TROPICAL FOREST CONSERVATION

This page intentionally left blank

TROPICAL FOREST
CONSERVATION
An Economic Assessment of the
Alternatives in Latin America

DOUGLAS SOUTHGATE

New York Oxford

Oxford University Press

1998
Oxford University Press
Oxford New York
Athens Auckland Bangkok Bogota Bombay
Buenos Aires Calcutta Cape Town Dar es Salaam
Delhi Florence Hong Kong Istanbul Karachi
Kuala Lumpur Madras Madrid Melbourne
Mexico City Nairobi Paris Singapore
Taipei Tokyo Toronto Warsaw
and associated companies in
Berlin Ibadan

Copyright © 1998 by Oxford University Press, Inc.

Published by Oxford University Press, Inc.
198 Madison Avenue, New York, New York 10016
Oxford is a registered trademark of Oxford University Press
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise,
without the prior permission of Oxford University Press.

Library of Congress Cataloging-in-Publication Data

Southgate, Douglas DeWitt, 1952-
Tropical forest conservation : an economic assessment of the alternatives in
Latin America / by Douglas Southgate.
p. cm.
Includes bibliographical references and index.
ISBN 0-19-510996-1
1. Rain forest conservation—Latin America. 2. Deforestation—
Control—Latin America. 3. Forest ecology—Latin America. 4. Non-
timber forest resources—Latin America. 5. Sustainable forestry—
Latin America. 6. Habitat conservation—Latin America.
7. Ecotourism—Latin America. I. Title.
SD414.L29S68 1998
333.75'16'0980913—dc21 97-34601

9 8 7 6 5 4 3 2 1
Printed in the United States of America
on acid free, recycled paper
Acknowledgments

T his book has benefited from the advice and sup-

port offered by a number of colleagues and
friends. First and foremost, William ("Jeff") Vaughn, of the
Inter-American Development Bank, hired me, as a consult-
ant, in 1995 to assess the contributions that nontimber
extraction, low-impact logging, genetic prospecting, and eco-
tourism can make to tropical forest conservation in Latin
America. The consultancy provided me with a unique op-
portunity to interview experts and to visit sites in Brazil,
Costa Rica, and Ecuador. All this greatly enriched the content
of chapters 4-7 of this book.
I deeply appreciate the comments on a draft text that I
received from Vaughn and five other individuals: Bruce Ayl-
ward (of the Tropical Science Center in San Jose, Costa Rica),
John Browder (of Virginia Polytechnic Institute and State
University), David Simpson (of Resources for the Future),
Steven Stone (of Cornell University), and Christopher Uhl (of
Pennsylvania State University). Also very useful were re-
views of my Inter-American Development Bank report pro-
vided by Glenn Prickett and John Reid (of Conservation
International) and four of Vaugin's associates: Sergio Ardila,
Arthur Darling, Kari Keipi, and Gil Nolet.
Among others, John Dixon (of the World Bank), Hans Gre-
gersen (of the University of Minnesota), and David Pearce (of
University College London) encouraged me to undertake an
examination of tropical forest conservation measures being
applied in Latin America. As work proceeded, Dixon and
three of his World Bank colleagues—Luis Constantino, John
Kellenberg, and Robert Schneider—furnished a great deal of
vi Acknowledgments

useful information. So did Marc Dourojeanni, Michelle Le-

may, and Raul Tuazon, all of the Inter-American Develop-
ment Bank. Several other people deserve special thanks,
including Alfredo Carrasco (of the Charles Darwin Founda-
tion), Howard Clark, Douglas McMeekin, Roberto Ulloa, and
Robert Vogel—all of Quito, Ecuador; Jaime Echeverria and
Joe Tosi of the Tropical Science Center; Adalberto Verissimo
and his colleagues at the Instituto do Homem e Meio Am-
biente da Amazonia, in Belem, Brazil; Jaime Acosta, Anabella
Larde de Palomo, and their associates at the Fundacidn Sal-
vadorena para el Desarrollo Economico y Social, in San Sal-
vador, El Salvador; George Carrington Wood, of Haverford,
Pennsylvania; Juan Carlos Quiroga of La Paz, Bolivia:
William Possiel (of The Nature Conservancy); Clovis Schar-
appe-Borges and Vitoria Miiller of the Sociedade de Pesquisa
em Vida Selvagem e Educagao Ambiental, in Curitiba, Brazil;
as well as Claudio Gonzalez-Vega, Brent Sohngen, and other
colleagues of mine at The Ohio State University. Two other
people at Ohio State also need to be acknowledged: Chad
Forster, my graduate assistant, spent many hours tracking
down literature and other materials, and Janice DiCarolis pre-
pared all the maps and line drawings.
Special thanks are due to Kirk Jensen and his colleagues
at Oxford University Press, for the excellent advice and sup-
port given at every stage of the writing and publication pro-
cess.
Of course, I am exclusively responsible for all the book's
errors and omissions, and the views and opinions expressed
herein are mine alone.
Finally, I am especially grateful to my wife, Myriam, and
my children, Elizabeth and Richard, for being patient as I
labored on this volume, which I dedicate to them.
Contents

Introduction: Responding to the Challenge of Habitat

Destruction in Latin America ix

Part I Deforestation and Its Causes and the Challenge of

Sustainable Forest-based Activities
1 Deforestation in the American Tropics: The Regional
and Global Stakes 3
The Magnitude of Land Use Change 5
The Impacts of Deforestation 8
2 The Causes of Excessive Habitat Destruction 11
Macroeconomic Crisis and Competition
for Space 11
The Contribution of Market Failure 13
The Role of Misguided Government Policies 17
Factors Contributing to Ecosystem Mining 21
3 Putting an End to Ecosystem Depletion 25
The Articulation of Formal Tenure 27
What Happens to Pioneer Nutrient Miners? 28
Saving Tropical Forests while Helping
the Rural Poor 30

Part II The Economic Returns of Environmentally Sound

Harvesting of Forest Products and of Nature-based
Tourism

4 Harvesting of Nontimber Products 43

The Movement to Establish Extractive Reserves 44
viii Contents

Impediments to Economically and Environmentally

Successful Extraction of Nontimber Products 45
Vegetable Ivory Production in Western Ecuador 49
Extraction of Nontimber Products and Rain Forest
Conservation 56
5 Environmentally Sound Timber Production 59
Logging in the Eastern Amazon 60
The Palcazu Forestry Project 70
Prospects for the Sustainable Development of
Tropical Timber Resources 78
6 Genetic Prospecting 83
What the Pharmaceutical Industry Might Be Willing
to Pay for Biological Raw Material 85
Controlling Access to Species-Rich Habitats 91
The Risk of Counting on Riches from
Bioprospecting 92
7 Nature-Based Tourism 95
The Ecotourism Boom in Costa Rica 98
Conservation, Tourism, and Local Interests
in the Galapagos 107
Ecotourism, Habitat Protection, and Local Economic
Development 118

Part III Key Elements of an Integrated Strategy for Habitat

Protection and Economic Progress

8 Another Approach to Habitat Conservation:

Agricultural Intensification 123
Agroforestry in Tropical Forest Settings 125
The Impacts of Sectorwide Improvements
in Productivity 127
9 Paying for Habitat Conservation and Investing in
Human and Social Capital 135
The Importance of Productivity-Enhancing
Investment 138
The Lessons To Be Learned in El Salvador about
Environmental Depletion and Conservation 143
The Inextricable Ties Binding Habitat Protection
to Economic Improvement in the Countryside 147

Abbreviations 151
References 153
Index 167
Introduction
Responding to the Challenge of Habitat
Destruction in Latin America

ropical deforestation has come to be an abiding

T international concern, and support remains strong
around the world for arresting encroachment on threatened
habitats in Latin America and other developing regions. Ex-
actly how to accomplish this task, however, is still a subject
of controversy.
Systems of national parks, like those that have been estab-
lished in wealthy nations, have proven difficult to transplant
to areas designated for special protection in Africa, Asia, and
Latin America. More often than not, these areas are already
inhabited, and local people, who tend to be quite poor, resent
being told by outsiders that they must relinquish certain eco-
nomic activities, move somewhere else, or both.
The difficulties of creating and maintaining an official pro-
tected area are illustrated by the case of Machalilla National
Park, along the Ecuadorian coast. The region has been inhab-
ited continuously for millennia and, when the reserve there
was set up in the early 1970s, local property owners were
promised payment for the land being taken from them. How-
ever, few of those individuals ever have received anything.
This would have been a simple, though regrettable, instance
of uncompensated confiscation had the government taken
effective action to safeguard its formal claims on resources.
But it did not. As of the early 1990s, only sixteen guards and
other employees were assigned to Machalilla Park, which
according to official maps takes in 467 square kilomet-
ers. Cattle grazing and fuelwood collection continue in
virtually every accessible part of the reserve. Indeed, the
argument could be made that, at least in Machalilla's case,

ix
x Introduction

public sector "protection" has accelerated environmental

degradation since resource users, many of whom formerly
had an ownership stake, now regard the area as a free, or
open-access, resource.
By the middle 1980s, the need for an alternative ap-
proach to the protection of natural habitats in the develop-
ing world was becoming obvious. Rather than trying to
suppress economic activity by local communities, those in-
volved in the implementation of integrated conservation
and development projects (ICDPs), as the new sort of initia-
tive came to be called, sought to address those communi-
ties' desires for improved standards of living and to secure
resource ownership, while at the same time meeting con-
servation objectives.
A typical ICDP has two main thrusts. The first is enhanced
park protection, conventionally understood to include
boundary demarcation, improvement of trails and facilities,
and the training and deployment of guards. The second
thrust is to enhance earnings in a surrounding buffer zone by
promoting the sustainable harvesting of forest products and,
quite often, nature-based tourism. The intention is for local
people to give up economic activities, like agricultural land
clearing, that result in substantial damage to natural re-
sources in favor of more environmentally benign alternatives.
Insofar as they do so, encroachment on parks should be di-
minished.
Although all this sounds attractive, very few ICDPs have
enjoyed much success. A review of twenty-three such pro-
jects in Africa, Asia, and Latin America, for example, re-
veals that enlisting the necessary cooperation from local
communities is often difficult (Wells and Brandon, 1993).
In addition, ICDPs have been criticized on more general
grounds. For one thing, ecotourism and other preferred ac-
tivities are not always environmentally benign. Also, the
roads and other improvements that are often needed for an
ICDP to function simultaneously enhance the profitability
of depletive lines of work. Southgate and Clark (1993)
point out that, where labor is underemployed, local popu-
lations can adopt ICDP activities without giving up what
they would do otherwise. For reasons such as these, ob-
servers like Dixon and Sherman (1990) have expressed
doubts that parks and reserves can truly be kept intact by
encouraging things like nature-based tourism or the sus-
tainable harvesting of forest products in surrounding buf-
fer zones.
Introduction xi

Outline of the Book

That ICDPs and related initiatives have not enjoyed great suc-
cess calls into question an approach to tropical forest con-
servation that has substantial prima facie appeal and has
attracted considerable moral and tangible support—namely,
the promotion of environmentally sound economic activities
in or near natural habitats that are under threat.
A good way to assess the economic viability of this ap-
proach would be to draw on assessments of ICDPs, them-
selves. Sad to say, though, this is not possible. In one of the
few contributions to the economic literature dealing with
those projects, Simpson and Sedjo (1996) complain that just
about all publicly available information can be described as
self-laudatory, because agencies responsible for administra-
tion or funding are not keen to report disappointing results,
because the projects involved do not have long track records,
or both. Even the "gray literature," consisting of internal
agency reports and evaluations, is not terribly informative,
presumably for exactly the same reasons.
This book contains an economic assessment of forest con-
servation strategies predicated on the fostering of sustainable
economic activities in and around threatened habitats. Little
reference is made here to the ICDP literature, such as it is.
Instead, activities often included in ICDPs are the subject of
direct examination. Four case-study chapters, each devoted
to a single type of activity, comprise the core of the book. The
case studies yield insights, offered in the last two chapters,
for a truly integrated strategy for ecosystem conservation and
economic progress in Latin America.
A proper point of departure for a study of what has become
a fashionable approach to the preservation of tropical forests
and other threatened habitats is a survey of the extent and
consequences of deforestation. Chapter 1 of this book con-
tains such a survey. Rates of encroachment on tree-covered
land vary, and there appears to have been a peak in the rates
in the Brazilian Amazon during the middle 1980s. However,
extensive tracts of land that have remained relatively unto-
uched up to now are in jeopardy. If there were widespread
conversion of forests into cropland and pasture land, the im-
pacts on local climate and global biodiversity would be se-
vere. Global warming could accelerate as well.
In chapter 2, various factors that contribute to land use
change are analyzed. Among these factors are population and
income growth, poorly articulated property rights, infrastruc-
xii Introduction

ture development, the tendency of deforestation agents to ne-

glect the environmental impacts of their activities, as well as
direct and indirect subsidies for land clearing. Above all, the
construction of new roads in previously inaccessible places
sets off a cycle of ecosystem depletion that is all but irresis-
tible for the actors directly involved.
The contribution that ICDPs and related initiatives might
make to habitat conservation, in the face of all this, is ad-
dressed in the chapter 3. In particular, the commercial po-
tential of extracting nontimber forest products and other
sustainable activities is examined in conceptual terms, with
special emphasis being placed on understanding the circum-
stances under which these activities actually benefit local
populations.
Chapter 4 is about the harvesting of nontimber forest prod-
ucts. Experience in the Amazon Basin indicates that there
are various impediments to that activity's economic and en-
vironmental success, including weak property rights, thin
markets, and production outside of forest settings. In addi-
tion, a study of extraction of nontimber products in western
Ecuador reveals a general tendency toward meager financial
returns for the households that engage in harvesting. By con-
trast, processors and exporters tend to capture most of what-
ever profits are generated by the exploitation of nontimber
resources.
Environmentally sound timber production is the focus of
chapter 5. Investigation of various modes of timber harvest-
ing and extraction in the eastern Amazon provides a clear
picture of how logging evolves in frontier regions, and also
yields the conclusion that sheer resource abundance dis-
courages the sort of investment required for sustainable re-
source management. The latter conclusion is corroborated by
experience gained in a sustainable forestry project carried out
in the Peruvian Amazon with financial and technical support
from the U.S. Agency for International Development (AID).
In chapter 6, there is a review of the empirical literature
on the value of tropical forests as a source of raw material for
biomedical research. Rudimentary estimates of that value,
contained in early contributions to the literature, turn out to
have been too high. The best available economic research
suggests that the economic returns of genetic prospecting
might be quite modest, particularly for forest dwellers. These
returns are almost certainly too small to justify the invest-
ment in property institutions that is required to establish ef-
ficient markets for genetic information collected in the wild.
Introduction xiii

Nature-based tourism in Costa Rica and the Galapagos Is-

lands is examined in chapter 7. Both places have drawn large
numbers of international visitors in the 1980s and 1990s, and
national economies have benefited enormously as a result.
By and large, local communities, which are poorly equipped
to provide the goods and services that ecotourists demand,
gain little from the visits being made to nearby parks and
reserves. In addition, there is a need to shore up the environ-
mental base for tourism's continued success. To accomplish
this task, entrance fees will have to be raised and other fi-
nancing mechanisms will have to be exploited.
As demonstrated in this book, ecotourism, extraction of
nontimber products, environmentally sound timber produc-
tion, and genetic prospecting can, under the right circum-
stances, contribute to habitat conservation and improved
living standards in selected areas. But in and of themselves,
these activities cannot serve as a sound centerpiece for an
integrated strategy for economic development and habitat
conservation.
Much more can probably be accomplished by raising crop
and livestock yields, so that agricultural land clearing is no
longer needed to satisfy increasing commodity demands. Of
even greater importance is human capital formation, which
reduces the number of people for whom converting natural
ecosystems into marginal farmland is an attractive employ-
ment option. Indeed, excessive encroachment on natural
ecosystems in Latin America can be interpreted as an envi-
ronmental manifestation of the more fundamental problem
of rural poverty, resulting from inadequate human capital
formation, and also from the lack of institutions—which
economists call social capital—that are needed for markets
to function robustly.
Available evidence, which is reviewed in chapters 8 and
9, suggests that a combination of sustainable agricultural in-
tensification and human and social capital investment allows
just about any country to raise material standards of living
while keeping natural habitats intact. The prospects for Latin
America's threatened ecosystems will be very bleak if accel-
erated development of the region's rural economy through
productivity-enhancing investment does not occur.
This page intentionally left blank
I
Deforestation and Its Causes and
the Challenge of Sustainable
Forest-based Activities
This page intentionally left blank
1
Deforestation in the
American Tropics

The Regional and Global Stakes

atin America's natural habitats have been receding

L for centuries. Before the 1970s, though, nearly
everyone in the region regarded the geographic expansion of
agriculture and other sectors of the rural economy as an un-
mitigated benefit. Well into the twentieth century, primary
forests and other relatively undisturbed ecosystems were ex-
tensive. Moreover, it was generally agreed that human dom-
ination of the landscape must increase as nations develop
and economies grow. Hence, deforestation was equated with
progress.
Economic assessments of widespread land use change oc-
curring through the early 1960s do not exist, mainly because
very few contemporary observers sensed that counting up
benefits and costs and analyzing causes would be worth the
trouble. Quite often, it must be said, a complacent view of
deforestation was entirely justified, as a pair of examples il-
lustrate.
First, clearing of agricultural land in the Guayas River Ba-
sin of western Ecuador began in the middle of the nineteenth
century and continued through the early 1900s, mainly to
accommodate a major expansion of cocoa production. After
a lull between the First and Second World Wars, deforesta-
tion resumed in the 1940s, when banana plantations were
first established (Bromley, 1981). With fertile soils and. abun-
dant rainfall, the watershed is where most of Ecuador's crops
and livestock are grown today.
Second, southern Brazil has experienced dramatic land
use change during the last few decades. Londrina, an urban
center in the state of Parana with a population of several hun-

3
4 Deforestation and Its Causes

dred thousand, had fewer than 20,000 inhabitants in the

1930s and was completely surrounded by forests. Coffee
farming predominated between the 1940s, when agricultural
land clearing began in earnest, and the middle 1970s, when
growers abandoned the area in the wake of a devastating
freeze. Soybeans soon became southern Brazil's major crop.
Only a few small woodlots now remain in the interior of Pa-
rana, which has become a center of oilseed production for
the country and the entire world.
Nearly all concerned, including local residents, agree that
more trees should be planted in places like the Guayas River
Basin and southern Brazil. Likewise, it is widely conceded
that soil erosion and chemical contamination must be re-
duced and that other environmental ills associated with ag-
ricultural development have to be curbed. However, no one
doubts that deforesting land that is well suited to crop and
livestock production has been, on balance, hugely beneficial.
Land use change around, say, Londrina is not much more
controversial than the migration of U.S. farmers into the Ohio
River Valley would have been in the early 1800s.
The same cannot be said of the conversion of Amazonian
forests into cropland and pasture, which a number of na-
tional governments began to encourage in the 1960s. Not-
withstanding the enthusiasm that some officials and many
settlers had for farming and ranching in the region, increased
agricultural output was never the main purpose of coloni-
zation efforts. Instead, the primary goals were to ease dem-
ographic pressure in more densely populated regions and to
solidify national territorial claims. General Emilio Garrasta-
zu Medici, who headed Brazil's military government from
November 1969 to March 1974, expressed both the objective
and the approach succinctly when he spoke of "bringing
men without land to a land without men" (Scartezini, 1985,
p. 10).
The difficulties of large-scale agricultural colonization in
Amazonia were apparent by the early 1970s. Nelson (1973)
was among the first to point out the impediments to success
in the humid tropics. Criticism was especially strong in re-
gard to an ambitious program to settle 70,000 families along
the Transamazon Highway, a road completed in 1972 that
runs south of, and approximately parallel to, the Amazon
River. Smith (1981) has found that only 3 percent of the
land originally designated for colonization is suitable for ag-
riculture. Furthermore, colonists suffered greatly from ma-
laria because deforestation created breeding grounds for the
The Regional and Global Stakes 5

Anopheles mosquito (Moran, 1983). Accordingly, the pro-

gram fell far short of planners' expectations; just 8,000 farm-
ers and their families were living along the Transamazon
Highway in 1980 (Smith, 1981).
Agricultural colonization, directed as well as spontane-
ous, has come under attack for being a threat to long-term
forest dwellers. It is significant that Fernando Belaunde-
Terry, who as Peru's president had championed settlement
of the Amazonian rainforests until being deposed in a 1968
military coup, encountered staunch opposition from indige-
nous groups and their domestic and international allies when
he tried to reinvigorate colonization projects after returning
to the presidency, in 1980 (see chapter 5). Likewise, rubber
tappers in the Brazilian Amazon have mounted a stiff resis-
tance to cattle ranchers' encroachment on tree-covered lands
(see chapter 4).
Of course, by the time that forest dwellers were starting to
have some voice in discussions about the future of the places
they live in and about the development of the resources that
surround them, the adverse environmental consequences of
land use change were starting to arouse substantial concern.
Myers (1984) offered early warnings about the species ex-
tinction brought about by the alteration of tropical habitats.
Also, deforestation appears to contribute to a rise in atmos-
pheric concentrations of greenhouse gases, like carbon di-
oxide (Dixon et al., 1994).

The Magnitude of Land Use Change

Most likely, alarm over habitat loss in Latin America and
other parts of the developing world peaked soon after 1987.
That year's dry season was unusually pronounced in many
parts of the Amazon Basin, which made it particularly easy
to burn off natural vegetation. In addition, many Brazilians
were eager to claim tree-covered land by putting it to an ag-
ricultural use because they sensed that impending constitu-
tional reform would circumscribe that age-old tenurial
convention.
With farmers and ranchers invading forests and reclearing
fields that had reverted to bush, satellite cameras were able
to detect an arc of flames extending south from Belem, a city
near the mouth of the Amazon River, and west to lowland
Bolivia. The consequences, it seemed, were global in scale
and followed quickly. The summer of 1988 turned out to be
one of the hottest that North America has ever experienced,
6 Deforestation and Its Causes

and the claim that the Earth's atmosphere is growing warmer

largely because of land use change close to the Equator was
expressed in congressional hearings, newscasts, and the
scholarly literature—and it acquired great credence. Many
are now firmly convinced that deforestation is the leading
cause of global warming.
As indicated in Table 1.1, few parts of the developing
world experienced more rapid land use change than Latin
America did during the 1980s. Expressed as a share of tree-
covered land, deforestation was especially high in Mexico
and Central America. But even in South America, which has
half the world's tropical forests, the ratio of newly cleared
land to remaining forests was comparable to what was being
observed in most of Africa and southern Asia.
Natural habitats continue to be lost at a distressing rate in
many parts of the American tropics. Myers (1988) has iden-
tified ten "hot spots" around the developing world, where
biodiversity is threatened severely. In each location, farmers,
ranchers, loggers, and miners are encroaching rapidly on for-
ests that are home to large numbers of species, many of which
are endemic to a small area. Two of the three most critical
hot spots are in Latin America. In western Ecuador and along
Brazil's Atlantic Coast, up to 90 percent of the original forest
has been either seriously degraded or lost entirely.
Clearing seemed to abate somewhat in the Amazon Basin
in the early 1990s. In Brazil, for example, deforestation was
less than a third of the maximum rate observed during the
middle 1980s, which might have been as high as 90,000
square kilometers per annum (World Resources Institute,
1990, p. 42). According to bulletins issued by the Brazilian
Institute of Geography and Statistics (IBGE), annual defor-
estation between 1992 and 1994 amounted to approximately
15,000 square kilometers.
However, land use change appears to have accelerated
since then. In 1995, the Brazilian Amazon had another pro-
nounced dry season, of the sort that occurred in 1987, and
there was a corresponding upswing in land clearing. Accord-
ing to one prominent environmental organization, annual de-
forestation there has risen 34 percent since 1992 (World
Wildlife Fund, 1996). Likewise, the World Bank's Inspection
Panel issued a report on the Planafloro Project, which is sup-
posed to conserve natural habitats in the southern portion of
the Brazilian Amazon, and the report contains an estimate of
deforestation in Rondonia (a state that borders Bolivia): 4,500
square kilometers per annum (International Bank for Recon-
The Regional and Global Stakes 7

Table 1.1 Natural Forests in 1990 and Average Deforestation

Rates for 1981 through 1990 in Selected Countries and Regions

Country/Region Area (thousand km2) Annual Loss (%)

Central Africa 2,041 0.53

Tropical southern Africa 1,459 0.84
West Africa 556 0.96
South Asia 639 0.79
Southeast Asia 2,106 1.33
Mexico 486 1.22
Central America 195 1.85
Brazil 5,611 0.61
Andean region and Paraguay 2,418 0.93
SOURCE: Data from WRI (1996): 218-219.

struction and Development, 1997). If that figure is correct,

then the pace of land clearing throughout the Amazon Basin
is much higher than what IBGE reported it to be in the early
1990s.
The threat to natural habitats will not ease any time soon.
Even in the Brazilian Amazon, where most primary forests
have remained inaccessible and therefore intact up to now,
widespread human alteration of extensive areas could be im-
minent. Uhl et al. (1997) point out that wood harvested in
the region already comprises well over half of what is con-
sumed throughout the country. Moreover, foreign interest in
Amazonian resources is growing, in part because of forest
depletion in southeast Asia. Within a few hundred kilome-
ters of roads and navigable waterways, logging is either tak-
ing place or can be expected to occur early in the next
millennium. Where timber extraction is moderate to heavy,
as it tends to be in areas close to the highway that runs from
Belern south to Brasilia, natural habitat recovery is often
slow. In addition, land made accessible by skidder tracks and
logging roads is frequently occupied by farmers and ranchers,
which further postpones the date when tree cover might be
reestablished (Uhl et al., 1997).
Uhl et al. (1997) do not project how much of the Amazon
Basin will be deforested by any particular date. But they do
provide a map indicating where timber harvesting, which is
becoming a primary initial catalyst for habitat loss, is hap-
pening or soon will be. Aside from the areas traversed by the
Belem-Brasilia Highway and other major roads, which are
8 Deforestation and Its Causes

where roundwood production is concentrated, there is a

wide swath of land south of the Amazon River where selec-
tive harvesting of mahogany (Swietenia macrophylla King)
takes place. Furthermore, Uhl et al. (1997) anticipate a major
expansion of riparian logging, both in vdrzea floodplains and
farther into the forest, throughout the watershed.
If farmers and ranchers follow loggers into many of these
areas, much of the Amazon Rain Forest will be lost.

The Impacts of Deforestation

If the conversion of forests and other natural habitats into
cropland and pasture is bound to continue in the American
tropics, then likely effects, both within the region and around
the world, need to be considered.
One impact has to do with watershed services, defined to
include reduced displacement and transport of sediments as
well as stream-flow regulation. As Hamilton and King (1983)
have stressed, the beneficial consequences of keeping a
catchment area covered with trees (or the opposite effect, the
watershed services lost because of deforestation) have been
exaggerated at times. In the same vein, Mahmood (1987)
claims that the reductions in reservoir sedimentation that re-
sult from forest conservation, tree planting, and related ac-
tivities are often overestimated because the magnitude of
"background" displacement and transport of soil and rocks,
which take place regardless of what human beings happen
to do, is not always fully appreciated.
Reviewing these and other contributions to the literature,
Chomitz and Kumari (1996) arrive at three conclusions. First,
deforestation has not been shown to be associated with large-
scale flooding. Second, the removal of tree cover does not
necessarily reduce water availability during the dry season
in downstream areas; indeed, an opposite relationship some-
times holds. Third, the linkage between upstream land use
and soil management and downstream sedimentation de-
pends greatly on the specific characteristics of the drainage
basin. In a large and relatively level basin, a long time is re-
quired for eroded soil and rock to reach a reservoir or a nav-
igable waterway, where sediments have to be dredged (at a
substantial expense), can interfere with the production of
various goods and services, or both. In smaller, more steeply
sloped watersheds, the damage is more immediate. However,
the poor performance of, say, hydroelectric projects located
in the latter sort of catchment frequently has as much to do
The Regional and Global Stakes 9

with inadequate accommodation of background sedimenta-

tion as it does with inadequate control of erosive human ac-
tivities, and it may have even more to do with the former.
Chomitz and Kumari (1996) also point out that the specific
effects of deforestation on the local climate are difficult to
determine. However, the evidence that those effects are sub-
stantial is mounting. In the Amazon Basin and other areas
where primary humid forests are extensive, transpiration
(i.e., the release of water into the air from trees and other
plants) is the source of most cloud formation and precipita-
tion (Salati and Vose, 1984). It follows, then, that removing
vegetation causes local rainfall to diminish.
The possibility exists in some places for emergence of a
mutually reenforcing cycle of deforestation and dehumidifi-
cation. As the local climate grows drier, because of reduced
tree cover, fire risks increase, due to the presence of dead
vegetative matter that no longer is too wet and is therefore
combustible. Although some trees can withstand a forest
blaze, others cannot. In addition, fire destroys many of the
seeds that lie on or in the ground, ready to sprout in the small
openings that constantly are being created in the forest can-
opy as mature trees are toppled by the elements (Uhl and
Kauffman, 1990). In other words, deforestation can lead to
drier conditions, which can result in fires, which destroy for-
ests and impede regeneration, and so on.
Climatic effects farther afield are the subject of much con-
cern and debate. As mentioned above, for at least a decade,
a causal linkage has been posited between deforestation and
global warming. But this does not mean that the issue is en-
tirely settled. For one thing, fossil fuel combustion, which
occurs mainly in affluent parts of the globe as well as in
China, Mexico, and other nations experiencing rapid indus-
trialization, accounts for the lion's share of greenhouse gas
emissions; tropical deforestation generally is reckoned to
contribute less than 15 percent of the world total (Dixon et
al., 1994).
Furthermore, it is not unanimously agreed that global
warming is truly under way, at least at the pace that some
have described. A commission of experts organized by the
United Nations has examined simulations of global climate
change obtained with the most advanced models available,
and has concluded that there is a modest warming trend that
cannot be ascribed entirely to natural causes (Houghton et
al., 1996). However, a number of scientists dissent from this
view. They point out that even the best global-climate models
10 Deforestation and Its Causes

have serious shortcomings. Skeptics also contend that ob-

served warming to date, which has been fairly modest, can-
not be attributed to the greenhouse effect (Seitx, 1996).
Regardless of how controversies over global warming and
deforestation's contributions to it are resolved, there is vir-
tually no doubt that the loss of natural habitats near the Equa-
tor is the single most important cause of declining global
biodiversity. Although they occupy little more than one-
twentieth of the world's land area, tropical forests harbor a
large share of all known plant and animal species, and many
more that are not yet named (Myers, 1984; Wilson, 1988).
Particularly in the hot spots that Myers (1988) has identified
in Africa, Asia, and Latin America, much of this biological
wealth is being extinguished at an alarming rate. This adds
up to a dramatic and irreversible change in the nature of life
on our planet.
Almost certainly, biodiversity conservation is the main
reason to protect tropical habitats.
2
The Causes of Excessive
Habitat Destruction

T here seems to be no end in sight to tropical defor-

estation, and the resulting environmental impacts
are generally viewed with grave concern. Accordingly, quite
a lot is being written about factors contributing to land use
change.
Pearce and Brown (1994), who have compiled a compre-
hensive review of the existing literature, have identified sev-
eral lines of inquiry into the causes of deforestation in the
developing world. Several investigators have examined the
impacts of the macroeconomic turmoil that beset nearly
every Latin American nation, and several countries in Africa
and Asia, during the 1980s and early 1990s. Other research
has to do with what Pearce and Brown (1994) refer to as com-
petition for space. The excessive land use change that has
resulted because farmers, ranchers, and other individual
agents do not internalize deforestation's environmental costs,
and because of inappropriate government policies, has been
studied as well.

Macroeconomic Crisis and Competition for Space

Linkages between land use conversion and developing coun-
tries' debt crisis and its aftermath are complex. On the one
hand, incentives to mine natural resources, including for-
ested ecosystems, so that debts could be serviced had to have
been strengthened. On the other hand, fiscal austerity has put
pressure on the public sector to diminish subsidies contrib-
uting to deforestation. At the same time, less money has been
available for building the roads needed to open up previously

11
12 Deforestation and Its Causes

inaccessible regions for colonization. Accordingly, it is hard-

ly surprising that some investigators are unable to discern
any statistically significant impact of indebtedness. One ex-
ception is Capistrano (1994), who has found in a cross-
national regression study that there has been a negative
relationship at times between debt-service ratios (i.e., interest
and principal payments divided by export revenues) and for-
est loss. She also reports that currency devaluation, which is
usually an essential feature of structural adjustment and
which increases the relative value of timber and other trad-
able commodities, appears to be a stimulus for deforestation.
Several statistical studies indicate that there is a posi-
tive connection between habitat destruction and population
growth, population density, and rising incomes. Each of
these three factors, Pearce and Brown (1994) point out, gives
rise to increased human occupation of environmental niches.
The space competition hypothesis, as it might be called,
does not square easily with some widely circulated views
about the causes of deforestation. In particular, the finding
that income growth and encroachment on natural habitats are
related can be seen as contradicting the claim that poverty is
an important contributing factor. The latter claim has been
repeated often. Recently, for example, the Consultative
Group for International Agricultural Research (CGIAR) re-
ported that poor farmers in developing countries clear nearly
160,000 square kilometers of tree-covered land each year
(Crossette, 1996).
Furthermore, the idea that demographic growth and rising
standards of living lead inexorably to habitat destruction is
hardly a reassuring one. Human fertility has fallen dramati-
cally in many developing countries, because of rising female
literacy, urbanization, and other factors. However, popula-
tions continue to increase since cohorts of women able to
have children are large. The space competition hypothesis,
interpreted simplistically, suggests that the only way to avoid
additional loss of natural ecosystems in the face of popula-
tion growth would be for Africans, Asians, and Latin Amer-
icans to grow poorer.
It is fortunate, then, that research bearing on the hypoth-
esis raises at least as many questions as it resolves. Consider,
for example, the fact that demographic increases in South
America lead to about five times the deforestation that results
as populations grow in Africa or Asia. In fact, an average of
0.75 hectares of forested land are lost when one person is
added to South America's population; the comparable figures
Excessive Habitat Destruction 13

for Africa and Asia are 0.16 and 0.13 hectares per person,
respectively. The reasons for the discrepancy between South
America and the rest of the developing world are not clear.
One might think that higher land-clearing coefficients for
Brazil and neighboring countries are the result of forests be-
ing converted mainly into low-density cattle pastures. But
the truth is that Asia is the place where pasture represents
the more prevalent new land use, with 92 percent of the de-
forested area being dedicated to livestock production, as op-
posed to 47 percent in South America and 29 percent in
Africa (Pearce, 1996).
Obviously, other things are going on, at least in the West-
ern Hemisphere.

The Contribution of Market Failure

Another reason why deforestation is excessive in Latin
America and other parts of the developing world is that in-
dividual agents of land use change do not consider all the
costs of their activities.
The problem is analogous to the inefficiency that would
arise if owners of steel mills were never liable for the soiled
laundry, pulmonary illnesses, and other impacts suffered be-
cause of air pollution. In such a situation, steel prices would
be too low and no market value at all would be placed on
clean air. In response to these incentives, steel production
and emissions would both be too high.
In the case of deforestation, farmers, ranchers, and other
agents of land use change regard the disutility resulting from
global warming and biodiversity loss as externalities. The
difference between the land clearing that they opt for and the
level where efficiency is achieved can be illustrated with a
very simple diagram, of the kind that can be found in any
standard text on environmental economics.
Depicted in Figure 2.1 are the marginal net private benefits
of agricultural colonization, defined as the difference be-
tween the market value of crops and livestock produced on
the last hectare cleared and the internal costs incurred on that
same hectare. The latter would include clearing expenses,
spending on fertilizer, vaccines, and other purchased inputs,
and the opportunity cost of labor and management inputs.
As a rule, marginal net benefits decline as deforestation in-
creases, because progressively more remote or less produc-
tive land is being occupied, wages rise as more labor is
employed, and so forth. Also shown in the same diagram are
14 Deforestation and Its Causes

Figure 2.1 Excessive deforestation resulting from market failure.

the marginal environmental costs of deforestation, which are

relatively modest if most land is left undisturbed, but which
increase along with the clearing rate.
For deforestation to be efficient, a proper balance needs to
be struck between net private benefits and environmental
costs. In particular, agricultural colonization should occur
only up to the level of DFE, which is where marginal net ben-
efits equal marginal environmental costs. But because farm-
ers and ranchers neglect the latter, they will opt for a higher
level—DFC, which is where their marginal net benefits equal
zero (Figure 2.1).
While the fundamental logic of the market failure expla-
nation of excessive tropical deforestation is straightforward,
quantifying the inefficiency depicted in Figure 2.1 is quite a
challenge. The main reason is that deforestation's environ-
mental costs, like most of the disutility resulting from air and
water pollution, are difficult to estimate.
For example, Swanson (1995) argues that the world's eco-
systems are becoming increasingly homogenized as an ex-
panding and more affluent human population exercises
progressively greater dominance of the landscape. At the
same time, the values of relatively undisturbed habitats are
rising because the diverse biological resources they contain
might turn out to be useful in buffering the homogenized eco-
systems (e.g., farms) on which people have come to depend
so much, following some unforeseen shock. However, there
Excessive Habitat Destruction 15

is no monetary transaction corresponding to what society, as

a whole, ought to be willing to pay for this service. Without
such compensation, a marginal environmental cost curve,
like that in Figure 2.1, cannot be plotted accurately, and
quantification of excessive destruction of biologically rich
habitats is imprecise.
There has been some empirical investigation of the values
that people attach to keeping the developing world's biodi-
verse forests intact. Pearce (1996) has reviewed a series of
debt-for-nature swaps effected in the late 1980s and early
1990s. In a typical transaction, an international conservation
organization purchased the foreign debt issued by a poor
country's government, usually at a substantial discount, and
turned the debt over to the government in exchange for a
certain amount of domestic currency to be made available for
park protection and related initiatives. Dividing funding by
the total area benefited, Pearce (1996) has found that the
financial resources generated by nearly all the swaps
amounted to $5 per hectare or less, and that the implicit per-
hectare payment for quite a few was below $1. A similar anal-
ysis of the budget for the Global Environmental Facility
(GEF), which the World Bank administers, yields implicit
payments of the same order of magnitude (Pearce, 1996).
Research carried out by Kramer, Mercer, and Sharma
(1996) suggests that the values that Pearce (1996) has calcu-
lated seriously understate what people really think the nat-
ural habitats in question are worth. Applying survey data to
a contingent valuation model, Kramer, Mercer, and Sharma
(1996) estimate that an average U.S. household would be
willing to make a onetime payment of $29 to $51 to extend
protection to an additional 5 percent of the world's tropical
forests. Assuming that all affluent households place an equal
value on the same outcome, Pearce (1996) concludes that the
citizens of all the world's rich countries would offer $11 bil-
lion to $23 billion. Dividing these amounts by the area to be
protected, which is 5 percent of approximately 20 million
square kilometers around the world, yields $110 to $230 per
hectare.
Obviously, these estimates of potential payments are
much higher than the amounts of money that have actually
changed hands through debt swaps and other arrangements.
However, it is not clear that potential payments, were they
to be made, would be large enough to cause a significant re-
duction in habitat loss. Schneider (1995) reports that real es-
tate along Brazil's agricultural frontiers is normally bought
16 Deforestation and Its Causes

and sold for $300 per hectare. Recently colonized parcels in

northwestern Ecuador fetch as much as $1,000 per hectare,
although the median value is below $500 per hectare. Since
real estate prices generally reflect the present value of the
future net returns that owners expect their holdings to gen-
erate, one doubts that, alone, biodiversity conservation pay-
ments of $110 to $230 per hectare would be enough to
dissuade many farmers and ranchers from encroaching on
tree-covered land.
An entirely different relationship appears to hold between
real estate prices and payments made to sequester carbon in
standing forests, so as to avoid changes in the world's cli-
mate. In an economic analysis of sea-level rise and other con-
sequences that could result from global warming, Fankhauser
and Pearce (1994) conclude that each ton of carbon emitted
into the atmosphere causes around $20 in damages. That
value can be applied to the emissions resulting from defor-
estation, which vary from approximately 100 tons per hec-
tare, when closed, secondary forests in the tropics are
converted to shifting cultivation or pasture, to more than 200
tons per hectare, when closed, primary tropical forests are
cleared to make way for permanent agriculture (Brown and
Pearce, 1994). The resulting cost estimates range from $2,000
to more than $4,000 tons per hectare.
Obviously, agricultural land clearing would be severely
curtailed if farmers and ranchers had to pay $2,000 to $4,000
in global-warming damages for every hectare they cleared.
Alternatively, landowners' interests in preserving tropical
forests would be greatly enhanced if a way could be found to
pay them for the carbon sequestered in undisturbed natural
vegetation.
The potential for mutually beneficial economic exchanges
between those in a position to keep carbon locked up in trop-
ical forests and parties that are inclined or obliged to do
something about global warming is intriguing. Members of
the former group would be willing participants if payments
received were at least equal to the income given up because
farming or ranching is forsworn. Along Latin America's ag-
ricultural frontiers, that opportunity cost translates into a
value of $5 to $10 per ton of sequestered carbon, which turns
out to compare favorably with the expense of diminishing
carbon emissions through other means. Payments to agents
of tropical forest conservation also would be less than the
taxes on emissions that some national governments, mainly
in Europe, have considered (Fankhauser and Pearce, 1994).
Excessive Habitat Destruction 17

It could be that an international market of sorts is devel-

oping for carbon sequestration services. In July 1996, for ex-
ample, the Costa Rican and Norwegian governments
announced that a prototypical agreement had been reached,
under which the Norwegians will make $2 million available
for reforestation and will receive, in return, bonds for the
carbon stored in the planted trees (Oficina Costarricense de
Implementacion Conjunta, 1996).
However, carbon sequestration deals continue to be a nov-
elty. One supposes that at least some participants are moti-
vated mainly by a desire for favorable publicity. A true
market for sequestration services will not emerge as long as
carbon emissions remain lightly taxed and regulated, as they
are around the world. That more serious measures to control
emissions have not yet been adopted at the national level
relates in part to doubts expressed in some influential quar-
ters about whether global warming is really a serious threat
(see chapter I). But modest taxes and regulations might also
reflect free riding. That is, some countries, hoping to benefit
from climatic stabilization without paying for it, could be
counting on other nations to raise domestic prices for fossil
fuels and to purchase forest conservation rights. Behavior of
this sort causes too little money to be raised for carbon se-
questration. As a result, converting tropical forests into crop-
land and pasture continues to be inordinately attractive for
rural people in the developing world.

The Role of Misguided Government Policies

Almost always, finding a way either to oblige people and
firms to take their activities' environmental costs into ac-
count or to reward them for supplying beneficial environ-
mental services is a considerable challenge. Building the
political consensus needed for effective environmental leg-
islation and regulations has taken many years in the United
States and other affluent nations. A particularly vexing prob-
lem in the developing world is that governments' capacities
to enforce any law, including something as basic as a traffic
code, are modest in the extreme. Under the circumstances,
many have concluded that attempting to tax or to regulate
agents of tropical deforestation would not be particularly
fruitful. A better alternative might be to focus on the reform
of existing government policies that encourage the waste and
misuse of timber resources, excessive agricultural land clear-
ing, or both.
18 Deforestation and Its Causes

The rationale for the latter approach can be described, as

Pearce and Brown (1994) have done, by slightly modifying
the standard environmental economics model of inefficien-
cies induced by noninternalized costs (Figure 2.1). The new
diagram (Figure 2.2) contains the same curve representing
marginal environmental costs. But in addition to the mar-
ginal-net-private-benefits function in Figure 2.1, there is a
new curve that indicates how those same benefits are aug-
mented by direct or indirect subsidies from the government.
As shown in Figure 2.2, deforestation is excessive for two
reasons. The first, which is represented here and in Figure
2.1 by the gap between DFE and DFC, is market failure. The
second, which economists call policy or intervention failure
on the government's part, corresponds to the horizontal dis-
tance between DFC and DFS. The latter level is opted for by
deforestation agents because that is where marginal net pri-
vate benefits, including subsidies, equal zero.
There is no question that intervention failure has an im-
pact. Repetto and Gillis (1988) contend that underpricing ac-
cess to government-owned forests has encouraged private
logging concessionaires to expand their activities beyond ec-
onomically efficient levels. Also, some of the deforestation
that has occurred in the Brazilian Amazon can be traced to
subsidies of various sorts.

Figure 2.2 Excessive deforestation resulting from intervention and

market failure.
Excessive Habitat Destruction 19

Some of the fiscal inducements contributing to deforesta-

tion in Brazil have been indirect ones. For example, Bin-
swanger (1989) has pointed out that the value which wealthy
operators of large farms attach to rural real estate is inflated
because agricultural income is taxed at a very low effective
rate and because agricultural losses (calculated using liberal
depreciation allowances and inflated estimates of operating
costs) can be credited against nonagricultural income. Also,
owning a large estate has tended to put one first in line to
receive agricultural credit from the public sector on favorable
terms. By contrast, most small operators pay little or no in-
come tax, and are therefore not very interested in agriculture
as a tax shelter. Also, their access to public sector credit is
minimal, which means that their decisions about buying land
are not influenced greatly by the prospect of getting a gov-
ernment loan that carries a low interest rate, that might be
forgiven entirely, or both.
Policy-induced distortions of the sort Binswanger (1989)
describes have contributed substantially to the concentration
of Brazilian real estate in relatively few hands. Currently, in
fact, 2.8 percent of the country's rural population owns 57
percent of the agricultural land. This in turn has precipitated
violent clashes with the Movimento Sem Terra and other
landless groups in recent years (Robinson, 1997). In addition,
many of the small farmers who find it difficult to compete in
the marketplace against wealthier individuals for prime
farmland end up relocating to less promising areas, in the
Amazon Basin, for example.
The competition for natural resources is unequal in many
other parts of Latin America. For example, Heath and Bin-
swanger (1996) blame the highly concentrated ownership of
Colombia's best agricultural land and the confinement of
small farmers to Andean hillsides and other marginal envi-
ronments on agriculture's preferential tax treatment and
policy-induced distortions in the financial sector.
Brazil is more unique in that, for a little more than two
decades, the government made subsidies available for large-
scale cattle ranching in the Amazon Basin. The system was
put in place in 1966, when livestock projects approved by
the Superintendency for Amazonian Development (SUDAM)
first became eligible for tax credits. From 1971 through 1987,
credits of nearly $3 billion (in constant 1990 dollars) were
disbursed to the livestock sector, which also received almost
$2 billion in loans carrying negative real interest rates
(Schneider, 1995).
20 Deforestation and Its Causes

No doubt, subsidies greatly enhanced the financial attrac-

tion of livestock production in the Amazon for quite a few
firms and individuals. Citing one enterprise-level study
(Hecht, Norgaard, and Possio, n.d.), Mahar (1989) concludes
that the typical SUDAM-approved project, which covered
more than 20,000 hectares, would not have been profitable
had tax credits not been available. Similarly, Browder (1988)
has found that subsidies amounted to a large part of the ac-
counting profits of many large cattle ranches.
However, it is important not to exaggerate the effects of
direct fiscal incentives. Out of 84,000 square kilometers ac-
counted for by all subsidized projects, 70,000 square kilo-
meters were in just two states in the eastern Amazon, Para
and Mato Grosso. Of those 70,000 square kilometers, 20,000
square kilometers had been converted to pasture as of the
middle 1980s. Since 156,000 square kilometers had been
cleared in Para and Mato Grosso during the preceding de-
cade, no more than 13 percent of total deforestation in the
two states took place on land owned by a person or firm re-
ceiving a subsidy (Schneider, 1995). If anything, the habitat
loss caused by fiscal inducements accounted for less than 13
percent of the total since some of the 20,000 square kilome-
ters would have been cleared even if no subsidies had been
available.
Schneider (1995) offers another perspective on the limited
impacts of direct subsidies, which the Brazilian government
had phased out by 1990. He cites agricultural census data
which show that, while the SUDAM program was still in ef-
fect, small- and medium-size ranchers were increasing their
herds more rapidly than large operators, who were the pro-
gram's main beneficiaries, were doing. Between 1980 and
1985, the number of cattle in Amazonian herds that had less
than 50 head grew by 70 percent; for herds with 50 to 500
head, overall growth during the same period was 38 percent.
Meanwhile, the number of cattle in herds larger than 500
head was just 17 percent higher for the region as a whole.
Particularly interesting was the case of Rondonia, where the
total number of cattle in herds larger than 200 head con-
tracted by 2 percent between 1980 and 1985 and the total
number in smaller herds rose by 80 percent during the same
period, with practically no help from SUDAM.
Brazil is not the only place where direct subsidies' true
contribution to agricultural land clearing is less than what
some have believed. For example, Ledec (1992) has found
that cheap credit for land clearing caused only 7-to-10 per-
Excessive Habitat Destruction 21

cent of cumulative deforestation in Panama through the

1980s. That share is considerably lower than the casual es-
timates which have been circulated at one time or another,
and continue to be cited.

Factors Contributing to Ecosystem Mining

In terms of yielding useful answers to the question of what
can be done to arrest encroachment on Latin America's nat-
ural habitats, much of the literature on the causes of tropical
deforestation seems to lead to a dead end. Simply observing
that either foreign debt, population growth, or poverty (or
improved standards of living) makes a contribution yields
few practical insights. Market failure certainly plays a role.
But mechanisms for effecting international transfers to pay
for carbon sequestration, biodiversity conservation, and
other environmental services that tropical forests provide are
at a nascent stage of development and are likely to remain so
for many years. In addition, crop and livestock production
in the Amazon Basin appears not to result primarily from the
sort of financial inducements that were the subject of stern
censure during the late 1980s.
Research that contributes most to an understanding of the
factors driving deforestation, and which reveals most clearly
what can be done to decelerate land use change, focuses on
what happens when transportation infrastructure is devel-
oped in previously inaccessible areas. Schneider (1995) re-
fers to this as road extensification, as opposed to road
intensification. The latter involves making improvements in
an accessible area's transportation network.
Needless to say, an important consequence of extended
road networks is that it increases the availability of relatively
cheap land in the rural economy. One response to this is ac-
celerated migration from areas where real estate is relatively
expensive to newly opened frontiers. For example, land
prices in southern Brazil increased dramatically after the
early 1970s, as a switch was being made in the region to large-
scale, mechanized soybean farming. (Preferential tax treat-
ment of agricultural income, credit subsidies, and other
policies analyzed by Binswanger [1989] promoted the tran-
sition.) At the same time, new highways were being con-
structed in Rondonia and other parts of the Amazon. As the
ratio of real estate values in Parana and neighboring states to
Amazonian land prices rose from two-to-one to ten-to-one or
higher, small farmers in the former region found it worth
22 Deforestation and Its Causes

their while to sell their holdings and to relocate to the fron-

tier. Tens of thousands chose to do exactly that (Schneider,
1992).
A very similar situation arose in Guatemala during the
1970s. Due to improved economic prospects for sugar pro-
duction, real estate prices were bid upward in the southern
coastal lowlands. This in turn prompted ranchers to move
their operations to areas in the northern part of the country
that were just then becoming accessible (International Bank
for Reconstruction and Development [IRBD], 1978, cited in
Kaimowitz, 1996).
Once in a frontier setting, many migrants find it worth
their while to engage in real estate speculation. For example,
Colchester and Lohmann (1993) report that land price in-
creases in northern Guatemala have consistently outstripped
inflation since the 1970s. For at least part of this period, this
trend reflected a long-term rise in international beef prices,
which is important since Central America exported meat in
significant quantities from the 1960s through the early 1980s
and since most deforested land in the region has been dedi-
cated to livestock production (Kaimowitz, 1996). In addition,
real estate speculation is profitable where road networks are
being put in place. For example, Ledec (1992) reports that
infrastructure development in a part of Panama's Los Santos
province caused local land prices to rise from $50 per hectare
to $100.
Something else that almost always takes place in frontier
regions, where renewable resources are relatively abundant,
is "nutrient mining," as Schneider (1995) calls it. This pro-
cess often commences with logging, which makes land more
accessible in addition to yielding timber. Since previously
uncultivated soils are relatively fertile, especially after veg-
etation has been burned, and free of pests, new colonists tend
to spend a few years raising crops. But as nutrients are de-
pleted and pests become more numerous, crop yields fall. At
some point, it makes sense to take up extensive cattle ranch-
ing. If livestock yields fall below a certain level, the colonized
parcel might be abandoned entirely.
Very important though it is, road extensification is by no
means the only cause of ecosystem depletion along Latin
America's agricultural frontiers. Most agents of depletion are
poor people who have little or no access to formal financial
markets; any credit they receive from informal sources tends
to carry very high real interest rates (Schneider, 1995). At
Excessive Habitat Destruction 23

these high rates, depleting environmental wealth is much

more remunerative than investing in land improvements is.
In addition, frontier tenurial arrangements are consistent
with, and often promote, nutrient mining. Where there are
no formal property rights, informal agricultural use rights are
the norm. Furthermore, the latter regime was for many years
codified in colonization laws, which were enforced by agrar-
ian reform agencies. For example, Macdonald (1981) reports
that, during the 1970s, the indigenous community of Pasu
Urcu, in the Ecuadorian Amazon, abandoned swidden agri-
culture, which involves managing the successive growth of
a wide variety of useful plants as well as periodic fallowing,
and converted forests to pasture. They did this because gov-
ernment officials informed them that, if the change were not
made, their land would be declared "idle" and therefore
open to occupation by agricultural colonists who were 50
kilometers away at the time.
Settlers moving in large numbers to Latin American rain
forests have felt no less obliged to clear land than Amerin-
dians had, in order to safeguard formal or informal tenure.
As Mahar (1989) points out, the "improvements" that a Bra-
zilian colonist must make on his or her holding in order for
the state to recognize his or her property rights have long
been equated with deforestation.
It cannot be denied that ecosystem depletion involves a
substantial drawing down of environmental wealth. This
does not mean, though, that converting forests into cropland
and pasture is inherently inviable. It would not make much
sense, for example, to spend $300 per hectare on pasture re-
generation that would lead to a tripling of net returns (not
counting investment expenditures) if the same outcome
could be obtained by purchasing more land at a per-hectare
price of $150 or less. Instead, allowing yields to fall and then
relocating to a new site is an economically rational response
to land abundance: "With accessible land sufficiently cheap,
it is more profitable to move the farm to the nutrient-rich,
pest-free environment, than to import the fertilizers and pes-
ticides to the farm" (Schneider, 1995, p. 16).
The cycle of depletion, abandonment, and relocation is
given new life every time a new road is constructed through
a virgin territory. It ceases only after all accessible land has
been colonized, by which point real estate prices are rising,
in response to increased resource scarcity, and productivity-
enhancing investment is starting to be financially attractive.
This page intentionally left blank
3
Putting an End to
Ecosystem Depletion

r) eginning in the 1960s, infrastructure development

D created myriad opportunities to exploit environ-
mental wealth in the American tropics. For example, there
were only 300 kilometers of paved roads in the Brazilian Am-
azon in 1960 (Mahar, 1989). In 1964, a federal highway tra-
versing the eastern fringes of the watershed was completed
to link Brasilia and Belem. During the next dozen years, as
many as 320,000 people migrated to the road's zone of influ-
ence (Katzman, 1977), where conditions for crop and live-
stock production are, for the most part, better than conditions
found in the watershed's interior.
Similar events have unfolded in southern and western
Amazonia and in other parts of the Western Hemisphere. The
population of Rondonia, which has a larger area than many
European nations have, was just 70,000 in 1960, and the out-
side world could be reached from there only by boat or by a
small airplane. Since then, a federal highway and a number
of feeder roads have been built, tens of thousands of migrants
have arrived, and statehood has been achieved. Cumulative
deforestation was approaching 25 percent in the late 1980s,
and has surpassed that level since then (Mahar, 1989). North-
eastern Ecuador was an inaccessible backwater until the late
1960s, when commercial petroleum deposits were discov-
ered there. The infrastructure development that ensued has
led since then to migration and agricultural land clearing on
a large scale (Bromley, 1981). Likewise, the extensive defor-
estation that Central America has experienced has much to
do with the construction of all-weather roads in the region;

25
26 Deforestation and Its Causes

overall highway length increased from 8,350 kilometers in

1953 to 26,700 kilometers in 1978 (Williams, 1986).
The pace of road extensification in Latin America dimin-
ished considerably after the middle 1980s. Fiscal austerity,
brought on by the debt crisis, constrained investment in pub-
lic works. Also, the World Bank and the Inter-American De-
velopment Bank (IDE) grew reluctant to support projects that
might harm the environment, in large part because of pres-
sure brought by supporters of indigenous rights and by in-
ternational conservation organizations.
Private banks are assuming a much larger role in the fi-
nancing of infrastructure development in Latin America. The
corresponding decline in the importance of multilateral lend-
ing institutions is a consequence of the length of time they
usually take to design a project, to carry out social and en-
vironmental reviews, and to win approval from boards of di-
rectors and host governments. Regardless, it is entirely
possible that roads and other high-impact projects will be
built without any multilateral bank involvement or environ-
mental review whatsoever.
A case in point is the Pangue Dam, on Chile's Bio-Bio
River, which was originally supported by the International
Finance Corporation (IFC). According to press accounts, a
confidential report commissioned by the president of the
World Bank, with which the IFC is affiliated, was critical
about the lack of enforcement of environmental conditions.
When an attempt was made to rectify this, Chile's dam-
building agency responded by paying off the IFC loan in full
and borrowing from Dresdner Bank. Not only did the private
source impose fewer environmental restrictions, but it
charged a lower interest rate as well (Economist, 1997a).
Those who lobby in Washington, D.C., where the World
Bank and the IDB are located, should keep this experience,
which is hardly unique, in mind as they decide how best to
attach environmental conditions to infrastructure develop-
ment in Latin America. One thing that is perhaps working in
their favor where highway construction is concerned is that
many of the areas which remain inaccessible do not have the
sort of agricultural potential needed to justify major expen-
ditures on highways and related improvements. Investing in
such areas should hardly be appealing to a bank or any other
institution interested in maximizing profits. For this reason,
road extensification is not expected to accelerate dramati-
cally in most Latin American hinterlands, at least for the time
being.
Putting an End to Ecosystem Depletion 27

The Articulation of Formal Land Tenure

As frontiers close, the attractions of land improvement grow
relative to those of ecosystem mining. This, in turn, causes
property rights to evolve where colonization already has
taken place. That is, increased resource scarcity induces the
development of formal tenurial institutions, in which gov-
ernment assumes responsibility for protecting landowners
from trespassers. In more remote areas, where formal prop-
erty rights do not yet exist, landowners must fend for them-
selves, sometimes threatening interlopers with violence.
Progress toward stronger property rights is not always
steady. In particular, people still draw on traditions of agri-
cultural use rights to justify their attempts to acquire land for
free, even after changes have been made in agrarian reform
legislation and other legal arrangements that justify this sort
of action. In August 1996, for example, there was a take-over
of several hundred hectares in northwestern Ecuador owned
by the country's leading wood products firm, which with full
governmental approval and encouragement has been apply-
ing various measures to limit environmental damages and to
enhance future timber production at the site. It is safe to as-
sume that the participants felt that they could take advantage
of the institutional flux that normally accompanies a change
in the national government, of the sort taking place at the
time.
Incidents such as these, and there have been many, dem-
onstrate.that the rule of law is not as strong as it should be
in the Latin American countryside. Nonagricultural uses of
land, like forestry, are greatly discouraged where agricultural
use rights are the sine qua non of traditional tenurial arrange-
ments. Even intensified crop or livestock production is dis-
couraged since the effort devoted to raising productivity
could instead be used to acquire more land, by clearing it of
course.
Nevertheless, the more general trend is toward a property
rights regime favoring investment, which Alston, Libecap,
and Schneider (1996) have documented for the case of Para.
Furthermore, investment is indeed starting to occur where
formal property institutions have been established. Recent
trends in the livestock industry in a part of the state southeast
of Belem, where livestock production got under way in the
1960s, are illustrative in this regard.
Mattos and Uhl (1994) have found that most pastures in
the area are not being allowed to deteriorate over time as a
28 Deforestation and Its Causes

prelude to abandonment and relocation. Instead, medium-

and large-scale cattle producers have been rehabilitating
their holdings, spending $260 per hectare, on average, on
plowing, fertilizing, and reseeding; these measures raise an-
nual live-weight production from 45-to-65 kilograms per hec-
tare to 150-to-250 kilograms per hectare. Given the prices
prevailing around 1990, the return on investment is 13 per-
cent to 14 percent per annum. On a per-hectare basis, spend-
ing on pasture regeneration by smaller operators, who tend
to specialize in calf and dairy production, is slightly lower;
but the annual return on their investment is a little higher,
16 percent.

What Happens to Pioneer Nutrient Miners?

Cattle ranchers, many of whom operate on a large scale, have
not been the only beneficiaries of agricultural settlement in
the American tropics. A study of 440 land settlement projects
undertaken by Brazil's National Institute for Colonization
and Agrarian Reform (INCRA) in every part of the country
reveals that the Amazonian projects have enjoyed some eco-
nomic success. To be sure, the participating households have
endured hardships, as indicated by high rates of infant mor-
tality; in addition, they move from place to place more often
than do farmers in the rest of Brazil (FAO/UNDP/MARA,
1992, cited in Schneider, 1995). But at the same time, the
financial rewards are, for them, relatively good. Learning
about crop and livestock production in a new environment
can, and often does, require a lot of time and trial and error
(Moran, 1989). However, the beneficiaries of INCRA's Ama-
zonian program have enjoyed higher incomes and rates of
asset accumulation than have their counterparts in all other
parts of the country, with the exception of the prosperous
southern states (FAO/UNDP/MARA, 1992, cited in Schnei-
der, 1995, p. 12).
It is of paramount importance, though, to keep in mind an
observation of Moran's (1989), which is that the people who
thrive once property rights have grown stronger and human
capital, communications infrastructure, and improved trans-
portation networks have all begun to accumulate are usually
quite different from those who initially colonize the frontier.
To be specific, the beneficiaries of institutional and economic
development tend to be better educated and often live in ur-
ban areas; they also have reasonably good access to govern-
mental agencies and formal financial markets. By contrast,
Putting an End to Ecosystem Depletion 29

pioneer nutrient miners are poorly educated and they are apt
to find that dealing with an agency, bank, or any other insti-
tution is a frustrating experience.
Where formal property rights have not been established,
the advantages of people who are more affluent and well con-
nected count for little. Also, the opportunity costs of the time
they devote to defending a claim in a remote region, where
the government's presence is negligible, are not covered by
that activity's returns. However, those same returns are suf-
ficiently attractive to colonists, whose time carries a low op-
portunity cost.
The tables turn once government begins to guarantee land
tenure. Landowners are no longer obliged to occupy their
holdings at all times so as to keep trespassers in check. In
addition, the articulation of property arrangements makes it
possible for formal financial institutions to function. When
this happens, people with good access to those institutions—
which lend money at interest rates substantially below what
informal lenders charge—can borrow money to buy real es-
tate, quite often from the original occupants. Wealthier peo-
ple with better connections have other advantages. Their
management skills tend to be superior. Furthermore, they
find it relatively easy to arrange for government to provide
them with land protection services.
Moran (1989) has documented that institutional and eco-
nomic development in a frontier region is accompanied by a
transition from one type of owner to another. Also, the survey
of INCRA land settlement projects reveals that turnover of
holdings has been high in the Amazon Basin (FAO/UNDP/
MARA, 1992, cited in Schneider, 1995, p. 12), which is con-
sistent with the transition in land ownership.
After selling out or otherwise giving ground in areas un-
dergoing development, pioneer nutrient miners' choices are
very limited. Indeed, those choices are not appreciably dif-
ferent from those facing the rural poor throughout Latin
America. If they do not migrate, their rudimentary skills per-
mit them to hold only menial, low-paying jobs. Seasonal farm
labor and casual employment in the wood products sector
would be two examples. Another choice, which does not ex-
clude the first, is to support oneself and one's family as best
as one can on a small holding, located perhaps in an area
where property rights remain weak or nonexistent.
Another option is to abandon the countryside, and then to
crowd into the urban slums where hundreds of thousands of
people already live in Belem, Manaus, and a number of me-
30 Deforestation and Its Causes

dium-size cities. As Browder and Godfrey (1997) indicate,

this is the most frequently exercised option, which explains
why the urban population of the Brazilian Amazon
amounted to nearly 61 percent of the region's total in 1990
and continues to grow.

Saving Tropical Forests while Helping the

Rural Poor
If institutional and economic development did not displace
pioneer nutrient miners, the fate of Latin America's natural
habitats would be determined in a relatively straightforward
manner by economic incentives. That is, individuals with
formal property rights could be counted on to behave no dif-
ferently than, say, their counterparts in rural areas of the
United States. With the state guaranteeing transferable and
exclusive rights for resources, no discount would be applied
to forestry earnings, reflecting dispossession risks associated
with nonagricultural uses of land. Instead, landowners'
choices would proceed from a direct comparison of future
discounted net returns for various land use options. If, for
example, the present value of earnings from forestry ex-
ceeded the present value of ranching income, then land
would be dedicated to timber production. Likewise, a hike
in real interest rates would lessen the relative appeal of land
use options, like forestry, that yield revenues only after many
years have passed.
Furthermore, manipulating land use in places where for-
mal property rights are respected would involve little more
than the imposition or withdrawal of taxes or subsidies. For
example, more tree-covered land would be maintained if car-
bon sequestration payments were offered and if landowners
found that the discounted sum of those payments and for-
estry income exceeded the present value of ranching income.
Another way to arrest deforestation would be to remove
whatever subsidies may exist for crop or livestock produc-
tion.
However, the displacement of pioneer colonists that re-
sults from institutional and economic development means
that habitat protection requires much more than just getting
property arrangements and prices "right." Many, if not most,
poor people from the countryside who do not move to a major
city can be counted among what might be called a reserve
population of potential nutrient miners, which stands ready
to take advantage of any opportunity to engage in ecosystem
Putting an End to Ecosystem Depletion 31

depletion resulting from renewed road extensification, from

appreciating values of open-access forest resources, or both.
It is widely accepted that habitats will never be safe as long
as the rural poor are neglected. Indeed, there is a general
sense that they must be full participants in the transition
from nutrient mining to ecosystem management that must
take place if natural habitats are to remain intact in places
like the Amazon Basin. This sentiment is reflected in policies
adopted recently by multilateral lending institutions. For ex-
ample, the World Bank has stated that it will support for-
estry only if it is sustainable and only where there has been
a "clear definition of the roles and rights o f . . . forest dwell-
ers" (IBRD, 1991, p. 66). Likewise, the IDE has committed
itself to taking advantage of "opportunities to aid in the con-
servation of biological diversity" while also making sure that
local communities share in the "benefits of sustainable forest
management" (IDB, 1994, p. 34).
Nongovernmental environmental organizations also rec-
ognize the need to reconcile habitat protection with the en-
hanced well-being of local people. One such organization,
which is active throughout the Western Hemisphere, is The
Nature Conservancy (TNC). Its president, John Sawhill, has
stated that he and his colleagues are "concentrating more on
strategies that address . . . the conservation issue of the
1990s: integrating economic growth with environmental pro-
tection" (Howard and Magretta, 1995, p. 111).
There are various ways to protect the environment while
at the same time helping impoverished local communities.
One approach, which is examined more closely in chapter 9
of this book, is to improve education and public health serv-
ices so that individuals who otherwise would engage in nu-
trient mining can instead compete successfully for jobs that
pay better and at the same time create less resource depletion.
Human capital formation has been a feature of many of the
habitat conservation projects that development agencies,
multilateral lending institutions, and international environ-
mental organizations have financed and/or implemented.
But at least as much emphasis has been placed on encour-
aging individuals and groups living in or near threatened
habitats to take up environmentally benign economic activ-
ities. These include the harvesting of nontimber products,
logging in ways that do little damage to soils and remaining
vegetation, collecting biological materials used in pharma-
ceutical and other research, and nature-based tourism.
Each of these four activities is analyzed in the next four
chapters. Except under special circumstances, none turns out
32 Deforestation and Its Causes

to be especially remunerative for local communities. One rea-

son for this is that opting for something like sustainable for-
estry is unlikely to lead to an appreciable increase in the
wages received by unskilled local labor. Another reason is
that just about all the natural resources that environmentally
benign economic activities draw on are abundant. As a rule,
timber and other forest resources are not economically
scarce. Likewise, there is no general shortage of ecotourism
sites, although it is true enough that a few travel destinations,
like the Galapagos Islands, are highly unique and desirable
places. Except where a local community is fortunate enough
to control access to an unusually valuable resource, local
well-being will not be affected very much by the promotion
of sustainable alternatives to ecosystem mining.
The challenge to a strategy for habitat conservation and
local development that is predicated on the sustainable ex-
ploitation of renewable forest resources can be put in prop-
er perspective by applying a conceptual framework origin-
ally developed by von Thiinen (1866). The exceptional case
where the strategy makes a great deal of economic sense
will be considered first. After that, the more typical case, in
which sustainable forestry is not very appealing, can be ex-
amined.

Opting for Sustainable Forestry

Close to a Marketing Center
Even though, as already mentioned, aggregate resource stocks
are more than ample, individual holdings can still be valu-
able because they are unusually productive. This possibility
is not examined here. Instead, the focus is on another deter-
minant of scarcity, which has to do with location. In partic-
ular, it can be shown that forests located close to a market
can have value if transporting the products they yield is ex-
pensive and if no competing land use is more profitable.
If moving a given quantity of output a certain distance is
indeed expensive, locational values are attached to advan-
tageously situated resources. These values can be illustrated
in the context of a very simple rain forest economy, one with
a single commodity produced at a constant average cost, C.
With the expense of transporting a unit of output one kilo-
meter being defined as T, marginal cost, MC, is a simple linear
function of distance from the market, D:
MC = C + Tx D. (3.1)
Putting an End to Ecosystem Depletion '.

Spatial equilibrium in the rain forest economy with one

good is depicted in Figure 3.1 A. At the extensive margin, D*,
the price that clears the market for the jungle commodity, P*,
equals per-unit production costs plus the expense of deliv-
ering output to buyers, nothing more and nothing less:
P* = C + T x D*. (3.2)
Beyond the extensive margin, combined costs exceed P*,
which means that production will not take place. But at in-
framarginal locations (i.e., those within D*), sales revenues
exceed the sum of production and transportation expenses.
In other words, a rental premium, comprising the difference
between the market-clearing price, P*, and C + T X D, is
received for every unit of output.
If T is sizable, which would be the case if the jungle com-
modity were perishable, bulky, or both, then the MC curve

Figure 3.1 Spatial equilibrium. (A) The single-product case, with

high per-unit transportation costs; (B) Land use succession for the
two-sector case.
34 Deforestation and Its Causes

will be steeply sloped, as illustrated in Figure 3.1 A. Produc-

ers located close to the market will collect sizable per-unit
rents. Moreover, those rents will be directly manifested in
resource values, since prices of inframarginal real estate will
be bid upward in direct proportion to the gap between P*
and the costs of production and transportation. For example,
what someone would pay for one year's use of an inframar-
ginal hectare of forest yielding Q units of output would be:
yearly land rents = (P* - C - TD) Q. (3.3)
With T being large, resource value declines precipitously as
one moves away from the market. Yearly land rents at the
extensive margin are nil. Beyond D*, resources would be oc-
cupied only if a subsidy were offered, to cover the negative
difference between P* and C + TD.
To complete the analysis of forests' locational values (or,
to be more precise, the economic circumstances under which
property owners decide to maintain tree cover), one must
reckon with the possibility of competing land uses. In partic-
ular, a series of land uses, not just the declining value of a
single use, observed at progressively less advantageous lo-
cations must be identified. For simplicity's sake, suppose that
ranching, which requires the clearing away of natural vege-
tation, is the only alternative to harvesting the jungle com-
modity. We are interested, then, in determining the distance
from the market at which there is a succession in land use
from forestry to livestock production.
To find the succession point, D, a rent triangle, which in-
dicates yearly resource values (defined in equation 3.3 for the
case of forestry) at varying distances from the market, must
be plotted out for each of the two sectors. One corner of the
triangle for forestry is found at D* on the horizontal axis (see
Figure 3.IB). The height of the same triangle corresponds to
the rents captured by the most centrally located producer,
who, by definition, incurs no transportation costs:
(P* - C) Q. (3.4)
A rent triangle for ranching is plotted as well in Figure 3.1B.
Its gentler slope is a signal that per-unit transportation costs
in the livestock sector are lower than what they are in the
market for the jungle commodity.
By definition, ranching and forestry are equally profitable
at the succession point. Beyond D, livestock production rep-
resents the dominant land use, since it yields higher resource
rents than forestry does. At lesser distances, property owners
Putting an End to Ecosystem Depletion 35

opt for production of the jungle commodity, and forests are

left standing. (They also remain intact beyond ranching's ex-
tensive margin, which occurs where the livestock sector's
rent triangle intersects the horizontal axis.)
Since time and effort are needed to delineate property
rights, formal tenure will be established first for inframar-
ginal land dedicated to uses that have a relatively high value.
In the two-sector model, with CPR standing for the annualized
per-hectare cost of identifying and enforcing property rights,
the frontier of institutional development occurs DPR kilome-
ters from the market. Beyond that frontier, claims on re-
sources, which are worth less than CPR, are not guaranteed
by the state. At less remote locations, property rights are re-
spected, regardless of whether land is used to raise cattle or
whether forests are conserved, because annual returns to
land exceed CPR.

Forestry's Locational Values When

Transportation Costs Are Low
It has to be repeated at this point that Figure 3.1's depiction
of locational values and land use succession is based on the
assumption that moving forest commodities from place to
place is expensive, which implies that the MC curve in Fig-
ure 3.1 A and the hypotenuse of the forestry rent triangle in
Figure 3.1B are both steeply sloped.
To be sure, transportation costs for some commodities ex-
tracted from Latin America's tropical forests are high. A few
jungle fruits, like the acai (Euterpe olemcea Mart.) harvested
near Belem (see Chapter 4), appear to be cases in point. But
most forest commodities are no more difficult to move
around than other goods are, and quite a few are relatively
easy to transport. Neither timber nor latex, for example, is as
perishable or bulky as most crops are. The implication of this
is that hypotenuses of rent triangles for various parts of the
forestry sector have much gentler slopes than what is de-
picted in Figure 3.I.B.
In addition, many forestry rent triangles are of modest
height. This frequently occurs because the prices of close
substitutes for forest commodities (e.g., timber or latex har-
vested on plantations) are low, which prevents market prices
for those forest commodities from rising very high. Also, pro-
duction costs in forest settings are often high, because of lim-
ited infrastructure, difficult terrain, and so forth. Something
else that keeps forestry rents in check is the mobility of other
36 Deforestation and Its Causes

factors of production. For example, small mills for cutting

logs into boards are not very difficult to move from one site
to another. If stumpage values (i.e., the prices paid for stand-
ing timber) rise in one place, there is not much to stop those
mills from being shifted to locations where raw materials re-
main cheap. For all these reasons, forestry tends not to be a
land use characterized by sizable locational rents.
What all this means is that, different from the situation
depicted in Figure 3.IB, tree cover is more likely to be main-
tained outside, rather than within, the frontier of property
rights development. Indeed, the succession from agricultural
land use to forestry often occurs far beyond that frontier. As
illustrated in Figure 3.2, the locational rents captured by nu-
trient miners exceed the locational value of forestry for the
considerable distance lying between DPR and the succession
point, D. This implies that harvesting of forest products is
confined to the most remote reaches of a rural economy.

Returns for Forest Dwellers' Labor

Where nature is bounteous, or transport costs are low, or fac-
tors of production other than environmental inputs are mo-
bile, or there is some combination of the three, forest dwellers
do not stand to gain very much by controlling access to the
natural resources that surround them. It is not inconceivable,
though, for those people to receive high incomes in the for-
estry sector by being paid a lot for their work and expertise.
Of course, wages and salaries in any labor market are de-
termined by the interplay of demand, which is a function of
workers' productivity and the price of whatever is being pro-
duced, and labor availability. Where output prices are low,

Figure 3.2 The succession from nutrient mining to forestry beyond

the frontier of property rights development.
Putting an End to Ecosystem Depletion 37

high earnings can occur only if productivity is high and the

supply of labor is inelastic.
All available evidence suggests that the productivity of la-
bor employed in logging, extraction of nontimber products,
and other resource-based enterprises in Latin America's trop-
ical forests is low. Biodiversity, itself, can be a cause of mod-
est returns for labor. In forests featuring a great variety of
species, for example, it is usually the case that useful organ-
isms are widely dispersed, which implies that workers spend
much of their time looking for something, as opposed to ex-
tracting it.
At the same time, the supply of labor tends to be elastic in
places like the Amazon Basin, mainly because workers' mo-
bility across regions and economic sectors is high. Romanoff
(1981, cited in Browder, 1992b) has found, for example, that
it is common for Amazonian rubber-tapper households to mi-
grate every few years in the hope of gaining some sort of im-
provement in their meager earnings. Even when relocation
does not occur, rural labor can move into and out of the cash
economy (i.e., out of and into subsistence farming, hunting,
and gathering) with ease.
When labor supply is highly elastic, wages cannot be ex-
pected to rise much above subsistence levels, even if major
productivity increases occur in logging, extraction of nontim-
ber products, and other forest-based occupations. Nor is the
demand growth resulting from the opening up of new lines
of work, in ecotourism or genetic prospecting, for example,
likely to lead to substantially higher earnings for those living
in or close to tropical forests.
The economic agents who are in the best position to ben-
efit if launching or expanding an enterprise in a remote lo-
cation proves to be remunerative are those furnishing factors
of production that have low, as opposed to high, supply elas-
ticities. Management, organizational, and marketing exper-
tise is much more apt to fall into this category than are natural
resources and unskilled labor. In addition, the firms that pro-
vide this expertise often face limited local competition, and
sometimes none at all. No doubt, this enhances their ability
to capture whatever profits are generated by activities like
nature-based tourism and genetic prospecting.

Distributional Concerns and

Human Capital Formation
By no means would a finding that local people gain little or
nothing from an activity yielding positive environmental
38 Deforestation and Its Causes

spillovers, like diminished pressure on natural habitats,

mean that the activity should not be pursued. If the activity's
benefits, including what people in other parts of the country
or the world are willing to pay for spillovers, exceed its costs,
then it satisfies the standard criterion of economic efficiency,
and is worth pursuing.
Even if implementation of an efficient project benefits
mainly people who are relatively affluent, incorporating re-
distributional measures into the project is not warranted, if
mechanisms, like a progressive income tax, are already in
place to reduce disparities between the rich and the poor.
Transfers to impoverished local communities are an im-
portant component of project design, though, if they are
needed to win the behavioral changes required to secure en-
vironmental benefits. For example, providing tree seedlings
free of charge might be a good way to accelerate the refores-
tation of upper watersheds. Likewise, furnishing financial
and technical assistance for non-timber products extraction,
genetic prospecting, and so forth might diminish encroach-
ment on rain forests and other species-rich tropical habitats.
It hardly needs to be said that the opportunity costs of
effort dedicated to saving threatened ecosystems in the de-
veloping world need to be given close attention. This is es-
pecially true when the central thrust of a project is to build
up local human capital that is specifically tailored to envi-
ronmentally sustainable commercial activities. For example,
investing in the specific skills that forest dwellers need in
order to run sustainable forestry ventures is questionable
when and where private enterprises show little interest in
managing stands of trees so as to enhance future yields, es-
tablishing market contacts, and related pursuits. Even if the
private sector has exhibited interest, preparing local popu-
lations to take on the latter tasks might not make economic
sense.
One criticism of sector-specific human capital formation
has been offered by Simpson and Sedjo (1996). In particular,
they contend that a much larger area could be protected if
the money used to subsidize specific activities like extraction
of nontimber products were instead used simply to pay forest
dwellers to keep natural habitats intact.
Just as there might be better ways to use scarce funds and
technical assistance to save endangered ecosystems, sector-
specific training may not be the most effective measure for
raising forest dwellers' standards of living. By and large,
those people derive more benefit from education that is more
Putting an End to Ecosystem Depletion 39

generally applicable, and which prepares them to take ad-

vantage of whatever well-paying employment opportunities
that might come their way.
The spillovers generated by human capital formation,
even of the sector-specific variety, cannot be ignored. Some
of the skills that a worker acquires by training for and holding
a job in, say, the ecotourism sector can be transferred to other
occupations. Also, preparing people for work in nature-based
tourism, forest management, and genetic prospecting can,
under the right circumstances, diminish the cost of policing
park boundaries.
The findings presented in chapters 4-7 indicate that sec-
tor-specific human capital formation is unlikely to lead to
substantial income growth for large numbers of people living
in or close to threatened habitats. Unless the skills forest
dwellers acquire as a result of sector-specific training apply
readily to other parts of the economy, the rationale for ap-
proaches to conservation and development that have become
fashionable could turn out to be quite shaky.
This page intentionally left blank
II
The Economic Returns of
Environmentally Sound Harvesting
of Forest Products and of
Nature-based Tourism
This page intentionally left blank
4
Harvesting of Nontimber Products

ollecting fruit, nuts, latex, and other nontimber

C products occupies at least some of the time of
large numbers of people in the Amazon Basin and in other
forested regions of Latin America. In northern Guatemala, for
example, there are more than 7,000 chicle (Manilkara zapota)
collectors, whose work generates $4 million in annual ex-
ports (Nations, 1989, cited in Salafsky, Dugelby, and Ter-
borgh, 1992). At the time of the 1980 census, 68,000
households engaged in the collection of wild rubber (Hevea
brasiliensis) in the Brazilian Amazon (FIBGE, 1982, cited in
Allegretti, 1990). According to local newspaper reports, the
harvesting, processing, and marketing of agai and palm
hearts currently employ nearly 30,000 people and generate
an annual cash flow of up to $300 million in the Amazon
estuary (Pollak, Mattos, and Uhl, 1995). In the whole river
basin, 500,000 to 1,500,000 rural people derive a significant
portion of their income from the extraction of Brazil nuts
(Bertholletia excelsa), fruits like aguaje (Mauritia flexuosa),
and other commodities (Gradwohl and Greenberg, 1988;
Schwartzman, 1989).
With very few exceptions, however, this work is not finan-
cially rewarding. All available research indicates that most
extractor households are desperately poor, even by the mod-
est standards of Latin America's forested hinterlands. Citing
research carried out in Bolivia and in two places in Brazil,
Browder (1992b) contends that households that collect non-
timber products tend to be nomadic and illiterate, suffer from
high rates of infant mortality, and are frequently in debt. In
addition, they have often found it difficult to win govern-

43
44 Economic Returns

mental recognition of their natural resource rights when they

are faced with the competing claims of loggers, ranchers, and
agricultural colonists (Allegretti, 1990).

The Movement to Establish Extractive Reserves

It was out of struggles over land that the movement arose to
create extractive reserves, defined as communal holdings
where people support themselves by harvesting nontimber
products. During the 1970s, cattle ranching expanded in
the Brazilian Amazon, and large numbers of rubber tappers
were dispossessed. In response, they began to organize them-
selves, with assistance from agrarian unions and Catholic
social activists (Allegretti, 1990). By the late 1980s, the move-
ment had a high political profile and had succeeded in at-
tracting international backing.
Support for the rubber tappers was galvanized in Decem-
ber 1988, when ranchers killed Francisco ("Chico") Mendes
Filho and two of his associates in the state of Acre, in western
Brazil. Mendes had been an adept and charismatic union or-
ganizer whose international accolades included the Global
500 Prize of the United Nations. After his murder, the mo-
mentum to establish extractive reserves grew inexorable.
As far as collector households and their representatives are
concerned, the new sort of landholding was never intended
to exist for extraction of nontimber products alone. A Brazil-
ian anthropologist who has worked with the rubber tappers
since the 1970s has conceded that it is only to be expected
that the inhabitants of extractive reserves will raise crops and
livestock and cut down and sell timber whenever it suits
them to do so (Allegretti, 1990).
Browder (1992a) confirms that crop and livestock produc-
tion and logging are major sources of livelihood for the typ-
ical collector household. He also has warned those foreign
environmentalists who have lent support to rubber tappers
and other extractor populations against forgetting that the lat-
ter groups' agenda is primarily social—aimed at achieving
legal recognition of their informal tenure in forested land—
and only secondarily environmental.
The potential discord between international conservation
interests and forest dwellers' desires for land rights was ob-
scured, in a way, by a two-page article published in a pres-
tigious scientific journal less than a year after the Mendes
assassination. The article described a case study that ap-
peared to demonstrate that environmentally sound extrac-
Harvesting of Nontimber Products 45

tion of nontimber products can be much more profitable than

any economic alternative available to forest dwellers in the
Amazon Basin.
The case study involved the estimation of sustainable pro-
duction of aguaje fruit and other commodities on a one-
hectare site near the town of Mishana, in the Peruvian
Amazon. Output was multiplied by market prices in Iquitos,
a nearby port city with more than 250,000 inhabitants located
30 kilometers downriver and to the northeast, and harvesting
and transport costs were deducted. The resulting estimate of
potential annual income was $422 per hectare. When the net
returns generated by a selective timber cut every twenty years
were taken into account, the present value of exploiting the
site for fifty years was found to be $6,820 per hectare, assum-
ing a real interest rate of 5 percent (Peters, Gentry, and Men-
delsohn, 1989).
These estimates compare very favorably with the returns
of other land uses, which usually create more environmental
destruction. As Peters, Gentry, and Mendelsohn (1989) in-
dicate, the potential income associated with nontimber-
pro ducts extraction alone is two to three times greater than
the per-hectare gross revenues that cattle ranchers in the
Brazilian Amazon, for instance, are accustomed to earning.
Likewise, the combined present value of incomes from all
estimated extractive and logging operations turns out to be
more than a dozen times the price at which rural real estate
in the Amazon Basin normally changes hands. Of course, this
implies that, in a perfectly free market, individuals engaged
in nontimber-products extraction would end up owning
land, and it would not be possible to cover the opportunity
cost of other land uses.
The authors of the Mishana case study concluded that
"without question, the sustainable exploitation of non-wood
forest resources represents the most immediate and profitable
method for integrating the use and conservation of Amazon
forests" (Peters, Gentry, and Mendelsohn, 1989, p. 656).

Impediments to Economically and Environmentally

Successful Extraction of Nontimber Products
Especially during the first year or two after Peters, Gentry,
and Mendelsohn (1989) published their findings, many
groups and organizations that were working to arrest tropical
deforestation seemed to be convinced that the case for ex-
tractive reserves was irrefutable. But as Browder (1992a and
46 Economic Returns

1992b), Redford (1992), and others were quick to stress, es-

timates of potential earnings at one site in eastern Peru could
never be treated as conclusive proof that a way had been
found to keep all rain forests intact while simultaneously
raising forest dwellers' standards of living.
A number of issues were beyond the scope of the Mishana
case study. For example, the production increases that would
result if large forested tracts were dedicated to commercial
harvesting of nontimber products would undoubtedly drive
down prices, and hence extraction income. Peters, Gentry,
and Mendelsohn (1989) left the analysis of this impact to
other researchers. Likewise, they did not examine the mar-
keting impediments that various entrepreneurs have encoun-
tered as they have tried to sell nontimber products in markets
outside the Amazon Basin. Nor is it clear that postharvest
losses were factored into the analysis. If, say, one-third of the
potential harvest were never consumed or sold, then annual
extraction income would drop from $422 per hectare to less
than $200.
Something else to keep in mind is that Mishana is advan-
tageously situated in two important respects. First, it is close
to a sizable market; Iquitos is by far the largest urban center
in the Peruvian Amazon. Moreover, the city is somewhat iso-
lated, without good road connections to the outside world,
and a large share of the population is descended from Indians
and from the colonists who came to the area during the
Rubber Boom (in the early 1900s); these people are therefore
familiar with and like food and other things that can be gath-
ered in rain forests.
Access to markets is critical to the commercial viability of
nontimber-products extraction elsewhere. For example,
much of Brazil's, and the world's, palm hearts are produced
close to the mouth of the Amazon River (Anderson and loris,
1992). A crucial advantage of the area is that there are quite
a few exporting firms in the nearby port of Belem. There is
also a very strong market in that city for the fruit of the mul-
tistemmed agai palm, which is the source of palm hearts in
the Amazon estuary. Due to its perishability, ago/fruit is rel-
atively expensive to transport. Hence, locational rents, of the
sort depicted in Figure 3.1, are captured by producers situ-
ated near market hubs.
The second advantage that Mishana enjoys is that it is on
a floodplain that happens to be dominated by trees bearing
aguaje and a few other commercial fruits. Like market access,
a location in a riparian area subject to periodic inundation is
Harvesting of Nontimber Products 47

an advantage shared with the Amazon estuary, where agai

palms are abundant. Pointing out that only 2 percent of the
Amazon Basin can be categorized as floodplain forests, Brow-
der (1992a, p. 228) quotes Richard Howard, a former vice
president of the New York Botanical Garden: "The most im-
portant point is how poorly the Peters et al. [Gentry and Men-
delson, 1989] Peruvian Amazon hectare represents the
tropical forests [of the entire region]; it is the rare exception
rather than the rule." The Mishana investigators, it must be
said, have been careful to avoid giving the impression that
their site is representative of Amazonia as a whole (Peters,
1990).
Even if extraction of nontimber products at a site with fa-
vorable growing conditions has commercial potential, actu-
ally realizing this potential is far from a straightforward
matter. Whenever demand is strong and harvesting practices
are damaging, deterioration of the resource is all but assured
in areas beyond the frontier of property rights consolidation.
For example, one of the investigators in the Mishana case
study is quoted as warning that wild fruit populations "are
being rapidly depleted by destructive harvesting techniques
as market pressure begins to build" (Vasquez and Gentry,
1989, p. 350).
Rapid expansion of extraction of nontimber products, in
response to high demand for the plant in question, has often
been followed by resource depletion and a collapse of the
activity. Cascarilla roja (Cinchona spp.), which is the natural
source of quinine, is a case in point. During the nineteenth
century, the British decided to establish plantations of the
species in India, so that there would be a ready source of
medicine to combat malaria. But at least one of the botanists
who was sent to gather seedlings along the western slopes of
the Andes reported that his work was made difficult by the
depredations of extractors, who often stripped medicinal
bark in ways that killed trees (Spruce, 1970, pp. 240-241).
On the basis of detailed field observation of palm heart
extraction, Pollak, Mattos, and Uhl describe why those who
harvest nontimber products in what is in effect an open-
access forest are not interested in conservation:
Consider the extractor, who has to wade through mud, hack-
ing away vines and other undergrowth, to reach the base of an
ago/ clump. Once there, he is inclined to cut all stems, even
the small ones, before trudging through the forest in search of
the next clump. . . . [There] are certain advantages to cutting
smaller stems. . . . [In addition] delaying palm heart extraction
48 Economic Returns

may mean losing the palm hearts altogether to another inter-

ested party. (1995: 376-377)

A sure sign that open-access predation is taking a toll on re-

source quality is that the average diameter of harvested palm
hearts has declined dramatically since the early 1970s (Pol-
lak, Mattos, and Uhl, 1995). It was during that period when
much of the Brazilian palm heart industry relocated to Ama-
zonia from southern Brazil, where destructive extraction had
led to the depletion of Euterpe edulis Mart. (Ferreira and Pas-
choalino, 1987, cited in Pollak, Mattos, and Uhl, 1995). Un-
like its Amazonian cousin, the latter plant is single-stemmed,
and therefore does not recover easily after a harvest.
Aside from the depletion of open-access resources, there
are other threats to the long-term viability of extraction of
nontimber products. Agricultural domestication might oc-
cur. Also, synthetic substitutes can be developed. The most
famous episode of domestication occurred during the early
1900s, when plants smuggled out of the Amazon Basin were
used to establish rubber plantations in Asia. Since produc-
tion costs were lower on those plantations than they were in
Amazon rain forests, world prices fell and the South Amer-
ican Rubber Boom came to an end once Asian production
began. Also, neither extraction of cascarilla roja in the wild
nor plantation production proved to be competitive once
synthetic quinine began to be manufactured.
If a close look is taken at successful examples of extraction
of nontimber products, one is apt to detect many, if not all,
of the key elements of agriculture, or agroforestry, to be more
precise. For example, Pollak, Mattos, and Uhl (1995) have
determined that the most profitable way to produce palm
hearts in the Amazon estuary is in managed stands. In par-
ticular, management involves dispersal of apa/seeds and pe-
riodic thinning of undergrowth and canopy species. At
current prices, applying such a system for twenty years yields
a net present value of $119 per hectare, assuming a 6 percent
real interest rate. Another benefit would be that processing
facilities would no longer have to relocate every few years to
progressively more remote locations, as a result of resource
depletion (Pollak, Mattos, and Uhl, 1995).
Even though it creates little or no environmental damage,
managing agai stands, like what can be done with an aguaje
"orchard," is not what some advocates of nontimber prod-
ucts extraction have in mind. However, those individuals
should keep in mind that harvesting in unmanaged forests is
Harvesting of Nontimber Products 49

rarely an alternative, if the purpose is commercial production

as opposed to the gathering of forest products for household
use. (The latter, of course, is an important feature of rural life
in many forested areas of the world.) Reviewing historical
experience in the Peruvian Amazon, Coomes (1995) observes
a consistent pattern of harvesting far in excess of sustainable
levels, followed by some combination of resource depletion,
agricultural domestication (usually outside the region), and
synthesis. Invariably, the earnings of extractor households
have been marginal even during periods of peak activity.
Once the peak has passed, those earnings dwindle, often
evaporating entirely.
Extraction income continues to be very low. A financial
analysis in one Brazilian community, for example, reveals
that extraction's net returns at the household level are infe-
rior net returns for intensive or extensive agroforestry (An-
derson, 1989, cited in Browder, 1992a). In a survey of 164
rubber tapper households that he conducted in the Bolivian
Amazon in 1981, Romanoff (1981, cited in Browder, 1992b)
found that 96 percent were in debt and 71 percent suffered
periodic food deficiencies. Significantly, Brazilian rubber
tappers' earnings have been similarly meager in spite of the
national government's efforts to maintain prices above inter-
national levels (Allegretti, 1990).

Vegetable Ivory Production in Western Ecuador

Insights into the true commercial potential of extraction of
nontimber products, and how the income generated by this
activity ends up being distributed, can be gained by exam-
ining the collection, transport, processing, and selling of veg-
etable ivory in western Ecuador.
Vegetable ivory, which is used mainly to make buttons, is
obtained from the seeds of a hardy tagua palm species (Phy-
telephas aequatorialis) that grows in humid and seasonally
dry tropical forests below an altitude of 1500 meters in north-
western South America (Acosta-Solis, 1944; Barfod, 1991).
Ecuador began to ship tagua disks to button manufacturers
in Italy and other countries around 1900. According to re-
ports issued by the country's Central Bank, these exports
peaked during the 1920s and 1930s, totaling nearly $20 mil-
lion (in 1995 dollars) in 1925. Shortly after the Second World
War, plastic buttons were introduced, which greatly reduced
demand. For three decades beginning in the early 1950s, for-
50 Economic Returns

eigners' purchases of vegetable ivory were negligible (Coles-

Ritchie, 1996).
The tagua industry has rebounded in recent years. Exports
rose from $1.5 million in 1987 to $3.5 million in 1988; ship-
ments hit $6 million in 1991 and, for the time being, are stay-
ing at or above $4 million a year. (The true value of exports
is probably much higher since under-invoicing of exports is
a common practice in the country.) Italy continues to be the
main importer, having purchased 81 percent of Ecuador's
production in 1991. However, button manufacturers in other
parts of the world are showing an interest in tagua, which is
more appealing than plastic to many upscale clothing buyers.
In addition, the small handicraft market for vegetable ivory
has strengthened somewhat because of the ban on interna-
tional trade in products derived from elephant tusks.

Modes of Production and Marketing Channels

Available research suggests that tagua can live for more than
a century and that its seeds can lie dormant for well over a
year (Acosta-Solis, 1944; Barfod, 1991). Practically all pro-
ductive stands date back to the 1920s or 1930s and were es-
tablished through secondary succession, not planting. Since
the plant yields various useful products, including roofing
materials and livestock fodder, few people were energetic
about uprooting trees during the years when Ecuador ex-
ported almost no vegetable ivory. Maintenance, such as it is,
involves little more than the occasional removal of dead
fronds.
After being collected by rural households, tagua is sold to
intermediaries. There are no significant barriers to entry in
this sort of business and, as Southgate, Coles-Ritchie, and
Salazar-Canelos (1996) have found, marketing margins are in
line with what is observed in any competitive industry. By
contrast, the top end of the domestic marketing chain is much
more concentrated. Only a few firms slice dried tagua seeds
into disks. The two largest processor-exporters accounted for
roughly 45 percent of total shipments in 1991 and another
three firms shipped 30 percent (Southgate, Coles-Ritchie, and
Salazar-Canelos, 1996).
Lack of competition in processing and exporting has noth-
ing to do with economies of scale in production since any
business can expand capacity simply by installing more slic-
ing machines, each of which is run by a single operator, and
by enlarging the yard where raw tagua is dried. Instead, con-
Harvesting of Nontimber Products 51

centration is a consequence of barriers to entry on the mar-

keting side. Historically, it has been all but impossible to get
into the tagua exporting business without having good con-
tacts among Italy's button manufacturers. A current initiative
to promote sustainable tagua production in Ecuador (Calero-
Hidalgo, 1992), which has been supported by Conservation
International (CI), seeks to develop new markets. This is a
challenge since clothing manufacturers that have not used
tagua in the past need to be assured that large volumes of
high-quality vegetable ivory will be available before they stop
using buttons made of other materials.

A Survey of Vegetable Ivory Producers

In order to determine the distribution of net benefits from
tagua production as well as possible responses to price
changes, Southgate, Coles-Ritchie, and Salazar-Canelos
(1996) conducted surveys of collector households, interme-
diaries, and processor-exporters in 1993.
The research began with the identification of firms en-
gaged in the manufacture and international marketing of
tagua disks. Interviews with representatives of five enter-
prises, with offices in Manta (a coastal port city), Quito (the
national capital), or in both (see Figure 4.1), began in January
and were concluded by May. In addition to yielding the in-
formation needed to estimate processor-exporter income,
those interviews shed light on the domestic marketing of raw
tagua, which was needed to plan surveys of intermediaries
and collector households.
Instruments used in those surveys were developed in Feb-
ruary and March of 1993. The Foundation for Training and
Investment for Socio-Environmental Development (CIDESA),
which is responsible for local implementation of the CI Tagua
Initiative, and which has substantial field experience in the
places where tagua is produced, was intimately involved in
this work. It also arranged for a pretest of questionnaires in
the middle of March, in which twenty collector households
and three intermediaries in the province of Esmeraldas par-
ticipated (see Figure 4.1).
After a minor modification of the questionnaires, survey-
ing was carried out in Esmeraldas in April. Fifty-nine house-
holds were interviewed in three areas identified by CIDESA.
Due to an interruption of funding, surveying of 22 house-
holds in the province of Manabi did not take place until No-
vember. Twenty-one intermediaries in both Esmeraldas and
52 Economic Returns

Figure 4.1 Centers of Taqua harvesting and processing in Ecuador.

Manabi were questioned during the collector pretest and sur-

vey. The intermediary sample was comprised of truckers and
other small businessmen operating in or close to the com-
munities where members of the collector sample reside. As
already pointed out, Southgate, Coles-Ritchie, and Salazar-
Canelos (1996) found that intermediaries' earnings are simi-
lar to what one finds in other competitive markets.

Sample Descriptions The collector sample is representative of

the rural poor who populate northwestern Ecuador and other
areas where nontimber forest commodities are produced. The
typical household comprises 4.4 individuals in Esmeraldas
and 6.4 people in Manabi. Four-fifths of all the household
heads were born in the community where they now live and
most of the rest came from some other part of the same prov-
Harvesting of Nontimber Products 53

ince. Education levels are low, adults in Esmeraldas and

Manabi having completed 4.2 years and 3.0 years, respec-
tively, of primary school.
No interviewee identified tagua collection as his main
livelihood. Instead, 83 percent of the household heads de-
scribed themselves as farmers. The other primary occupa-
tions reported were fishing and small-time commerce. Six
percent of the sample worked at off-farm jobs, receiving ap-
proximately 5,000 sucres (equivalent in 1993 to $2.63) a day,
plus lunch.
Most households in the sample have one tagual, averaging
9.7 hectares, where family members, and no one else, collect
vegetable ivory. A third of the sample possesses two sites and
13 percent has three. Agroforestry is practiced on nine-tenths
of the sites, with bananas, cocoa, coffee, and oranges (in Man-
abi) interplanted with tagua. There is no such interplanting
on 14 percent of the Esmeraldas taguales. Five percent of the
Manabi sites were described by the households that exploit
them as communal forests.
As discussed below, patterns of tagua collection in Es-
meraldas are distinct from those in Manabi. In addition, there
are minor differences in local marketing. Some households
in Esmeraldas take their product to an intermediary in order
to be paid a price that is, on average, 22 percent higher than
what would be received at the farm gate. The typical inter-
mediary is the owner of a small business, who contracts with
truckers to deliver loads of vegetable ivory to processor-
exporters. In Manabi, where the roads are better, slightly
more than a third of the households sell directly to truckers;
the balance is marketed to local merchants, in much the same
way it is done farther to the north. There is no appreciable
difference between farm-gate prices and what intermediaries
pay for tagua delivered to their places of business (Southgate,
Coles-Ritchie, and Salazar-Canelos, 1996).

Cains from Collecting, Marketing, and Processing Vegetable Ivory

Southgate, Coles-Ritchie, and Salazar-Canelos (1996) used
household survey data to estimate an implicit daily payment
for time spent collecting vegetable ivory.
The survey revealed that production is higher in Esmer-
aldas (400 pounds are obtained from a representative site
during a single peak-season harvest) than it is in Manabi (200
pounds per harvest). Also, there are differences in labor in-
puts between the two provinces: 2.9 person-days per harvest
in Esmeraldas versus 2.0 person-days in Manabi. Further-
54 Economic Returns

more, producer-level tagua prices rose between April 1993,

when Esmeraldas households were surveyed, and November,
when interviews were carried out in Manabi. CIDESA em-
ployees and individuals involved in the vegetable ivory busi-
ness suggest that prices went up in both provinces by roughly
80 percent—from 4,900 sucres ($2.58) to 8,800 sucres ($4.63)
per hundred pounds in Manabi, for example.
Daily payments to labor employed in tagua collection
were estimated for median peak-season harvests in Esmer-
aldas and Manabi. Calculations were made with both the
lower prices that prevailed in April and the higher values
observed in late 1993. In addition, all estimates reflect a small
deduction for nonlabor inputs (e.g., the burlap bags used to
carry tagua to market). As the estimates reported in Table 4.1
indicate, median daily returns to labor employed in vegetable
ivory harvesting ($2.36 in Esmeraldas and $2.32 in Manabi)
compared poorly with off-farm wages— $2.63 a day plus
lunch (as noted above)—before prices rose. After the price
increase, daily returns—$4.40 in Esmeraldas and $4.37 in
Manabi—were above rural wages.
Whenever the daily returns for harvesting exceed the op-
portunity cost of labor, tagua resources have implicit scarcity
value. Interestingly, the de facto owners of these resources
seem willing to respond to positive tagua resource values by
dedicating more effort to management. In particular, 70 per-
cent of the collectors surveyed in Esmeraldas indicated that
they would respond to higher prices by trying to increase the
productivity of tagua stands. This could be done by pruning
more carefully and frequently. (The other Esmeraldas house-
holds reported that additional earnings would be used to try
to raise agricultural output [15 percent of the sample], to in-
crease consumption [11 percent], and to pay for the educa-
tion of children [4 percent].) One factor that has held down

Table 4.1 Daily Payments for Peak-Season

Tagua Collection in Ecuador
Province
Calculation Month Esmeraldas Manabi
At April 1993 prices $2.36 $2.32
At November 1993 prices 4.40 4.37
SOURCE: Southgate, Coles-Ritchie, and Salazar-Canelos (1996):
76. Reprinted by permission of CAB International.
Harvesting of Nontimber Products 55

household-level earnings in the past has been limited com-

petition among processor-exporters. These firms certainly ap-
pear to be highly profitable, as made clear in Table 4.2's
account of what a representative enterprise would have
earned, given April 1993's raw material prices.
The analysis of vegetable ivory processing and exporting
carried out by Southgate, Coles-Ritchie, and Salazar-Canelos
(1996) suggests that the industry has not achieved long-run
equilibrium, with earnings well above the norm for the Ecu-
adorian economy as a whole. Understandably, then, entry
into the industry has been taking place. Several new enter-
prises have begun to operate in Manta and Quito during the
last two to three years. Without a doubt, greater competition
helps to explain recent increases in raw material prices at the
household level. Similar adjustments can be expected in the
future as long as processor-exporters continue to earn super-
normal profits. Interpretation of Table 4.2 suggests that, even
if raw tagua prices rose to three times what they were in April
1993, profits would still equal 33 percent of revenues.
Since Southgate, Coles-Ritchie, and Salazar-Canelos
(1996) completed their study, real tagua prices have not
stayed at the levels that prevailed in late 1993. In November
1995, the prices producers in Manabi received were between
10,000 sucres ($3.45) and 12,000 sucres ($4.14) per hundred
pounds. In Esmeraldas, where tagua seeds tend to be larger,
producer-level prices have varied between 15,000 sucres

Table 4.2 Revenues, Costs, and Profits in Tagua

Processing, April 1993.
Gross Revenues $645,880
Sales of 225,000 pounds of disks 640,497
Sales of tagua flour and other by- 5,383
products
Expenses $255,079
Purchases of raw tagua 89,905
Wages and salaries 102,337
Administrative, electricity, and other 62,837
costs
Profits $390,801
Profits as a share of revenues 61%
SOURCE: Southgate, Coles-Ritchie, and Salazar-Canelos (1996):
77. Reprinted by permission of CAB International.
56 Economic Returns

($5.17) and 16,000 ($5.52) per hundred pounds. The director

of CIDESA attributes price weakness to diminished demand
for premium clothing in Europe and other affluent parts of
the world (Rodrigo Calero-Hidalgo, personal communica-
tion, 1995).

Extraction of Nontimber Products and

Rain Forest Conservation
There are a few tropical forests in Latin America where har-
vesting nontimber commodities truly is commercially viable.
In the floodplains of the Amazon River and its tributaries, for
example, useful species, like aguaje and agai, are not too
widely dispersed, and access to major urban markets, such
as Belem and Manaus, is relatively good (Goulding, Smith,
and Mahar, 1996).
But in most places, the earnings associated with extraction
of nontimber products are much more modest. Through the
early 1990s, for example, the net returns for tagua collection
were no better than the opportunity cost of unskilled rural
labor in one of the poorest parts of South America. The same
has been true of rubber tapping and other forms of nontimber-
products extraction in the Amazon Basin, both now and in
the past. By contrast, siipernormal profits are probably being
captured by the few Ecuadorian firms that process and export
vegetable ivory. This is similar to what happened on a larger
scale during the Rubber Boom. The Manaus Opera House, for
example, is lasting evidence of the wealth that was lodged at
the top of the domestic rubber marketing chain at the turn of
the century.
It is fortunate that tagua palms suffer no damage as a result
of harvesting, as has been the case for aguaje and cascarilla
roja extraction. However, vegetable ivory collection is rep-
resentative of commercial nontimber-products extraction in
that it usually does not take place in locations that are bio-
logically diverse. As previously mentioned, most tagua is
harvested in stands that have emerged as a result of second-
ary succession. The users of these stands, needless to say,
weed out other species that have no household or commer-
cial value. The typical tagual bears little resemblance to an
undisturbed primary forest. Instead, it represents a transition
to agricultural domestication, which along with synthesis is
what usually happens when demand for a forest product
stays strong. Once domestication or synthesis is commer-
cially successful, plants in the wild continue to have value
Harvesting of Nontimber Products 57

only insofar as there is a need for repeated genetic improve-

ment.
Outside of a few economic niches, characterized by good
access to markets and relatively favorable harvesting condi-
tions, the environmental wealth on which extractive activity
is based is not economically scarce. This means that, even if
resource rights were to be strengthened (which would help
to forestall wasteful harvesting of some products), resource
owners would probably not derive much rental income and
their incentives to engage in intensive management (i.e.,
quasi farming) would remain weak. Likewise, the skills
needed to gather nontimber products are not particularly
unique. All this implies that few forest dwellers are likely to
earn very much from nontimber-products extraction. That
line of work, then, comprises a very shaky foundation for an
integrated strategy for habitat conservation and local eco-
nomic development.
Since peaking in the late 1980s, enthusiasm for the extrac-
tion of nontimber-products has ebbed considerably. Knowl-
edgeable experts, like Browder (1992a and 1992b), have
provided convincing evidence of the activity's limitations.
As a result, it has become rare for the view to be expressed
that harvesting products like agai and aguaje can save trop-
ical forests.
Likewise, the activity no longer plays as prominent a role
as it once did in habitat conservation initiatives. Sponsoring
development agencies have grown skeptical about the claims
made by the few remaining enthusiasts. The typical venture
aimed at promoting commercially viable harvesting of non-
timber products is a small-scale one, aimed at taking advan-
tage of a limited niche characterized by an existing and
accessible market as well as favorable growing conditions.
There is not room in Latin America for many such projects.
This page intentionally left blank
5
Environmentally Sound
Timber Production

M any of those who hope to arrest the loss of threat-

ened habitats in Latin America by promoting
sustainable development of selected forest resources, like la-
tex and aguaje, staunchly resist any form of timber harvest-
ing. Their aversion to the latter activity is easy to understand
since, for many years, most of the region's loggers have ex-
hibited little if any concern for environmental impacts.
There is considerable scope, though, for containing the
damage that results from logging. New systems that recon-
cile the growing and harvesting of commercial timber with
biodiversity protection and watershed conservation are be-
ing developed. Better felling and skidding practices can be
applied as well, thereby eliminating many of logging's long-
term environmental impacts. In order to enhance regenera-
tion after an initial selective harvest, extraction proceeds
only after inventories and zoning, which includes the de-
marcation of sensitive areas that are to be left undisturbed.
After that, vines are cut and directional felling is practiced
in order to foster the regeneration of commercial species.
Also, roads and skidder trails are designed and built with
an eye toward avoiding soil erosion.
Research carried out by investigators affiliated with the
Institute for Man and the Environment of the Amazon (IM-
AZON) reveals that the adoption of improved practices in the
eastern Amazon is often hampered in forested hinterlands by
low stumpage values, which are mainly a consequence of
bounteous supplies of timber from lands that are not man-
aged. That research is reviewed in the first part of this chap-
ter. The second part contains an analysis of a project

59
60 Economic Returns

undertaken in the Peruvian Amazon, which yielded an in-

novative harvesting and management system. No attempt
was made to continue applying the system after external sup-
port for the project ended. Outside technical assistance
proved to have been vital. In addition, the attractions of sus-
tainable timber production were diminished by the disrup-
tions associated with guerrilla activity. Also having an
impact were low stumpage values, which had to do with ad-
verse governmental policies, inefficient marketing, as well as
resource abundance.
The findings presented in this chapter suggest that, just as
there are limited opportunities to keep natural habitats intact
and to raise forest dwellers' standards of living by promoting
the extraction of nontimber products, it is important not to
exaggerate the financial appeal of environmentally sound
logging. Like nontimber resources, timber is cheap in many
places, which makes it difficult to justify the sorts of invest-
ment needed for sustainable management. Furthermore, pro-
jected increases in global roundwood prices will not be large
enough to make a significant improvement in the commercial
prospects of sustainable tropical forestry.

Logging in the Eastern Amazon

The challenge involved in sustainable development of stand-
ing tropical forests in Latin America is brought into sharp
focus by a group of four studies carried out by IMAZON, a
private, nonprofit research institute located in Belem. Each
study addresses one contemporary mode of timber produc-
tion or another in the state of Para (see Figure 5.1), which was
where 87 percent of all the Brazilian Amazon's roundwood
was produced in 1988 (Verrfssimo et al., 1992). Taken to-
gether, the four pieces of research furnish an idea of the for-
estry sector's evolution over time and across the landscape
from an initial stage, in which timber is readily available and
mechanization is minimal, to a situation characterized by
higher capital intensity and incipient resource scarcity re-
sulting from cumulative depletion.

Riparian Logging
One IMAZON study, carried out by Barros and Uhl (1995),
addresses the oldest form of forest exploitation in the Brazil-
ian Amazon. In the seventeenth century, the two investiga-
tors point out, high-quality timber began to be harvested
Environmentally Sound Timber Production 61

Figure 5.1 The Eastern Amazon.

close to the banks of navigable rivers, for shipment to Europe.

The first sawmills and veneer factories were built in the Am-
azon estuary during the 1950s. These plants were interna-
tionally financed and powered by steam. Two species, Virola
surinamensis and Carapa guianensis, accounted for most of
the production, which was destined primarily for foreign
markets. An active program of road building, which began in
the 1960s (see chapters 2 and 3), has made it feasible to ex-
ploit forest resources in places away from navigable water-
ways. Nevertheless, timber harvesting close to the Amazon
River and its navigable tributaries continues today.
One sort of riparian logging activity is not very different
from what could have been observed during the colonial era.
In vdrzea lands that have built up over time on floodplains,
small independent teams, normally comprised of three men,
fell timber and manually drag or float it to nearby riverbanks.
There it is sold and assembled into log rafts that are pushed
to mills by small boats. Clear evidence of low mechanization
in this business has been obtained in a survey of sixty-three
62 Economic Returns

plants—both cottage facilities with circular saws and mills

with band saws that employ thirty people, on average—that
process vdrzea timber in the Amazon estuary and around
Santarem (see Figure 5.1). In particular, Barros and Uhl
(1995) found logs that had been cut with chain saws at just
twelve of those sixty-three plants. The rest had been felled
with axes.
Logging is more mechanized in riparian areas away from
floodplains. The teams of five men or so that operate in these
areas always use chain saws, and trucks are used to carry logs
to rivers or roads. The scale of such operations is more than
twice the scale of a typical vdrzea enterprise (2,311 cubic
meters per annum versus 873 cubic meters) and annual pro-
duction per person is higher in the former (492 cubic meters)
than it is in the latter (265 cubic meters). The more mecha-
nized system carries higher average costs as well. But the
additional expenditures are compensated because timber
harvested away from floodplains is typically of higher qual-
ity, and therefore fetches a higher price (see Table 5.1).
Loggers who operate in vdrzeas and adjacent dry lands
tend to go about their business with little regard for forest
regeneration. Barros and Uhl (1995) contend, though, that
sustainable production is possible, especially in floodplains.
As mentioned in the preceding chapter, these areas feature a
less diverse array of species. In addition, vdrzeas are well
stocked with commercial timber, growth rates are twice what
is observed in dry land forests, and logging involves less
damage to the canopy and ground because there are fewer

Table 5.1 Riparian Logging Costs in the Amazon Estuary,

Early 1990s.
Vdrzea Enterprise Dry Land Enterprise
Cost Item (873 m3 per annum) (2311 m3 per annum)
Labor expenses $3.83/m3 $2.06/m3
Stumpage payments 2.90 5.85
Equipment and fuel for felling 0.00 0.48
Equipment and fuel for transport
to river's edge 0.00 5.93
Combined costs $6.73/m3 $14.32/m3

Payment received at river's edge $9.00/m3 $18.00/m3

SOURCE: Data from Barros and Uhl (1995): 92.
Environmentally Sound Timber Production 63

vines connecting trees being cut down to those in which log-

gers have no interest (Barros and Uhl, 1995).
Barros and Uhl (1995) also identify three prerequisites for
the sustainable development ofvdrzea forest resources. First,
the environmental knowledge that local people have must be
complemented by training in inventorying, felling, extrac-
tion, and thinning techniques so that they can become effec-
tive resource managers. Second, their property rights in
forested land must be stronger. Third, stumpage values must
rise.
A small investment in technical assistance and modest
policy and institutional reforms would satisfy the first two
prerequisites for sustainable forestry development. However,
meeting the third condition is apt to be more of a problem.
A ban on log exports, instituted to protect wood processors,
would have to be eliminated. However, it is possible, even
likely, that the main impact of such a change would be the
encouragement of logging in more remote areas. Barros and
Uhl (1995) have found that the cost of transporting one cubic
meter of harvested timber for a distance of 100 kilometers is
as follows: $30.02, using a truck; $7.94, with a barge; and just
$1.02 when logs are rafted. With transportation costs this low
and with standing timber as abundant as the two IMAZON
investigators report it to be, stumpage values in a free and
competitive market are unlikely to rise very high, even in the
most favorably located vdrzea forests. Instead, the main con-
sequence of increased demand for stumpage, resulting from
the lifting of the log export ban or from some other event,
would be the acceleration of loggers' progress up the Amazon
River and its tributaries, which Barros and Uhl (1995) report
is already under way.

Extracting Mahogany in Primary Forests

without Roads
The high cost of reaching the outside world and its markets
has always hampered commerce in the Amazon Basin. To
this day, economic activity remains largely confined to areas
served by roads and navigable waterways, with only the most
valuable forest products being taken from less accessible
places. One such product is mahogany, a wood prized for its
durability, workability, and attractive color.
Commercial harvesting of the species did not begin in ear-
nest in the Brazilian Amazon until the 1960s, when the na-
tional road network started to extend into the region from the
64 Economic Returns

south. At that time, a standard pattern of operations for ma-

hogany firms was established, which continues to hold as the
industry has spread to the north and west. Instead of being
comprised of large numbers of independent firms and house-
holds, each specializing in logging, timber transport, or proc-
essing, the industry consists of integrated businesses. A
typical enterprise employs crews to search for and harvest
mahogany; arranges to move logs to mills (which the same
company owns); and sells the boards produced at its mills to
foreign buyers, mainly in the United States and Great Britain
(Verrissimo et al., 1995).
Since mahogany is so valuable, it is common for scouting
and harvesting crews to operate at sites several hundred kil-
ometers from their employer's mills, and far away from pub-
lic roads and highways. This mode of operations is
expensive, since it usually requires road construction and
delivering supplies by air. In the second of the four IMAZON
studies of the eastern Amazon's timber industry, Verrissimo
et al. (1995) have examined the operations of a single inte-
grated firm in southern Para, to the west of the Belem-Brasilia
Highway and to the south of the Transamazon Highway (see
Figure 5.1).
As indicated in Table 5.2, raw material costs incurred by
the firm depend greatly on where mahogany is harvested in
relation to processing facilities. When extraction takes place
600 kilometers away from the mill, stumpage payments drop
to zero, because locational rents are negligible, and because
there often is no owner around who can demand money for
standing timber. By contrast, transport costs are more than
an order of magnitude higher than what they are for a logging
operation carried out at a distance of 50 kilometers.
For all intents and purposes, mahogany logging in primary
forests amounts to mining. In inventories of three 100-hectare
plots in southern Para where felling and extraction had oc-
curred in the recent past, Verrissimo et al. (1995) found
virtually no representatives of the species with a diameter at
breast height (dbh) of 10 to 30 centimeters; saplings were rare
as well. Similar findings were obtained at four other sites in
the region where mahogany had been extracted recently. The
IMAZON investigators also cited evidence obtained in Mex-
ico, Central America, and other parts of the Amazon Basin
that showed poor regeneration of mahogany after harvesting,
due to the absence of large seed trees and of extensive light-
rich openings in the forest (Lamb, 1966; Snook, 1993). They
conclude that more than 100 years might need to pass be-
Environmentally Sound Timber Production 65

Table 5.2 Mahogany Felling, Extraction, and Transport Costs in

Southern Para, Early 1990s.

Cost Item 50 Kilometers 600 Kilometers

from Mill from Mill

Stumpage payments $70.00/m3 $0.00/m3

Reconnaissance and felling 4.60 4.60
Extracting logs from 23.80 $23.80
forests
Transporting logs to mill 12.00 144.00
Taxes on logs 3.00 3.00
Interest charges 4.60 4.60
Combined costs $118.00/m3 $180.00/m3
SOURCE: Data from Verrissimo et at. (1995): 48.

tween an initial mahogany harvest in natural forests and a

second cut.
Because of poor regeneration, the mahogany industry can
be expected to continue along its current track of venturing
ever farther into primary forests in search of undisturbed
stands of timber, at a progressively higher cost. As the second
column of figures in Table 5.2 indicates, the expense of de-
livering a cubic meter of timber to a mill approaches $200 as
the distance between logging and processing sites exceeds
600 kilometers. This expense is easy to justify, since mahog-
any boards fetch very high prices in world markets. But at
some point, the option of raising mahogany outside a natural
forest setting is certain to be commercially appealing. That
point appears to have been reached in southern Para, where
a substantial amount of mahogany is starting to be raised,
often in fields where pepper or rapidly growing timber spe-
cies are planted, too (Steven Stone, personal communication,
1996).
Mahogany does not grow very rapidly, which reduces
landowners' interest in raising the species in a plantation
setting. A more serious problem is the harm caused by the
shoot-borer moth (Hypsipyla grandella). In a financial anal-
ysis, Browder, Matricardi, and Abdala (1996) have found
that, if there were no insect damage, plantation production
would be profitable if stumpage could be sold for $150 per
cubic meter. Table 5.2's figures suggest that a mill should be
willing to pay such a price in order to avoid having to truck
in timber from remote forests.
Regardless of which path the mahogany industry takes in
the future—either continued ecosystem mining or tree farm-
66 Economic Returns

ing—forest dwellers will be in a poor position to derive much

benefit from it. Except for those fortunate enough to possess
advantageously located stands, resource owners are unlikely
to receive high onetime stumpage payments. Once harvest-
able stocks in the wild are sufficiently depleted and mahog-
any begins to be raised on plantations, the industry's interest
in natural forests will relate only to its needs for fresh genetic
material.

Logging along Active Frontiers of Colonization

When mahogany was being harvested primarily in areas be-
yond the reach of ranchers, farmers, and other settlers, en-
vironmental impacts consisted mainly of the eradication, for
several decades, of commercial stocks of that same species,
as far as Vemssimo et al. (1995) and other researchers have
been able to determine. But otherwise, the IMAZON team has
found that forest recovery, including the regeneration of
some commercial timber species, was well under way a short
time after mahogany loggers had left the scene.
The same sort of forest recovery can occur in more acces-
sible places, as the third of the four IMAZON studies makes
clear. Uhl et al. (1991) analyzed timber production in the late
1980s around Tailandia (see Figure 5.1), a town that was con-
nected to Belem, which is about 200 kilometers to the north,
with an asphalted state road in 1985. As in other places ex-
periencing rapid settlement and expansion of the wood prod-
ucts industry, loggers working in the vicinity of Tailandia
tended to be highly selective. At most, they were interested
in twenty species; on average, just two trees, each with a use-
ful volume of 6.2 cubic meters, were being extracted per hec-
tare in the late 1980s. Even though felling and skidding
resulted in substantial damage to trees left in the forest, in-
cluding commercial species, timber was regenerating rap-
idly. On ten plots, each measuring 5 meters by 15 meters and
located in separate harvest openings, an average of 14.3 small
individuals (standard deviation, 6.7) from species that log-
gers in the region were harvesting were counted, although
Manilkara huberi, which is the most commonly extracted
kind of timber, proved to be rare (Uhl et al., 1991).
The depletion of commercial stocks is certainly a concern
when and where the intention is to use natural forests as a
long-term source of wood. Also, as indicated in chapter 1,
logging can increase the risk of forest fires, sometimes dra-
matically (Uhl and Kauffman, 1990). However, there is an-
Environmentally Sound Timber Production 67

other aspect of timber production in frontier regions that has

more of an impact on the environment. As Uhl et al. (1991)
documented in the case of Tailandia, 69 percent (in terms of
linear distance) of all side roads suitable for vehicular traffic
in the area were built entirely or partially by loggers. Con-
structing these roads, which were and continue to be a com-
mercial lifeline for local farmers and ranchers, has often led
to the conversion of forests into cropland and pasture.
That the upgrading of transportation infrastructure was
linked to rapid agricultural land clearing around Tailandia
had much to do with low timber values. In addition to what-
ever improvements loggers might have left behind, resource
owners received only about $.80 per cubic meter for their
stumpage. At prices this low, the present value of a few sea-
sons of crop production or of a few years of cattle ranching
exceeded the present value of a second timber harvest twenty
years or so after loggers' initial intervention. For example,
Uhl et al. (1991) estimated that crops harvested during a sin-
gle season were worth $460 per hectare in the late 1980s. By
contrast, the value of timber extracted from a secondary forest
after twenty years of good regeneration would have been
$770 per hectare, assuming that the price paid for felled tim-
ber in 1989, $15.10 per cubic meter, exhibited no real growth
over time (Uhl et al., 1991). Obviously, managing forests after
intervention so as to enhance future timber production
would not have been remunerative, even at a very low real
interest rate.
Uhl et al. (1991) contend that, during the late 1980s, there
was some scope for raising timber prices, and thus for en-
hancing resource owners' interests in sustainable forest man-
agement in the vicinity of Tailandia. In particular, they argue
that the mills that were purchasing timber and converting it
into boards could have paid more than $18 per cubic meter
for their inputs. Resource owners, they also observe, could
have used the additional monies to pay for improved man-
agement of production forests. The IMAZON researchers
point out that a typical mill's monthly operating and main-
tenance costs ($23,440), denned not to include depreciation
and interest expenses, amounted to 81 percent of its monthly
revenues ($28,800). Also the industry was not using raw
materials very efficiently, two cubic meters being lost in proc-
essing for every cubic meter of finished product manufac-
tured.
These observations do not add up to convincing proof that
the wood-products industry's profits were excessive. Indeed,
68 Economic Returns

net returns were actually quite modest, particularly after al-

lowances were made for depreciation and other capital
charges. Profit margins were certainly exceeded by what has
been earned by firms that process and export vegetable ivory
in western Ecuador, for example. Also, the competitive mar-
ket forces containing profits appeared to be strong, there
having been forty-eight sawmills in and around Tailandia in
1989 (Uhl et al., 1991). In addition, low processing efficien-
cies were more a consequence, than a cause, of low stumpage
values. After all, there was little reason to make the invest-
ments required to utilize a higher portion of raw materials
when those materials were abundant, and therefore cheap.

Timber Production in More Settled Areas

As of the late 1980s, some of the Tailandia landowners sur-
veyed by Uhl et al. (1991) were refraining from selling stump-
age in the hope that prices would rise appreciably over time.
These expectations could well have been ill-founded, if the
findings obtained in the fourth IMAZON case study (Verris-
simo et al., 1992) are any guide.
The setting for the latter research was the area around Par-
agominas (see Figure 5.1), which is on the Belem-Brasilia
Highway. This road having been in operation since the late
1960s, it is accurate to describe the region as an old frontier,
one in which settlement processes, of the sort that were in
full swing around Tailandia in the late 1980s, have largely
run their course.
Much of what Verrissimo et al. (1992) have to report about
timber production in the vicinity of Paragominas is similar
to what Uhl et al. (1991) have found near Tailandia. Logging
takes a toll on the forest; on average, twenty-seven trees
greater than or equal to 10 centimeters dbh are damaged for
every tree extracted. However, natural regeneration is robust,
4,300 seedlings and saplings of commercial species being
registered per harvested hectare.
As one might expect in a place where both agricultural
land clearing and depletion of timber resources have reached
an advanced cumulative stage, timber values have started to
rise. In 1989, when Verrissimo et al. (1992) conducted their
field research, small logging firms, typically comprised of
thirteen individuals with a pair of chain saws, a bulldozer
and a log loader, as well as three trucks, were receiving
$27.50 for a cubic meter of roundwood. This was 50 percent
higher than what mills around Tailandia were paying for tim-
Environmentally Sound Timber Production 69

ber at exactly the same time (as noted above) and 50 percent
higher than the price for logs extracted from dry lands close
to navigable rivers a year or two later (see Table 5.1). Re-
sponding to higher prices, wood processing firms had chosen
to operate at somewhat improved levels of efficiency. Instead
of three cubic meters of raw materials being used to produce
one cubic meter of output, as was the case in Tailandia. 47
percent of the roundwood going into a typical mill was
emerging as finished product (Verrissimo et al., 1992).
Stumpage prices, however, were not much higher in the
old frontier region than they were where resources were more
abundant. As of 1989, a cubic meter of uncut timber in the
Paragominas region was worth $1.84 (Verrissimo et al.,
1992), compared to $.80 around Tailandia. One reason why
prices remained low was that trucking in logs from other lo-
cations is not prohibitively expensive. In addition, process-
ing facilities (e.g., small band-saw mills) can be moved from
place to place without great difficulty. So can labor. In gen-
eral, a high level of factor mobility diminished location-
al rents in the late 1980s, and has continued to do so to
this day.
Since prices of standing timber have not risen very much
in the vicinity of Paragominas, resource owners and loggers
have been slow to adopt better management practices. Ver-
rissimo et al. (1992) carried out an economic analysis of two
such practices. The first is vine removal at least eighteen
months before a harvest, which reduces collateral damage to
commercial species left in the forest. The second is the prac-
tice of refinement thinnings, which promotes the growth of
those same species after logging has taken place. Together,
these two practices were estimated to cost $120 per hectare.
But even this modest expenditure was found not to be re-
munerative under market conditions that prevailed in the
late 1980s.
Since that time, deforestation has continued in the Para-
gominas region, which in turn has caused timber to become
even more scarce. Stumpage prices currently exceed $5 per
cubic meter (Paulo Barreto, personal communication, 1996),
and mills are now paying more than $35 per cubic meter for
roundwood. A recent IMAZON-sponsored survey of the re-
gion's wood-products industry indicates that several smaller
mills, which can relocate fairly easily, have departed, pre-
sumably for areas where agricultural land clearing has not
reached such an advanced stage. But less mobile plants, like
veneer factories, are responding to higher prices by making
70 Economic Returns

the investments needed to raise processing efficiency. At the

same time, at least some landowners are sufficiently sure of
their property rights to begin considering forestry as a per-
manent land use option (Stone, 1996).
Technology for improved forest management is available,
and several measures for softening logging's environmental
impacts appear to be profitable. In another IMAZON study,
Barreto et al. (in press) offer an economic assessment of var-
ious measures that are a part of planned timber harvesting.
These include mapping out roads and skidder trails so as to
reduce soil erosion, vine removal, and directional felling.
Collecting data on a 105-hectare plot where all of this was
done, and on a 75-hectare parcel where logging proceeded in
the usual way with no planning whatsoever, the investigators
have found that applying a plan to limit damages results in
a 15 percent increase in logging crews' hourly output and
reduces average machine time requirements by 2 7 percent to
37 percent. Furthermore, the failure to plan logging opera-
tions results in the waste of 26 percent of all felled timber—7
percent due to poor harvesting techniques, and 19 percent
because logs that have been cut down are simply left in the
forest. Between productivity impacts and the reduced waste
of timber, planning results in financial returns of $3.70 for
every cubic meter felled. These benefits do not reflect any
payoff associated with leaving a better-stocked stand of trees
behind after timber extraction has been completed.
Barreto et al. (in press) lament that no effective regulations
are in place to oblige loggers to adopt improved practices.
Under the Brazilian forestry law, harvesting and management
plans are supposed to be prepared, officially approved, and
applied in the field. But the public sector institutions re-
sponsible for enforcing this system are severely understaffed
and poorly funded. In practice, any logging scheme featuring
the removal of all large commercial timber, which really
amounts to high-grading, receives governmental permission.
Like many others, the IMAZON investigators would like
to see improved regulations and tighter enforcement. How-
ever, they also acknowledge that the adoption of sustainable
forestry is seriously impeded because of a general lack of
technical information and qualified personnel in the sector.

The Palcazu Forestry Project

IMAZON's program of research reveals that, except where
deforestation has largely run its course and timber values
Environmentally Sound Timber Production 71

have begun to rise, incentives for managing natural forests

for wood production are weak in the Amazon Basin. Never-
theless, natural forest management has been promoted in
many parts of the region. Under the auspices of one project,
carried out during the 1980s in the Palcazu Valley of central
Peru, an innovative harvesting and processing system was
developed and applied, with the participation of local indig-
enous communities. That system was abandoned, however,
soon after outside financial and technical assistance was
withdrawn. A fundamental reason for the project's collapse
was low stumpage values, which had much to do with re-
source abundance.

Setting and Background

Located approximately 400 kilometers northeast of Lima, the

Palcazu Valley is typical of the places in Colombia, Ecuador,
Peru, and Bolivia where the Andes give way to Amazonia.
Elevation there varies dramatically, from 3800 meters above
sea level, in the western reaches of the watershed, to 270
meters where the Palcazu River flows into a larger tributary
of the Amazon River (Hartshorn, 1990). The climate is hot
and wet. Average annual temperature is 23.6 degrees Celsius,
and precipitation in a typical year, which features no pro-
nounced dry season, is 6300 millimeters. Under these con-
ditions, natural vegetation is luxuriant (Hartshorn, 1990).
Aside from a few places where fertile alluvial soils have
built up over time, and which the indigenous population has
used for generations as sites for raising beans, maize, and
other crops (Benavides and Pariona, 1995), the Palcazu
Valley is largely inhospitable toward agriculture. The red
clay soils that predominate are extremely acidic, with a pH
of 3.8 to 4.5; feature high concentrations of aluminum; and
are prone to erode when exposed to driving rains. Moreover,
they are highly leached, and consequently lack major nutri-
ents like calcium, phosphorus, and potassium. Even less fer-
tile are the white sandy-clay loam soils that border many
rivers and streams. No more than 8 percent of the entire val-
ley is suitable for crop production (Hartshorn, 1990).
Notwithstanding these limitations, agricultural coloniza-
tion was the main thrust of central government policy for
places like the Palcazu Valley for many years. Shortly after
President Fernando Belaunde Terry, who had been deposed
during a military coup in 1968, returned to office in 1980, his
government announced plans to construct roads; to establish
72 Economic Returns

wood-processing and other industries; and to settle 150,000

colonists in the region, under the auspices of the Pichis-
Palcazii Special Project (PEPP). Keen to support Peru's return
to civilian government, the U.S. Agency for International De-
velopment (AID) promised funding and technical assistance.

The AID Project

From the outset, PEPP met with fierce opposition from the
indigenous Yanesha (also known as Amuesha) communities
and their Peruvian and foreign allies. Responding to these
criticisms, and the cautionary advice of its own consultants
(Smith, 1982, cited in Benavides and Pariona, 1995), AID
decided not to back colonization. Instead, $22 million, in-
cluding $4 million for technical assistance and project de-
velopment, was allotted to the Central Selva Resource Man-
agement Project (CSRMP). A protected reserve was to be set
up and managed and a system for sustainable timber exploi-
tation was to be developed and applied. Environmentally
sound agricultural production was to be promoted and pub-
lic health services were to be upgraded as well.
As stressed by an expert affiliated with Costa Rica's Trop-
ical Science Center (TSC), which provided technical assis-
tance to the CSRMP's forestry component, the challenges of
sustaining timber resource development in a place like the
Palcazu Valley were considerable (Hartshorn, 1990). Govern-
ment policies have accelerated agricultural land clearing and
public sector institutions have had little or no capacity for
furnishing useful advice to those with a potential interest in
forest management. Low concentrations of commercial tim-
ber and high extraction costs were additional problems. In
addition, there was a negligible level of understanding of
tropical forest dynamics and the regeneration requirements
of canopy tree species.
The CSRMP did not address policy issues or attempt a
thorough overhaul of institutions engaged in forestry re-
search and extension. Instead, primary emphasis was placed
on developing and promoting an alternative to the usual pat-
tern of unplanned high-grading of a limited number of spe-
cies, which is what takes place throughout the Brazilian
Amazon (as noted previously) and elsewhere in the Ameri-
can tropics. TSC investigators were convinced that an alter-
native approach was viable since, along with national and
international demand for fine tropical woods, local and na-
tional markets exist for a wide variety of species. The impli-
Environmentally Sound Timber Production 73

cation of this is that the volume extracted from any given

place could be increased appreciably, which would in turn
diminish extraction costs and raise revenues (Hartshorn,
1990).
Due to the requirements of project implementation, there
was no time for a thorough analysis of forest dynamics in the
Palcazii Valley. However, the TSC assistance team was able
to draw on observations of critical importance made in Costa
Rica. Scientists at the La Selva Biological Station (see Figure
7.2) had documented that the growth of shade-intolerant spe-
cies in gaps that open, when large trees are pushed over by
the elements, is an intrinsic feature of tropical forest ecology.
Half of all the tree varieties at La Selva, and 63 percent of
the canopy species, require well-lit areas for regeneration
(Hartshorn, 1978; Hartshorn, 1990). Gap-phase dynamics
therefore became the central feature of the TSC management
plan for the CSRMP (Hartshorn, 1990).
Experimentation carried out to identify the proper dimen-
sions and orientation of gaps to be clear-cut began in 1985.
A test strip 20 meters across proved to be too narrow for the
emergence of large-crowned trees and it was determined that
the growth of weedy species would be excessive on 50-meter-
wide strips. It was decided to make production blocks, which
vary in length from 200 to 500 meters, 30 to 40 meters in
width.
Tree regeneration in these blocks, which are supposed to
be logged every thirty to forty years, has been excellent for
two reasons. First, neither burning nor farming, which de-
stroy seeds, ever take place. Second, living seed sources are
always nearby since strips that will be logged are always in
the midst of intact forests. In addition, vines are removed
before the harvest and enrichment thinning is done after tree
emergence in order to enhance site quality. Also, soil loss is
minimized by rotating logging sites and by running strips
along topographic contours (Hartshorn, 1990; Hartshorn, Si-
meone, and Tosi, 1986; Tosi, 1986).
The TSC scheme called for extracting everything with a
diameter of 5 centimeters or more. This represents a dramatic
departure from standard practice in and around the Palcazii
Valley. Like their counterparts in Para, loggers in eastern
Peru rarely cut down more than ten mature trees from a hec-
tare of primary tropical forest. All other vegetation remains,
often in a damaged state because of careless harvesting. In-
dustry sources report that per-hectare extraction rates in
the region are below 15 cubic meters and that high-quality
74 Economic Returns

hardwoods, cut with chain saws into crudely dimensioned

planks, comprise most of the output (Southgate andElgegren,
1995).
Of course, this pattern of forest exploitation makes sense
where logging, transport, and processing costs are high. Elec-
tricity, for example, is considerably more expensive in east-
ern Peru than in other parts of the country. Since inefficient
diesel-powered generators are the primary source of supply
in the Amazon, prices there average $0.20 per kilowatt-hour,
versus a national average of $0.05 per kilowatt-hour. Under
these conditions, payments for electricity amount to as much
as a fifth of wood processing costs and investments in proc-
essing capacity are minimal (Southgate and Elgegren, 1995).
In spite of adverse economic conditions, making use of
timber with a small diameter was judged to be important to
the functioning of the harvesting scheme based on gap-phase
dynamics. Accordingly, a small mill was installed to convert
timber of varying dimensions into different sorts of wood
products: treated utility poles and fence posts, charcoal, and
the sawed lumber normally exported from the region (Hart-
shorn, 1990; Simeone, 1990).
Another distinctive aspect of the CSRMP forestry compo-
nent was the importance given to participation by indigenous
communities. An early decision was made not to involve col-
onists, who had already converted most of their respective
holdings to pasture and cropland and who lacked the social
cohesion of the Yanesha. Work with the latter group began
with participatory land use capability assessments. The
Yanesha Forestry Cooperative, Limited (COFYAL) was estab-
lished, and plans to extract timber from some forests and to
set aside other areas as reserves were adopted democratically
(Simeone, 1990). A major benefit that the native population
associated with the project was stronger formal property
rights; each of the fourteen indigenous communities in the
valley received a communal land title (Benavides and Par-
iona, 1995).

Project Performance
Reflecting subsequently on the forestry activities he carried
out with COFYAL, Simeone (1990) observed that outside
technical assistance would be needed for many years if the
production, harvesting, and milling scheme and marketing
initiatives were to succeed. Poor performance of the system
Environmentally Sound Timber Production 75

in the years immediately following AID's departure proved

him right.
Foreign technicians and scientists paid by the U.S. gov-
ernment withdrew in 1989, largely in response to guerrilla
activity in the vicinity of the Palcazu Valley, but not among
the Yanesha themselves. Before that time, up to a dozen such
individuals were active in the project. During the next four
years, the World Wildlife Fund (WWF) and the Peruvian
Foundation for the Conservation of Nature (FPCN), an envi-
ronmental organization based in Lima, provided limited
support. This allowed for four advisers, one each in admin-
istration, forest management, forest extension, and market-
ing, to continue working with COFYAL (Benavides and
Pariona, 1995). But in 1993, even this assistance came to an
end.
COFYAL's performance before 1993 compared poorly
with what had been expected when the CSRMP was being
designed, ten years earlier. Using data obtained from AID re-
ports and other sources, Elgegren (1993) replicated the ex
ante financial analysis of the forestry component. He esti-
mated the base-case internal rate of return to be 20 percent,
and also found that profitability was especially sensitive to
changes in output prices and production costs. He also eval-
uated the cooperative's forestry operations in 1991, when
harvesting took place on three strips with a combined area
of 2.87 hectares. Significantly, he found that revenues
($5,491.83 per hectare) were lower than costs ($5,614.89 per
hectare).
One reason why earnings were disappointing was that the
prices received for COFYAL timber were low. On average,
hardwood boards, which accounted for 40 percent of total
production, were sold for $88.98 per cubic meter locally and
for $135.59 per cubic meter in Lima. These prices were well
below FOB border values, which exceeded $500 per cubic
meter at the time (Southgate and Elgegren, 1995).
Uneven quality and marketing mistakes contributed to the
low payments received for COFYAL output. On at least one
occasion, for example, a buyer in the United Kingdom com-
plained to the FPCN that there was too much empty space in
shipping containers (Southgate and Elgegren, 1995). Need-
less to say, this drove down the payments received by the
cooperative.
In addition, public policy tended to depress timber values.
By the early 1990s, the Peruvian government was not regu-
76 Economic Returns

lating or taxing the export of unprocessed lumber. However,

exporters were obliged to deposit foreign currency earnings
with the Central Bank and then wait for several weeks to be
paid back in Peruvian soles, at exchange rates set at the time
of deposit. During 1991, when Peru suffered one of the high-
est rates of inflation in Latin America, this arrangement di-
minished the revenues received by wood exporters by 30 to
35 percent, on average (Southgate and Elgegren, 1995).
Depressed revenues were also a consequence of low pro-
duction. Overall timber yields, which approached 45 cubic
meters per harvested hectare, were three times what is ex-
tracted when normal logging practices are employed (see the
earlier description of selective timber extraction). However,
practically all of the difference was accounted for by the util-
ity poles and fence posts (with yields of 55.40 and 188.85
units per harvested hectare, respectively, and with a com-
bined volume of less than 30 cubic meters) manufactured
from smaller timber; also, production of sawed tropical hard-
wood amounted to only 18.68 cubic meters per harvested
hectare (Elgegren, 1993). The latter yield, in addition to being
little more than what is obtained with standard extraction
techniques, compares poorly with inventories of standing
timber in the Palcazii Valley: 150 cubic meters of sawed logs
per hectare and 90 cubic meters of posts and poles per hec-
tare (Hartshorn, 1990).
It is revealing that the net economic losses ($123.06 per
harvested hectare) incurred by COFYAL as it tried to imple-
ment the TSC system exceeded the net losses suffered by a
private logging firm operating in adjoining lands. The latter,
which were calculated taking all capital and operating and
maintenance expenses into account and using data provided
by the firm, amounted to $34.57 per harvested hectare (El-
gegren, 1993).
Apparently cognizant of the financial advantages of the
selective extraction techniques that prevail in central and
eastern Peru and throughout the Amazon Basin, COFYAL de-
cided to apply those same techniques on some of its lands at
the same time that strips were being harvested in a way gen-
erally consistent with TSC guidelines. For example, only 46
percent of the timber that the cooperative produced for saw-
milling in 1991 actually came from strips. The balance was
extracted in much the same way that local loggers would
have done it if given the opportunity (Southgate and Elge-
gren, 1995).
Environmentally Sound Timber Production 77

After WWF and FPCN support came to an end, the Yane-

sha disbanded the cooperative, which had never operated
profitably. The latest word from the Palcazu Valley is that
strip harvesting has ceased, as has the utilization of small-
diameter timber. Both high-grading of timber and agricultural
land clearing are taking place in the areas set aside for sus-
tainable timber production by COFYAL (Benavides and Par-
iona, 1995].

Lessons Learned

Abandonment of the CSRMP does not mean that the efforts

of AID, environmental organizations, technical assistance
contractors, and COFYAL were entirely futile. The regener-
ation that is occurring on harvested strips suggests that the
logging scheme developed under the project's auspices is bi-
ologically sound. Economic performance was indeed much
less encouraging, although a conclusive test of feasibility was
preempted by the disruptions caused by guerrillas and by
adverse public policy.
To be sure, domestic prices for lumber and other tradable
goods would have been higher, and incentives to apply the
TSC harvesting and processing system would have been
stronger, had exporters been able to choose when to convert
foreign earnings into domestic currency. It can even be ar-
gued that, without price distortions, the opportunity cost of
land dedicated to sustainable forestry would have been cov-
ered. Suppose, for example, that payments to COFYAL in
1991 had not been depressed by 30 percent (i.e., that reve-
nues had been $7,700 instead of $5,500 per harvested hec-
tare). With no improvements in the efficiency of timber
extraction or milling, average annual income on a 40-hectare
site, where a 40-year, TSC-style rotation was being followed,
would have been $52.50 per hectare (equal to one-fortieth of
the difference between $7,700 in revenues and $5,600 in
costs). At a real interest rate of 10 percent, the present value
of maintaining this income level indefinitely is $525 per hec-
tare. If anything, this amount exceeds average farmland val-
ues in and around the Palcazu Valley (Elgegren, 1993).
But whether or not the TSC system is truly viable in the
Peruvian Amazon remains in doubt. For the feasibility issue
to be settled, one would have to examine not just the oppor-
tunity costs of land, labor, and other factors obtained locally,
but also the scarcity values of management, marketing, and
78 Economic Returns

technical expertise brought in from the outside. COFYAL ex-

perience demonstrates just how critical the latter inputs
were; operating the mill, for example, proved to be beyond
the reach of the local community (Benavides and Pariona,
1995). The possibility that needs to be faced squarely is that,
unless and until standing timber grows much more scarce, it
will not make economic sense to devote a great deal of time
or effort to forest management and related tasks in places like
the Palcazii Valley.

Prospects for the Sustainable Development

of Tropical Timber Resources
Covering the opportunity costs of all the inputs needed for
forest management, timber processing, and the marketing of
wood products is proving to be a major challenge in other
sustainable forestry initiatives. Performance of a Bolivian
project that is very similar to the CSRMP is a case in point—
one of the participants argues that it would be impossible to
practice good silviculture while at the same time carrying out
processing and marketing operations without financial and
technical assistance from the outside (Olivera, 1995).
Attracting management, technical, and marketing exper-
tise to environmentally sound forestry ventures is sometimes
made more difficult by public policy. The recent experience
in Costa Rica of one of the TSC scientists who worked in the
Palcazii Valley is instructive in this regard. That individual
is currently trying to upgrade timber quality on a 600-hectare
forested parcel, located in humid lowlands about 100 kilo-
meters north of San Jose (see Figure 7.2). To promote regen-
eration at the site, where logging has taken place in the past,
large specimens of the most valuable species will be left to
stand for at least a few years so that they can serve as seed
sources; only the smaller trees are to be harvested and ex-
tracted, preferably with oxen so that erosion and damage to
remaining vegetation are minimized. However, the TSC sci-
entist complains that this sort of innovation is being frus-
trated by public officials, who encourage high-grading, just
as their counterparts in Brazil (Barreto et al., in press) and
other Latin American countries do. In the specific case of
Costa Rica, officials are accustomed to officially sanctioning
selective logging (which normally involves the removal of all
commercial timber larger than 60 centimeters dbh) and to
granting permits to transport logs to mills in and around the
capital city (which can process only raw material with a large
Environmentally Sound Timber Production 79

diameter). He also complains that forest taxes, which amount

to 10-to-15 percent of the value of harvested timber, diminish
stumpage values, and therefore resource owners' interest in
conservation (Joseph Tosi, personal communication, 1995).
The rules and procedures of international development
banks and agencies likewise inhibit the free flow of skills and
expertise to forestry ventures in Latin America and other
parts of the developing world. Just about any project that AID
proposes to carry out in a tropical forest setting must be pre-
ceded by a thorough environmental impact statement. Like-
wise, the Inter-American Development Bank (IDE) offers only
limited support for the sector.
In primary tropical forest settings, the Bank may support op-
erations to enhance the ability of responsible agencies to man-
age forestry resources in a sustainable manner. However, the
Bank does not finance commercial logging in these forests, nor
the purchase of equipment for such purposes. (IDE, 1994: 34)

Foresters employed by the World Bank, which has similar

requirements, indicate that they would be reluctant to spend
a great deal of time working on a promising venture for de-
veloping timber resources in either a primary or a secondary
forest. Their concern is that it might end up being too difficult
to prove that a project is not running a substantial risk of
unacceptable environmental damage.
One place where the ingredients for sustainable forestry
seem to be present is Quintana Roo, in southeastern Mexico.
Local communities there have organized to solidify their con-
trol over tree-covered land and to develop alternative mar-
keting channels, so as to receive higher prices for timber
(Bray et al., 1993). In addition, Quintana Roo holds major
advantages over alternative sources of mahogany, which is
the region's main wood product. For example, the cost of
delivering logs to local sawmills has been estimated to be $80
per cubic meter (Richards, 1991), which is considerably less
than the expenses that mahogany companies in the Brazilian
Amazon face (see Table 5.2).
Even in Quintana Roo, though, there is no guarantee that
environmentally sound development of timber resources can
truly pay for itself. Germany's Agency for Technical Assis-
tance (GTZ) has been providing advice and support since the
early 1980s. Also, private foundations are supporting local
forestry cooperatives' attempts to tap fledgling markets for
tropical timber that one nonprofit group or another has cer-
tified as having been produced on the basis of sustainable
80 Economic Returns

forestry. But at the same time, Guatemalan production and

exports, much of which are illegal, are exerting downward
pressure on mahogany prices. Furthermore, a 25-year cutting
cycle has been selected in Quintana Roo (Bray et al., 1993;
Richards, 1991). Since all available research suggests that
mahogany requires much more time to regenerate, the latter
decision presumably reflects a finding that a longer cycle is
unprofitable from the perspective of resource owners.
To admit that sustainable development of forest resources
might not really be taking place in Quintana Roo would be
discouraging for anyone pinning his or her hopes on a re-
placement for the wasteful and destructive logging that pre-
dominates throughout the American tropics. That AID's
efforts in the Palcazii Valley were largely unsuccessful can
be blamed on poor roads, guerrillas, and other exogenous
factors. Also, the challenge to sustainable forestry is obvious
in a place like Para. Brazil's second-largest state is nearly as
large as California, Oregon, Washington, Nevada, and Idaho
combined and, since its timber resources are virtually
boundless, market forces are stacked strongly against conser-
vation. But the situation is different in Quintana Roo. The
region is relatively accessible; social cohesion is better there
than in many other places; and mahogany, which is widely
sought around the world, grows well there. Thus, if outside
technical assistance, which has been provided for more than
fifteen years, will have to continue indefinitely in order to
maintain a production system that does not allow for com-
plete resource regeneration between harvests, the prospects
for environmentally sound commercial forestry just about
anywhere else must be considered dim, indeed.
Richard Rice, an economist employed by CI, which is in-
volved in vegetable ivory production in Ecuador and is im-
plementing a number of forestry projects in Latin America,
has in fact concluded that promoting the sustainable devel-
opment of timber in primary tropical forests is futile. Using
data obtained in a CI project in the Bolivian Amazon, he has
estimated that annual accumulation of commercial mahog-
any stocks in a well-managed stand is 4 percent. In addition,
inflation-adjusted stumpage prices are increasing 1 percent
per annum, which implies that the overall returns for ma-
hogany management are 5 percent (McRae, 1997). Since fi-
nancial instruments yield much more than 5 percent in
Bolivia and many other countries, a resource owner can grow
wealthier by harvesting timber and investing the proceeds
than by practicing sustainable forestry.
Environmentally Sound Timber Production 81

Rather than trying to convince people to do something that

is not in their best interests, Rice argues that it would be
better to let forest owners proceed with selective extraction.
Immediately afterward, land could be purchased at low
prices, reflecting the depletion of standing commercial tim-
ber, and placed in protected reserves. Theodore Gullison, a
tropical forest ecologist at London's Imperial College who
has been working with Rice in Bolivia, suggests that such an
approach might actually have less of an impact on biological
resources than attempts at sustainable mahogany production
do. As others have done, Gullison observes that large open-
ings must be maintained in the forest canopy so that shade-
intolerant mahogany saplings can mature. Furthermore, his
own work in the field leads him to believe that biodiversity
is being diminished due to the thinning of species that com-
pete with mahogany in the same openings (McRae, 1997).
At the February 1997 meeting of the American Association
for the Advancement of Science (AAAS), where Rice and
Gullison presented their findings, the general proposition
that alternatives to the sort of selective logging that one finds
throughout the Western Hemisphere are neither rewarding
for resource owners nor beneficial in terms of enhanced bio-
diversity was sharply contested. For example, Richard Don-
ovan, who directs Smartwood—which is a program that
certifies environmentally sound logging enterprises around
the world—pointed out that mahogany extraction in Latin
America's primary forests is much less intense than logging
operations are in many parts of Asia and Africa, where it is
common for half a dozen species or more to be exploited. He
is sure that giving the latter operations free rein would be to
invite disaster. Donovan also strongly doubts that extensive
tracts of tree-covered land in the developing world can be
placed under effective protection, even with the major infu-
sions of cash that Rice appears to be contemplating (McRae,
1997).
Almost certainly, the prescription not to interfere with se-
lective logging is not universally applicable. Nevertheless, it
probably reconciles quite well with market conditions that
will prevail for many years to come in Latin America's for-
ested hinterlands. Drawing on an analysis of long-term trends
in global roundwood markets, Sohngen et al. (1997) offer pro-
jections of stumpage values in various parts of the world. Of
particular relevance to frontier regions in the tropics is their
finding that timber demand growth, which is expected to be
substantial, will be satisfied primarily through increased out-
82 Economic Returns

put in existing centers of production (e.g., the southeastern

United States and the Pacific Northwest), and through a ma-
jor expansion of tree plantations in the developing world,
including Latin America. More timber will be extracted from
the world's unmanaged forests, in boreal regions as well as
the tropics. However, the latter output will continue to com-
prise a small portion of global production. Moreover, stump-
age prices along the fringes of unmanaged primary forests in
places like the Amazon Basin are unlikely to rise appreciably,
since roundwood harvested in more accessible places, where
management is more intense, will grow nearly as rapidly as
consumption will (Sohngen et al., 1997).
These findings do not bode well for sustainable develop-
ment of timber products other than fine tropical hardwoods
in Latin America's primary forests. Market forces will not
favor the sorts of management improvements that AID tried
to introduce in the Palcazii Valley, and which CI has been
promoting in the Bolivian Amazon. Sohngen et al. (1997) ex-
pect that unmanaged timber extraction will remain the norm.
Without a doubt, local environmental damage, of the sort in
Belize documented recently (Ito and Loftus, 1997), will often
be severe.
The good news, though, is that waste and destruction will
be confined mainly to the fringes of unmanaged forests. Other
than those seeking mahogany and other highly valued spe-
cies in trackless jungles, loggers will operate mainly in the
vicinity of navigable rivers and passable roads. Where infra-
structure remains undeveloped, natural vegetation ought to
remain relatively untouched. The encouragement offered to
the protagonist in the film Field of Dreams remains a clear
warning about the consequences of highway construction in
tropical forests: "If you build it, they will come."
6
Genetic Prospecting

ven though it is not particularly scarce, standing

E timber is the most valuable commercial asset
found in tropical forests. For humankind as a whole, though,
timber is probably of less importance than genetic resources.
Although they occupy little more than 5 percent of the
world's land surface, rain forests near the equator are the
home of at least half of all plant and animal species (Myers,
1984; Wilson, 1988). Many of these species are threatened
with extinction. As mentioned in chapter 1 of this book, My-
ers (1988) has identified several hot spots in Africa, Asia, and
Latin America where tree-covered land rich in endemic flora
and fauna is being encroached on rapidly.
None of the consequences of biodiversity loss is easy to
quantify or to evaluate. It is conceivable that vital ecosystem
functions, like hydrological cycle regulation, could break
down if species diversity fell below some critical threshold.
However, the risk of such an outcome is extremely remote in
most places. By contrast, it is entirely reasonable to worry
that avenues of research needed to cure cancer, AIDS, and
other diseases might be hindered because too much biologi-
cal raw material has perished irreversibly. It is worth remem-
bering, for example, that the rosy periwinkle (Catharanthus
roseus), which is the source of two uniquely effective drugs
for treating leukemia, grows in the forests of Madagascar,
which comprise one of the hot spots that Myers (1988) has
identified.
Convincing though the story of the rosy periwinkle may
be, circulating anecdotes is no substitute for thorough em-
pirical study of the value of tropical forests as a source of

83
84 Economic Returns

inputs to pharmaceutical investigation. Without value esti-

mates, it is hard to determine when and where deforestation
is inefficient. Analysis of what the flow of biological raw ma-
terial is worth to laboratories and other research venues is
also needed to guide the articulation of legal and institutional
arrangements for governing access to genetic information.
The best available evidence, which is surveyed in this
chapter, suggests that there is no strong reason for pharma-
ceutical companies to pay a great deal to maintain supplies
of genetic inputs from tropical forests. To be sure, society as
a whole, either now or in later years, might attach a very high
value to the lives saved because specimens can be gathered
in the wild. But none of this has much of an impact in the
marketplace.
Moreover, forest dwellers are not likely to derive substan-
tial income from whatever medicinal products might be ob-
tained from natural habitats. This is true even when they
grow or collect a substance that, with limited processing, is
consumed directly by people, rather than just being used in
research. Consider the case of the sangre de drago (Croton
spp.) tree, which grows in humid tropical areas throughout
the Western Hemisphere, and which produces a sap that has
been used for generations to cure various maladies. The prod-
uct is starting to be sold in health food stores in Europe. Also,
Shaman Pharmaceuticals Inc., of San Francisco, has been
conducting clinical trials to test the safety and effectiveness
of the substance as a topical treatment for drug-resistant her-
pes. Its usefulness in combating a respiratory virus that af-
flicts children is being investigated as well (Burton, 1994).
As a result of international interest, the sangre de drago
market in Ecuador and several other countries has strength-
ened a great deal. A local entrepreneur who works with Sha-
man Pharmaceuticals reports that a hectare in the Ecuadorian
Amazon with 100 trees (a 10-meter-by-10-meter spacing)
yields approximately 300 liters of sap eleven years or so after
the trees are planted, at a cost of $2.00 apiece, or emerge of
their own accord (Douglas McMeekin, personal communi-
cation, 1994). As of January 1994, the producer-level price of
the sap was $4.25 per liter. If that price were to hold steady
in real terms over time, then the present value of a single
planting and harvesting cycle would be a little less than $250
per hectare, assuming a real discount rate of 10 percent. That
value is comparable to the opportunity cost of land in the
region.
Genetic Prospecting 85

Harvesting a medicinal product like sangre de drago is

about as remunerative as collecting tagua, aguaje, or any
other nontimber product is in a South American forest. Since
the net returns for the latter activity, which are modest, are
examined in chapter 4, this chapter focuses exclusively on
the value of tropical forests as a source of biological inputs
to pharmaceutical research and development. The literature
addressing that value is reviewed. Also described are the dif-
ficulties facing any party, be it an individual landowner or a
national government, attempting to benefit by controlling ac-
cess to habitats used in bioprospecting.

What the Pharmaceutical Industry Might Be Willing

to Pay for Biological Raw Material
Various approaches have been used to evaluate the resources
exploited by genetic prospectors. Laird (1993), for example,
reports that payments for resource samples range from $50
to $200 per dry kilogram. But as Simpson, Sedjo, and Reid
(1996) correctly point out, the people receiving these pay-
ments often have no property rights for the plants they col-
lect, which means that prices would not fully reflect in situ
values. To arrive at the latter, of course, labor and other ex-
penses, which can be sizable, would have to be deducted.
A more commonly used approach for estimating the gross
value of wild genetic resources is to draw on the pharmaceu-
tical industry's experience with medicines derived from
plants. In particular, the probability that an individual spe-
cies will yield something commercially useful (i.e., the suc-
cess rate) is multiplied by the returns associated with
success. The resulting product suggests what an untested
species is worth. But practitioners of this approach, and there
have been quite a few, do not always distinguish between the
gross and net values of a discovery. This exaggerates the re-
turns to success since pharmaceutical research and devel-
opment is usually very expensive. Overestimation also
occurs because a distinction is not always made between av-
erage values of past discoveries and marginal values of new
discoveries. This is a significant oversight because, as Ayl-
ward (1993) stresses, the gap between the two values is large
and probably widening because the value of medicines de-
rived from plants seems to be diminishing.
Farnsworth and Soejarto (1985) were among the first to
base value estimates on industry experience. Farnsworth had
86 Economic Returns

analyzed the origins of drugs prescribed in the United States

from 1959 through 1973, finding that plants were the source
of one or more active agents in 25.4 percent of those drugs.
Multiplying that share by the average price of a prescription
in 1980 and by the number of prescriptions filled each year
in the United States, Farnsworth and Soejarto (1985) con-
cluded that the gross value of all prescriptions derived
wholly or partly from plants was a little more than $8 billion:
25.4 percent: the proportion of all medicines with botanical
ingredients
X 4 billion U.S. prescriptions per annum
X $8 per prescription
$8.128 billion: gross value of plant-derived prescriptions in the
United States.

They also reported that, as of the early 1980s, only 5,000 spe-
cies had been examined thoroughly and that, of that number,
40 were found to contain commercially useful medicinal in-
gredients. Applying an implicit rate of 1 research success per
125 species to one-fortieth of the gross value of plant-derived
prescriptions, Farnsworth and Soejarto (1985) contended
that the worth in the United States of an untested species is
$1.63 million, in 1980 dollars:
1/125 X 1/40 X $8.128 billion =
$1.63 million per untested species.

Principe (1989) carried out a study much like the one by

Farnsworth and Soejarto (1985), but focused on all countries
that belong to the Organization for Cooperation and Devel-
opment (OECD), and not just the United States. Much of the
data used by Principe was the same as the data Farnsworth
and Soejarto had used, although he assumed much lower
success rates, between 1 and 10 successes per 10,000 species.
For a rate of 5 per 10,000, values in the OECD market, which
is three times the size of the U.S. market, turned out to av-
erage a little more than $300,000 per species (Principe, 1989):
0.0005 success rate
X $203 million per R&D success in the United States
X 3 (ratio of the OECD market to the U.S. market)
$304,500 in the OECD per untested species.

This estimate, like the larger figure Farnsworth and Soejarto

(1985) had arrived at a few years earlier, is a flawed indicator
Genetic Prospecting 87

of in situ value because research and development costs were

not deducted. In addition, it really amounts to an average
historical value, not the marginal worth of an additional un-
tested species.
In an evaluation of biodiversity values associated with a
forest protection project in Cameroon, Ruitenbeek (1989,
cited in Aylward, 1993) improved considerably on the no-
tions of economic worth underlying the estimates provided
by Farnsworth and Soejarto (1985) and by Principe (1989).
The values of patented discoveries, rather than the prices
charged for retail drugs, were examined. Aylward (1993) sug-
gests that patent values may not be the best possible indicator
to use in this sort of evaluation, but concedes that new dis-
coveries are an appropriate measure of the output of genetic
prospecting and of pharmaceutical research and develop-
ment. To get at in situ values, specifically, it was assumed
that only 10 percent of the value of new discoveries would
accrue to the country in which the genetic material was orig-
inally found. Assuming that ten research successes, each
worth $7,500, could be expected to result each year because
of project implementation, the benefit to Cameroon would be
$7,500 per annum (Ruitenbeek 1989, cited in Aylward,
1993):
$7,500 per research success
X 10 successes per annum
X 10 percent captured by the country where the sample was
collected
$7,500 in annual biodiversity protection value captured
domestically.
Aylward (1993) observes that, if the forest that is to be pro-
tected by the project contains 500 species, then the average
annual return that the country collects because of species
conservation is $15 per species:
$7,500 X 1/500= domestic benefits of $15 per protected
species per annum.
The approach developed for evaluation of the Cameroon
project has been applied, with modifications, in other set-
tings. For example, Pearce and Puroshothaman (1995) sought
to evaluate the biodiversity losses resulting from tropical de-
forestation. They used the range of success rates that Principe
(1989) had identified—from 1 to 10 per 10,000—as well as
rough estimates that he had also provided of the value of lives
that would probably be lost in the United States because of
88 Economic Returns

the extinction of species with medicinal properties. Suppos-

ing as well that 60,000 plants are at risk in the world's trop-
ical forests, which extend across I billion hectares, Pearce
and Puroshothaman (1995) conclude that, from a U.S. per-
spective, the expected cost of lost genetic resources resulting
from deforestation is between $0.012 and $21.00 per hectare:

60,000 tropical forest plant species at risk

X 0.0001 to 0.001—success rate
X $390 million to $7 billion per success
X 0.5 to 5 percent of the value captured by the host country
1 billion hectares of tropical forests
$0.012 to $21.00 in value of genetic resources per hectare.

A comparably wide range of species values has been es-

timated by Reid et al. (1993). They have modeled the species
screening process in a more detailed fashion than other econ-
omists have done. In particular, they assume that there is a
l-in-10,000 chance that a biotic sample will contain a lead
compound, and that only 25 percent of all such compounds
will end up yielding commercial pharmaceuticals. Notwith-
standing this insight, available data allow them only to con-
clude that the value of untested material is somewhere
between $52.50 and $46,000 per species.
Aylward (1993) reports that more recent studies of biodiv-
ersity values reflect a better understanding of genetic pros-
pecting and pharmaceutical research and development, and
that the sort of conceptual errors that characterized earlier
contributions to the literature are now being avoided. Not
coincidentally, claims about what biodiversity is worth to
drug companies and their customers have grown more mod-
est over the years.
However, Simpson, Sedjo, and Reid (1996) argue that
there is an upward bias in any estimate of the value of an
untested species that is obtained simply by multiplying an
expected success rate times the value of a new discovery,
regardless of how accurately these two factors have been
measured. Such an approach neglects the redundancy that
can exist between one organism's genetic information and
what another organism contains. Among the reasons why re-
dundancy can occur is that the same useful compound might
be found in more than one species. Also, different com-
pounds can have the same curative properties.
Genetic Prospecting 89

Redundancy and the marginal value of untested species

are closely interrelated. As Simpson, Sedjo, and Reid (1996)
point out, a low level of redundancy implies that the chances
are small that examining a given species will reveal anything
useful. But marginal value is also low if the redundancy level
is high since there is little chance of finding something new.
Between these two extremes are levels of redundancy at
which the probability of discovering something that is truly
useful and unique is higher.
Taking redundancy into account, Simpson, Sedjo, and
Reid (1996) offer a theoretical analysis of the linkages among
screening costs, the value of a significant find, and the size
of the sample of unscreened genetic material. Among other
things, they find that, as the sample size increases, the mar-
ginal value of untested material grows small, even if screen-
ing costs are not very high. They also estimate an upper limit
on the value of what they label the "marginal species." Using
data from the pharmaceutical industry, they assume an av-
erage screening cost of $3,600 per sample, and that the av-
erage value of a research and development success is $450
million. Given some assumptions made about the distribu-
tion of useful compounds, the optimal success rate turns out
to be 0.000012. At this rate, there is a 95 percent chance that
a collection of 250,000 samples will yield a success, and the
marginal value of the untested species is $9,431. However,
this value is highly sensitive to the success rate. For example,
if the rate falls by one-third, to 0.000008, then the marginal
value dwindles to zero. The marginal value also declines if
the success rate rises above its optimal level. At a rate of
0.00040, for instance, an untested species is worth only $67.
Simpson, Sedjo, and Reid (1996) show that the marginal
value is also highly sensitive to sample size, the returns as-
sociated with a success, and other variables. In addition, they
address the costs resulting from encroachment on natural ec-
osystems by combining estimates of the marginal value of
untested species with a model of island biogeography (Mac-
Arthur and Wilson, 1967), which relates species extinction
to habitat loss. They find that western Ecuador is the place
where the pharmaceutical industry would be willing to pay
the most to halt the conversion of forests into cropland and
pasture. Myers (1988) considers the same region to be among
the world's most active hot spots of threatened biodiversity
since a huge number of species, many of them endemic, con-
tinue to survive there even though cumulative deforestation
90 Economic Returns

has reached an advanced stage. However, Simpson, Sedjo,

and Reid's (1996) maximum estimate of the marginal value
to the pharmaceutical industry of habitat protection in west-
ern Ecuador, $20.63 per hectare, is less than a tenth of what
the region's farmers and ranchers routinely pay for cleared
land (see chapter 2).
The remnants of natural habitats on three islands, where
levels of species endemism are high, also feature relatively
high biodiversity values at the margin: southwestern Sri
Lanka ($16.84 per hectare), New Caledonia ($12.43 per hec-
tare), and Madagascar ($6.86 per hectare). Next come three
hot spots with biodiversity values just below $5.00 per mar-
ginal hectare: India's Western Ghats ($4.77), the Philippines
($4.66), and the Atlantic Coast of Brazil ($4.42). Elsewhere,
saving one forested hectare is worth a couple of dollars or
less to the pharmaceutical industry (Simpson, Sedjo, and
Reid, 1996).
The industry certainly behaves as if tropical forests are not
worth very much, in terms of the commercial (as opposed to
the social) value of medicines they might eventually yield.
Much has been made of the agreement that Merck and Com-
pany signed in September 1991 with Costa Rica's National
Biodiversity Institute (INBio), under which the pharmaceu-
tical manufacturer pledged to pay the latter agency $1 mil-
lion per annum in exchange for plants and other raw
materials, plus royalties on useful products derived from
those inputs (Harvard Business School, 1992). However, one
might speculate that Merck's interest in the deal had more to
do with a desire to win favorable publicity than with sample
collection, pure and simple. In any event, the amount of com-
pensation involved is negligible in relation to what many
commentators claim biodiversity values to be. (INBio has
signed similar agreements, with comparable payments in-
volved, with a few other companies.)
Other firms have tried to make similar arrangements in
other countries. For example, in early 1995, Pfizer Inc. began
negotiating access to Ecuador's biodiversity with the Ecu-
adorian Institute of Forestry, Natural Areas, and Wildlife
(INEFAN). Originally, the company proposed to spend
$998,000 on the acquisition and management of three small
holdings and on a laboratory at which sample extracts would
be prepared for export to foreign research and development
facilities. INEFAN requested that the funds, instead of being
used to buy land, be made available for the management of
existing public parks and nature reserves. Also, both parties
Genetic Prospecting 91

agreed that a 1 percent royalty would be paid for any pat-

ented veterinary drugs derived from Ecuadorian materials
and a 2 percent royalty for human medicines (Roberto Ulloa
and Joseph Vogel, personal communication, 1996).
Pfizer stood to gain less public acclaim for saving species-
rich environments in 1995 than Merck had received four
years earlier. Pfizer even received a great deal of unfavorable
publicity, from the radical fringes of the environmental
movement in Ecuador and around the world. In response, the
firm decided in December 1995 to break off negotiations,
hinting that it would shift its operations to Brazil.

Controlling Access to Species-rich Habitats

As the actions being taken by INBio, INEFAN, and their coun-
terparts in other countries demonstrate, national govern-
ments in the developing world are attempting to control
access to diverse genetic resources indigenous to the tropics.
INBio, for example, is trying to become a monopoly supplier
of the biological samples that Costa Rica ships to laboratories
around the world.
This interest in biological diversity contrasts sharply with
the total lack of regard for it exhibited by national govern-
ments not so very long ago. For example, the late Julian Stey-
ermark, a U.S. citizen and a longtime Venezuelan resident,
put together and edited a comprehensive multivolume study
of Venezuela's plant life for the country's botanical institute;
which is a branch of the Directorate of Renewable Natural
Resources (DRNR). (When the work was published [Lasser,
1974], the institute's director, not Steyermark, was listed as
the series editor.) In 1980, the DRNR was transferred to a
newly created environmental ministry. Although this reor-
ganization was supposed to signal the national government's
enhanced environmental commitment, Steyermark's inven-
tory of genetic resources was soon mispalaced, and did not
emerge from a government warehouse for several years (How-
ard Clark, personal communication, 1996).
Although episodes like this cause one to doubt that de-
veloping country governments should be designated as the
sole custodians of the bounteous genetic resources that rain
forests and other tropical habitats contain, this is exactly
what is now being established. Local communities' owner-
ship of traditional environmental knowledge is recognized
in Article 8J of the 1992 Global Convention on Biodiversity.
But the same treaty states that national governments have
92 Economic Returns

sovereign rights over germplasm and its derivatives (e.g., pro-

teins and alkaloids).
The legal and institutional structures that need to be
erected for such a regime to operate effectively are monu-
mental. Vogel (1994), who favors applying the existing sys-
tem of intellectual property rights, including 15 percent
royalties, to genetic material, acknowledges that a "gargan-
tuan" data base would have to be developed and maintained
for the expanded system to work. Such a data base would
comprise far more than simple botanical and zoological in-
ventories. In particular, detailed information on the spatial
incidence of each species would have to be included, so that
royalties generated by any plant-derived drug could be dis-
tributed fairly.
As Vogel (1994) emphasizes, nothing less than this sort of
system would suffice for the emergence of robust markets for
genetic information. Such markets, of course, would generate
the sorts of price signals required for the efficient develop-
ment of that information. Likewise, a well-articulated data
base of the sort he describes would be a necessary, though
not a sufficient, condition for the successful functioning of a
cartel, made up of governments in tropical countries, that
would control access to biodiversity in the wild—such a car-
tel, Asebey and Kempenaar (1995) indicate, is a possibility.
Investing in the detailed mapping of the world's biodiv-
ersity resources, so that a royalty system or a cartel can work,
does not seem viable at present. The social benefits associ-
ated with using medicines derived from species collected in
the wild might be substantial, reflecting, as they do, human
beings' willingness to pay a great deal to extend life spans
and to escape the ravages of disease. But until it can be con-
clusively demonstrated that those benefits exceed the legal
and institutional costs of a royalty system or a cartel, neither
of these two arrangements will be economically practical.

The Risk of Counting on Riches from Bioprospecting

Along with a film or two, a number of books and articles extol
the medicinal wonders that come from plants and other bi-
ological resources found in tropical rain forests and known
to people who live in this setting. Mark Plotkin, an ethno-
botanist, has been particularly energetic about documenting
what indigenous shamans in the American tropics know
about preparing skin rash treatments, the curare extract used
Genetic Prospecting 93

as a muscle relaxant, aphrodisiacs, and other products made

from materials found in the jungle (Plotkin, 1994).
Plotkin contends that the pharmaceutical industry could
come up with products of enormous value to all humankind
by tapping forest dwellers' environmental knowledge; and
the public imagination is captivated by stories of lone sci-
entists working with tribal people to bring a cure, for AIDS
or cancer, out of the rain forest, often doing this in spite of
the depredations of loggers and miners. The tale of the rosy
periwinkle, which nearly everyone who speaks or writes
about biodiverse tropical forests feels obliged to repeat, is an
essential part of this lore.
But the reality of pharmaceutical research and develop-
ment appears to be quite at odds with romantic ideas about
ethnobotany. Rather than sending scores of experts, either
anthropologists who know about biology or biologists who
are familiar with native cultures, into the field with video
recorders, industry has chosen to invest heavily in the labo-
ratory facilities needed for rapid assessment of large numbers
of samples, many of which are microbial as opposed to being
actual plants and most of which are collected without con-
sulting local people. For example, Amgen Inc. boasts that it
is now applying "newly emerging techniques in robotics
and miniaturization" so that it can "synthesize and test
thousands of potential drug compounds in a fraction of the
time that would have been required in the recent past" (1997,
page 17). These innovations comprise the cornerstone of a
company strategy to have it evolve from a large biotechnol-
ogy enterprise, with sales of two products accounting for
more than 85 percent of total revenues, into a major player
in the pharmaceutical industry.
Fundamental industry trends, as exemplified by the direc-
tion Amgen is taking, could be depriving humankind of cures
for various diseases insofar as lines of ethnobotanical inves-
tigation that Plotkin (1994) has mapped out remain unex-
ploited. But accepting industry trends as a given, one must
concede that the models which economists like Aylward
(1993) and Simpson, Sedjo, and Reid (1996) have developed
and applied are reasonably accurate, as are their estimates of
the marginal commercial values of untested species and spe-
cies-rich habitats.
Those estimates suggest that, as a rule, genetic prospecting
will not create a financial bonanza for countries with species-
rich habitats. Assuming a 10 percent discount rate and a
94 Economic Returns

l-in-10,000 success rate in pharmaceutical research and de-

velopment, Aylward (1993) finds that the present value of
royalty revenues provided by Costa Rica's 600,000 hectares
of parks and reserves, which can be expected to yield 1,000
samples a year for testing, is approximately $4 million. This
benefit estimate, which is consistent with what Simpson,
Sedjo, and Reid (1996) have found concerning tropical hot
spots' bioprospecting values, compares poorly with the op-
portunity costs of the same 600,000 hectares, which exceed
$200 million (Aylward, 1993).
Certainly, the relationship between benefits and costs
would be greatly altered if a "blockbuster" plant, like the rosy
periwinkle, were discovered in a Costa Rican park. However,
counting on such an event, which would have to be consid-
ered an exceptional case, is hardly a prudent financial basis
for a habitat conservation strategy.
7
Nature-Based Tourism

f all the economic alternatives contemplated for

O threatened habitats in the developing world,
none appear to hold as much commercial promise as tourism.
The business of accommodating people who wish to expe-
rience different planes firsthand has grown rapidly and, since
the late 1980s, has accounted for 7 percent of all international
trade in goods and services (Whelan, 1991). Formerly the es-
oteric pursuit of a few people, vacations taken for the purpose
of viewing and learning about exotic flora and fauna in their
natural settings now comprise an important part of the tour-
ism market in many places (Boo, 1990).
Dramatic ecotourism growth in Costa Rica and the Gala-
pagus Islands, which are the geographic focus of this chapter,
has not been entirely free of adverse environmental conse-
quences. Mountfort (1974) reports that careless photogra-
phers in the Galapagos occasionally interfere with the
breeding of birds; and de Groot (1983) complains of people
chasing marine iguanas (Amblyrhynchus cristatus), who
must lie still under the tropical sun to recover body heat after
emerging from the cold ocean. In addition, ships and boats
that carry tourists around the islands routinely discharge gar-
bage and sewage. Even at Costa Rica's Monteverde Cloud For-
est Biological Preserve, which many reckon is one of the
best-run protected areas in Latin America, soil sometimes
erodes along visitors' trails. Also, tourists occasionally tap on
the nests of resplendent quetzals (Pharomachrus mocinno),
which are rare and finicky birds, to get them to take flight
(Rovinski, 1991).

95
96 Economic Returns

Except when threatened species, like the quetzal, are dis-

turbed, impacts such as these are very localized, since the
nature-based tourism sector does not make use of extensive
tracts of land. For example, Huber (1996) reports that the total
area managed by ecotourism enterprises close to Manaus,
which is a mecca for foreign visitors to the Brazilian Amazon,
is 4,000 hectares. This is a very small share of all the land in
the region with ecotourism potential.
But if the environmental damages resulting from nature-
based tourism are not of great concern, its local economic
impacts are not very significant, either. It is almost always
the case that, of all the money that visitors spend, only a
small portion reaches local communities. Admonitions to in-
crease local benefits, through planning, consultation, and
other measures, are standard fare in the literature (Boo, 1990;
Drake, 1991). But notwithstanding a few ballyhooed success
stories, those measures have not been very effective in Latin
America. Nor is it readily apparent that making the invest-
ments needed to enhance the local economic impacts of na-
ture-based tourism would truly be efficient.
Local communities are not the only interest claiming more
of the money spent by foreign visitors. In Costa Rica, the Ga-
lapagos, and elsewhere, maintenance and protection activi-
ties in the natural habitats that make ecotourism possible fall
well short of what is required. In recent years, some of the
money used to run parks and reserves, and to hire the guards
needed to discourage encroachment, has been raised through
debt-for-nature swaps and other mechanisms. Another way
to put parks and reserves on sounder financial footing is to
raise the fees paid by tourists and by the firms that serve them
(Dixon and Sherman, 1990).
It is generally conceded that access to protected areas has
been underpriced throughout Latin America, as it has been
in many other parts of the world. However, caution must be
exercised when fees are being readjusted for sites that have
close substitutes. As David Simpson, an economist at Re-
sources for the Future who has studied bioprospecting values
in the developing world, puts it, "Every nice little hill and
valley isn't going to be able to charge a lot for admission,
since there are so many nice little hills and valleys that one
could visit" (personal communication, 1996).
Simpson's warning is particularly relevant for those places
in Latin America where price discrimination is being at-
tempted. The potential gains from such a policy are illus-
trated in Figure 7.1, which shows the marginal costs (MC),
Nature-Based Tourism 97

Figure 7.1 Price discrimination: The two-group case.

including environmental damages, associated with, total pa-

tronage (Q) of a hypothetical park, and both national demand
(DN) and foreign demand (DF) for access to that venue.
If everyone is charged the same entrance fee, Ps, then do-
mestic and foreign visits will amount to Qns and QFS, respec-
tively. Total costs will consist of the area under the MC curve
between 0 and Qg (i.e., the sum of QNS and QFS). These costs
will be less than total revenues, denned either as Ps times Qg
or as the sum of Ps times QNS plus Ps times QFS.
But with price discrimination, a price is set in each market
by finding the level of visits at which marginal revenues (MR)
equal (a lower) MC. As shown in Figure 7.1, revenues col-
lected in the domestic market are not much affected: Ps times
QNS is about the same as PND times QND. However, price dis-
crimination results in more revenues being collected from
foreigners, for whom demand is assumed to be less elastic;
as can be seen in Figure 7.1, PFD times QFD greatly exceeds Ps
times QFS. At the same time, costs are lower if price discrim-
ination is practiced since total visits are consequently dimin-
ished.
As can be readily appreciated from this conceptual dis-
cussion, the viability of price discrimination depends on
there being significant differences among various groups'
elasticities of demand for access to a park. There are some
places where such differences exist, the Galapagos and the
Monteverde Preserve in Costa Rica being prime examples.
But elsewhere, national park services are attempting to apply
the pricing strategy with little or no idea of what the true
relationship between entrance fees and visits is. The results
of the strategy can be disappointing.
98 Economic Keiuma

The Ecotourism Boom in Costa Rica

Few places in the world are in a better position than Costa

Rica (Figure 7.2) is to benefit from the interest that people
have in visiting tropical habitats. The country is compact,
with a land area of just 51,000 square kilometers. Within its
borders, one can experience a wide variety of environments,
ranging from beaches to jungles to mountains. Costa Rica is
also home to a uniquely diverse mix of flora and fauna since
it sits on the land bridge connecting North and South Amer-
ica. As of the early 1990s, 850 bird species and 208 mammal
species had been identified in the country, as had countless
insects and plants (Umafia and Brandon, 1992).
A serious effort has been made to promote travel to Costa
Rica, which is easy to reach from North America or Europe.
The response, as indicated by growing international arrivals,
has been impressive, and tourism now makes a sizable con-
tribution to the national economy.
Problems arise because all but a few of the communities
near national parks and private reserves gain little from visits

Figure 7.2 Costa Rica.

Nature-Based Tourism 99

made by hikers, bird-watchers, and others. Moreover, the

continued success of ecotourism, which accounts for an ap-
preciable share of international arrivals to Costa Rica, re-
quires major improvements in transportation infrastructure
and protected areas. The national government has made
imaginative use of various financing mechanisms and has
raised park entrance fees, especially those paid by foreigners.
However, the issue of how to fund improvements is not any-
where close to being fully resolved.

Tourism's Recent Expansion and

Economic Significance
Long-term trends in Costa Rican tourism cannot be plotted
with a high degree of precision, mainly because of how peo-
ple who arrived from neighboring countries during the 1970s
and 1980s opted to describe the purpose of their visit to im-
migration officials. Many Central Americans, of course, were
trying to escape civil strife and economic dislocation, but
most claimed that tourism was their reason for spending time
in Costa Rica.
There is no doubt, however, that arrivals from the United
States and other wealthy nations, which had been growing
slowly for several years, began to skyrocket in the late 1980s.
As armed hostilities in Nicaragua and El Salvador drew to a
close, North Americans and Europeans no longer sensed that
a trip to Costa Rica involved any great danger. As a result,
many more of them chose to vacation in the country. Indeed,
1989 was the first time when Canadian, U.S., and Mexican
visitors first outnumbered the combined total of visitors ar-
riving from Central America (see Table 7.1).
Costa Rican tourism continues to expand, albeit at a more
modest pace. Since 1992, when arrivals from the United
States were 23 percent higher than what they had been a year
earlier, growth rates have fallen, from 14 percent in 1993 to
11 percent in 1994. Data collected by the Costa Rican Tour-
ism Institute (ICT) indicate that the number of foreign air
passengers landing in San Jose from January through Septem-
ber of 1995 was only 3.7 percent higher than arrivals for the
first nine months of 1994.
Even though tourism might be growing more slowly, the
sector continues to be a mainstay of the national economy.
In 1994, the money spent in Costa Rica by international vis-
itors was equivalent to 28 percent of the value of all the coun-
try's exports. Indeed, neither the value of the bananas sold
100 Economic Returns

Table 7.1 Number of Foreign Tourists in Costa Rica, 1985 through

1994
Year Canada, U.S., and Mexico Central America Europe Total

1985 89,825 112,623 28,179 261,552

1986 93,105 106,825 29,026 260,840
1987 104,841 108,543 32,354 277,861
1988 123,551 124,728 41,396 329,386
1989 153,112 135,376 45,355 375,951
1990 191,284 139,913 57,177 435,037
1991 223,126 164,809 67,319 504,649
1992 274,061 187,790 88,301 610,591
1993 302,741 193,512 113,943 684,005
1994 332,602 221,384 129,580 761,448

SOURCE: Data from ICT (1995): 33.

overseas ($522 million) nor of coffee exports ($300 million)

exceeded the $623 million in foreign exchange that tourism
generated (ICT, 1995).

Importance of Nature-Based Tourism

Most international visitors choose to be in Costa Rica in De-
cember, January, February, or March, when precipitation is
at an ebb. There are also a large number of arrivals in July
and August, when North Americans and Europeans take a
summer break (ICT, 1995). Especially during the dry months,
when country roads are more passable, one can choose
among a wide array of options for recreation.
Surveys reveal that, during the 1995 peak season, 82 per-
cent of all visitors from the United States, which is Costa
Rican tourism's primary market, were in the country on a
tourist visa. Of this group, 76 percent went to the beach; 28
percent engaged in surfing, diving, and other water sports;
17 percent fished; and 11 percent went river-rafting. Many
participated in one form of ecotourism or another: natural
history (38 percent), bird-watching (35 percent), photo-
graphic tours (7 percent), and "tropical adventures" (38 per-
cent). Some peak-season tourists from the United States also
took advantage of their time in Costa Rica to learn Spanish,
to attend a conference, to do some business, or to take in a
museum (ICT, 1994).
The preceding data indicate only what foreigners end up
doing while in Costa Rica, not their motivations for traveling
Nature-Based Tourism 101

there in the first place. The typical vacationer engages in

more than one kind of activity, and it is important to gain
other perspectives on the degree to which the recent tourism
boom has been driven by visitors' desires to see rain forests,
to watch birds, and so forth.
One alternative indicator of ecotourism's importance is
the number of visits to protected areas, which rose dramati-
cally in the late 1980s and early 1990s. As one official of the
National Park Service (SPN) has documented, 404,342 for-
eigners entered Costa Rica's national parks in 1993, which
was about five times the number recorded six years earlier
(Bermudez, 1992 and 1995, cited in Chase, 1995). There has
been a decline during the past couple of years, in part be-
cause of entrance fee increases. However, patronage remains
much higher than what it used to be. Rovinski (1991) and
Aylward et al. (1996) report that similar changes have oc-
curred over time in visits to privately owned areas, like the
La Selva Biological Field Station and the Monteverde Pre-
serve (see Figure 7.2).
Aylward et al. (1996) observe that the purpose of many
of the visits to Costa Rican national parks might not be con-
sistent with some of the narrower definitions of ecotourism
that are used by some researchers and activists. The
second-smallest park, Manuel Antonio, which only takes
in 683 hectares, is consistently one of the most heavily
used protected areas (it drew 94,102 international and
33,921 national visitors in 1994), mainly because it
boasts spectacular beaches and because there are dozens
of hotels and inns in the vicinity. The same year, 143,822
foreigners and 181,448 Costa Ricans went to Volcan Poas
and Volcan Irazu. Although these two sites contain high-
altitude cloud forests that are important for watershed
protection and interesting to biologists, they attract tour-
ists almost exclusively because each offers views of an active
volcano.
Subtracting the patronage of Manuel Antonio, Volcan
Poas, Volcan Irazu, and a few other sites, Aylward et al.
(1996) estimate that there were 73,000 paid visits by foreign
ecotourists to Costa Rican national parks in 1992, when
330,000 entries were recorded for the system as a whole.
They contend that, when use of privately owned sites is taken
into account, approximately 100,000 foreigners—one in
every four vacationers from the United States, Canada, and
Europe (ICT, 1995)—participated in nature-based tourism in
Costa Rica in 1992.
102 Economic Returns

Local Economic Impacts

Hard evidence concerning the local, as opposed to national,
economic impacts of ecotourism is spotty. Enthusiastic
claims are sometimes made about the benefits flowing to in-
dividual communities. But by and large, such claims are ex-
aggerated.
It is widely accepted that the typical international vaca-
tioner spends a little more than $2,000 getting to, around, and
back from Costa Rica. For example, a survey of 575 foreign
visitors to the Monteverde Preserve, carried out in 1991 by
the TSC, which owns and manages the site, revealed total
expenditures of $2,207 per person, of which $1,273 were
made somewhere within the country. For those individuals
who indicated that seeing Monteverde was a prime motiva-
tion for being in Costa Rica, average overall and within-the-
country expenditures were $1,961 and $1,131, respectively
(Aylward et al., 1996).
Monteverde is indeed one place where ecotourism makes
a significant contribution to the local economy. For the most
part, relations are good between the preserve and the neigh-
boring community, which Quakers from Alabama estab-
lished in 1949. In addition, the town of Monteverde is
pleasant, with clean restaurants and comfortable hotels, and
the majority of foreign and national visitors to the area choose
to eat their meals and to sleep there. Based on the finding,
from the 1991 TSC survey, that the average duration of a for-
eigner's visit is three days, and using a recent ICT estimate
of international tourists' average expenditures ($86 per day),
Figueroa (1995) concludes that local spending by the 27,748
foreign visitors in 1994 was nearly $7,160,000. He also points
out that more than half of what those visitors paid to enter
the preserve and for souvenirs at its gift shop—$616,111 in
1994—goes directly to wages for local workers and to pur-
chases of goods in the surrounding area.
But as a rule, money spent close to parks and reserves com-
prises a small portion of tourists' expenditures in Costa Rica.
It is interesting to note that, notwithstanding the recent ex-
pansion of ecotourism, very little new lodging capacity has
been constructed in the vicinity of ecotourism destinations.
An ICT official reports that the spatial distribution of the
country's hotel rooms is as follows: 31 percent in San Jose,
21 percent in other urban areas, and 31 percent along the
Pacific or Caribbean coasts (Martin Quesada, personal com-
munication, 1995). Even where there are lodging facilities of
Nature-Based Tourism 103

the sort that most foreign tourists (or at least those who are
prepared to spend more than token sums of money) would
care to patronize, those facilities are often self-contained,
providing direct benefits to nearby communities only in the
form of the low wages paid to waiters, maids, and other un-
skilled laborers.
For some protected areas, local economic impacts are
practically nonexistent. For example, nearly everyone who
visits Volcan Poas or Volcan Irazu does so on a half-day ex-
cursion from San Jose, which any hotel or guesthouse can
easily arrange. Most visitors purchase little or nothing from
people living around the park. Even at La Selva, the local
returns are not all that great; Rovinski (1991) reports that the
13,000 individuals who visited the station in 1989 spent just
$291,000 in nearby settlements.

Challenges to Continued Success

Now that international arrivals are no longer increasing as
rapidly as they did a few years ago, impediments to the future
expansion of Costa Rica's tourism sector, generally, and of
ecotourism, specifically, are being given close scrutiny. If
more tourists do not choose to visit the country and its parks,
the chances are small that local economic benefits can grow
appreciably.
Some barriers to expansion have to do with economic de-
velopment in the broadest sense. Particularly outside the
central valley where San Jose and most other cities are lo-
cated, the quality of major thoroughfares and secondary
roads is not what it should be. For example, the Panamerican
Highway between the valley and Punta Arenas (see Figure
7.2), which is Costa Rica's main port on the Pacific Ocean, is
a winding, bumpy, two-lane affair. Its only virtue is that it
helps to steel the traveler for what awaits him or her on the
byways that serve rural areas. These roads are not much bet-
ter than country lanes in the poorest parts of Latin America.
Just as they are a constant subject of lament among Costa
Rica's farmers, rural roads that are muddy during the rainy
season and rutted in dry months are roundly criticized by
nearly every travel writer who visits the country (Ottaway,
1995).
Crime is another worry for foreign visitors. In 1994, nearly
4,000 international tourists were robbed in Costa Rica and a
few suffered violent attacks of one sort or another. In January
1996, the kidnapping of two women—a German tourist and
104 Economic Returns

the Swiss owner of a travel agency in San Jose—near Puerto

Viejo attracted global press coverage; along with demanding
a $1 million ransom, the perpetrators threatened to attack a
U.S. family and insisted on freedom for individuals impris-
oned after the 1993 seizure of the Costa Rican supreme court
building. Since the criminal targeting of foreign tourists
could weaken the country's reputation for being Central
America's lone haven of peace, policing is being improved
in the capital city and elsewhere.
Habitat conservation is still another concern for the eco-
tourism industry. In past years, international development
agencies and private organizations based in the United States
and in other wealthy nations have provided large sums to
Costa Rica for expansion and improved management of na-
tional parks and private reserves. Additional projects are in
the pipeline. However, the country faces increased compe-
tition for funds from other parts of the world, including its
Central American neighbors, where systems of protected nat-
ural areas are much less developed.
At the same time, the demands inherent in managing ex-
isting national parks appear to be straining the budgetary and
human resources that have been allocated to the SPN. A
benchmark for evaluating the adequacy of those resources is
Monteverde. Over the years, the preserve has expanded to
approximately 10,000 hectares. In 1994, the TSC employed
38 people there and spent more than 80 million colones
($509,327) on management (Figueroa, 1995). (This amount
does not include purchases of merchandise for a souvenir
shop and a refreshment stand.) By contrast, in 1995 the SPN
assigned 58 employees and allotted 35 million colones
($198,999) to a region in the central highlands that takes in
Volcan Irazii (2,309 hectares), Volcan Poas (5,600 hectares),
two other national parks (with a combined area of 60,157
hectares), and a small national monument.
Support for the SPN and partner institutions in the public
and private sectors will have to be increased substantially if
proposals to expand protected areas in Costa Rica are
adopted. For example, a team of consultants that worked for
the World Bank has concluded that management of existing
parks is successful in the sense that conspicuous fauna, with
the exception of some large birds and a few other species,
seem to be thriving. However, they advocate a 50 percent
increase in the area under the direct control of the SPN, with
special emphasis to be placed on the acquisition of lowland
properties and on the protection of corridors among existing
parks so as to allow for the unmolested migration of wildlife
Nature-Based Tourism 105

(DHV Consultants BV, 1992). Even more ambitious would be

the full implementation of the scheme to create nine conser-
vation areas around the country, each to contain one or more
national parks and to be locally administered (Umana and
Brandon, 1992].
That proposals of this sort are being given serious consid-
eration suggests that the recent growth in ecotourism has
made habitat conservation a much higher priority in Costa
Rica than it would otherwise have been.

Pricing Issues

Over the years, many innovative approaches to financing

habitat protection have been applied in Costa Rica. The coun-
try was one of the first to take advantage of debt-for-nature
swaps. Through 1991, nearly $80 million of the country's
foreign debt had been converted into $42 million in local
currency bonds to benefit national parks and private reserves,
at a cumulative cost of $12 million (Umana and Brandon,
1992). Endowments to support the management of some pro-
tected areas have been established as well.
Since the very beginning of Costa Rica's ecotourism boom,
economists have argued that habitat protection could be
partly financied by raising park entrance fees. Several studies
show that visitors' willingness to pay for access to protected
habitats exceeds the nominal admission prices charged in
September 1994. The fee increases that went into effect at
that time seem to have been predicated on the assumption
that demand for access to the country's national parks is
price-inelastic across all major groups of visitors, especially
foreigners who are now being asked to pay the most.
Baldares and Laarman (1990) made one of the first at-
tempts to estimate the value that people place on visits to
protected areas in Costa Rica. In a survey conducted by the
SPN in 1989, when the entrance fee was 25 colones ($0.31),
860 visitors to Manuel Antonio, Volcan Poas, a national park
on the Caribbean coast, and the Monteverde Preserve were
asked what they thought daily payments should be. There
was support for charging international visitors more than
Costa Ricans were being charged, although foreigners pre-
ferred a smaller gap. Both groups favored raising the price
for Costa Ricans to 50 colones and the international charge
to 100 colones or so (Baldares and Laarman, 1990).
Two studies have addressed the value that visitors place
on the Monteverde Preserve. In a travel-cost analysis, Tobias
and Mendelsohn (1991) found that visitors' willingness to
106 Economic Returns

pay for access to the site amounted to about $12.5 million,

which was more than an order of magnitude greater than
what all of them paid for admission. Using data collected in
the 1991 TSC survey mentioned above and contingent valu-
ation techniques, Echevema, Hanrahan, and Soldrzano
(1995) estimated that Costa Ricans were willing to pay $137
apiece, besides admission charges, to keep the reserve intact.
On average, international visitors, who gain less from \vhat-
ever watershed protection and local climatic stabilization re-
sults from conservation of Mont ever de's forests, were willing
to pay $119 apiece for the reserve's continued existence.
Findings such as these helped to justify the fee increases
for foreigners that the SPN adopted in September 1994. The
uniform daily price, 200 colones ($1.25), did not go up for
Costa Ricans, but charges for international guests were raised
markedly. The top fee, paid by those simply showing up at
a park gate, was $15.00, paid in either dollars or colones,
while someone making arrangements a day ahead of his or
her visit was charged $10.00. Travel agencies were allowed
to purchase tickets for $5.00 apiece. No discounts were made
available to international students, which probably had a big
impact on visits to privately owned sites, like the Monteverde
Preserve.
Needless to say, a brisk trade in discounted tickets
emerged quickly. Also, this evasion of the maximum fee was
tolerated by some park administrators, particularly those re-
sponsible for areas that received relatively few visitors before
and after the price increase (Chase, 1995). But the policy
change still had a major effect on park use. During the pre-
vious peak tourism season, from December 1993 through
March 1994, paid international visits had totaled 199,408; by
contrast, only 113,461 foreigners were admitted to national
parks during the next peak season, immediately after the
price increase.
That a fourfold-to-twelvefold increase in daily admission
charges for foreigners caused visits to decline only by 43 per-
cent seems to indicate that international demand for access
to Costa Rica's national parks is price-inelastic and, hence,
the policy change raised revenues for the SPN. But Chase
(1995) cautions that there might be a greater response to
higher entrance fees in the long term. In particular, at least
some tourists can be expected to choose destinations other
than Costa Rica, and growth in SPN revenues might not be
as spectacular as what some might have been expecting.
Aylward et al. (1996) contend that international tourists
have not waited very long to revise travel plans. Since ad-
Nature-Based Tourism 107

mission prices at the Monteverde Preserve—a token amount

for Costa Rican students, $1.50 or so for citizens and resi-
dents of the country, $4.00 for foreign students, $8 for for-
eigners not on a package tour, and $16 for international
participants in a tour—were not adjusted in the wake of the
September 1994 increases in national park entrance fees, it
should have experienced a major increase in visits. Indeed,
from January through April of 1995, 9 percent more foreign-
ers came to the area than those who came during the first four
months of 1994. However, this was mainly because there was
an 84 percent rise in visits by international students (who, to
repeat, received no discount from normal admission prices
at the national parks), which more than compensated for si-
multaneous reductions in the numbers of foreigners paying
the $8 or $16 fee.
The impacts of park entrance fees on visits to Monteverde,
and all that they might imply about substitute relationships
between ecotourism in Costa Rica and ecotourism elsewhere,
should not be exaggerated since prices for lodging and other
services have been going up throughout Costa Rica as well.
In addition, Chase's (1995) econometric analysis suggests
that there is considerable scope for using price variations to
direct tourists away from heavily used sites, like Manuel An-
tonio and Volcan Poas, to areas that could accommodate ad-
ditional patronage.
However, the point that international tourists are cost-
conscious should never be forgotten. The possibility that
price increases might be causing more than a few interna-
tional tourists to stay away from the country and its parks
has not been ignored by the SPN, which modified its fee
schedule in July 1995. Foreigners not buying tickets ahead
of time will continue to pay $15.00 and the advance-
purchase tickets to heavily visited sites will still cost $10.00;
however, the price of advance-purchase tickets to other parks
has been lowered to as little as $5.00.

Conservation, Tourism, and Local Interests

in the Galapagos
As in Costa Rica, habitat protection and park pricing issues
have become more important as tourism has expanded in the
Galapagos (see Figure 7.3), the group of twenty-two islands
and scores of smaller land formations, all of recent volcanic
origin, 1,000 kilometers west of mainland Ecuador.
With a combined area of 800,000 hectares, the archipelago
is home to a unique and largely endemic mix of flora and
108 Economic Returns

Figure 7.3 The Galapagos.

fauna. Among noteworthy creatures are the gigantic land tor-

toises (Geochelone elephantopus) for which the islands are
named, prehistoric-looking marine iguanas, various birds,
and sea lions (Zalophus californianus) and fur seals (Arcto-
cephalus galapagoensis), all of which show little or no fear
of humans. About 60 percent of the native plant and animal
species live nowhere else in the world (de Groot, 1983).
Of course, anyone who has taken a biology class knows
about the five weeks that Charles Darwin spent in the Gala-
pagos in 1835. The observations he made there eventually
flowered into his theory that new species arise because of
natural selection (Darwin, 1859). The islands continue to be
an ideal natural setting for studying evolution. The thirteen
forms of Darwin's finches (Geospiza spp.), for example, pro-
vide textbook examples of rapid adaptation to variations in
the environment (Weiner, 1994).
Even by the time of Darwin's excursion, though, endemic
species were in jeopardy. It was the misfortune of adult tor-
toises, which have no natural predators, that they were able
to stay alive for up to a year in a ship's hold. Since there were
few other ways to have fresh meat during a long voyage, it
was a common practice for whaling ships and other vessels
Nature-Based Tourism 109

to pass by the Galapagos to pick up a few large specimens.

Intensive tortoise hunting took place for several decades dur-
ing the nineteenth century, which caused some island pop-
ulations to disappear.
For the most part, the threat to endemic plants and animals
has involved less overt action by people. Mainland organ-
isms, which easily supplant or prey on native flora and fauna,
probably began to establish themselves in the archipelago not
long after humans arrived on the scene, in the 1530s. The
widely circulated notion that goats and pigs were introduced
by sailors hoping to guarantee a food supply during subse-
quent visits is probably apocryphal. However, it is undenia-
ble that domestic animals that have become feral and rats
have adapted readily to local conditions. In addition, indig-
enous vegetation is losing ground to plants brought in, pur-
posefully or inadvertently, from the mainland (Jackson,
1990).
The introduction of nonnative species, which continues to
this day, is the lasting legacy of sporadic attempts by people
to settle in the Galapagos (Southgate and Whitaker, 1994).
Island ecosystems have been thrown far off their equilibrium
as a result and, if the archipelago were simply left alone,
many more indigenous organisms would become extinct. To
avoid the outcome of a desolate landscape inhabited mainly
by burros, goats, rats, cats, and dogs, rehabilitative action,
which is expensive, is essential.

Conservation Initiatives
Ecuador first showed interest in protecting Galapagos wild-
life as the centennial of Darwin's visit approached. Two of
the country's earliest environmental laws relate directly to
the archipelago. A nature sanctuary was established there in
1934 and the hunting of selected island species was prohib-
ited two years later. However, these laws amounted to little
more than an expression of virtuous intent since effective
enforcement was precluded by the remoteness of the Gala-
pagos.
International involvement was catalyzed two decades later
by the United Nations Educational, Scientific, and Cultural
Organization (UNESCO), which dispatched a field mission
in 1957, and by the International Union for the Conservation
of Nature (IUCN). The Charles Darwin Foundation was cre-
ated in 1959 to take charge of protecting endangered species,
and received official authorization to operate in the Galapa-
110 Economic Returns

gos. Its Charles Darwin Station, located on the outskirts of

Puerto Ayora (see Figure 7.3), was dedicated in January 1964.
The station coordinates scientific research in the islands, and
its programs to revive land tortoise populations have enjoyed
much success.
The presence of a research facility, combined with grow-
ing national and worldwide interest in habitat protection, led
to the creation of the Galapagos National Park, the limits of
which were demarcated in 1969 and 1970. Except for a little
more than 3 percent of the land area previously occupied by
the military, towns, and farms, the entire archipelago was
placed inside the park. The first superintendent was ap-
pointed in 1972 (Southgate and Whitaker, 1994).

The Growth and Economic Significance

of Ecotourism
In just about any accessible place in the world where natural
scientists venture, tourists are sure to follow. Certainly, the
Galapagos are not an exception in this regard.
An international scientific project mounted in 1964 dem-
onstrated the feasibility of transporting relatively large
groups of people to the islands. For the dedication of the
Darwin Station, sixty-six scientists sailed from California and
the Ecuadorian Air Force flew in dozens of government of-
ficials, diplomats, students, and special guests from the Ecu-
adorian mainland. After this event, a small but steady stream
of visitors was attracted by the facility's presence.
In the late 1960s, international cruise operators sought to
bring groups to the Galapagos, and contacted Ecuadorian
travel firms to make local arrangements. By that time, the
Ecuadorian Air Force had acquired larger planes, which
made it possible to fly passengers to an airfield that the
United States had constructed on Baltra during the Second
World War. The travel industry soon linked up with the mil-
itary's flight operations and the era of what would later be
called ecotourism began (Southgate and Whitaker, 1994).
According to the Galapagos National Park Service (SPNG),
the number of visitors has increased more than tenfold, from
5,000 in 1970 to more than 55,000 in 1995 (see Table 7.2).
There has been no pronounced trend since the late 1980s in
the number of Ecuadorian tourists. By contrast, visits by for-
eigners have grown steadily over time. Variation in the up-
ward trend occurs when wars or civil unrest cause North
Nature-Based Tourism 111

Americans or Europeans to fear traveling to Latin America

(or elsewhere), and when additional capacity is authorized
for the islands' fleet of cruise vessels. In the past, peaks in
authorizations have coincided with changes in the national
government, probably because incoming or outgoing officials
feel obliged or able to respond positively to requests made
by tourism operators during periods of transition.
Some of the economic impacts of Galapagos tourism are
relatively easy to gauge. De Miras (1994) obtained a measure
of total spending in 1993 by multiplying visitor numbers—
10,136 Ecuadorians and 36,682 foreigners (see Table 7.2)—
by average estimated expenditures for each group: $505.61
and $1,336.82, respectively. The total value he obtained was
$54,162,135.
This spending, however, represents just a part of the na-
tional economic significance of Galapagos tourism, since all
foreign visitors must pass through Quito or Guayaquil on
their way to and from the archipelago, and many of those
people choose to spend a few days or weeks on the mainland.

Table 7.2 Number of Ecuadorian and Foreign

Visitors to the Galapagos, 1979 to 1995
Ecuadorian
Year Nationals Foreigners Total

1979 2,226 9,539 11,765

1980 3,980 13,465 17,445
1981 4,036 12,229 16,265
1982 6,067 11,056 17,123
1983 7,254 10,402 17,656
1984 7,627 11,231 18,858
1985 6,279 11,561 17,840
1986 12,126 13,897 26,023
1987 17,769 14,826 32,595
1988 17,192 23,553 40,745
1989 15,133 26,766 41,899
1990 15,549 25,643 41,192
1991 14,815 25,931 40,746
1992 12,855 26,655 39,510
1993 10,136 36,682 46,818
1994 13,357 40,468 53,825
1995 15,483 40,303 55,782

SOURCE: Unpublished data from the Galapagos National Park

Service (SPNG).
112 Economic Returns

At least some of their expenditures during that time must be

regarded as another direct benefit that Ecuador derives from
the islands.
There is no denying that spending by visitors to the Ga-
lapagos is sizable. In a survey conducted in August 1995, the
government's Ecuadorian Tourism Corporation (CETUR)
found that 24 percent of all foreign vacationers arriving by
air identified the islands as their primary destination; another
large portion of the sample included a stop in the Galapagos
in their itinerary. Moreover, foreign exchange earned because
people from other countries visit or live in Ecuador has in-
creased at a rapid pace in recent years, almost entirely be-
cause of the expansion of tourism. In the first six months of
1995, those earnings amounted to $144 million, which was
equivalent to 6.6 percent of what the country received during
the same period for its exports of petroleum, agricultural
commodities, and other products (BCE, 1995, p. 53). Clearly,
many millions of dollars, marks, and yen flow into Ecuador
because of the Galapagos.
One reason why this is so is that the typical foreign visitor
is affluent. Most of the people who make their way to the
Galapagos are from rich countries; of the 457 tourists that
Machlis, Costa, and Cardenas-Salazar (1990) interviewed in
the archipelago in July and August of 1990, 70 percent were
from North America and Europe. These investigators found
that virtually all the foreigners had at least some university
education, and quite a few had graduate or professional de-
grees. One's country of origin and educational attainment
are, of course, closely related to one's standard of living. Ed-
wards (1991), for example, found that the average income for
360 foreign and Ecuadorian tourists interviewed in 1986 was
$32,000.
The Galapagos are steadily becoming a premium nature
tourism destination, visited mainly by people who are well
off. Hotel and restaurant operators observe that fewer mo-
chileros—young foreigners, who carry their belongings in
backpacks (mochilas) and who watch their meager budgets
with great care—are getting out to the islands. In addition,
more Ecuadorian tourists visited each year during the late
1980s than during the early 1990s (see Table 7.2). Demand
for visits among both these groups is fairly sensitive to air-
fares, expenditures on lodging and meals, park entrance fees,
and other costs, many of which have been rising. By contrast,
demand among affluent residents of wealthy nations appears
Nature-Based Tourism 113

to be price-inelastic, if the steady growth in their numbers in

the face of rising costs is any indication.

Local Economic Impacts

Before ecotourism arrived on the scene, the Galapagos were
one of the most isolated places on the face of the Earth. Since
Ecuador laid claim to the islands, in 1832, settlement was
attempted on various occasions. With some minor excep-
tions, though, the farmers, fishermen, soldiers, and romantics
involved in these ventures all gave up after just a few years
(Southgate and Whitaker, 1994).
The challenges settlers faced were considerable. Floreana
and San Cristobal (see Figure 7.3) are the only two islands
with fairly dependable supplies of fresh water, and mainland
markets for fish and other produce are distant. It is little won-
der, then, that the Galapagus had only 1,346 inhabitants at
the time of Ecuador's first national census, in 1950, and 2,391
a dozen years later, when the second census was conducted
(INEC, 1992).
The infrastructure development that has accompanied
ecotourism has made it much easier to travel to and from the
archipelago. The airport on Baltra has been improved and
another facility was opened in 1982 next to Puerto Baquerizo,
the provincial capital. A third facility, with an 1800-meter
landing strip, came on line in Isabela Island in February
1996.
To be sure, the possibility of finding employment in ho-
tels, restaurants, or shops, or on boats or ships, also has stim-
ulated immigration, which has been completely unrestricted
since the Galapagos became a province in 1973. Annual pop-
ulation growth has averaged 5 percent, year in and year out,
due mainly to arrivals from the mainland. The total number
of inhabitants, concentrated almost entirely in Puerto Ayora
and Puerto Baquerizo, was 9,785 when the last census was
taken, in 1990 (INEC, 1992).
The attractions of the Galapagos remain strong for many
potential migrants. Virtually every household in the two larg-
est towns is connected to potable water and sewage systems
and has electricity (INEC, 1992). Actual delivery of water and
power may be subject to interruption, but no urban center
along the Ecuadorian coast can boast of similar service. How-
ever, there are also disadvantages to living in the islands.
Prices for food and other consumer goods, nearly all of which
114 Economic Returns

are brought in from the mainland, are high. Furthermore,

many island residents have not found it easy to benefit from
ecotourism.
Limited local earnings have to do primarily with how af-
fluent visitors like to get around the archipelago. Different
from mochileros and Ecuadorians, who often stay in hotels
or inns and travel to different sites in small boats during the
day, premium tourists generally prefer a cruise, either on
ships that carry up to ninety passengers or on smaller craft
that provide their six-to-twelve guests with more flexible itin-
eraries. Operation of the larger vessels, in particular, requires
few inputs from the local economy. The problem is com-
pounded by a law that mandates a higher minimum wage
and more restrictive work rules for Galapagos residents. Re-
sponding to that law, cruise ship operators do everything
they can to hire crews on the mainland.
De Miras (1994) estimates that, of the $1,337 that an av-
erage foreign tourist spends getting to, around, and back
from the archipelago, $102 goes into the local economy, in
the form of hotel and restaurant charges, receipts for sou-
venir sales, and moorage fees. Additional benefits, such as
wages and salaries paid to park guards and cruise ship em-
ployees, are not very large. A sophisticated analysis, of the
sort needed to estimate the total impacts on the local econ-
omy resulting from these stimuli, has not been carried out
in the Galapagos. One would be surprised, though, if those
impacts amounted to as much as one-third of what well-off
foreigners are spending.

The Role of Public Policy

Local impacts could be enhanced by changing government
policy. In particular, new development initiatives in the is-
lands could be funded by raising the taxes and fees paid by
tourism operators and the people they serve. There is some
scope for doing exactly this since a few operators appear to
be earning supernormal profits. For example, the two airlines
that operate in the archipelago—one private and the other
linked to the armed forces—have exploited their monopolis-
tic position to engage in exactly the sort of price discrimi-
nation depicted in Figure 7.1. In December 1995, for
example, when the price charged Ecuadorians for a round-
trip ticket was 558,000 sucres ($192), international passen-
gers were paying $377 for exactly the same service.
Nature-Based Tourism 115

The government could also collect more money directly

from tourists. Edwards (1991) has carried out research which
shows that sizable revenues are being foregone because of
current pricing arrangements, in which every visitor pays the
same price regardless of the length of his or her stay. He es-
timated that imposing a daily fee of $214 in 1986, when the
flat price for foreigners was $40, would have caused average
trip duration to decline by 50 percent. Regulatory controls
on the number of visitors could have been dispensed with
entirely without any net environmental impact because the
doubling in visitor numbers that would have occurred would
have coincided with the reductions in trip length associated
with charging daily fees. In addition, park revenues would
have risen from $40 to $770 per person (Edwards, 1991).
The government has never seriously considered institut-
ing a daily price. However, entrance charges were raised sub-
stantially in 1993, from $40 to $80 for foreigners and from
600 sucres ($0.35) to 12,000 sucres ($7.00) for Ecuadorians.
In addition, there was an increase in the fees levied on pas-
senger vessels. As a result of many years of high inflation,
the annual charge (patente) assessed on each passenger berth
had fallen below $10. Large ships, capable of carrying ninety
guests, only had to pay about $600 apiece in 1992. By con-
trast, surveys carried out by Bruce Epler, who was a re-
searcher working in Ecuador for the University of Rhode
Island at the time, revealed that vessels of that sort were earn-
ing as much as $4 million in annual gross revenues (personal
communication, 1992). No data are available, but no knowl-
edgeable observer thinks that annual operation and mainte-
nance expenses exceed $2.5 million for any ship carrying
passengers around the Galapagos.
INEFAN, which holds the administrative responsibility
for all the country's parks, responded to these findings by
raising berth fees. In general, vessels equipped to accommo-
date foreigners in reasonable comfort pay the "luxury" rate
of $200 per berth per year. Annual patentes of $100 or $150
are charged tourist craft that either are smaller or lack an
escape boat or advanced fire control measures. Day boats pay
$30 per annum.
As a result of the higher entrance fees and patente in-
creases, revenues have risen substantially. In 1993, $2.2 mil-
lion were collected. Because of growth in international
arrivals, 1994 and 1995 revenues were $3.2 and $3.7 million,
respectively. The portion of this money going to the islands,
116 Economic Returns

however, has been small. INEFAN and predecessor agencies

have always used Galapagos tourism to help pay for the man-
agement of mainland parks, none of which comes close to
being financially self-supporting. In 1993, for example, the
SPNG kept approximately 30 percent of the $2.1 million it
collected; the rest was spent in other parks.
To capture a larger share of SPNG revenues, local govern-
ments will have to develop specific and viable proposals for
development projects—which has not happened so far. It
could also be that they will no longer have to compete just
with INEFAN for funds collected in the Galapagos. In March
1996, the Ministry of Finance announced that, after consid-
eration of INEFAN's requests to budget 3.5 billion sucres
(worth approximately $1.2 million at the time) for the SPNG
and 5 billion sucres ($1.7 million or so) for mainland parks,
only 850 million sucres (roughly $300,000) and 2.7 billion
sucres (about $900,000) were actually to be allocated during
the year. This action could have resulted from the need to
settle bills incurred during the small war that Ecuador fought
with Peru in early 1995, or to pay for other projects to which
the government had assigned a higher priority. However, the
central government's decision to capture a higher portion of
the entrance fees and patentes collected in the Galapagos
might signify that, in the future, local governments, the
SPNG, and mainland parks will face new, and more power-
ful, competition for scarce financial resources.
Finally, it must be conceded that the fee increases insti-
tuted by INEFAN since 1992 might have been detrimental to
islanders' immediate financial interests since higher fees,
along with increases in airfares and other prices, have helped
to discourage visits by mochileros, Ecuadorians, and other
less affluent tourists who tend to patronize on-shore facili-
ties. Higher costs, it is fair to say, have accelerated the gen-
trification of Galapagos tourism, which is diminishing the
earnings of many of the islands' hotels, restaurants, and
shops.

Crises Coming to a Head

Benefiting in only a limited way from international visitors'
interest in the Galapagos, residents have been quick to seize
whatever economic alternatives present themselves. At the
same time, they have responded eagerly to calls for political
action to close the gap between what they think they should
receive from ecotourism and what actually comes their way.
Nature-Based Tourism 117

Some alternative economic activities, like fishing, involve

substantial environmental damage. Since the government
has virtually no capacity for effective regulation, intense har-
vesting has led to severe depletion of various species. The
latest episode of boom-to-bust exploitation of an open access
fishery began in 1991, when people and firms responsible for
depleting sea cucumber (Isostichopus fuscus) populations
along the coast of mainland Ecuador started to transfer their
operations to the Galapagos. The sea-bottom-dwelling rela-
tives of the starfish, which are sold as an aphrodisiac in China
and elsewhere in Asia, proved to be particularly abundant in
the waters around Isabela. Furthermore, the island is far be-
yond the reach of the SPNG, the Subsecretariat of Fishery
Resources (SRP), or the Ecuadorian Navy. As a result, indi-
vidual fishermen and the oceangoing transport ships that
routinely pass by to make purchases have gone about their
business with little official interference. During the early
stages of the sea cucumber boom, a fisherman on Isabela
could make as much as $700 a week, which stimulated a
major increase in harvest rates as well as migration. By the
middle of 1992, daily production for the entire archipelago
was averaging 70,000 to 110,000 sea cucumbers. In 1994,
when a ban on harvesting was declared (but never really en-
forced), approximately 116 fisherman had settled on Isabela
(Zador, 1994).
The collapse of natural stocks due to overfishing could be
a catastrophe for the entire island ecosystem. There is no way
to know, for example, which kinds of birds might depend
directly or indirectly for their survival on sea cucumbers,
which are the most plentiful form of marine life in the ar-
chipelago. Fishermen, of course, show little or no regard for
these consequences and have vigorously resisted any attempt
to regulate their activities.
Islanders (both those born in the Galapagos and those who
have lived there for several years) and sea cucumber fisher-
men are largely, though not entirely, distinct populations. In
addition, many islanders either have a stake in the success
of ecotourism or hope to have one in the future, so they have
some interest in ecosystem integrity. However, they have
given enthusiastic support to Eduardo Veliz, who was
elected in 1994 to represent the archipelago in the National
Congress, and other political leaders who are demanding a
larger share of tourism's benefits for the islands. For example,
a general strike was called in September 1995 to protest the
Ecuadorian president's veto of Veliz's bill, which among
118 Economic Returns

other things would have increased automatic appropriations

by the central government to local governments and projects
and forced all tourists in the Galapagos to spend at least one
night ashore. The strike succeeded in closing most private
businesses and interfering with access to the Darwin Station
(Lemonick, 1995).
Responding to the concerns of Galapagos residents, as the
Ecuadorian government has committed itself to doing, in-
volves some serious conundrums. For one thing, undertaking
more projects to benefit local communities could stimulate
additional migration to the archipelago. That such migration
would dissipate the benefits accruing to individual house-
holds is widely recognized. Even Veliz has expressed some
sympathy for coupling migration controls, which his prede-
cessor supported (Brooke, 1993), with increased financial
support from the central government. Such support, of
course, could come at the expense of budgets for habitat man-
agement and species protection.
Adding to the difficulty of choices facing the Ecuadorian
government is the international scrutiny of what takes place
in the Galapagos, which has intensified in recent years. Vis-
itor numbers fell sharply in 1995 in response to the islanders'
strike. Since then, ecotourism has rebounded, but now
UNESCO appears poised to declare the archipelago a "world
heritage in danger." Responding to the damage that such a
declaration would do to national prestige and foreign
exchange earnings, Ecuador's interim president recently is-
sued a decree addressing various issues, including pollution,
migration, illegal fishing, and the introduction of species.
Though encouraging, this action cannot take the place of
comprehensive legislation addressing conservation and local
development issues. After years of debating various propos-
als, the Ecuadorian Congress has put the future of the Gala-
pagos back on its agenda. However, there are no guarantees
regarding either the content or timing of a new law (Econo-
mist, 1997b).

Ecotourism, Habitat Protection, and

Local Economic Development
There is no denying that Costa Rica and Ecuador have been
highly effective at responding to the burgeoning demand, in
places like the United States and Germany, among those who
wish to visit rain forests and other tropical habitats. This suc-
cess is manifested in increased patronage of protected areas
Nature-Based Tourism 119

in Costa Rica and in the status the Galapagos now enjoy as a

prime ecotourism destination. In turn, demand for access to
natural habitats appears, in some places, to have strength-
ened the case for conserving them.
Other than backpackers and a few others, people who
spend the time and money required to reach a place like
Monteverde or the Galapagos are pretty exacting about the
quality of the goods and services they purchase. More often
than not, local communities find it a challenge to supply ap-
petizing food, clean sheets, cold beverages, and, more im-
portant than anything else, safe and reliable transportation.
Consequently, the firms that provide these services, which
by and large are based in capital cities or foreign countries,
receive most of what ecotourists spend.
The distribution of the gains from nature-based tourism
could be altered by investing in human capital and local in-
frastructure. Before making such an investment, though,
some fundamental economic questions relating to opportu-
nity costs would have to be faced. The returns local people
are likely to enjoy as a result of investment specifically tai-
lored to ecotourism would have to be compared to the returns
they would capture if, for example, spending on general hu-
man capital—which is applicable across all sectors of the
economy—were increased. Likewise, other investment op-
tions, in protected areas, for instance, would have to be eval-
uated as well.
Opportunity cost questions are of central concern because,
as emphasized in this chapter, the environmental base for
nature-based tourism is by no means secure in Costa Rica and
the Galapagos. Substantial sums need to be spent so that the
habitats and species that attract international visitors do not
disappear, due to encroachment by farmers, ranchers, and
loggers, or because of competition from newly introduced
flora and fauna. Those sums greatly exceed national park
services' existing financial resources.
Both Costa Rica and Ecuador are trying to raise more
money by increasing the fees paid by park visitors and, in
the case of the Galapagos, by increasing fees paid by the busi-
nesses that cater to visitors. In some cases, price discrimi-
nation and varying fees from site to site have yielded
dividends. At the Monteverde Preserve, for example, as
noted previously, different fees are charged domestic stu-
dents, foreign students, international tourists visiting in
groups, and others. In Ecuador, foreigners who are not stu-
dents pay $80 to visit the Galapagos; meanwhile, entrance
120 Economic Returns

charges are modest for those mainland parks that receive few
visitors. By contrast, the Costa Rican government has only
just begun to move away from a policy of charging all for-
eigners one uniform price for admission to any park and all
citizens of the country another price. Setting fees that better
reflect the differences among various sorts of visitors and
among various parks would allow the country to capture
more revenues, while at the same time redirecting people
away from places, like Manuel Antonio, where carrying ca-
pacities are probably being exceeded. Likewise, linking en-
trance fees to trip duration should be a central feature of
pricing policy, in the Galapagos and elsewhere.
Even if countries with appealing ecotourism destinations
price access efficiently, it will be rare for revenues generated
to exceed the expenses of protecting natural areas. Even at
Monteverde, the costs of intense maintenance and vigilant
protection, which are needed to secure the reserve, exceed
entrance fee revenues by a wide margin. Without donations
from foreign individuals and foundations, the site would de-
teriorate.
The Galapagos appear to be one of those unusual places
with the potential to be a net generator of funds, defined
broadly to consist of foreign donations as well as users' fees.
As noted in this chapter, it has long been a policy to use
monies collected from visitors and tourism businesses to
help pay for the management of mainland parks. If the recent
decisions by the Ministry of Finance are any guide, central
government officials are now casting an envious eye toward
those monies as well.
But even in the archipelago, it is important not to exag-
gerate the size of the financial surplus that ecotourism can
be expected to generate, year in and year out. Once habitat
protection has been adequately provided for, not much is
likely to be left over for local development initiatives or any-
thing else.
Ill
Key Elements of an Integrated
Strategy for Habitat Protection
and Economic Progress
This page intentionally left blank
8
Another Approach to
Habitat Conservation

Agricultural Intensification

nder the right set of circumstances, nature-based

U tourism and the harvesting of forest products are
commercially promising and can be carried out in ways that
benefit local communities and keep renewable resources in-
tact.
For example, conditions for the sustainable production of
agai appear to be favorable in the Amazon estuary. The re-
gion's waterways provide easy access to Belem, where sev-
eral palm heart exporters are located and where demand is
strong for agai fruit. In addition, output can be increased
through the application of relatively simple stand improve-
ment measures, such as seed dispersal and periodic thinning
(Pollak, Mattos, and Uhl, 1995). Accordingly, quite a few
smallholders in the estuary find it rewarding to raise acai,
which involves little or no damage to natural resources.
Another place where economic development has been en-
vironmentally benign as well as socially broad-based is the
Monteverde Cloud Forest Biological Preserve. The park is un-
doubtedly one of the best-managed of its kind in the world,
and visitors spend millions of dollars each year on meals,
lodging, and souvenirs in neighboring communities; it is
widely hailed as an example of what ecotourism can accom-
plish.
No one would argue against taking full advantage of the
opportunities being offered in places like the Amazon estu-
ary and Monteverde. However, the evidence presented in this
book suggests that those opportunities are not all that com-
mon. For example, promising sites for the collection of non-
timber forest commodities in the Brazilian Amazon are

123
124 An Integrated Strategy for Habitat Protection and Economic Progress

confined almost entirely to the estuary and floodplains

(Goulding, Smith, and Mahar, 1996), which together account
for just 2 percent of the entire watershed. Elsewhere, har-
vesting costs tend to be high, because beneficial species are
not concentrated. Also, unless a road or a waterway is close
at hand, transportation expenses are prohibitive.
Likewise, Monteverde is a unique place. The Alabama
Quakers who first settled in the area a half-century ago ap-
plied various conservation measures, including the protec-
tion of primary forests. In addition, the TSC, which manages
the reserve, has pursued habitat conservation and local de-
velopment initiatives that are frequently advocated but rarely
implemented.
If opportunities for sustainable tropical forest develop-
ment occur mainly in special environmental and market
"niches," they are best exploited by implementing small-
scale projects with a narrow geographic focus. An example
of such a project is one undertaken in the late 1980s and early
1990s to increase agai fruit output in the Amazon estuary.
With support from the Ford Foundation and the British Over-
seas Development Agency (ODA), a plant biologist was as-
signed to work with local communities on the improvement
of thinning practices, the establishment of a nursery to pro-
duce better seedlings, and related pursuits. Also investigated
were techniques, like pinching off flower buds, for delaying
fruit production until the off-season, when prices are higher
(John Rhomboid, personal communication, 1996). Dozens of
families, at least, benefited directly from this initiative.
In contrast with the performance of small, niche-specific
efforts, large-scale undertakings have experienced serious
shortcomings. During the late 1980s and early 1990s, a strong
case was made for promoting activities like the extraction of
nontimber products, which would conserve forests while si-
multaneously improving local living standards (see chapter
4). AID and other development agencies responded by fund-
ing a number of multimillion-dollar projects in the region,
many of which were managed by environmental organiza-
tions. By and large, these projects' accomplishments have
been disappointing, with nontimber extraction, ecotourism,
and other activities having taken hold only here and there.
Heeding this experience, implementing and funding agen-
cies have become more cautious. Efforts to promote sustain-
able economic activities in tropical forests and other
threatened ecosystems now tend to be quite limited in scope.
Moreover, it is coming to be accepted that nontimber extrac-
Agricultural Intensification 125

tion, ecotourism, logging with minimal environmental im-

pacts, and genetic prospecting can only make a modest
contribution, and that the search for effective habitat conser-
vation measures must continue.

Agroforestry in Tropical Forest Settings

A sensible response to an understanding of the limitations of
economic activities examined in chapters 4-7 of this book is
to pursue a more varied array of initiatives for the purpose
of saving habitats and promoting local development. One al-
ternative that holds promise is agroforestry, which involves
the interplanting of crops or forages with trees that yield con-
struction materials, fuelwood, and other useful commodities.
Denevan et al. (1984) point out that countless generations
of indigenous communities in the Amazon Basin, like their
counterparts in other tropical settings, have practiced one or
another form of agroforestry for millenia. Such systems are
environmentally sound in that they mimic the succession of
plants that emerge in a small forest clearing. At the same
time, careful management is often needed, in order to maxi-
mize useful output.
Indigenous peoples are not agroforestry's only practition-
ers. Many of the ethnic Japanese who have migrated to the
Brazilian Amazon have enjoyed considerable success in in-
terplanting black pepper with a variety of other annual and
perennial crops as well as trees (Subler and Uhl, 1990). Also,
Browder (1997) reports that colonists who have settled in
Rondonia are interested in agroforestry; particularly enthu-
siastic are those who have acquired an appreciation for the
limitations of ecosystem mining, as a result of having farmed
in the state for several years. Smith et al. (1995) note that
there is a pronounced trend toward agroforestry throughout
the region, which they say should lead to diminished pres-
sure on forested ecosystems.
Given the system's management demands, successful
agroforestry can require a substantial investment in human
capital. This investment is accomplished among the ethnic
Japanese through producers' associations and community
networks, which channel technical and market information
to their members (Subler and Uhl, 1990). Agroforestry pro-
jects supported by development agencies seek to establish
the same sort of institutional infrastructure, sometimes
within the public sector but more and more often in civil
society.
126 An Integrated Strategy for Habitat Protection and Economic Progress

Obviously, commercial viability also has a lot to do with

favorable market conditions, as an economic study of agro-
forestry in the Ecuadorian Amazon makes clear (Southgate et
al., 1992). In this region, it is quite common for farmers to
allow laurel (Cordia alliodora) and other commercial timber
species to grow in fields where coffee has been planted
(Uquillas, Ramirez, and Sere, 1991). Research conducted dur-
ing the 1980s, under the auspices of an AID project, yielded
several recommendations for the refinement of this practice
(Peck, 1990), as well as the technical coefficients needed for
a financial analysis of the improved system.
The results of the analysis, expressed as per-hectare an-
nual net returns, are reported in Table 8.1. Each estimated
net return corresponds to a certain wage, coffee price, stump-
age value, and real interest rate. The lower and upper daily
wages, $3.00 and $3.50, are consistent with what unskilled
laborers earn in the study area. Likewise, the two coffee
prices used in the study, $0.10 and $0.15 per kilogram, are
in line with the farm-level realities of recent years. By con-
trast, both of the stumpage values, $10 and $15 per cubic
meter, are better than what resource owners typically receive.
Similarly, the real interest rates they pay on loans tend to be
well above the 7.5-to-10-percent range. Of course, using
higher stumpage values and lower interest rates enhances
profitability estimates. However, the parameters used in the
study were chosen because the price of standing timber
would be higher if Ecuadorian wood product markets were

Table 8.1 Annual Net Returns per Hectare for

Agroforestry in the Ecuadorian Amazon, late 1980s.
Real interest rate

7.5% 10.0%

Daily wage = $3 .00

lau P = $10/m 3 & cof p = So,. 10/kg S48 .40 $36 .32
lau P = $10/m 3 & cof p = $0.,15/kg 164, .19 154,.83
lau P = Sl5/m 3 & cof p = $0.10/kg $61, ,01 $45, ,69
lau p = $15/m 3 & cof p = $0.,15/kg 176,,81 164,.20
Daily wage = $3 .50
lau P = $10/m3 & cof p = $0. 10/kg $15, ,76 $2,.07
lau P = $10/m 3 & cof p = $0. 15/kg 131, ,55 120,,58
lau P = $15/m 3 & cof p = SO. 10/kg $28, ,88 $11. .45
lau P = $15/m 3 & cof p = $0. 15/kg 144..17 129..96

SOURCE: Data from Southgate et al. (1992): 42.

Agricultural Intensification 127

more competitive and because real interest rates would be

lower if financial intermediation in the country were more
efficient (Southgate et al., 1992).
The financial analysis indicates that agroforestry's profit-
ability is not affected very much by interest rate fluctuations.
This is because most revenues, which are associated with
coffee sales, are captured early in the agroforestry rotation.
Furthermore, the system is less risky than, say, establishing
and managing a tree plantation. This is because multiple
commodities, rather than a single commodity, are produced
(Southgate et al., 1992).
By contrast, profitability is highly sensitive to wage levels
since the improved system is labor-intensive. Of even greater
importance is the farm-level value of the main product of the
system in eastern Ecuador, which is coffee. At the higher
wage and the lower coffee price, agroforestry compares
poorly with alternative land uses, including extensive cattle
ranching. Another reason that the system is unprofitable is
linked to the diseases that afflict coffee trees, thereby dimin-
ishing yields (Southgate et al., 1992).

The Impacts of Sectorwide Improvements

in Productivity
As is the case with just about any line of commerce, agrofo-
restry's profitability is inextricably linked to prices paid for
output. Of course, the best prices are received by those pro-
ducers who enjoy good access to markets. This is true
whether the commodity in question is coffee or black pepper
produced on a farm where interplanting is practiced, or
whether it is agai extracted from a riparian forest. Appreci-
ating all this, Smith et al. (1995) emphasize that the com-
mercial prospects for agroforestry are especially favorable
where there is a passable road or a navigable waterway
nearby that leads to a marketing center of one kind or an-
other.
It is not to be expected, then, that agroforestry will take
place first in remote hinterlands. The same holds true for
intensified crop and livestock production, in general,
which results from spending more on research and exten-
sion systems and other investments in the agricultural
economy. Intensification, which is related to technological
improvement, can be expected to diminish human en-
croachment on natural habitats. Quite often, the impacts
are dramatic.
128 An Integrated Strategy for Habitat Protection and Economic Progress

The effects that higher productivity has along agricultural

frontiers are depicted in Figure 8.1.. Investment and tech-
nological change lower costs of production for all farmers
and ranchers, including those located in remote areas. This
enterprise-level consequence is represented in Figure 8.1 by
a downward shift of marginal-cost (MC) and average-cost
(AC) curves. But at the same time, commodity supplies in-
crease, as shown in the figure. This causes the market-
clearing price to fall.
It is possible for the impact of lower production to domi-
nate the impact of lower prices. Suppose that better agricul-
tural land-clearing techniques are developed. This would
result in substantial displacement of enterprise-level cost
curves along agriculture's extensive margin. Provided there
are no major changes in market-level supply and price, fron-
tier producers would raise output and would also find it prof-
itable to deforest more land.
Much more typical, however, are investment and techno-
logical change that benefit inframarginal producers more
than those in remote areas. In this case, which is illustrated
in Figure 8.1, frontier producers incur losses because price
declines exceed reductions in per-unit costs (AC). In the
short run, migration to agriculture's extensive margin ceases,
since additional deforestation is unprofitable. In time, people
who already have settled on the frontier choose to abandon
their farms and ranches, which are reclaimed by natural veg-
etation.
Agricultural intensification clearly has caused the sector's
extensive margin to recede in the United States. The area
used to raise crops and livestock in the country reached a
maximum around 1920. (In New England, the peak occurred
about 150 years ago.) Since then, productivity has skyrock-
eted, due to the development of hybrid corn and other tech-
nological improvements, and because of investments in
roads and other infrastructure. At prevailing crop prices,
which reflect low production costs in the midwestern bread-
basket, production of corn, soy beans, and other commodities
is uneconomical from New York to Alabama. Accordingly,
much of the land in that part of the country which was
farmed up to a generation ago is now covered with maturing
forests.
The inverse relationship between intensified crop and
livestock production and expansion of the agricultural fron-
tier holds in other parts of the world as well. Had it not been
Agricultural Intensification 129

Figure 8.1 Agricultural intensification and the agricultural frontier.

for the Green Revolution, which has boosted rice and wheat
yields, tropical deforestation in Asia would now be dis-
cussed mostly in the past tense. In the absence of intensifi-
cation, farmers and ranchers would have had to occupy every
available square meter in order to meet burgeoning food de-
mands, resulting from population growth and higher stan-
dards of living.
Results of a regression analysis of the causes of defores-
tation in the Brazilian Amazon (Reis and Guzman, 1994)
could be used to bolster the claim that intensification can
lead to land clearing. Indeed, those results reveal a link be-
tween increased crop output and deforestation at the county
[municipio] level. But in no way does this disprove the fact
that raising agricultural productivity tends to arrest farmers
and ranchers' overall encroachment on natural habitats. Con-
sistent with what Reis and Guzman (1994) have found, in-
tensification should cause cropland and pasture to expand at
the expense of forests in inframarginal areas—centrally lo-
cated municipios, in the case of the Brazilian Amazon. But
at the same time, agricultural land use should be diminishing
in more remote places. If more land is left undisturbed in
frontier areas than is cleared in inframarginal areas, then the
overall relationship between agricultural productivity in-
creases and deforestation is negative.
That there is, in fact, a negative linkage is indicated by the
findings of another regression study, one in which national-
level data are used (Southgate, 1994). In that study, annual
percentage growth in cropland and pasture during the mid-
dle 1980s (AGLNDGRO) is the dependent variable, and the
130 An Integrated Strategy for Habitat Protection and Economic Progress

independent variables include annual percentage growth in

population, exports, and crop yields (POPGRO, EXPGRO,
and YLDGRO, respectively). Also included is a dummy fac-
tor, reflecting a comparison of actual and potential land use
during the early 1980s (Southgate, 1994), that indicates that
land not yet cleared is poorly suited to agriculture (NO-
LAND).
Using data for twenty-three countries, ordinary-least-
squares estimation of the regression model yields the results
that follow.

AGLNDGRO = 0.463 + 0.249 POPGRO + 0.031 EXPGRO

(0.161) (0.066) (0.014)

- 0.198 YLDGRO - 0.641 NOLAND

(0.033) (0.205)

ADJ fl2 = 0.669 SSR = 3.489 F = 12.098

All regression coefficients are statistically significant (the

numbers given in parentheses are standard deviations), and
each parameter estimate is consistent with what should be
expected. For example, frontier expansion is positively re-
lated to population growth and increased exports. Also, in a
country where natural conditions do not favor frontier ex-
pansion (i.e., where the value of NOLAND is 1 instead of 0),
annual growth in cropland and pasture should be 0.641 per-
centage points lower than what would occur if soils lending
themselves well to crop or livestock production were unoc-
cupied.
Of special relevance to a discussion of deforestation and
agricultural intensification is the negative and statistically
significant coefficient of crop yield growth. (YLDGRO is a
satisfactory indicator of productivity trends for the sector as
a whole in this model since most deforested land is used for
livestock production and, therefore, there is not much chance
of lower crop yields in frontier areas having a major impact
on nationwide averages.) Comparison of regression coeffi-
cients indicates that a 1 percent increase in yields can offset
nearly four-fifths of the agricultural land clearing induced by
a 1 percent increase in population. Alternatively, a 1 percent
yield increase can compensate for the deforestation that
would result from 6 percent growth in agricultural exports.
In addition to being revealed by regression analysis, the
relationship between agricultural intensification and habitat
Agricultural Intensification 131

conservation is clarified by the experience of two Latin

American nations: Chile and Ecuador. During the 1980s,
commodity demand rose dramatically in Chile, due to pop-
ulation growth that averaged 1.7 percent a year and an in-
crease in agricultural exports at a 17.5 percent annual rate.
Nevertheless, there was practically no geographic expansion
in Chile's agricultural economy because yields were rising
by 3.6 percent per annum. Meanwhile, in Ecuador, where
annual population growth averaged 2.7 percent and com-
modity exports were rising by 11.4 percent a year, yields ac-
tually fell slightly. As a result, the country experienced the
second-highest rate of agricultural land clearing in the hemi-
sphere: 2.0 percent per annum between 1982 and 1987
(Southgate, 1994).
Commodity demand in Latin America is bound to increase
for many years to come. Notwithstanding recent declines in
human fertility, population growth is continuing. Also, in-
comes are beginning to rise, and agriculture and other sectors
of the rural economy are growing because discriminatory pol-
icies, like currency overvaluation, are largely a thing of the
past. Without intensification, continued deforestation is all
but inevitable.
In essence, raising crop and livestock yields requires the
building up of nonenvironmental wealth in the sector. Good
access to markets, made possible by the articulation of local
road networks, is essential, as an econometric analysis of fac-
tors contributing to agricultural land clearing in Belize makes
clear. Chomitz and Gray (1995), who carried out that analy-
sis, conclude that agricultural development can be achieved,
with minimal loss of natural habitats, by improving trans-
portation infrastructure, including secondary and feeder
roads, in areas that have been settled already and where nat-
ural conditions favor crop and livestock production.
Another requirement for the sustainable intensification of
agriculture is investment in the sector's science and technol-
ogy base, which has been singularly deficient in Ecuador.
Having declined 7.3 percent a year from 1975 through 1988,
real spending on agricultural research is now less than 0.2
percent of sectoral GDP, which is low even by the standards
of neighboring countries. Support for agricultural extension
is similarly inadequate (Southgate and Whitaker, 1994).
An important reason why too little support has been made
available for research and extension is that too much money
has been spent on irrigation subsidies, which yield a poor
economic return, benefit a small minority consisting of rela-
132 An Integrated Strategy for Habitat Protection and Economic Progress

lively affluent individuals, and result in the waste and mis-

use of natural resources. On average, the prices that the
beneficiaries of Ecuador's public irrigation projects pay for
water are around 5 percent of what it costs the public sector
to deliver the input. The money needed to cover the other 95
percent, which amounts to two-fifths of the government's
budget for agriculture, dwarfs the funding for the develop-
ment and dissemination of improved technology for crop and
livestock production. Furthermore, no beneficiary of a public
irrigation project has a strong reason to adopt costly conser-
vation measures. Accordingly, water is wasted and misused
on a grand scale and land is sometimes ruined because of
salinization, which is brought about by excessive irrigation
and poor drainage. Furthermore, Ecuador's policy of cheap
irrigation water benefits no more than 10 percent of the coun-
try's producers, who tend to be much better off than those
with no connection to a public irrigation system (Southgate
and Whitaker, 1994).
Since the pace at which improved crop varieties and pro-
duction techniques are being developed and disseminated
has been too slow in Ecuador, yields have stagnated, and
commodity demand growth has had to be accommodated pri-
marily by increasing cropland and pasture, at the expense of
tropical forests and other natural habitats. Only in Surinam,
a lightly populated country with very little cropland and pas-
ture in the early 1980s, was annual percentage growth in ag-
ricultural land higher than what Ecuador experienced
between 1982 and 1987 (Southgate, 1994).
Obviously, human pressure on forested hinterlands can be
eased considerably by intensifying crop and livestock pro-
duction. The same relationship holds true for the wood prod-
ucts sector. As mentioned at the end of chapter 5, future
increases in global roundwood demand will be met mainly
in two ways. The first is improved management in forests that
are already accessible. The second response will be the es-
tablishment of tree plantations in Latin America and other
parts of the developing world, and, again, in relatively ac-
cessible places. By contrast, loggers are not expected to ven-
ture far into primary forests that lack transportation
infrastructure (Sohngen et al., 1997).
Intensification is not a panacea, either in agriculture or in
the wood products sector. Consider the case of a small, open
agricultural economy, one in which commodity prices are
determined exclusively at the international level. This im-
plies that the price impacts of technological change, which
Agricultural Intensification 133

are illustrated in Figure 8.1, are bound to be negligible. If

input elasticities are also very high, then improved agricul-
tural productivity always leads to increased deforestation.
Kaimowitz (1996) contends that this is an accurate descrip-
tion of the livestock sector's geographic expansion in Central
America.
Another problem, which the econometric study by Reis
and Guzman (1994) addresses, is that agricultural intensifi-
cation is often accompanied by various environmental dam-
ages. These include soil erosion and the contamination of
waterways with fertilizers and pesticides. Moreover, local
biodiversity is diminished insofar as intensification leads to
the destruction of the last remaining patches of forests in in-
framarginal areas. Pagiola et al. (1997) point out that this lo-
cal biodiversity can be an important buffer for agricultural
systems in which one or a few species are predominant. To
prevent its loss, they advocate measures like the avoidance
of monoculture and the protection of forest remnants.
Notwithstanding their call to limit environmental risks,
however, Pagiola et al. (1997) subscribe to the emerging con-
sensus that intensification plays a pivotal role in arresting
encroachment by farmers and ranchers on tropical forests
and other natural habitats.
This page intentionally left blank
9
Paying for Habitat Conservation
and Investing in Human and
Social Capital

A .n important merit of increased agricultural pro-

/Vductivity is that it facilitates other habitat protec-
tion measures. Insofar as technological change causes
commodity prices to decline more than farming and ranching
expenses along the extensive margins of crop and livestock
production (see Figure 8.1), real estate values fall. Accord-
ingly, it is cheaper to set aside forested hinterlands to se-
quester carbon, to harbor threatened species, or for any other
purpose.
Of course, intensification can have precisely the opposite
effect in inframarginal areas. As Reis and Guzman's (1994)
econometric analysis demonstrates and as Pagiola et al.
(1997) warn, higher crop and livestock yields strengthen in-
centives to clear tree-covered land that is close to markets
and that lends itself well to farming or ranching. Balanced
against this, however, would be increased payments to keep
accessible forests intact, which would be forthcoming in a
society that has grown wealthier because of productivity
gains in agriculture and other sectors of the economy.
Although the fate of forests within easy reach of farmers
and ranchers is a legitimate concern, it is clear that intensi-
fying crop and livestock production arrests aggregate defor-
estation (Southgate, 1994). As indicated in the preceding
chapter, enhanced agricultural productivity has allowed for
a resurgence of forests in the eastern United States, and the
Green Revolution has reduced the extent of agricultural land
clearing in many poor countries. Likewise, widespread in-
tensification is absolutely essential if the visionary plans,

135
136 An Integrated Strategy for Habitat Protection and Economic Progress

currently being discussed, to step up habitat protection in

the developing world are to succeed.
For example, Bezanson and Mendez (1995) propose the
establishment of a global body under the auspices of the in-
ternational Convention on Biological Diversity (CBD). It
would raise money by levying charges on transnational ac-
tivities that make use of the global commons, broadly denned
to include oceans beyond sovereign states' exclusive eco-
nomic zones, biodiversity, and other resources. Much of that
money would be used to save tropical forests. Similar, though
less ambitious, is a proposed scheme to use the proceeds of
a tax on the international timber trade to pay for habitat con-
servation. For example, a l-to-3 percent duty on tropical tim-
ber imported by the European Union would raise $30 million
to $90 million per annum, without significantly distorting
markets (NEI, 1989).
Better use can be made of existing mechanisms for financ-
ing park protection, such as entrance fees and the GEF. In
addition, interest is growing in trust funds, which usually are
created through debt-for-nature swaps arranged to benefit na-
tional park systems or specific protected areas. Among other
advantages, such funds allow for the recurring costs of park
management to be met and can also be used to attract com-
plementary support from development agencies, philan-
thropic foundations, and other sources (Newcombe, 1995).
But putting national parks and reserves on a sounder fi-
nancial footing will never be any more than a partial solution.
A survey of forests in eighty countries around the world re-
veals that only 6 percent of them are officially protected
(WWF, 1996). Even if the share in Latin America were in-
creased to 10 percent and the money needed for effective
management were raised, the vast majority of natural habitats
would still be exposed to encroachment. It should not be for-
gotten, after all, that attempts in recent years to conserve hab-
itats by fostering ecotourism and the sustainable harvesting
of forest products grew out of an appreciation of the short-
comings of conservation strategies that were predicated ex-
clusively on park protection.
To deal more directly with encroachment, payments re-
flecting the value of environmental services provided by for-
ests could be made to agents of land use change. Available
evidence of the costs of habitat destruction, summarized in
chapter 2 of this book, suggests that deforestation would not
be affected a great deal if farmers and ranchers internalized
the downstream impacts of hydrological cycle disruptions
Investing in Human and Social Capital 137

and accelerated soil erosion. Similarly, payments that the

pharmaceutical industry might offer to conserve areas where
genetic raw material is collected are well below the value of
newly cleared cropland or pasture along agricultural fron-
tiers in the Western Hemisphere (see chapters 2 and 6).
By contrast, the economic disutility associated with the
global climatic change induced by deforestation could well
outweigh the benefits that might result from the geographic
expansion of farming and ranching in the American tropics.
If releasing a ton of carbon into the atmosphere creates $10
in damages and if clearing a single hectare results in 100 to
200 tons of emissions, then leaving natural vegetation undis-
turbed is worth $1,000 to $2,000 per hectare, over and above
whatever values are attached to watershed services and bio-
diversity conservation. Amounts in this range greatly exceed
the prices at which frontier holdings normally change hands.
It would seem that the potential exists for mutually ben-
eficial trade between firms and individuals who gain when
forests are left standing and those who benefit from land use
change. To be specific, manufacturers and public utilities
facing taxes or regulations on carbon emissions might find
it worthwhile to offer the levels of financial inducement
needed to convince landowners in places like the Amazon
Basin to refrain from clearing away natural vegetation.
One alternative would be to engage in the sort of bidding
exercise that the U.S. Soil Conservation Service (SCS) has
used at times to identify farmer-participants in the Conser-
vation Reserve Program (CRP). In auctions in which rights to
develop tree-covered land are bought and sold, each partic-
ipating landowner would state the minimum price at which
he or she would agree not to remove natural vegetation from
his or her holding. As a rule, such a price would comprise
the difference between the present value of agricultural in-
come and clearing costs and the present value of income de-
rived from sustainable harvesting of timber and other forest
products. On the other side of the market, bids offered by
prospective buyers of development rights would reflect the
cost of reducing carbon emissions in other ways, by switch-
ing to alternative energy sources, for instance.
In any market, there is a gap, reflecting bargaining and in-
formation costs, between the price paid by buyers and the
price received by sellers. However, communications tech-
nology has advanced so far that the gap in the market for
transferable forest development rights need not be very large.
Bidding could be computerized, which would make the cost
138 An Integrated Strategy for Habitat Protection and Economic Progress

of arriving at a market equilibrium negligible. Also, plotting

the boundaries of areas that would be covered by carbon se-
questration agreements would be a straightforward matter,
requiring the use of hand-held global positioning system
(GPS) receivers that cost a few hundred dollars and that are
accurate within a meter or two.
Nor would verifying compliance be much of a problem.
A good indicator of the expense of monitoring is provided
by a contract that the World Bank's office in Brasilia re-
cently awarded for land use assessment in five locations.
Each of the locations is in the Amazon Basin and is more
than 100,000 hectares in size. Since total expenditures for
acquiring and analyzing satellite images and engaging in
"ground-truthing" (i.e., verifying what the images appear to
show) will be approximately $90,000, per-hectare costs will
be less than $0.20 (Robert Schneider, personal communica-
tion, 1996). Obviously, monitoring expenses in this range
would never impede a mutually beneficial exchange be-
tween landowners and those who want natural vegetation
to be left undisturbed.
All the technology required for creating efficient and
large-scale markets for tropical forest development rights is
available. The reason why no such market exists at present
is that, as a rule, carbon emissions are lightly taxed and reg-
ulated. Unless and until this situation changes, buyers of
development rights will remain scarce. Carbon sequestra-
tion agreements will continue to be a novelty, involving an
environmental organization here, a well-intentioned com-
pany there, either of which makes limited amounts of
money available for the protection of relatively small tracts
of land.

The Importance of Productivity-Enhancing

Investment
Funding for habitat protection, channeled through the GEF,
a CBD secretariat (if and when it is established), or some other
organization, could be increased substantially and a true
market could be set up for carbon sequestration services. Cer-
tainly in relation to overall capital flows, though, the impacts
of all this would be minor. This is a problem because if any-
thing less than a great deal of money is raised the impetus
for deforestation will not diminish significantly.
Investing in Human and Social Capital 139

Although the number of books and articles addressing sus-

tainable economic development is growing large, the reasons
why destruction of forests and other natural resources occurs
on a grand scale in poor countries remain the subject of active
debate. One explanation, which is stressed in the environ-
mental economics literature, is that standard indicators of
macroeconomic performance, which are used routinely by
decision-makers in and outside government, are flawed.
As a number of economists have observed, one of the most
commonly used macroeconomic indicators, net national
product (NNP), does not treat physical capital evenhandedly
(Solow, 1992). To arrive at NNP, deductions are made for the
depreciation of equipment, buildings, and other assets made
by human beings. However, depletion of oil deposits, stand-
ing timber, and other kinds of environmental wealth is en-
tirely neglected. This means that a country can raise NNP,
conventionally defined, in the short run by exhausting its
natural resources, even though its long-term economic pros-
pects might be severely impaired in the process. To illustrate
the bias of standard macroeconomic measures, Repetto et al.
(1989) have demonstrated that much of the Indonesian econ-
omy's apparent expansion through the middle 1980s resulted
from deforestation, soil erosion, and, above all else, the de-
pletion of fossil fuel deposits; when natural resource ac-
counting is done, actual rates of growth turn out to have been
much lower.
There have been other refinements in the conceptualiza-
tion and measurement of economic progress. Although the
role of natural resources is not stressed in many of the recent
contributions to the literature on economic development,
those contributions provide a framework for understanding
why environmental degradation, generally, and habitat loss,
specifically, are often excessive.
For example, Romer (1986) has examined the shortcom-
ings of neoclassical growth models, originally developed
during the 1950s. Underlying the neoclassical approach is
the idea that improvements in living standards over time are
linked to the accumulation of productive assets, which is rea-
sonable enough. Moreover, capital has usually been treated
as a homogeneous quantity, something that is manufactured
and physical more often than not. However, the empirical
models derived from theorists' work lack explanatory power.
That is, attempts to estimate linkages between different na-
tions' growth rates and their respective endowments of ma-
140 An Integrated Strategy for Habitat Protection and Economic Progress

chinery and other manufactured capital have not met with

much success. Romer (1986) contends that poor empirical re-
sults can often be traced to the neglect of less tangible assets,
like human capital.
No cause of poor economic performance in Latin America
and other parts of the developing world is more important
than inadequate human capital formation. More often than
not, providing too little support for education and health and
sanitation services, which are all especially deficient in rural
areas, has resulted from a fundamental misallocation of pub-
lic sector resources. Over the years, enormous sums have
been spent in Latin America on consumption subsidies (most
of which are captured by relatively affluent households), the
military, parastatal corporations, and so forth. Inevitably, ed-
ucation, health, and sanitation expenditures have suffered.
Consequently, there are now large numbers of people in the
region who do not have the skills required for more remu-
nerative and productive employment.
Aside from poverty and lost opportunities for economic
growth, there is an environmental price to be paid where ed-
ucation and related human capital investments are not what
they should be. Environmentally sound production of agri-
cultural commodities and timber, it must be remembered,
often requires the sophisticated understanding and manage-
ment of biological processes. Obviously, adoption of im-
proved systems is discouraged if knowledge is lacking. For
example, Barreto et al. (in press) identify scarce technical
information as a major constraint on improved forest man-
agement in the eastern Amazon. Similarly, Browder (1997)
emphasizes that accelerated training will be needed if agro-
forestry is to take hold in Rondonia and other parts of Brazil.
At a more general level, human capital formation is essential
to achieve the sustainable yield increases needed to forestall
the agricultural economy's expansion onto land now covered
with natural vegetation.
Environmental impacts also arise because inadequate sup-
port for education and health and sanitation services in rural
areas impedes the movement of labor out of agriculture and
into other sectors of the economy, typically into lines of work
that involve less exploitation of natural resources. In partic-
ular, many rural people whose meager skills circumscribe
their nonfarm employment possibilities find that their best
option is to resort to raising crops and perhaps some livestock
along hillsides, in rain forests, and in other fragile environ-
ments.
Investing in Human and Social Capital 141

It bears repeating that mining nutrients in areas where

forests are giving way to cropland and pasture is a rela-
tively attractive option for Latin America's rural poor. By
no means does this imply that the returns they derive
from agricultural colonization are high. Except in advan-
tageously located settlement projects, colonists' household
incomes in the Brazilian Amazon exceed the minimum
wage, which in rural areas amounts to less than $100 per
month, only by a factor of two or three (FAO/UNDP/
MARA, 1992, cited in Schneider, 1995, p. 51). The returns
captured by agents of deforestation in northwestern Ecua-
dor are similarly marginal. In a survey of approximately
165 colonists carried out in 1991, Southgate et al. (1992)
found that annual net cash flows in the region averaged
just $25 per hectare. Of those settlers willing to name a
price, the vast majority indicated that they would part
with their holdings if offered just $300 per hectare.
That colonists indeed find it worth their while to migrate
to remote frontiers because they lack the human capital re-
quired for more lucrative employment is revealed by surveys
carried out in Ecuador. In the northeastern part of the coun-
try, average educational attainment for rural adults amounts
to less than three years (Pichon, 1997). This is comparable to
what Southgate et al. (1992) have found among colonist
households in northwestern Ecuador.
Although it is of paramount importance, human capital
scarcity is neither the only cause of rural poverty, nor the
only reason why overexploitation of renewable resources is
a primary feature of economic activity in the Latin American
countryside. Resource degradation also results from the un-
derdevelopment of what economists refer to as social capital,
which is defined to include the rule of law, courts, and re-
lated institutions that exist to make sure that contracts and
property rights are respected. Without those institutions, of
course, markets function anemically or not at all.
The role of social capital has received a lot of attention in
the recent literature on growth and development. For exam-
ple, Olson (1996) has observed that discrepancies between
poor countries' economic performance and how rich coun-
tries are faring cannot be explained only in terms of average
educational levels, access to technology, the availability of
machinery, and other kinds of human and physical capital.
Slow growth and low living standards in places like the for-
mer Soviet Union, he argues, are caused mainly by the waste
and misallocation of existing assets, which in turn have to
142 An Integrated Strategy for Habitat Protection and Economic Progress

do with the feebleness of capitalism's undergirding institu-

tions.
Although Olson (1996) does not focus specifically on nat-
ural resources, his work furnishes an intriguing perspective
on why there is not enough environmentally sustainable eco-
nomic progress in the Latin American countryside. There is
no denying that the resource endowments of Brazil, Vene-
zuela, and many other countries in the Western Hemisphere
are superior to what one finds in many parts of Asia and
Africa, and of Europe, for that matter. However, social capital
remains poorly articulated, which inevitably has environ-
mental consequences.
The contribution that weak institutions make to tropical
deforestation has been examined by Deacon (1994). He points
out that poor countries' governments are often unstable, as
reflected by frequent coups d'etat and constitutional changes,
and tend to do a poor job of enforcing property rights. Fur-
thermore, lawlessness, plus a general lack of governmental
accountability, leads to excessive habitat loss, as an econo-
metric analysis of land clearing in 120 countries demon-
strates. Deacon's (1994) results corroborate the findings of a
more narrowly focused study, in which settlers in the Ecu-
adorian Amazon were found to respond to the government's
slowness in processing applications for formal land tenure
by claiming land in the time-honored way—by clearing away
natural vegetation in order to exercise agricultural use rights
(Southgate, Sierra, and Brown, 1991).
By contrast, strong institutions make sustainable resource
management possible. For example, the success that ethnic
Japanese communities have enjoyed with agroforestry in the
Brazilian Amazon is related in part to the cooperative organ-
izations they have established, which provide technical
assistance, credit, market information, and other services
(Subler and Uhl, 1990).
Pearce and Warford (1993) emphasize that the factors driv-
ing habitat loss and other forms of environmental degrada-
tion in poor countries are highly complex and require much
more investigation. No simple explanation (e.g., that poverty
causes resource depletion, or vice versa) comes close to being
universally applicable. It is obvious, though, that there is a
common set of causes leading both to poor economic perfor-
mance (and rural poverty, in particular) and to the deterio-
ration of renewable natural resources. Assigning specific
weights to inadequate human capital and underdeveloped
Investing in Human and Social Capital 143

social capital may not be possible. Suffice it to say that the

two are interrelated, and that they jointly lead to the exces-
sive depletion of natural resources just as surely as they com-
bine to induce the waste and misallocation of physical
capital made by human hands.

The Lessons to Be Learned in El Salvador about

Environmental Depletion and Conservation
Nowhere are the consequences of meager human and social
capital more clearly on display than in El Salvador, which,
with a national territory of only 21,041 square kilometers, is
the smallest country on the American mainland.
Well into the 1980s, human fertility, which is closely
linked to female illiteracy, was very high in El Salvador. But
because of emigration, population growth was fairly modest,
averaging just 1.8 percent per annum between the 1971 and
1992 censuses (DGEC, 1977 and 1995). But since the civil war
ended, fewer people have left the country and demographic
expansion has accelerated. From 1992 through 1997, the an-
nual growth rate has been a little less than 3 percent (DGEC,
1996). Even though one out of every six or seven Salvadorans
now lives in the United States, the country is more densely
populated, with 250 people per square kilometer, than any
other independent state in the Western Hemisphere, aside
from a few small Caribbean islands (Panayotou, Paris, and
Restrepo, 1997).
Economic performance in El Salvador has been positive in
recent years, with structural adjustment having begun in ear-
nest around 1990. Responding to the commercial opportu-
nities created by freer markets, the Salvadoran economy grew
by 6 percent per annum through 1995; the rate of annual
expansion reached a peak of more than 7 percent immedi-
ately after the 1992 peace accords, which brought an end to
a war that had lasted for more than a decade (Panayotou,
Paris, and Restrepo, 1997).
However, economic growth is showing signs of flagging,
the rate for 1996 amounting to only 3 percent. Some of the
blame can be assigned to the exchange rate distortions re-
sulting from remittances sent home by emigrants. These re-
mittances exceed what the country earns from overseas sales
of coffee, which is the leading national export (other than
people), and, according to local bankers, have caused the Sal-
vadoran colon to be overvalued by approximately 30 percent.
144 An Integrated Strategy for Habitat Protection and Economic Progress

As a result, incentives to compete in international markets

have been weakened and imported substitutes for locally
produced commodities are artificially cheap.
But emigration, remittances, and exchange rate distortions
do not fully explain why El Salvador's domestic economy is
not growing much more rapidly than the country's popula-
tion is. Especially in rural areas, major inefficiencies result
because misguided public policies and a dearth of social cap-
ital cause markets for various factors of production to be un-
duly rigid. Human capital scarcity takes a severe toll as well.
One public policy that has turned out to have adverse con-
sequences is agrarian reform. Undertaken to alleviate social
ferment in the countryside, it has had the effect of keeping
prime farmland tied up in relatively large, unproductive
holdings. About 30 percent of the country's agricultural real
estate is controlled by agrarian reform cooperatives estab-
lished during the 1980s (Shaw, 1997). Nevertheless, an ag-
ricultural policy expert who works on an AID project in El
Salvador confirms the widely circulated claim that less than
10 percent of the country's crop and livestock output is pro-
duced by cooperatives (Hugo Ramos, personal communica-
tion, 1997). Shaw (1997) reports that 20 percent of El
Salvador's very best farmland is cooperatively owned, and
that three-tenths of that real estate currently sits idle.
Left to their own devices, members of at least some co-
operatives could be counted on to sell off their holdings,
thereby allowing land to be transferred to firms and individ-
uals able to use rural real estate more productively. However,
this sort of exchange is preempted by various laws and reg-
ulations. Only in 1996 and 1997 have sales of cooperative
land been permitted, but only to pay off debts incurred dur-
ing the course of land acquisition (Shaw, 1997).
Another option would be for cooperatives to begin func-
tioning more like regular business enterprises, like corpo-
rations, for example. But for this to happen, major improve-
ments in social capital would be required. As they currently
stand, contract and property law in the Salvadoran country-
side would not provide much "assurance" to shareholders in
a corporation that had evolved from a cooperative. That is,
owners could not count on having their best interests re-
spected consistently by managers and workers, given the
•weakness of courts and related institutions.
Currently, cooperative members appear to be responding
to the social capital vacuum by refusing to invest and by sell-
ing whatever the law says is marketable (i.e., machinery but
Investing in Human and Social Capital 145

not land). When driving through the flat, fertile valleys east
of San Salvador, the national capital, one is struck by the
sight of primary irrigation canals leading up to the borders
of extensive cooperatives, which have not organized their
members to dig the secondary and tertiary canals needed to
distribute water to individual fields. Likewise, farm equip-
ment seems to have become extremely scarce on the coastal
plains, where rich volcanic soils once supported a thriving
cotton industry but where cooperatives now dominate and
the land is severely underutilized.
A lack of social capital also appears to explain some of the
inefficiencies and rigidities that characterize financial inter-
mediation in El Salvador. The level of participation in rural
credit markets is very low in the country, with only 31 per-
cent of farmers and rural households surveyed in early 1996
reporting having taken out a loan during the preceding five
years (Lopez, 1997). Government regulation is not the prob-
lem, now that interest-rate controls are a thing of the past.
Instead, limited lending in rural areas has a lot to do with
poor business prospects in the agricultural sector, which in
turn are related to an overvalued currency, inflexible real es-
tate markets, and other distortions. Furthermore, restricted
financial intermediation seems to reflect the fact that credit
sources, both formal and informal, are not confident that bor-
rowers can be obliged to repay. This is perhaps inevitable in
a country that has gone through the sort of conflict and social
dislocation that El Salvador recently experienced. (Of course,
the chance that a recalcitrant borrower owns a weapon, or
can easily get hold of one, has to cross a lender's mind from
time to time.)
Undertaking both the policy reforms and the social capital
improvements needed to make factor markets work better
would reduce rural poverty and ease pressure on the natural
resources on which poor people have come to depend. If land
and capital markets were more efficient, real estate would
end up in the hands of its highest and best users, including
productive small farmers, and more investment would take
place. Employment would increase, both on commercial
farms and in the agribusinesses that serve them. This would
benefit the rural poor directly, and would cause at least a few
of them to retire from the hillside plots where they now raise
corn, beans, and a few other crops.
Notwithstanding its importance, however, improving fac-
tor market efficiency is not the only condition needed for
rural development that is socially broad-based as well as en-
146 An Integrated Strategy for Habitat Protection and Economic Progress

vironmentally sustainable. Accelerated human capital for-

mation, too, has to take place.
El Salvador's record in this area is poor. In 1990-91, health
expenditures amounted to only 1.5 percent of gross domestic
product (GDP). At 1.8 percent of GDP, spending on education
was slightly higher, but still compared poorly with the sup-
port being provided in other countries. For example, school-
ing expenditures as a portion of GDP were 3.7 percent in
Chile, 3.8 percent in Thailand, and 6.9 percent in Malaysia
in 1990-91. Indeed, these three countries supported human
capital formation better in 1960, when their standards of liv-
ing were barely better than what El Salvador's are now; ed-
ucation expenditures as a portion of GDP at that time were
2.7 percent in Chile, 2.3 percent in Thailand, and 2.9 percent
in Malaysia (Panayotou, Paris, and Restrepo, 1997).
El Salvador's poor showing in 1990—91 probably had a lot
to do with its elevated military expenditures. Since then, hu-
man capital formation has claimed a greater share of govern-
ment funds, and local control of schools has been increased
so that teacher absenteeism can be reduced and other mea-
sures for enhancing educational quality can be applied. How-
ever, the legacy of underinvestment endures. The country's
human development index, which the United Nations cal-
culates using data on average educational attainment, life
expectancy and nutritional status, and related factors, is in-
distinguishable from the index for Honduras, which is the
poorest country in Central and South America. Only in Haiti
are conditions appreciably worse (Panayotou, Faris, and Res-
trepo, 1997).
Education and health and sanitation services are espe-
cially deficient outside San Salvador and other major cities,
and the incidence of poverty in rural areas is well above the
national rate, which is 50 percent (Lopez, 1997). An econo-
metric analysis suggests that meager human capital does not
greatly affect the income received either by small farmers or
landless agricultural laborers (Lopez, 1997). In addition, Pa-
giola and Dixon (1997) contend that the shortage of skills in
rural areas does not greatly impede the adoption of conser-
vation tillage and other erosion control measures. Of far
greater importance are the impacts on intersectoral labor
movements as well as on small farmers' locational decisions.
Lopez's (1997) econometric study reveals that the chances of
finding nonagricultural work, which pays better, are posi-
tively related to educational attainment.
Rural families with limited employment prospects are par-
ticularly likely to make a go of farming on a small hillside
Investing in Human and Social Capital 147

plot. More of them are now doing exactly that in northern El

Salvador, which was the scene of some of the worst fighting
during the war, and which is where the steepest land in the
country is found. Since the 1992 peace accords were signed,
small farmers have been drifting back to the area, which had
experienced a certain amount of land abandonment and for-
est regeneration.
Information about demographic trends in El Salvador is
spotty. Especially outside cities and towns, carrying out a
census in 1992, just as the civil war was ending, was difficult.
In addition, several rural administrative districts, called
municipios, have been recategorized as urban districts. Thus,
census results indicating that a contraction in the rural pop-
ulations of El Salvador's three northern departments oc-
curred in the years between the last two censuses are almost
certainly exaggerated (see Table 9.1). Regardless, it is clear
that there has been a reversal since 1992. Although separate
projections are not available for urban and rural areas (DGEC,
1996), it is reasonable to conclude that some of the popula-
tion growth that has taken place in Cabanas, Chalatenango,
and Morazan reflects an increase in the farming population.
The increase may not be as great as rural demographic ex-
pansion has been in other parts of the country, but it is taking
place nonetheless (see Table 9.1).
Resettlement of the countryside is allowing El Salvador to
continue holding the distinction of being the Western Hemi-
sphere's second most denuded country, after Haiti. Almost
no tree-covered land is left for farmers to clear. Current mea-
sures of forest area include coffee plantations, where shade
trees have been planted or have grown of their own accord.
Not counting these plantations, which occupy 10 percent of
the national territory, no more than 2 percent of El Salvador
is forested (Panayotou, Faris, and Restrepo, 1997).

The Inextricable Ties Binding Habitat Protection to

Economic Improvement in the Countryside
Bleak though the situation in El Salvador is in many respects,
environmentalists can draw a positive lesson of great impor-
tance from the country's problems, which is that improving
the economic prospects of the rural poor can contribute sub-
stantially to the conservation of Latin America's natural hab-
itats. That is, investing in human and social capital in rural
areas should enable more individuals to find more remuner-
ative jobs, usually outside agriculture; fewer people should
find themselves relegated to subsisting on a small hillside
148 An Integrated Strategy for Habitat Protection and Economic Progress

Table 9.1 Annual Population Growth (%) in El

Salvador's Three Northern Departments
1971-1992 1992-1997

Department Urban Rural Total Total

Cabanas 3.24 -0.54 0.33 1.64

Chalatenango 1.15 -0.31 0.13 1.66
Morazan 1.86 -0.35 0.14 1.30
Entire nation 2.97 0.80 1.77 2.91

SOURCES: Data from DGEC, 1977; DGEC, 1995; DGEC, 1996

holding, migrating to an urban slum in the hope of finding

some sort of employment (informal more often than not), or
settling on an agricultural frontier.
By the same token, one cannot be optimistic about eco-
nomic progress and habitat conservation in the Latin Amer-
ican countryside if human and social capital remain scarce.
In general, poverty will continue to be widespread in the
countryside and the rural poor will take advantage of each
and every opportunity that comes their way to make money
by depleting renewable resources. Even where it is made
available, full advantage will not be taken of technology re-
quired for the sustainable intensification of crop and live-
stock production.
Nor will it be possible to deal with the market failure of
deforestation if the plight of the rural poor is not addressed
more effectively. Efforts to pay for carbon sequestration and
other environmental services provided by forests will be se-
verely hampered by the costs of preventing encroachment by
agricultural colonists. Likewise, initiatives to police park
boundaries and to promote nature-based tourism and the sus-
tainable development of forest resources will not withstand
the human onslaught unleashed if poverty continues to grip
the countryside.
Those who suppose that habitats can be protected merely
by establishing nature reserves should keep in mind the ge-
ographic lengths to which poverty has driven the Salvadoran
peasantry in recent decades. During the 1960s, tens of
thousands pushed their way into Honduras in order to carve
farms out of the neighboring country's forests; military action
was required to expel them. Twenty years later, a much larger
migration to the United States occurred, in spite of the efforts
Investing in Human and Social Capital 149

of the Immigration and Naturalization Service. If rural people

who are desperate to support themselves and their families
pay little heed to national frontiers, how much respect are
they ever likely to show for park boundaries?
The economic activities examined in this book can help
put a halt to deforestation. Under certain circumstances, na-
ture-based tourism and the sustainable harvesting of timber
and other forest products have a role to play. Certainly, peo-
ple who have moved to agricultural frontiers should be
encouraged to adopt agroforestry and other activities char-
acterized more by resource management than by resource
mining. However, natural habitats will not be safeguarded
merely by fostering environmentally sound lines of work in
tree-covered hinterlands. Indeed, since protection of the non-
market environmental services that tropical forests yield is a
paramount concern, perhaps it is best to reduce, rather than
to enhance, the incentives to engage in all classes of eco-
nomic activity, sustainable as well as depletive, in those ar-
eas. The aim, in other words, should be to make natural
habitats less, not more, appealing as venues for subsistence
and commerce.
Undertaking the investments required for agricultural in-
tensification, which allows for more crops and livestock to
be produced on less land, will lower the returns of frontier
colonization. Complementary formation of human capital,
which permits more people to find productive employment
that does little or no damage to natural resources, raises the
opportunity costs associated with being an agent of defores-
tation. Of course, strengthening capitalism's underpinning
institutions and other forms of social capital stimulates all
kinds of productivity-enhancing investment, which is abso-
lutely essential for sustainable economic progress in the
Latin American countryside and anywhere else in the world.
This page intentionally left blank
Abbreviations

AAAS American Association for the Advancement of

Science
AID U.S. Agency for International Development
CBD Convention on Biological Diversity
CETUR Corporacion Ecuatoriana de Turismo
(Ecuadorian Tourism Corporation)
CGIAR Consultative Group for International
Agricultural Research
CI Conservation International
CIDESA Fundacion de Capacitacion e Inversion para el
Desarrollo Socio-Ambiental (Foundation for
Training and Investment for Socio-
Environmental Development)
COFYAL Cooperativa Forestal Yanesha Limitada
(Yanesha Forestry Cooperative Limited)
CRP Conservation Reserve Program
CSRMP Central Selva Resource Management Project
dbh diameter (of a tree) at breast height
DRNR Direccion de Recursos Naturales Renovables
(Venezuelan Directorate of Renewable Natural
Resources)
FPCN Fundacion Peruana para la Conservacion de la
Naturaleza (Peruvian Foundation for the
Conservation of Nature)
GDP gross domestic product
GEF Global Environmental Facility
GPS global positioning system
GTZ Gesellschaft fur Technische Zusammenarbeit
GmbH (German Agency for Technical
Assistance)

151
152 Abbreviations

IBGE Institute Brasileiro de Geografia e Estatistica

(Brazilian Institute of Geography and Statistics)
ICDP Integrated Conservation and Development
Project
ICT Institute Costarricense de Turismo (Costa Rican
Tourism Institute)
IDE Inter-American Development Bank
IFC International Finance Corporation
IMAZON Institute do Homem e Meio Ambiente da
Amazonia (Institute for Man and the
Environment of the Amazon)
INBio Institute Nacional de Biodiversidad (National
Biodiversity Institute of Costa Rica)
INCRA Institute Nacional de Colonizagao e Reforma
Agraria (Brazil's National Institute for
Colonization and Agrarian Reform)
INEFAN Institute Ecuatoriano Forestal, de Areas
Naturales, y de Vida Silvestre (Ecuadorian
Institute of Forestry, Natural Areas, and
Wildlife)
INS U.S. Immigration and Naturalization Service
IUCN International Union for the Conservation of
Nature
NNP net national product
ODA Overseas Development Agency
OECD Organization for Economic Cooperation and
Development
PEPP Proyecto Especial Pichis-Palcazii (Pichis-
Palcazu Special Project)
SCS U.S. Soil Conservation Service
SPN Servicio de Parques Nacionales (Costa Rica's
National Park Service)
SPNG Servicio del Parque Nacional Galapagos
(Galapagos National Park Service)
SRP Subsecretaria de Recursos Pesqueros (Ecuador's
Subsecretariat of Fishery Resources)
SUDAM Superintendencia de Desenvolvimento da
Amazonia (Superintendency for Amazonian
Development)
TNC The Nature Conservancy
TSC Tropical Science Center
UNESCO United Nations Educational, Scientific, and
Cultural Organization
WWF World Wildlife Fund
References

Acosta-Solis, M. 1944. La Tagua. Quito: Editorial Ecuador.

Allegretti, M. 1990. "Extractive Reserves: An Alternative for Rec-
onciling Development and Environmental Conservation in
Amazonia" in A. Anderson (ed.), Alternatives to Deforestation:
Steps toward Sustainable Use of the Amazon Rain Forest. New
York: Columbia University Press.
Alston, L., G. Libecap, and R. Schneider. 1996. "The Determinants
and Impact of Property Rights: Land Titles on the Brazilian
Frontier," Journal of Law, Economics, and Organization 12:1,
pp. 25-61.
Amgen Inc. 1997. 1996 Annual Report. Thousand Oaks, Calif.
Anderson, A. 1989. "Land Use Strategies for Successful Extractive
Economies." Paper read at National Wildlife Federation Sym-
posium on Extractive Economies in Tropical Forests, Washing-
ton, D.C.
Anderson, A. and E. loris. 1992. "The Logic of Extraction: Resource
Management and Income Generation by Extractive Populations
in the Amazon Estuary" in K. Redford and C. Padoch (eds.),
Conservation of Neotropical Forests: Working from Traditional
Resource Use. New York: Columbia University Press.
Asebey, E. and J. Kempenaar. 1995. "Biodiversity Prospecting: Ful-
filling the Mandate of the Biodiversity Convention," Vanderbilt
Journal of Transnational Law 28:4, pp. 703-754.
Aylward, B. 1993. "The Economic Value of Pharmaceutical Pros-
pecting and its Role in Biodiversity Conservation" (discussion
paper 93—05), London Environmental Economics Centre.
Aylward, B., K. Allen, J. Echeverria, and J. Tosi. 1996. "Sustainable
Ecotourism in Costa Rica: The Monteverde Cloud Forest Pre-
serve," Biodiversity Conservation 5:3, pp. 315—343.
Baldares, M. and J. Laarman. 1990. "Derechos de Entrada a las Areas
Protegidas de Costa Rica," Ciencias Economicas 10:1, pp. 63-
82.

153
154 References

Banco Central del Ecuador (BCE). 1995. Information Estadistica

Mensual. Quito.
Barfod, A. 1991. "A Monographic Study of the Subfamily Phytele-
phantoidae," Opera Botanica 105, pp. 1-73.
Barreto, P., P. Amaral, E. Vidal, and C. Uhl. In press. "Costs and
Benefits of Forest Management for Timber Production in East-
ern Amazonia," Forest Ecology and Management.
Barros, A. and C. Uhl. 1995. "Logging along the Amazon River and
Estuary: Patterns, Problems, and Potential," Forest Ecology and
Management 77, pp. 87-105.
Benavides, M. and M. Pariona. 1995. "The Yanesha Forestry Co-
operative and Community-Based Management in the Central
Peruvian Forest" in Proceedings of Symposium on Forestry in
the Americas: Community-Based Management and Sustaina-
bility. Madison: University of Wisconsin, Land Tenure Center.
Bermudez, F. 1992. "Evolucion del Turismo en las Areas Silvestres,
Periodo 1982—1991," Ministerio de Recursos Naturales, Ener-
gia, y Minas, Servicio de Parques Nacionales, San Jose, Costa
Rica.
Bermudez, F. 1995. Unpublished national parks visitation data,
from Ministerio de Recursos Naturales, Energia, y Minas, Ser-
vicio de Parques Nacionales, San Jose.
Bezanson, K. and R. Mendez. 1995. "Alternative Funding: Looking
beyond the Nation-State," Futures 27:2, pp. 223-229.
Binswanger, H. 1989. "Brazilian Policies that Encourage Defores-
tation in the Amazon" (working paper 16), Environment De-
partment, World Bank, Washington, B.C.
Boo, E. 1990. Ecotourism: The Potentials and Pitfalls. Washington,
D.C.: World Wildlife Fund.
Bray, D., M. Carreon, L. Merino, and V. Santos. 1993. "On the Road
to Sustainable Forestry," Cultural Survival Quarterly 17:1, pp.
38-41.
Bromley, R. 1981. "The Colonization of Humid Tropical Areas in
Ecuador," Singapore Journal of Tropical Geography 2:1, pp.
15-26.
Brooke, J. 1993. "Galapagos Burden: Goats, Pigs, and Now People,"
New York Times, 30 September, p. A4.
Browder, J. 1988. "Public Policy and Deforestation in the Brazilian
Amazon" in R. Repetto and M. Gillis (eds.), Public Policies and
the Misuse of Forest Resources. Cambridge: Cambridge Univer-
sity Press.
Browder, J. 1992a. "Extractive Reserves and the Future of the Am-
azon's Rainforests: Some Cautionary Observations" in S. Coun-
sell and T. Rice (eds.), The Rainforest Harvest: Sustainable
Strategies for Saving Tropical Forests. London: Friends of the
Earth Trust.
Browder, J. 1992b. "The Limits of Extractivism: Tropical Forest
Strategies beyond Extractive Reserves," Bioscience 42:3, pp.
174-181.
References 155

Browder, J. 1997. "Planting the Seeds of Change in the Ashes of the

Rainforest: The Rondonia Agroforestry Pilot Project," College
of Architecture and Urban Studies, Virginia Polytechnic Insti-
tute and State University, Blacksburg.
Browder, J. and B. Godfrey. 1997. Rainforest Cities: The Urban
Transformation of Brazilian Amazonia. New York: Columbia
University Press.
Browder, J., E. Matricardi, and W. Abdala. 1996. "Is Sustainable
Tropical Timber Production Financially Viable? A Compara-
tive Analysis of Mahogany Silviculture among Small Farmers
in the Brazilian Amazon," Ecological Economics 16, pp. 147-
159.
Brown, K. and D. Pearce. 1994. "The Economic Value of Non-
Market Benefits of Tropical Forests: Carbon Storage" in J. Weiss
(ed.), The Economics of Project Appraisal and the Environment.
London: Edward Elgar.
Burton, T. 1994. "Drug Company Looks to 'Witch Doctors' to Con-
jure Profits," Wall Street Journal, 7 July, pp. Al and A8.
Calero-Hidalgo, R. 1992. "The Tagua Initiative in Ecuador: A Com-
munity Approach to Tropical Rain Forest Conservation and De-
velopment" in M. Plotkin and L. Famolare (eds.), Sustainable
Harvest and Marketing of Rain Forest Products. Washington,
D.C.: Island Press.
Capistrano, A. 1994. "Tropical Forest Depletion and the Changing
Macroeconomy, 1967-85" in K. Brown and D. Pearce (eds.) The
Causes of Tropical Deforestation: The Economic and Statistical
Analysis of Factors Giving Rise to the Loss of the Tropical For-
ests. London: University College London Press.
Chase, L. 1995. "Capturing the Benefits of Ecotourism: The Econom-
ics of National Park Entrance Fees in Costa Rica" (M.S. thesis),
Department of Agricultural, Resource, and Managerial Econom-
ics, Cornell University, Ithaca, N.Y.
Chomitz, K. and D. Gray. 1995. "Roads, Lands, Markets, and Defor-
estation: A Spatial Model of Land Use in Belize" (policy re-
search working paper 1444), World Bank, Washington, D.C.
Chomitz, K. and K. Kumari. 1996. "The Domestic Benefits of Trop-
ical Forests: A Critical Review Emphasizing Hydrological Func-
tions" (policy research working paper 1601), World Bank,
Washington, D.C.
Colchester, M. and L. Lohmann. 1993. The Struggle for Land and
the Fate of the Forests. London: Zed Books.
Coles-Ritchie, M. 1996. "Analysis of Non-Timber Extractive Prod-
ucts from Tropical Forests: The Tagua Example in Ecuador"
(M.S. thesis), Graduate School of International Studies, Bard
College, Annandale-on-Hudson, N.Y.
Coomes, O. 1995. "A Century of Rain Forest Use in Western Ama-
zonia: Lessons for Extraction-Based Conservation of Tropical
Forest Resources," Forest and Conservation History 39:3, pp.
108-120.
156 References

Crossette, B. 1996. "Report Blames Poor Farmers for Depleting

World Forests," New York Times, 4 August, p. 8.
Darwin, C. 1859. On the Origin of the Species by Means of Natural
Selection. London: John Murray.
Deacon, R. 1994. "Deforestation and the Rule of Law in a Cross-
Section of Countries," Land Economics 70:4, pp. 414-430.
de Groot, R. 1983. "Tourism and Conservation in the Galapagos Is-
lands," Biological Conservation 26:4, pp. 291-300.
de Miras, C. 1994. "Las Islas Galapagos: Un Reto Economico y Tres
Contradicciones Basicas," Institut Francais de Recherche
Scientifique pour le Developpement en Cooperacion, Quito.
Denevan, W., J. Treacy, J. Alcorn, C. Padoch, J. Denslow, and S.
Flores-Paitan. 1984. "Indigenous Agroforestry in the Peruvian
Amazon: Bora Indian Management of Swidden Fallows," Inter-
ciencia 9, pp. 346-357.
DHV Consultants BV. 1992. "Biodiversity Protection and Invest-
ment Needs for the Minimum Conservation System in Costa
Rica" (report to the World Bank), Amersfoort.
Direccion General de Estadistica y Censos (DGEC). 1977. IV Censo
Nacional de Poblacion, 1971, San Salvador: Ministerio de
Economia.
Direccion General de Estadistica y Censos (DGEC). 1995. V Censo
Nacional de Poblacion, 1992. San Salvador: Ministerio de
Economia.
Direccion General de Estadistica y Censos (DGEC). 1996. Proyec-
ciones de la Poblacion de El Salvador, 1995-2025. San Salva-
dor: Ministerio de Economia.
Dixon, J. and P. Sherman. 1990. Economics of Protected Areas: A
New Look at Benefits and Costs. Washington, D.C.: IslandPress.
Dixon, R., S. Brown, R. Houghton, A. Solomon, M. Trexler, and J.
Wisniewski. 1994. "Carbon Pools and Flux of Global Forest Ec-
osystems," Science 263, pp. 185-190.
Drake, S. 1991. "Local Participation in Ecotourism Projects" in T.
Whelan (ed.), Nature Tourism: Managing for the Environment.
Washington, D.C.: Island Press.
Echeverria, J., M. Hanrahan, and R. Solorzano. 1995. "Valuation of
Non-Priced Amenities Provided by the Biological Resources
within the Monteverde Cloud Forest Preserve, Costa Rica," Ec-
ological Economics 13, pp. 43-52.
Economist. 1997a. "A Flood of Fiascos," 19 April, p. 80.
Economist. 1997b. "Galapagos: Too Many People," 10 May, p. 44.
Edwards, S. 1991. "The Demand for Galapagos Vacations: Estima-
tion and Application to Conservation," Coastal Management
19:2, pp. 155-169.
Elgegren, J. 1993. "Desarrollo Sustentable y Manejo de Bosques Na-
turales en la Amazonia Peruana: Un Estudio Economico-
Ambiental del Sistema de Manejo Forestal en Fajas en el Valle
del Palcazu" (M.S. thesis), Facultad Latinoamericana de Cien-
cias Sociales, Quito.
References 157

Fankhauser, S. and D. Pearce. 1994. "The Social Costs of Green-

house Gas Emissions" in The Economics of Climate Change.
Paris: Organization for Economic Cooperation and Develop-
ment.
Farnsworth, N. and D. Soejarto. 1985. "Potential Consequences of
Plant Extinction in the United States on the Current and Future
Availability of Prescription Drugs," Economic Botany 39:3, pp.
231-240.
Ferreira, V. and J. Paschoalino. 1987. "Pesquisa sobre Palmito no
Institute de Tecnologia de Alimentos" in Proceedings from the
First National Conference of Researches on Palm Hearts. Cur-
itiba: Impresa Brasileira de Pesquisa Agropecuaria.
Fundagao Instituto Brasileiro de Geografia e Estatistica (FIBGE).
1982. Censo Demogrdfico de 1980: Acre, Amazonas, Para, Ro-
raima, Amapd, Rondonia. Rio de Janeiro.
Figueroa, L. 1995. "Analisis del Impacto Economico del Turismo
sobre la Comunidad y sobre la Reserva Biologica Bosque Nu-
boso Monteverde" (report to Tropical Science Center), Servi-
cios Corporativos Emanuel S.A., San Jose.
Food and Agriculture Organization of the United Nations, United
Nations Development Program, and Ministerio da Agricultura
e Reforma Agraria (FAO/UNDP/MARA). 1992. Principals In-
dicadores Socio-Economicos dos Assentamentos de Reforma
Agraria (project report BRA-87/022). Brasilia.
Goulding, M., N. Smith, and D. Mahar. 1996. Floods of Fortune:
Ecology and Economy along the Amazon. New York: Columbia
University Press.
Gradwohl, J. and R. Greenberg. 1988. Saving the Tropical Forests.
London: Earthscan Publications.
Hamilton, L. and P. King. 1983. Tropical Forested Watersheds: Hy-
drologic and Soils Response to Major Uses or Conversions.
Boulder: Westview Press.
Hartshorn, G. 1978. "Tree Falls and Tropical Forest Dynamics" in
P. Tomlinson and M. Zimmermann (eds.), Tropical Trees as
Living Systems. Cambridge: Cambridge University Press.
Hartshorn, G. 1990. "Natural Forest Management by the Yanesha
Forestry Cooperative in Peruvian Amazonia" in A. Anderson
(ed.), Alternatives to Deforestation: Steps toward Sustainable
Use of the Amazon Rain Forest. New York: Columbia Univer-
sity Press.
Hartshorn, G., R. Simeone, and J. Tosi. 1986. "Manejo para el Ren-
dimiento Sostenido de Bosques Naturales: Un Sinopsis del
Proyecto Desarrollo del Palcazu en la Selva Central de la Ama-
zonia Peruana," Tropical Science Center, San Jose.
Harvard Business School. 1992. "INBio/Merck Agreement: Pio-
neers in Sustainable Development" (case study NI-593-015),
Boston.
Heath, J. and H. Binswanger. 1996. "Natural Resource Degradation
Effects of Poverty and Population Growth Are Largely Policy
158 References

Induced: The Case of Colombia," Environment and Develop-

ment Economics 1:1, pp. 65-84.
Hecht, S., R. Norgaard, and G. Possio. N.d. "The Economics of Cattle
Ranching in Eastern Amazonia," Graduate School of Architec-
ture and Urban Planning, University of California, Los Angeles.
Houghton, }., L. Meiro-Filho, B. Callander, N. Harris, A. Kattenberg,
and K. Maskell. 1996. Climate Change 1995: The Science of
Climate Change. Cambridge: Cambridge University Press.
Howard, A. and J. Magretta. 1995. "Surviving Success: An Interview
with The Nature Conservancy's John Sawhill," Harvard Busi-
ness Review 73:5, pp. 109-118.
Huber, R. 1996. "Case Studies Showing Costs and Benefits of Eco-
tourism and Cultural Heritage Protection," Sixth Caribbean
Conference on Ecotourism, Point-a-Pitre, Guadeloupe.
Institute Costarricense de Turismo (ICT). 1994. Encuesta Aerea de
Extranjeros: Epoca Alta Turistica, 1994. San Jose.
Institute Costarricense de Turismo (ICT). 1995. Anuario Estadistico
de Turismo, 1994. San Jose.
Institute Nacional de Estadistica y Censos (INEC). 1992. "Andlisis
de los Resultados Definitives del V Censo de Poblacion y IV de
Vivienda, Provincia de Galapagos." Quito.
Inter-American Development Bank (IDB). 1994. "Eighth General In-
crease in the Financial Resources of the Inter-American Devel-
opment Bank" (document AB-1704). Washington, D.C.
International Bank for Reconstruction and Development (IBRD).
1978. "Guatemala: Livestock Development Project, Loan 722-
GU" (project termination report), Washington, D.C.
International Bank for Reconstruction and Development (IBRD).
1997. "Inspection Panel Finds Mixed Results in Brazilian Am-
azon Project" (press release) 10 April, Washington, D.C.
International Bank for Reconstruction and Development (IBRD).
1991. The Forestry Sector: A World Bank Policy Paper. Wash-
ington, D.C.
Ito, T. and M. Loftus. 1997. "Cutting and Dealing: Asian Loggers
Target the World's Remaining Rain Forests," U.S. News and
World Report 122:9, pp. 19-41.
Jackson, M. 1990. Galapagos: A Natural History Guide. Calgary:
University of Calgary Press.
Kaimowitz, D. 1996. "Livestock and Deforestation in Central Amer-
ica in the 1980s and 1990s: A Policy Perspective," Center for
International Forestry Research, Jakarta.
Katzman, M. 1977. Cities and Frontiers in Brazil. Cambridge: Har-
vard University Press.
Kramer, R., E. Mercer, and N. Sharma. 1996. "Valuing Tropical Rain
Forest Protection Using the Contingent Valuation Method" in
W. Adamowicz, P. Boxall, M. Luckert, W. Phillips, and W.
White (eds.), Forestry, Economics, and the Environment. Wal-
lingford, England: CAB International.
Laird, S. 1993. "Contracts for Biodiversity Prospecting" in W. Reid,
References 159

S. Laird, C. Meyer, R. Gamez, A. Sittenfeld, D. Jansen, M. Gollin,

and C. Juma (eds.), Biodiversity Prospecting: Using Genetic Re-
sources for Sustainable Development. Baltimore: World Re-
sources Institute Publications.
Lamb, F. 1966. Mahogany of Tropical America: Its Ecology and
Management. Ann Arbor: University of Michigan Press.
Lasser, T. 1974. Flora de Venezuela, 16 vols. Caracas: Institute Bo-
tanico de la Direccion de Recursos Naturales Renovables.
Ledec, G. 1992. "The Role of Bank Credit for Cattle Raising in Fi-
nancing Tropical Deforestation: An Economic Case Study"
(Ph.D. dissertation), Department of Wildlife Resource Science,
University of California, Berkeley.
Lemonick, M. 1995. "Can the Galapagos Survive?," Time Magazine,
30 October, pp. 80-82.
Lopez, R. 1997. "Rural Poverty in El Salvador: A Quantitative Anal-
ysis" (report 16253-ES), World Bank, Washington, D.C.
MacArthur, R. and E. Wilson. 1967. The Theory of Island Biogeog-
raphy. Princeton: Princeton University Press.
Macdonald. T. 1981. "Indigenous Responses to an Expanding Fron-
tier: Jungle Quechua Economic Conversion to Cattle Ranching"
in N. Whitten (ed.), Cultural Transformations and Ethnicity in
Modern Ecuador. Urbana: University of Illinois Press.
Machlis, G., D. Costa, and J. Cardenas-Salazar. 1990. "Estudio del
Visitante a las Islas Galapagos," Fundacion Charles Darwin,
Quito.
Mahar, D. 1989. "Government Policies and Deforestation in Brazil's
Amazon Region," World Bank, Washington, D.C.
Mahmood, K. 1987. "Reservoir Sedimentation: Impact, Extent, and
Mitigation" (technical paper 71), World Bank, Washington, D.C.
Mattos, M. and C. Uhl. 1994. "Economic and Ecological Perspec-
tives on Ranching in the Eastern Amazon in the 1990s," World
Development 22:2, pp. 145-158.
McRae, M. 1997. "Is 'Good Wood' Bad for Forests?," Science 275,
pp. 1868-1869.
Moran, E. (ed.). 1983. The Dilemma of Amazonian Development.
Boulder: Westview Press.
Moran, E. 1989. "Adaptation and Maladaptation in Newly Settled
Areas" and "Government-Directed Settlement in the 1970s: An
Assessment of Transamazon Highway Colonization" in D.
Schumann and W. Partridge (eds.), The Human Ecology of
Tropical Land Settlement in Latin America. Boulder: Westview
Press.
Mountfort, G. 1974. "The Need for Partnership: Tourism and Con-
servation," Development Forum 2:3, pp. 6-7.
Myers, N. 1984. The Primary Source. New York: Norton.
Myers, N. 1988. "Threatened Biotas: Hotspots in Tropical Forests,"
Environmentalist 8:3, pp. 1—20.
Nations, J. 1989. "La Reserva del Biosfera Maya, Peten: Estudio Tec-
nico," Consejo Nacional de Areas Protegidas, Guatemala City.
160 References

Nelson, M. 1973. The Development of Tropical Lands. Baltimore:

Johns Hopkins University Press.
Netherlands Economic Institute (NEI). 1989. "An Import Surcharge
on the Import of Tropical Timber in the European Community:
An Evaluation," Rotterdam.
Newcombe, K. 1995. "Financing Innovations and Instruments Con-
tribution of the Investment Portfolio of the Pilot Phase of the
Global Environmental Facility" in McNeely, J. (ed.), Biodivers-
ity Conservation in the Asia and Pacific Region. Manila: Asian
Development Bank.
Oficina Costarricense de Implementacion Conjunta (OCIC). 1996.
Costa Rica-Norway Reforestation and Forest Conservation AIJ
Pilot Project. San Jose.
Olivera, A. 1995. "Forestry Project of the Indigenous Chiquitano
Communities of Lomerio" in Proceedings of Symposium on
Forestry in the Americas: Community-Based Management and
Sustainability. Madison: University of Wisconsin, Land Tenure
Center.
Olson, M. 1996. "Big Bills Left on the Sidewalk: Why Some Nations
Are Rich and Others Poor," Journal of Economic Perspectives
10:2, pp. 3-24.
Ottaway, M. 1995. "Pick of the Bunch: Costa Rica is Central America
at its Very Best" Sunday Times (London), 19 November, pp.
O • J.—O ,tL,

Pagiola, S. and J. Dixon. 1997. "Land Degradation Problems in El

Salvador" (report 16253-ES), World Bank, Washington, D.C.
Pagiola, S. and J. Kellenberg with L. Vidaeus and J. Srivastava. 1997.
"Mainstreaming Biodiversity in Agricultural Development: To-
ward Good Practice" (environment paper 15), World Bank,
Washington.
Panayotou, T., R. Paris, and C. Restrepo. 1997. El Desafio Salvador-
eno: De la Paz al Desarrollo Sostenible. San Salvador: Funda-
cion Salvadorena para el Desarrollo Economico y Social.
Pearce, D. 1996. "Global Environmental Value and the Tropical For-
ests: Demonstration and Capture" in W. Adamowicz, P. Boxall,
M. Luckert, W. Phillips, and W. White (eds.), Forestry, Econom-
ics, and the Environment. Wallingford, England: CAB Interna-
tional.
Pearce, D. and K. Brown. 1994. "Saving the World's Tropical For-
ests" in K. Brown and D. Pearce (eds.) The Causes of Tropical
Deforestation: The Economic and Statistical Analysis of Factors
Giving Rise to the Loss of the Tropical Forests. London: Uni-
versity College-London Press.
Pearce, D. and S. Puroshothaman. 1995. "The Economic Value of
Plant-Based Pharmaceuticals" in T. Swanson (ed.),Intellectural
Property Rights and Biodiversity Conservation: An Interdisci-
plinary Analysis of the Values of Medicinal Plants. Cambridge:
Cambridge University Press.
Pearce, D. and J. Warford. 1993. World without End: Economics,
References 161

Environment, and Sustainable Development. New York: Oxford

University Press.
Peck, R. 1990. "Promoting Agroforestry Practices among Small Pro-
ducers: The Case of the Coca Agroforestry Project in Amazonian
Ecuador" in A. Anderson (ed.), Alternatives to Deforestation:
Steps toward Sustainable Use of the Amazon Rain Forest. New
York: Columbia University Press.
Peters, C. 1990. "Population Ecology and Management of Forest
Fruit Trees in Peruvian Amazonia" in A. Anderson (ed.), Al-
tematives to Deforestation: Steps toward Sustainable Use of
the Amazon Rain Forest. New York: Columbia University
Press.
Peters, C., A. Gentry, and R. Mendelsohn. 1989. "Valuation of an
Amazon Rainforest," Nature 339, pp. 655-656.
Pichon, F. 1997. "Colonist Land Allocation Decisions: Land Use and
Deforestation in the Ecuadorian Amazon Frontier," Economic
Development and Cultural Change, 45:4, pp. 707-744.
Plotkin.M. 1994. Tales of a Shaman's Apprentice: An Ethnobotanist
Searches for New Medicines in the Amazonian Rainforest. New
York: Viking Penguin.
Pollak, H., M. Mattos, and C. Uhl. 1995. "A Profile of Palm Heart
Extraction in the Amazon Estuary" Human Ecology, 23:3, pp.
357-385.
Principe, P. 1989. "The Economic Significance of Plants and their
Constituents as Drugs" in H. Wagner, H. Hikino, and N. Farns-
worth (eds.), Economic and Medicinal Plant Research, vol. 3.
London: Academic Press.
Redford, K. 1992. "The Empty Forest," Bioscience 42:6, pp. 412-
422.
Reid, W., S. Laird, R. Gamez, A. Sittenfeld, D. Jansen, M. Gollin, and
C. Juma. 1993. "A New Lease on Life" in W. Reid, S. Laird, C.
Meyer, R. Gamez, A. Sittenfeld, D. Jansen, M. Gollin, and C.
Juma (eds.), Biodiversity Prospecting: Using Genetic Resources
for Sustainable Development. Baltimore: World Resources In-
stitute Publications.
Reis, E. and R. Guzman. 1994. "An Econometric Model of Amazon
Deforestation" in K. Brown and D. Pearce (eds.) The Causes of
Tropical Deforestation: The Economic and Statistical Analysis
of Factors Giving Rise to the Loss of the Tropical Forests. Lon-
don: University College London Press.
Repetto, R. and M. Gillis (eds.). 1988. Public Policies and the Misuse
of Forest Resources. Cambridge: Cambridge University Press.
Repetto, R., W. Magrath, M. Wells, C. Beer, and F. Rossini. 1989.
Wasting Assets: Natural Resources in National Income Ac-
counts. Washington, D.C.: World Resources Institute.
Richards, E. 1991. "The Forest Ejidos of Southeast Mexico: A Case
Study of Community-Based Sustained Yield Management,"
Commonwealth Forestry Review 70:4, pp. 290—311.
Robinson, L. 1997. "Latin Robin Hood: Brazil's Squatter Movement
162 References

Pits Church against State," U.S. News and World Report 122:
24, pp. 30-32.
Romanoff, S. 1981. "Analisis de las Condiciones Socioeconomicas
para el Desarrollo Integral de la Amazonia Boliviana" (consult-
ing report), Organization of American States, Washington, B.C.
Romer, P. 1986. "Increasing Returns and Long-Run Growth" Journal
of Political Economy 94:5, pp. 1002-1037.
Rovinski, Y. 1991. "Private Reserves, Parks, and Ecotourism in
Costa Rica" in T. Whelan (ed.), Nature Tourism: Managing for
the Environment. Washington, B.C.: Island Press.
Ruitenbeek, H. 1989. "Social Cost-Benefit Analysis of the Korup
Project, Cameroon" (consulting report), Worldwide Fund for
Nature, London.
Salafsky, N., B. Bugelby, and J. Terborgh. 1992. "Can Extractive Re-
serves Save the Rainforest?," Buke University Center for Trop-
ical Conservation, Burham, N.C.
Salati, E. and P. Bose. 1984. "Amazon Basin: A System in Equilib-
rium," Science 225, pp. 129-138.
Scartezini, A. 1985. Segredos de Medici. Sao Paulo: Editora Marco
Zero.
Schneider, R. 1992. "Brazil: An Analysis of Environmental Prob-
lems in the Amazon" (report 9104-Br), World Bank, Washing-
ton, B.C.
Schneider, R. 1995. "Government and the Economy on the Amazon
Frontier" (environment paper 11), World Bank, Washington,
B.C.
Schwartzman, S. 1989. "Extractive Reserves: The Rubber Tap-
pers' Strategy for Sustainable Use of the Amazon Rain For-
est" in J. Browder (ed.), Fragile Lands of Latin America:
Strategies for Sustainable Development. Boulder, Colorado:
Westview Press.
Seitz, F. 1996. "A Major Beception on Global Warming," Wall Street
Journal, 12 June, p. A16.
Shaw, C. 1997. "Rural Land Markets" (report 16253-ES), World
Bank, Washington, B.C.
Simeone, R. 1990. "Land Use Planning and Forestry-Based Econ-
omy: The Case of the Amuesha Forestry Cooperative," Tebiwa:
The Journal of the Idaho Museum of Natural History 24, pp. 7-
12.
Simpson, B. and R. Sedjo. 1996. "Paying for the Conservation of
Endangered Ecosystems: A Comparison of Birect and Indirect
Approaches," Environment and Development Economics 1:2,
pp. 241-257.
Simpson, D., R. Sedjo, and J. Reid. 1996. "Valuing Biodiversity: An
Application to Genetic Prospecting," Journal of Political Econ-
omy 104:1, pp. 163-185.
Smith, N. 1981. "Colonization Lessons from a Tropical Forests,"
Science 214, pp. 755-761.
Smith, N., T. Fik, P. Alvim, I. Falesi, and E. Serrao. 1995. "Agrofo-
References 163

restry Developments and Potential in the Brazilian Amazon,"

Land Degradation and Rehabilitation 6, pp. 251-263.
Smith, R. 1982. "Las Comunidades Nativas y el Mito del Gran Vacio
Economico: Un Analisis de Planificacion para el Desarrollo en
el Proyecto Especial Pichis Palcazu" (consulting report), U.S.
Agency for International Development, Lima.
Snook, L. 1993. "Stand Dynamics of Mahogany (Swietenia Macro-
phylla King) after Fire and Hurricanes in the Tropical Forests
of the Yucatan Peninsula, Mexico." (Ph.D. dissertation), School
of Forestry, Yale University, New Haven.
Sohngen, B., R.Mendelsohn, R. Sedjo, andK. Lyon. 1997. "An Anal-
ysis of Global Timber Markets" (discussion paper 97-37), Re-
sources for the Future, Washington.
Solow, R. 1992. "An Almost Practical Step toward Sustainability."
Lecture presented at Resources for the Future, Washington, D.C.
Southgate, D. 1994. "Tropical Deforestation and Agricultural De-
velopment in Latin America" in K. Brown and D. Pearce (eds.),
The Causes of Tropical Deforestation: The Economic and Sta-
tistical Analysis of Factors Giving Rise to the Loss of Tropical
Forests. London: University College, London Press.
Southgate, D. and H. Clark. 1993. "Can Conservation Projects Save
Biodiversity in South America?," Ambio 22:2-3, pp. 163-166.
Southgate, D., M. Coles-Ritchie, and P. Salazar-Canelos. 1996. "Can
Tropical Forests Be Saved by Harvesting Non-Timber Products?
A Case Study for Ecuador" in W. Adamowicz, P. Boxall, M.
Luckert, W. Phillips, and W. White (eds.), Forestry, Economics,
and the Environment. Wallingford, England: CAB International.
Southgate, D. and J. Elgegren. 1995. "Development of Tropical Tim-
ber Resources by Local Communities: A Case Study from the
Peruvian Amazon," Commonwealth Forestry Review 74:2, pp.
142-146.
Southgate, D., with M. Hanrahan, M. Bonifaz, M. Camacho, M. Ca-
rey, and L. Chase. 1992. "The Economics of Agricultural Land
Clearing in Northwestern Ecuador," Institute de Estrategias
Agropecuarias, Quito.
Southgate, D., R. Sierra, and L. Brown. 1991. "A Statistical Analysis
of the Causes of Deforestation in Eastern Ecuador," World De-
velopment 19:9, pp. 1145-1151.
Southgate, D. and M. Whitaker. 1994. Economic Progress and the
Environment: One Developing Country's Policy Crisis. New
York: Oxford University Press.
Spruce, R. 1970. Notes of a Botanist on the Amazon and Andes,
Volume II. London: Johnson Reprint Corp.
Stone, S. 1996. "Economic Trends in the Timber Industry of the
Brazilian Amazon: Evidence from Paragominas" (CREED work-
ing paper 6), International Institute for Environment and De-
velopment, London.
Subler, S. and C. Uhl. 1990. "Japanese Agroforestry in Amazonia:
A Case Study in Tome-Agu, Brazil" in A. Anderson (ed.), Al-
164 References

ternatives to Deforestation: Steps toward Sustainable Use of the

Amazon Rain Forest. New York: Columbia University Press.
Swanson, T. 1995. "Diversity and Sustainability: Evolution, Infor-
mation, and Institutions" in T. Swanson (ed.), Intellectual Prop-
erty Rights and Biodiversity Conservation: An Interdisciplinary
Analysis of the Values of Medicinal Plants. Cambridge: Cam-
bridge University Press.
Tobias, D, and R. Mendelsohn. 1991. "Valuing Ecotourism in a
Tropical Rain Forest Preserve," Ambio 20:2, pp. 91-93.
Tosi, J. 1986. "Natural Forest Management for the Sustained Yield
of Forest Products," Tropical Science Center, San Jose.
Uhl, C., P. Barreto, A. Verissimo, E. Vidal, P. Amaral, A. Barros, C.
Souza, J. Johns, and J. Gerwing. 1997. "Natural Resource Man-
agement in the Brazilian Amazon: An Integrated Approach,"
Bioscience47:3, pp. 160-168.
Uhl, C. and J. Kaufiman. 1990. "Deforestation, Fire Susceptibility,
and the Potential Response of Tree Species to Fire in the Eastern
Amazon," Ecology 71:2, pp. 437-449.
Uhl, C., A. Verissimo, M. Mattos, Z. Brandino, and I. Vieira. 1991.
"Social, Economic, and Ecological Consequences of Selective
Logging in an Amazon Frontier: The Case of Tailandia," Forest
Ecology and Management 46, pp. 243-273.
Umana, A. and K. Brandon. 1992. "Inventing Institutions for Con-
servation: Lessons from Costa Rica" in S. Annis (ed.), Poverty,
Natural Resources, and Public Policy in Central America. New
Brunswick, N.J.: Transaction Publishers.
Uquillas, J., A. Ramirez, and C. Sere. 1991. "Are Modern Agrofo-
restry Practices Economically Viable? A Case Study in the Ecu-
adorian Amazon," Workshop on the Economics of Agroforestry
Systems, the Nitrogen Tree Fixing Association, Honolulu.
Vasquez, R. and A. Gentry. 1989. "Use and Misuse of Forest-
Harvested Fruits in the Iquitos Area," Conservation Biology 3:
4, pp. 350-361.
Verissimo, A., P. Barreto, M. Mattos, R. Tarifa, and C. Uhl. 1992.
"Logging Impacts and Prospects for Sustainable Forest Man-
agement in an Old Amazonian Frontier: The Case of Paragom-
inas," Forest Ecology and Management 55, pp. 169-199.
Verissimo, A., P. Barreto, R. Tarifa, and C. Uhl. 1995. "Extraction
of a High-Value Natural Resource in Amazonia: The Case of
Mahogany," Forest Ecology and Management 72, pp. 39-60.
Vogel, J. 1994. Genes for Sale: Privatization as a Conservation Pol-
icy. New York: Oxford University Press.
von Thiinen, J. 1866. The Isolated State. New York: Pergamin Press.
Weiner, J. 1994. "The Handy-Dandy Evolution Prover," New York
Times Magazine, 8 May, pp. 40-41.
Wells, M. and K. Brandon. 1993. People and Parks: Linking Pro-
tected Area Management with Local Communities. Washing-
ton, D.C.: World Bank.
Whelan, T. 1991. "Ecotourism and its Role in Sustainable Devel-
References 165

opment" in T. Whelan (ed.), Nature Tourism: Managing for the

Environment. Washington, B.C.: Island Press.
Williams, R. 1986. Export Agriculture and the Crisis in Central
America. Chapel Hill: University of North Carolina Press.
Wilson, E. (ed.). 1988. Biodiversity. Washington, B.C.: National
Academy Press.
World Resources Institute (WRI). 1990. World Resources 1990-91.
Washington, B.C.
World Resources Institute (WRI). 1996. World Resources 1996-97.
Washington, B.C.
World Wildlife Fund (WWF). 1996. "First-of-Its-Kind World Forest
Map Reveals 94 Percent of Earth's Forests Have No Formal Pro-
tection" (press release), Washington, B.C.
Zador, M. 1994. "Galapagos Marine Resources Reserve: A Pre-
Investment Analysis for the Parks in Peril Program," The Nature
Conservancy, Washington, B.C.
This page intentionally left blank
Index

AAAS (American Association and sustainable forestry

for the Advancement of projects in Peruvian
Science), 81 Amazon, xii, 72, 75, 77, 80,
acai, 35, 43, 46-48, 56, 57, 123, 82 (see also CSRMP)
124 Amazon Basin
agricultural intensification, 131— and agricultural settlement, 4-
132 5, 19-21, 27-29, 31, 37,
land use impacts of, xiii, 127- 140-141, 142
131, 132-135, 149 (see also agroforestry in, 125-127, 142
Brazil; Chile) deforestation in, xi, 5-8, 18—
See also Ecuador 19,21,25, 31, 129, 137-
agroforestry systems 138, 142
agricultural production in, harvesting of non-timber
125-127 products in, xii, 43—49, 56—
and habitat conservation, 57, 123, 124
125, 149 medicinal plants in, 84-85
and human and social nature-based tourism in, 96
capital's formation, 140,142 timber production in, xii, 7,
nontimber forest products 59-82
from, 48, 49, 53, 56 See also agroforestry systems;
timber production in, 65, 126 Brazil; Ecuador; Peru
(see also laurel) Amgen Inc., 93
See also Brazil; Ecuador;
Peru Baltra Island, 108, 110, 113
aguaje, 45-48, 56, 57 Belaunde-Terry, Fernando, 5, 71
AID (U.S. Agency for Belem, 5, 27, 29
International Development) logging in the vicinity of, 60,
in Ecuadorian Amazon, 126 61, 66
in El Salvador, 144 nontimber products
support for forest product harvesting near, 35, 46, 56,
initiatives, 79, 124 123

167
168 Index

Belem-Brasilia Highway, 7, 25, rural landlessness in, 19 (see

61, 64, 68 also Movimento Sem
Belize, 82, 131 Terra)
Bio Bio River, 26. See also subsidies leading to
Dresdner Bank excessive habitat loss in,
biodiverse habitats, 14-16, 83, 18-20 (see also SUDAM)
86 timber production in, 7-8,59-
as a source of 70, 72, 78 (see also
Pharmaceuticals, xii, 83— IMAZON; mahogany;
84, 89-91, 93-94 Paragominas; Santarem;
threatened with destruction, Tailandia)
5, 6, 10, 83, 89-90 See also Belem; Manaus;
See also Brazil; Ecuador; Medici, Emilio Garrastazu;
genetic prospecting; habitat Para; Parana; Rondonia
conservation; India; Brazil nut, 43
Madagascar; New
Caledonia; Philippines; Cabanas, 147, 148
plant-derived medicines, Canada, 99, 100, 101
value of; Sri Lanka carbon sequestration
bioprospecting. See genetic agreements, 16-17, 21, 30,
prospecting 137-138, 148
Bolivia, 5, 43, 49, 78, 80-81 See also Costa Rica; Norway
Brazil, 29-30, 90, 96, 142 cascarilla roja, 47, 48, 56
agricultural colonists in, 4—5, CBD (Convention on Biological
19, 21-23, 28-29, 141 (see Diversity), 136, 138
also INCRA) Central America, 6, 22, 25-26,
agroforestry in, 48, 49, 65, 64, 99, 133. See also
125, 140, 142 Belize; Costa Rica; El
cattle ranching in, 5, 13, 19— Salvador; Guatemala;
20, 27-28, 44, 45 Honduras; Panama
ethnic Japanese farmers in, CETUR (Ecuadorian Tourism
125, 142 Corporation), 112
deforestation in, xi, 3—5, 6—8, CGIAR (Consultative Group for
13, 23, 25 (see also IBGE; International Agricultural
Londrina; Mato Grosso) Research), 12
land use impacts of Chalatenango, 147, 148
agriculture in, 129, 135 Charles Darwin Research
nontimber forest products in, Foundation and Station,
35, 43-44, 46, 47-48, 49, 109-110, 118
56, 123-124 (see also agai) chicle, 43
property rights development Chile, 26, 131, 146. See also
in, 15-16, 27, 28-30 Bio Bio River
road construction in, 21, 25 China, 9, 117
(see also Belem- Brasilia CI (Conservation International),
Highway; Transamazon 51, 80, 82
Highway) CIDESA (Foundation for
rubber tappers' movement in, Training and Investment
5, 44, 49 (see also Mendes, for Socio-Environmental
Francisco "Chico") Development), 51, 54, 56
Index 169

COFYAL (Yanesha Forestry deforestation, 9, 13-17, 21-22

Cooperative, Limited), 74, in Africa, 7, 12-13
75, 76-78 agricultural returns of, 3-4,
Colombia, 19 22,23, 28
Costa Rica in Asia, 7, 12-13
crime problems in, 103—104 biodiversity loss from, 5, 10,
(see also Puerto Viejo) 14
forestry research and and foreign debt problems,
development in, 73, 78-79 11-12, 26
genetic prospecting and forest fires, 9
agreements, carbon and frontier property
sequestration agreements, arrangements, 23, 27, 142
and debt-for-nature swaps global climatic impacts
in, 17, 90, 91, 105 (see also linked to, 5-6, 9-10, 16,
INBio; Merck; Norway) 137 (see a/so global
nature-based tourism in, xiii, warming)
98-107, 118-119 (see also in Latin America, xi, 5—8, 12—
Canada, United States) 13
park protection and and population and income
management in, 96, 104— growth, 12, 129-131
107, 119-120 (see also and poverty in rural areas,
SPN) 12, 140-141, 147-149 (see
roads in, 103 (see also also CGIAR)
Panamerican Highway; and road construction, xii,
Punta Arenas) 21, 25-26
species and biodiverse and government policies, 17-
habitats in, 93-94, 95, 98 21 (see also Colombia)
(see also quetzal, watershed stability affected
resplendent) by, 8-9, 136-137
See also ICT; La Selva See a/so Brazil; Costa Rica;
Biological Station: Manuel biodiverse habitats;
Antonio National Park; ecosystem depletion;
Monteverde Cloud Forest El Salvador; Ecuador;
Biological Reserve; San Jose; habitat conservation;
Volcan Irazu; Volcan Poas Surinam
CRP (U.S. Conservation Donovan, Richard, 81
Reserve Program), 137 DRNR (Venezuela's Directorate
See a/so SCS of Renewable Natural
CSRMP (Central Service Resources), 91. See also
Resource Management Steyermark, Julian
Project), 72-78. Dresdner Bank, 26
See a/so AID; COFYAL;
FPCN; PEPP; TSC; WWF ecosystem depletion, xii, 22-
23. See also timber
Darwin, Charles, 108 production
Darwin's finch, 108 ecotourism. See nature-based
debt-for-nature swaps, 15, 96, tourism
105, 136. See also Costa Ecuador, 131-132, 141
Rica agroforestry in, 53, 126-127
170 Index

Ecuador (continued) local impacts of tourism in,

biodiverse habitats in, 6, 89— xiii, 113, 114, 116-118,
90, 95, 96, 107-109 119 (see also Veliz,
deforestation in, 3, 23, 25, Eduardo)
27, 142 (see also Guayas nature-based tourism in, xiii,
River Basin) 111-112, 118 (see also
genetic prospecting in, 84,90— Guayaquil; Quito)
91 (see also Pfizer Inc.; over-fishing in, 117 (see also
sangre de drago) sea cucumber; SRP)
and nontimber forest park protection and
products, 49-56 (see also management in, 96,
tagua) 109-110, 119 (see also
nature-based tourism in, ix- INEFAN)
x, 109-120 (see also scientific interest in, 108,109-
INEFAN; Machalilla 110 (see cufso Charles
National Park; Galapagos Darwin Research
National Park and SPNG) Foundation and Station;
and real estate values in, Darwin, Charles; IUCN;
16 UNESCO)
See also INEFAN; Galapagos tourism operators and clients
Islands in, xiii, 32, 97, 110-113,
El Salvador, 143-149. See also 115-116, 119 (see also
Cabanas; Chalatenango; CETUR; Epler, Bruce)
Honduras; Morazan; San See a/so Galapagos National
Salvador; United States Park and SPNG; Isabela
Epler, Bruce, 115 Island
Esmeraldas, 51, 52, 53, 54, 55 Galapagos National Park
and SPNG (Galapagos
Floreana Island, 108, 113 National Park Service),
Ford Foundation, 124 110
forest conservation. See habitat GEF (Global Environmental
conservation Facility), 15, 136, 138
FPCN (Peruvian Foundation for genetic prospecting, 84, 90-94.
the Conservation of See also Costa Rica;
Nature), 75, 77 Ecuador; plant-derived
fur seal, 108 medicines, value of;
Plotkin, Mark
Galapagos Islands, 95, 109, Germany, 118
114 global warming, 5-6, 9-10, 16,
flora and fauna of, 95, 107- 137. See a7so carbon
109 (see also Darwin's sequestration agreements;
finch; fur seal; iguana; land China
tortoise; sea lion) Great Britain, 64, 75
history of human presence GTZ (Germany's Agency for
in, 108-109, 113 (see also Technical Assistance), 79
Baltra Island; Floreana Guatemala, 22, 43, 80
Island; Puerto Ayora; Guayaquil, 111
Puerto Baquerizo Moreno; Guayas River Basin, 3, 4
San Cristobal Island) Gullison, Theodore, 81
Index 171

habitat conservation INBio (Costa Rica's National

and agricultural Biodiversity Institute), 90,
intensification, xiii, 127- 91
131, 132-135, 149 (see a7so INCRA (Brazil's National
Chile) Institute for Colonization
and development, 30, 142, and Agrarian Reform), 28,
147, 149 29
and human capital formation, India, 47, 90
xiii, 30-31, 140-141, 147- Indonesia, 139
149 INEFAN (Ecuadorian Institute
and nature-based tourism, of Forestry, Natural Areas,
xiii, 104-105, 123-125, and Wildlife), 90, 91, 115,
135-138, 149 (see also 116
CBD; carbon sequestration INS (U.S. Immigration and
agreements; debt-for-nature Naturalization Service), 149
swaps; park protection and Iquitos, 45, 46
management) Isabela Island, 108, 113, 117
and sustainable development Italy, 49, 50, 51
of forest resources, 123- IUCN (International Union for
124, 149 the Conservation of
See also Costa Rica; Nature), 109
deforestation; Ecuador;
Galapagos Islands land tortoise, 108-109, 110
Haiti, 146, 147 La Selva Biological Station, 73,
Honduras, 146, 148 101, 103
human capital formation, xiii, laurel, 126
36-39, 140-141, 147-149. logging. See timber production
See also Chile; Ecuador; El Londrina, 3-4
Salvador; Haiti; Malaysia; Los Santos, 22
Thailand
Machalilla National Park, ix-x
IBGE (Brazilian Institute of Madagascar, 83, 90
Geography and Statistics), mahogany, 8, 63-66, 79-81.
6, 7 See also Bolivia; Brazil;
ICDPs (Integrated Conservation shoot-borer moth
and Development Projects), Malaysia, 146
x-xi, xii Manabi, 51, 52, 53, 54, 55
ICT (Costa Rican Tourism Manaus, 29, 56, 61, 96
Institute), 99, 102 Manta, 51, 55
IDE (Inter-American Manuel Antonio National Park,
Development Bank), 26, 98, 101, 105, 107, 120
31, 79 Mato Grosso, 20
IFC (International Finance Medici, Emilio Garrastazu, 4
Corporation), 26 Mendes, Francisco ("Chico"),
iguana, 95, 108 44
IMAZON (Institute for Man and Merck, 90, 91
the Environment of the Mexico, 6, 9, 64, 99, 100. See
Amazon), 59, 60, 64, 66, a7so Quintana Roo
67, 68, 69, 70 Mishana, 45-47
172 Index

Monteverde Cloud Forest engaging in, 43-44, 49, 54,

Biological Reserve, 95,105— 57
107, 119 production outside of forest
nature-based tourism in, 97, settings displacing, xii, 48,
101, 102, 123 49-50
park management in, 104, projects to promote, 44-45,
120, 124 57, 124-125 (see also Ford
Morazan, 147, 148 Foundation; ODA)
Movimento Sem Terra, 19 prospects in accessible
floodplains, 46-47, 56,123-
natural resource values, 32-36. 124
See also biodiversity; See also agai; aguaje; Brazil;
nature-based tourism; cascarilla roja; chicle;
nontimber forest products, Ecuador; Peru; rubber and
harvesting of; plant- rubber tappers; Rubber
derived medicines, value Boom; tagua
of; timber values Norway, 17
nature-based tourism nutrient mining. See ecosystem
adverse environmental depletion
impacts of, x, 95-96
and dependence on habitat ODA (British Overseas
protection, xiii, 104-105, Development Agency), 124
120 OECD (Organization for
growth of, xiii, 95, 99, 110- Economic Cooperation and
113, 118-119 (see also Develpment) countries,
Germany, United States) value of plant-derived
habitat conservation medicines in, 86
resulting from, xiii, 96,
124, 149 Palcazu Valley, 71-80. See a/so
local economic impacts of, COFYAL; CSRMP; PEPP
xiii, 37, 39, 96, 102-103, Panama, 20-21, 22. See also
113-114,116-120, 123 Los Santos
national benefits of, xiii, 99— Panamerican Highway, 98, 103
101, 111-112, 114 Para, 20, 27, 60, 61, 64, 65, 80
scarcity value of sites for, 32, Paragominas, 61, 68, 69
124 (see also Simpson, Parana, 3-4, 21
David) park protection and
See also Costa Rica; management
Galapagos Islands; Manaus; challenge of in Latin
park protection and America, ix, x, 39, 109-
management 110, 118, 123 (see also
New Caledonia, 90 Machalilla National Park)
nontimber forest products, financing of, 15, 90-91, 93-
harvesting of, xii, 47-49, 94, 96-97, 105-107, 115-
53,56-57 116, 136 (see also debt-for-
employment generated by, 43 nature swaps; SPN)
(see also Brazil nut) role of in habitat
poverty of most households conservation strategies, x,
Index 173

118-120, 136, 148-149 (see quetzal, resplendent, 95, 96

also El Salvador; ICDPs) Quintana Roo, 79-80. See also
See also Costa Rica; GTZ
Galapagos Islands; Quito, 51, 55, 111
Galapagos National Park
and SPNG; habitat Rice, Richard, 80-81
conservation; INEFAN; Rondonia, 6, 20, 21, 25, 125,
Manuel Antonio National 140
Park; Monteverde Cloud rosy periwinkle, 83, 93, 94
Forest Biological Reserve; Rubber Boom, 48, 56
nature-based tourism rubber tapping and rubber
Peru tappers, 5, 37, 43, 44, 49,
agricultural colonization in, 56. See also Bolivia; Brazil;
5, 71-72 (see also Mendes, Francisco
Belaunde-Terry, Fernando; ("Chico")
PEPP)
harvesting of nontimber San Cristobal Island, 108, 113
forest product in, 45-47, sangre de drago, 84. See a7so
49 (see also aguaje; Iquitos; Shaman Pharmaceuticals
Mishana] San Jose, 78, 98, 99, 102, 103,
sustainable forestry 104
initiatives in, xii, 60, 70— San Salvador, 146
78, 82 Santarem, 61, 62
traditional agroforestry SCS (U.S. Soil Conservation
systems in, 125 Service), 137
See also COFYAL; CSRMP; sea cucumber, 117
FPCN; Palcazii Valley sea lion, 108
Pfizer Inc., 90, 91 Shaman Pharmaceuticals Inc.,
PEPP (Pichis-Palcazu Special 84
Project), 72 shoot-borer moth, 65
Philippines, 90 Simpson, David, 96
Planafloro Project, 6 social capital formation, xiii,
plant-derived medicines, value 141-142, 144-145, 147,
of, 85-91, 92-94. See also 149. See a7so El Salvador;
Amgen Inc.; OECD Soviet Union, former
countries, value of plant- Soviet Union, former, 141-142
derived medicines in; SPN (Costa Rica's National
Plotkin, Mark; rosy Park Service), 101, 104,
periwinkle; sangre de 105, 106, 107
drago; United States Sri Lanka, 90
Plotkin, Mark, 92-93 SRP (Ecuador's Subsecretariat
property rights development in of Fishery Resources), 117
frontier areas, 27-29, 35 Steyermark, Julian, 91
Puerto Ayora, 108, 110, 113 SUDAM (Brazil's
Puerto Baquerizo Moreno, 108, Superintendency for
113 Amazonian Development),
Puerto Viejo, 98, 104 19-20
Punta Arenas, 98, 103 Surinam, 132
174 Index

sustainable economic policies, 60, 63, 75-76, 78-

development, 139-142. See 79
also Indonesia depressed by illegal logging, 80
social capital investment logging practices and forest
required for, 141-142 management affected by,
69-70, 81-82
tagua, xii, 49-57. See also CI; long-term trends in, 81-82
CIDESA; Esmeraldas; Italy; resource abundance and
Manabi; Manta; Quito factor mobility holding
Tailandia, 61, 66-68, 69 down, xii, 59-60, 63, 67-
Thailand, 146 68, 68-69 (see a7so Rice,
timber production Richard)
adverse environmental transportation and marketing
impacts of, 22, 66-67 costs influencing, 35—36,
development agencies' 63, 75
policies on, 79 (see also TNC (The Nature
AID; IDE World Bank) Conservancy), 31
forest regeneration after, 7, 9, Transamazon Highway, 4—5,
62-63, 64-65, 66, 68, 73- 61,64
74, 81 (see also Gullison, TSC (Tropical Science Center)
Theodore) forestry research in Costa
in forests along agricultural Rica, 78-79
frontiers, 66-70 nature-based tourism
in forests near navigable programs of, 102, 104, 106,
rivers, 7, 8, 60-63 124
gap-phase dynamics Peruviah projects of, 72, 73,
approach to, 72-75, 76-78 76, 77
(see also Palcazu Valley; See also Monteverde Cloud
TSC) Forest Biological Reserve
governmental regulation of,
70, 71, 78-79 UNESCO (U.N. Educational,
outside of natural forests, 65- Scientific, and Cultural
66 Organization), 109, 118
property rights' regimes United States
influencing, 63, 70, 74 ecotourists from, 99, 100,
in remote areas, 63-65 101, 104, 118
techniques for reducing enviromental legislation in, 17
environmental damages of, future timber production in,
59, 63, 69-70, 79-80 (see 81-82
also Donovan, Richard) land use in, 30, 128
See also agroforestry systems; migration from El Salvador
Bolivia; Brazil; CSRMP; to, 143, 148 (see also INS)
Great Britain; IMAZON; plant-derived
mahogany; Quintana Roo; Pharmaceuticals used in,
timber values; United 86, 87
States soil conservation programs
timber values of, 137 (see also CRP)
affected by government tropical timber buyers in, 64
Index 175

World War II air base in Washington, B.C., 26

Galapagos, 110 (see also World Bank
Baltra) environmental programs and
See also AID projects of, 6, 15, 104, 138
(see also PlanafLoro Project)
vegetable ivory. See tagua environmental review
Veliz, Eduardo Veliz, 117 requirements of, 26, 31, 79
Venezuela, 91, 142. See also See also GEF, IFC
DRNR WWF (World Wildlife Fund),
Volcan Irazu, 98, 101, 103, 75, 77
104
Volcan Poas, 98, 101, 103, 104, Yanesha Indians. See COFYAL;
105, 107 Palcazii Valley; PEPP