IUCN/UNEP/WWF (1980) World Conservation Strategy. Living
resource conservation for sustainable development. IUCN, Gland,
Switzerland.
[97] IUCN/UNEP/WWF (1991) Caring for the Earth: a strategy for sustainable living. IUCN, Gland, Switzerland.
[98] Janzen, D. H. (1994) Wildland biodiversity management in the tropics:
Where are we now and where are we going?. Vida Silvestre Neotropical,
3: 3-15.
[99] Johansson, P.-O. (1988) On the properties of supply-side option value.
Land Economics, 64: 86-97.
[100] Katz, E. G. (2000) Social capital and natural capital: A comparative
analysis of land tenure and natural resource management in Guatemala.
Land Economics, 76(1): 114-132.
195
[101] Kock, M. D. (1996) Zimbabwe: a model for the sustainable use of
wildlife and the development of innovative wildlife management pratices. In V. J. Taylor and N. Dunstone (Eds.) The Exploitation of Mammal Populations. Chapman & Hall, London, pp. 229-249.
[102] Kramer, R. A. and C. P. van Schaik (1997) Preservation Paradigms and
Tropical Rain Forests. In R. Kramer, C. P. van Schaik and J. Johnson (Eds.) Last Stand: Protected Areas and the Defense of Tropical
Biodiversity. Oxford University Press, pp. 3-14.
[103] Kramer, R., C. P. van Schaik and J. Johnson (Eds.) (1997) Last Stand:
Protected Areas and the Defense of Tropical Biodiversity. Oxford University Press.
[104] Krautkraemer, J. A. (1995) Incentives, Development and Population:
A Growth-Theoretic Perspective. In T. M. Swanson (Ed.), The Economics and Ecology of Biodiversity Decline: The Forces Driving Global
Change. Cambridge University Press, pp. 13-24.
[105] Krutilla, J. V. (1967) Conservation reconsidered. American Economic
Review, 57: 777-786.
[106] Lant, C. L. (1994) The role of property rights in economic research on
U.S. wetlands policy. Ecological Economics, 11: 27-33.
[107] Laurance, W. F., H. L. Vasconcelos and T. E. Lovejoy (2000) Forest
loss and fragmentation in the Amazon: implications for wildlife conservation. Oryx, 34(1): 39-45.
[108] Lavigne, D. M., C. J. Callaghan and R. J. Smith (1996) Sustainable
utilization: the lessons of history. In V. J. Taylor and D. Dunstone
(Eds.), The Exploitation of Mammal Populations, Chapman & Hall,
London, pp. 250-265.
[109] Lee, R. J. (2000) Impact of subsistence hunting in north Sulawesi, Indonesia, and conservation options. In J. G. Robinson and E. L. Bennett
(Eds.), Hunting for Sustainability in Tropical Forests. Columbia University Press, New York, pp. 455-472.
[110] Lewis, D. M. and A. Phiri (1998) Wildlife snaring-an indicator of community response to a community-based conservation project. Oryx,
32(2): 111-121.
196
[111] Li, C.-Z. and K.-G. Löfgren (1998) A dynamic model of biodiversity
preservation. Environment and Development Economics, 3: 157-172.
[112] Li, T.-M. and H.-C. Tang (1999) Forever Danayiku: retrospect in ShanMei community development. Unpublished research report of Chia-Yi
County and the Political Institute of National Chung-Cheng University,
Taiwan. (In Chinese)
[113] Lin, L. (2000) Nature reserves and national parks. Quarterly Journal
of Construction, 9(4): 13-30. (In Chinese)
[114] Lin, Y,-S. (2000) The treaty of biodiversity (5): protected areas. The
Nature, 69: 110-115. (In Chinese)
[115] Liu, J.-S. (2000) Promoting indigenous economy through sustainable
use of natural resources. The Nature, 67: 34-41. (In Chinese)
[116] Lovejoy, T. E. (1980) A Projection of Species Extinction. In G. Barney
(Ed.), The Global 2000 Report to the President. Council on Environmental Quality: Washington, D. C..
[117] Lovejoy, T. E., R. O. Bierregaard Jr., A. B. Rylands, J. R. Malcolm,
C. E. Quintela, L. H. Harper, K. S. Brown, Jr., A. H. Powell, G.
V. N. Powell, H. O. R. Schubart, and M. B. Hays (1986) Edge and
other effects of isolation on Amazon forest fragments. In M. E. Soulé
(Ed.) Conservation Biology: The Science of Scarcity and Diversity.
Sunderland, Mass.: Sinauer, pp. 257-285.
[118] Lu, D.-J. (1999) Participation, Institutions and Protected Area
Management- A Qualitative Analysis of the Wildlife Refuges in Taiwan. Unpublished dissertation of University of Wales, Aberystwyth.
[119] Ludwig, D., R. Hilborn and C. Walters (1993) Uncertainty, resource
exploitation and conservation: lessons from history. Science, 260(17):
36.
[120] Lugo, A. E., J. A. Parrotta and S. Brown (1993) Loss of species caused
by tropical deforestation and their recovery through management. AMBIO, 22(2-3): 106-109.
[121] MacArthur, R. M. and E. O. Wilson (1967) The Theory of Island
Biogeography. Monographs in Population Biology. Princeton University
Press, Princeton, New Jersey.
197
[122] Machlis, G. E. and D. L. Tichnell (1987) Economic development and
threats to National Parks: a preliminary analysis. Environmental Conservation, 14(2): 151-156.
[123] MacKinnon, K. (1997) The ecological foundations of biodiversity protection. In R. Kramer, C. V. Schaik and J. Johnson (Eds.) Last Stand:
Protected Areas and the Defense of Tropical Biodiversity. Oxford University Press, pp. 36-63.
[124] Madhusudan, M. D. and K. U. Karanth (2000) Hunting for an answer:
Is local hunting compatible with large mammal conservation in India?
In J. G. Robinson and E. L. Bennett (Eds.), Hunting for Sustainability
in Tropical Forests. Columbia University Press, New York, pp. 339-355.
[125] Mangel, M., L. M. Talbot, G. K. Meffe, M. T. Agardy, D. L. Alverson, J. Barlow, D. B. Botkin, G. Budowski, T. Clark, J. Cooke, R.
H. Crozier, P. K. Dayton, D. L. Elder, C. W. Fowler, S. Funtowicz,
J. Giske, R. J. Hofman, S. J. Holt, S. R. Kellert, L. A. Kimball, D.
Ludwig, K. Magnusson, B. S. Malayang III, C. Mann, E. A. Norse, S.
P. Northridge, W. F. Perrin, C. Perrings, R. M. Peterman, G. B. Rabb,
H. A. Regier, J. E. Reynolds III, K. Sherman, M. P. Sissenwine, T. D.
Smith, A. Starfield, R. J. Taylor, M. F. Tillman, C. Toft, J. R. Twiss,
Jr., J. Wilen, and T. P. Young (1996) Principles for the conservation
of wild living resources. Ecological Applications, 6: 338-362.
[126] McCullough, D. R. (1984) Lessons from the George Reserve, Michigan.
In L. K. Halls (ed.), White-tailed deer: Ecology and management.
Harrisburg, Pa.: Stackpole.
[127] McNeely, J. A. (1988) Economics and biological diversity: Developing
and using economic incentives to conserve biological resources. Gland,
Switzerland: IUCN.
[128] McNeely, J. A., J. R. Miller, W. V. Reid, et al. (1990) Conserving the
World´s Biological Diversity. IUCN, Gland, Switzerland.
[129] Medellı́n, R. A. (1999) Sustainable Harvest for Conservation. Conservation Biology, 13(2): 225.
[130] Meffe, G. K. and C. R. Carroll (1994) Principles of Conservation Biology. Sinauer Associates, Inc., Sunderland, Massachusetts.
198
[131] Milner-Gulland, E. J. and N. Leader-Williams (1992) Illegal exploitation of wildlife. In T. M. Swanson and E. B. Barbier (Eds.), Economics
for the Wilds. Earthscan, London.
[132] Mitchell, J. G. (1994) Our National Parks. National Geographic,
186(4): 2-55.
[133] Mittermeier, R. A. and T. B. Werner (1990) Wealth of plants and
animals unites ´megadiversity´countries. Tropicus, 4: 1, 4-5.
[134] Murphree, M. W. (1994) The role of institutions in community-based
conservation. In D. Western and R. M. Wright (Eds.), Natural Connections: Perspectives in Community-based Conservation. Island Press,
Washington, D. C., pp. 403-427.
[135] Myers, N. (1988) Threatened biotas: ´hotspots´ in tropical forests.
The Environmentalists, 8: 1-20.
[136] Myers, N. (1994) Global Biodiversity II: Losses. In G. K. Meffe and
C. R. Carroll (Eds.), Principles of Conservation Biology. Sinauer Associates, Inc., Sunderland, Massachusetts, pp. 110-140.
[137] Noss, A. J. (1997) Challenges to nature conservation with community
development in central African forests. Oryx, 31(3): 180-188.
[138] Noss, R. F. (1991) Sustainability and wilderness. Conservation Biology,
5:120-122.
[139] Oates, J. F. (1999) Myth and Reality in the Rain Forest: How Conservation Strategies Are Failing in West Africa. University of California
Press.
[140] Orians, G. H. (1994) Global Biodiversity I: Patterns and Processes.
In G. K. Meffe and C. R. Carroll (Eds.), Principles of Conservation
Biology. Sinauer Associates, Inc., Sunderland, Massachusetts, pp. 78109.
[141] Owen, D. F. (1992) Chapter 6: The abundance and biomass of forest
animals. In F. B. Golley (Ed.), Ecosystems of the world 14A: Tropical
rain forest ecosystems. Elsevier Scientific Publishing Company.
[142] Pearce, D. W. and R. K. Turner (1990) Economics of Natural Resources
and the Environment. Harvester Wheatsheaf.
[143] Pei, K. J. C. (2001) Dancing with Wildlife. (In Chinese)
199
[144] Peres, C. A. (2000) Evaluating the impact and sustainability of subsistence hunting at multiple Amazonian Forest sites. In J. G. Robinson and E. L. Bennett (Eds.), Hunting for Sustainability in Tropical
Forests. Columbia University Press, New York, pp. 31-56.
[145] Perrings, C. and D. Pearce (1994) Threshold effects and incentives
for the conservation of biodiversity. Environmental and Resource Economics, 4: 13-28.
[146] Phillips, A. and J. Harrison (1999) The Framework for International
Standards in Establishing National Parks and Other Protected Areas.
In S. Stolton and N. Dudley (Eds.), Partnerships for Protection: New
Strategies for Planning and Management for Protected Areas. Earthscan Publications Ltd., pp. 13-17.
[147] Pimbert, M. P. and J. N. Pretty (1997) Parks, People and Professionals:
Putting ´Participation´ into Protected Area Management. In K. B.
Ghimire and M. P. Pimbert (Eds.) Social Change and Conservation.
Environmental Politics and Impacts of National Parks and Protected
Areas. Earthscan, London, pp. 297-330.
[148] Powell, G. V. N., J. Barborak and M. Rodriguez S. (2000) Assessing
representativeness of protected natural areas in Costa Rica for conserving biodiversity: a preliminary gap analysis. Biological Conservation,
93: 35-41.
[149] Prescott-Allen, R. and C. Prescott-Allen (1996) The good, the bad,
and the neutral: assessing the sustainability of uses of wild species. In
R. and C. Prescott-Allen (Eds.) Assessing the Sustainability of Uses
of Wild Species. Occational Paper of the IUCN Species Survival Commission No. 12. Cambridge, UK, pp. 81-101.
[150] Primack, R. B. (1998) Essentials of Conservation Biology, Second Edition. Sinauer Associates, Inc., Massachusetts, U.S.A..
[151] Prins, H. H. T., J. G. Grootenhuis, and T. T. Dolan (Eds.) (2000)
Wildlife Conservation by Sustainable Use. Kluwer Academic Publishers, Boston.
[152] Rasker, R., M. V. Martin ,and R. L. Johnson (2000) Economics: Theory versus Practice in Wildlife Management. In M. A. Michael (Ed.)
Preserving Wildlife: An International Perspective. Humanity books,
New York, pp. 239-262.
200
[153] Raven, P. H. (1988) Our Diminishing Tropical Forests. In E. O. Wilson
(Ed.), Biodiversity. Washington, D. C., pp. 119-122.
[154] Redford, K. H. (1991) The Ecologically Noble Savage. Cultural Survival
Quarterly, 15(1): 46-48.
[155] Reid, W. and K. Miller (1989) Keeping Options Alive, World Resources
Institute: Washington, D. C..
[156] Reid, W. (1992) How Many Species Will There Be? In T. C. Whitmore
and J. A. Sayer (Eds.) Tropical Deforestation and Species Extinction.
Chapman and Hall, London.
[157] Reid, W. V., J. A. McNeely, D. B. Tunstall, D. A. Bryant and M.
Winograd (1993) Biodiversity Indicators for Policy-Makers. World Resources Institute, Washington, D. C..
[158] Ricklefs, R. E. (1990) Ecology. W. H. Freeman and Company, New
York.
[159] Robinson, J. G. (1993) The limits to caring: Sustainable living and the
loss of biodiversity. Conservation Biology, 7; 20-28.
[160] Robinson, J. G. (2000) Appendix: Calculating maximum sustainable
harvests and percentage offtakes. In J. G. Robinson and E. L. Bennett (Eds.), Hunting for Sustainability in Tropical Forests. Columbia
University Press, New York, pp. 521-524.
[161] Robinson, J. G. and E. L. Bennett (2000) Carrying capacity limits to
sustainable hunting in tropical forests. In J. G. Robinson and E. L. Bennett (Eds.), Hunting for Sustainability in Tropical Forests. Columbia
University Press, New York, pp. 13-30.
[162] Robinson, J. G. and K. H. Redford (1986) Intrinsic rate of natural
increase in neotropical forest mammals: Relationship to phylogeny and
diet. Oecologia, 68: 516-520.
[163] Robinson, J. G. and K. H. Redford (1991) Sustainable harvest of
neotropical forest mammals. In J. G. Robinson and K. H. Redford
(Eds.), Neotropical Wildlife Use and Conservation. Chicago: University of Chicago Press, pp. 415-429.
[164] Romer, P. M. (1990) Endogenous technological change. Journal of Political Economy, 98(5): S71-S102.
201
[165] Roth, H. H. (1997a) Section 2.3.8. Elephants (P roboscidea). In H. H.
Roth and G. Merz (Eds.), Wildlife Resources: A Global Account of
Economic Use. Springer Verlag, pp. 244-256.
[166] Roth, H. H. (1997b) Section 2.3.10. Suiform Ungulates
(Nonruminantia : Artiodactyla). In H. H. Roth and G. Merz
(Eds.), Wildlife Resources: A Global Account of Economic Use.
Springer Verlag, pp. 264-271.
[167] Sattler, P. (1992) Planning Towards Consolidation of Queensland´s
National Park Estate. In J. H. M. Willison, C. Drysdale, T. B. Herman,
et al. (Eds.), Science and Management of Protected Areas. Elsevier,
Amsterdam, p. 107-116.
[168] Sayer, J. A. and S. Stuart (1988) Biological diversity and tropical
forests. Environmental Conservation, 15: 193-194.
[169] Schmalensee, R. (1972) Option demand and consumer´s surplus: Valuing price changes under uncertainty. American Economic Review,
62(1): 813-824.
[170] Scott, A. D. (1955) The fishery: The objectives of sole ownership.
Journal of Political Economy, 63: 116-124.
[171] Scott, J. M., B. Csuti, J. D. Jacobi, et al. (1987) Species richness.
BioScience, 37: 782-788.
[172] Shafer, C. L. (1999) History of selection and system planning for US
natural area national parks and monuments: beauty and biology. Biodiversity and Conservation, 8: 189-204.
[173] Shah, A. (1995) The Economics of Third World National Parks. Edward Elgar, UK.
[174] Shaw, J. H. (1991) The outlook for sustainable harvests of wildlife in
Latin America. In J. G. Robinson and K. H. Redford (Eds.), Neotropical Wildlife Use and Conservation. Chicago: University of Chicago
Press, pp. 24-34.
[175] Sherman, P. B. (1989) Market Failure and the Underprovision of Parks
and Protected Areas. Unpublished Ph. D. dissertation, University of
Hawaii.
202
[176] Siegfried, W. R. (1989) Preservation of Species in Southern African
Nature Reserves. In B. J. Huntley (Ed.), Biotic Diversity in Southern
Africa. Oxford University Press, Cape Town.
[177] Simberloff, D. (1986) Are we on the Verge of an Mass Extinction in
Tropical Rain Forests?. In D. Elliot (Ed.), Dynamics of Extinction,
John Wiley: New York.
[178] Simberloff, D. (1998) Flagships, umbrellas, and keystones: is singlespecies management passé in the landscape era?. Biological Conservation, 83(3): 247-257.
[179] Skonhoft, A. and J. T. Solstad (1996) Wildlife management, illegal
hunting and conflicts: A bioeconomic analysis. Environment and Development Economics 1: 165-181.
[180] SMCDS (1994) Financial Report of Shan-Mei Community Development Society 1994. (In chinese)
[181] SMCDS (1995) Financial Report of Shan-Mei Community Development Society 1995. (In chinese)
[182] SMCDS (1996) Financial Report of Shan-Mei Community Development Society 1996. (In chinese)
[183] SMCDS (1997) Financial Report of Shan-Mei Community Development Society 1997. (In chinese)
[184] SMCDS (1998) Financial Report of Shan-Mei Community Development Society 1998. (In chinese)
[185] SMCDS (1999) Financial Report of Shan-Mei Community Development Society 1999. (In chinese)
[186] SMCDS (2001) Financial Report of Shan-Mei Community Development Society 2000. (In chinese)
[187] Smith, V. L. (1968) Economics of production from natural resources.
American Economic Review, 58: 409-431.
[188] Smith, V. L. (1969) On models of commercial fishing. Journal of Political Economy, 77: 181-198.
[189] Solow, A., S. Polasky and J. Broadus (1993) On the measurement of
biological diversity. Journal of Environmental Economics and Management, 24: 60-68.
203
[190] Songorwa, A. N., T. Bührs, and K. F. D. Hughey (2000) CommunityBased Wildlife Management in Africa: A Critical Assessment of the
Literature. Natural Resources Journal, 40: 603-643.
[191] Spinage, C. (1998) Social change and conservation misrepresentation
in Africa. Oryx, 32(4): 265-276.
[192] Stevens, S. (1997) The Legacy of Yellowstone. In S. Stevens (Ed.)
Conservation Through Cultural Survival: Indigenous Peoples and Protected Areas. Island Press, pp. 13-32.
[193] Stiling, P. D. (1992) Introductory Ecology. Prentice Hall, Englewood
Cliffs, NJ.
[194] Struhsaker, T. T. (1998) A biologist´s perspective on the role of sustainable harvest in conservation. Conservation Biology, 12(4): 930-932.
[195] Sung, B.-M. (1999) The strategy for the impacts on indigenous culture
from national parks in Taiwan. Journal of National Park, 9(1): 65-80.
(In Chinese)
[196] Swanson, T. M. (1994) The International Regulation of Extinction.
Macmillan, London.
[197] Swallow, B. M. and D. W. Bromley (1995) Institutions, governence
and incentives in common property regimes for african rangelands. Environmental and Resource Economics, 6: 99-118.
[198] Terborgh, J. and C. P. van Schaik (1997) Minimizing Species Loss:
The Imperative of Protection. In R. Kramer, C. P. van Schaik and
J. Johnson (Eds.) Last Stand: Protected Areas and the Defense of
Tropical Biodiversity. Oxford University Press, pp. 15-35.
[199] The Nature (1991) Newsletter, 32, p.121. (In Chinese)
[200] The Nature (1993) Newsletter, 40, p.118. (In Chinese)
[201] The Nature (1994a) Newsletter, 42, p.120. (In Chinese)
[202] The Nature (1994b) Newsletter, 43, p.114. (In Chinese)
[203] The Nature (1995a) Newsletter, 47, p.120. (In Chinese)
[204] The Nature (1995b) Newsletter, 49, p.114. (In Chinese)
[205] The Nature (1996) Newsletter, 53, p.120. (In Chinese)
204
[206] The Nature (1999) Newsletter, 63, p.118. (In Chinese)
[207] The Nature (2000) An Interview with the Chief Director of the Construction and Planning Administration, 67: 104-111. (In Chinese)
[208] Turner, A. M., C. D. A. Rubec and E. B. Wiken (1992) Canadian
Ecosystems: A Systems Approach to Their Conservation. In J. H. M.
Willison, C. Drysdale, T. B. Herman, et al. (Eds.), Science and Management of Protected Areas. Elsevier, Amsterdam, p. 117-127.
[209] Udvardy, M. D. F. (1975) A Classification of the Biogeographical
Provinces of the World, Occasional Paper 18, IUCN, Gland, Switzerland.
[210] Wacker, H. and J.E. Blank (1998) Ressourcenökonomik, Band I:
Einführung in die Theorie regenerativer natürlicher Ressourcen. Oldenbourg Verlag, München.
[211] Wade, R. (1987) The management of common property resources: Collective action as an alternative to privatisation or state regulation.
Cambridge Journal of Economics, 11(2): 95-106.
[212] Wood, D. (1995) Conserved to death: Are tropical forests being overprotected from people? Land Use Policy, 12(2): 115-135.
[213] World Conservation Monitoring Centre (WCMC) (1992) Global Biodiversity: Status of the Earth´s Living Resources. London: Chapman &
Hall.
[214] Weisbrod, B. (1964) Collective-consumption services of individualconsumption goods. Quarterly Journal of Economics, 78: 471-477.
[215] Weitzman, M. L. (1992) On diversity. Quarterly Journal of Economics,
107: 363-406.
[216] Wen, I.-J. (2000) Conservation of Danayiku and the community development of Shan-Mei. In China Times Foundation (Ed.), Rivers and
Community. China Times Press, Taipei, Taiwan. pp. 171-184.
[217] Western, D. and R. M. Wright (1994) The background to communitybased conservation. In D. Western and R. M. Wright (Eds.), Natural
Connections: Perspectives in Community-based Conservation. Island
Press, Washington, D. C., pp. 1-12.
205
[218] Whittaker, R. H. (1975) Communities and Ecosystems, 2d ed.. Macmillan, New York.
[219] Wildlife Conservation Law (WCL) (1994) The Central Government of
the Republic of China.
[220] Wilson, E. O. (1988) The current state of biological diversity. In E. O.
Wilson (Ed.), Biodiversity. National Academy Press, Washington, pp.
3-27.
[221] Wilson, E. O. (1992) The Diversity of Life. The Belknap Press of Harvard University Press, Cambridge, MA.
[222] Woodruff, D. S. (1989) The Problems of Conserving Genes and Species.
In D. Western and M. C. Pearl (Eds.), Conservation for the TwentyFirst Century, Oxford University Press, New York, pp. 76-88.
[223] World Resources Institute (WRI) (1994) World Resources 1994-95. Oxford University Press, Oxford, p.192.
206
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Related manuals

Download PDF

advertisement